An affiliate of Elsevier Inc. © 2006, Elsevier Inc. All rights reserved. First published 2006 First edition 1982 Second edition 1990 Third edition 1996 Fourth Edition 2003 No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the Publishers. Permissions may be sought directly from Elsevier’s Health Sciences Rights Department, 1600 John F. Kennedy Boulevard, Suite 1800, Philadelphia, PA 19103-2899, USA: phone: (+1) 215 239 3804; fax: (+1) 215 239 3805; or, e-mail:
[email protected]. You may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com), by selecting ‘Support and contact’ and then ‘Copyright and Permission’. ISBN-13: 978-1-4160-3258-8 ISBN-10: 1-4160-3258-4 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress
Notice Medical knowledge is constantly changing. Standard safety precautions must be followed, but as new research and clinical experience broaden our knowledge, changes in treatment and drug therapy may become necessary or appropriate. Readers are advised to check the most current product information provided by the manufacturer of each drug to be administered to verify the recommended dose, the method and duration of administration, and contraindications. It is the responsibility of the practitioner, relying on experience and knowledge of the patient, to determine dosages and the best treatment for each individual patient. Neither the Publisher nor the author assume any liability for any injury and/or damage to persons or property arising from this publication. The Publisher
Printed in Canada Last digit is the print number: 9
8
7
6
5
4
Working together to grow libraries in developing countries www.elsevier.com | www.bookaid.org | www.sabre.org
3
2
1
To the memory of Ralph Wright, MA, MD, DPhil, FRCP. A leading first generation hepatologist, clinical scientist, teacher and father of one of the editors.
Contributors Paul C Adams MD Professor of Medicine Department of Medicine London Health Sciences Centre – University Campus London, ON, Canada Aijaz Ahmed MD Gastroenterology Fellow Hepatology Stanford University Medical Center Palo Alto, CA, USA Karl E Anderson MD Professor of Preventive Medicine and Community Health, Department of Preventive Medicine and Community Health University of Texas Medical Branch Galveston, TX, USA Miguel R Arguedas MD MPH Assistant Professor Gastroenterology – School of Medicine Birmingham, AL, USA Vicente Arroyo MD Professor of Medicine, Director Institute for Digestive Disease Hospital Clinic i Provincial Universitat de Barcelona Barcelona, Spain Veronica A Arteaga MD Doctor of Medicine Department of Surgery Cedars-Sinai Medical Center Los Angeles, CA; Clinical Researcher New England Hepatobiliary Disease Center Dartmouth-Hitchcock Medical Center Lebanon, NH, USA
Heike Bantel MD Department of Gastroenterology and Hepatology Hannover Medical School Hannover, Germany
Jordi Bruix MD Director Liver Unit IMD Hospital Clinic Provincial Catalonia, Spain
Angeline Bartholomeusz MD Research and Molecular Development Victorian Infectious Diseases Reference Laboratory North Melbourne, VIC, Australia
Kathleen M Campbell MD Assistant Professor Division of Gastroenterology, Hepatology and Nutrition Cincinnati Children’s Hospital Medical Center Cincinnati, OH, USA
Marina Berenguer MD Adjunct Professor Department of Digestive Medicine Hospital Universitari La Fe Valencia, Spain Annika Bergquist MD PhD Clinical Assistant Department of Gastroenterology and Hepatology Karolinska University Hospital Huddinge, Sweden Henri Bismuth MD Director Henri Bismuth Hepatobiliary Institute Villejuif, France Herbert L Bonkovsky MD Director of Clinical Research, the Lowell P. Weicker, Jr. General Clinical Research Center, and the Clinical Trials Unit Professor of Medicine and Molecular, Microbial, and Structural Biology Farmington, CT, USA Thomas D Boyer MD Director Arizona Liver Institute University of Arizona Tucson, AZ, USA
Rizwan Aslam MD ChB MRCP FRCR Assistant Clinical Professor Department of Radiology University of California, San Francisco San Francisco, CA, USA
Ulrika Broomé MD PhD Associate Professor of Medicine Department of Gastroenterology and Hepatology Karolinska University Hospital Huddinge, Sweden
Soon Koo Baik MD Associate Professor of Medicine Department of Medicine Yonsei University Wonju College of Medicine Wonju, Korea
Robert S Brown MD MPH Chief of Clinical Hepatology Medical Director Center for Columbia University College of Physicians and Surgeons New York, NY, USA
William F Balistreri MD Director Division of Paediatric Gastroenterology / Nutrition University of Cincinnati Children’s Hospital Medical Centre Cincinatti, OH, USA
Concepció Bru MD Senior Consultant, Associate Professor of Radiology BCLC Group Diagnosis Imaging Centre Barcelona, Spain
Martin Caselitz MD Consultant Medical Clinic II Klinikum Deggendorf Deggendorf, Germany John P Cello MD Professor of Medicine and Surgery Division of Gastroenterology, Hepatology and Clinical Nutrition San Francisco General Hospital San Francisco, CA, USA Naga Chalasani MD Associate Professor of Medicine Division of Gastroenterology/ Hepatology Indiana University School of Medicine Indianapolis, IN, USA Judy Chang BSc Department of Microbiology and Immunology The University of Melbourne Parkville, VIC, Australia Linda J Chen MD Clinical Instructor Division of Transplantation, Department of Surgery Stanford University School of Medicine Stanford, CA, USA Xin Chen PhD Assistant Professor Department of Biopharmaceutical Sciences University of California, San Francisco San Francisco, CA, USA Massimo Colombo MD Professor of Gastroenterology and Endocrinology Maggiore Hospital and University of Milan Milan, Italy Diane W Cox PhD CCMG FRSC Professor and Chair Department of Medical Genetics University of Alberta Edmonton, AB, Canada
ix
Contributors
Oscar W Cummings MD Associate Professor of Pathology Department of Pathology and Laboratory Medicine Indiana University School of Medicine Indianapolis, IN, USA John T Cunningham MD Professor of Clinical Medicine Section of Gastroenterology and Hepatology University of Arizona School of Medicine Tucson, AZ, USA Christopher P Day MD PhD FRCP Professor of Liver Medicine Department of Gastroenterology and Hepatology The Medical School University of Newcastle upon Tyne Newcastle upon Tyne, UK Laurie D DeLeve MD PhD Professor of Medicine Division of Gastrointestinal and Liver Diseases University of Southern California Keck School of Medicine Los Angeles, CA, USA R Brian Doctor PhD Associate Professor Division of Gastroenterology and Hepatology Department of Medicine University of Colorado Health Sciences Center Denver, CO, USA Scott A Elisofon MD Advanced Hepatology Fellow Division of Gastroenterology Hunnewell Ground Children’s Hospital Boston Boston, MA, USA Eric Esrailian MD MPH Clinical Instructor of Medicine Division of Digestive Diseases David Geffen School of Medicine at UCLA Los Angeles, CA, USA Carlos O Esquivel MD PhD The Arnold and Barbara Silverman Professor of Pediatric Transplantation Professor of Surgery and Chief, Division of Transplantation Stanford University School of Medicine Stanford, CA, USA Gregory T Everson MD Professor of Medicine; Director of Hepatology University of Colorado School of Medicine and Health Sciences Denver, CO, USA Michael B Fallon MD Associated Professor of Medicine Med – Gastroenterology University of Alabama at Birmingham Birmingham, AL, USA Diana M Flynn MB Bch Consultant in Paediatric Gastroenterology John Radcliffe Hospital Oxford, UK
x
Scott L Friedman MD Professor of Medicine Division of Liver Diseases Mount Sinai School of Medicine New York, NY, USA Hans Fromm MD Professor of Medicine; Director Dartmouth-Hitchcock Hepatopancreaticobiliary Disease Center Lebanon, NH, USA Paul J Gaglio MD Associate Clinical Professor of Medicine (in Surgery) Center for Liver Disease and Transplantation Columbia University College of Physicians and Surgeons New York, NY, USA Guadalupe Garcia-Tsao MD Professor of Medicine Section of Digestive Diseases Yale University School of Medicine Yale, CT, USA Fayez K Ghishan MD Horace W. Steele Endowed Chair in Pediatric Research Professor and Head, Department of Pediatrics Director, Steele Memorial Children’s Research Center University of Arizona Health Sciences Center Tucson, AZ, USA Kathleen M Giacomini PhD Professor and Chair Department of Biopharmaceutical Sciences University of California, San Francisco San Francisco, CA, USA Pere Ginés MD Associate Professor of Medicine Liver Unit Institute for Liver Research Hospital Clinic i Provincial Universitat de Barcelona Barcelona, Spain
Mónica Guevara MD Associate Investigator Liver Unit Institute of Digestive Diseases Hospital Clinic Institut d’Investigacions Biomediques (IDIBAPS) Barcelona, Spain Françoise Imbert-Bismut MD Department of Biochemistry Hôpital de la Salpétrière Paris, France Sanjeev Gupta MD Professor of Medicine and Pathology Departments of Medicine and Pathology Albert Einstein College of Medicine Bronx, NY, USA Elizabeth J L Heathcote MD Professor of Medicine Toronto Hospital (Western) Toronto, ON, Canada Kristel Hunt MD Gastroenterology Fellow Division of Liver Diseases Mount Sinai School of Medicine New York, NY, USA John Hunter MD Professor and Chairman Department of Surgery, L223 Oregon Health Services University Portland, OR, USA Hartmut W Jaeschke PhD Professor of Pharmacology Associate Director Liver Research Institute University of Arizona College of Medicine Tucson, AZ, USA Peter L M Jansen MD Professor of Medicine Liver Center Division of Gastroenterology and Hepatology Academic Medical Center Amsterdam, The Netherlands Birgir Johannsson MD Infectious Disease Fellow Division of Infectious Diseases The University of Iowa Iowa City, IA, USA
Steven Goldschmid MD Associate Professor of Clinical Medicine Chief Section of Gastroenterology and Hepatology University of Arizona School of Medicine Tucson, AZ, USA
Maureen M Jonas MD Associate Professor of Pediatrics Division of Gastroenterology Harvard Medical School Children’s Hospital Boston, MA, USA
Gregory J Gores MD Professor of Medicine Mayo Medical School Rochester, MN, USA
Dean P Jones PhD Professor of Medicine Department of Biochemistry Emory University Atlanta, GA, USA
Albert K Groen PhD Associate Professor Department of Medical Biochemistry Academic Medical Center Amsterdam, The Netherlands
Emmet B Keeffe MD Professor of Medicine, Chief of Hepatology, Co-Director Stanford University Medical Center Palo Alto, CA, USA
Contributors
Deirdre A Kelly MD FRCP FRCPI FRCPH MB BA Professor of Paediatric Hepatology The Liver Unit Birmingham Children’s Hospital Birmingham, UK Percy A Knolle MD Professor of Molecular Medicine and Immunology Institute for Molecular Medicine and Experimental Immunology University of Bonn Bonn, Germany Jina Krissat MD Surgical Registrar Hepatobiliary and Pancreatic Surgery Unit The Royal London Hospital London, UK Manoj Kumar MD DM Senior Research Associate Department of Gastroenterology GB Pant Hospital New Delhi, India Douglas R LaBrecque MD Director, Liver Services Internal Medicine Liver Services University of Iowa Hospitals and Clinics Iowa City, IA, USA Konstantinos Lazaridis MD Assistant Professor of Medicine Center for Basic Research in Digestive Diseases Division of Gastroenterology and Hepatology Mayo Clinic College of Medicine Rochester, MN, USA Samuel S Lee MD Professor of Medicine Liver Unit Department of Medicine University of Calgary Calgary, AB, Canada Jay H Lefkowitch MD Professor of Clinical Pathology College of Surgeons and Physicians of Columbia University New York, NY, USA Riccardo Lencioni MD Associate Professor of Radiology Department of Oncology, Transplants and Advanced Technologies in Medicine Division of Diagnostic and Interventional Radiology University of Pisa Pisa, Italy Sharon Lewin MD Director, Infectious Diseases Unit; Professor, Department of Medicine Monash University The Alfred Hospital Melbourne, VIC, Australia
Keith D Lindor MD Professor of Medicine Division of Gastroenterology and Hepatology Mayo Medical School, Clinic and Foundation Rochester, MN, USA
Brent Neuschwander-Tetri MD Professor of Internal Medicine Division of Gastroenterology and Heptology Saint Louis University Liver Center St. Louis, MO, USA
Josep M Llovet, MD Senior Research Associate BCLC Group. Liver Unit, Digestive Disease Institute Mount Sinai School of Medicine Barcelona, Spain
Matthew Nichols MD Fellow in Gastroenterology and Hepatology University of Colorado Health Sciences Center Denver, CO, USA
Stephen Locarnini MD Divisional Head Research & Molecular Development Victorian Infectious Diseases Reference Laboratory North Melbourne, VIC, Australia Robert S McCuskey PhD Professor and Head of Cell Biology and Anatomy; Professor of Physiology; Professor of Pediatrics Department of Cell Biology and Anatomy College of Medicine, University of Arizona Tucson, AZ, USA Michael P Manns MD Professor of Medicine Head, Department of Gastroenterology and Hepatology Zentrum Innere Medizin and Dermatologie Medizinische Hochschule Hannover Hannover, Germany Enrique J Martinez MD FACP Associate Professor of Clinical Medicine Center for Liver Diseases University of Miami Miami, FL, USA Darius Moradpour MD Associate Professor of Medicine Division of Gastroenterology and Hepatology Centre Hospitalier Universitaire Vaudois Lausanne, Switzerland Kevin D Mullen MB FRCPI Professor of Medicine at Case Western Reserve University; Consultant of Gastroenterology Gastroenterology and Hepatology Division MetroHealth Medical Center Cleveland, OH, USA Satheesh Nair MD Medical Director of Liver Transplantation Ochsner Clinic Foundation New Orleans, LA, USA Russell Nash MD Assistant Professor Department of Pathology University of Colorado Health Sciences Center Denver, CO, USA James Neuberger DM FRCP Consultant Physician, Professor of Medicine Liver Unit Queen Elizabeth Hospital Birmingham, UK
David H Perlmutter MD Professor Pediatrics/Cell Biology Department of Pediatrics Washington University Pittsburgh, PA, USA Robert P Perrillo MD Director, Gastroenterology and Hepatology Section of Gastroenterology and Hepatology Ochsner Clinic Foundation New Orleans, LA, USA Thierry Poynard MD PhD Professor of Medicine Department of Hepato-Gastroenterology University of Paris VI Paris, France Jorge Rakela MD Professor of Medicine Department of Internal Medicine Mayo Clinic Scottsdale, AZ, USA Charles M Rice PhD Maurice R and Corinne P Greenberg Professor; Head, Laboratory of Virology and Infectious Disease; Scientific and Executive Director Center for the Study of Hepatitis C The Rockefeller University New York Presbyterian Hospital New York, NY, USA Mario Rizzetto MD Professor of Gastroenterology Department of Gastroenterology University of Torino – Molinette Torino, Italy Eve A Roberts MD FRCPC Professor of Paediatrics, Medicine and Pharmacology Division of Gastroenterology, Hepatology and Nutrition The Hospital for Sick Children Toronto, ON, Canada Don C Rockey MD Chief, Division of Digestive and Liver Diseases; Professor of Medicine University of Texas at Southwestern Medical Center Dallas, TX, USA Juan Rodés MD Professor of Medicine Hospital Clinic Barcelona, Spain
xi
Contributors
Hector Rodriguez-Luna MD Associate Consultant Division of Transplantation Medicine Mayo Clinic Phoenix, AZ, USA
Maria H Sjogren MD MPH Chief Department of Clinical Investigation Walter Reed Army Medical Center Washington, DC, USA
Sammy Saab MD MPH Associate Professor of Medicine and Surgery Division of Digestive Diseases David Geffen School of Medicine at UCLA Los Angeles, CA, USA
Jack T Stapleton MD Professor and Director Division of Infectious Diseases The University of Iowa and The Iowa City VA Medical Center Iowa City, IA, USA
Arun J Sanyal MBBS MD Charles Caravati Professor of Medicine Chairman: Division of Gastroenterology, Hepatology and Internal Medicine Medical College of Virginia Richmond, VA, USA S K Sarin MD DM FNA FNASc President, Asian Pacific Association Study of Liver; Adjunct Professor, Molecular Medicine, JNU; Professor and Head Department of Gastroenterology GB Pant Hospital New Delhi, India Thomas D Schiano MD Associate Professor of Medicine; Medical Director, Adult Liver Transplantation; Director of Clinical Hepatology Division of Liver Diseases Mount Sinai School of Medicine New York, NY, USA Leonard B Seeff MD Senior Scientist for Hepatitis Research Liver Disease Research Branch National Institute of Diabetic Digestive and Kidney Diseases (NIDDK) Bethesda, MD, USA Shobha Sharma MD Assistant Professor of Pathology Department of Pathology Emory University Hospital Atlanta, GA, USA Steven I Shedlofsky MD Marcos Lins Andrade Professor Division of Digestive Diseases and Nutrition University of Kentucky Lexington, KY, USA Oren Shibolet MD Lecturer in Medicine The Liver Unit Hadassah University Hospital Jerusalem, Israel Daniel Shouval MD Professor of Medicine; Director, Liver Unit Hadassah University Hospital Jerusalem, Israel
xii
Stephen F Stewart MBChB BSc PhD MRCP Consultant Hepatologist Department of Gastroenterology and Hepatology School of Clinical Medical Sciences The Medical School University of Newcastle upon Tyne Newcastle upon Tyne, UK Doris B Strader MD Associate Professor of Medicine Division of Gastroenterology/Hepatology Fletcher Allen Health Care Burlington, VT, USA
Dominique-Charles Valla MD Professor of Hepatology, University of Paris 7 Head, Federation Medico-Chirurgicale d’HepatoGastroentologie Hospital Beaujon Paris, France Rebecca W Van Dyke MD Professor of Medicine Gastroenterology Division University of Michigan School of Medicine Ann Arbor, MI, USA Hugo E Vargas MD Associate Professor of Medicine Mayo Clinic College of Medicine Phoenix, AZ, USA Siegfried Wagner MD Professor and Physician-in-Chief Medical Clinic II Klinikum Deggendorf Deggendorf, Germany
Christian P Strassburg MD Assistant Professor in Experimental Gastroenterology (Privatdozent) Department of Gastroenterology, Hepatology and Endocrinology Hannover Medical School Hannover, Germany
Jack R Wands MD Jeffrey and Kimberly Greenberg-Artemis and Martha Joukowsky Professor in Gastroenterology and the Professor of Medical Science Brown University; Director, Division of Gastroenterology and the Liver Research Center Rhode Island Hospital Providence, RI, USA
R Todd Stravitz MD Associate Professor of Medicine Section of Hepatology Virginia Commonwealth University Richmond, VA, USA
C Mel Wilcox MD Professor of Medicine Division of Gastroenterology and Hepatology University of Alabama at Birmingham Birmingham, AL, USA
Yee-Li Sun MD Resident Department of Radiology University of California, San Francisco San Francisco, CA, USA
Teresa L Wright MD Staff Physician Veterans’ Affairs Medical Center San Francisco, CA, USA
Jayant A Talwalkar MD MPH Assistant Professor of Medicine Mayo Clinic College of Medicine Rochester, MN, USA Ruedi Thoeni MD Chief GI Radiology University of California, San Francisco San Francisco, CA, USA Christian Trautwein MD Professor of Medicine Department of Gastroenterology, Hepatology and Endocrinology Zentrum Innere Medizin Hannover, Germany Daniel Tseng MD Minimally Invasive Surgery Fellow Oregon Health and Science University Portland, OR, USA
Judy Yee MD Associate Professor and Vice Chairman Department of Radiology University of California, San Francisco; Chief, Radiology Service Veterans’ Affairs Medical Center San Francisco, CA, USA Mahmoud M Yousfi MD Assistant Professor of Medicine Department of Internal Medicine Mayo Clinic Scottsdale, AZ, USA
Preface
The first edition of Hepatology appeared in 1982. The creation of the book came from a desire of Drs Zakim and Boyer to write a textbook on liver disease in which both pathophysiology and clinical material were presented in a manner that allowed the reader to understand current issues but, more importantly, to prepare the reader for new developments in the discipline of hepatology. David Zakim was the driving force behind the first edition and his expertise in understanding the basic science behind the clinical diseases was critical to the success of the book. With his leadership Hepatology met the goals set in the first and subsequent editions. David Zakim has now retired from academic medicine and for this book to continue to be one of the leading books in the field of liver disease it was essential that new editors be added. Drs Teresa Wright and Michael Manns are world leaders in hepatology with broad clinical expertise and outstanding research credentials. They continue the tradition established by Dr Zakim that a textbook should help in understanding a patient’s disease by providing fundamental knowledge of the pathophysiology of the disease process. Thus, their addition as editors allows Zakim and Boyer’s Hepatology to continue to evolve in this fifth edition. With the advent of the electronic age and ready availability of summaries of published works there has been a move away from reading source material and an increasing reliance on opinion articles frequently published with the support of the pharmaceutical industry. These ‘reviews’ are brief and cover the essentials but lack depth. In this environment books such as Zakim and Boyer’s Hepatology play an increasingly important role. The authoritative chapters in the book cover a subject in depth and give the reader both the basic and clinical information they need to grasp the area of interest. With a thorough understanding of an area gained by reading one or two chapters, the reader can then better understand new arti-
cles and judge the quality of ‘reviews’. Hepatology and the figures will be available on the Elsevier web site in a searchable and downloadable format to those who purchase the book. This edition continues the tradition of changing authors and adding new chapters to cover areas not previously discussed in detail in the previous edition. Fifty-one of the chapters have new or additional authors compared to the previous edition. We also have added 20 new chapters that cover subjects such as hepatotoxicity from herbal preparations, pediatric viral hepatitis and pharmacogenomics that were not in the previous edition. These new chapters reflect our commitment to keeping Zakim and Boyer’s Hepatology current and providing the reader with the latest advancements in liver disease as well as the influence of the new editors. Although we continue to have chapters on basic pathophysiology, we are focusing more and more on the clinical aspects of the discipline of hepatology. This evolution of the past 25 years reflects the advances in the field and the increasing number of treatment options available to the practicing physician for the management of hepatobiliary disorders. Lastly, the presentation of the book has been enhanced by drawing many of the figures in color, making the tables more pleasing to the eye and placing the color plates within the chapters rather than grouping them throughout the book. We believe these latter changes will make the book easier to use and more readable. The editors hope that this book will be of help to modern hepatologists and gastroenterologists around the world, as well as to physicians of other specialties and fellows in training, in order to increase their knowledge for the benefit of their patients. Thomas D. Boyer Teresa L. Wright Michael P. Manns
xiii
Acknowledgements A book of this size involves numerous individuals in its creation. We are grateful to all of the contributors for their timely delivery of the manuscripts and the excellence of the chapters that they have written. This book would never have been published without the professional staff at Elsevier. Karen Bowler has been instrumental in the creation of this edition of Zakim and Boyer’s Hepatology. She was there at the first meetings with the new editors and has been there to solve problems and encourage all of us to make the best book possible. Claire Bonnett and Kathryn Mason have been of
equal importance during the production of Zakim and Boyer’s Hepatology. Their work has lead to a book with a new and exciting appearance, from the cover by illustrator Richard Tibbitts, to the tables, to the high quality color plates that are now distributed within the chapters. We realize that there are numerous other people at Elsevier who have contributed to Zakim and Boyer’s Hepatology and we are all grateful for their efforts as well. Lastly, we would like to thank our families who have supported us during the genesis of this book.
xv
Section I: Pathophysiology of the Liver
1
ANATOMY OF THE LIVER Robert S. McCuskey Abbreviations CGRP calcitonin gene-related peptide Da daltons DNA deoxyribonucleic acid GERL granules or secondary lysosomes HD high density HMS hepatic microvascular subunits ICAM intercellular adhesion molecule
IgA LAL LD LGL IL NK nNOS
immunoglobulin A liver-associated lymphocytes low density large granular lymphocytes interleukin natural killer neuronal nitric oxide
INTRODUCTION OVERVIEW OF THE STRUCTURE AND FUNCTION OF THE LIVER The liver is the largest organ in the body. In humans it is separated incompletely into lobes, covered on their external surfaces by a thin connective tissue capsule. The liver is composed of several cell types, which interact with each other but are adapted to performing specific functions. The principal cell type is the hepatic parenchymal cell, loosely referred to as the hepatocyte, which accounts for 60% of the total cell population and 80% of the volume of the organ. Hepatocytes are organized into plates or laminae that are interconnected to form a continuous three-dimensional lattice (Figure 1-1). Between the plates of hepatocytes are spaces occupied by hepatic sinusoids, the large-bore fenestrated capillaries of the liver that nourish each parenchymal cell on several sides (Figure 1-1). The sinusoidal space, and non-parenchymal cells associated with sinusoids, comprises the majority of the remaining liver volume. The non-parenchymal cells include sinusoidal endothelial cells, perisinusoidal stellate cells (fat-storing cells of Ito), and intraluminal Kupffer cells. An interconnecting network of minute intercellular channels form bile canaliculi, which course between adjacent hepatocytes (Figure 1-1A). These receive the bile secreted from hepatocytes and then drain through short bile ductules (cholangioles) partially lined by cuboidal epithelial cells to bile ducts. Hepatocytes carry out most of the functions generally associated with the liver. They extract and process nutrients and other materials from the blood, and they produce both exocrine and endocrine secretions, as follows:
Bile Synthesis and Secretion Hepatocytes synthesize bile acids from cholesterol; these function in the lumen of the small intestine to emulsify fats. Bilirubin, a toxic metabolite generated from the breakdown of hemoglobin, is excreted by hepatocytes as follows. Insoluble bilirubin is produced as a byproduct of red blood cell breakdown in the spleen; it circulates in the blood complex to albumin, and is taken up from the blood hepatocytes, conjugated to a soluble form, then secreted into bile canaliculi.
NPY RER SER SOM SP Tf VIP
neuropeptide Y rough endoplasmic reticulum smooth endoplasmic reticulum somatostatin substance P transferrin vasoactive intestinal peptide
Protein Synthesis Hepatocytes synthesize proteins for hepatic and non-hepatic use. Proteins for hepatic use include a wide variety of liver-specific enzymes that carry out the many synthetic and detoxifying functions of the liver. Proteins secreted by hepatocytes include all of the major plasma proteins except immunoglobulins (synthesized by plasma cells), e.g. albumin, transferrin, prothrombin, fibrinogen, lipoproteins and complement proteins.
Glucose Homeostasis Hepatocytes help to maintain blood glucose levels. In response to pancreatic islet hormones hepatocytes synthesize glycogen from glucose or break down glycogen and release glucose (glycogenolysis); hepatocytes can also synthesize glucose from other sugars (e.g. fructose) and from amino acids (gluconeogenesis).
Metabolism of Drugs and Toxins Hepatocyte enzymes metabolize drugs and toxins delivered to the liver from the gut via the portal circulation. The functions of the hepatic non-parenchymal cells are:
Kupffer Cells • Phagocytosis of bloodborne toxicants and particulates such as bacteria from the circulation. • Secretion of mediators (e.g. inflammatory mediators) that affect the function of adjacent cells and cells in distant sites. • Production of beneficial and toxic substances that contribute to host defense as well as liver injury.
Sinusoidal Endothelial Cells These form a leaky barrier between the parenchymal cells and the blood flowing in sinusoids. The endothelial cells are fenestrated and act as a sieve to prevent red blood cells and other cellular components from interacting with hepatocytes, while allowing rapid access to the other substances in the blood.
Stellate Cells • Storage of vitamin A and other fat-soluble vitamins. • Stellate cells when activated synthesize collagen, thus they are important in the development of cirrhosis.
3
Section I. Pathophysiology of the Liver
A
B
Figure 1-1. Laminae of hepatic parenchymal cells (H) interconnected to form a three-dimensional lattice containing a labyrinth of spaces occupied by sinusoids (S). BC, bile canaliculus; KC, Kupffer cell. (Modified from McCuskey RS. Functional morphology of the liver with emphasis on its microvasculature. In: Tavoloni N, Berk PD, eds. Hepatic transport and bile secretion. New York, Raven Press, 1993: 2, ©1993, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
The structure of the liver at the tissue, cellular and molecular levels has evolved to subserve the above functions and is the subject of the remainder of this chapter.
GROSS ANATOMY The mature liver lies mainly in the right upper quadrant of the abdominal cavity, is attached to the diaphragm, and is protected by the ribcage. Its morphology has been extensively reviewed.1,2 Briefly, in adults the healthy liver weighs approximately 1500 g3 and extends along the midclavicular line from the right fifth intercostal space to just inferior to the costal margin. From there, the anterior border of the liver extends medially crossing the midline just inferior to the xiphoid process. A small portion of the organ projects across the midline and lies in the upper left abdominal quadrant. The liver has a dual blood supply which enters the liver at its hilus (porta hepatis) accompanied by the hepatic bile duct, lymphatics and nerves. Approximately 80% of the blood entering the liver is poorly oxygenated and is supplied by the hepatic portal vein. This is the venous blood flowing from the intestines, pancreas, spleen, and gallbladder. The remaining 20% of the blood supply is well oxygenated and delivered by the hepatic artery. Anatomically the liver is divided into right and left lobes by the falciform ligament, which is a peritoneal fold connecting the liver to the anterior abdominal wall and the diaphragm (Figure 1-2). The right lobe is further subdivided inferiorly and posteriorly into two smaller lobes – the caudate and quadrate lobes. The functional division, however, is a plane that passes through the gallbladder and
4
inferior vena cava which defines the halves of the liver supplied by the right and left branches of the portal vein and hepatic artery, together with biliary drainage into the right and left hepatic ducts.4 As a result, the quadrate lobe and a large portion of the caudate lobe belong functionally to the left hemiliver. Further functional subdivision of the liver into eight segments having independent vascular and biliary supplies has been reported (Figure 1-3) and is important when liver resection is required.1,2,5,6 The liver is encapsulated by a thin connective tissue layer (Glisson’s capsule) consisting mostly of regularly arranged type I collagen fibers, scattered type III fibers, fibroblasts, mast cells, and small blood vessels. On the surfaces facing the abdominal cavity this connective tissue layer is covered by the simple squamous mesothelial cells of the peritoneal lining. At the attachment of the falciform ligament to the liver, the two leaves of the ligament separate to form an area devoid of peritoneum, the ‘bare area’, on the superior surface of the liver. The right and left leaves of the falciform then merge with reflections of the peritoneum coming off the diaphragm, forming respectively the triangular and the coronary ligaments.
DEVELOPMENT OF THE LIVER The development of the liver has been extensively described1,2,7,8 and is illustrated in Figure 1-4. Briefly, the liver primordium appears in human embryos during the third week of gestation as an endodermal bud from the ventral foregut just cranial to the yolk sac. This bud becomes the hepatic diverticulum as it enlarges, elongates, and develops a cavity contiguous with the foregut. The hepatic
Chapter 1 ANATOMY OF THE LIVER
Diaphragmatic area Right lobe
Left triangular ligament
Coronary ligament
Lesser omentum
Right triangular Inferior vena cava Bare area ligament
Caudate lobe Left lobe
Renal area Portal vein
Porta hepatis
Hepatic duct Pyloric area
Cystic duct
Falciform ligament Colic areas Ligamentum teres
Falciform ligament Gallbladder Ligamentum teres
Gallbladder
Quadrate lobe
Figure 1-2. Lobes, surfaces, and ligaments of the liver viewed anteriorly (left) and from a posteroinferior perspective (right). (Modified from Moore KL, Dalley AF. Clinically oriented anatomy, 4th edn. Philadelphia: LWW, 1999: 264, ©1999, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
Inferior vena cava Left and middle hepatic veins Right hepatic vein
Figure 1-3. Segmentation of the liver based on principal divisions of the portal vein and hepatic artery. (Modified from Moore KL, Dalley AF. Clinically oriented anatomy, 4th edn. Philadelphia: LWW, 1999: 268, © 1999, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
II VII
VIII
I IV III
VI
Right and left branches of hepatic artery
V
Portal vein Gallbladder
Bile duct
Portal triad
Hepatic artery
diverticulum grows into the septum transversum, a plate of mesenchyme that incompletely separates the pericardial and peritoneal cavities, and separates into an hepatic portion that forms the hepatic parenchymal cells as well as the intrahepatic bile ducts, a cystic portion that forms the gall bladder, and a ventral portion that forms the head of the pancreas.
During the fourth week of development, buds of epithelial cells extending from the hepatic diverticulum into the mesenchyme of the septum transversum as thick, anastomosing cords several cells thick become interspersed within the developing anastomotic network of capillaries arising from the vitelline veins, thus beginning to establish the close relationship of hepatic parenchymal cells to
5
Section I. Pathophysiology of the Liver
A
Mesenchymal lining of coelomic tract
B Common cardinal vein
Sinus venosus
Left hepato-cardiac channel
Liver
Right hepato-cardiac channel
Left umbilical vein
Left umbilical vein
Septum transversum
Hepatic sinusoidal plexus Vitelline veins
C
Gut
D Inferior vena cava
Sinus venosus
Right umbilical vein
Right hepatic vein Hepatic sinusoidal plexus
Middle hepatic vein Left hepatic vein Ductus venosus (Dia. = 600μm)
Left umbilical vein Origin of hepatic bud Portal vein (Dia. = 100μm)
Left umbilical vein (Dia. = 600μm)
Right and left viteline veins Figure 1-4. Development of the liver. A Section through the region of the hepatic bud of a human embryo of 25 somites (26 days). B Vascular channels associated with the developing liver in a human embryo of 30 somites. C Vascular channels at a later stage showing development of the sinusoidal network. D Portal hepatic circulation in a human embryo of 17 mm (7 weeks). (Reproduced from MacSween RNM, Desmet VJ, Roskams T, Scothorne RJ. Functional morphology of the liver with emphasis on its microvasculature. In: MacSween NM, Burt AD, et al., eds. Pathology of the liver, 4th edn. London: Churchill Livingstone, 2002: 4, ©2002, with permission of Elsevier.)
the sinusoids.9 The anastomotic pattern of both multicellular cords of parenchymal cells and sinusoids persists until several years after birth, by which time cords two or more parenchymal cells thick bounded on several sides by sinusoids have become plates consisting of single parenchymal cells bounded on at least two sides by sinusoids, particularly in the centrilobular region.10,11 By 7 weeks the vitelline veins unite to form the portal vein. The hepatic artery is derived from the celiac axis and its ingrowth into the hepatic primordium closely follows that of the bile ducts.12–14 Between the sixth week and birth the fetal liver serves as a hematopoietic organ and as the primary site for fetal blood formation until the third trimester, when most hemopoietic sites disappear as the bone marrow develops.
6
MICROSCOPIC ANATOMY VASCULATURE, BILIARY SYSTEM, INNERVATION Vasculature Both the hepatic portal vein and the hepatic artery, together with afferent nerves, enter the liver at the hilus, where efferent bile ducts as well as lymphatics and nerves also exit the organ (see below). Branches of the hepatic artery, hepatic portal vein, main bile duct and main lymphatic vessel travel together in portal tracts through the liver parenchyma (Figure 1-5). Portal tracts are sometimes referred to as portal triads, because, of the five elements present, the lymphatic vessel is usually collapsed and inconspicuous, as are the autonomic nerves, resulting in only three elements being visible
Chapter 1 ANATOMY OF THE LIVER Figure 1-5. Hepatic microvasculature as determined by in vivo microscopic studies. PV, portal venule; HA, hepatic arteriole; L, lymphatic; BD, bile ductule; N, nerve; CV, central venule; SLV, sublobular hepatic vein. Arrows indicate direction of flow. (Modified from McCuskey RS. Functional morphology of the liver with emphasis on its microvasculature. In: Tavoloni N, Berk PD, eds. Hepatic transport and bile secretion. New York: Raven Press, 1993: 2, ©1993, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
CV
Bile canaliculus
Sinusoid Perisinusoidal space
Inlet sphincter
Intersinusoidal sinusoid and sphincter Outlet sphincter
Arterio-sinus twig
Bile canaliculus Hepatic cell Sinusoid
Arterio-portal anastomoses N
PV
L BD
HA SLV
in sections through portal tracts. After repeated branching, terminal branches of the blood vessels (portal venules and hepatic arterioles) supply blood to the sinusoids (Figure 1-6). Branches of hepatic arterioles also supply the peribiliary plexus of capillaries nourishing the bile ducts, and then drain into sinusoids (via arteriosinus twigs) (Figure 1-7) or occasionally into portal venules (arterioportal anastomoses). Because all these vessels are independently contractile, the sinusoids receive a varying mixture of portal venous and hepatic arterial blood.15,16 After flowing through the sinusoids, blood is collected in small branches of hepatic veins termed central venules (central veins, terminal hepatic venules) (Figure 1-6). These course independently of the portal tracts and drain via hepatic veins, which leave the liver on the dorsal surface and join the inferior vena cava. Lymphatic vessels originate as blind-ending capillaries in the connective tissue spaces within the portal tracts.17 The fluid contained in these lymphatics flows toward the hepatic hilus and eventually into the cisternae chili and thoracic duct. The perisinusoidal space of Disse is thought by some to function as a lymphatic space that channels plasma to the true lymphatics coursing in the portal tract. However, anatomic connections between the space of Disse and the portal tract have not been identified.17,18 Lymph also leaves the liver in small lymphatics associated with the larger hepatic veins into lymphatics along the wall of the inferior vena cava.1 Lymphatics in the
Figure 1-6. Vascular cast of the hepatic microvasculature illustrating the tortuous anastomotic sinusoids adjacent to the portal venule (PV) and the more parallel and larger sinusoids near the central venule (CV). (Modified from McCuskey RS. The hepatic microvascular system. In: Tavoloni N, Berk PD, eds. Hepatic transport and bile secretion. New York: Raven Press, 1993: 4, ©1993, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
7
Section I. Pathophysiology of the Liver
Figure 1-7. Terminal branches (arrowheads) from hepatic arteriole (HA) which frequently terminate in inlet venules or terminal portal venules where sinusoids originate. PV, portal venule; B, peribiliary plexus supplied by the adjacent hepatic arteriole. (Modified from McCuskey RS. Functional morphology of the liver with emphasis on its microvasculature. In: Arias IM, Boyer JL, et al., eds. The Liver: Biology and Pathobiology, 3rd edn. New York: Raven Press, 1994: 1095, ©1994, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
Figure 1-9. Intrahepatic aminergic innervation in the dog. Brightly fluorescent nerve fibers are adjacent to the portal vein (PV), hepatic artery (HA), and bile duct (not visible in this section) and are also distributed intralobularly along the sinusoids (arrows).
of the portal vein and hepatic artery in portal tracts. Bile ducts drain through larger left and right hepatic ducts, which exit the liver at the hilus to form the common bile duct. These ducts are lined with simple columnar epithelial cells. Branches of the hepatic artery supply an extensive peribiliary plexus of capillaries (Figure 1-7).22
Innervation
Figure 1-8. Bile canalicular network filled with dye injected retrograde into the bile duct.
hepatic capsule drain to vessels either at the hilum or around the hepatic veins and inferior vena cava.1
Biliary System Bile canaliculi are spaces 1–2 mm wide formed between adjacent hepatocytes (Figure 1-1A).19–21 They are interconnected and form a network of minute intercellular channels (Figure 1-8) which receive the bile secreted from hepatocytes. These minute biliary channels are specialized regions of adjacent hepatic parenchymal cells and will be discussed in more detail together with the ultrastructure of these cells. The bile canaliculi drain through short bile ductules (cholangioles) partially lined by cuboidal epithelial cells to bile ducts, lined with simple cuboidal epithelium, which course along with branches
8
Aminergic, peptidergic, and cholinergic nerves are contained in the portal tracts and affect both intrahepatic blood flow and hepatic metabolism.23,24 The role of neural elements in regulating blood flow through the hepatic sinusoids, solute exchange, and parenchymal function is incompletely understood. This is due in part to limited investigation in only a few species, whose hepatic innervation may differ significantly from that of humans. For example, most experimental studies have used rats and mice, whose livers have little or no intralobular innervation. In contrast, most other mammals, including humans, have aminergic and peptidergic nerves extending from the perivascular plexus in the portal space into the lobule (Figure 1-9), where they course in the space of Disse in close relationship to stellate cells and hepatic parenchymal cells (Figure 1-10). Although these fibers extend throughout the lobule, they predominate in the periportal region. Cholinergic innervation, however, appears to be restricted to structures in the portal space and immediately adjacent hepatic parenchymal cells. Neuropeptides have been colocalized with neurotransmitters in both adrenergic and cholinergic nerves. Neuropeptide Y (NPY) has been colocalized in aminergic nerves supplying all segments of the hepatic–portal venous and the hepatic arterial and biliary systems. Nerve fibers immunoreactive for substance P (SP) and somatostatin (SOM) follow a similar distribution. Intralobular distribution of all of these nerve fibers is species dependent and similar to that reported for aminergic fibers. Vasoactive intestinal peptide (VIP) and calcitonin gene-related peptide (CGRP) are reported to coexist in cholinergic and sensory afferent nerves innervating portal veins and hepatic
Chapter 1 ANATOMY OF THE LIVER
are mutually exclusive, have been proposed as follows and are illustrated in Figure 1-11. The classic hepatic lobule is a polygonal structure having as its central axis a central venule, with portal tracts distributed along its peripheral boundary.25 The peripheral boundaries of these lobules are poorly defined in most species, including man (Figure 1-12). In some species, e.g. pigs and seals, there is considerably more connective tissue in the liver and this is distributed along the peripheral boundary of classic lobules, thereby making them very distinct. Considerable sinusoidal anastomoses occur between adjacent lobules, and thus the blood collected by each central venule is supplied by several portal venules. The hepatic acinus26,27 is a unit having no distinct morphologic boundaries. Its axis is a portal tract and its peripheral boundary is circumscribed by an imaginary line connecting the neighboring terminal hepatic venules (central hepatic venules of the classic lobule), which collect blood from sinusoids. Contained within the acinus are three zones, each having different levels of oxygenation and metabolic function. In yet another model of lobular organization, the lobule is defined by bile drainage. So-called portal lobules28 have at their center a portal tract, with central veins present around the periphery of each lobule. Currently, the concept of subunits of the classic lobule forming functional units is the most consistent with existing evidence.29–32 In this model, each ‘classic’ lobule consists of several ‘primary lobules’. Each primary lobule is cone-shaped, having its convex surface at the periphery of the classic lobule supplied by terminal branches of portal venules and hepatic arterioles, and its apex at the center of the classic lobule drained by a central (terminal hepatic) venule.
arteries and their branches, but not the other vascular segments or the bile ducts. Nitrergic nerves immunoreactive for neuronal nitric oxide (nNOS) are located in the portal tract, where nNOS colocalizes with both NPY- and CGRP-containing fibers.
HEPATIC FUNCTIONAL UNITS The organization of each liver lobe into structural or functional units related to function and/or disease has been the subject of considerable debate during the past century. Several models, none of which
Figure 1-10. Nerve fiber (N) closely associated with a stellate cell (FSC) in the space of Disse in the dog. H, hepatic parenchymal cell; L, lipid droplet, C, collagen.
HA
Primary lobule
PV
Classic lobule A
B
C CV
HMS C B
Portal lobule
A
1
2
3
Acinus
Figure 1-11. Contiguous hepatic lobules illustrating the interconnecting network of sinusoids derived from two portal venules (PV). Note that the sinusoids become more parallel as they course toward the central venule (CV), which forms the axis of the classic lobule. Hepatic arterioles (HA) supply blood to sinusoids near the periphery of the lobule, usually by terminating in inlet venules or terminal portal venules. As a result, three zones (1, 2, 3) of differing oxygenation and metabolism have been postulated to compose a hepatic acinus, with its axis being the portal tract (lower left). Several acini would compose the portal lobule (lower right). Each classic lobule contains several cone-shaped subunits having convex surfaces fed by portal and arterial blood at the periphery and its apex at the central venule (upper left). A, B, and C represent hemodynamically equipotential lines in a ‘primary lobule.’ A recent modification further subdivides lobules into conical hepatic microcirculatory subunits (HMS), each being supplied by a single inlet venule. (Modified from McCuskey RS. Functional morphology of the liver with emphasis on its microvasculature. In: Tavoloni N, Berk PD, eds. Hepatic transport and bile secretion. New York: Raven Press, 1993: 4, ©1993, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
9
Section I. Pathophysiology of the Liver
Figure 1-12. The liver is composed of lobules each having a central venule (CV) as its axis and peripheral boundaries which are poorly defined (arrows) but contain branches of the portal vein (PV), hepatic artery and bile duct.
These ‘primary lobules’ were renamed as ‘hepatic microvascular subunits (HMS)’ and were demonstrated to consist of a group of sinusoids supplied by a single inlet venule and its associated termination of a branch of the hepatic arteriole from the adjacent portal space (Figure 1-12). Further confirmation of this HMS concept was obtained by studying their development in neonatal livers.33 Accompanying the HMS are hepatic parenchymal cells and the associated cholangioles and canaliculi. Hepatocellular metabolic gradients also have been demonstrated to conform to this proposed functional-unit concept.34,35
HEPATIC PARENCHYMAL CELLS Hepatic parenchymal cells, commonly referred to as hepatocytes, are polyhedral cells about 20–30 mm in size, have a volume of approximately 5000 mm3 , and are organized into anastomotic sheets (Figure 1-1).36–38 They are epithelial cells, and like other polarized epithelial cells they have distinct apical, lateral and basal surfaces. The basal surfaces of hepatocytes face the sinusoidal endothelium. Their plasma membranes have microvilli which extend into the space of Disse (the space between hepatocytes and endothelial cells) to increase surface area for the exchange of materials between hepatocytes and blood plasma. The apical surfaces of hepatocytes face adjacent hepatocytes and enclose the bile canaliculi, minute spaces forming a network of channels that carry the bile secretion (exocrine secretion) of hepatocytes (Figures 1-13 and 1-14). The apical surfaces also form microvilli to increase the surface area for secretion. This is also referred to as the canalicular domain of the plasma membrane. The lateral membranes of hepatocytes extend from the bile canaliculi to the space of Disse and form cell–cell junctions, including gap junctions which facilitate communication between hepatocytes, and tight junctions which seal the bile canalicular lumen from the interstitial space (Figures 1-13 and 1-14). These tight junctions are critical in that they prevent leakage of plasma into bile as well as backflow of bile from canaliculi into the blood. Functionally, the basal and lateral membranes are frequently considered a unit, the basolateral membrane.
10
Figure 1-13. Portions of three hepatic parenchymal cells having bile canaliculi (BC) located between adjacent cells. RER, rough endoplasmic reticulum; SER, smooth endoplasmic reticulum; M, mitochondria; L, lysosome; S, sinusoid; D, space of Disse; E, endothelial cell. Inset is a higher magnification of a tight junction (arrowhead) between adjacent parenchymal cells. DS, desmosome. (Reproduced from Jones AL. Anatomy of the normal liver. In: Zakim D, Boyer TD, eds. Hepatology: A Textbook of Liver Disease, 3rd edn. Philadelphia: WB Saunders, 1996: 22, ©1996, with permission of Elsevier.)
Hepatocytes have one, or frequently two, nuclei and their cytoplasm contains numerous mitochondria as well as a prominent Golgi apparatus located between the nucleus and the bile canaliculi, rough endoplasmic reticulum, and smooth endoplasmic reticulum, with associated rosettes of glycogen particles.38 They also contain numerous endosomes, lysosomes, and peroxisomes. Fat droplets also may be present.
PLASMA MEMBRANE The plasma membrane is a dynamic structure38–41 that has a variety of regions having specific functions and characteristics. The basal plasma membrane of each hepatocyte faces one or more sinusoids, where its surface area is greatly increased by microvilli that extend into the space of Disse (Figure 1-13) to facilitate the uptake of bloodborne substances into hepatocytes and the secretion of constitutively produced substances into the blood. This exchange of products across the plasma membrane in the space of Disse is further facilitated by the absence of a typical epithelial basal lamina; the sinusoidal endothelium also has a greatly reduced or absent basal
Chapter 1 ANATOMY OF THE LIVER
Figure 1-14. Two adjacent hepatic parenchymal cells and enclosed bile canaliculus (BC) and associated organelles. L, lysosome; G, Golgi; SER, smooth endoplasmic reticulum; Mb, microbody (peroxisome); M, mitochondria, g, glycogen; N, nucleus; arrowheads, tight junctions. (Reproduced from Jones AL. Anatomy of the normal liver. In: Zakim D, Boyer TD, eds. Hepatology: A Textbook of Liver Disease, 3rd edn. Philadelphia: WB Saunders, 1996: 22, ©1996, with permission of Elsevier.)
lamina. The apical surface of the plasma membrane is limited to the bile canaliculi, which are channels formed by tight junctions between adjacent hepatocytes (Figures 1-13 and 1-14). Microvilli extend into the bile canaliculi, expanding the surface area of the apical plasma membrane for secretion of bile. Communication between hepatocytes is provided by gap junctions, which are an assemblage of many connexons, membrane pores formed by the circular arrangement of six transmembrane proteins called connexins. Connexons in apposing plasma membranes are directly aligned with each other and form aqueous channels that allow the passage of ions and small molecules. Cellular metabolic products, as well as chemical and electrical signals, can pass from cell to cell. Hepatocytes express specific genes for their unique connexin proteins. Desmosomes, as well as ‘knob and groove’ or interdigitating undulations of adjacent plasma membranes, attach cells together in addition to the tight junctions forming bile canaliculi. The molecular structure of the hepatocyte plasma membrane includes specializations such as membrane proteins that are receptors for hormones, for example insulin and glucagon; and receptors that bind other substances, such as circulating immunoglobulin A (IgA), and also contribute the secretory component required for IgA function. Assorted carrier and channel protein membrane components regulate/facilitate the great variety of substances that enter and leave hepatocytes by ways other than receptors, endocytosis, and exocytosis. Hepatocyte uptake and the release of glucose affects the regulation of blood glucose levels, and also accounts for the variable intracellular glycogen deposits that have been characterized in a variety of physiological conditions.
NUCLEUS Hepatocytes have one or two spherical nuclei containing one or more prominent nucleoli (Figures 1-13 and 1-14).1,2,8,38 Some of the nuclei are polyploid and their number increases with age.42 Polyploid nuclei are characterized by their greater size, which is proportional
to their ploidy. Multinucleated hepatocytes and polyploidy are consistent with high cellular function and demands, and are mechanisms by which both nuclear and cytosomal ‘machinery’ are increased to meet these functional demands. The high level of hepatocellular activity is also reflected in the high percentage of nuclei that are euchromatic, which indicates that transcription of most of the genome is occurring continuously; thus, almost all of the deoxyribonucleic acid (DNA) is in the extended configuration, and little heterochromatin is observed. Hepatocytes engaged in the synthesis of many proteins have a large nucleolus (sometimes several) that can be recognized by light microscopy, and this characteristic is typical of hepatocytes. Electron microscopy reveals the nucleolus to consist of pale-staining areas of nucleolar organizer DNA, an electron-dense granular portion of ribonucleoprotein particles forming ribosomal subunits, and a fibrillar region of transcripts of rRNA. Heterochromatic nucleolar-associated chromatin is found at the nucleolus periphery. Nucleoli are the sites of the translation of rRNA into protein-rich ribosomal subunits that exit the nucleus through pores in the double membrane nuclear envelope.
ENDOPLASMIC RETICULUM, RIBOSOMES, AND GOLGI APPARATUS Rough endoplasmic reticulum (RER), smooth endoplasmic reticulum (SER), and Golgi complex are abundant in mammalian hepatocytes (Figures 1-13 and 1-14).1,2,38,43,44 Their functions are related mainly to the synthesis and conjugation of proteins, metabolism of lipids and steroids, detoxification and metabolism of drugs, and breakdown of glycogen. The endoplasmic reticulum forms a continuous three-dimensional network of tubules, vesicles, and lamellae.43,45 About 60% of the endoplasmic reticulum has ribosomes attached to its cytoplasmic surface and is known as the RER. The remaining 40% constitutes the SER, which lacks a coating of ribosomes. The membranes of the endoplasmic reticulum are 5–8 nm thick. The lumen of the RER is about 20–30 nm in width, whereas that of the SER is larger (30–60 nm). The morphological characteristics and amount of the endoplasmic reticulum may vary in the different zones of the liver lobule. RER is arranged in aggregates of flat cisternae that may be found in the whole cytoplasm. It is more frequently distributed in the perinuclear, pericanalicular, and vascular regions of hepatocytes, and it is more abundant in periportal cells than in centrilobular cells.46,47 The numerous attached membrane-bound ribosomes consist of a large and a small subunit; the former is the part found attached to the RER. Free ribosomes and polyribosomes are also present within the hepatocyte cytoplasm. Ribosomes contain RNA and ribosomal proteins and play a key role in the synthesis of proteins. SER is less common and has a more complex arrangement than RER.38 It is usually much more abundant in centrilobular than in periportal hepatocytes.46–48 The cytoplasm within the SER tubules is usually slightly more electron dense than the surrounding cytoplasm. SER membranes are irregular in size and present a tortuous course. They may be tubular or vesicular in type, with a width of 20–40 nm. SER is mainly distributed near the periphery of the cell. It is often in close relation to RER and Golgi membranes, as well as to glycogen inclusions.49 The ER is not the only site for the protein synthesis in hepatocytes. Abundant free ribosomes in the cytoplasm participate in the
11
Section I. Pathophysiology of the Liver
synthesis of some proteins that will be secreted, but especially of all structural proteins for the hepatocyte. Messages encoding proteins that are to remain within the cytoplasm or are destined to enter the nucleus, peroxisomes, or mitochondria are completely synthesized by free ribosomes. The Golgi complex is a three-dimensional structure in hepatocytes consisting of numerous membranes and vacuoles.8,38,43,44 Multiple Golgi complexes exist in each hepatic parenchymal cell. Whether or not these complexes are connected with each other (functionally forming a single large organelle) is uncertain. The Golgi generally is distributed near the bile canaliculus or nucleus. The Golgi apparatus presents a characteristic heterogeneity. It is usually formed by a stack of four to six parallel cisternae, often with dilated bulbous ends containing electron-dense material. The cisternae may show a size up to 1 mm in diameter with a lumen 30 nm wide. This structure shows a convex or proximal part facing the nucleus and the endoplasmic reticulum (cis Golgi) where small vesicles transfer proteins from the endoplasmic reticulum, and a concave part (trans Golgi) where vesicles and vacuoles (secretory granules) originate to transport the contained secretory proteins to the plasma membrane for discharge into the space of Disse. Cis and trans Golgi are connected by means of the medial Golgi. The latter is the intermediate station between endoplasmic reticulum and Golgi products, such as secretory granules or secondary lysosomes (GERL). This arrangement of Golgi stacks corresponds to its morphofunctional polarization related to the pathway of protein passage through this structure. Proteins in fact enter via the cis Golgi, pass through the medial Golgi and leave this structure via the exit pole (trans Golgi). Two main types of secretory vesicle can be considered within the Golgi apparatus: smaller presecretory granules of 50 nm in diameter and larger secretory granules of 400–600 nm in diameter containing proteins such as very low-density lipoproteins.50
MITOCHONDRIA Mitochondria are large organelles and are very numerous in hepatocytes (1–2000/cell) (Figures 1-13 and 1-14), making up about 18–20% of the cell volume.51 They play a role in the oxidative phosphorylation and oxidation of fatty acids and in all metabolic processes of the hepatocyte.38 Although the mitochondria are dispersed ubiquitously within hepatocytes, they are more concentrated near sites of ATP utilization52 and are often associated with the RER.53 Such a relationship seems to be important during the formation of cytoplasmic membranes (SER) and cytochromes.53 Mitochondria in hepatocytes may be round or elongated, have a width of 0.4–0.6 mm and a length of 0.7–1.0 mm. Longer (up to 4 mm) and larger (up to 1.5 mm in diameter) mitochondria are more numerous in periportal hepatocytes.46,51 Mitochondria are bounded by an outer and an inner membrane, each 5–7 nm thick. The outer membrane possesses special small pores, which allow the passage of molecules smaller than about 2000 daltons (Da). The inner membrane’s surface area is greatly increased by the presence of numerous cristae, which fold within the mitochondrial matrix. The space between inner and outer membranes presents a low-density matrix and ranges from about 7 to 10 nm in thickness. Mitochondria show a relatively low-density matrix in which lamellar or tubular cristae and a variable amount of small dense granules can be observed. The dense granules have a diameter of 20–50 nm. In addition, filaments
12
of the circular mitochondrial DNA about 3–5 nm in width and granules containing mitochondrial RNA of about 12 nm in diameter are also present. The DNA codes for some of the mitochondrial proteins that are synthesized in ribosomes within the organelle, but most of the mitochondrial protein is encoded by nuclear DNA. Mitochondria are self-replicating and have a half-life of approximately 10 days.
LYSOSOMES Lysosomes in hepatocytes (Figures 1-13 and 1-14) consist of a heterogeneous population of organelles containing hydrolytic enzymes that are morphologically and functionally interrelated.38,54,55 These organelles form rounded single-membrane-bound dense bodies, autophagic vacuoles, multivesicular bodies, coated vesicles, and the GERL. The latter is like a cytoplasmic pool of structures located proximal to the Golgi apparatus (but is not part of it), consisting of smooth-surfaced membranes (like a specialized area of smooth ER) with the same hydrolase activity of the lysosome (but without the typical morphology of spherical organelles) that probably have a major role in the formation of lysosomes and hepatocyte lipoprotein metabolism. Several classes of lysosome can be identified within the hepatocyte cytoplasm: (1) the primary lysosomes, small in size, which are considered from a functional point of view to be in a resting phase; (2) the secondary lysosomes, which are functionally activated; (3) the autophagic vacuoles, containing parts of the degrading cytoplasmic organelles, and which often are delimited by a double membrane; and, finally (4) the residual bodies, which are larger than primary and secondary lysosomes and are usually more frequent in older organisms. The residual bodies contain the residues of non-digested material or pigments such as lipofuscins (which are considered undigestible permanent residues). Lipofuscin granules are the most numerous lysosomal bodies present in human hepatocytes.47 Lysosomes are frequently found near the plasma membrane proximal to the bile canaliculus, forming the so-called ‘peribiliary dense bodies’ of early histological descriptions. The lysosomes in periportal hepatocytes are often larger and more positive for acid phosphatase than those in centrilobular hepatocytes.47,48
PEROXISOMES (MICROBODIES) Peroxisomes are subcellular organelles that are usually rounded or slightly oval in shape, surrounded by a single membrane (Figure 1-14), and participate mainly in oxidative processes.38,56,57 Each hepatocyte may contain 300–600 peroxisomes. These organelles are characteristically more numerous and larger in hepatocytes than in other mammalian cells.47 They contain a fine granular matrix in which in some species (but not in humans) a denser paracrystalline structure may be present. The peroxisome size ranges between 0.2 and 1.0 mm. They are often found grouped in clusters near the endoplasmic reticulum. However, the presence of direct connections (the so-called ‘tails’) with endoplasmic reticulum or other peroxisomes (peroxisomal reticulum) is still under investigation. Peroxisomes may be more numerous in pericentral hepatocytes, but they are generally homogeneously distributed within the hepatic lobule.47,48 The origin and formation of peroxisomes is still under debate. Nevertheless, they have been said to originate as a focal protrusion of the RER.
Chapter 1 ANATOMY OF THE LIVER
CYTOPLASMIC INCLUSIONS The hepatocyte is extremely rich in cytoplasmic inclusions. These are functionally related to the enhanced metabolic activity of the liver cells. The more frequently observed cytoplasmic inclusions are glycogen granules, lipid droplets, and pigments of various natures.38 Glycogen granules are the most abundant inclusions in normal hepatocytes (Figures 1-13 and 1-14).38,47 At the electron microscopy level they are stained by lead salts, and may occur either in the monoparticulate form (b particles, 15–30 nm in size) or, more frequently, as aggregates of smaller particles arranged to form ‘rosettes’ (a particles). Glycogen granules are dispersed in the cytoplasm, but are often associated with the SER. Glycogen is depleted during fasting, disappearing first from periportal hepatocytes and then from centrilobular cells. Upon refeeding, the sequence reverses. Lipid inclusions appear as empty vacuoles or osmiophilic droplets usually not surrounded by membranes. Fat droplets may vary in size and quantity, and correspond mainly to triglyceride levels in the hepatocyte.48 A variable amount of iron-containing granules may often be present within the hepatocyte cytoplasm. These are related to the apoferritin–ferritin system (the so-called ‘hepatic iron buffer’). Liver iron metabolism occurs in hepatocytes; nevertheless, the pathway of iron transport from the blood to the hepatocytes has not yet been fully elucidated. In addition to hepatocytes, liver endothelial cells and Kupffer cells58,59 also possess receptors for transferrin, a glycoprotein implicated in cellular iron uptake, thus suggesting that iron transport involves a transendothelial (transcytosis) mechanism. Hepatocytes contain iron in the form of ferritin particles, i.e. an iron-containing protein ultrastructurally characterized by a roughly spherical shape and comprising a protein shell (apoferritin) 11 nm in diameter and a central core of about 5 nm in diameter containing iron. Hepatocyte iron deposits may also occur as single-membrane-bound lysosomal bodies (residual bodies) forming aggregates of iron-containing electron-dense particles (siderosomes–hemosiderin granules).
CYTOSKELETON AND CYTOMATRIX The cytoskeleton is a structure that is thought to regulate the shape, subcellular organization and movements of the cells. In the hepatocyte, the cytoskeletal organization50,60,61 is dependent on the arrangement of the three main components of this structure: the microfilaments, the intermediate filaments, and the microtubules. These filament types are regularly distributed in the cytoplasm and characterize the cytomatrix, which together with other finer filaments (microtrabeculae) is thought to give the ‘gel’ consistency to the cell cytoplasm. Microfilaments, made of actin, and microtubules, consisting of tubulin, are both related to intracellular motility. Microtubules are thought to be involved in determining cell shape, in mitosis, and in regulating the intracellular transport of vesicles.62 Especially in the liver, these structures assume a relevant role in the secretion of lipoproteins, albumin, and the release of lipids into bile. Microfilaments are more directly related to bile secretion. In fact, they are normally found around the bile canaliculi (pericanalicular web). Many experimental studies have shown that microfilaments play an active role in the dilatation and contraction of bile canaliculi.63–65 Thus, they may control the bile canalicular
caliber and bile flow. Intermediate filaments show a more complex architecture. They correspond to the epithelial cell ‘tonofilaments’ of the old nomenclature. In the liver they show a relationship with the Mallory bodies (the structural marker of human alcoholic liver disease).66 They are located around the nucleus, near the cell border, in the cytoplasmic network and around the bile canaliculi. There is very little information on the presence of microtubules or microfilaments in differentiating hepatocytes. In mice, these structures have been recognized as dense bundles occurring near the nucleus and the plasma membrane in late developmental stages.67 Their presence could have some importance in bile canaliculus and desmosome differentiation.
NON-PARENCHYMAL CELLS The hepatic sinusoid is an unique, dynamic microvascular structure which serves as the principal site of exchange between the blood and the perisinusoidal space (of Disse), into which project the microvilli of the hepatic parenchymal cells that form the external lining of this space.15 The sinusoid is composed of non-parenchymal cells, of which there are four recognized types (Figures 1-15 and 1-16).15,68 These are fenestrated endothelial cells and phagocytic Kupffer cells, which form the sinusoid lining that is in contact with the blood; extraluminal fat-storing cells (of Ito), also referred to as stellate cells, lipocytes, or perisinusoidal cells, which serve as specialized pericytes extending processes throughout the space of Disse; and pit cells, which are immunoreactive natural killer (NK) cells that are attached to the luminal surface of the sinusoid and are part of a population of liver-associated lymphocytes (LAL).69 Additional cells and cell processes may be present in the perisinusoidal space (of Disse) of some species, most notably mast cells in the dog70 and adrenergic and peptinergic nerves in most mammalian species except mouse and rat.23 The perisinusoidal space is thought by some to function as a lymphatic space that channels plasma to the true lymphatics coursing in the portal tract. Although this hypothesis would help to explain the large efflux of lymph from the liver, it may not be valid as anatomic connections between the space of Disse and the portal tract have not been identified.17,18 For a review of intrahepatic lymphatics see Trutmann and Sasse.17 The majority of the non-parenchymal cells have been studied both in situ and in vitro. Together, sinusoidal cells represent about 6% of the total liver volume, but account for 30–35% of the total number of liver cells.71,72 The purpose of this chapter is to present an overview of the structural and functional features of these sinusoidal cells, which together provide a physical and selective barrier between the blood and the parenchyma that is dynamic and responsive to a wide variety of physical and chemical stimuli. Whereas sinusoidal lining cells have the capacity to divide and proliferate, especially when stimulated by immune system modifiers,73 sinusoidal macrophages and NK cells may also be increased in numbers by the respective recruitment and subsequent modification of monocytes and lymphocytes, principally of bone marrow origin.74
SINUSOIDAL ENDOTHELIAL CELLS Like endothelial cells in capillaries elsewhere in the body, contiguous sinusoidal endothelial cells in the liver form the basic tubular
13
Section I. Pathophysiology of the Liver
KC
E
SP
SD
SC
BC
HC
Figure 1-15. Sinusoid wall and contiguous hepatic parenchymal cells (HC). E, endothelium; KC, Kupffer cell; SD, space of Disse; SC, stellate cell; SP, sieve plate of fenestrae; BC, bile canaliculus. (Modified from McCuskey RS. In: Tavoloni N, Berk PD, eds. Hepatic transport and bile secretion. New York: Raven Press, 1993: 6, ©1993, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
14
Figure 1-17. Sinusoidal endothelial cell with limited perinuclear cytoplasm, containing a few organelles, such as mitochondria, a lysosome, and a few cisternae of endoplasmic reticulum. The endothelial cell rests on the microvilli filling the space of Disse. L, sinusoidal lumen; N, nucleus. (Modified from Wisse E, Braet F, et al. Structure and function of sinusoidal lining cells in the liver. Toxicol Pathol 1996;24:100–111, with permission.)
Figure 1-16. Sinusoid (S) lined by endothelial cells (SEC) having attenuated cytoplasm with Kupffer cell (KC) attached to the luminal surface and a stellate cell (SC) lying externally in the space of Disse. (Modified from McCuskey RS. Functional morphology of the liver with emphasis on its microvasculature. In: Tavoloni N, Berk PD, eds. Hepatic transport and bile secretion. New York: Raven Press, 1993: 6, ©1993, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
Figure 1-18. Sinusoid illustrating fenestrae organized in clusters as ‘sieve’ plates (arrowheads). SD, space of Disse; H, hepatic parenchymal cell. (Reproduced from McCuskey RS. Functional morphology of the liver with emphasis on its microvasculature. In: Tavoloni N, Berk PD, eds. Hepatic transport and bile secretion. New York: Raven Press, 1993: 7, ©1993, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
vessel for transvascular exchange between the blood and the surrounding tissue (Figure 1-15) and represent approximately 50% of the numbers and volume of sinusoidal cells.71,72 The morphology of hepatic sinusoidal endothelial cells has been reviewed by several authors.68,75 These cells are unique to the liver in that their extensive, attenuated cytoplasm contains numerous fenestrae, approximately 170 nm in diameter, which lack diaphragms and which are
clustered together in groups known as ‘sieve plates’76 (Figures 1-16–1-18). In addition, this specialized endothelium generally lacks a basal lamina during health, so that solutes and small particles have direct access to the perisinusoidal space containing processes of fat-storing cells and the microvilli of hepatic parenchymal cells. The endothelium of the sinusoids exhibits heterogeneity. The fenestrae are not uniform in size or distribution throughout the length
Chapter 1 ANATOMY OF THE LIVER
of the sinusoid, from its origin at the portal venule to its termination in the central venule. At the periportal end of the sinusoid the fenestrae are somewhat larger than those located centrilobularly, but their numbers are fewer which, when combined with the sinusoid having a smaller diameter at the periportal rather than the centrilobular end, results in a higher centrilobular endothelial porosity.76–78 The functional significance of these regional differences is unclear, but it is tempting to relate them to the functional metabolic heterogeneity that has been demonstrated for hepatocytes in different regions of the lobule,35,79–81 as well as the portal-to-central intralobular oxygen gradient.82 The fenestrae constitute only 6–8% of the surface area of the endothelial lining. They form a selective barrier between the blood and parenchyma that acts as a dynamic, selective sieve for particulates such as chylomicron remnants.76,83 Transport of particulates somewhat larger than the size of the fenestrae is postulated to be accomplished by the ‘forced sieving’ and ‘endothelial massage’ concomitant with the passage of blood cells, particularly leukocytes, through the sinusoids and the resulting interaction of these cells with the endothelial wall.76 The endothelial fenestrae are dynamic structures whose diameters are affected by luminal blood pressure, vasoactive substances, drugs, and toxins.76,83–85 The mechanism for active control of the diameters of these fenestrae appears to reside in actin-containing components of the cytoskeleton.86–89 Additional cytoskeletal components form rings that delineate both the fenestrae and the sieve plates.87,90 As a result, the fenestrae are thought to regulate the passage of large substances such as chylomicron remnants through the endothelium while allowing free exchange of plasma and large proteins between the blood and the space of Disse. Thus, the sinusoidal endothelial filter influences the fat balance between the liver and other organs, the cholesterol level in the plasma, and the delivery of retinoids to parenchymal and fat-storing cells. There is a reduction of the numbers of fenestrae with age.91–93 The surfaces of the sinusoidal endothelial cells are relatively smooth compared to that of Kupffer cells and are generally lacking in filopodia or lamellopodia (Figures 1-16–1-18). The perikaryon contains mitochondria, some scattered components of both smooth and rough endoplasmic reticulum, and a well developed Golgi apparatus. Throughout the cytoplasm are located numerous vacuoles and organelles associated with the uptake, transport and degradation of material. These include bristle-coated pits, which are invaginations from the cell membrane, bristle-coated micropinocytotic vesicles, endosomes, transfer tubules, and lysosomes.68,94 The fact that these endothelial cells contain 45% by volume of the pinocytotic vesicles in the liver as well as 14% of the lysosomes71,72 indicates the high degree of endocytotic activity present in these cells. The variety of substances known to be endocytosed by sinusoidal endothelial cells includes proteins, glycoproteins, lipoproteins, glycosaminoglycans95,96 and, under certain conditions, larger particulates which are phagocytosed in the absence of functional Kupffer cells.97 A number of receptors to accomplish this have been identified on the cell surface, including Fc receptors for immune complexes, transferrin (Tf) receptors, scavenger receptors, mannose, galactose, apo E and C-III receptors. Of these, the scavenger and apo-E receptors are particularly abundant on endothelial cells compared to Kupffer cells, as are mannose/N-acetyl glucosamine recep-
tors. The former indicate the important role played by the sinusoidal endothelial cells in the processing and metabolism of lipoproteins. Recently, they have been demonstrated to play a significant role in the removal of AGE molecules.93 The endothelial cells also are secretory and release interleukin (IL)-1, IL-6, and interferon.68,95 In addition, these cells produce eicosanoids, particularly PGI2 , PGE2 , and TXA2 , as well as endothelin and nitric oxide.68 Thus, along with Kupffer cells, the endothelium participates in host defense mechanisms and regulation of sinusoidal blood flow in the liver. Finally, sinusoidal endothelial cells constitutively express the intercellular adhesion molecule ICAM-1, which along with VCAM-1 is up-regulated by inflammatory stimuli either directly or by mediators released from stimulated Kupffer cells, resulting in increased adhesion of leukocytes to the endothelial surfaces.98
KUPFFER CELLS Kupffer cells constitute the largest population of fixed macrophages in most vertebrates. They are components of the walls of hepatic sinusoids and play a significant role in the removal by endocytosis of particulates and cells from the portal blood, as well as toxic, infective and foreign substances, particularly those of intestinal origin.96 Kupffer cells also are the source of a variety of beneficial, vasoactive, and toxic mediators which are thought to be involved in host defense mechanisms, as well as some disease processes in the liver.96,99 Included among the substances released are eicosanoids, free radicals, cytokines, interferon, platelet-activating factor, and lysosomal enzymes. The morphology of mammalian Kupffer cells, including those in humans, has been described and extensively reviewed.96,100 Kupffer cells are macrophages that constitute one of the cellular components of hepatic sinusoids (Figures 1-15, 1-16, 1-19, 1-20). In this site they are anchored to the luminal surface of the sinusoidal endothelium and thus are exposed to the bloodstream. Occasionally, Kupffer cells also are interdigitated between endothelial cells. However, Kupffer cells are unevenly distributed within hepatic lobules, with
Figure 1-19. Kupffer cell (KC) attached to luminal surface of sinusoidal endothelium by processes that penetrate fenestrae. (Modified from McCuskey RS. Functional morphology of the liver with emphasis on its microvasculature. In: Tavoloni N, Berk PD, eds. Hepatic transport and bile secretion. New York: Raven Press, 1993: 7, ©1993, with permission from Lippincott, Williams and Wilkins (http://lww.com).)
15
Section I. Pathophysiology of the Liver
Figure 1-20. Kupffer cell, having lysosomes with varying density and diameter, vacuoles, and a nucleus (N). Kupffer cells are sometimes seen in direct contact with the microvilli of the parenchymal cells (arrowhead). L, sinusoidal lumen; f, fenestrae; SD, space of Disse. (Modified from Wisse E, Braet F, et al. Structure and function of sinusoidal lining cells in the liver. Toxicol Pathol 1996;24:100–111, with permission.)
the majority being found in the periportal region, where they are larger and have greater phagocytic activity than Kupffer cells located in the centrilobular region of the lobule.73,101–103 In addition, Kupffer cells are often located at the junctions of sinusoids. As a result, the majority of Kupffer cells are strategically located to remove foreign materials as they enter the liver lobule. Kupffer cells often present a large irregular surface, caused by numerous microvilli, filopodia and lamellopodia extending from the cellular surface (Figures 1-16 and 1-19).100 Attachment to the endothelium appears to be by cytoplasmic processes which often penetrate the endothelial fenestrae to enter the space of Disse, where they may come in contact with fat-storing cells and, occasionally, parenchymal cells. Other processes frequently extend across the lumen to anchor in the opposite wall of the sinusoid. As a result, Kupffer cells often have a branched or ‘stellate’ appearance. Whereas Kupffer cells frequently contact other sinusoidal cell types, no organized junctions have been visualized between Kupffer cells and these contiguous cells. The surface of Kupffer cells is covered with a fuzzy coat of unknown composition which normally is not preserved by perfusion fixation with glutaraldehyde.104,105 It can, however, be seen coating the inner surface of the membranes of large pinocytotic vacuoles, and as a dense midline within membranous invaginations known as ‘worm-like’ bodies or vermiform processes.106 These structures are thought to be unique to Kupffer cells, as are annulate lamellae.106 The latter are sometimes found connected to the rough endoplasmic reticulum (RER) and are thought to represent a particular arrangement of the RER. These latter two structures, along with the nuclear membrane, stain positive for endogenous peroxidase. Although this is a specific marker for Kupffer cells in the rat liver105,107,108 it is not as useful in other species, because of a similar positivity in large numbers of endothe-
16
lial cells. More recently, monoclonal antibodies also have been used to identify macrophages and Kupffer cells.109–111 In addition to the above structures, the cytoplasm of Kupffer cells contains bristle-coated micropinocytotic vesicles and a number of clear vacuoles and dense bodies (lysosomes) which, along with the vermiform processes and fuzzy-coated vacuoles, are involved in the high level of endocytotic and digestive activity attributed to these cells.96,106 Additionally, the usual set of cellular organelles is also present in the cytoplasm, including mitochondria, RER, free ribosomes, Golgi apparatuses, microtubules, microfilaments, intermediate filaments, centrioles and a nucleolus.96,100,106 However, fat droplets, autophagic vacuoles, multivesicular bodies, peroxisomes, and smooth endoplasmic reticulum have not been reported in Kupffer cells in situ. The endocytotic mechanisms of Kupffer cells have been studied both in situ and, in greater detail, in isolated cultured cells. Four morphologically recognizable endocytotic mechanisms for Kupffer cells fixed in situ by perfusion have been described: bristle-coated micropinocytosis; pinocytosis veriformis; pinocytosis (fuzzy-coated vacuole); and phagocytosis.96,106 Of these, the principal endocytotic mechanisms, both in vivo and in vitro, are thought to be phagocytosis and bristle-coated micropinocytosis. Phagocytosis of particulates larger than 0.3–0.5 mm (e.g. latex, bacteria, etc.) is performed by hyaloplasmic pseudopodia, which extend from the cell surface to engulf the particulate. Phagocytosis of particulates >0.5 mm, e.g. latex, has been used as a marker to distinguish Kupffer cells from other sinusoidal lining cells under normal conditions.108 However, as noted previously the sinusoidal endothelium is also capable of phagocytosing latex particles if Kupffer cells are injured.97 Bristlecoated micropinocytosis is thought to be responsible for both receptor-mediated and non-receptor mediated fluid-phase endocytosis. Several receptors have been demonstrated on Kupffer cells, including Fc and C3 receptors, N-acetyl-D-galactosamine receptors, and N-acetyl-glucosamine/mannose receptors. The origin and cell kinetics of Kupffer cells continues to be debated between those who are proponents of a monocytic origin and those favoring self-replication.73,112–114 Taken together, the data seem to support both points of view. Kupffer cells during health have long residence times and slow rates of self-replication, augmented by some recruitment and transformation of monocytes. Monocyte recruitment becomes more important during stimulation of Kupffer cell function (e.g. zymozan, BCG, etc.).115–118
STELLATE CELLS External to the endothelium, perisinusoidal cells known as stellate cells (previously known as fat-storing cells, Ito cells or lipocytes) are located in the space of Disse (Figures 1-15, 1-16, 1-21), with a higher frequency in the periportal area.119–121 These cells contain fat droplets and are the major storage site of retinoids, including vitamin A, which emits a characteristic, rapidly quenched autofluorescence when excited with ultraviolet light at 328 nm. Two types of fat droplet are recognized, one with and one without a limiting membrane.120 The nuclear area of the stellate cell is frequently located in recesses between hepatic parenchymal cells, whereas the thin, multiple cytoplasmic processes of these cells course though the perisinusoidal space and extensively embrace the abluminal surfaces of
Chapter 1 ANATOMY OF THE LIVER
Figure 1-21. Stellate cell lying within the space of Disse, covered by the endothelial lining. Fat droplets (*) and cisternae of the endoplasmic reticulum are located in the cytoplasm. A small bundle of collagen fibers (arrow) is associated with the cell. L, sinusoidal lumen; f, fenestrae; N, nucleus; SD, space of Disse. (Modified from Wisse E, Braet F, et al. Structure and function of sinusoidal lining cells in the liver. Toxicol Pathol 1996;24:100–111, with permission.)
the endothelium that surrounds the sinusoid like a cylindrical basket.122 This close relationship of the processes of the stellate cell to the sinusoid wall, the presence of large numbers of cytoplasmic microtubules and microfilaments, the positive immunostaining of desmin and a-smooth muscle actin, and the close association of nerve fibers (Figure 1-10), coupled with the demonstration of contractile activity in these cells both in vivo and in vitro, strongly suggests that stellate cells play a role in the local regulation of blood flow through the hepatic sinusoids.123–126 In health, little or no basal lamina and collagen is associated with the sinusoidal endothelium. As a result, the sinusoid wall is a highly permeable structure that permits continuity of plasma between the blood and the hepatocyte. However, during certain types of liver injury, e.g. cirrhosis, basement membrane material and collagen fibrils accumulate in the perisinusoidal space, resulting in ‘capillarization’ of the sinusoid and impaired transvascular exchange.127 The perisinusoidal stellate cells are thought to be responsible for the synthesis of this material, following their transformation into myofibroblast-like cells having reduced numbers of fat droplets and vitamin A, as well as an increased capacity to secrete extracellular matrix materials, including collagen types I and III–VI, fibronectin, laminin, tenascin, undulin, hyaluronic acid, biglycan, decorin, syndecan-containing chondroitin sulfate, heparan, and dermatan sulfate.128
LIVER-ASSOCIATED LYMPHOCYTES 129
Pit cells are derived from circulating large granular lymphocytes (LGL)130 that become attached to the sinusoidal wall (Figure 1-22) and which possess natural killer (NK) activity and are part of a population of liver-associated lymphocytes (LAL).69,131,132 Pit cells contain azurophilic granules which stain for acid phosphatase, sug-
Figure 1-22. Pit cell with typical dense granules. This pit cell is in close contact with the endothelial lining and is seen to contact microvilli of the parenchymal cells (arrowhead). Ec, endothelial cell; f, fenestrae; L, sinusoidal lumen; N, nucleus; SD, space of Disse. (Modified from Wisse E, Braet F, et al. Structure and function of sinusoidal lining cells in the liver. Toxicol Pathol 1996;24:100–111, with permission.)
gesting that they are lysosomal in nature.133,134 In addition, the cytoplasm of these cells contains characteristic rod-cored vesicles as well as multivesicular bodies, a Golgi apparatus, and mitochondria, all of which exhibit polarity toward one side of an eccentric, indented nucleus. Although the majority of attachments to the sinusoidal wall are to endothelial cells, adhesion to Kupffer cells is not uncommon. Pit cells have been shown to spontaneously kill tumor cells as well as produce a cytolytic factor which is up-regulated by biological response modifiers such as zymosan, as well as by interleukin-2.131 These substances also induce proliferation of pit cells, as does partial hepatectomy, perhaps through the activation of Kupffer cells. Finally, two types of pit cell have been recognized: high density (HD) and low density (LD). The LD pit cells have a larger number of granules, which are also smaller than those in HD cells; in addition, LD cells exhibit more cytotoxicity.135
HETEROGENEITY Within the hepatic lobules, the parenchyma exhibits considerable heterogeneity along the portal venous–central venous axis, both ultrastructurally and in various enzyme activities. This results in an intralobular metabolic zonation, with different cellular functions represented in different zones within each lobule.79,80 For example, the key enzymes involved in glucose uptake and release and in the
17
Section I. Pathophysiology of the Liver
formation of urea and glutamine are reciprocally located with glucogenic and urea cycle enzymes, principally in the periportal zone, and glycolytic and glutaminogenic enzymes in the centrilobular zone. Mixed-function oxidation and glucuronidation are mainly centrilobular functions, whereas sulfation is principally a periportal function. This zonation of enzymatic functions also is reflected ultrastructurally in differences in mitochondria and smooth endoplasmic reticulum between different zones. As a result of this zonation, as well as the portal–central oxygen gradient, most toxicologic and pathologic events in the liver show a considerable degree of zonal preference. An example of toxicants eliciting periportal injury is allyl alcohol; carbon tetrachloride and acetaminophen elicit centrilobular injury. The sinusoids are composed of specialized non-parenchymal cells and also exhibit structural and functional heterogeneity.15,16 Near their origins from portal venules and hepatic arterioles, sinusoids are slightly narrower as well as being tortuous and anastomotic, forming interconnecting polygonal networks; farther away from the portal venules the sinusoids become organized as parallel vessels that terminate in central venules (terminal hepatic venules). Short intersinusoidal sinusoids connect adjacent parallel sinusoids. The volume of liver occupied by sinusoids in the periportal area is also greater than that surrounding central venules. However, because of the smaller size and the anastomotic nature of the periportal sinusoids, the surface available for exchange in this area (surface/volume ratio) is greater than in centrilobular sinusoids. The size and pattern of distribution of endothelial fenestrae differs along the length of the sinusoid. At the portal end, the fenestrae are larger but comprise less of the endothelial surface area than they do in the pericentral region. The functional significance of these regional differences is unclear, but relates to the functional metabolic heterogeneity that has been demonstrated for hepatocytes in different regions of the lobule. This, in turn, may depend on the recognized portal–central intralobular oxygen gradient.
REFERENCES 1. Saxena R, Zucker SD, Crawford JM. Anatomy and physiology of the liver. In: Zakim D, Boyer TD, eds. Hepatology: a textbook of liver disease. Philadelphia: WB Saunders, 2003: 3–30. 2. Wanless IR. Physioanatomic considerations. In: Schiff ER, Sorrell MF, Maddrey WC, eds. Schiff ’s diseases of the liver, 8th edn. Philadelphia: Lippincott-Raven, 1999: 3–37. 3. Boyd E. Normal variability in weight of the adult human liver and spleen. Arch Pathol 1933;16:350. 4. Hjortsjo C-H. The topography of the intrahepatic duct systems. Acta Anat 1951;11:599–615. 5. Bismuth H. Surgical anatomy and anatomical surgery of the liver. World J Surg 1982;6:3–9. 6. Blumgart LH, Hann LE. Surgical and radiologic anatomy of the liver and biliary tract. In: Blumgart LH, Fong Y, eds. Surgery of the liver and biliary tract. Vol. 1. London: Harcourt, 2000: 1–34. 7. Zaret KS. Embryonic development of the liver. In: Boyer JL, Fausto N, Chisari FV, Schachter D, eds. The liver: biology and pathobiology, 4th edn. Philadelphia: Lippincott Williams & Wilkins, 2001: Chapter 2. 8. MacSween RNM, Scothorne RJ. Developmental anatomy and normal structure. In: MacSween RNM, Burt AD, Portmann BC, et al., eds. Pathology of the liver, 4th edn. London: Churchill Livingstone, 2002: 1–66.
18
9. Wilson JW, Groat CS, Leduc EH. Histogenesis of the liver. Ann NY Acad Sci 1963;111:8–24. 10. Marks C. Developmental basis of the portal venous system. Am J Surg 1969;117:671–681. 11. Morgan JD, Hartroft WS. Juvenile liver. Arch Pathol 1961;71:86–88. 12. Haratake J, Hisaoka M, Furuta A, et al. A scanning electron microscopic study of postnatal development of rat peribiliary plexus. Hepatology 1991;14:1196–1200. 13. Nakanuma Y, Hoso M, Sanzen T, Sasaki M. Microstructure and development of the normal and pathologic biliary tract in humans, including blood supply. Microsc Res Tech 1997;38:552–570. 14. Vijayan V, Tan CE. The developing human biliary system in three dimensions. Anat Rec 1997;249:389–398. 15. McCuskey RS. The hepatic microvascular system. In: Arias IM, Boyer JL, Fausto N, et al., eds. The liver: biology and pathobiology, 3rd edn. New York: Raven Press, 1994: 1089–1106. 16. McCuskey RS. Morphologic mechanisms for regulating blood flow through hepatic sinusoids. Liver 2000;20:3–7. 17. Trutmann M, Sasse D. The lymphatics of the liver. Anat Embryol 1994;190:201–209. 18. Niiro GK, O’Morchoe CCC. Pattern and distribution of intrahepatic lymph vessels in the rat. Anat Rec 1986;215:351–360. 19. Ludwig J, Ritman EL, LaRusso NF, et al. Anatomy of the human biliary system studied by quantitative computer-aided threedimensional imaging techniques. Hepatology 1998;27:893–899. 20. Motta PM. Scanning electron microscopy of the liver. Prog Liver Dis 1982;7:1–16. 21. Steiner JW, Carruthers JS. Studies on the fine structure of the terminal branches of the biliary tree. Am J Pathol 1961;38:639–661. 22. Takasaki S, Hano H. Three dimensional observation of the human hepatic artery (arterial system in the liver). J Hepatol 2001;34:455–466. 23. McCuskey RS. Distribution of intrahepatic nerves: an overview. In: Shimazu T, ed. Liver innervation. London: John Libbey, 1996: 17–22. 24. McCuskey RS. Anatomy of efferent hepatic nerves. Anat Rec 2004;280A:821–826. 25. Kiernan F. The anatomy and physiology of the liver. Trans Roy Soc Lond 1883;123:711–770. 26. Rappaport AM. The microcirculatory hepatic unit. Microvasc Res 1973;6:218–228. 27. Rappaport AM, Borowy ZJ, Lougheed WM, Lotto WN. Subdivision of the hexagonal liver lobules into a structural and functional unit. Anat Rec 1954;119:11–34. 28. Mall FP. On the origin of the lymphatics in the liver. A study of the structural unit of the liver. Am J Anat 1906;5:227–308. 29. Matsumoto T, Kawakami M. The unit-concept of hepatic parenchyma – a reexamination based on angio architectural studies. Acta Pathol Jpn 1982;32:285–314. 30. Ekataksin W, Kaneda K. Liver microvascular architecture: an insight into the pathophysiology of portal hypertension. Semin Liver Dis 1999;19:359–382. 31. Ekataksin W, Wake K. New concepts in biliary and vascular anatomy of the liver. In: Boyer JL, Ockner RK, eds. Progress in liver diseases. Vol. XV. Philadelphia: WB Saunders, 1997:1–30. 32. Ekataksin W, Zou Z, Wake K, et al. The hepatic microcirculatory subunits: An over-three-century-long search for the missing link between the exocrine unit and an endocrine unit in mammalian liver lobules. In: Motta PM, ed. Recent advances in microscopy of cells, tissues and organs. Rome: University of Rome ‘La Sapienza,’ 1997:407–412. 33. McCuskey RS, Ekataksin W, LeBouton AV, et al. Hepatic microvascular development in relation to the morphogenesis of
Chapter 1 ANATOMY OF THE LIVER
34.
35.
36. 37. 38.
39.
40.
41.
42. 43.
44.
45.
46. 47. 48.
49.
50.
51.
52.
53.
hepatocellular plates in neonatal rats. Anat Rec 2003;275A:1019–1030. Teutsch HF, Altemus J, Gerlach-Arbeiter S, Kyander-Teutsch TL. Distribution of 3-hydroxybutyrate dehydrogenase in primary lobules of the rat liver. J Histochem Cytochem 1992;40:213–219. Teutsch HF, Schuerfeld D, Groezinger E. Three-dimensional reconstruction of parenchymal units in the liver of the rat. Hepatology 1999;29:494–505. Elias H, Sherrick JC. Morphology of the liver. New York: Academic Press, 1969. Motta P, Muto M, Fujita T. The liver cell. Tokyo: Igaku-Shoin, 1978: 30–33. Cardell EL, Cardell RR. Structure and function of hepatic parenchymal cells. In: McCuskey RS, Earnest DL, eds. Comprehensive toxicology. Vol. 9. Hepatic and gastrointestinal toxicology. Oxford: Elsevier Science, 1997: 11–34. Schachter D. The hepatocyte plasma membrane: organisation and differentiation. In: Arias IM, Jakoby WB, Popper H, et al., eds. The liver: biology and pathobiology. New York: Raven Press, 1988: 131–140. Schachter D. The hepatocyte plasma membrane: organization, differentiation, biogenesis, and turnover. In: Arias IM, Boyer JL, Fausto N, et al., eds. The liver: biology and pathobiology, 4th edn. Philadelphia: Lippincott Williams & Wilkins, 2001: 77–81. Doyle D, Bujanover Y, Petell JK. Plasma membrane: biogenesis and turnover. In: Arias I, Jakoby WB, Popper H, et al., eds. The liver: biology and pathobiology. New York: Raven Press, 1988: 61–69. Feldmann G. Liver ploidy. J Hepatol 1992;16:7–10. Lippincott-Schwarz J. The endoplasmic reticulum–Golgi membrane system. In: Arias IM, Boyer JL, Fausto N, et al., eds. The liver, biology and pathobiology, 3rd edn. New York: Raven Press, 1994: 215–228. Lippincott-Schwarz J. The endoplasmic reticulum and Golgi complex in secretory membrane transport. In: Arias IM, Boyer JL, Fausto N, et al., eds. The liver: biology and pathobiology, 4th edn. Philadelphia: Lippincott Williams & Wilkins, 2001: 119–131. DePierre JW, Andersson G, Dallner G. Endoplasmic reticulum and Golgi complex. In: Arias IM, Jakoby WB, Popper H, et al., eds. The liver: biology and pathobiology. New York: Raven Press, 1988: 165–187. Ma MH, Biempica L. The normal human liver cell: cytochemical and ultrastructural studies. Am J Pathol 1971;62:353–370. Tanikawa K. Ultrastructural aspects of the liver and its disorders, 2nd edn. New York: Igaku-Shoin, 1979. Weibel ER, Staubli W, Gnagi HR, Hess FA. Correlated morphometric and biochemical studies on the liver cell: i. morphometric model, stereologic methods, and normal morphometric data for rat liver. J Cell Biol 1969;42:68–91. Cardell RRJ. Smooth endoplasmic reticulum in rat hepatocytes during glycogen deposition and depletion. Int Rev Cytol 1977;48:221–279. Phillips MJ. Biology and pathobiology of actin in the liver. In: Arias IM, Boyer JL, Fausto N, et al., eds. The liver: biology and pathobiology, 3rd edn. New York: Raven Press, 1994: 19–32. Hinkle PC. Mitochondria. In: Arias IM, Boyer JL, Fausto N, et al., eds. The liver: biology and pathobiology. New York: Raven Press, 1994: 309–317. Johnson LV, Walsh ML, Chen LB. Localization of mitochondria in living cells with rhodamine 123. Proc Natl Acad Sci USA 1980;77:990–994. Cascarano J, Montisano DF, Pickett CB, James TW. Rough endoplasmic reticulum–mitochondrial complexes from rat liver – an extramitochondrial succinic dehydrogenase associated with this rough endoplasmic reticulum. Exp Cell Res 1982;139:39–50.
54. Novikoff AB, Novikoff PM. Lysosomes. In: Arias IM, Jakoby WB, Popper H, et al., eds. The liver: biology and pathobiology. New York: Raven Press, 1988: 227–239. 55. Sahagian GG, Novikoff PM. Lysosomes. In: Arias IM, Boyer JL, Fausto N, et al., eds. The liver: biology and pathobiology. New York: Raven Press, 1994: 275–291. 56. Lazarow PB. Peroxisomes. In: Arias IM, Boyer JL, Fausto N, et al., eds. The liver: biology and pathobiology, 3rd edn. New York: Raven Press, 1994: 293–307. 57. Sotto U, Rapp S, Gorgas K, Just WW. Peroxisomes and lysosomes. In: LeBouton AV, ed. Molecular and cell biology of the liver. Boca Raton: CRC Press, 1994:; 181–262. 58. Kishimoto T, Tavassoli M. Transendothelial transport (transcytosis) of iron-transferrin complex in the rat liver. Am J Anat 1987;178:241–249. 59. Soda R, Tavassoli M. Blood 1984;63:270–276. 60. Feldmann G. The cytoskeleton of the hepatocyte. Structure and functions. J Hepatol 1989;8:380–386. 61. Philips MJ, Satir P. The cytoskeleton of the hepatocyte: organisation, relationships and pathology. In: Arias IM, Jakoby WB, Popper H, et al., eds. The liver: biology and pathobiology. New York: Raven Press, 1988: 11–27. 62. Crawford J. The role of vesicle-mediated transport pathways in hepatocellular bile secretion. Semin Liver Dis 1996;16:169–189. 63. Philips MJ, Oshio C, Miyami M, et al. A study of bile canalicular contractions in isolated hepatocytes. Hepatology 1982;2:763–768. 64. Kawahara H, French SW. Role of cytoskeleton in canalicular contraction in cultured differentiated hepatocytes. Am J Pathol 1990;136:521–532. 65. Watanabe N, Tsukada N, Smith CR, Phillips MJ. Motility of bile canaliculi in the living animal: implications for bile flow. J Cell Biol 1991;113:1069–1080. 66. Phillips MJ. A study of bile canalicular contractions in isolated hepatocytes. Lab Invest 982;47:311–313. 67. Sugisaki T, Sagakuchi T. Intracytoplasmic tonofilaments: a desmosome-like structure in the mouse fetal liver cell. J Ultrastruct Res 1977;59:178–184. 68. Wisse E, Braet F, Luo DZ, et al. Structure and function of sinusoidal lining cells in the liver. Toxicol Pathol 1996;24:100–111. 69. Winnock M, Barcina MG, Lukomska B, et al. Liver-associated lymphocytes: Role in tumor defense. Semin Liver Dis 1993;13:81–92. 70. McCuskey PA. Electron and fluorescence microscopic study of mast cells and adrenergic innervation in Beagle dog liver. In: Wisse E, Knook DL, Decker K, eds. Cells of the hepatic sinusoids. Vol. 2. Leiden: Kupffer Cell Foundation, 1989: 260–265. 71. Blouin A. Morphometry of the liver sinusoidal cells. In: Wisse E, Knook DL, eds. Kupffer cells and other liver sinusoidal cells. Amsterdam: Elsevier Biomedical, 1977: 61–71. 72. Blouin A, Bolender RP, Weibel ER. Distribution of organelles and membranes between hepatocytes and non hepatocytes in the rat liver parenchyma. A stereological study. J Cell Biol 1977;72:441–455. 73. Bouwens L, Baekeland M, Zanger RD, Wisse E. Quantitation, tissue distribution and proliferation kinetics of Kupffer cells in normal rat liver. Hepatology 1986;6:718–722. 74. Bouwens L, Knook DL, Kuppen PJK, et al. Electron microscopic observations on the accumulation of large granular lymphocytes (pit cells) and Kupffer cells in the liver of rats treated with continuous infusion of interleukin-2. Hepatology 1990;12:1365– 1370. 75. Brouwer A, Knook DL, Wisse E. Sinusoidal endothelial cells and perisinusoidal fat-storing cells. In: Arias IM, Jakoby WB, Popper H, et al., eds. Liver: biology and pathobiology. New York: Raven Press, 1988: 665–682. 76. Wisse E, DeZanger RB, Jacobs R, et al. The liver sieve: consideration concerning the structure and function of
19
Section I. Pathophysiology of the Liver
77.
78.
79.
80. 81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95. 96.
20
endothelial fenestrae, the sinusoid wall and the space of Disse. Hepatology 1985;5:683–692. Horn T, Henriksen JH, Christoffersen P. The sinusoidal lining cells in ‘normal’ human liver. A scanning electron microscopic investigation. Liver 1986;6:98–110. Vidal-Vanaclocha F, Barbera-Guillem E. Fenestration patterns in endothelial cells of rat liver sinusoids. J Ultrastruct Res 1985;90:115–123. Gumucio JJ, Bilir BM, Moseley RH, Berkowitz CM. The biology of the liver cell plate. In: Arias I, Boyer JL, Fausto N, et al., eds. The liver: biology and pathobiology, 3rd edn. New York: Raven Press, 1994: 1143–1163. Jungermann K. Metabolic zonation of liver parenchyma. Semin Liver Dis 1988;8:329–341. Teutsch HF. Regionality of glucose-6-phosphate hydrolysis in the liver lobule of the rat: Metabolic heterogeneity of ‘portal’ and ‘septal’ sinusoids. Hepatology 1988;8:311–317. Lemasters JJ, Ji S, Thurman RG. Centrilobular injury following hypoxia in isolated, perfused rat liver. Science 1981; 213:661–663. Fraser R, Day WA, Fernando NS. Review: The liver sinusoidal cells. Their role in disorders of the liver, lipoprotein metabolism and atherogenesis. Pathology 1986;18:5–11. Fraser R, Dobbs BR, Rogers GWT. Lipoproteins and the liver sieve: The role of the fenestrated sinusoidal endothelium in lipoprotein metabolism, atherosclerosis, and cirrhosis. Hepatology 1995;21:863–874. Oda M, Azuma T, Watanabe N, et al. Regulatory mechanism of hepatic microcirculation: Involvement of contraction and dilatation of sinusoids and sinusoidal endothelial fenestrae. In: Hammersen MK, ed. Progress in applied microcirculation. Vol. 17. Basel: Karger, 1990: 103–128. Arias IM. The biology of hepatic endothelial fenestrae. In: Schaffner F, Popper H, eds. Progress in liver diseases. Vol. IX. Philadelphia: WB Saunders, 1990: 11–26. Braet F, DeZanger R, Baekeland M, et al. Structure and dynamics of the fenestrae-associated cytoskeleton of rat sinusoidal endothelial cells. Hepatology 1995;21:180–189. Oda M, Han JY, Yokomori H. Local regulators of hepatic sinusoidal microcirculation: recent advances. Clin Hemorheol Microcirc 2000;23:85–94. Oda M, Yokomori H, Han JY, et al. Hepatic sinusoidal endothelial fenestrae are a stationary type of fused and interconnected caveolae. In: Wisse E, Knook DL, DeZanger R, et al., eds. Cells of the hepatic sinusoid. Vol. 8. Leiden: Kupffer Cell Foundation, 2001: 94–98. Braet F, Spector I, Zanger RD, Wisse E. A novel structure involved in the formation of liver endothelial cell fenestrae revealed by using the actin inhibitor misakinolide. Proc Natl Acad Sci USA 1998;95:13635–13640. Cogger VC, Warren A, Fraser R, et al. Hepatic sinusoidal pseudocapillarization with aging in the non-human primate. Exp Gerontol 2003;38:1101–1107. LeCouteur DG, Fraser R, Cogger VC, McLean AJ. Hepatic pseudocapillarisation and atherosclerosis in ageing. Lancet 2002;359:1612–1615. Smedsrod B, Melkko J, Araki N, et al. Advanced glycation end products are eliminated by scavenger-receptor-mediated endocytosis in hepatic sinusoidal Kupffer and endothelial cells. Biochem J 1997;322:567–573. Wisse E. An electron microscopic study of the fenestrated endothelial lining of rat liver sinusoids. J Ultrastruct Res 1970;31:125–150. Smedsrod B, DeBleser PJ, Braet F, et al. Cell biology of liver endothelial and Kupffer cells. Gut 1994;35:1509–1516. Wisse E, Braet F, Luo D, et al. Structure and function of sinusoidal lining cells in the liver. Toxicol Pathol 1996;24:100–111.
97. Steffan A-M, Gendrault J-L, McCuskey RS, et al. Phagocytosis, an unrecognized property of murine endothelial liver cells. Hepatology 1986;6:830–836. 98. VanOosten M, Bilt Evd, Vries HEd, et al. Vascular adhesion molecule-1 and intercellular adhesion molecule-1 expression on rat liver cells after lipopolysaccharide administration in vivo. Hepatology 1995;22:1538–1546. 99. Decker K. Biologically active products of stimulated liver macrophages. Eur J Biochem 1990;192:245–261. 100. McCuskey RS, McCuskey PA. The fine structure and function of Kupffer cells. J Electron Microsc Tech 1990;14:237–246. 101. McCuskey PA, McCuskey RS. In vivo and electron microscopic study of the development diabetic cerebral microangiopathy. Microcirc Endothel Lymphatics 1984;1:221–224. 102. McCuskey RS, McCuskey PA, Urbaschek R, Urbaschek B. Species differences in Kupffer cells and endotoxin sensitivity. Infection and Immunity 1984;45:278–280. 103. Sleyster ECH, Knook DL. Relation between localization and function of rat liver Kupffer cells. Lab Invest 1982;47: 484–490. 104. Emeis JJ. Morphologic and cytochemical heterogeneity of the cell coat of rat liver Kupffer cells. J Reticuloendothelial Soc 1976;20:31–50. 105. Wisse E. Observations on the fine structure and peroxidase cytochemistry of normal rat liver Kupffer cells. J Ultrastruct Res 1974;46:393–426. 106. Wisse E, Knook DL. The investigation of sinusoidal cells: a new approach to the study of liver function. In: Popper H, Schaffner F, eds. Progress in liver diseases. Vol. VI. New York: Grune & Stratton, 1979: 153–171. 107. Fahimi HD. The fine structural localization of endogenous and exogenous peroxidase activity in Kupffer cells of rat liver. J Cell Biol 1970;47:247–262. 108. Widmann JJ, Cotran RS, Fahimi HD. Mononuclear phagocytes (Kupffer cells) and endothelial cells. Identification of two functional cell types in rat liver sinusoids by endogenous peroxidase activity. J Cell Biol 1972;52:159–170. 109. Bodenheimer HC, Faris RA, Charland C, Hixson DC. Characterization of a new monoclonal antibody to rat macrophages and Kupffer cells. Hepatology 1988;8:1667–1672. 110. Malorny U, Michels E, Sorg C. A monoclonal antibody against an antigen present on mouse macrophages and absent from monocytes. Cell Tissue Res 1986;243:421–428. 111. Ramadori G, Dienes H, Burger R, et al. Expression of Iaantigens on guinea pig Kupffer cells. Studies with monoclonal antibodies. J Hepatol 1986;2:208–217. 112. Bouwens L, Baekeland M, Wisse E. Cytokinetic analysis of the expanding Kupffer-cell population in rat liver. Cell Tissue Kinet 1986;19:217–226. 113. Crofton RW, Dulk MMCD-D, Furth RV. The origin, kinetics, and characteristics of the Kupffer cells in the normal steady state. J Exp Med 1978;148:1–17. 114. Volkman A. Disparity in origin of mononuclear phagocyte populations. J Reticuloendothelial Soc 1976;19:249–268. 115. Bouwens L, Baekeland M, Wisse E. Importance of local proliferation in the expanding Kupffer cell population of rat liver after zymosan stimulation and partial hepatectomy. Hepatology 1984;4:213–229. 116. Bouwens L, Knook DL, Wisse E. Local proliferation and extrahepatic recruitment of liver macrophages (Kupffer cells) in partial-body irradiated rats. J Leukocyte Biol 1986;39:687–697. 117. Bouwens L, Wisse E. Proliferation, kinetics, and fate of monocytes in rat liver during a zymosan-induced inflammation. J Leukocyte Biol 1985;37:531–543. 118. Deimann W, Fahimi H. The appearance of transition forms between monocytes and Kupffer cells in the liver of rats treated with glucan. J Exp Med 1979;149:883.
Chapter 1 ANATOMY OF THE LIVER
119. Wake K. Sternzellen in the liver: Perisinusoidal cells with special reference to storage of vitamin A. Am J Anat 1971;132:429–461. 120. Wake K. Development of vitamin A-rich lipid droplets in multivesicular bodies of rat liver stellate cells. J Cell Biol 1974;63:683–691. 121. Wake K. Perisinusoidal stellate cells (fat-storing cells, interstitial cells, lipocytes), their related structure in and around liver sinusoids, and vitamin A-storing cells in extrahepatic organs. Int Rev Cytol 1980;66:303–353. 122. Wake K, Motomatsu K, Dan C, Kaneda K. Three-dimensional structure of endothelial cells in hepatic sinusoids of the rat as revealed by the Golgi method. Cell Tissue Res 1988;253:563–571. 123. Kawada N, Tran-Thi TA, Klein H, Decker K. The contraction of hepatic stellate (Ito) cells stimulated with vasoactive substances. Possible involvement of endothelin 1 and nitric oxide in the regulation of the sinusoidal tonus. Eur J Biochem 1993;213:815–822. 124. Pinznai M, Faili P, Ruocco C, et al. Fat-storing cells as liverspecific pericytes: Spatial dynamics of agonist-stimulated intracellular calcium transients. J Clin Invest 1992;90:642–646. 125. Zhang JX, Pegoli W, Clemens MG. Endothelin-1 induces direct constriction of hepatic sinusoids. Am J Physiol 1994;266:G624–G632. 126. Rockey DC. Hepatic blood flow regulation by stellate cells in normal and injured liver. Semin Liver Dis 2001;21:337–349.
127. LeBail B, Bioulac-Sage P, Senuita R, et al. Fine structure of hepatic sinusoids and sinusoidal cells in disease. J Electron Microsc Tech 1990;14:257–282. 128. Gressner AM. Perisinusoidal lipocytes and fibrogenesis. Gut 1994;35:1331–1333. 129. Wisse E, Noordende JMVt, Meulen JVD, Daems WT. The pit cell: Description of a new type of cell occurring in rat liver and peripheral blood. Cell Tissue Res 1976;173: 423–435. 130. Vanderkerken K, Bouwens L, Neve WD, et al. Origin and differentiation of hepatic natural killer cells (pit cells). Hepatology 1993;18:919–925. 131. Bouwens L, Wisse E. Pit cells in the liver. Liver 1992;12:3–9. 132. Kaneda K, Dan C, Wake K. Pit cells as natural killer cells. Biomed Res 1983;4:567–576. 133. Bouwens L, Remels L, Baekeland M, et al. Large granular lymphocytes or pit cells from rat liver: Isolation, ultrastructural characterization and natural killer cell activity. Eur J Immunol 1987;17:37–42. 134. Kaneda K, Wake K. Distribution and morphological characteristics of the pit cells in the liver of the rat. Cell Tissue Res 1983;233:485–505. 135. Vanderkerken K, Bouwens L, Wisse E. Characterization of a phenotypically and functionally distinct subset of large granular lymphocytes (pit cells) in rat liver sinusoids. Hepatology 1990;12:70–75.
21
Section I: Pathophysiology of the Liver
2
LIVER REGENERATION Christian Trautwein Abbreviations APC adenomatous polyposis coli APR acute-phase response CCL4 carbontetrachloride C/EBPs CCAAT/enhancer-binding proteins CDK cyclin-dependent kinases CKIS cyclin-dependent kinase inhibitors CNTF ciliary neurotropic factor CT-1 cardiotropin 1 DSH dishevelled EGF epidermal growth factor EGFR/ epidermal growth factor receptor Erbb1 ERK/ extracellular signal-regulated ECM extracellular matrix FADD Fas-associated death domain FOXM1B forkhead box M1B transcription factor Gab1 growth-factor-receptor-bound protein 2 (Grb 2)-associated binder 1
GSK glycogen synthase-3b HB-EGF heparin-binding EGF HGF hepatocyte growth factor IL interleukin IKK I-kB kinase JAKs Janus kinases JNK Jun terminal kinase LIF leukemia inhibitory factor LRP lipoprotein receptor-related protein MAPkinase kinase/mitogen-activated protein kinase MAPKKK mitogen-activated protein kinase kinase kinase NEMO NF-kB Essential Modulator NF-kB nuclear factor-kB NK natural killer OSM oncostatin M Pak p21-activated protein kinase Pl3K phosphotidylinositol3-kinase
INTRODUCTION The liver stands out for its unique ability to regenerate and thereby restore its original mass after tissue loss. Major progress has been achieved during the last 50 years in understanding the mechanisms involved in controlling this process. However, this phenomenon was already known in Greek mythology: the Titan Prometheus stole the fire from Zeus and took it to mankind. Zeus punished Prometheus by chaining him to a rock in the Caucasus mountains. Every day an eagle came and ate from his liver, which regenerated overnight. The liver has a large metabolic task to perform, and normally hepatocyte proliferation in the liver is a rare event. However, liver regeneration is induced following different mechanisms of injury. In humans, examples are acute liver damage after virus infection or liver resection. Moreover, in recent years liver transplantation, and especially split liver transplantation, has become of major importance owing to organ shortage. Therefore, models to study liver regeneration are of direct relevance to a better understanding of the mechanisms that are important also in patients. Additionally, the direct clinical application of split liver transplantation allows further proof of the concepts developed in animal models. Different animal models have been established to study liver regeneration, mainly in rats and mice. More recently, mouse models have been favored because genetic manipulation in this species allows us to directly address the function of specific genes in hepatocyte proliferation after injury. The best-studied model to investigate the mechanisms involved in liver regeneration is that of partial
PKC PKB PPAR RIP Shp2 SODD SOS Stat TGF TNF TRADD TRAF2 TR u-PA
proteinkinase C protein kinase B peroxisome proliferator-activated receptor receptor-interacting protein SH2-domain containing protein tyrosine phosphatase 2 silencer of death domains son-of-sevenless signal transducer and activator of transcription transforming growth factor tumor necrosis factor TNF receptor-associated death domain TNF-R-associated factor 2 thyroid hormone receptor urokinase-type plasminogen activator
hepatectomy. In mice or rats at least 60% of the liver is surgically removed and the impact on cell cycle progression of parenchymal and non-parenchymal liver cells can be investigated. In this classic model, a first wave of hepatocyte proliferation is found in rats after 24 hours and in mice after 40 hours. Non-parenchymal cells follow hepatocyte proliferation several hours later (Figure 2-1).1 In this simple model of partial hepatectomy, basic mechanisms have become evident in the control of hepatocyte proliferation. The correct liver/body-weight ratio can be restored very rapidly, which takes between 7 and 10 days in rodents. This ratio is relatively constant and reflects the balance between liver function and the body’s demands. After resection the removed liver lobes are not replaced, but hepatocytes in the remaining lobes proliferate. Therefore liver regeneration can also be described as compensatory hyperplasia. After liver mass is restored, hepatocytes receive signals that lead to a halt in proliferation. Thus liver regeneration is a tightly regulated process where hepatocytes enter the cell cycle and become quiescent again. Besides the proliferative response after resection or injury, the liver also has the capacity to proliferate without loss of tissue, and this is known as direct hyperplasia. Different agents, e.g. nuclear receptors, have been characterized that trigger direct hyperplasia.2 However, this chapter will focus on the mechanism of liver regeneration known as compensatory hyperplasia. In the last 50 years different aspects of liver regeneration have been covered. In the beginning, the model of partial hepatectomy was established and morphological and metabolic changes during hepatocyte proliferation were studied. Subsequently it became obvious that growth factors are involved in controlling the exact
23
Section I. Pathophysiology of the Liver
created. Before mitosis of a somatic cell can occur an interphase has to be established where certain cellular processes are performed in order to allow cell division. The interphase can be divided into a Gap1 (G1) phase, a DNA-synthesis (S) phase and a Gap2 (G2) phase (Figure 2-2). The G1 and G2 phases are characterized by increasing cell volume and processes that are essential to prepare the cell for chromosomal doubling in S phase or chromosomal segregation during mitosis. Challenging events during cell cycle progression, such as incomplete DNA replication, DNA damage, and depletion of growth factors or mismatching of metaphase chromosomes, result in blocking of cell cycle progression at so-called checkpoints. These are found at late G1 phase (restriction checkpoint), at the end of G2 (G2/M checkpoint) and during mitosis (spindle checkpoint). These checkpoints are crucial in order to stop cell cycle progression until defects have been repaired or, alternatively, cells activate a suicide program leading to the elimination of cells with a dysregulated cell cycle in order to avoid uncontrolled growth. In the adult liver hepatocytes are in a resting state, which is also called G0 (Figure 2-2). The different phases of cell cycle progression are coordinated by proteins called cyclins. Accordingly different members of this family are expressed in a defined manner throughout the cell cycle. Cyclins are the regulatory subunits of cyclin-dependent kinases (CDK) and activate kinases in order to specifically phosphorylate substrates that are crucial in regulating the different phases of cell cycle progression. Members of the cyclin D family (D1, D2, and D3) are expressed in early and middle G1 phase, where they bind to CDK4 and CDK6. Expression of D-cyclins is dependent on stimulation by growth factors. Therefore D-cyclins represent sensors that are needed for the interaction of the cell with the extracellular environment. During this period after partial hepatectomy the induction of cyclin D1 in particular plays a pivotal role in triggering the proliferation of hepatocytes.3,4 As a consequence c-myc expression is induced, which is important to trigger hepatocyte proliferation. Additionally, there is evidence that overexpression of c-myc alone is sufficient to stim-
timing of cell cycle progression of hepatocytes. In further studies the intracellular events, especially in the nucleus, were investigated, which resulted in an analysis of changes in the expression and activity of transcription factors. Meanwhile, through the powerful tools offered by genetically manipulated mice, the complex interacting pathways of liver regeneration have been investigated, and these tools help us discover the essential mechanisms required to restore liver mass after injury.
MECHANISMS OF CELL CYCLE REGULATION Eukaryotic cells have the capacity to divide. This process is called mitosis, and during the event two identical daughter cells are
30 0 Percent labelled cells
Hepatocytes Biliary ductular cells Kupffer and lto cells Sinusoidal endothelial cells
20
10
0 0
1
2
3
4
5
6
7
8
9
10
Days after partial hepatectomy Figure 2-1. Time kinetics of DNA synthesis in different liver cell types during liver regeneration after partial hepatectomy. The four major types of liver cell undergo DNA synthesis at different time points after partial hepatectomy. In rats, hepatocyte DNA synthesis peaks at 24 hours, whereas the other cell types proliferate later. Regenerating hepatocytes produce growth factors that can function as mitogens for these cells. This has suggested that hepatocytes stimulate proliferation of the other cells via a paracrine mechanism. (Reprinted from Science 1977; 276:60–66; Michalopoulos GK, DeFrances MC. Liver regeneration.)
G0
P18ink4C P21WAF1 P27KIP1
P21WAF1 P27KIP1
Cyclin D-CDK 4/6
Cyclin A-CDK 1/2
G1
S
Cyclin A/B-CDK 1
G2
M
Cyclin E-CDK 2/3
P21WAF1 P27KIP1 Restrictioncheck point
24
G2/Mcheck point Spindelcheck point
Figure 2-2. Cell cycle progression. Cell cycle progression is dependent on the orchestrated expression and activation of specific catalytic enzymes (CDKS) with their regulatory units (cyclins). The resting hepatocyte (G0) is activated by different stimuli to enter the cell cycle. During G1 phase cyclin D–CDK4/6 complexes become activated, followed by cyclin E–CDK2/3 at G1/S phase. During S phase the cyclin A–CDK1/2, and during G2/M cyclin A/B–CDK1, are required. Additionally, cell cycle progression is controlled through a complex network of proteins where the cyclin-dependent kinase inhibitors (CKIs) play an important role. They interact with the specific cyclin–Cdk complexes and thus can manipulate their activity. Using knockout animals in hepatocytes, so far a role especially for p21, p27 and p18I has been shown in manipulating cell cycle progression during liver regeneration.
Chapter 2 LIVER REGENERATION
ulate proliferation and oncogenic transformation of hepatocytes in vivo, whereas blocking its expression reverses this effect.5 At late G1 phase control of cell cycle progression by the cyclin D family becomes less relevant and is followed by the expression of cyclin E (E1, E2). During the transition, the strength and timing of DNA synthesis after partial hepatectomy is strongly controlled by the expression and cooperation of CDK inhibitors, which are involved in controlling CDK activity. In particular, p18, p21 and p27 have been shown to play an important role during this period.6–8 After passing the restriction point cell cycle progression will become independent of growth factors, meaning that after this point the depletion of growth factors will have no impact on cell cycle progression, as these cells will complete mitosis. After the restriction point, expression of E-cyclins will increase until G1-/S-phase transition, whereas at later time points the protein will become degraded. Cyclins E1 and E2 build complexes with CDK2 and CDK3, respectively. It has long been thought that the activity of these complexes is necessary to coordinate the mechanisms essential for G1-/S-phase progression. The fundamental understanding has recently been questioned by the work of different groups. Experiments using knockout animals for CDK2 or double-knockout mice for cyclins E1 and E2 indicated that the cyclin E/CDK2 complex might be less essential than at first thought for G1-/ S-phase transition, and as originally described in Drosophilz melanogaster.9–12 However, at present it is not clear whether these findings are also important for liver regeneration, as experiments that address the question whether a resting cell in G0 phase may need the complex in order to go into mitosis are lacking. In S phase CDK2 and CDK1 interact with cyclin A, whereas in G2 complexes consisting of cyclins A and B with CDK1 become relevant to drive cell cycle progression. After reaching the G2-/Mphase passage members of the cyclin B family alone take over to control mitosis. In recent years Costa’s group used the partial hepatectomy model to further characterize the role of the forkhead box M1B transcription factor (FoxM1B) – a winged helix transcription factor – for cell cycle progression. This transcription factor is expressed in all replicating cells and plays a role especially after the G1-/S-phase checkpoint, although it is not expressed in terminally differentiated cells.13 The group demonstrated that hepatocyte-specific deletion of FoxM1B results in strong reduction of DNA synthesis. Lack of FoxM1B expression is associated with increased protein expression of the CDK inhibitor p21 and reduced Cdc25A phosphatase levels, which leads to decreased CDK2 activation and progression into S phase.14 In addition, FoxM1B has been shown to correlate with the regenerative capacity of the aging liver. Re-establishing FoxM1B expression in the aging liver reverses proliferative capacity after partial hepatectomy, indicating that FoxM1B might be an important limiting factor for hepatocyte proliferation in the aging liver.15 The links of FoxM1B to cell cycle progression are CDKspecific phosphorylation sites in the transactivation domain of FoxM1B. Phosphorylation of these sites results in target gene transcription of genes involved in S- or M-phase progression.16 The results as found for FoxM1B demonstrate that the partial hepatectomy model is also useful to study general cell cycle-dependent mechanisms in vivo.
MECHANISMS CONTROLLING TRANSITION OF RESTING HEPATOCYTES INTO MITOSIS Mitosis in an uninjured liver is a rare event. After loss of tissue mass different growth factors, cytokines and molecular events were characterized for their role to mediate transition of the resting hepatocytes into mitosis. As outlined above, cells are sensitive to extracellular stimuli until they pass the G1 restriction point. In order to better define the events that drive hepatocytes during liver regeneration from G0 until the G1 restriction point, two time periods have been defined: the priming and the transition phase. The priming phase reflects the first hours after it becomes evident that the liver needs to restore its mass. This is the period that triggers hepatocytes to leave G0 and enter G1 phase. The transition phase describes the period after entry into G1 until the G1 restriction time point. In these two phases different growth factors and intracellular events become relevant, and so these will be outlined in more detail. The time period of the two phases may vary between mouse and rat and their respective strains. However, the first priming period is short and the events take place in the first 1–3 hours after injury. In the priming phase the extracellular environment and, as a consequence, also the intracellular pathways that become activated, changes dramatically. Among the mechanisms that are relevant to prime the hepatocytes are changes in the extracellular matrix, and activation of growth factors, cytokines and chemokines. These different components interact with each other. Interestingly, lack of expression of one of the genes frequently has a rather mild phenotype, as a redundancy of the systems appears to exist so that pathways can at least in part compensate for each other. Additionally, in nature the artificial differentiation between the priming and the transition periods is less strict, as during liver regeneration many steps occur in parallel.
CHANGES IN THE EXTRACELLULAR MATRIX Besides parenchymal and non-parenchymal cells, the extracellular matrix is important for liver homeostasis and plays an important role in initiating liver regeneration. In order to allow cell proliferation the extracellular matrix (ECM) has to be modified. Changes in the extracellular matrix have different effects. They allow the hepatocyte to be less attached to surrounding cells, and thus give the cells space for mitosis. Additionally, these events trigger intracellular pathways, which are important for G0/G1 transition of hepatocytes. One of the first ECM components to become activated is the urokinase-type plasminogen activator (u-PA). One minute after partial hepatectomy higher u-PA activity can already be detected, and continuously increases during the first 60 minutes.17 u-PA is one of the major initiators of the metalloproteinase cascade leading to ECM degradation and proteolysis, and in addition is an activator of plasminogen and hepatocyte growth factor (HGF). Consecutively a
25
Section I. Pathophysiology of the Liver
higher conversion rate of plasminogen to plasmin and increased fibrinogen breakdown can be detected in the first minutes after partial hepatectomy.17 Matrix degradation results in the release of matrix-bound factors such as HGF. Furthermore, the event is also involved in activating HGF: the inactive single chain of HGF is transformed into the mitotically active two-chain form, which is important to trigger the intracellular pathways required for hepatocyte proliferation.18–21 An important finding is the observation that disruption of the interaction between the cell adhesion molecules E-cadherin and bcatenin is an early step during liver regeneration. In the first 5 minutes after partial hepatectomy nuclear translocation of b-catenin increases, which results in target gene transcription.22 These are genes involved in mediating cell proliferation, such as cyclin D1 and c-myc. E-cadherin shows an opposite regulation to b-catenin. Together these results indicate that modulation of cell adhesion molecules at the outer surface of hepatocytes activates b-catenin, which is involved in contributing to prime hepatocytes. Besides the interaction with cell adhesion molecules, growth factors also have an impact on disruption of the interaction between b-catenin and E-cadherin (Figure 2-3). These are, for example, the HGF/C-metand the EGF/EGF-receptor systems.23
Wnt Cadherin LRP
Frizzled
The pathways interacting with b-catenin are even more complex, as this system also allows interaction with growth factors. Besides binding to the intracellular part of E-cadherin, b-catenin is also present in the cytoplasm, where it directly interacts with other proteins that direct the molecule to the ubiquitin degradation system. After binding of Wnt to its receptor, frizzled degradation of bcatenin is inhibited and as a result the protein translocates to the nucleus, where it activates target gene transcription. Therefore, bcatenin can be activated via two independent mechanisms – Ecadherin and Wnt – which also interact with each other through a complex network.23
LIGAND–RECEPTOR INTERACTION ACTIVATED DURING LIVER REGENERATION Several growth factors, cytokines and chemokines have been shown to interact with receptors on the cell membrane, activating the intracellular signaling pathways involved in target gene activation in hepatocytes. The growth factors were characterized by their mitogenic capacity. Therefore, at present HGF, EGF and transforming growth factor (TGF)-a can be differentiated as mitogens, whereas insulin and epinephrine are classified as co-mitogens. Additionally, growth inhibitory factors – TGF-b and activin A – have been described that are downregulated during the early phase of liver regeneration, but are important to inhibit cell proliferation after liver mass has been restored.
b-catenin a-catenin Dsh b-catenin
Axin GSK
APC b-catenin
b-catenin
TCF
Figure 2-3. Schematic of b-catenin activation via Wnt and cadherin. Binding of Wnt to its receptors Frizzled and LRP (lipoprotein receptor-related protein) results in Dsh activation and the accumulation of b-catenin and interaction with TCF that regulates target gene transcription. In unstimulated cells the level of b-catenin is kept low through degradation by the proteosome system involving Axin, adenomatous polyposis coli (APC) and glycogen synthase-3b (GSK). Dsh (dishevelled) uncouples b-catenin from this protein complex. Additionally, the cytoplasmic domains of type I cadherin binds b-catenin and thus links the protein via a-catenin to the actin cytoskeleton. The interaction of these molecules is controlled by phosphorylation. In general activation of tyrosine kinases, e.g. by growth factors, results in loss of cadherin-mediated cell–cell adhesion and thus increases b-catenin expression and gene transcription. Both possibilities result in the activation of processes involved in cell adhesion and cell migration that play a role during liver regeneration.
26
MITOGENIC GROWTH FACTORS HEPATOCYTE GROWTH FACTOR (HGF) HGF is the best-characterized mitogenic growth factor involved in stimulating liver regeneration and was first isolated and purified from the serum of a patient with fulminant hepatic failure and from rats after partial hepatectomy.24,25 Meanwhile, it has become evident that HGF and the scatter factor, or HGF, are the same molecules,26 and thus the protein also has tasks in cell tissues besides the liver. Activation of intracellular pathways via HGF occurs after binding to its receptor c-met (Figure 2-427). Crucial for the downstream activation of c-met-dependent pathways is the phosphorylation of two tyrosine residues at its intracellular domain. Tyrosine phosphorylation creates docking sites for substrates. The most relevant partner is Gab1. Through a specific Met-binding-site, Gab1 interacts with c-met and becomes phosphorylated. Phosphorylated Gab1 binds signal molecules such as the SH2 domain-containing protein tyrosine phosphatase 2 (Shp2), PI3K, phospholipase C and CRK. One of the major downstream pathways to become activated through c-met/Gab1/Shp2, is the ERK/MAPK pathway that triggers transcription factors such as ETS/AP1 and adhesion molecules. This mechanism is directly involved in mediating cell proliferation, whereas PI3K, via Akt/protein kinase B, confers cell survival. Besides these main signaling pathways c-met also activates Jun terminal kinase (JNK), signal transducer and activator of transcription (Stat) 3, nuclear factor-kB (NF-kB) and b-catenin (for review, see 27).
Chapter 2 LIVER REGENERATION
HGF a b c-met p p
Grb2 Sos
Rac1
Cdc42
Ras
Gab1
Crk
Shp2
Raf
PI3K
C3G Akt/PKB
Pak
Rap1
ERK/MAPK
Cadherins
pRB Cdk6 p27
Cell polarity Motility
Proliferation Cell-cycle
uPA MMPs Fibronectin
FAK Integrins
Migration Cell junction
c-Met can also interact with other membrane receptors on the cell surface, e.g. E-cadherin, b4-integrin or Fas. This receptor cross-talk may also have a direct effect on the cellular response of the cell (for review, see 27). Therefore, HGF/c-met can interact on different levels with cell cycle progression during liver regeneration. HGF serum concentrations are elevated after partial hepatectomy, and in particular the active two-chain form can be detected.28 Activation of HGF/c-met-dependent signaling is found directly after partial hepatectomy, but also at later time points. Phosphorylation of c-met was detected within the first 5 minutes after surgery.29 The role of c-met in cell cycle progression during liver regeneration has recently been further clarified by two independent groups using conditional c-met knockout mice.30,31 These experiments provided evidence that after partial hepatectomy ERK activation is selectively mediated via the HGF/c-met system, which is associated with a reduction in DNA synthesis.30 Also after carbontetrachloride (CCL4) injury impaired regeneration was found and inflammatory changes in the livers of these animals were more prolonged. Additionally, the animals, which lacked c-met expression, showed higher sensitivity versus Fas-induced apoptosis.31 Therefore, using conditional ablation of the c-met receptor in hepatocytes shows that the signaling pathway is essential for providing proliferative and protective signals during liver regeneration.
EPIDERMAL GROWTH FACTOR (EGF) AND TRANSFORMING GROWTH FACTOR-a (TGF-a) The epidermal growth factor receptor (EGFR/Erbb1) belongs to a family of structurally related tyrosine kinase receptors, including
Figure 2-4. HGF/c-met-dependent signaling. After ligand binding tyrosines at the intracellular part of the c-met receptor become phosphorylated and serve as docking sites for different adapter molecules, e.g. Gab1, Grb2, phosphotidylinositol 3kinase and others. As a consequence, specific signaling molecules such as Ras, Shp2 and Crk become activated that trigger downstream pathways. These cascades are essential in stimulating gene expression and/or functions involved in, for example, proliferation, migration or survival. Some of the prominent pathways that are activated are the Ras/Raf/ERK/MAPkinase or PI3K/Akt/PKB pathways. Abbreviations: ERK/MAPkinase, extracellular signal-regulated kinase/mitogen-activated protein kinase; Gab1, growth-factor-receptor-bound protein 2 (Grb2)-associated binder 1; Pak, p21-activated protein kinase; PI3K, phosphotidylinositol3kinase; PKB, protein kinase B; Shp2, SH2-domain containing protein tyrosine phosphatase 2; Sos, son-of-sevenless; met a and b receptor subunit (modified from 27).
Bad Caspase-9
Survival
Erbb2/neu, Erbb3 and Erbb4. After ligand binding the receptor can bind homo- or heterodimers. Several growth factors, such as EGF, TGF-a, amphiregulin, heparin-binding EGF (HB-EGF), b-cellulin and epiregulin, can bind EGFR and induce receptor dimerization. Consecutive activation of the intrinsic tyrosine kinase induces complex downstream pathways (Figure 2-5).32–34 The mitogenic role of the EGFR in liver regeneration is thought to be mediated by stimulating MAP kinase activity. Studies in hepatocytes revealed that only EGFR and Erbb3 are expressed in the adult liver. No expression of Erbb2 and Erbb4 can be detected. Additionally, the expression of EGFR and Erbb3 does not change in hepatocytes during liver regeneration.35 For two ligands – TGF-a and EGF – a potential role during liver regeneration has been suggested. The two molecules show around 35% homology. Good evidence for the important role of EGF during liver regeneration first came from experiments in sialoadenectomized rats. In these animals DNA synthesis is reduced after 48 hours and the application of EGF can rescue the phenotype. An effect on DNA synthesis was evident when sialoadenectomy was performed 3 hours after partial hepatectomy, indicating that EGF is not required during the priming phase.36 Additionally, recent experiments using a transgenic mouse that overexpressed heparin-binding (HB)-EGF-like growth factors showed that after partial hepatectomy a stronger impact on DNA synthesis could be found. As HB-EGF also signals via EGFR, these results further provide evidence that factors binding to EGF receptors are able to stimulate DNA synthesis after partial hepatectomy.37 However, there is also evidence that tyrosine phosphorylation of EGFR in the liver is constitutive and the status
27
Section I. Pathophysiology of the Liver
NOTCH/JAGGED SIGNALING
EGF-R1 (Erbb1) Erbb2/Erbb3/Erbb4 EGF TGF-a
PTK P
P
SATs
Ras PKC
PI3K
Raf
AKT
MAPK
PTEN
FKHR
c-myc c-fos c-jun
GSK-3b
Cyclin D1 Cyclin E
Figure 2-5. Signaling via the epidermal growth factor (EGF) receptor. Binding of EGF or TGF-a to the EGF receptor (EGF-R/Erbb1) or other members of the family (Erbb2/neu, Erbb3, and Erbb4) results in intracellular receptor homo- or heterodimerization. This event stimulates intrinsic tyrosine kinases and phosphorylation of the intracellular receptor domains. As a consequence, the Ras/Raf/MAPkinase cascade, but also other signaling pathways, e.g. PI3K, proteinkinase c (PKC) and Stat proteins, are activated which translate the different EGF-dependent functions at the cellular level.
The Notch/Jagged signaling pathway is relevant in different systems for cell growth and differentiation. Notch genes encode for a family of transmembrane receptors. After ligand binding of Jagged, proteolytic cleavage of intracellular domains occurs. One part of the cytoplasmic domain (NICD) translocates to the nucleus, where NICD binds to the transcription factor CBF1/RBP-Jk, which turns the complex from a repressor into a transactivator protein. Recent evidence demonstrates that NCID nuclear translocation is an early event that occurs in the first 15 minutes after partial hepatectomy.48 When Notch or Jagged expression is blocked before partial hepatectomy via siRNA, this results in a reduction of DNA synthesis in the animals. Thus the Notch/Jagged system seems to be directly involved in triggering cell proliferation during liver regeneration in the priming phase.
ROLE OF CYTOKINES DURING LIVER REGENERATION Besides the direct role of growth factors, it also became evident that cytokines such as TNF-a and interleukin (IL)-6 are involved in priming hepatocytes towards cell cycle progression. Initially it was found that there is a cascade of events leading first to elevated TNFa, and subsequently to increased IL-6 serum levels. Additionally, experiments indicated that anti-TNF antibodies inhibit the proliferation of hepatocytes, again indicating that the molecule might be involved in controlling cell cycle progression.
TUMOR NECROSIS FACTOR-a (TNF-a)
does not change during liver regeneration, indicating that there is no further activation of the receptor after partial hepatectomy.29 Serum levels of TGF-a increase after partial hepatectomy,38 although this increase is modest (twofold).39 TGF-a levels in the serum directly correlate with hepatocyte proliferation after partial hepatectomy.40 TNF has been discovered to be a potential regulator of TGF-a expression after partial hepatectomy.41 During liver regeneration TNF stimulates the metalloproteinase TACE, also known as ADAM 17. TACE cleaves TGF-a at the surface and increases its serum expression. Transgenic animals overexpressing TGF-a in the liver show an increased liver/body-weight ratio.42 However, TGF-a -/- animals have no defect in liver regeneration, indicating that TGF-a in physiological doses is dispensable for liver regeneration.43 Knockout mice constitutive for Erbb2, Erbb3 and Erbb4 die during mid gestation.44–46 Knockout animals for EGFR die at the latest 20 days after birth.47 However, so far no liver phenotype has been reported for any of these mice. As only EGFR and Erbb3 are expressed in hepatocytes, conditional knockout experiments will ultimately resolve whether the pathway is required under normal condition to restore liver mass.
28
TNF signals through two distinct cell surface receptors, TNF-R1 and TNF-R2, of which TNF-R1 initiates the majority of TNF’s biological activities in hepatocytes. Binding of TNF to its receptor leads to the release of the inhibitory protein silencer of death domains (SODD) from TNF-R1’s intracellular domain. This leads to the recognition of the intracellular TNF-R1 domain by the adapter protein TNF receptor-associated death domain (TRADD), which recruits additional adapter proteins: receptor-interacting protein (RIP), TNF-R-associated factor 2 (TRAF2), and Fas-associated death domain (FADD). These proteins then activate distinct signaling cascades (Figure 2-6). FADD recruits caspase 8, which can trigger apoptosis. TRAF2 is upstream of several cascades. It activates cIAP-1 and -2, a mitogenactivated protein kinase kinase kinase (MAPKKK) which ultimately activates c-Jun NH2-terminal kinase (JNK). Additionally, TRAF2 is involved in NF-kB activation. Here also RIP is required, but it does not need its enzymatic activity (for review, see 49). Especially for NF-kB, but also for JNK, a role in liver regeneration has been reported. Activation of NF-kB by TNF requires a complex network of kinases. First, the IKK complex interacts with TRAF2 and RIP. Upon activation the IKK kinase phosphorylates I-kB, which results in its degradation, and as a consequence NF-kB is released to the nucleus, where target gene transcription starts.
Chapter 2 LIVER REGENERATION
TNFa TNF-R1 FADD
Pro-Caspase 8
TRADD
RIP
IKK1 HSP90
CDC3
Caspase 8
cIAP
TRAF2
Anti-apoptosis IKK2
Figure 2-6. TNF-dependent signal transduction. Binding of TNF to its cognate receptor TNF-R1 results in the release of SODD and the formation of a receptor-proximal complex containing the important adapter proteins TRADD, TRAF2, RIP and FADD. These adapter proteins in turn recruit additional key pathway-specific enzymes (for example caspase-8 and IKK2) to the TNF-R1 complex, where they become activated and initiate downstream events leading to apoptosis via caspase 8, NF-B activation involving the IKK-complex, and Jun kinase (JNK) activation.
Mitochondria
NEMO ASK1
RelA
p50
Caspase 9
PP IKBa
IKBa
RelA
RelA
p50
RelA
p50
MKK4/6
Cytochrome C Apaf-1
p50
Proteosome
JNK
c-jun, ATF-2, Elk-1 other substrates
Caspase 3 Cytoplasmic targets
Nuclear targets
Inflammatory/antiapoptotic genes
The high molecular weight IKK complex that mediates the phosphorylation of I-kB has been purified and characterized. This complex consists of three tightly associated I-kB kinase (IKK) polypeptides: IKK1 (also called IKK-a) and IKK2 (IKK-b) are the catalytic subunits of the kinase complex and have very similar primary structures with 52% overall similarity.50–52 Moreover, it contains a regulatory subunit called NEMO (NF-kB Essential Modulator), IKK-g or IKKAP-1.53,54 In vitro, IKK1 and IKK2 can form homoand heterodimers.55 Both IKK1 and IKK2 are able to phosphorylate I-kB in vitro, but IKK2 has a higher kinase activity in vitro than IKK1.56–58 The IKK complex phosphorylates I-kBs at the N-terminal domain at two conserved serines (S32 and S36 in human I-kBa). After phosphorylation, the I-kBs undergo a second post-translational modification: polyubiquitination by a cascade of enzymatic reactions, mediated by the b-TrCP–SCF complex (or the E3IkB ubiquitin ligase complex). This process is followed by the degradation of I-kB proteins by the proteasome, thus releasing NF-kB from its inhibitory IkB-binding partner, so it can translocate to the nucleus and activate transcription of NF-kB-dependent target genes.51,59 Because the enzymes that catalyze the ubiquitination of I-kB are constitutively active, the only regulated step in NF-kB activation appears in most cases to be the phosphorylation of I-kB molecules. NF-kB was first identified in the liver as a factor that is rapidly activated within 30 minutes after partial hepatectomy.60 The importance of NF-kB and TNF signaling was further confirmed by the
fact that liver regeneration is defective in TNF-receptor1 knockout mice that do not show hepatic NF-kB activation after partial hepatectomy.61 The question remained as to whether NF-kB is able to directly promote hepatocyte proliferation in this model. NF-kB has been shown to be capable of directly stimulating the transcription of genes that encode G1-phase cyclins, and a kB-site is present within the cyclin D1 promoter.62,63 Additionally, experiments using an adenovirus of non-degradable I-kBa super-repressor, which blocks NF-kB activation, indicated that NF-kB activation after partial hepatectomy is required for liver regeneration. Animals treated with the virus showed a lack of hepatocyte proliferation and increased apoptosis.64 In contrast, Chaisson et al. used transgenic mice that expressed the non-degradable I-kBa super-repressor specifically in hepatocytes, but only 60% of the hepatocytes expressed the transgene. These mice – in contrast to the adenovirus experiments – showed normal hepatocyte proliferation after partial hepatectomy.65 However, both systems, which were used to block NF-kB activation, have some experimental problems. Therefore, at present it is unclear which level of NF-kB activation is required to allow normal liver regeneration after partial hepatectomy. TNF also triggers Jun kinase (JNK) activity and c-Jun activation during liver regeneration.66,67 Both factors are essential for cell cycle progression after partial hepatectomy. Inhibition of JNK activity results in reduced hepatocyte proliferation and G0/G1 transition of
29
Section I. Pathophysiology of the Liver
hepatocytes. However, no impact on apoptosis was observed.68 Conditional knockout mice for c-Jun have a severe phenotype after partial hepatectomy as half of the mice died, or showed impaired regeneration, increased cell death and lipid accumulation in hepatocytes.69 Together these results demonstrate that JNK/c-Jun activation is crucial to stimulate liver regeneration after partial hepatectomy. Via FADD, TNF can trigger apoptosis via caspase 8 activation. Fas can use the same pathway. However, unlike TNF, hepatocytes are more sensitive to Fas-induced apoptosis as the counterbalancing effect of NF-kB activation is missing.70 During liver regeneration after partial hepatectomy, hepatocytes are less sensitive to Fasinduced apoptosis. Additionally, Fas stimulation enhances hepatocyte proliferation, indicating that the FADD/caspase 8 pathway during liver regeneration induces pro-proliferative effects.71
INTERLEUKIN-6 Interleukin-6 (IL-6) belongs to a family comprising IL-6, IL-11, leukemia inhibitory factor (LIF), oncostatin M (OSM), ciliary neurotropic factor (CNTF) and cardiotropin 1 (CT-1), all of which need the gp130 molecule for signal transduction.72,73 Cytokines of the IL6 family interact with a receptor complex on the cell surface in which gp130 is the central molecule, as it is used by several family members for signal transduction. IL-6 first binds the IL-6 receptor (gp80) and then interacts with gp130. Subsequently, dimerization of two gp130 molecules activates Janus kinases (Jaks), which phosphorylate specific tyrosine residues of gp130 and thus activate the SHP2/ERK/MAP pathways or the transcription factors STAT1 and STAT3 (Figure 2-7).72,73
IL-6
JAKs (Jak1/2/Tyk2) Y2 Shp2 Y3 Y4 Y5 Y6 Y1
IL-6R/ gp80 gp130
STAT3 activation
STAT1 Ras/Mapactivation pathway
Figure 2-7. Interleukin-6-dependent signaling. On the cell surface interleukin6 (IL-6) first interacts with the IL-6 receptor (IL-6R)/gp80. This complex interacts with gp130 molecules and in turn triggers intracellular dimerization. Receptorbound Janus kinases (JAKs: Jak1/2/Tyk2) became activated and phosphorylate tyrosines as the intracellular part of gp130. The phosphorylated tyrosines are essential to activate downstream pathways. Although phosphorylation of the second tyrosine is important to trigger the Ras/Map pathway via SH2-domaincontaining protein tyrosine phosphatase 2 (Shp2), the four distal tyrosines are essential to activate Stat transcription factors.
30
Shortly after the STAT transcription factors were identified74 it became evident that there is transient IL-6-dependent STAT3 activation after partial hepatectomy, which is restricted to the first hours as in turn its inhibitor SOCS3 is immediately induced and thus limits its activity.75–77 The ultimate proof for the relevance of IL-6 for liver regeneration came from experiments with IL-6-/- mice. First experiments published by Taub’s group demonstrated that these animals had a defect in hepatocyte proliferation after partial hepatectomy. Significantly more of the IL-6-/- animals died than did the wt control mice.78 The relevance of these findings was further emphasized as the defect in liver regeneration found in TNFR-1-/- mice could be reversed by IL-6 injection.61 Through these two findings the hypothesis was raised that IL-6 is an essential factor in driving the resting hepatocyte into the cell cycle. Further experiments were aimed at better defining the pathways activated by IL-6 that are essential for liver regeneration. The most prominent factor activated by IL-6 in hepatocytes is STAT3. Treatment of IL-6-/- mice after partial hepatectomy with stem cell factor restored Stat3 activation and DNA synthesis.79 As Stat3 knockout mice are embryonal lethal,80 conditional knockout mice with a hepatocyte-specific knockout for STAT3 were used to study the role of IL-6/gp130-dependent STAT3 activation during liver regeneration. These animals also showed impairment in liver regeneration, resembling the results of IL-6-/- animals.81 Therefore these results suggested that the STAT3 pathway in particular seems necessary for liver regeneration following partial hepatectomy. However, in these animals there was strong STAT1 activation, which is normally not found after partial hepatectomy. STAT1 is known to mediate effects opposite to those of STAT3. Therefore this experimental setting has major problems to solve the role of STAT3 during liver regeneration. Blindenbacher et al. performed a careful study in IL-6-/- mice to better define the role of IL-6 during liver regeneration.82 They tested to see whether IL-6 has a direct impact on hepatocyte proliferation or body homeostasis. Using intravenous or subcutaneous IL-6 injection the authors found that the role of IL-6 seems not to be directly involved in stimulating hepatocyte proliferation, but in maintaining body homeostasis in order to allow normal liver regeneration. These results were further confirmed in conditional knockout animals for gp130. These animals showed normal liver regeneration compared to wt animals.83 However, after LPSinjection – mimicking bacterial infection – more of the gp130-/animals died than did controls, and showed impaired hepatocyte proliferation. Taken together, the work of these groups indicates that IL-6/gp130 is involved in contributing to liver regeneration through mechanisms that are not directly related to cell cycle control. At present the pathways relevant to mediate this effect are not completely understood. However, in recent years several reports have demonstrated that IL-6 activates antiapoptotic pathways also in hepatocytes. Earlier experiments by Kovalovich et al. demonstrated that IL-6 can activate BcL-xL expression, and a role for activating Akt has also been suggested.84,85 Therefore these results indicate that IL-6/gp130 might be relevant to directly protect hepatocytes during cell cycle progression. Additionally, IL-6 induces pathways involved in mediating immune-dependent mechanisms. IL-6 via STAT3 is the major cytokine to induce the acute-phase response (APR) in the liver. The
Chapter 2 LIVER REGENERATION
APR is a first line of defense in the body, but is also involved in the regulation of other pathophysiologic mechanisms, e.g. macrophage activation and interaction with the complement system.86 Besides controlling APR expression, IL-6 contributes to the regulation of the Th1/Th2 response.87 Therefore, these IL-6-dependent tasks could also be relevant in contributing to body homeostasis after partial hepatectomy.
INTERFERON-g Intracellular pathways activated via interferon-g comprise distinct pathways. However, the most prominent one is STAT1. STAT1 has been shown to directly activate the cyclin-dependent kinase inhibitors p21 and IRF-1. For both factors a direct role in inhibiting cell cycle progression has been demonstrated.88,89 After partial hepatectomy overexpression of p21 in hepatocytes results directly in inhibition of liver regeneration.90 Also for IRF-1 there is in vitro evidence in hepatocytes that IFN-g-induced expression results in cell cycle arrest.91 During liver regeneration after partial hepatectomy natural killer (NK) cells are activated, which results in secretion of interferon-g. Induction of NK cell activity, for example by pIpC or murine CMV infection, results in a strong release of IFN-g by NK cells, which is associated with inhibition and retardation of hepatocyte proliferation. Additionally, stronger DNA synthesis after partial hepatectomy was found in interferon-g-/- animals.92 Therefore, interferon-g is a physiological inhibitor of liver regeneration, and a strong immunological response, e.g. during immune-mediated liver disease, might be associated with a lack of liver regeneration.
CHEMOKINES TNF is able to induce multiple intracellular functions. Among these is the induction of several members of the CXC chemokine family. These are heparin-binding proteins with four conserved cysteine amino acids, where the first two cysteines are separated by one nonconserved amino acid.93 Members of this family are IL-8, MIP-2, ENA-78, IP-10 and others. They are best known because of their chemotactic properties. ENA-78 and MIP-2 in particular can be produced after TNF stimulation also by hepatocytes, and therefore the question emerged as to whether these chemokines could be responsible for some of the TNF-dependent effects during liver regeneration.94,95 Hepatic ENA-78 and EPA levels are increased after partial hepatectomy,96 and manipulating both chemokine-dependent effects has a major impact on liver regeneration. Blockage of either of the two chemokines impairs liver regeneration, whereas stimulation experiments with MIP-2 results in enhanced hepatocyte proliferation after partial hepatectomy.95 As shown for IL-6, MIP-2 via its receptor CXCR2 also activates STAT3,96,97 suggesting that this could be one of the mechanisms by which MIP-2 enhances cell cycle progression of hepatocytes. A positive effect on liver regeneration has also been demonstrated for IP-10. IP-10 activates intracellular signals as shown for MIP-2 via the CXC-R2 receptor.98 Therefore it is likely that all members of the CXC chemokine family are able to stimulate hepatocyte proliferation, at least to a certain extent, using the CXC-R2 receptor.
ENERGY METABOLISM DURING LIVER REGENERATION In the early phase of liver regeneration after partial hepatectomy the liver accumulates large amounts of triglycerides.99 Fat accumulation is reduced at the start of DNA synthesis. Therefore fat accumulation in hepatocytes seems important during the process of liver regeneration. In order to better characterize the relevance of this observation, Shteyer et al. blocked fat accumulation after partial hepatectomy by two different approaches and in both cases liver regeneration was impaired. These results demonstrate that the tight regulation of fat accumulation is essential to allow proliferation of hepatocytes.100 The molecular mechanisms behind these observations are at present not completely understood. However, in recent years mainly two families of transcription factors – CCAAT/enhancer-binding proteins (C/EBPs) and PPAR – that are important in regulating genes involved in lipid metabolism have been studied during liver regeneration. The C/CBP family consists of several members and a role for hepatocyte proliferation both in vivo and in vitro has been demonstrated.101,102 After partial hepatectomy an up-regulation of C/EBPb and a down-regulation of C/EBPa can be observed which triggers changes in target gene expression.103 These changes occur before the start of DNA synthesis. Lack of C/EBPb expression in knockout animals results in impaired DNA synthesis and is associated with hypoglycemia and a lack of immediate early gene expression.104 Additionally, lack of expression of insulin-like growth factor-binding protein also blocks C/EBPb induction after partial hepatectomy and a phenotype related to the phenotype found in C/EBPb-/- mice.105 Therefore, direct cross-talk between metabolic functions and C/EBPb expression is obviously important during liver regeneration, and these changes are also essential regulators during adipocyte differentiation.106 Thus changes in C/EBP activity may also determine adipogenic changes found during liver regeneration. PPARa (peroxisome proliferator-activated receptor a) belongs to the family of nuclear receptors and there is emerging evidence that it might be a critical modulator controlling the energy flux important for repair of liver damage. PPARa especially controls the constitutive expression of genes involved in mitochondrial fatty acid b-oxidation, including carnitine palmitoyltransferase-1.107,108 Mitochondrial b-oxidation of fatty acids has been shown critical for energy metabolism during liver regeneration.109 Therefore, partial hepatectomy experiments in PPARa-/- mice were interesting as they showed delayed and reduced DNA synthesis. These findings were associated with a lack of cyclin D1 and c-myc expression, indicating that energy metabolism is critical in the early priming phase, allowing hepatocytes to leave the resting G0 state.110 Besides its role in energy metabolism there is also evidence that PPARa directly induces cyclin D1 expression. Experiments with nafenopin, a peroxisome proliferator, and a PPARa ligand show that via this mechanism this drug can induce hyperplasia of the liver.111 In recent years a second class of nuclear receptor thyroid hormone receptor (TR) has been shown to stimulate liver regeneration after partial hepatectomy via its ligand thyroid hormone (T3), apart from its effect on direct hyperplasia. As shown for PPARs, T3 directly induces cyclin D1 expression112 and thus enhances liver regenera-
31
Section I. Pathophysiology of the Liver
tion after partial hepatectomy, especially during the first 24 hours after surgery.113 At present there are no data concerning whether T3 may also have an effect on the energy metabolism of hepatocytes during liver regeneration. Together these results indicate that hepatocytes require a tightly regulated network controlling energy metabolism in order to allow cell cycle progression. Further work in this area might lead to the possibility of stimulating the regenerative capacity of hepatocytes in diseased livers.
INHIBITION OF HEPATOCYTE PROLIFERATION DURING LIVER REGENERATION Several factors have been defined that are involved in mediating hepatocyte proliferation. In contrast, the mechanisms leading to cessation of liver growth are incompletely understood. The most prominent factor that seems important is transforming growth factor-b (TGF-b). TGF-b is a multifunctional cytokine involved in different mechanisms, e.g. growth and development. Three forms of TGF-b (TGFb 1–3) are known in mammals, which have 80% identity on the amino acid level. All TGF-b forms bind directly or via co-receptors to the TGF-b type II receptor. In turn they recruit, bind and transphosphorylate type I receptors, thereby stimulating their
protein kinase activity. The activated type I receptors phosphorylate Smad2 or Smad3, which then bind to Smad4. The resulting Smad complex translocates to the nucleus in order to interact in a cellspecific manner with various transcription factors to regulate target gene transcription (Figure 2-8) (for review see 114). In many cells TGF-b inhibits cell proliferation in G1 as it stimulates cyclin-dependent protein kinase inhibitor p15 and blocks the function or production of essential cell cycle regulators, e.g. cyclin-dependent protein kinases 2 and 4 and cyclins D1 and D4.115 A role for TGF-b in hepatocytes was first detected in vitro, where it has strong antiproliferative activity.116 After partial hepatectomy TGF-b mRNA increases immediately117, and infusion of TGF-b after partial hepatectomy transiently delays the start of DNA synthesis.118 Additionally, during liver regeneration hepatocytes acquire a transient resistance against TGF-b by down-regulating TGF-b receptors119 or up-regulating inhibitors of the TGF-b signaling pathway.120 Therefore, these results suggest that TGF-b-dependent signaling is directly involved in controlling liver regeneration at different stages. At the beginning the pathway is down-regulated in order to allow hepatocytes to enter the cell cycle, whereas after DNA synthesis TGF-b sensitivity is restored so as to limit hepatocyte proliferation and terminate liver regeneration.121 The concept that TGF-b-dependent signaling is especially involved in the early phase of liver regeneration has been further confirmed in hepatocyte-specific knockout mice for TGF-b receptor II (TGF-RII). These animals show an earlier and increased DNA
Figure 2-8. TGF-b-dependent signaling. Binding of TGF-b induces phosphorylation and activation of TGF-b-receptor1 (TGFbR1) by the TGF-b-receptor2 (TGFbR2). TGFbR1 phosphorylates Smad2 and Smad3. Both factors interact with Smad4 in the cytoplasm or nucleus and regulate gene transcription in several ways. This includes binding and interaction with other transcription factors, interacting with co-repressors, and binding to factors such as CBP and p300 involved in mediating gene transcription. Smad7 represses signaling by other Smads in order to down-regulate the cascades. Besides activating Smads TGF-b also induces the ERK/MAPkinase cascade that is involved in modulating/inhibiting Smad proteins.
TGFb TGFbR2
TGFbR1 PP Smad3
P
PP Smad2
Smad7
P P Smad2 Smad3
Smad4
Smad4
PP Smad2
PP Smad4
PP Smad2
PP Smad2 PP
Smad2
PP Smad3
Smad2 Smad2
PP
PP
Smad3
PP Smad3
Smad4
32
Chapter 2 LIVER REGENERATION
synthesis. Additionally, the weight of the regenerating liver is increased after partial hepatectomy. However, there was no major difference in cessation of DNA synthesis between TGF-RII-/animals and controls, indicating that additional pathways are involved in blocking DNA synthesis after partial hepatectomy.122 A second member of the TGF-b superfamily, activin A, has also been suggested for blocking hepatocyte proliferation in vitro and after partial hepatectomy.123 Activin A induces intracellular Smad activation through its type II receptor, comparable to TGF-b. Therefore similar roles as for TGF-b in terminating liver regeneration have been suggested. Kogure and co-workers infused follistatin – an activin A receptor antagonist – during liver regeneration and demonstrated that hepatocyte proliferation and liver weight are induced.124 In the TGF-RII-/- animals activin A expression was increased compared to controls. Additionally, after follistatin infusion in TGFRII-/- hepatocyte proliferation was increased after 120 hours, again indicating that the activin A-induced pathway is involved in terminating liver regeneration.122 However, as activin A and TGF-b activate very similar intracellular pathways there is a high chance that the two pathways are able to compensate for each other in both ways. Besides TGF-b and activin A, alternative signaling cascades have been discussed that might be involved in terminating liver regeneration. An attractive candidate is sphingosine 1-phosphate (S1P), as interaction with the G-protein-coupled endothelial differentiation gene Edg-5 activates Rho activity in hepatocytes, which is growth inhibitory. As shown for TGF-b, Edg-5 also increases 24–72 hours after partial hepatectomy, and administration of S1P during liver regeneration increases Rho activity and inhibits DNA synthesis.125 Together the results of the inhibitor pathways of hepatocyte proliferation demonstrate that different pathways do exist which are involved in inhibiting hepatocyte proliferation. Further work in this area will probably help to develop new treatment options to limit the uncontrolled growth of hepatocytes.
REFERENCES 1. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997;276:60–66. 2. Columbano A, Shinozuka H. Liver regeneration versus direct hyperplasia. FASEB J 1996;10:1118–1128. 3. Rickheim DG, Nelsen CJ, Fassett JT, et al. Differential regulation of cyclins D1 and D3 in hepatocyte proliferation. Hepatology 2002;36:30–38. 4. Nelsen CJ, Rickheim DG, Timchenko NA, et al. Transient expression of cyclin D1 is sufficient to promote hepatocyte replication and liver growth in vivo. Cancer Res 2001;61:8564–8568. 5. Shachaf CM, Kopelman AM, Arvanitis C, et al. MYC inactivation uncovers pluripotent differentiation and tumour dormancy in hepatocellular cancer. Nature 2004;431: 1112–1117. 6. Albrecht JH, Poon RY, Ahonen CL, et al. Involvement of p21 and p27 in the regulation of CDK activity and cell cycle progression in the regenerating liver. Oncogene 1998;16:2141–2150. 7. Luedde T, Rodriguez ME, Tacke F, et al. p18(INK4c) collaborates with other CDK-inhibitory proteins in the regenerating liver. Hepatology 2003;37:833–841.
8. Kossatz U, Dietrich N, Zender L, et al. Skp2-dependent degradation of p27kip1 is essential for cell cycle progression. Genes Dev 2004;18:2602–2607. 9. Geng Y, Yu Q, Sicinska E, et al. Cyclin E ablation in the mouse. Cell 2003;114 :431–443. 10. Roberts JM, Sherr CJ. Bared essentials of CDK2 and cyclin E. Nature Genet 2003;35:9–10. 11. Ortega S, Prieto I, Odajima J, et al. Cyclin-dependent kinase 2 is essential for meiosis but not for mitotic cell division in mice. Nature Genet 2003;35:25–31. 12. Parisi T, Beck AR, Rougier N, et al. Cyclins E1 and E2 are required for endoreplication in placental trophoblast giant cells. EMBO J 2003;22:4794–4803. 13. Ye H, Kelly TF, Samadani U, et al. Hepatocyte nuclear factor 3/fork head homolog 11 is expressed in proliferating epithelial and mesenchymal cells of embryonic and adult tissues. Mol Cell Biol 1997;17:1626–1641. 14. Wang X, Kiyokawa H, Dennewitz MB, et al. The Forkhead Box m1b transcription factor is essential for hepatocyte DNA replication and mitosis during mouse liver regeneration. Proc Natl Acad Sci USA 2002;99:16881–16886. 15. Wang X, Krupczak-Hollis K, Tan Y, et al. Increased hepatic Forkhead Box M1B (FoxM1B) levels in old-aged mice stimulated liver regeneration through diminished p27Kip1 protein levels and increased Cdc25B expression. J Biol Chem 2002;277:44310–44316. 16. Major ML, Lepe R, Costa RH. Forkhead box M1B transcriptional activity requires binding of Cdk–cyclin complexes for phosphorylation-dependent recruitment of p300/CBP coactivators. Mol Cell Biol 2004;24:2649–2661. 17. Kim TH, Mars WM, Stolz DB, et al. Extracellular matrix remodeling at the early stages of liver regeneration in the rat. Hepatology 1997;26:896–904; Comment in: Hepatology 1997;26:1069–1071. 18. Mars WM, Zarnegar R, Michalopoulos GK. Activation of hepatocyte growth factor by the plasminogen activators uPA and tPA. Am J Pathol 1993;143:949–958. 19. Mars WM, Kim TH, Stolz DB, et al. Presence of urokinase in serum-free primary rat hepatocyte cultures and its role in activating hepatocyte growth factor. Cancer Res 1996;56: 2837–2843. 20. Naldini L, Tamagnone L, Vigna E, et al. Extracellular proteolytic cleavage by urokinase is required for activation of hepatocyte growth factor/scatter factor. EMBO J 1992;11:4825–4833. 21. Naldini L, Vigna E, Bardelli A, et al. Biological activation of proHGF (hepatocyte growth factor) by urokinase is controlled by a stoichiometric reaction. J Biol Chem 1995;270:603–611. 22. Monga SP, Pediaditakis P, Mule K, et al. Changes in WNT/beta-catenin pathway during regulated growth in rat liver regeneration. Hepatology 2001;33:1098–1109. 23. Nelson WJ, Nusse R. Convergence of Wnt, beta-catenin, and cadherin pathways. Science 2004;303:1483–1487. 24. Gohda E, Tsubouchi H, Nakayama H, et al. Purification and partial characterization of hepatocyte growth factor from plasma of a patient with fulminant hepatic failure. J Clin Invest 1988;81:414–419. 25. Michalopoulos GK. Liver regeneration: molecular mechanisms of growth control. FASEB J 1990;4:176–187. 26. Naldini L, Weidner KM, Vigna E, et al. Scatter factor and hepatocyte growth factor are indistinguishable ligands for the MET receptor. EMBO J 1991;10:2867–2878. 27. Birchmeier C, Birchmeier W, Gherardi E, et al. Met, metastasis, motility and more. Nature Rev Mol Cell Biol 2003;4:915–925. 28. Pediaditakis P, Lopez-Talavera JC, Petersen B, et al. The processing and utilization of hepatocyte growth factor/scatter factor following partial hepatectomy in the rat. Hepatology 2001;34:688–693.
33
Section I. Pathophysiology of the Liver
29. Stolz DB, Mars WM, Petersen BE, et al. Growth factor signal transduction immediately after two-thirds partial hepatectomy in the rat. Cancer Res 1999;59:3954–3960. 30. Borowiak M, Garratt AN, Wustefeld T, et al. Met provides essential signals for liver regeneration. Proc Natl Acad Sci USA 2004;101:10608–10613. 31. Huh CG, Factor VM, Sanchez A, et al. Hepatocyte growth factor/c-met signaling pathway is required for efficient liver regeneration and repair. Proc Natl Acad Sci USA 2004;101:4477–4482. 32. Olayioye MA, Neve RM, Lane HA, et al. The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J 20003;19:3159–3167. 33. Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell 2000;103:211–225. 34. Yarden Y, Sliwkowski MX. Untangling the ErbB signalling network. Nature Rev Mol Cell Biol 2001;2:127–137. 35. Carver RS, Stevenson MC, Scheving LA, et al. Diverse expression of ErbB receptor proteins during rat liver development and regeneration. Gastroenterology 2002;123:2017–2027. 36. Lambotte L, Saliez A, Triest S, et al. Effect of sialoadenectomy and epidermal growth factor administration on liver regeneration after partial hepatectomy. Hepatology 1997;25:607–612. 37. Kiso S, Kawata S, Tamura S, et al. Liver regeneration in heparinbinding EGF-like growth factor transgenic mice after partial hepatectomy. Gastroenterology 2003;124:701–707. 38. Tomiya T, Ogata I, Fujiwara K. Transforming growth factor alpha levels in liver and blood correlate better than hepatocyte growth factor with hepatocyte proliferation during liver regeneration. Am J Pathol 1998;153:955–961. 39. Russell WE, Dempsey PJ, Sitaric S, et al. Transforming growth factor-alpha (TGF alpha) concentrations increase in regenerating rat liver: evidence for a delayed accumulation of mature TGF alpha. Endocrinology 1993;133:1731–1738. 40. Tomiya T, Ogata I, Yamaoka M, et al. The mitogenic activity of hepatocyte growth factor on rat hepatocytes is dependent upon endogenous transforming growth factor-alpha. Am J Pathol 2000;157:1693–1701. 41. Gallucci RM, Simeonova PP, Toriumi W, et al. TNF-alpha regulates transforming growth factor-alpha expression in regenerating murine liver and isolated hepatocytes. J Immunol 2000;164:872–878. 42. Webber EM, Wu JC, Wang L, et al. Overexpression of transforming growth factor-alpha causes liver enlargement and increased hepatocyte proliferation in transgenic mice. Am J Pathol 1994;145:398–408. 43. Russell WE, Kaufmann WK, Sitaric S, et al. Liver regeneration and hepatocarcinogenesis in transforming growth factor-alphatargeted mice. Mol Carcinog 1996;15:183–189. 44. Gassmann M, Casagranda F, Orioli D, et al. Aberrant neural and cardiac development in mice lacking the ErbB4 neuregulin receptor. Nature 1995;378:390–394. 45. Lee KF, Simon H, Chen H, et al. Requirement for neuregulin receptor erbB2 in neural and cardiac development. Nature 1995;378:394–398. 46. Riethmacher D, Sonnenberg-Riethmacher E, Brinkmann V, et al. Severe neuropathies in mice with targeted mutations in the ErbB3 receptor. Nature 1997;389:725–730. 47. Sibilia M, Wagner EF. Strain-dependent epithelial defects in mice lacking the EGF receptor. Science 1995;269:234–238. 48. Kohler C, Bell AW, Bowen WC, et al. Expression of Notch-1 and its ligand Jagged-1 in rat liver during liver regeneration. Hepatology 2004;39:1056–1065. 49. Chen G, Goeddel DV. TNF-R1 signaling: a beautiful pathway. Science 2002;296:1634–1635. 50. DiDonato JA, Hayakawa M, Rothwarf DM, et al. A cytokineresponsive IkappaB kinase that activates the transcription factor NF-kappaB. Nature 1997;388:548–554.
34
51. Karin M. How NF-kappaB is activated: the role of the IkappaB kinase (IKK) complex. Oncogene 1999;18:6867–6874. 52. Regnier CH, Song HY, Gao X, et al. Identification and characterization of an IkappaB kinase. Cell 1997;90:373–383. 53. Rothwarf DM, Zandi E, Natoli G, et al. IKK-gamma is an essential regulatory subunit of the IkappaB kinase complex. Nature 1998;395:297–300. 54. Yamaoka S, Courtois G, Bessia C, et al. Complementation cloning of NEMO, a component of the IkappaB kinase complex essential for NF-kappaB activation. Cell 1998;93:1231–1240. 55. Zandi E, Chen Y, Karin M. Direct phosphorylation of IkappaB by IKKalpha and IKKbeta: discrimination between free and NF-kappaB-bound substrate. Science 1998;281: 1360–1363. 56. Delhase M, Hayakawa M, Chen Y, et al. Positive and negative regulation of IkappaB kinase activity through IKKbeta subunit phosphorylation. Science 1999;284:309–313. 57. Woronicz JD, Gao X, Cao Z, et al. IkappaB kinase-beta: NFkappaB activation and complex formation with IkappaB kinasealpha and NIK. Science 1997;278:866–869. 58. Zandi E, Rothwarf DM, Delhase M, et al. The IkappaB kinase complex (IKK) contains two kinase subunits, IKKalpha and IKKbeta, necessary for IkappaB phosphorylation and NF-kappaB activation. Cell 1997;91:243–252. 59. Yamamoto Y, Gaynor RB. IkappaB kinases: key regulators of the NF-kappaB pathway. Trends Biochem Sci 2004;29:72–79. 60. Cressman DE, Greenbaum LE, Haber BA, et al. Rapid activation of post-hepatectomy factor/nuclear factor kappa B in hepatocytes, a primary response in the regenerating liver. J Biol Chem 1994;269:30429–30435. 61. Yamada Y, Kirillova I, Peschon JJ, et al. Initiation of liver growth by tumor necrosis factor: deficient liver regeneration in mice lacking type I tumor necrosis factor receptor. Proc Natl Acad Sci USA 1997;94:1441–1446. 62. Guttridge DC, Albanese C, Reuther JY, et al. NF-kappaB controls cell growth and differentiation through transcriptional regulation of cyclin D1. Mol Cell Biol 1999;19:5785–5799. 63. Hinz M, Krappmann D, Eichten A, et al. NF-kappaB function in growth control: regulation of cyclin D1 expression and G0/G1to-S-phase transition. Mol Cell Biol 1999;19:2690–2698. 64. Limuro Y, Nishiura T, Hellerbrand C, et al. NFkappaB prevents apoptosis and liver dysfunction during liver regeneration. J Clin Invest 1998;101:802–811. 65. Chaisson ML, Brooling JT, Ladiges W, et al. Hepatocyte-specific inhibition of NF-kappaB leads to apoptosis after TNF treatment, but not after partial hepatectomy. J Clin Invest 2002;110:193–202. 66. Diehl AM, Yin M, Fleckenstein J, et al. Tumor necrosis factoralpha induces c-jun during the regenerative response to liver injury. Am J Physiol 1994;267:G552–561. 67. Westwick JK, Weitzel C, Leffert HL, et al. Activation of Jun kinase is an early event in hepatic regeneration. J Clin Invest 1995;95:803–810. 68. Schwabe RF, Bradham CA, Uehara T, et al. c-Jun-N-terminal kinase drives cyclin D1 expression and proliferation during liver regeneration. Hepatology 2003;37:824–832. 69. Behrens A, Sibilia M, David JP, et al. Impaired postnatal hepatocyte proliferation and liver regeneration in mice lacking c-jun in the liver. EMBO J 2002;21:1782–1790. 70. Galle PR, Hofmann WJ, Walczak H, et al. Involvement of the CD95 (APO-1/Fas) receptor and ligand in liver damage. J Exp Med 1995;182:1223–1230. 71. Desbarats J, Newell MK. Fas engagement accelerates liver regeneration after partial hepatectomy. Nature Med 2000;6:920–923. 72. Taga T, Kishimoto T. Gp130 and the interleukin-6 family of cytokines. Annu Rev Immunol 1997;15:797–819.
Chapter 2 LIVER REGENERATION
73. Heinrich PC, Behrmann I, Muller-Newen G, et al. Interleukin6-type cytokine signalling through the gp130/Jak/STAT pathway. Biochem J 1998;334:297–314. 74. Zhong Z, Wen Z, Darnell JE Jr. Stat3: a STAT family member activated by tyrosine phosphorylation in response to epidermal growth factor and interleukin-6. Science 1994;264:95–98. 75. Cressman DE, Diamond RH, Taub R. Rapid activation of the Stat3 transcription complex in liver regeneration. Hepatology 1995;21:1443–1449. 76. Campbell JS, Prichard L, Schaper F, et al. Expression of suppressors of cytokine signaling during liver regeneration. J Clin Invest 2001;107:1285–1292. 77. Trautwein C, Rakemann T, Niehof M, et al. Acute-phase response factor, increased binding, and target gene transcription during liver regeneration. Gastroenterology 1996;110:1854–1862. 78. Cressman DE, Greenbaum LE, DeAngelis RA, et al. Liver failure and defective hepatocyte regeneration in interleukin-6-deficient mice. Science 1996;274:1379–1383. 79. Ren X, Hogaboam C, Carpenter A, et al. Stem cell factor restores hepatocyte proliferation in IL-6 knockout mice following 70% hepatectomy. J Clin Invest 2003;112: 1407–1418. 80. Takeda K, Noguchi K, Shi W, et al. Targeted disruption of the mouse Stat3 gene leads to early embryonic lethality. Proc Natl Acad Sci USA 1997;94:3801–3804. 81. Li W, Liang X, Kellendonk C, et al. STAT3 contributes to the mitogenic response of hepatocytes during liver regeneration. J Biol Chem 2002;277:28411–28417. 82. Blindenbacher A, Wang X, Langer I, et al. Interleukin 6 is important for survival after partial hepatectomy in mice. Hepatology 2003;38:674–682. 83. Wuestefeld T, Klein C, Streetz KL, et al. Interleukin6/glycoprotein 130-dependent pathways are protective during liver regeneration. J Biol Chem 2003;278:11281–11288. 84. Kovalovich K, Li W, DeAngelis R, et al. Interleukin-6 protects against Fas-mediated death by establishing a critical level of antiapoptotic hepatic proteins FLIP, Bcl-2, and Bcl-xL. J Biol Chem 2001;276:26605–26613. 85. Streetz KL, Wustefeld T, Klein C, et al. Lack of gp130 expression in hepatocytes promotes liver injury. Gastroenterology 2003;125:532–543. 86. Strey CW, Markiewski M, Mastellos D, et al. The proinflammatory mediators C3a and C5a are essential for liver regeneration. J Exp Med 2003;198:913–923. 87. Betz UA, Bloch W, van den Broek M, et al. Postnatally induced inactivation of gp130 in mice results in neurological, cardiac, hematopoietic, immunological, hepatic, and pulmonary defects. J Exp Med 1998;188:1955–1965. 88. Chin YE, Kitagawa M, Su WC, et al. Cell growth arrest and induction of cyclin-dependent kinase inhibitor p21 WAF1/CIP1 mediated by STAT1. Science 1996;272:719–722. 89. Kroger A, Koster M, Schroeder K, et al. Activities of IRF-1. J Interferon Cytokine Res 2002;22:5–14. 90. Wu H, Wade M, Krall L, et al. Targeted in vivo expression of the cyclin-dependent kinase inhibitor p21 halts hepatocyte cell-cycle progression, postnatal liver development and regeneration. Genes Dev 1996;10:245–260. 91. Kano A, Haruyama T, Akaike T, et al. IRF-1 is an essential mediator in IFN-gamma-induced cell cycle arrest and apoptosis of primary cultured hepatocytes. Biochem Biophys Res Commun 1999;257:672–677. 92. Sun R, Gao B. Negative regulation of liver regeneration by innate immunity (natural killer cells/interferon-gamma). Gastroenterology 2004;127:1525–1539. 93. Miller MD, Krangel MS. Biology and biochemistry of the chemokines: a family of chemotactic and inflammatory cytokines. Crit Rev Immunol 1992;12:17–46.
94. Thornton AJ, Ham J, Kunkel SL. Kupffer cell-derived cytokines induce the synthesis of a leukocyte chemotactic peptide, interleukin-8, in human hepatoma and primary hepatocyte cultures. Hepatology 1991;14:1112–1122. 95. Ren X, Carpenter A, Hogaboam C, et al. Mitogenic properties of endogenous and pharmacological doses of macrophage inflammatory protein-2 after 70% hepatectomy in the mouse. Am J Pathol 2003;163:563–570. 96. Colletti LM, Green M, Burdick MD, et al. Proliferative effects of CXC chemokines in rat hepatocytes in vitro and in vivo. Shock 1998;10:248–257. 97. Mellado M, Rodriguez-Frade JM, Manes S, et al. Chemokine signaling and functional responses: the role of receptor dimerization and TK pathway activation. Annu Rev Immunol 2001;19:397–421. 98. Koniaris LG, Zimmers-Koniaris T, Hsiao EC, et al. Cytokineresponsive gene-2/IFN-inducible protein-10 expression in multiple models of liver and bile duct injury suggests a role in tissue regeneration. J Immunol 2001;167:399–406. 99. Murray AB, Strecker W, Silz S. Ultrastructural changes in rat hepatocytes after partial hepatectomy, and comparison with biochemical results. J Cell Sci 1981;50:433–448. 100. Shteyer E, Liao Y, Muglia LJ, et al. Disruption of hepatic adipogenesis is associated with impaired liver regeneration in mice. Hepatology 2004;40:1322–1332. 101. Greenbaum LE, Cressman DE, Haber BA, et al. Coexistence of C/EBP alpha, beta, growth-induced proteins and DNA synthesis in hepatocytes during liver regeneration. Implications for maintenance of the differentiated state during liver growth. J Clin Invest 1995;96:1351–1365. 102. Luedde T, Duderstadt M, Streetz KL, et al. C/EBP beta isoforms LIP and LAP modulate progression of the cell cycle in the regenerating mouse liver. Hepatology 2004;40:356–365. 103. Trautwein C, Rakemann T, Pietrangelo A, et al. C/EBP-beta/LAP controls down-regulation of albumin gene transcription during liver regeneration. J Biol Chem 1996;271:22262–22270. 104. Greenbaum LE, Li W, Cressman DE, et al. CCAAT enhancerbinding protein beta is required for normal hepatocyte proliferation in mice after partial hepatectomy. J Clin Invest 1998;102:996–1007. 105. Leu JI, Crissey MA, Craig LE, et al. Impaired hepatocyte DNA synthetic response posthepatectomy in insulin-like growth factor binding protein 1-deficient mice with defects in C/EBP beta and mitogen-activated protein kinase/extracellular signal-regulated kinase regulation. Mol Cell Biol 2003;23:1251–1259. 106. Rosen ED, Walkey CJ, Puigserver P, et al. Transcriptional regulation of adipogenesis. Genes Dev 2000;14:1293–1307. 107. Aoyama T, Peters JM, Iritani N, et al. Altered constitutive expression of fatty acid-metabolizing enzymes in mice lacking the peroxisome proliferator-activated receptor alpha (PPARalpha). J Biol Chem 1998;273:5678–5684. 108. Kersten S, Seydoux J, Peters JM, et al. Peroxisome proliferatoractivated receptor alpha mediates the adaptive response to fasting. J Clin Invest 1999;103:1489–1498. 109. Lai HS, Chen WJ. Alterations of remnant liver carnitine palmitoyltransferase I activity and serum carnitine concentration after partial hepatectomy in rats. J Surg Res 1995;59:754–758. 110. Anderson SP, Yoon L, Richard EB, et al. Delayed liver regeneration in peroxisome proliferator-activated receptor-alphanull mice. Hepatology 2002;36:544–554. 111. Rao MS, Reddy JK. Hepatocarcinogenesis of peroxisome proliferators. Ann NY Acad Sci 1996;804:573–587. 112. Pibiri M, Ledda-Columbano GM, Cossu C, et al. Cyclin D1 is an early target in hepatocyte proliferation induced by thyroid hormone (T3). FASEB J 2001;15:1006–1013. 113. Malik R, Mellor N, Selden C, et al. Triiodothyronine enhances the regenerative capacity of the liver following partial hepatectomy. Hepatology 2003;37:79–86.
35
Section I. Pathophysiology of the Liver
114. Shi Y, Massague J. Mechanisms of TGF-beta signaling from cell membrane to the nucleus. Cell 2003;113:685–700. 115. Alexandrow MG, Moses HL. Transforming growth factor beta 1 inhibits mouse keratinocytes late in G1 independent of effects on gene transcription. Cancer Res 1995;55:3928–3932. 116. Nakamura T, Tomita Y, Hirai R, et al. Inhibitory effect of transforming growth factor-beta on DNA synthesis of adult rat hepatocytes in primary culture. Biochem Biophys Res Commun 1985;133:1042–1050. 117. Braun L, Mead JE, Panzica M, et al. Transforming growth factor beta mRNA increases during liver regeneration: a possible paracrine mechanism of growth regulation. Proc Natl Acad Sci USA 1988;85:1539–1543. 118. Russell WE, Coffey RJ Jr, Ouellette AJ, et al. Type beta transforming growth factor reversibly inhibits the early proliferative response to partial hepatectomy in the rat. Proc Natl Acad Sci USA 1988;85:5126–5130. 119. Chari RS, Price DT, Sue SR, et al. Down-regulation of transforming growth factor beta receptor type I, II, and III during liver regeneration. Am J Surg 1995;169:126–131. 120. Macias-Silva M, Li W, Leu JI, et al. Up-regulated transcriptional repressors SnoN and Ski bind Smad proteins to antagonize
36
121.
122.
123.
124.
125.
transforming growth factor-beta signals during liver regeneration. J Biol Chem 2002;277:28483–28490. Nishikawa Y, Wang M, Carr BI. Changes in TGF-beta receptors of rat hepatocytes during primary culture and liver regeneration: increased expression of TGF-beta receptors associated with increased sensitivity to TGF-beta-mediated growth inhibition. J Cell Physiol 1998;176:612–623. Oe S, Lemmer ER, Conner EA, et al. Intact signaling by transforming growth factor beta is not required for termination of liver regeneration in mice. Hepatology 2004;40:1098–1105. Yasuda H, Mine T, Shibata H, et al. Activin A: an autocrine inhibitor of initiation of DNA synthesis in rat hepatocytes. J Clin Invest 1993;92:1491–1496. Kogure K, Omata W, Kanzaki M, et al. A single intraportal administration of follistatin accelerates liver regeneration in partially hepatectomized rats. Gastroenterology 1995;108:1136–1142. Ikeda H, Satoh H, Yanase M, et al. Antiproliferative property of sphingosine 1-phosphate in rat hepatocytes involves activation of Rho via Edg-5. Gastroenterology 2003;124:459–469.
Section I: Pathophysiology of the Liver
3
MECHANISMS OF LIVER CELL DESTRUCTION Hartmut Jaeschke Abbreviations AIF apoptosis-inducing factor AMAP 3¢-hydroxyacetanilide A-SMase acidic sphingomyelinase Apaf-1 apoptosis protease-activating factor-1 APAP acetaminophen CAD caspase-dependent DNase CXC cys-X-cys c-FLIP cellular FLICE-inhibitory protein DD death domain DED death effector domain DIABLO direct inhibitor of apoptosis-binding protein with low PI DISC death-inducing signaling complex FADD Fas-associated death domain
FAN FasL IAPs ICAD IL-6 IKK iNOS KC MAT1 MCP-1 MIP-2 MPT mtDNA NAPQI
factor associated with N-SMase Fas ligand inhibitors of apoptosis proteins inhibitor of caspase-dependent DNase interleukin-6 inhibitor of NF-kB kinase inducible nitric oxide synthase keratinocyte-derived chemokine methionine adenosyltransferase 1A monocyte chemoattractant protein-1 macrophage inflammatory protein-2 membrane permeability transition pores mitochondrial DNA N-acetyl benzoquinone imine
INTRODUCTION Most acute and chronic liver diseases are characterized by an excessive rate of death of hepatocytes and other types of liver cell. Therefore, investigating the mechanisms of how liver cells die under various pathophysiological conditions is critically important for the development of therapeutic intervention strategies. In recent years much progress has been made in understanding the intracellular signaling mechanisms that cause cell death. This led to the elucidation of signaling pathways of ‘self-destruction’ (apoptosis) and to increased insight into mechanisms of ‘forced destruction’ (oncotic necrosis). In addition, it is recognized that there is a balance between liver cell death and regeneration. This chapter will focus on recent advances in our understanding of apoptotic and necrotic signaling mechanisms in liver cells.
APOPTOSIS OR ONCOTIC NECROSIS Until about 10 years ago almost all research on mechanisms of cell death focused on necrosis or oncotic necrosis. However, more recently, apoptosis signaling pathways became the center of attention. As a result, many previous conclusions regarding necrotic cell death were challenged,1,2 and today there is a controversy over which is the predominant pathway in most forms of liver disease.3–5 Given the overlap in the signaling pathways between the two modes of cell death, it appears less relevant to argue about the label but preferable instead to focus on the intracellular signaling events. Nevertheless, the term apoptosis should not be used unless the critical
NK NKT cells NO N-SMase PARP-1 RIP1 Smac tBid TGF-a TRADD TRAF2 TUNEL
natural killer NK cells with T-cell receptors nitric oxide neutral sphingomyelinase poly-(ADP-ribose)-polymerase-1 TNF-receptor-interacting protein 1 second mitochondria-derived activator of caspases truncated Bid transforming growth factor-a TNFR1-associated death domain protein TNF-receptor associated factor 2 transferase-mediated deoxyuridine triphosphate nick-end labeling
morphological characteristics of apoptotic cell death are fulfilled, i.e. cell shrinkage, chromatin condensation, and margination and formation of apoptotic bodies (Table 3-1).3 Apoptosis is generally a single cell event and does not per se cause an inflammatory response. Apoptotic bodies are removed without a trace by neighboring cells or tissue macrophages. On the other hand, if the cell swells, shows evidence of vacuolation, karyolysis and karyorrhexis, together with the release of cell contents, it is said to undergo oncotic necrosis. In general, oncotic necrosis affects large numbers of cells clustered together. In addition, oncotic necrosis triggers a significant inflammatory response. Although most of these cell death characteristics can be identified in a tissue (Table 3-1), there are a number of important issues to consider.
SEVERITY OF INSULT DETERMINES MODE OF CELL DEATH Not every type of cellular stress triggers a uniform response with a predetermined mode of cell death. In fact, a very mild insult affecting only a few cellular organelles, e.g. mitochondria, may not cause cell death but instead lead to removal of the damaged mitochondria in the surviving cell by autophagy.6 If the insult is more severe, more mitochondria are involved and the cell undergoes apoptosis. A moderate insult may lead to the release of enough cytochrome c and other mediators into the cytosol to propagate the cell death signal and, on the other hand, leave enough mitochondria intact to maintain cellular ATP levels. However, if the insult is too severe and most of the mitochondria are damaged, cellular ATP levels drop and the cell undergoes oncotic necrosis.3,7 This concept was recently supported in an in vivo model. Short periods of hemorrhage, which
37
Section I. Pathophysiology of the Liver
Table 3-1. Morphological Characteristics and Biochemical Assays for Apoptosis and Oncotic Necrosis
Morphological characteristics Cell morphology
Nuclear morphology (H&E, DAPI) Mitochondrial morphology Inflammation (mainly neutrophil accumulation) Biochemical parameters TUNEL Assay (DNA strand breaks in nucleus and detection of large DNA fragments in cytosol) Antihistone ELISA (detection of small DNA fragments in cytosol) Internucleosomal DNA damage (DNA ladder on agarose gels) Caspase enzyme activities (fluorescence assays are most sensitive)
APOPTOSIS
ONCOTIC NECROSIS
Cell shrinkage, eventually break-up into apoptotic bodies; single cell event even if many cells are affected Chromatin condensation and margination No relevant changes initially; later: swelling and rupture of outer membrane Negative; can be positive during excessive apoptotic cell death and secondary necrosis
Cell swelling, eventually cell contents release; affects large numbers of clustered cells
Nuclear and cytoplasmic staining when nucleus is degraded
>> 10-fold increase over baseline
3–10-fold increase over baseline during karyolysis Positive during karyolysis
Positive Strongly positive (up to >100-fold increase over baseline) Strongly positive; shows decline of proenzymes and the appearance of active fragments
Caspase processing, e.g. active caspase-3 (immunohistochemistry) Cleavage of caspase substrates, e.g. PARP-1 (Western blot) Release of mitochondrial proteins, e.g. cytochrome c, Smac/DIABLO, endonuclease G, AIF Translocation of proteins to mitochondria, e.g. Bax, Bid, tBid.
Strongly positive
Cell contents release, e.g. ALT, AST, caspases, etc. Nuclear staining with propidium iodide, trypan blue etc. Annexin V staining (phosphatidyl serine externalization) Mitochondrial depolarization (detected by membrane potential-indicating probes, e.g. rhodamine 123, JC-1 and others)
Strong inflammatory response can aggravate the existing injury
Nuclear staining
Caspase processing (Western blot)
Death receptor and death ligand expression
Karyolysis, karyorrhexis Swelling and rupture of outer membrane
Strongly positive; shows decline of intact protein and the appearance of fragment Positive; best detected as increase of proteins levels in the cytosol; decline of protein levels in mitochondria is less sensitive Positive, increase of protein levels in mitochondria is more sensitive to detect than reduction of concentration in cytosol Death receptors are constitutively expressed on hepatocytes and nonparenchymal cells; increased expression is no evidence for apoptosis Initially negative; increase in liver enzyme activities in plasma is observed during secondary necrosis Initially negative; will become positive during secondary necrosis when cell membrane permeability is increased Positive; early event in apoptosis Positive; initially affects a small number of mitochondria; progresses during secondary necrosis
Negative or minor increase (<3-fold increase over baseline potentially due to activation of other proteases) Negative; shows no change of the proenzyme levels compared to untreated controls and no active fragments Negative Negative (can be positive if protein may be substrate for other proteases) Positive; release of these proteins can be similar to apoptosis Positive; translocation of these proteins can also occur during oncotic necrosis Increased expression may be related to the inflammatory response Highly positive; in general large increase in liver enzymes in plasma during acute injury phase Positive as soon as cell membrane becomes permeable Negative; can be positive under certain circumstances Positive; affects large numbers of mitochondria in the cells; key feature of necrosis
AIF, apoptosis-inducing factor; ALT, alanine aminotransferase; AST, aspartate aminotransferase; DAPI, 4¢,6-diamidino-2-phenylindole; ELISA, enzyme-linked immunosorbent assay; H&E, hematoxylin and eosin; JC-1, 5,5¢,6,6¢-tetrachloro-1,1¢,3,3¢-tetraethylbenzimidazolyl carboxycyanine iodide; PARP-1, poly-(ADP-ribose) polymerase; Smac/DIABLO, second mitochondriaderived activator of caspases/direct inhibitor of apoptosis-binding protein with low PI; TUNEL, terminal deoxynucleotidyl-transferase-mediated deoxyuridine triphosphate nick-end labeling assay.
allowed a full recovery of hepatocellular ATP levels during resuscitation, induced mainly apoptotic cell death.8 On the other hand, when tissue ATP levels do not recover well after more prolonged hemorrhagic shock, most liver cells undergo oncotic necrosis during resuscitation.8 Thus, the major mode of cell death after a certain insult may vary according to the severity of the stress (Figure 3-1). In addition to the concept that the severity of the insult can determine the mode of cell death, many insults on liver cells not only
38
have a direct effect but also generate an immune response, which can act as a second hit. These different insults may cause an additive effect on the same mode of cell death. In addition, they may trigger apoptosis and oncotic necrosis at the same time. Carbon tetrachloride and galactosamine, which are frequently used model hepatotoxins, induce oncotic necrosis and caspase-dependent apoptosis in a similar number of hepatocytes as independent events.9–11 It is hypothesized that the direct effect of galactosamine on hepa-
Chapter 3 MECHANISMS OF LIVER CELL DESTRUCTION
mild
Autophagy of damaged cell organelles
ATP
Cellular stress e.g.; ischemia / reperfusion, drugs / chemicals
Cell survival
Apoptotic bodies
Phagocytosis of apoptotic bodies
Secondary necrosis
Regeneration
Apoptosis moderate ATP severe Oncotic necrosis severe acute
moderate chronic
Inflammation organ failure
Inflammation wound repair, fibrosis
Figure 3-1. Apoptotic and necrotic cell death in response to cellular stress.
tocytes, together with a Kupffer cell-mediated oxidant stress, is responsible for the oncosis.9 On the other hand, bacterial translocation due to increased gut permeability induces TNF-a formation, which in combination with galactosamine can trigger death receptor-mediated apoptosis in hepatocytes.12 In this case, both modes of cell death can be independently modulated.9
APOPTOSIS AND SECONDARY NECROSIS Only a limited number of cells can actually undergo complete apoptosis, which ends with phagocytosis of the apoptotic bodies. Thus, true apoptosis has to be a relatively rare, single cell event in order not to overwhelm the capacity of neighboring cells. Given the high regenerative capacity of the liver, these cells are then readily replaced. A low rate of liver cell apoptosis is essential for normal tissue turnover, and even a moderately increased rate of apoptosis is of no real pathophysiological consequence. However, if an insult is severe enough and large numbers of cells are affected, the apoptotic process may be aborted owing to the precipitous drop in cellular ATP levels, and cells rapidly lose viability (Figure 3-1). This process, which is called ‘secondary necrosis’, occurs in most pathophysiological processes, both in vivo and in virtually all cell culture experiments.13–19 However, a characteristic feature of secondary as opposed to oncotic necrosis is that it is preceded unequivocally by apoptotic hallmarks at earlier time points. Thus, secondary necrosis can be completely prevented by inhibiting the earlier apoptosis.14–16 Given the overlapping signaling events between apoptosis and oncotic necrosis, the switch from an oncotic process to apoptosis should also be possible. Indeed, there are several examples where in cultured hepatocytes the oncotic process is inhibited, and the cells then undergo apoptosis.18,19 During hepatic ischemia–reperfusion18,20 and acetaminophen hepatotoxicity,19,21,22 hepatocytes die predominantly through oncotic necrosis in vivo and in cell culture. However, inhibition of the oncotic process by fructose and glycine, which respectively maintain cellular ATP levels and prevent loss of cell viability, caused a delayed apoptotic cell death dependent on caspase activation.18,19 Although this supports the concept of a unified
mode of cell death (‘necrapoptosis’),3,7 it must be borne in mind that this switch from oncosis to apoptosis has so far only been shown in cell culture. Thus, the degeneration of the apoptotic cell death pathway into secondary necrosis is a common event both in vivo and in vitro. However, a transition from oncotic necrosis to apoptosis requires an active intervention from outside to prevent the oncotic cell death. The cell may then either recover and survive, or it may still be stressed enough to initiate an apoptotic signaling pathway (Figure 3-1).
APOPTOSIS, NECROSIS AND INFLAMMATION Excessive cell death in the liver causes an acute inflammatory response, with neutrophils being the predominant cell type involved.23–25 Because neutrophils can extravasate into the parenchyma and actively attack target cells,23–25 neutrophils can – but will not always26,27 – dramatically aggravate the existing injury, as has been demonstrated for ischemia–reperfusion injury and obstructive cholestasis.25,28,29 Neutrophils are recruited into the liver by a variety of mechanisms. Release of cell contents can activate complement factors in plasma, and resident macrophages (Kupffer cells) may generate cytokines, both of which can systemically activate neutrophils and cause neutrophil accumulation in sinusoids and postsinusoidal venules.30 In addition, cytokines can trigger chemokine formation, including chemoattractants for neutrophils, e.g. macrophage inflammatory protein-2 (MIP-2) or keratinocyte derived chemokine (KC).31,32 Neutrophils kill target cells by generating reactive oxygen species, mainly hypochlorous acid, and by the release of proteases.25,33 As a result, highly chemotactic lipid peroxidation products, e.g. lipid aldehydes, are formed, which can amplify the inflammatory response.23,24 Because of the generation of multiple mediators, neutrophil recruitment is a prominent feature of oncotic cell death, a process that can be further accelerated and amplified by the infiltrating leukocytes (Figure 3-1). Although apoptosis is thought not to cause an inflammatory response, more recent evidence suggest that neutrophil extravasation can be triggered by excessive apoptotic cell death during endotoxemia.34–36 Close proximity of extravasated neutrophils to
39
Section I. Pathophysiology of the Liver
apoptotic hepatocytes during alcoholic hepatitis suggests that similar mechanisms may be operative, both in human liver37,38 and in experimental models39,40 of alcohol-induced liver injury. The mechanism of this effect is still unclear. Apoptotic cell death can induce cys-Xcys (CXC) chemokine formation and recruit neutrophils into the liver vasculature.41 However, chemokines are not responsible for neutrophil extravasation during endotoxemia.32 On the other hand, Kupffer cell activation and cytokine formation can be triggered during the phagocytosis of apoptotic bodies, and thereby promote an inflammatory response.42 Another possibility is activation of neutrophils through direct contact with apoptotic hepatocytes. Recognition of phosphatidylserine on the surface of cells undergoing apoptosis is a major stimulus for phagocytes to engulf apoptotic bodies.43 Endothelial cells gaps, which are common features of various insults to the liver,44,45 may facilitate this contact through pseudopods of the neutrophils.46 Together, these emerging data suggest that not only oncotic necrosis but also apoptotic cell death can promote an inflammatory response, and consequently induce neutrophil-mediated oncotic cell death.16,34,35
BIOCHEMICAL MARKERS OF APOPTOSIS AND NECROSIS The increased knowledge of intracellular signaling mechanisms of apoptosis revealed a number of biochemical markers for this mode of cell death (Table 3-1). In contrast to the more cumbersome manual counting of cells with morphological characteristics of apoptosis, most biochemical markers are easier to quantify.3,4 In particular, fluorescence-based enzyme activity measurements of caspases, e.g. caspase-3, are highly sensitive and relatively specific for apoptosis. This can be combined with Western blot analyses of procaspases and their active fragments. In addition, active caspases and the cleavage of caspase substrates, e.g. cytokeratin-18, can be detected by immunohistochemistry. The characteristic caspase-dependent cleavage of other targets, such as poly-(ADP-ribose)-polymerase-1 (PARP-1), can be assessed by Western blot analysis. Moreover, in hepatocytes, the mitochondrial release of cytochrome c and other intermembrane proteins, including endonuclease G, apoptosisinducing factor (AIF), and second mitochondria-derived activator of caspases (Smac)/direct inhibitor of apoptosis-binding protein with low PI (DIABLO), can be evaluated. On the other hand, cytosolic cleavage of Bid and the translocation of truncated Bid (tBid) and other proapoptotic members of the Bcl-2 family (e.g. Bax) to mitochondria can indicate apoptotic cell death. Activation of caspase-3 and other effector caspases leads to degradation of the inhibitor of caspase-dependent DNase (ICAD) and the liberation and nuclear translocation of CAD, which then induces internucleosomal DNA cleavage with its characteristic DNA fragmentation pattern (DNA ladder). The cells should then stain positive for the terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling (TUNEL) assay. In addition, small DNA fragments are present in the cytosol detectable by an antihistone ELISA. On the other hand, increased expression of cell death receptors, e.g. Fas and its ligand, provide only indirect support for potential apoptotic cell death; such parameters need to be interpreted with caution. In contrast, oncotic necrosis is characterized by swelling of the entire liver, individual cells and cell organelles, by massive release of cell
40
contents, and in most cases by nuclear fragmentation (karyolysis) (see Table 3-1). Despite the large number of parameters that can be used to characterize cell injury and death, a major drawback is the fact that, with the exception of morphology, virtually none of the biochemical parameters is specific for apoptosis or oncotic necrosis.3,4 Thus, any characterization of the extent of cell death in a given liver disease has to start with the identification of the mode of cell death, followed by its quantification using one or several of the parameters mentioned. To identify apoptosis unequivocally as the major mode of cell death, there has to be convincing morphological evidence of apoptotic cells together with a consistent increase in a substantial number of biochemical parameters. However, it is most important to recognize that changes in these parameters must correlate quantitatively with the overall postulated number of cells undergoing apoptosis. The use of ‘positive controls’, i.e. models where apoptotic cell death is undisputed, is valuable, allowing correlation of the number of apoptotic cells with the extent of caspase activation, DNA fragmentation, etc. In addition, pan-caspase inhibitors completely eliminate caspase-dependent apoptosis both in vivo and in cultured cells. These compounds are important experimental tools to assess the extent of apoptotic cell death.
APOPTOTIC CELL DEATH PATHWAYS Extensive investigations of apoptotic cell death mechanisms during the last 10–15 years have resulted in a substantially increased insight into apoptotic signaling pathways in many cell types, including hepatocytes. However, there are many examples where the mode of cell death is wrongly assumed to be apoptosis.4 Therefore, the current discussion will focus mainly on Fas receptor- and TNF receptor type 1-mediated signal transduction mechanisms of apoptosis (‘extrinsic pathway’) in experimental models where the mode of cell death is undisputed.
FAS RECEPTOR (CD95)-MEDIATED APOPTOSIS IN HEPATOCYTES Fas receptor-induced apoptosis is thought to be important in viral hepatitis and other liver disease processes.47,48 When Fas ligand (FasL, CD178) located on lymphocytes or an agonistic antibody binds to the Fas receptor it causes trimerization of that receptor, which allows binding of the adapter molecule Fas-associated death domain (FADD) to the death domain (DD) of the cytoplasmic tail of the receptor (Figure 3-2).48 Once FADD is bound, it can recruit through its death effector domain (DED) procaspase-8 or -10 to form the death-inducing signaling complex (DISC) at the receptor.48 The formation of the DISC results in the autocatalytic activation (processing) of procaspase-8 and -10. If sufficient quantities of caspase-8 are formed at the DISC, caspase-8 will process procaspase-3 and other executioner caspases to cause apoptotic cell death (type I cells).49 However, hepatocytes are cells that generate only small amounts of caspase-8 at the receptor, and therefore require amplification of the apoptosis signal through the mitochondrial pathway (type II cells).49 In hepatocytes, caspase-8 cleaves the Bcl2 family member Bid, whose truncated form (tBid) inserts into the outer mitochondrial membrane (Figure 3-2).47 Together with other
Chapter 3 MECHANISMS OF LIVER CELL DESTRUCTION
FAS-L
EndoG
FLIP
Cas 8
FADD FADD
DD
FADD FADD
Cas 10
FADD FADD
Bax
Bax
dATP C CARD
Bid
Casp8
C C
Diablo sIAP
CARD C dATP
EndoG
Casp10
Bcl-2
C
Casp 9 Apaf1 Apaf1 Casp 9
Diablo
C
C C
C
FADD
AIF
C
C
Bax
EndoG
DISC
Diablo
EndoG
C
C
R1
C
C
Bax
AIF
C
C C
C EndoG EndoG
Diablo
C
FAS-L
FAS-L
R1
DD
DD
R1
FADD
R1
DD
R1
DD
FAS-L
cIAP
Casp 3 Casp 6 Casp 7
Caspase 3 substrates
AIF Figure 3-2. Fas receptor-induced signaling pathway of apoptosis. AIF, apoptosis-inducing factor; Apaf1, apoptosis protease-activating factor-1; CARD, caspaseactivating and -recruiting domain; c, cytochrome c; cIAP, cellular inhibitor of apoptosis proteins; DD, death domain; DIABLO, direct inhibitor of apoptosis-binding protein with low pi; DISC, death-inducing signaling complex; EndoG, endonuclease G; FADD, Fas-associated death domain; FLIP, FLICE-inhibitory protein.
Bcl-2 members, such as Bax, Bak and Bad, tBid forms pores in the outer membrane and triggers the release of cytochrome c.47 tBid can also induce the mitochondrial membrane permeability transition (MPT) pore formation in the inner membrane, with mitochondrial swelling and rupture of the outer membrane.50 The MPT can trigger the release of cytochrome c and other proteins from the mitochondrial intermembrane space.50–52 However, the MPT only accelerates cytochrome c release after Fas receptor activation, but is not mandatory.51 Recent data suggest that Bid can induce a mitochondrial oxidant stress, which is the actual mediator that induces MPT.13 However, a significant role for an oxidant stress in receptormediated apoptosis could not be confirmed in vivo.53 Independent of the molecular mechanism of release, once in the cytosol, cytochrome c forms together with apoptosis protease-activating factor-1 (Apaf-1), dATP and procaspase-9 the apoptosome, a complex which induces caspase-9 processing and activation. The amplification loop is completed when caspase-9 processes procaspase-3 and other downstream caspases (Figure 3-2). As procaspase-8 can also be processed and activated by caspase-3 and -6, these downstream caspases can initiate another amplification
cycle through mitochondria. This feedback amplification is an important feature of Fas receptor-mediated apoptosis in hepatocytes and endothelial cells,14,15 leading to the generation of sufficient caspase-3 and -6 activity to complete the apoptotic process. The executioner caspases are then degrading a large number of different intracellular proteins.54 Although the consequences of their proteolytic cleavage remain unclear for most substrates, there are a number of examples where the pathophysiological effects are well studied.54 Among others, caspase-3 cleaves the inhibitor (ICAD) of the caspase-dependent DNase (CAD), which causes the liberation of the active DNase and its translocation to the nucleus.55 CAD is responsible for internucleosomal DNA cleavage, leading to the characteristic DNA ladder.55 Another prominent substrate of caspase-3 is poly (ADP-ribose) polymerase-1 (PARP-1), an enzyme involved in DNA repair. Cleavage of PARP-1 prevents futile attempts of the cells to repair the DNA damage caused by CAD. Because excessive activation of PARP-1 consumes cellular NAD+ and ATP, PARP-1 inactivation ensures that cellular ATP levels are not depleted below the levels necessary to maintain apoptotic signaling. Cleavage of many of the proteins involved in cell adhesion, cytoskeletal and
41
Section I. Pathophysiology of the Liver
structural proteins, and nuclear structural proteins is likely to be responsible for the characteristic morphologic changes of cells undergoing apoptosis, such as shrinkage, membrane blebbing, and chromatin condensation.54
INTERNAL REGULATORS OF FAS-INDUCED APOPTOSIS The signaling mechanism of the Fas receptor is regulated at different intracellular levels to prevent accidental triggering of the apoptosis pathway (Figure 3-2). Cellular FLICE-inhibitory protein (c-FLIP) has structural similarities to procaspase-8 and can compete with the procaspase for binding to FADD.48 Thus, increased c-FLIP expression confers resistance against Fas-induced apoptosis.48,56 However, a splice variant of c-FLIP, c-FLIP(L), was shown to dimerize with procaspase-8 and contribute to the activation of caspase-8.57 In addition to c-FLIP, there is a family of inhibitors of apoptosis proteins (IAPs) that can inhibit active caspases in the cytosol (Figure 3-2).58 In order to transmit the apoptotic signal effectively these inhibitors need to be inactivated by other proteins derived from the intermembrane space of the mitochondria. Smac/DIABLO and HtrA2/Omi are released from the mitochondria through pores formed by tBid, Bax, and other proapoptotic Bcl-2 family members, bind to IAPs in the cytosol, and target
DD
DD
R1
DD RIP
R1
Hepatocellular apoptosis by TNF-a is initiated by binding of the cytokine to the TNF receptor type 1 (Figure 3-3). Ligand binding triggers a clustering of several TNFR1 molecules, which then allows
Figure 3-3. TNF receptor type 1-induced NF-kB and AP-1 activation. AP-1, activating protein-1; ASK1, apoptosis signal-regulating kinase 1; cIAP, cellular inhibitor of apoptosis proteins; DD, death domain; IKK, IkB kinases; IkB, inhibitor of nuclear factor-kB (NF-kB); JNK, c-Jun Nterminal kinase; MKK and MEKK, MAP/ERK kinase kinase; NIK, NF-kB-inducing kinase; R1 receptor type 1; RIP, receptor interacting protein; TRADD, TNF receptor 1-associated death domain; TRAF2, TNF receptor-associated factor 2.
TRADD
R1
TNF RECEPTOR-MEDIATED NF-KB ACTIVATION AND APOPTOSIS
TNF
TRADD
R1
DD
TNF
these inhibitors for degradation.59 Thus, hepatocytes require Bid activation and the mitochondrial pathway to induce apoptosis because the low caspase-8 activities generated at the DISC are not sufficient to overcome the block by IAPs.47 Another regulation point is the increased expression of Bcl-2 itself and other antiapoptotic family members, e.g. Bcl-xL. These antiapoptotic proteins form heterodimers with Bax, Bak, or Bad and prevent pore formation and the release of cytochrome c, Smac/DIABLO and HtrA2/Omi, thereby inhibiting the progression of the apoptotic signal through the mitochondria. Reduced Fas-mediated hepatocellular apoptosis after treatment with cytokines such as hepatocyte growth factor, interleukin-6 (IL-6) and transforming growth factora (TGF-a) is due to increased expression of antiapoptotic Bcl-2 family members such as Bcl-xL.47,56,60 In addition to the modulation by intracellular proteins, Fas receptor-mediated apoptosis can be regulated by an increase in Fas receptor expression, or by shedding of the receptor.61
Complex I
cIAP cIAP TRAF2 TRAF2 IKKb IKKa MEKK3 IKKg NIK IKBa NFKB
NFKB
P65
P52
ASK1
MKK4/7 NFKB1
NFKB
P50
RelB
JNK
NFKB P65/50
42
NFKB P52/RelB
AP-1
Chapter 3 MECHANISMS OF LIVER CELL DESTRUCTION Figure 3-4. TNF receptor type 1-induced apoptosis. CARD, caspase-activating and -recruiting domain; cIAP, cellular inhibitor of apoptosis proteins; DD, death domain; FADD, Fas-associated death domain; R1 receptor type 1; RAID, RIP associated Ich/CED homologous protein with death domain; RIP1, receptor interacting protein 1; TRADD, TNF receptor 1-associated death domain; TRAF2, TNF receptorassociated factor 2.
TNF
TRAF2 RIP1
FADD
CARD
TRAF2 TRADD RIP1
RIP1 TRAF2 TRADD
Modifications of adapter proteins and dissociation from receptor
TRADD
TRAF2 RIP1
Recruitment of FADD and procaspase
RAID
cIAP TRAF2
TRADD
Complex II
FADD
cIAP TRAF2
TRADD
DD RIP
Complex I
TRADD
R1
Apoptosis
the recruitment of adapter molecules to the cytoplasmic tail of the receptor.62,63 However, in contrast to the Fas receptor, the default response of TNFR1 ligation is activation of NF-kB.62,64 This effect is caused by the preferential recruitment of the adapter molecule TNFR1-associated death domain protein (TRADD), which allows binding of additional mediators such as TNF-receptor-interacting protein 1 (RIP1), TNF-receptor associated factor 2 (TRAF2) and c-IAP1 (Figure 3-3).63,65 This complex I does not contain FADD or caspase-8, but triggers the recruitment and activation of the inhibitor of NF-kB kinase (IKK). The IKK complex is responsible for the phosphorylation of IkB-a, which causes its separation from the p65/p50 heterodimer of NF-kB and its degradation by the proteasome pathway. The active NF-kB enters the nucleus, binds to DNA, and induces the transcription of a large number of proinflammatory genes and antiapoptotic proteins.63 Interestingly, once complex I is assembled at the receptor, post-translational modifications of the adapter proteins TRADD and RIP1 occur within 1–2 hours.65 These modifications, which may involve ubiquitination,66 could be responsible for the dissociation of complex I from the receptor (Figure 3-4).65 The dissociated complex again has the death domain of TRADD available for interactions with other death domain-containing proteins such as FADD, which then recruit procaspase-8 or -10 to the complex (complex II) (Figure 3-4). The reason why stimulation of TNFR1 does generally not cause apoptosis is that NF-kB-dependent antiapoptotic genes are up-regulated.
These proteins, which can include c-FLIP, IAPs, A20, A1, Bcl-xL, iNOS, and others,62 are all inhibitors of apoptosis signaling. Consequently, the NF-kB-dependent overexpression of c-FLIP,56 Bcl-xL,67 Bcl-2,68 A20,69 iNOS,70,71 and IAPs72,73 protects against apoptosis induced by TNF receptor family members. Therefore, TNF-a can only induce apoptosis when the formation of these antiapoptosis proteins is inhibited, by preventing either NF-kB activation or the use of blockers of transcription or translation.62 If complex II triggers the activation of caspase-8 and -10, the subsequent signaling pathway in hepatocytes is similar to that of Fas signaling, and includes Bid activation and translocation to the mitochondria, cytochrome c release from mitochondria, apoptosome formation, and activation of caspase-9 and -3. However, based on some differential responses to inhibitors of caspase-8 versus those of caspase3,15 there must be other modulating factors.
MODULATORS OF TNF-INDUCED APOPTOSIS In addition to the intracellular modulators of apoptosis signaling discussed earlier, additional mediators may specifically affect TNF-mediated apoptosis. One of these pathways that can regulate TNF-mediated apoptosis is activation of c-Jun N-terminal kinase (JNK), a mitogen-activating protein kinase (MAPK).74 JNK activation requires TRAF2 on the TNF receptor (Figure 3-3).75 In a rat hepatocyte cell line, inhibition of NF-kB caused a prolonged activation of JNK and AP-1.76 Inhibition of JNK blocked TNF-induced
43
Section I. Pathophysiology of the Liver
mitochondrial cytochrome c release, caspase activation, and cell death, indicating that JNK acted upstream of mitochondria.76 These findings were confirmed in primary rodent hepatocytes.77 Inhibition of JNK activation decreased TNF receptor but not Fas receptorinduced caspase activation, DNA fragmentation, and cell death.77 Although not yet confirmed in hepatocytes, it was proposed that the role of JNK activation in TNF-mediated apoptosis is to induce caspase-8-independent cleavage of Bid to a specific fragment named jBid.78 JBid specifically releases Smac/DIABLO from mitochondria, which interrupts the inhibitory TRAF2-cIAP1 complex and thus enables caspase-8 activation.78 These proapoptotic mechanisms of JNK activation are different from the recently described involvement of JNK in activation of AP-1-mediated cyclin D1 expression and the promotion of proliferation after partial hepatectomy.79 On the other hand, blocking JNK activation in hepatoma cells actually increased TNF-mediated apoptosis.75 The suggested mechanism of JNK-mediated protection in hepatoma cells includes the formation of antiapoptotic proteins that interfere with FADD-dependent signaling of caspase activation and the mitochondrial pathway.75 Thus, TNF-induced JNK activation may be proapoptotic, antiapoptotic or pro-proliferative, depending on the liver cell types involved and the pathophysiological conditions employed. Other mediators known to modulate TNF-mediated apoptosis in the liver are sphingolipids. Intermediates such as ceramides are generated by sphingomyelin degradation through two major enzymes, the neutral sphingomyelinase (N-SMase) and the acidic sphingomyelinase (A-SMase).80 A-SMase can be activated by TNF-receptor ligation, leading to the formation of ceramide and the ganglioside GD3 in hepatocytes.81 Ceramide and GD3 can translocate to the mitochondria and trigger an increased mitochondrial oxidant stress, MPT pore opening and cytochrome c release.80,82 In addition, exogenous GD3 sensitizes hepatocytes to TNF-a-induced apoptosis by inhibiting NF-kB activation and the formation of survival proteins.83 Consistent with the importance of sphingolipid formation, galactosamine/endotoxin-treated A-SMase-deficient mice showed reduced sensitivity to TNF-mediated hepatocellular apoptosis.84 This protection against TNF-induced apoptosis correlated with the absence of DG3 translocation to mitochondria.84 The addition of exogenous A-SMase to hepatocytes isolated from A-SMase-deficient mice restored mitochondrial DG3 translocation and sensitivity to TNF-mediated apoptosis.84 In addition, A-SMase-generated ceramide down-regulates methionine adenosyltransferase 1A (MAT1), which causes depletion of intracellular levels of Sadenosyl methionine and GSH.85 This effect sensitizes the mitochondria to MPT induction by GD3 and enhances susceptibility to TNF-mediated apoptosis.85 Interestingly, N-SMase was still activated by TNF in A-SMase-deficient mice.84 Activation of NSMase and ceramide formation after TNF receptor 1 ligation is mediated through the adapter protein factor associated with NSMase (FAN).86 However, consistent with findings in A-SMasedeficient mice, FAN-deficient mice did not show reduced TNF-mediated apoptosis and liver injury after galactosamine/ endotoxin treatment, despite the fact that intracellular ceramide levels were reduced.86 Galactosamine-sensitized Fan-deficient mice had a lower mortality rate at low doses of TNF-a, which correlated with reduced IL-6 formation.86 Overall, these emerging findings suggest that only ceramide generated in the acidic compartment
44
of the lysosomes may be important for TNF-induced apoptosis in hepatocytes.87 Ligation of the TNF receptor also causes the release of lysosomal proteases such as cathepsin B.88 Inhibition of cathepsin B enzyme activity or deletion of the gene reduced mitochondrial cytochrome c release and protected against TNF-mediated apoptosis, both in cultured hepatocytes and in vivo.88,89 These findings suggest that cathepsin B may act upstream of mitochondria. However, the molecular target of cathepsin B is currently unknown. On the other hand, lysosomal cathepsin B release in response to TNF was inhibited in hepatocytes from Bid-deficient animals, and in hepatoma cells transfected with dominant-negative FADD, or treated with a caspase-8 inhibitor.90 These results support the conclusion that cathepsin B release requires caspase-8/Bid activation. Interestingly, dominant-negative FAN also prevented caspase-8 activation and cathepsin B release, suggesting that FAN activation is critical for cathepsin mobilization.90 However, these data obtained in hepatoma cells are in contrast to results obtained in vivo with FAN-deficient mice.86 Thus, more work is necessary to investigate in more detail the role of cathepsin B in TNF-mediated apoptosis in hepatocytes.
INTRINSIC (MITOCHONDRIAL) PATHWAY OF APOPTOSIS In contrast to the previously discussed ‘extrinsic’ or receptor-mediated pathway of apoptosis, the ‘intrinsic’ or mitochondrial pathway is initiated independent of members of the TNF receptor family, and does not involve DISC formation, caspase-8 activation, or Bid cleavage. On the other hand, events downstream of mitochondria are similar. The key difference is that the cytotoxic stress, e.g. radiation or chemicals, can cause DNA damage and trigger p53 activation.91 The tumor suppressor p53 acts as transcription factor to induce increased expression of proapoptotic Bcl-2 family members such as Bax,92 which then induce the release of cytochrome c, endonuclease G, and AIF.93 Involvement of the intrinsic pathway of apoptotic cell death has been suggested for increased apoptosis in the livers of older animals,94 after prolonged treatment with alcohol,95 or in hepatoma cells after benzo(a)pyrene96 or acetaminophen exposure.97,98 However, as discussed later, similar signaling events can occur during oncotic necrosis. Therefore, it is critical to verify that the cells actually undergo apoptosis by demonstrating caspase activation, cell shrinkage, and nuclear fragmentation, together with the absence of cell contents release and inflammation.
MECHANISMS OF ONCOTIC NECROSIS In the past, it was generally assumed that necrotic cell death was caused by catastrophic events that were incompatible with cell survival. These mechanisms included the formation of reactive metabolites with covalent binding to cellular proteins, excessive protein sulfhydryl oxidation, massive lipid peroxidation with cell membrane failure, and/or intracellular Ca2+ accumulation with mitochondrial dysfunction and activation of phospholipases. However, this simplistic view of oncotic necrosis is challenged by newer findings,
Chapter 3 MECHANISMS OF LIVER CELL DESTRUCTION
GSH
1 P450 NAPQI
Protein arylation
?
2 Protein nitration
NO O2-
?
Bax
AAP
ONOO-
Bax
MPT ATP
Mitochondria Bid
Caspase activation
Figure 3-5. Acetaminophen-induced oncotic necrosis. AAP, acetaminophen; AIF, apoptosis-inducing factor; GSH, glutathione; MPT, mitochondrial membrane permeability transition; NAPQI, N-acetyl benzoquinone imine; PARP, poly-(ADP-ribose)-polymerase1; Smac, second mitochondriaderived activator of caspases.
tBid
Cyt c / smac
AIF
Endonuclease G
ATP
Chromatin condensation
DNAstrandbreaks ONOOH H2O2
DNAfragmentation Cell necrosis
ATP NAD+ depletion
PARP activation
DNA repair Nucleus
Protein arylation and nitration
which demonstrate multiple signaling events that eventually lead to dysfunction of mitochondria and a bioenergetic catastrophe for the cell. These emerging concepts also explain why mechanistically diverse interventions can prevent necrotic cell death after the same insult. Because acetaminophen (APAP) hepatotoxicity is the most frequent cause of drug-induced liver failure in the US and the UK,99 and as APAP overdose is a widely used experimental model for druginduced liver toxicity, newer insights into the pathophysiology of oncotic cell death will be discussed using this example (Figure 3-5).
MITOCHONDRIAL DYSFUNCTION AND APAP-INDUCED CELL DEATH More than 90% of an ingested dose of APAP is enzymatically conjugated with sulfate or glucuronic acid and excreted. The remainder is metabolized by the cytochrome P450 system (mainly Cyp2E1) to form a reactive metabolite, presumably N-acetyl benzoquinone imine (NAPQI).100 NAPQI reacts rapidly with glutathione and causes massive depletion of this sulfhydryl reagent in hepatocytes, and then covalently binds to cellular proteins (Figure 3-5). Although these events are critical for APAP toxicity, covalent modifications of proteins alone cause only partial inactivation of enzymes, and can explain neither the extent of cell death nor the
Survival regeneration
rapid progression to cell necrosis observed after APAP overdose.101 Thus, covalent binding to intracellular proteins is an initiating event that needs to be amplified to cause cell death. APAP and its nonhepatotoxic isomer 3¢-hydroxyacetanilide (AMAP) cause similar overall protein binding of reactive metabolites, but no hepatotoxicity with the latter.100 However, APAP – but not AMAP – binds to mitochondrial proteins102 and causes inhibition of the mitochondrial respiratory chain and hence mitochondrial oxidant stress.101 The oxidant stress occurs immediately after GSH depletion and clearly precedes cell death.22 Although the oxidant stress is critical for cell injury101,103 it does not cause massive lipid peroxidation in vivo.104 In contrast, superoxide derived from the mitochondrial respiratory chain reacts with nitric oxide (NO) to form the potent oxidant and nitrosylating agent peroxynitrite (Figure 3-5).101,105,106 Interestingly, staining for nitrotyrosine protein adducts, a footprint for peroxynitrite formation in the tissue, occurs almost exclusively in mitochondria.107 When animals were treated with GSH 1.5–2.5 hours after the administration of APAP, the accelerated recovery of mitochondrial GSH effectively scavenged peroxynitrite without affecting NAPQI formation or the mitochondrial oxidant stress.108 Because this delayed GSH treatment attenuated cell death and promoted cell cycle activation,108,109 these data support the hypothesis that peroxynitrite is a critical mediator of the injury. Reduced
45
Section I. Pathophysiology of the Liver
injury in mice that lack the inducible nitric oxide synthase (iNOS)110 further supports the importance of peroxynitrite as a cytotoxic mediator, and suggests that iNOS is at least partly responsible for NO formation. The oxidant stress and peroxynitrite formation within the mitochondria is mainly responsible for the opening of the mitochondrial membrane permeability transition pores (MPT), which leads to the breakdown of the mitochondrial membrane potential and necrotic cell death (Figure 3-5).19,111 Blocking of the MPT by ciclosporin A prevents cell death both in vitro19,111 and in vivo.112,113 In addition to the oxidant stress, ferrous iron released from lysosomes appears to be involved in the mechanism of APAPinduced MPT and cell death.114 Another target of the intramitochondrial oxidant stress is mitochondrial DNA (mtDNA), which is progressively depleted after APAP overdose.107 Scavenging peroxynitrite with delayed GSH treatment partially reversed the mtDNA depletion, indicating that in addition to peroxynitrite, hydrogen peroxide or other reactive oxygen species may be involved in mtDNA modifications.107 Although the ramifications of mtDNA damage for APAP-induced cell death are not clear, it can be assumed that this effect may compromise the long-term survival and recovery of hepatocytes. Although there is little doubt that mitochondrial dysfunction, oxidant stress, and eventual collapse of the mitochondrial membrane potential are important events in the mechanism of APAP-induced oncotic necrosis, the signals triggering mitochondrial dysfunction remain less well defined. Based on the observation that only the reactive metabolite of APAP, but not that of AMAP, covalently modifies mitochondrial proteins,102 it can be hypothesized that a direct binding of NAPQI to mitochondria may be at least partly responsible for the inhibition of mitochondrial respiration and the formation of reactive oxygen.100,101 The inhibition of Ca2+ pumps in the plasma membrane through covalent binding causes an increase in cytosolic Ca2+ levels, which then may uncouple mitochondria through futile cycling of Ca2+. More recently, the translocation of Bax and tBid from the cytosol to the mitochondria was described.115–117 These proteins, together with other Bcl-2 family members, can form pores in the outer mitochondrial membrane and facilitate the release of cytochrome c and other intermembrane proteins (Figure 3-5).93,118 Indeed, recent experiments using Bax gene knockout mice showed a substantial delay in APAP-induced liver injury, which was accompanied by a delay in cytochrome c and Smac/DIABLO release into the cytosol, and a strong delay in nuclear DNA fragmentation.117 However, Bax translocation had no effect on mitochondrial oxidant stress and peroxynitrite formation.117 These data suggest that mitochondrial Bax translocation is involved in the early release of intermembrane proteins from the mitochondria during APAP-induced oncotic necrosis. Eliminating Bax – and potentially other Bcl-2 family members – can only delay the cell death process: the oxidant stress-dependent opening of the MPT with swelling of the mitochondria and rupture of the outer membrane will ultimately release the same proteins.
NUCLEAR DNA DAMAGE AND APAP-INDUCED CELL DEATH Nuclear DNA damage indicated by DNA fragmentation (DNA ladder, antihistone ELISA) and TUNEL staining is a well-known
46
phenomenon after APAP overdose (Figure 3-5).101 Because of a lack of caspase activation after APAP treatment,21,115,119 DNA degradation is unlikely to be caused by the caspase-dependent DNase (CAD). This conclusion is supported by the different staining pattern of the TUNEL assay and the different extent of DNA fragmentation, as measured by an antihistone ELISA,21,107,115,120 suggesting that during APAP-induced karyolysis larger DNA fragments are produced than during caspase-mediated apoptosis (see Table 3-1).107 The appearance of predominantly larger DNA fragments in the plasma of APAP-treated mice compared to animals with Fas-mediated apoptosis121 provides direct evidence for this concept. Thus, the mechanism of APAP-induced nuclear DNA damage is clearly different from that observed during receptor-mediated apoptosis. One possible mechanism is that the DNA damage is caused by apoptosis-inducing factor (AIF) and/or endonuclease G, both of which can be released from mitochondria.122,123 Indeed, APAP induces the translocation of cytochrome c and Smac/DIABLO from mitochondria into the cytosol,115–117,120 and of AIF and endonuclease G from mitochondria into the nucleus.117 Although it remains unclear whether AIF, endonuclease G, or any other DNase activity is responsible for nuclear DNA fragmentation, the reduced nuclear DNA damage in Bax gene knockout mice after APAP treatment suggests a connection with mitochondrial dysfunction (Figure 3-5). Because DNA strand breaks lead to activation of PARP-1, which consumes large amounts of NAD+ and ATP,124 it was hypothesized that excessive PARP activation may trigger necrotic cell death by further enhancing the demand on the already compromised cellular energy production.125 However, most PARP activation during APAP overdose occurs after many hepatocytes have already lost viability.126 In addition, PARP-1 gene-deficient mice are not protected from APAP-induced liver injury.126 In contrast, PARP gene knockout mice showed a moderately increased injury, suggesting that PARP activation does not trigger necrotic cell death but may actually be beneficial, by initiating DNA repair and contributing to the recovery of a number of stressed cells (Figure 3-5).126
INTRACELLULAR PROTEASES AND APAP-INDUCED ONCOTIC NECROSIS There is growing evidence that intracellular proteases such as caspases, calpains, and cathepsins play key roles in the mechanisms of cell death under various pathophysiological conditions. Mammalian cysteine proteases are well known proteolytic enzymes with a broad range of functions in the cell. As discussed, proteolytic activation of caspases is critical for apoptosis. However, the reports in the literature are consistent that there is no relevant caspase activation after APAP treatment,21,115,116,119,120,127 and treatment with a pan-caspase inhibitor after APAP exposure did not protect from liver cell injury.119 Cathepsins, including cathepsin B, are lysosomal cysteine proteases that are involved in protein catabolism.128 Cathepsin B has been implicated as a major factor in liver injury after obstructive cholestasis in vivo129 and hepatic ischemia–reperfusion injury.130 Although it was hypothesized by the authors that cathepsin B induces apoptotic cell death,129,130 there is overwhelming evidence that the dominant mode of cell death in these models is oncotic necrosis.3,4,20,131–133 Thus, the activation and release of cathepsin B from lysosomes contributes to oncotic necrotic cell death. In APAP
Chapter 3 MECHANISMS OF LIVER CELL DESTRUCTION
hepatotoxicity, cathepsin B is also released from lysosomes (Jaeschke, unpublished). However, a cathepsin inhibitor was not protective in this model, suggesting that these lysosomal proteases are not involved in APAP-induced cell death.116 Calpains are a family of Ca2+-dependent cytosolic cysteine proteases that include m- and m-calpain as the two major ubiquitous isoenzymes.134 Calpains catalyze only limited proteolysis of proteins, and among other activities are involved in cytoskeletal remodeling, cell cycle regulation, and signal transduction. Calpains have been implicated in the pathophysiology of ischemia–reperfusion injury135 and AAP hepatotoxicity.136 Mehendale and coworkers136 suggested that calpains are responsible for the propagation of tissue injury because they are released by injured cells and may cause damage to neighboring cells. Consistent with the potential activation of Ca2+dependent cytosolic proteases is the early elevation of cytosolic Ca2+ levels after AAP treatment, which is caused by the alkylation and inhibition of the plasma membrane Ca2+-ATPase.100 Dysregulation of cellular Ca2+ homeostasis is linked to APAP-induced cell injury both in vivo and in vitro.100 Interestingly, both calpains and cathepsins can cleave the proapoptotic signaling molecule Bid,137,138 which could explain Bid cleavage after AAP overdose without caspase activation116 and provide a link between covalent binding, increase in cytosolic Ca2+ levels, calpain activation, and mitochondrial dysfunction.
INFLAMMATION AND APAP-INDUCED ONCOTIC NECROSIS Inflammation is a common response to extensive cell death through oncotic necrosis. The main purpose is to remove cell debris and make room for new cells. Unfortunately, under certain conditions the resident liver macrophages (Kupffer cells) and newly recruited macrophages, neutrophils, and T lymphocytes may actually aggravate tissue injury. Activation of hepatic macrophages after APAP treatment has been described.139–141 However, the pathophysiological consequence remains controversial. It was originally postulated that Kupffer cells may be responsible for the oxidant stress and peroxynitrite formation during APAP hepatotoxicity.139,140,142 However, functional inhibition of Kupffer cells by gadolinium chloride143–145 and eliminating the enzyme responsible for superoxide formation by Kupffer cells (NADPH oxidase)146 did not reduce the APAPinduced oxidant stress, and had either no or a very limited protective effect. In contrast, elimination of Kupffer cells actually enhanced APAP-induced liver injury.147 The protection was due to the formation of anti-inflammatory cytokines such as IL-10, which attenuated the induction of iNOS and thus limited intracellular peroxynitrite formation.148 Consistent with this finding, the increased susceptibility of TNF gene knockout mice to APAP was correlated with increased iNOS expression and enhanced peroxynitrite formation.149 TNF-a formation is also a critical factor in enhancing antioxidant defense mechanisms150 and the promotion of regeneration after APAP hepatotoxicity.151 In addition, the later recruitment of activated macrophages by monocyte chemoattractant protein-1 (MCP-1) contributes to the resolution of the APAP-induced tissue injury.152 Thus, Kupffer cells and newly recruited macrophages have a predominantly beneficial effect on the overall injury process after APAP overdose.
Neutrophils are also well known to accumulate in the liver in response to APAP-induced liver cell injury.26 In contrast to other insults,25,28 neutrophils are not systemically activated after APAP overdose, and antibodies against adhesion molecules on neutrophils (anti-CD18 antibodies) are not protective.26 In addition, mice with no functional NADPH oxidase, which is also the main superoxide-generating enzyme in neutrophils, are not protected from APAP injury.146 Although these data do not support the hypothesis that neutrophils actively contribute to APAP-induced cell injury in mice, if the right conditions come together there is a possibility that the accumulating neutrophils can cause additional tissue injury.27 Recently, it was shown that natural killer (NK) cells and NK cells with T-cell receptors (NKT cells) may play a role in APAP-induced hepatotoxicity.153 These resident lymphocytes generate a number of cytokines and chemokines, which are responsible for hepatic neutrophil recruitment.153 On the other hand, the inflammatory mediators trigger up-regulation of Fas ligands, and animals deficient in the Fas receptor (lpr mice) or the Fas ligand (gld mice) have reduced liver injury after APAP treatment.153 Despite the fact that there is no morphological evidence of apoptosis,21 the data are consistent with a previous report showing that down-regulation of the Fas receptor with antisense oligonucleotides attenuated APAP hepatotoxicity.154 On the other hand, subliminal stimulation of the Fas receptor by an agonistic antibody aggravated APAP-induced hepatotoxicity.155 Fas receptor stimulation triggered expression of the inducible nitric oxide synthase, which led to enhanced peroxynitrite formation.155 Thus, the Fas/FasL system may aggravate the APAPinduced liver injury independent of apoptosis by enhancing mitochondrial dysfunction, peroxynitrite formation, and ATP depletion.155 Consistent with these findings, Fas receptor-deficient lpr mice show a reduced inflammatory response and less necrotic cell injury in response to bile duct ligation.132 Inflammation is an inevitable response to excessive cell injury in the liver. Resident and accumulating inflammatory cells are activated to resolve the tissue injury. However, under certain circumstances these cells can aggravate the existing injury. This may become more likely if the animal or person is exposed to a second hit, in particular when this insult triggers a systemic inflammatory response.156 An excessive inflammatory response may explain some of the late progressive injury and liver failure seen in humans after APAP overdose.157
SUMMARY During the past decade substantial progress has been made in the understanding of apoptotic and necrotic signaling mechanisms in hepatocytes. However, there is increasing evidence that both modes of cell death share many common features, especially the central role of mitochondrial dysfunction, membrane permeability transition pore opening, and the release of pro-cell death mediators from the intermembrane space. In addition, nuclear DNA damage occurs in both apoptosis and oncotic necrosis. Because of these commonalities, it is even more important not to indiscriminately call every form of cell death ‘apoptosis’, but to reserve this term for mechanisms where cells satisfy the morphological criteria.
47
Section I. Pathophysiology of the Liver
REFERENCES 1. Rust C, Gores GJ. Apoptosis and liver disease. Am J Med 2000;108:567–574. 2. Rudiger HA, Graf R, Clavien PA. Liver ischemia: apoptosis as a central mechanism of injury. J Invest Surg 2003;16:149–159. 3. Jaeschke H, Lemasters JJ. Apoptosis versus oncotic necrosis in hepatic ischemia/reperfusion injury. Gastroenterology 2003;125:1246–1257. 4. Jaeschke H, Gujral JS, Bajt ML. Apoptosis and necrosis in liver disease. Liver Int 2004;24:85–89. 5. Guicciardi ME, Gores GJ. Cholestatic hepatocellular injury: what do we know and how should we proceed. J Hepatol 2005;42:297–300. 6. Rodriguez-Enriquez S, He L, Lemasters JJ. Role of mitochondrial permeability transition pores in mitochondrial autophagy. Int J Biochem Cell Biol 2004;36:2463–2472. 7. Lemasters JJ, Qian T, He L, et al. Role of mitochondrial inner membrane permeabilization in necrotic cell death, apoptosis, and autophagy. Antioxid Redox Signal 2002;4:769–781. 8. Paxian M, Bauer I, Rensing H, et al. Recovery of hepatocellular ATP and ‘pericentral apoptosis’ after hemorrhage and resuscitation. FASEB J 2003;17:993–1002. 9. Gujral JS, Farhood A, Jaeschke H. Oncotic necrosis and caspasedependent apoptosis during galactosamine-induced liver injury in rats. Toxicol Appl Pharmacol 2003;190:37–46. 10. Sun F, Hamagawa E, Tsutsui C, et al. Evaluation of oxidative stress during apoptosis and necrosis caused by D-galactosamine in rat liver. Biochem Pharmacol 2003;65:101–107. 11. Sun F, Hamagawa E, Tsutsui C, et al. Evaluation of oxidative stress during apoptosis and necrosis caused by carbon tetrachloride in rat liver. Biochim Biophys Acta 2001;1535:186–191. 12. Stachlewitz RF, Seabra V, Bradford B, et al. Glycine and uridine prevent D-galactosamine hepatotoxicity in the rat: role of Kupffer cells. Hepatology 1999;29:737–745. 13. Ding WX, Ni HM, DiFrancesca D, et al. Bid-dependent generation of oxygen radicals promotes death receptor activation-induced apoptosis in murine hepatocytes. Hepatology 2004;40:403–413. 14. Bajt ML, Lawson JA, Vonderfecht SL, et al. Protection against Fas receptor-mediated apoptosis in hepatocytes and nonparenchymal cells by a caspase-8 inhibitor in vivo: evidence for a postmitochondrial processing of caspase-8. Toxicol Sci 2000;58:109–117. 15. Bajt ML, Vonderfecht SL, Jaeschke H. Differential protection with inhibitors of caspase-8 and caspase-3 in murine models of tumor necrosis factor and Fas receptor-mediated hepatocellular apoptosis. Toxicol Appl Pharmacol 2001;175: 243–252. 16. Gujral JS, Hinson JA, Farhood A, Jaeschke H. NADPH oxidasederived oxidant stress is critical for neutrophil cytotoxicity during endotoxemia. Am J Physiol Gastrointest Liver Physiol 2004;287:G243–G252. 17. Ishii H, Adachi M, Fernandez-Checa JC, et al. Role of apoptosis in alcoholic liver injury. Alcohol Clin Exp Res 2003;27:1207–1212. 18. Kim JS, Qian T, Lemasters JJ. Mitochondrial permeability transition in the switch from necrotic to apoptotic cell death in ischemic rat hepatocytes. Gastroenterology 2003;124: 494–503. 19. Kon K, Kim JS, Jaeschke H, Lemasters JJ. Mitochondrial permeability transition in acetaminophen-induced necrosis and apoptosis of cultured mouse hepatocytes. Hepatology 2004;40:1170–1179. 20. Gujral JS, Bucci TJ, Farhood A, Jaeschke H. Mechanism of cell death during warm hepatic ischemia–reperfusion in rats: apoptosis or necrosis? Hepatology 2001;33:397–405.
48
21. Gujral JS, Knight TR, Farhood A, et al. Mode of cell death after acetaminophen overdose in mice: apoptosis or oncotic necrosis? Toxicol Sci 2002;67:322–328. 22. Bajt ML, Knight TR, Lemasters JJ, Jaeschke H. Acetaminopheninduced oxidant stress and cell injury in cultured mouse hepatocytes: protection by N-acetyl cysteine. Toxicol Sci 2004;80:343–349. 23. Jaeschke H, Gores GJ, Cederbaum AI, et al. Mechanisms of hepatotoxicity. Toxicol Sci 2002;65:166–176. 24. Jaeschke H. Reactive oxygen and mechanisms of inflammatory liver injury. J Gastroenterol Hepatol 2000;15: 718–724. 25. Jaeschke H. Molecular mechanisms of hepatic ischemia–reperfusion injury and preconditioning. Am J Physiol Gastrointest Liver Physiol 2003;284:G15–G26. 26. Lawson JA, Farhood A, Hopper RD, et al. The hepatic inflammatory response after acetaminophen overdose: role of neutrophils. Toxicol Sci 2000;54:509–516. 27. Jaeschke H, Cover C, Bajt ML. Inflammation and drug hepatotoxicity: aggravation of injury or clean-up mission? Hepatology 2005;41:1176–1178. 28. Gujral JS, Farhood A, Bajt ML, Jaeschke H. Neutrophils aggravate acute liver injury during obstructive cholestasis in bile duct-ligated mice. Hepatology 2003;38:355–363. 29. Gujral JS, Liu J, Farhood A, et al. Functional importance of ICAM-1 in the mechanism of neutrophil-induced liver injury in bile duct-ligated mice. Am J Physiol Gastrointest Liver Physiol 2004;286:G499–G507. 30. Bajt ML, Farhood A, Jaeschke H. Effects of CXC chemokines on neutrophil activation and sequestration in hepatic vasculature. Am J Physiol Gastrointest Liver Physiol 2001;281: G1188–G1195. 31. Okaya T, Lentsch AB. Cytokine cascades and the hepatic inflammatory response to ischemia and reperfusion. J Invest Surg 2003;16:141–147. 32. Dorman RB, Gujral JS, Bajt ML, et al. Generation and functional significance of CXC chemokines for neutrophilinduced liver injury during endotoxemia. Am J Physiol Gastrointest Liver Physiol 2005;288:G880–G886. 33. Jaeschke H. Role of reactive oxygen species in hepatic ischemia–reperfusion injury and preconditioning. J Invest Surg 2003;16:127–140. 34. Jaeschke H, Fisher MA, Lawson JA, et al. Activation of caspase 3 (CPP32)-like proteases is essential for TNF-alpha-induced hepatic parenchymal cell apoptosis and neutrophil-mediated necrosis in a murine endotoxin shock model. J Immunol 1998;160:3480–3486. 35. Lawson JA, Fisher MA, Simmons CA, et al. Parenchymal cell apoptosis as a signal for sinusoidal sequestration and transendothelial migration of neutrophils in murine models of endotoxin and Fas-antibody-induced liver injury. Hepatology 1998;28:761–767. 36. Jaeschke H. Inflammation in response to hepatocellular apoptosis. Hepatology 2002;35:964–966. 37. Natori S, Rust C, Stadheim LM, et al. Hepatocyte apoptosis is a pathologic feature of human alcoholic hepatitis. J Hepatol 2001;34:248–253. 38. Ziol M, Tepper M, Lohez M, et al. Clinical and biological relevance of hepatocyte apoptosis in alcoholic hepatitis. J Hepatol 2001;34:254–260. 39. Deaciuc IV, D’Souza NB, de Villiers WJ, et al. Inhibition of caspases in vivo protects the rat liver against alcohol-induced sensitization to bacterial lipopolysaccharide. Alcohol Clin Exp Res 2001;25:935–943. 40. Jaeschke H. Neutrophil-mediated tissue injury in alcoholic hepatitis. Alcohol 2002;27:23–27. 41. Faouzi S, Burckhardt BE, Hanson JC, et al. Anti-Fas induces hepatic chemokines and promotes inflammation by an NF-kappa
Chapter 3 MECHANISMS OF LIVER CELL DESTRUCTION
42.
43.
44.
45.
46.
47.
48. 49. 50.
51.
52.
53.
54.
55.
56.
57. 58.
59.
60.
61.
B-independent, caspase-3-dependent pathway. J Biol Chem 2001;276:49077–49082. Canbay A, Feldstein AE, Higuchi H, et al. Kupffer cell engulfment of apoptotic bodies stimulates death ligand and cytokine expression. Hepatology 2003;38:1188–1198. Fadok VA, de Cathelineau A, Daleke DL, et al. Loss of phospholipid asymmetry and surface exposure of phosphatidylserine is required for phagocytosis of apoptotic cells by macrophages and fibroblasts. J Biol Chem 2001;276:1071–1077. McCuskey RS, Bethea NW, Wong J, et al. Ethanol binging exacerbates sinusoidal endothelial and parenchymal injury elicited by acetaminophen. J Hepatol 2005;42:371–377. DeLeve LD, Ito Y, Bethea NW, et al. Embolization by sinusoidal lining cells obstructs the microcirculation in rat sinusoidal obstruction syndrome. Am J Physiol Gastrointest Liver Physiol. 2003;284:G1045–G1052. Lawson JA, Burns AR, Farhood A, et al. Pathophysiologic importance of E- and L-selectin for neutrophil-induced liver injury during endotoxemia in mice. Hepatology 2000;32:990–998. Yin XM, Ding WX. Death receptor activation-induced hepatocyte apoptosis and liver injury. Curr Mol Med 2003;3:491–508. Peter ME, Krammer PH. The CD95 (APO-1/Fas) DISC and beyond. Cell Death Differ 2003;10:26–35. Barnhart BC, Alappat EC, Peter ME. The CD95 type I/type II model. Semin Immunol 2003;15:185–193. Zhao Y, Ding WX, Qian T, et al. Bid activates multiple mitochondrial apoptotic mechanisms in primary hepatocytes after death receptor engagement. Gastroenterology 2003;125:854–867. Hatano E, Bradham CA, Stark A, et al. The mitochondrial permeability transition augments Fas-induced apoptosis in mouse hepatocytes. J Biol Chem 2000;275:11814–11823. Feldmann G, Haouzi D, Moreau A, et al. Opening of the mitochondrial permeability transition pore causes matrix expansion and outer membrane rupture in Fas-mediated hepatic apoptosis in mice. Hepatology 2000;31:674–683. Bajt ML, Ho YS, Vonderfecht SL, Jaeschke H. Reactive oxygen as modulator of TNF and Fas receptor-mediated apoptosis in vivo: studies with glutathione peroxidase-deficient mice. Antioxid Redox Signal 2002;4:733–740. Fischer U, Janicke RU, Schulze-Osthoff K. Many cuts to ruin: a comprehensive update of caspase substrates. Cell Death Differ 2003;10:76–100. Nagata S, Nagase H, Kawane K, et al. Degradation of chromosomal DNA during apoptosis. Cell Death Differ 2003;10:108–116. Kovalovich K, Li W, DeAngelis R, et al. Interleukin-6 protects against Fas-mediated death by establishing a critical level of antiapoptotic hepatic proteins FLIP, Bcl-2, and Bcl-xL. J Biol Chem 2001;276:26605–26613. Boatright KM, Deis C, Denault JB, et al. Activation of caspases8 and -10 by FLIP(L). Biochem J 2004;382:651–657. Liston P, Fong WG, Korneluk RG. The inhibitors of apoptosis: there is more to life than Bcl2. Oncogene 2003;22:8568– 8580. Li S, Zhao Y, He X, et al. Relief of extrinsic pathway inhibition by the Bid-dependent mitochondrial release of Smac in Fasmediated hepatocyte apoptosis. J Biol Chem 2002;277:26912–26920. Kanda D, Takagi H, Toyoda M, et al. Transforming growth factor alpha protects against Fas-mediated liver apoptosis in mice. FEBS Lett 2002;519:11–15. Xia M, Xu C, Ji S, Xue S. Shedding of Fas ectodomain that affects apoptosis of hepatocytes occurring in regenerative liver. J Gastroenterol 2002;37:1042–1047.
62. Ding WX, Yin XM. Dissection of the multiple mechanisms of TNF-alpha-induced apoptosis in liver injury. J Cell Mol Med 2004;8:445–454. 63. Baud V, Karin M. Signal transduction by tumor necrosis factor and its relatives. Trends Cell Biol 2001;11:372–377. 64. Barnhart BC, Peter ME. The TNF receptor 1: a split personality complex. Cell 2003;114:148–150. 65. Micheau O, Tschopp J. Induction of TNF receptor I-mediated apoptosis via two sequential signaling complexes. Cell 2003;114:181–190. 66. Lee JC, Peter ME. Regulation of apoptosis by ubiquitination. Immunol Rev 2003;193:39–47. 67. Zender L, Hutker S, Mundt B, et al. NFkappaB-mediated upregulation of bcl-xl restrains TRAIL-mediated apoptosis in murine viral hepatitis. Hepatology 2005;41:280–288. 68. Van Molle W, Denecker G, Rodriguez I, et al. Activation of caspases in lethal experimental hepatitis and prevention by acute phase proteins. J Immunol 1999;163:5235–5241. 69. Arvelo MB, Cooper JT, Longo C, et al. A20 protects mice from D-galactosamine/lipopoly-saccharide acute toxic lethal hepatitis. Hepatology 2002;35:535–543. 70. Hatano E, Bennett BL, Manning AM, et al. NF-kappaB stimulates inducible nitric oxide synthase to protect mouse hepatocytes from TNF-alpha- and Fas-mediated apoptosis. Gastroenterology 2001;120:1251–1262. 71. Kim YM, Kim TH, Chung HT, et al. Nitric oxide prevents tumor necrosis factor alpha-induced rat hepatocyte apoptosis by the interruption of mitochondrial apoptotic signaling through Snitrosylation of caspase-8. Hepatology 2000;32:770–778. 72. Schoemaker MH, Ros JE, Homan M, et al. Cytokine regulation of pro- and anti-apoptotic genes in rat hepatocytes: NF-kappaBregulated inhibitor of apoptosis protein 2 (cIAP2) prevents apoptosis. J Hepatol 2002;36:742–750. 73. Schoemaker MH, Gommans WM, de la Rosa LC, et al. Resistance of rat hepatocytes against bile acid-induced apoptosis in cholestatic liver injury is due to nuclear factor-kappa B activation. J Hepatol 2003;39:153–161. 74. Czaja MJ. The future of GI and liver research: editorial perspectives. III. JNK/AP-1 regulation of hepatocyte death. Am J Physiol Gastrointest Liver Physiol 2003;284: G875-G879. 75. Liedtke C, Plumpe J, Kubicka S, et al. Jun kinase modulates tumor necrosis factor-dependent apoptosis in liver cells. Hepatology 2002;36:315–325. 76. Liu H, Lo CR, Czaja MJ. NF-kappaB inhibition sensitizes hepatocytes to TNF-induced apoptosis through a sustained activation of JNK and c-Jun. Hepatology 2002;35:772–778. 77. Schwabe RF, Uchinami H, Qian T, et al. Differential requirement for c-Jun NH2-terminal kinase in TNFalpha- and Fas-mediated apoptosis in hepatocytes. FASEB J 2004;18:720–722. 78. Deng Y, Ren X, Yang L, Lin Y, Wu X. A JNK-dependent pathway is required for TNFalpha-induced apoptosis. Cell 2003;115:61–70. 79. Schwabe RF, Bradham CA, Uehara T, et al. c-Jun-N-terminal kinase drives cyclin D1 expression and proliferation during liver regeneration. Hepatology 2003;37:824–832. 80. Pettus BJ, Chalfant CE, Hannun YA. Ceramide in apoptosis: an overview and current perspectives. Biochim Biophys Acta 2002;1585:114–125. 81. Garcia-Ruiz C, Colell A, Morales A, et al. Trafficking of ganglioside GD3 to mitochondria by tumor necrosis factor-alpha. J Biol Chem 2002;277:36443–36448. 82. Garcia-Ruiz C, Colell A, Paris R, Fernandez-Checa JC. Direct interaction of GD3 ganglioside with mitochondria generates reactive oxygen species followed by mitochondrial permeability transition, cytochrome c release, and caspase activation. FASEB J 2000;14:847–858.
49
Section I. Pathophysiology of the Liver
83. Colell A, Garcia-Ruiz C, Roman J, et al. Ganglioside GD3 enhances apoptosis by suppressing the nuclear factor-kappa Bdependent survival pathway. FASEB J 2001;15:1068–1070. 84. Garcia-Ruiz C, Colell A, Mari M, et al. Defective TNF-alphamediated hepatocellular apoptosis and liver damage in acidic sphingomyelinase knockout mice. J Clin Invest 2003;111:197–208. 85. Mari M, Colell A, Morales A, et al. Acidic sphingomyelinase downregulates the liver-specific methionine adenosyltransferase 1A, contributing to tumor necrosis factor-induced lethal hepatitis. J Clin Invest 2004;113:895–904. 86. Malagarie-Cazenave S, Segui B, Leveque S, et al. Role of FAN in tumor necrosis factor-alpha and lipopolysaccharide-induced interleukin-6 secretion and lethality in D-galactosaminesensitized mice. J Biol Chem 2004;279:18648–18655. 87. Morales A, Colell A, Mari M, et al. Glycosphingolipids and mitochondria: role in apoptosis and disease. Glycoconj J 2004;20:579–588. 88. Guicciardi ME, Deussing J, Miyoshi H, et al. Cathepsin B contributes to TNF-alpha-mediated hepatocyte apoptosis by promoting mitochondrial release of cytochrome c. J Clin Invest 2000;106:1127–1137. 89. Guicciardi ME, Miyoshi H, Bronk SF, Gores GJ. Cathepsin B knockout mice are resistant to tumor necrosis factor-alphamediated hepatocyte apoptosis and liver injury: implications for therapeutic applications. Am J Pathol 2001;159:2045–2054. 90. Werneburg N, Guicciardi ME, Yin XM, Gores GJ. TNF-alphamediated lysosomal permeabilization is FAN and caspase 8/Bid dependent. Am J Physiol Gastrointest Liver Physiol 2004;287:G436-G443. 91. Sheikh MS, Fornace AJ Jr. Role of p53 family members in apoptosis. J Cell Physiol 2000;182:171–181. 92. Miyashita T, Reed JC. Tumor suppressor p53 is a direct transcriptional activator of the human bax gene. Cell 1995;80:293–299. 93. Saelens X, Festjens N, Vande Walle L, et al. Toxic proteins released from mitochondria in cell death. Oncogene 2004;23:2861–2874. 94. Zhang Y, Chong E, Herman B. Age-associated increases in the activity of multiple caspases in Fisher 344 rat organs. Exp Gerontol 2002;37:777–789. 95. Deaciuc IV, D’Souza NB, Burikhanov R, et al. Alcohol, but not lipopolysaccharide-induced liver apoptosis involves changes in intracellular compartmentalization of apoptotic regulators. Alcohol Clin Exp Res 2004;28:160–172. 96. Ko CB, Kim SJ, Park C, et al. Benzo(a)pyrene-induced apoptotic death of mouse hepatoma Hepa1c1c7 cells via activation of intrinsic caspase cascade and mitochondrial dysfunction. Toxicology 2004;199:35–46. 97. Boulares AH, Zoltoski AJ, Stoica BA, et al. Acetaminophen induces a caspase-dependent and Bcl-XL sensitive apoptosis in human hepatoma cells and lymphocytes. Pharmacol Toxicol 2002;90:38–50. 98. Kass GE, Macanas-Pirard P, Lee PC, Hinton RH. The role of apoptosis in acetaminophen-induced injury. Ann NY Acad Sci 2003;1010:557–559. 99. Lee WM. Acetaminophen and the US. Acute Liver Failure Study Group: lowering the risks of hepatic failure. Hepatology 2004;40:6–9. 100. Nelson SD, Bruschi SA. Mechanism of acetaminophen-induced liver disease. In: Kaplowitz N, DeLeve LD, eds. Drug-induced liver disease. New York: Marcel-Dekker, 2003: 287–325. 101. Jaeschke H, Knight TR, Bajt ML. The role of oxidant stress and reactive nitrogen species in acetaminophen hepatotoxicity. Toxicol Lett 2003;144:279–288. 102. Qiu Y, Benet LZ, Burlingame AL. Identification of hepatic protein targets of the reactive metabolites of the nonhepatotoxic regioisomer of acetaminophen, 3¢-
50
103.
104.
105. 106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
hydroxyacetanilide, in the mouse in vivo using two-dimensional gel electrophoresis and mass spectrometry. Adv Exp Med Biol 2001;500:663–673. Knight TR, Kurtz A, Bajt ML, et al. Vascular and hepatocellular peroxynitrite formation during acetaminophen toxicity: role of mitochondrial oxidant stress. Toxicol Sci 2001;62:212–220. Knight TR, Fariss MW, Farhood A, Jaeschke H. Role of lipid peroxidation as a mechanism of liver injury after acetaminophen overdose in mice. Toxicol Sci 2003;76:229–236. James LP, Mayeux PR, Hinson JA. Acetaminophen-induced hepatotoxicity. Drug Metab Dispos 2003;31:1499–14506. Laskin JD, Heck DE, Gardner CR, Laskin DL. Prooxidant and antioxidant functions of nitric oxide in liver toxicity. Antioxid Redox Signal 2001;3:261–271. Cover C, Mansouri A, Knight TR, et al. Peroxynitrite-induced mitochondrial and endonuclease-mediated nuclear DNA damage in acetaminophen hepatotoxicity. J Pharmacol Exp Ther 2005; 315:879–887. Knight TR, Ho YS, Farhood A, Jaeschke H. Peroxynitrite is a critical mediator of acetaminophen hepatotoxicity in murine livers: protection by glutathione. J Pharmacol Exp Ther 2002;303:468–475. Bajt ML, Knight TR, Farhood A, Jaeschke H. Scavenging peroxynitrite with glutathione promotes regeneration and enhances survival during acetaminophen-induced liver injury in mice. J Pharmacol Exp Ther 2003;307:67–73. Gardner CR, Laskin JD, Dambach DM, et al. Reduced hepatotoxicity of acetaminophen in mice lacking inducible nitric oxide synthase: potential role of tumor necrosis factor-alpha and interleukin-10. Toxicol Appl Pharmacol 2002;184:27–36. Reid AB, Kurten RC, McCullough SS, et al. Mechanisms of acetaminophen-induced hepatotoxicity: role of oxidative stress and mitochondrial permeability transition in freshly isolated mouse hepatocytes. J Pharmacol Exp Ther 2005;312:509– 516. Haouzi D, Cohen I, Vieira HL, et al. Mitochondrial permeability transition as a novel principle of hepatorenal toxicity in vivo. Apoptosis 2002;7:395–405. Masubuchi Y, Suda C, Horie T. Involvement of mitochondrial permeability transition in acetaminophen-induced liver injury in mice. J Hepatol 2005;42:110–116. Kon K, Kim JS, Jaeschke H, Lemasters JJ. Increase of cytosolic ferrous iron induces the mitochondrial permeability transition in acetaminophen-induced toxicity to mouse hepatocytes [abstract]. Hepatology 2004;40:647A. Adams ML, Pierce RH, Vail ME, et al. Enhanced acetaminophen hepatotoxicity in transgenic mice overexpressing BCL-2. Mol Pharmacol 2001;60:907–915. El-Hassan H, Anwar K, Macanas-Pirard P, et al. Involvement of mitochondria in acetaminophen-induced apoptosis and hepatic injury: roles of cytochrome c, Bax, Bid, and caspases. Toxicol Appl Pharmacol 2003;191:118–129. Bajt ML, Lemasters JJ, Jaeschke H. Role of mitochondrial Bax translocation in acetaminophen-induced hepatic necrosis. Toxicol Sci 2005;84(Suppl 1): 215. Droin NM, Green DR. Role of Bcl-2 family members in immunity and disease. Biochim Biophys Acta 2004;1644:179–188. Lawson JA, Fisher MA, Simmons CA, et al. Inhibition of Fas receptor (CD95)-induced hepatic caspase activation and apoptosis by acetaminophen in mice. Toxicol Appl Pharmacol 1999;156:179–186. Knight TR, Jaeschke H. Acetaminophen-induced inhibition of Fas receptor-mediated liver cell apoptosis: mitochondrial dysfunction versus glutathione depletion. Toxicol Appl Pharmacol 2002;181:133–141. Jahr S, Hentze H, Englisch S, et al. DNA fragments in the blood plasma of cancer patients: quantitations and evidence for
Chapter 3 MECHANISMS OF LIVER CELL DESTRUCTION
122.
123.
124.
125. 126.
127.
128. 129.
130.
131.
132.
133.
134. 135.
136.
137.
138.
139.
140.
their origin from apoptotic and necrotic cells. Cancer Res 2001;61:1659–1665. Cande C, Cecconi F, Dessen P, Kroemer G. Apoptosis-inducing factor (AIF): key to the conserved caspase-independent pathways of cell death? J Cell Sci 2002;115:4727–4734. van Loo G, Schotte P, van Gurp M, et al. Endonuclease G: a mitochondrial protein released in apoptosis and involved in caspase-independent DNA degradation. Cell Death Differ 2001;8:1136–1142. Meyer-Ficca ML, Meyer RG, Jacobson EL, Jacobson MK. Poly(ADP-ribose) polymerases: managing genome stability. Int J Biochem Cell Biol. 2005;37:920–926. Virag L, Szabo C. The therapeutic potential of poly(ADP-ribose) polymerase inhibitors. Pharmacol Rev 2002;54:375–429. Cover C, Fickert P, Knight TR, et al. Pathophysiological role of poly(ADP-ribose) polymerase (PARP) activation during acetaminophen-induced liver cell necrosis in mice. Toxicol Sci 2005;84:201–208. Ferret PJ, Hammoud R, Tulliez M, et al. Detoxification of reactive oxygen species by a nonpeptidyl mimic of superoxide dismutase cures acetaminophen-induced acute liver failure in the mouse. Hepatology 2001;33:1173–1180. Guicciardi ME, Leist M, Gores GJ. Lysosomes in cell death. Oncogene 2004;23:2881–2890. Canbay A, Guicciardi ME, Higuchi H, et al. Cathepsin B inactivation attenuates hepatic injury and fibrosis during cholestasis. J Clin Invest 2003;112:152–159. Baskin-Bey ES, Canbay A, Bronk SF, et al. Cathepsin B inactivation attenuates hepatocyte apoptosis and liver damage in steatotic livers after cold ischemia-warm reperfusion injury. Am J Physiol Gastrointest Liver Physiol. 2005;288:G396– G402. Fickert P, Trauner M, Fuchsbichler A, et al. Oncosis represents the main type of cell death in mouse models of cholestasis. J Hepatol 2005;42:378–385. Gujral JS, Liu J, Farhood A, Jaeschke H. Reduced oncotic necrosis in Fas receptor-deficient C57BL/6J-lpr mice after bile duct ligation. Hepatology 2004;40:998–1007. Selzner M, Rudiger HA, Sindram D, et al. Mechanisms of ischemic injury are different in the steatotic and normal rat liver. Hepatology 2000;32:1280–1288. Goll DE, Thompson VF, Li H, et al. The calpain system. Physiol Rev. 2003;83:731–801. Kohli V, Madden JF, Bentley RC, Clavien PA. Calpain mediates ischemic injury of the liver through modulation of apoptosis and necrosis. Gastroenterology 1999;116:168–178. Limaye PB, Apte UM, Shankar K, et al. Calpain released from dying hepatocytes mediates progression of acute liver injury induced by model hepatotoxicants. Toxicol Appl Pharmacol 2003;191:211–226. Chen M, He H, Zhan S, et al. Bid is cleaved by calpain to an active fragment in vitro and during myocardial ischemia/reperfusion. J Biol Chem 2001;276:30724–30728. Stoka V, Turk B, Schendel SL, et al. Lysosomal protease pathways to apoptosis. Cleavage of bid, not pro-caspases, is the most likely route. J Biol Chem 2001;276:3149–3157. Laskin DL, Laskin JD. Role of macrophages and inflammatory mediators in chemically induced toxicity. Toxicology 2001;160:111–118. Laskin DL, Gardner CR. Nonparenchymal cells, inflammatory macrophages, and hepatotoxicity. In: Kaplowitz N, Deleve LD,
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
eds. Drug-induced liver disease. New York: Marcel Dekker, 2003: 183–211. Jaeschke H. Role of inflammation in the mechanism of acetaminophen-induced hepatotoxity. Expert Opin Drug Metab Toxicol 2005;1:389–397. Michael SL, Pumford NR, Mayeux PR, et al. Pretreatment of mice with macrophage inactivators decreases acetaminophen hepatotoxicity and the formation of reactive oxygen and nitrogen species. Hepatology 1999;30:186–195. Knight TR, Jaeschke H. Peroxynitrite formation and sinusoidal endothelial cell injury during acetaminophen-induced hepatotoxicity in mice. Comp Hepatol 2004;3 (Suppl 1):S46. Ito Y, Abril ER, Bethea NW, McCuskey RS. Ethanol binging enhances hepatic microvascular responses to acetaminophen in mice. Microcirculation 2004;11:625–632. Ito Y, Bethea NW, Abril ER, McCuskey RS. Early hepatic microvascular injury in response to acetaminophen toxicity. Microcirculation 2003;10:391–400. James LP, McCullough SS, Knight TR, et al. Acetaminophen toxicity in mice lacking NADPH oxidase activity: role of peroxynitrite formation and mitochondrial oxidant stress. Free Radic Res 2003;37:1289–1297. Ju C, Reilly TP, Bourdi M, et al. Protective role of Kupffer cells in acetaminophen-induced hepatic injury in mice. Chem Res Toxicol 2002;15:1504–1513. Bourdi M, Masubuchi Y, Reilly TP, et al. Protection against acetaminophen-induced liver injury and lethality by interleukin 10: role of inducible nitric oxide synthase. Hepatology. 2002;35:289–298. Gardner CR, Laskin JD, Dambach DM, et al. Exaggerated hepatotoxicity of acetaminophen in mice lacking tumor necrosis factor receptor-1. Potential role of inflammatory mediators. Toxicol Appl Pharmacol 2003;192:119–130. Chiu H, Gardner CR, Dambach DM, et al. Role of p55 tumor necrosis factor receptor 1 in acetaminophen-induced antioxidant defense. Am J Physiol Gastrointest Liver Physiol 2003;285:G959-G966. Chiu H, Gardner CR, Dambach DM, et al. Role of tumor necrosis factor receptor 1 (p55) in hepatocyte proliferation during acetaminophen-induced toxicity in mice. Toxicol Appl Pharmacol 2003;193:218–227. Dambach DM, Watson LM, Gray KR, et al. Role of CCR2 in macrophage migration into the liver during acetaminopheninduced hepatotoxicity in the mouse. Hepatology 2002;35:1093–1103. Liu ZX, Govindarajan S, Kaplowitz N. Innate immune system plays a critical role in determining the progression and severity of acetaminophen hepatotoxicity. Gastroenterology 2004;127:1760–1774. Zhang H, Cook J, Nickel J, et al. Reduction of liver Fas expression by an antisense oligonucleotide protects mice from fulminant hepatitis. Nature Biotechnol 2000;18:862–867. Tinel M, Berson A, Vadrot N, et al. Subliminal Fas stimulation increases the hepatotoxicity of acetaminophen and bromobenzene in mice. Hepatology 2004;39:655–666. Ganey PE, Luyendyk JP, Maddox JF, Roth RA. Adverse hepatic drug reactions: inflammatory episodes as consequence and contributor. Chem Biol Interact 2004;150:35–51. Kaplowitz N. Acetaminophen hepatotoxicity: what do we know, what don’t we know, and what do we do next? Hepatology 2004;40:23–26.
51
Section I: Pathophysiology of the Liver
4
PHARMACOGENETICS: FOCUS ON THE LIVER Kathleen M. Giacomini and Xin Chen Abbreviations 5-FU 5-fluorouracil 6-MP 6-mercaptopurine ABC ATP binding cassette AUC plasma concentration time curve bp base-pair cSNPs coding SNPs CYP cytochrome P450 CYP2C9 cytochrome P450 2C9 CYP2C19 cytochrome P450 2C19
CYP2D6 DPD EM GST HMG-CoA IM NAT2 NSAIDs
cytochrome P450 2D6 dihydropyrimidine dehydrogenase extensive metabolizers glutathione S-transferase hydroxymethylglutaryl-coenzyme A intermediate metabolizers N-acetyltransferase 2 non-steroidal anti-inflammatory drugs
INTRODUCTION Pharmacogenetics is the study of the genetic basis for variation in drug response.1,2 Pharmacogenomics is a newer term for pharmacogenetics and may also include the use of proteomic or genomic information, such as mRNA expression of multiple genes to explain variations in response to drugs. Before the Human Genome Project the field of pharmacogenetics focused largely on single gene traits. Notably, early pharmacogenetic studies centered on drugs that were eliminated by a single enzyme. Typically in such studies, an adverse drug event was associated with a mutation in a single gene, often an enzyme that metabolized the drug (e.g. thiopurine methyltransferase deficiency). In the wake of the Human Genome Project it is now possible to learn more about the genetic basis for variations in drug response by identifying the common alleles that lead to such variations; extending pharmacogenetic studies from single to multiple gene traits; and performing genotype-to-phenotype studies in which new variants identified in large-scale sequencing projects are studied as covariates of drug response. Modern pharmacogenetic studies often use a pathway approach to identify the genetic basis for variation in response to a drug. That is, a diagram of the drug response pathway is constructed based on information from studies in the literature. Pathways are of two major types: those that depict the pharmacokinetics and those that depict the pharmacodynamics of the drug. Pharmacokinetic pathways show the proteins that function in the absorption, distribution, metabolism and excretion of the drug. Pharmacodynamic pathways show the proteins involved in both therapeutic and adverse drug responses, and include the drug target. Variants in the genes that encode proteins in pharmacokinetic and pharmacodynamic pathways are candidates for variation in drug response and can be tested in association studies with drug response. Pharmacokinetic and pharmacodynamic pathways for many drugs are represented in
OATPs PM SLC SNPs TPMT UGT UGT1A1 UM
organic anion transporting polypeptides poor metabolizers solute carrier single nucleotide polymorphisms thiopurine methyltransferase UDP-glucuronosyltransferases UDP-Glucuronosyltransferase 1 ultrarapid metabolizers
the Pharmacogenetics and Pharmacogenomics Knowledge Base, along with genes and variants in these genes (see http://www. pharmgkb.org). A variation in the DNA sequence that is present at an allele with a frequency of 1% or greater in a population is termed a polymorphism. Single base-pair (bp) substitutions that occur at these frequencies are termed single nucleotide polymorphisms (SNPs) and are the most common form of DNA sequence variation. In general, SNPs are present in the human genome at approximately 1 in every 500 bp, but their frequency in the genome depends upon the gene region. SNPs in the coding regions, termed coding SNPs (cSNPs), are less frequent than SNPs in intergenic regions. cSNPs can be either non-synonymous (or missense), if the base pair substitution results in an amino acid substitution, or synonymous (or sense), if the base pair substitution does not change the amino acid. Often synonymous cSNPs occur in the third or wobble position of a codon, as this position is the most degenerate. Nonsense mutations are base pair substitutions that lead to a stop codon. These are extremely rare. Insertions and deletions, in which one or more base pairs are inserted or deleted, are much less frequent in the genome than are SNPs. A genetic variant may have any frequency and may refer to either an SNP or an insertion or deletion. SNPs and other DNA variations differ in their frequencies in human populations. Selective pressures act to preserve the functional activity of proteins and hence the amino acid sequence. Thus, selective pressures keep non-synonymous SNPs at low frequencies, whereas synonymous SNPs, which do not affect the amino acid sequence of proteins, are often present at higher frequencies. Frequencies of SNPs in human populations may differ. Cosmopolitan SNPs, present in all ethnic groups, are usually found at higher allele frequencies than are population-specific SNPs, which occur in only one ethnic group.3 Geographical isolation of human populations, such that random mating occurred within but not between the isolated populations, led to the presence of ethnic or race-specific
53
Section I. Pathophysiology of the Liver
SNPs. Although such SNPs may reach high frequencies in particular populations because they confer some kind of an advantage, it is likely that they are simply neutral, conferring no advantage or disadvantage to a population. Africans are considered the oldest population and therefore have more SNPs, which have arisen over time.3 A haplotype refers to a group of variants or SNPs that occur together on a single chromosome. These variants are in linkage disequilibrium, that is, they are linked to one another and recombination has not yet occurred to split them apart. Obviously variants that are closer to one another on a chromosome are more likely to be in linkage disequilibrium than variants that are farther apart. Many haplotypes are named as Gene Name *XY, where X is an Arabic number and Y is a letter.4 In general, X is designated 1 if the haplotype represents the common haplotype in a population. Numbers 2, 3, 4, etc. are used to designate other major haplotypes of the gene. Each major haplotype may have additional variations that occur within the context of that haplotype. These minor haplotypes are designated A, B, C, etc. In some cases the ancestral haplotype of a gene is designated *1. The ancestral haplotype is the oldest haplotype of a gene and is often the closest in sequence to the chimpanzee ortholog. Coding region variants of a membrane transporter, MDR1 (encoded by ABCB1), identified in 247 ethnically diverse DNA samples (100 African-Americans, 100 European-Americans, 30 Asians, 10 Mexicans and 7 Pacific Islanders) are shown in Figure 4-1.5,6 Shown are non-synonymous and synonymous SNPs.
Population-specific non-synonymous cSNPs are indicated in the figure. Like most genes, MDR1 has fewer non-synonymous variants than synonymous variants. Also, the allele frequencies of the synonymous variants are greater than those of the non-synonymous variants6 (see also www.pharmgkb.org). This chapter focuses primarily on pharmacogenetics that involve the liver. Genetic variations in enzymes and transporters that are primarily expressed in the liver are described. These proteins play major roles in drug elimination and are therefore critical determinants of systemic drug levels. In addition, the genetic basis for variation in the response to cholesterol-lowering drugs in which the liver is the target organ is discussed.
PHASE I ENZYMES The cytochrome P450 (CYP) system is responsible for the first phase in the metabolism and elimination of many endogenous and exogenous compounds. P450 enzymes convert these substances into electrophilic intermediates, which may then be conjugated by phase II enzymes. As illustrated at http://www.imm.ki.se/cypalleles/, polymorphisms have been identified in all the major CYPs that control drug metabolism. Many of these polymorphisms alter enzymatic activity and therefore have important clinical implications. Here, we will focus on three members of the cytochrome P450 family: CYP2D6, CYP2C9 and CYP2C19.
Extracellular
Cytoplasm
Figure 4-1. Coding region variants of MDR1, termed P-glycoprotein (ABCB1). Secondary structure was rendered using the transmembrane protein display software TOPO (SJ Johns, University of California, San Francisco, and RC Speth, Washington State University, Pullman), transmembrane protein display software available at the University of California, San Francisco Sequence Analysis Consulting Group web site, www.sacs.ucsf.edu/TOPO/topo. html). Coding region variants were identified in 247 ethnically diverse DNA samples (100 African-Americans, 100 European Americans, 30 Asians, 10 Mexicans and 7 Pacific Islanders). Red circles show amino acid variants (or non-synonymous variants); blue circles show synonymous variants; blue areas indicate signature sequences (Walker Domains) in MDR1.
54
Chapter 4 PHARMACOGENETICS: FOCUS ON THE LIVER
CYTOCHROME P450 2D6, CYP2D6 7,8
CYP2D6 is one of the major hepatic cytochrome P450s. The gene has been mapped to chromosome 22q13.1 and contains nine exons. CYP2D6 has an important function in drug metabolism: it is estimated that approximately 19% of all drugs used clinically involve CYP2D6-dependent metabolic pathways for elimination. These include antidepressants, antipsychotics, b-adrenergic blocking agents and antiarrhythmics (Table 4-1). Differences in CYP2D6 activity
Table 4-1. Effect of polymorphisms in enzymes and transporters on drug response Enzyme or transporter
Drugs
Effect of polymorphisms
References
CYP2C9 CYP2C19
Warfarin Mephenytoin Omeprazole Hexobarbital Propranolol Proguanil Phenytoin Debrisoquine Nortriptyline Codeine Sparteine Fluorouracil
Enhanced effect Increased levels Enhanced effect Increased levels Increased levels Increased levels Increased levels Enhanced effect Enhanced effect Reduced effect Enhanced effect Toxic effects
Weinshilboum, 20032 Weinshilboum, 20032 Weber, 199758
NAT2
Isoniazid hydralazine sulfonamides
GSTM1
Several anticancer agents
Neurotoxicity Lupus erythematosus Increased levels Increased susceptibility to cancers
TPMT
Mercaptopurine Azathioprine Irinotecan bilirubin Pravastatin
CYP2D6
DPYD
UGT1A1 SLCO1B1
UM
Increased toxicity of drugs Increased toxicity Increased levels
Weinshilboum, 20032
Milano et al., 199946; Mattison et al., 200247 Weber, 199758; Weinshilboum, 20032
Weber, 199758; van Poppel et al., 199263; Strange et al., 199162 Weinshilboum, 20032; Evans and Relling, 20041 Innocenti et al., 200269 Tirona et al., 200192; Nishizato et al., 200394 Niemi et al., 200495
EM
have profound influences on pharmacokinetics and the clinical responses of individuals to drugs. Based on CYP2D6 activity, people can be classified phenotypically as ultrarapid metabolizers (UM), extensive metabolizers (EM), intermediate metabolizers (IM), or poor metabolizers (PM) (Figure 4-2). To date, more than 75 different allelic variants of CYP2D6 have been identified and characterized (http://www.imm.ki.se/CY Palleles/cyp2d6.htm).9 These polymorphisms can be classified as null alleles, alleles with decreased enzymatic activity, and alleles with increased activity. The relationships between these alleles, metabolizer phenotype and drug response are shown in Figure 4-3. The most common alleles of CYP2D6 are depicted in Table 4-2. The frequency of poor metabolizers (PM) is approximately 7% in European Caucasians, whereas only about 1% of Asians are classified as PMs.10 PMs are caused by null alleles of CYP2D6. Null alleles may be caused by a variety of mutational events: small insertions/deletions that lead to premature termination of the CYP2D6 protein (for example CYP2D6*3 and CYP2D6*4); base pair substitutions (SNPs) that result in full-length non-functional proteins (for example CYP2D6*7); or deletion of the CYP2D6 locus from chromosome 22 (for example CYP2D6*5) (Table 4-2). Proteins encoded by these null alleles are non-functional, with no detectable enzymatic activity. Most PMs can be identified by screening for the CYP2D6*3, CYP2D6*4, CYP2D6*5 and CYP2D6*6 alleles. CYP2D6*4 is the most frequent of these null alleles in Caucasians, occurring with a frequency of about 25%. It accounts for approximately 70–90% of all PMs in Caucasians. CYP2D6*4 contains a GÆA mutation at a splice site, resulting in a spliced mRNA with one additional base and premature termination of the protein reading frame. The CYP2D6*4 allele occurs in approximately 6% of Africans and 1% of Asians. Assuming that the PM phenotype is observed in individuals who are homozygous for CYP2D6*4 alleles, and assuming Hardy–Weinberg equilibrium, these allele frequencies would suggest that only 0.36% (0.06 ¥ 0.06) of Africans and 0.01% of Asians would be PMs. The low allele frequency of CYP2D6*4 in these populations in comparison to its very high frequency in Caucasians
IM
PM
Number of patients
60
Figure 4-2. Hypothetical histogram showing the activity of CYP2D6 in human populations. Individuals are categorized as ultrarapid metabolizers (UM), extensive metabolizers (EM), intermediate metabolizers (IM) or poor metabolizers (PM). MR, metabolism ratio.
30
0 –1
0
2 log (MRs)
55
Section I. Pathophysiology of the Liver
Phenotype
PM
Percentage of population
Genotype
5 – 10
or
Pharmacokinetics
Clinical Response
C
Adverse effects
0 1 Time C
Exaggerated Response IM
10 – 15
0
or
Adverse effects
1 Time C
EM
65 – 80
or
or
Expected response
0 1 Time C
UM
5 – 10
or
No response
0 1 Time
Null allele
Allele with normal activity
Allele with decreased activity
Allele with increased activity
Figure 4-3. Relationships between clinical phenotypes (pharmacokinetics and clinical response) and genotype of CYP2D6 (depicted as the activity of an allele) in human populations.
probably explains the differences in PM phenotypes among the three populations. Intermediate metabolizers (IM) are particularly common among Asians and Africans. IMs may be either heterozygous for one of the null alleles or homozygous for alleles associated with decreased CYP2D6 enzymatic activity.11 The CYP2D6*10 allele is an example of the polymorphisms that are associated with a decreased metabolism of CYP substrates. A C100T mutation in the CYP2D6*10 gene leads to a ProÆSer substitution. In biochemical and genetic studies CYP2D6*10 is associated with decreased CYP2D6 expression levels but some residual enzymatic activity.12,13 In Caucasians, the CYP2D6*10 allele frequency is about 2%. It accounts for approximately 10–20% of Caucasians with the IM phenotype. In Asians, the CYP2D6*10 allele frequency is over 50%.10
56
Ultra rapid metabolizers (UM) often have multiple copies of CYP2D6 alleles. Some individuals in this category have 13 copies of CYP2D6 arranged as tandem repeats; however, a single gene duplication event is more common.14,15 Increased DNA copy number has been shown to lead to significantly increased CYP2D6 expression and increased enzymatic activity.12,16 CYP2D6 gene duplication/amplification occurs with allele frequencies of 1–5% in Europeans, but can occur with much higher frequency in Africans.9 Recent studies have linked several other CYP2D6 polymorphisms with the UM phenotype. For example, CYP2D6*35 has a G31A base pair substitution in exon 1, resulting in a Val11Met mutation.17 However, the molecular mechanisms by which such polymorphisms produce the UM phenotype remain unclear.
Chapter 4 PHARMACOGENETICS: FOCUS ON THE LIVER
CYTOCHROME P450 2C9, CYP2C9 CYP2C9 is an important hepatic metabolic enzyme. It constitutes approximately 20% of the total P450 content in human hepatic microsomes. CYP2C9, with nine exons, is located at chromosome 10q24. Approximately 16% of all clinically used drugs are eliminated by CYP2C9. These drugs include anticoagulants (for example warfarin), hypoglycemics (for example glipizide) and some nonsteroidal anti-inflammatory drugs (NSAIDs) (Table 4-1).18 Thirteen polymorphisms of CYP2C9 have been described to date (http://www.imm.ki.se/CYPalleles/cyp2c9.htm). Among these, CYP2C9*2 and CYP2C9*3 are the most common (Table 4-2). CYP2C9*2 is generated by a CÆT mutation in exon 3, resulting in an Arg144Cys substitution in the protein. CYP2C9*2 has not been found in Asian populations but shows allele frequencies of 3% in Africans, and 8–19% in Caucasians.19–21 CYP2C9*3 results from an AÆC substitution in exon 7, resulting in an Ile359Leu amino acid substitution. CYP2C9*3 is present at an allele frequency of 1–3% in Asians, 1% in Africans, and 3–16% in Caucasians. Other CYP2C9 variants occur at much lower allele frequencies.19–21 For example, CYP2C9*5, a CÆT mutation in exon 7 which results in an Asp360Glu amino acid substitution, is only found in 1% of DNA samples from African-Americans and Hispanic-Americans. Both in vivo and in vitro studies demonstrate that allelic variants of the CYP2C9 gene generally exhibit a reduced enzymatic activity
compared with the reference allele. For example, it was shown that both CYP2C9*2 and CYP2C9*3 lead to reduced catalytic activity, with increased Kms and/or decreased Vmaxs, reducing the apparent intrinsic clearance of substrates (Vmax/Km).22–24 CYP2C9*3 in particular appears to confer the largest reduction in CYP2C9 enzymatic activity. It has been shown that the clearance of the anticoagulant warfarin was reduced by 66% in people who are heterozygous, and by 90% in those who are homozygous for CYP2C9*3 alleles. Patients carrying CYP2C9*2 or CYP2C9*3 alleles require lower maintenance doses of warfarin, take more time to achieve stable dosing, and have a higher risk of bleeding than patients who are homozygous for the reference allele.25–28 It is therefore suggested that for drugs with narrow therapeutic indices such as warfarin, genotyping CYP2C9 prior to drug initiation may improve safety. However, the clinical utility of the pharmacogenetic strategy has yet to be established.20 Although the study of CYP2C9 polymorphisms has increased our understanding of some differences in patients’ drug responses, much still needs to be understood. For example, it has been found that Asian patients require smaller average maintenance doses of warfarin than do Caucasian patients,29,30 and the adverse effects of CYP2C9 substrate drugs such as warfarin and glipizide appear to be more common in Africans than in Caucasians.31,32 However, the
Table 4-2. Molecular Effect of Selected Alleles of Phase I Drug Matabolizing Enzymes Gene
Allele
Major nucleotide changes
Effect
Enzyme activity
CYP2D6 (Zanger et al., 200412)
CYP2D6*1 CYP2D6*2 CYP2D6*3 CYP2D6*4 CYP2D6*5 CYP2D6*6 CYP2D6*7 CYP2D6*10 CYP2D6*35
Wild type C2850T; G4180C A2549del G1846A CYP2D6 locus deleted from chromsome 22 T1707del A2935C C100T G31A; G1661C; C2850T; G4180C
R296C; S486T Frameshift and protein truncation Splicing defect and protein truncataion CTP2D6 deleted Frameshift and protein truncation H324P P34S V11M; R296C; S486T
normal normal None None None None None decreased activity increased activity
CYP2C9*1 CYP2C9*2 CYP2C9*3 CYP2C9*5
Wild type C430T A1075C C1080G
R144C I359L D360E
normal decreased activity decreased activity decreased activity
CYP2C19*1 CYP2C19*2 CYP2C19*3 CYP2C19*4 CYP2C19*5 CYP2C19*6 CYP2C19*7 CYP2C19*8
Wild type G681A G636A A1G C1297T G395A IVS5 + T2A T358C
Splicing defect and protein truncataion Stop codon and protein truncation GTG initiation codon R433W R132Q Splicing defect W120R
normal None None None None None None None
DPYD*1 DPYD*2A DPYD*5A DPYD*6 DPYD*8 DPYD*9A DPYD*10 DPYD*13
Wild type Intron 14 G1A A1627G G2194A C703T T85C G2983T T1679G
Exon 14 skipping I543V V732I R235W C29R V995F I560S
normal decreased activity normal, decreased activity normal, decreased activity decreased activity normal, decreased activity normal, decreased activity decreased activity
CYP2C9 (Lee et al., 200221)
CYP2C19 (Klotz et al., 200435)
DPYD (Mattison et al., 200247)
57
Section I. Pathophysiology of the Liver
common variants of CYP2C9, namely CYP2C9*2 and CYP2C9*3, are both relatively uncommon in Asian and Africans compared to Caucasians. The data therefore imply that there may be other unknown functionally significant polymorphisms in CYP2C9. These polymorphisms may, for example, occur in introns, or in promoter/enhancer regions that may regulate CYP2C9 expression or activity. It is also possible that other genes in the pathways involved in the anticoagulation effects of warfarin may be polymorphic, and may explain the differences in response to this drug.
CYTOCHROME P450 2C19, CYP2C19 Like CYP2C9, CYP2C19 is located at chromosome 10q24. The CYP2C19 gene encodes 490 amino acids and only differs from CYP2C9 by 43 amino acids (8%). However, CYP2C9 and CYP2C19 have completely different enzymatic activities. CYP2C19 encodes the enzyme S-mephenytoin hydroxylase, which was first identified as deficient in some individuals who are poor metabolizers of the anticonvulsant mephenytoin.33,34 CYP2C19 metabolizes diverse and pharmacologically important drugs, including proton pump inhibitors (for example omeprazole),35 antidepressants (for example imipramine),36 and some barbiturates37 (Table 4-1). To date, at least 16 variants of CYP2C19 have been identified (http://www.imm.ki.se/CYPalleles/cyp2c19.htm). Among them, alleles numbered CYP2C19*2 to CYP2C19*8 have been extensively studied (Table 4-2). At least seven variants have been found to be null alleles of coding regions that result in no enzymatic activity. For example, CYP2C19*2 is caused by a GÆA mutation in exon 5 that creates an aberrant splice site. This change alters the reading frame of the mRNA, starting with amino acid 215, and produces a premature stop codon 20 amino acids downstream. The result is a truncated, non-functional CYP2C19 protein. CYP2C19*3 is caused by GÆA mutation in exon 4 which results in a stop codon. On the basis of their ability to metabolize substrate probes of CYP2C19, individuals can be classified as extensive metabolizers (EM), intermediate metabolizers (IM) and poor metablizers (PM). The frequency of CYP2C19 PMs differs among ethnic groups, being 18–23% in Asians and 2–5% in Caucasians.38,39 It was demonstrated that EMs are homozygous for the wild-type (or reference) allele, IMs are heterozygous for one reference allele and one mutant allele, and PMs are homozygous for mutant alleles. CYP2C19 genotypes have important clinical implications. For example, studies have shown that the CYP2C19 genotype may affect the cure rates for Helicobacter pylori infection in ulcer patients. It was found that the oral clearance (defined as the oral dose/area under the plasma concentration time curve) of CYP2C19 substrates, including diazepam and omeprazole, was 1.5 to two times higher in Caucasians than in Chinese.40–42 In one study of Japanese patients who were treated with omeprazole and amoxicillin, the rate of eradication of H. pylori and cure of gastric and duodenal ulcers was 100% in CYP2C19 homozygotes for mutant alleles, 60% in heterozygotes having one mutant allele, and only 29% in patients who are homozygous for reference alleles.43,44 In another study focusing on pantoprazole it was found that CYP2C19 PMs have a statistically significant greater area under the plasma concentration time curve (AUC) and a longer half and lower apparent oral clearance for pantoprazole than do CYP2C19 EMs.45 Therefore, it has been proposed that genotyping tests for variants of CYP2C19
58
in patients may be useful for the optimal prescription of CYP2C19 substrate drugs.
DIHYDROPYRIMIDINE DEHYDROGENASE (DPD) Dihydropyrimidine dehydrogenase (DPD) is involved in the metabolism of pyrimidine nucleosides and nucleoside analogs. It is particularly important in the inactivation of the anticancer drug 5-fluorouracil (5-FU).46–49 Molecular studies indicate that DPD is a homodimeric enzyme consisting of 1025 amino acids. The protein is encoded by DYPD, which contains 23 exons and is found on chromosome 1p22.50 DPD levels in the liver are subject to circadian rhythms, with high levels in the early morning and low levels in the early afternoon. Although it is an older drug, 5-FU remains one of the major chemotherapeutic agents in the treatment of various cancers, including colorectal and breast cancers. Toxicities associated with 5-FU therapy are frequent, with more than 30% of patients experiencing grade 3 and 4 toxicities to the drug. The major toxicity is myelosuppression. The importance of DPD to the pharmacogenetics of 5-FU was noted in 1985, when Tuchman et al. determined that family members of a patient with toxicity to 5-FU exhibited excess pyrimidines in their urine and blood.51 These studies, together with later studies from the Diasio laboratory of the families of patients with 5-FU toxicity and pyrimidinemia, demonstrated that the trait was heritable and related to deficiencies in DPD activity.47 The relationship between mutations in the gene DYPD and reduced activity of the protein DPD was determined in 1995, when Meinsma et al. found that a patient with DPD deficiency had an allelic variant of DPYD that was present in other family members.52 This variant, now designated as DPYD*2A, has been found in other studies of 5FU toxicity.47,48 The major pathway of degradation or inactivation of 5-FU is catalyzed by three enzymes sequentially: DPD, dihydropyrimidinase, and b-alanine synthase (Figure 4-4). DPD catalyzes the first and rate-limiting step in the inactivation of 5-FU. Given that this threestep pathway catalyzes 85% of 5-FU elimination, and that DPD catalysis is the rate-limiting step, reduced activity of DPD can result in profoundly increased levels of 5-FU and cytotoxicity (Figure 4-4). Conflicting results demonstrating a relationship between the activity of DPD in peripheral mononuclear cells and the blood levels or clearance of 5-FU have been obtained. Because some patients with partial DPD activity in their peripheral mononuclear cells have suffered lethal events following 5-FU therapy, a threshold for DPD activity of 70% in mononuclear cells has been proposed.47 However, there is controversy about this level, because DPD activity in peripheral mononuclear cells may not be an appropriate surrogate indicator for activity in the liver, the tissue in which 5-FU is inactivated. The pharmacogenomics of DPD and 5-FU in tumor cells has been evaluated. In brief, studies have shown that the lower the activity of DPD, the greater the sensitivity of the cancer cell lines to 5-FU.47 Further, a relationship between mRNA levels, 5-FU activity, DPD protein level and immunohistochemical scores supports the use of DPD as a predictor of tumor sensitivity to 5-FU. Some studies suggest that measuring DPD mRNA levels along with mRNA levels
Chapter 4 PHARMACOGENETICS: FOCUS ON THE LIVER
Catabolism FBAL
FUPA
Anabolism FUH2
DPD
5 -F U
5-FU Nucleotides
85%
Cytotoxicity
1 – 3%
5 – 20% Urine A
Catabolism
FBAL
FUPA
Anabolism
FUH2
DPD
5 -F U
5-FU Nucleotides
Cytotoxicity
Urine B Figure 4-4. Pathway of 5-fluorouracil activation and inactivation. The enzyme dihydropyrimidine dehydrogenase (DPD) catalyzes the first and rate-limiting step in the inactivation of 5-fluorouracil. Panel A shows the normal catabolism and anabolism (activation) pathways and Panel B shows the shift in the pathway that is taken by 5-fluorouracil when DPD is inactive. (Adapted from Mattison et al., 2002, with permission.47)
of other enzymes in the activation/inactivation pathways of 5-FU (thymidylate phosphorylase and thymidylate synthase) may provide better predictive measures of sensitivity. A number of genetic polymorphisms of DPD have been identified.47,48 Table 4-2 lists some of the common reduced-function allelic variants of DPYD. Many of these variants have been associated with 5-FU toxicity. However, it is important to note that the functional activities of many of the allelic variants of DPYD remain controversial, and that the population frequencies of several of the alleles are not known.
PHASE II ENZYMES THIOPURINE METHYLTRANSFERASE (TPMT) Thiopurine S-methyltransferase (TPMT) catalyzes the transfer of a methyl group to endogenous thiopurines, as well as to 6mercaptopurine (6-MP) and azathioprine, two synthetic thiopurine analogs used in the treatments of cancer and autoimmune diseases such as Crohn’s disease.2,53,54 TPMT is located on chromosome 6p22.3 and consists of 10 exons. The gene is 34 kb long and encodes a 245 amino acid enzyme that is expressed in many tissues. TPMT activity can be readily measured in red cells. It has been shown that the Caucasian population exhibits a trimodal distribution of TPMT activity. Individuals with low TPMT activity are homozygous for reduced functional TPMT alleles. In the Caucasian population, the most common reduced function allele (TPMT*3A) contains two single nucleotide polymorphisms resulting in two
amino acid changes, Ala154Thr and Tyr240Cys (Table 4-2). TPMT*3A is found at an allele frequency of 4% in the Caucasian population, but it does not appear to be present in individuals from China, Japan or Korea. The most common mutant allele in Africans is TPMT*3C, which contains only a single amino acid change, Tyr240Cys. Detailed molecular studies of TPMT have shown that the mechanism responsible for the reduced function of the TPMT*3A allele is enhanced TPMT protein degradation. That is, the variant TPMT protein is likely to be rapidly degraded in the cell. 6-MP can be used in combination with other drugs in the treatment and cure of childhood leukemia.1 Its major side effect is acute myelosuppression, which can be fatal or result in prolonged hospital stays and significant morbidity. 6-MP is metabolized by TPMT to an inactive 6-methylmercaptopurine (Figure 4-5). However, a competing pathway may activate it to form 6-thioguanine nucleotides, which inhibit DNA replication. In the presence of the reduced-function alleles of TPMT, the rate of TPMT inactivation is reduced. As a result, the drug is shunted through the activation pathway and the concentrations of 6-thioguanine nucleotides increase. Myelosuppression becomes a major problem as a result of the increased levels of 6-thioguanine nucleotides. Children with TPMT deficiency can still be treated with 6-MP, but doses should be adjusted downward. It is recommended that individuals who are homozygous for the deficiency alleles of TPMT (e.g. TPMT*3A) should receive one-tenth the normal dose. The dose to give heterozygotes is controversial, although some studies recommend half the normal dose.
59
Section I. Pathophysiology of the Liver
S
CH3
S
N
N
CH NH
N
HN
CH NH
TPMT H2N
H2N N 6-methylmercaptopurine INACTIVE
N
ACTIVE 6-thioguanine nucleotides
6-mercaptopurine
A
S H
N
CH3
S
N CH NH
N
HN
CH NH
TPMT H2N
H2N N 6-methylmercaptopurine INACTIVE
N
6-mercaptopurine
ACTIVE 6-thioguanine nucleotides ACTIVE 6-thioguanine nucleotides ACTIVE 6-thioguanine nucleotides ACTIVE 6-thioguanine nucleotides
Figure 4-5. Competing pathways of activation and inactivation of 6-mercaptopurine. TPMT (thiopurine methyltransferase) catalyzes the transfer of a methyl group to 6-mercaptopurine, resulting in the formation of the inactive 6methylmercapatopurine. This inactivation pathway competes with the activation pathway to 6-thioguanine nucleotides (Panel A). Defects in TPMT result in the drug being shunted through the activation pathway (Panel B). High levels of thioguanine nucleotides lead to serious adverse drug events, in particular life-threatening myelosuppression.
B
Rapid Acetylators
Slow Acetylators
25
Figure 4-6. Frequency distribution of plasma levels of isoniazid obtained 6 hours after administering the same dose of isoniazid to 267 individuals (see Evans et al., 196055).
Frequency
20
15
10
5
0 0
3
6
9
12
Plasma isoniazid concentration ( μg/ml)
Genetic tests for TPMT are available, but their use remains controversial. Opponents suggest that the cost of identifying the 1:300 individuals who is homozygous for deficiency alleles are too high. Proponents argue that the life-threatening severity of the adverse reactions should be prevented with a simple genetic test.
N-ACETYLTRANSFERASE 2 (NAT2) Bimodal and trimodal frequency distributions of acetylation activity were demonstrated in early work using the antituberculosis drug isoniazid as a probe (Figure 4-6).55 This work led to several insights into drug and other xenobiotic toxicities, including the toxicities of isoniazid, procainamide, hydralazine and sulfonamides. For example,
60
isoniazid is associated with severe peripheral neuropathy, which is variable among patients. It was discovered that individuals who were slow metabolizers (or acetylators) of isoniazid were more prone to the drug-induced peripheral neuropathy than individuals who metabolized the drug more rapidly.56,57 The isoniazid studies led to studies of twins in which acetylation status was found to be highly heritable and autosomal.58 Pedigree analysis suggested that the slow acetylator trait was recessive. In studies of various ethnic groups, a high proportion (40–50%) of Caucasians and African-Americans have been shown to be slow acetylators, whereas a much lower proportion (10–20%) of Japanese and Canadian Eskimos are slow acetylators. Egyptians and certain Jewish populations have a very high proportion of slow acetylators (more
Chapter 4 PHARMACOGENETICS: FOCUS ON THE LIVER
than 80%). The trend is for populations in northern climates to have a much lower frequency of slow acetylators. In 1990 Blum et al. identified two chromosomal loci for N-acetyltransferase activity at 8p22.59 These are NAT1 and NAT2, each of which contains a single exon that encodes a protein of approximately 33 kDa. NAT1 and NAT2 are about 87% homologous in amino acid sequence. NAT1 appears to be less variable in human populations than NAT2. NAT2 has been identified as coding for the enzyme that is responsible for the heterogeneity of the acetylation of isoniazid and sulfamethazine. To date, seven non-synonymous variants have been identified in NAT2.56,57 These are G191A (Arg64Gln), T341C (Ile114Thr), A434C (Gln145Pro), G590A (Arg197Gln), A803G (Lys268Arg), A845C (Lys282Thr), and G857A (Gly286Glu). These variants occur in the context of various haplotypes (Table 42). In general, homozygous reference alleles of NAT2 are considered rapid acetylators, whereas heterozygous individuals with one reference and one non-synonymous variant allele have intermediate activity. Compound heterozygotes are individuals who carry two different alleles, each of which contains one of the non-synonymous variants described above. Slow acetylation activity is found in individuals who are homozygotes or compound heterozygotes for the haplotypes that contain the non-synonymous polymorphisms. In general, the adverse effects of drugs that are eliminated by acetylation via NAT2 are thought to be greater in individuals who are slow acetylators. These individuals are expected to exhibit higher levels of the parent compound after therapeutic doses than individuals who are rapid or intermediate acetylators. Parent drug levels are particularly elevated for drugs for which acetylation constitutes a primary route of elimination. For example, isoniazid is eliminated almost entirely by acetylation via NAT2. Therefore, individuals who are slow acetylators will eliminate the drug slowly and will be more susceptible to adverse reactions to isoniazid. These adverse effects include peripheral neuropathy and drug-induced hepatotoxicity. Several drugs that are eliminated by acetylation produce druginduced systemic lupus erythematosus. These include the antileprosy drug dapsone, the antiarrhythmia drug procainamide, the vasodilator hydralazine, and isoniazid. Some studies have found an association of drug-induced systemic lupus erythematosus with a slow acetylator phenotype.58 However, this has not been uniformly found.
GLUTATHIONE S-TRANSFERASES The enzymes of the glutathione S-transferase (GST) family catalyze the transfer of glutathione to a variety of xenobiotics, including many chemical carcinogens and environmental toxins.60 Based on their amino acid similarities, mammalian cytosolic GSTs are divided into six classes: a, m, k, q, p and s.58 In this chapter we will focus on GST1, also termed GSTM1, found in the m class of GSTs. This gene is located in a cluster of five GST genes of the m class (GSTM1, GSTM2, GSTM3, GSTM4 and GSTM5) on the short arm of chromosome 1. It encodes one of the two most active GSTs found in the liver. GSTM1 is also found in other tissues, including kidney, adrenal gland and stomach. GSTM1 has a null allele, which is present at high frequencies in human populations.61 This allele represents a 15 kb deletion that appears to have originated from a homologous crossover of GST genes on chromosome 1. The null
allele is present at frequencies of approximately 50% in many populations. For example, in Caucasians the frequency of GSTM1 null alleles is 40–45%. Because of the high frequency null allele, GSTM1 has been the subject of many association studies.62–64 The null allele may be a susceptibility allele for cancer because it may result in an individual’s inability to inactivate chemical carcinogens. For example, in a group of patients with colorectal cancer it was found that 56% had GSTM1 null alleles, compared to 41% of controls. GSTM1 null alleles have also been associated with risk for lung cancer. Further studies have suggested that aromatic-DNA adduct levels are higher in individuals who are GSTM1 null. Other studies have suggested that aplastic anemia, which may be due to exposure to drugs, carcinogens or environmental toxins, is associated with a significantly higher frequency of GSTM1 deletions in patients compared to healthy controls.65 A similar association has been observed for individuals with the GSTT1 deletion, a highfrequency GST deletion in the GST q family.
UDP-GLUCURONOSYLTRANSFERASE 1, UGT1A1 UDP-glucuronosyltransferases (UGT) catalyze the addition of a glycosyl group from a nucleotide sugar to a small hydrophobic molecule, i.e. glucuronidation. Glucuronidation represents a major pathway that enhances the elimination of many lipophilic xenobiotics and endobiotics to more water-soluble compounds. UGT1A1 (UDP-glycosyltransferase 1 family, polypeptide A1) is located at chromosome 2q13. It is expressed in the liver, bile ducts, stomach and colon. UGT1A1 is the major hepatic UGT. To date, 33 variant UGT1A1 alleles have been identified.66–69 Some of these have been shown to be responsible for both type I and type II Crigler–Najjar syndromes, and for the more common mild hyperbilirubinemia known as Gilbert syndrome.70–72 One of the common variants, UTG1A1*28, has a TA insertion in the (TA)6 element of the UGT1A1 promoter region (Figure 4-7). Because transcriptional activity decreases as the number of TA repeats increases, UGT1A1*28 has reduced expression and activity compared to the reference (TA)6 allele.71,73 This promoter variant is much more common in Caucasians and African-Americans than in Asians. Irinotecan (CPT-11) is an antitumor agent that inhibits topoisomerase-1 activity. Its potent activity has been demonstrated against many types of solid tumor, including gastrointestinal and pulmonary cancers.74,75 For example, the addition of irinotecan to fluorouracil and leucovorin has been shown to improve survival in patients with advanced colorectal cancer.76 However, irinotecan has significant side effects, including severe diarrhea, neutropenia, and a vascular syndrome. The drug toxicity has been linked with a high mortality rate in patients receiving the combination of irinotecan with bolus fluorouracil and leucovorin during the first 60 days of therapy.77,78 Irinotecan is a prodrug and must be converted to SN-38 (7-ethyl10-hydroxycamptothecin), which has 100–1000-fold higher antitumor activity (Figure 4-7). SN-38 is glucuronidated by hepatic UGTs into the inactive metabolite SN-38G (Figure 4-7). Diarrhea, a major dose-limiting toxicity of irinotecan, is believed to be secondary to the biliary excretion of SN-38, the extent of which is determined by SN-38 glucuronidation.
61
Section I. Pathophysiology of the Liver
Irinotecan
CYP3A4/5
Oxidated metabolites
Carboxylesterase 2
SN-38
UGT1A1
SN-38G
A
UGT1A1*1 ((TA)6) (Normal function)
TATATATATATA
UGT1A1*28 ((TA)n) TATATATATATATA (Decreased function) B Figure 4-7. A Pathway of irinotecan activation and inactivation. UGT1A1 is involved in the glucuronidation of the active metabolite of irinotecan, SN38. B Gene structure of UGT1A1 (upper) as well as promoter region variant, UGT1A1*28 (lower).
The hepatic isoform UTG1A1 is primarily responsible for the glucuronidation and detoxification of SN-38 to its inactive SN38G.79 The common polymorphism UGT1A1*28 has been associated with a significant decrease of SN-38 glucuronidation.80 Thus, patients who have the UGT1A1*28 allele will have a reduced detoxification capacity, leading to a higher risk of irinotecan side effects. For example, in a study of 20 patients having solid tumors that were treated with irinotecan, the patients who were heterozygous or homozygous for UGT1A1*28 were much more likely to experience severe grades of neutropenia and diarrhea. These patients also showed significantly lower SN-38 glucuronidation rates than those with the six TA repeat allele.81 These studies have suggested that it would be possible to avoid the severe side effects of irinotecan in patients who have UGT1A1*28 polymorphisms by varying the dose based on genetic testing. Irinotecan therefore represents an excellent candidate for individualized therapy.
MEMBRANE TRANSPORTERS Figure 4-8 depicts the membrane transporters that are involved in drug disposition in the liver.82–84 These transporters are from two major superfamilies, the solute carrier (SLC) superfamily and the ATP binding cassette (ABC) superfamily.85–87 SLC transporters in the liver are influx proteins that mediate the uptake of drugs and other xenobiotics into the hepatocyte. SLC transporters facilitate the movement of their substrates across the lipid bilayer down their electrochemical gradient. Sometimes these transporters may be sec-
62
ondarily active. In such cases, the movement of substrates against their electrochemical gradient is coupled to the movement of a solute (usually a sodium or proton) down its electrochemical gradient. For example, the Na+ bile acid transporter couples the transport of bile acids into the hepatocyte, against their concentration gradient, with the movement of Na+ down its concentration gradient into the hepatocyte. The concentration gradient of Na+ is established by Na+/K+ ATPase. Transporters in two major SLC families are involved in the uptake of xenobiotics in the liver.88,89 These are SLC22, which includes the organic cation transporter OCT1 (SLC22A1) and the organic anion transporter OAT2 (SLC22A7), and SLCO. The SLCO family encodes the organic anion transporting polypeptides (OATPs). OCT1 transporters play key roles in the hepatic uptake of structurally diverse organic cations.90,91 OAT2 and OATP transporters play key roles in the hepatic uptake of diverse anions. After uptake into the hepatocyte, the drug and/or its metabolites are removed from the liver by efflux transporters in the ABC superfamily. In contrast to the facilitated transporters in the SLC superfamily, transporters in the ABC superfamily are active and can move their substrates against a concentration gradient. ABC transporters rely on the hydrolysis of ATP to actively pump their substrates from the hepatocyte to the blood (e.g. MRP1) or bile (e.g. MRP2) (Figure 4-8). By controlling the influx and efflux of drugs and metabolites in the hepatocyte, transporters in the liver ultimately control the rates of drug metabolism, which depend on the intracellular concentrations of the drugs. Further, transporters may directly control the rate of elimination for drugs that are eliminated unchanged in the bile. Thus, hepatic transporters play a key role in drug disposition. It follows that genetic variants in these transporters are important determinants of variation in drug disposition, levels and response. A large pharmacogenetics research project has identified a number of variants in SLC and ABC transporters (see www.pharmgkb.org and www.pharmacogenetics.ucsf.edu) in ethnically diverse human populations.5 For OCT1 and OATP1B1, the functional activities of the variants have been tested in cell-based studies.92,93 Some of the variants have been shown to exhibit altered functional activity (see Figure 4-9 for OCT1). Variants with altered functions may cause variations in drug disposition or response. In fact, several reducedfunction variants of OATP1B1 have been associated clinically with higher blood levels of pravastatin, an HMG-CoA reductase inhibitor used to reduce blood cholesterol levels.94,95 The higher blood levels are presumably due to a reduced hepatic elimination in individuals with OATP1B1 variants.
HEPATIC PHARMACODYNAMICS OF HMG-COA REDUCTASE Hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductase is the rate-limiting enzyme in cholesterol biosynthesis. Located at chromosome 5q13-14 with 20 exons, the gene encodes an approximately 97 kDa protein with 888 amino acids. Statins are competitive inhibitors of HMG-CoA reductase that induce the expression of LDL receptors in the liver. The expression of LDL receptors increases the catabolism of plasma LDL and
Chapter 4 PHARMACOGENETICS: FOCUS ON THE LIVER
MRP1 (ABCC1)
MRP3 (ABCC3)
MRP4 (ABCC4)
OATP1B3 (OATP8)
(A MDR BC 1 B1 )
Hepatocyte
R3 ) MD CB4 B (A
OAT1P1B1 (OATP2/OATP-C)
OATP2B1 (OATP2/OATP-C)
Figure 4-8. Current model of hepatic transport of xenobiotics. Drugs and other xenobiotics move from blood into the hepatocyte via uptake transporters on the basolateral membrane, OCT1 (SLC22A1), OAT2 (SLC22A?) and various members of the SLCO family. In the hepatocyte, drugs or their metabolites move into bile or back into the blood primarily by transporters in the ABC superfamily.
Bile Canaliculus
MRP2 (ABCC2)
RP ) BC CG2 B A (
OCT1 Sinusoidal Membrane NTCP
(A BSE BC P B1 1)
OAT2
Canalicular Membrane
MPP+ Uptake (% of reference OCT1) ni n O jec C te T1 d R S1 ef 4 R6 F 1C L8 F1 5F 6 S1 0L 89 G L 22 P3 0V 4 R3 1L 24 G H 40 M 1S M 408 42 V 0d M el 44 V4 0I G 61I 4 R4 65R 88 M
Blood
200
100
U
0
Figure 4-9. Functional activity of the reference organic cation transporter, OCT1, and all of the protein altering variants that were identified in ethnically diverse populations in a study by Shu et al., 200393. OCT1 and its variants, constructed by site-directed mutagenesis were expressed in oocytes and the uptake of the model substrate, MPP+ (N-methylpyridinium) was studied. Four of the variants (indicated by the arrows) exhibit significantly reduced function and one variant exhibits a gain in function.
lowers the plasma concentration of cholesterol.96 Over the past several decades, large clinical trials have shown the beneficial effects of statins in preventing coronary artery disease.97,98 However, there is considerable interindividual variation in response to statin therapy. Recent studies have demonstrated that the genetic basis for response to statin therapy is complicated. Several polymorphisms may be involved: for example, the e4 allelle in apoE; the G-455A SNP in b-fibrinogen; C514T polymorphism in hepatic lipase; and the Asp9Asn mutation in lipoprotein lipase.99–101
HMG-CoA reductase, which is the main target for statins, has been investigated for its association with statin responsiveness in a recent study by Chasman et al.102 of cholesterol reduction and the effects of statin therapy among 1536 individuals. Two common and tightly linked SNPs in the gene encoding HMG-CoA reductase were found to be significantly associated with the reduced efficacy of pravastatin therapy in lowering cholesterol levels. The SNPs, designated SNP12 and SNP29, are located in introns 5 and 15, respectively. The haplotype they define has been designated haplotype 7. Compared with individuals homozygous for the major allele of the SNPs, individuals with a single copy of the minor allele had a 22% smaller reduction in total cholesterol. The result therefore suggests that genetic polymorphisms of HMG-CoA reductase may contribute to the clinical variations in statin responsiveness.
CONCLUSIONS Significant progress using pharmacogenetics has been made in understanding why certain groups of patients appear to respond to drugs or be more susceptible to adverse drug reactions. This has been especially true of drugs whose toxicities are related to the concentration of a toxic product of a known metabolic pathway. To date pharmacogenetics has been less helpful in defining patients at risk for idiosyncratic drug reactions that occur at a very low frequency. Hopefully, as we gain more information about polymorphisms in genes that may contribute to these rare drug reactions, screening tests will be developed to prevent the administration of these types of drug to this susceptible population, or chemical modifications to the drug will prevent it from entering the toxic pathway.
63
Section I. Pathophysiology of the Liver
REFERENCES 1. Evans WE, Relling MV. Moving towards individualized medicine with pharmacogenomics. Nature 2004;429:464–468. 2. Weinshilboum R. Inheritance and drug response. N Engl J Med 2003;348:529–537. 3. Stephens JC, Schneider JA, Tanguay DA, et al. Haplotype variation and linkage disequilibrium in 313 human genes. Science 2001;293:489–493. 4. Nebert DW, Wain HM. Update on human genome completion and annotations: gene nomenclature. Hum Genomics 2003;1:66–71. 5. Leabman MK, Huang CC, DeYoung J, et al. Natural variation in human membrane transporter genes reveals evolutionary and functional constraints. Proc Natl Acad Sci USA 2003;100:5896–5901. 6. Kroetz DL, Pauli-Magnus C, Hodges LM, et al. Sequence diversity and haplotype structure in the human ABCB1 (MDR1, multidrug resistance transporter) gene. Pharmacogenetics 2003;13:481–494. 7. Wolf CR, Smith G. Cytochrome P450 CYP2D6. IARC Sci Publ 1999;148;209–29. 8. Vermeulen NP. Prediction of drug metabolism: the case of cytochrome P450 2D6. Curr Top Med Chem 2003;3:1227–1239. 9. Zanger UM, Raimundo S, Eichelbaum M. Cytochrome P450 2D6: overview and update on pharmacology, genetics, biochemistry. Naunyn Schmiedebergs Arch Pharmacol 2004;369:23–37. 10. Ozawa S, Soyama A, Saeki M, et al. Ethnic differences in genetic polymorphisms of CYP2D6, CYP2C19, CYP3As and MDR1/ABCB1. Drug Metab Pharmacokinet 2004;19: 83–95. 11. Raimundo S, Fischer J, Eichelbaum M, et al. Elucidation of the genetic basis of the common ‘intermediate metabolizer’ phenotype for drug oxidation by CYP2D6. Pharmacogenetics 2000;10:577–581. 12. Zanger UM, Fischer J, Raimundo S, et al. Comprehensive analysis of the genetic factors determining expression and function of hepatic CYP2D6. Pharmacogenetics 2001;11:573–585. 13. Yu A, Kneller BM, Rettie AE, Haining RL. Expression, purification, biochemical characterization, and comparative function of human cytochrome P450 2D6.1, 2D6.2, 2D6.10, and 2D6.17 allelic isoforms. J Pharmacol Exp Ther 2002;303:1291–1300. 14. Bertilsson L, Dahl ML, Sjoqvist F, et al. Molecular basis for rational megaprescribing in ultrarapid hydroxylators of debrisoquine. Lancet 1993;341:63. 15. Lundqvist E, Johansson I, Ingelman-Sundberg M. Genetic mechanisms for duplication and multiduplication of the human CYP2D6 gene and methods for detection of duplicated CYP2D6 genes. Gene 1999;226:327–338. 16. Dalen P, Dahl ML, Ruiz ML, Nordin J, Bertilsson L. 10Hydroxylation of nortriptyline in white persons with 0, 1, 2, 3, and 13 functional CYP2D6 genes. Clin Pharmacol Ther 1998;63:444–452. 17. Lovlie R, Daly AK, Matre GE, Molven A, Steen VM. Polymorphisms in CYP2D6 duplication-negative individuals with the ultrarapid metabolizer phenotype: a role for the CYP2D6*35 allele in ultrarapid metabolism? Pharmacogenetics 2001;11:45–55. 18. Miners JO, Birkett DJ. Cytochrome P4502C9: an enzyme of major importance in human drug metabolism. Br J Clin Pharmacol 1998;45:525–538. 19. Xie HG, Prasad HC, Kim RB, Stein CM. CYP2C9 allelic variants: ethnic distribution and functional significance. Adv Drug Del Rev 2002;54:1257–1270.
64
20. Takahashi H, Echizen H. Pharmacogenetics of CYP2C9 and interindividual variability in anticoagulant response to warfarin. Pharmacogenomics J 2003;3:202–214. 21. Lee CR, Goldstein JA, Pieper JA. Cytochrome P450 2C9 polymorphisms: a comprehensive review of the in-vitro and human data. Pharmacogenetics 2002;12:251–263. 22. Rettie AE, Wienkers LC, Gonzalez FJ, et al. Impaired (S)warfarin metabolism catalysed by the R144C allelic variant of CYP2C9. Pharmacogenetics 1994;4:39–42. 23. Haining RL, Hunter AP, Veronese ME, et al. Allelic variants of human cytochrome P450 2C9: baculovirus-mediated expression, purification, structural characterization, substrate stereoselectivity, and prochiral selectivity of the wild-type and I359L mutant forms. Arch Biochem Biophys 1996;333:447–458. 24. Takanashi K, Tainaka H, Kobayashi K, et al. CYP2C9 Ile359 and Leu359 variants: enzyme kinetic study with seven substrates. Pharmacogenetics 2000;10:95–104. 25. Steward DJ, Haining RL, Henne KR, et al. Genetic association between sensitivity to warfarin and expression of CYP2C9*3. Pharmacogenetics 1997;7:361–367. 26. Takahashi H, Kashima T, Nomoto S, et al. Comparisons between in-vitro and in-vivo metabolism of (S)-warfarin: catalytic activities of cDNA-expressed CYP2C9, its Leu359 variant and their mixture versus unbound clearance in patients with the corresponding CYP2C9 genotypes. Pharmacogenetics 1998;8:365–373. 27. Ogg MS, Brennan P, Meade T, Humphries SE. CYP2C9*3 allelic variant and bleeding complications. Lancet 1999;354:1124. 28. Scordo MG, Pengo V, Spina E, et al. Influence of CYP2C9 and CYP2C19 genetic polymorphisms on warfarin maintenance dose and metabolic clearance. Clin Pharmacol Ther 2002;72:702–710. 29. Yu HC, Chan TY, Critchley JA, Woo KS. Factors determining the maintenance dose of warfarin in Chinese patients. Q J Med 1996;89:127–135. 30. Takahashi H, Wilkinson GR, Caraco Y, et al. Population differences in S-warfarin metabolism between CYP2C9 genotype-matched Caucasian and Japanese patients. Clin Pharmacol Ther 2003;73:253–263. 31. Shorr RI, Ray WA, Daugherty JR, Griffin MR. Individual sulfonylureas and serious hypoglycemia in older people. J Am Geriatr Soc 1996;44:751–755. 32. White RH, Beyth RJ, Zhou H, Romano PS. Major bleeding after hospitalization for deep-venous thrombosis. Am J Med 1999;107:414–424. 33. Wrighton SA, Stevens JC, Becker GW, VandenBranden M. Isolation and characterization of human liver cytochrome P450 2C19: correlation between 2C19 and S-mephenytoin 4¢hydroxylation. Arch Biochem Biophys 1993;306:240–245. 34. Goldstein JA, Faletto MB, Romkes-Sparks M, et al. Evidence that CYP2C19 is the major (S)-mephenytoin 4¢-hydroxylase in humans. Biochemistry 1994;33:1743–1752. 35. Klotz U, Schwab M, Treiber G. CYP2C19 polymorphism and proton pump inhibitors. Basic Clin Pharmacol Toxicol 2004;95:2–8. 36. Kirchheiner J, Nickchen K, Bauer M, et al. Pharmacogenetics of antidepressants and antipsychotics: the contribution of allelic variations to the phenotype of drug response. Mol Psychiatry 2004;9:442–473. 37. Goldstein JA. Clinical relevance of genetic polymorphisms in the human CYP2C subfamily. Br J Clin Pharmacol 2001;52:349–355. 38. Nakamura K, Goto F, Ray WA, et al. Interethnic differences in genetic polymorphism of debrisoquin and mephenytoin hydroxylation between Japanese and Caucasian populations. Clin Pharmacol Ther 1985;38:402–408.
Chapter 4 PHARMACOGENETICS: FOCUS ON THE LIVER
39. Ward SA, Goto F, Nakamura K, et al. S-mephenytoin 4hydroxylase is inherited as an autosomal-recessive trait in Japanese families. Clin Pharmacol Ther 1987;42:96–99. 40. Zhang YA, Reviriego J, Lou YQ, et al. Diazepam metabolism in native Chinese poor and extensive hydroxylators of Smephenytoin: interethnic differences in comparison with white subjects. Clin Pharmacol Ther 1990;48:496–502. 41. Caraco Y, Tateishi T, Wood AJ. Interethnic difference in omeprazole’s inhibition of diazepam metabolism. Clin Pharmacol Ther 1995;58:62–72. 42. Caraco Y, Lagerstrom PO, Wood AJ. Ethnic and genetic determinants of omeprazole disposition and effect. Clin Pharmacol Ther 1996;60:157–167. 43. Furuta T, Ohashi K, Kamata T, et al. Effect of genetic differences in omeprazole metabolism on cure rates for Helicobacter pylori infection and peptic ulcer. Ann Intern Med 1998;129:1027–1030. 44. Furuta T, Ohashi K, Kosuge K, et al. CYP2C19 genotype status and effect of omeprazole on intragastric pH in humans. Clin Pharmacol Ther 1999;65:552–561. 45. Tanaka M, Ohkubo T, Otani K, et al. Metabolic disposition of pantoprazole, a proton pump inhibitor, in relation to Smephenytoin 4¢-hydroxylation phenotype and genotype. Clin Pharmacol Ther 1997;62:619–628. 46. Milano G, Etienne MC, Pierrefite V, et al. Dihydropyrimidine dehydrogenase deficiency and fluorouracil-related toxicity. Br J Cancer 1999;79:627–630. 47. Mattison LK, Soong R, Diasio RB. Implications of dihydropyrimidine dehydrogenase on 5-fluorouracil pharmacogenetics and pharmacogenomics. Pharmacogenomics 2002;3:485–492. 48. van Kuilenburg AB, De Abreu RA, van Gennip AH. Pharmacogenetic and clinical aspects of dihydropyrimidine dehydrogenase deficiency. Ann Clin Biochem 2003;40:41–45. 49. van Kuilenburg AB, Meinsma R, Zonnenberg BA, et al. Dihydropyrimidinase deficiency and severe 5-fluorouracil toxicity. Clin Cancer Res 2003;9:4363–4367. 50. Wei X, Elizondo G, Sapone A, et al. Characterization of the human dihydropyrimidine dehydrogenase gene. Genomics 1998;51:391–400. 51. Tuchman M, Stoeckeler JS, Kiang DT, et al. Familial pyrimidinemia and pyrimidinuria associated with severe fluorouracil toxicity. N Engl J Med 1985;313:245–249. 52. Meinsma R, Fernandez-Salguero P, Van Kuilenburg AB, et al. Human polymorphism in drug metabolism: mutation in the dihydropyrimidine dehydrogenase gene results in exon skipping and thymine uracilurea. DNA Cell Biol 1995;14:1–6. 53. Weinshilboum R. Thiopurine pharmacogenetics: clinical and molecular studies of thiopurine methyltransferase. Drug Metab Dispos 2001;29:601–605. 54. Weinshilboum R, Wang L. Pharmacogenomics: bench to bedside. Nature Rev Drug Discov 2004;3:739–748. 55. Evans DA, Manley KA, McKusick VA. Genetic control of isoniazid metabolism in man. Br Med J 1960;5197:485–491. 56. Hein DW, Doll MA, Fretland AJ, et al. Molecular genetics and epidemiology of the NAT1 and NAT2 acetylation polymorphisms. Cancer Epidemiol Biomarkers Prev 2000;9:29–42. 57. Hein DW. Molecular genetics and function of NAT1 and NAT2: role in aromatic amine metabolism and carcinogenesis. Mutat Res 2002;506–507:65–77. 58. Weber WW. Pharmacogenetics. New York: Oxford University Press, Inc., 1997. 59. Blum M, Grant DM, McBride W, et al. Human arylamine Nacetyltransferase genes: isolation, chromosomal localization, and functional expression. DNA Cell Biol 1990;9:193–203. 60. Mannervik B. The isoenzymes of glutathione transferase. Adv Enzymol Relat Areas Mol Biol 1985;57:357–417.
61. Board PG. Biochemical genetics of glutathione-S-transferase in man. Am J Hum Genet 1981;33:36–43. 62. Strange RC, Matharoo B, Faulder GC, et al. The human glutathione S-transferases: a case–control study of the incidence of the GST1 0 phenotype in patients with adenocarcinoma. Carcinogenesis 1991;12:25–28. 63. van Poppel G, de Vogel N, van Balderen PJ, Kok FJ. Increased cytogenetic damage in smokers deficient in glutathione Stransferase isozyme mu. Carcinogenesis 1992;13:303–305. 64. Zhong S, Wyllie AH, Barnes D, et al. Relationship between the GSTM1 genetic polymorphism and susceptibility to bladder, breast and colon cancer. Carcinogenesis 1993;14:1821–1824. 65. Lee KA, Kim SH, Woo HY, et al. Increased frequencies of glutathione S-transferase (GSTM1 and GSTT1) gene deletions in Korean patients with acquired aplastic anemia. Blood 2001;98:3483–3485. 66. Tukey RH, Strassburg CP. Human UDP-glucuronosyltransferases: metabolism, expression, and disease. Annu Rev Pharmacol Toxicol 2000;40:581–616. 67. Tukey RH, Strassburg CP, Mackenzie PI. Pharmacogenomics of human UDP-glucuronosyltransferases and irinotecan toxicity. Mol Pharmacol 2002;62:446–450. 68. Miners JO, McKinnon RA, Mackenzie PI. Genetic polymorphisms of UDP-glucuronosyltransferases and their functional significance. Toxicology 2002;181–182:453–456. 69. Innocenti F, Grimsley C, Das S, et al. Haplotype structure of the UDP-glucuronosyltransferase 1A1 promoter in different ethnic groups. Pharmacogenetics 2002;12:725–733. 70. Francoual J, Rivierre A, Mokrani C, et al. Crigler–Najjar syndrome type I in Tunisia may be associated with a founder effect related to the Q357R mutation within the UGT1 gene. Hum Mutat 2002;19:570–571. 71. Beutler E, Gelbart T, Demina A. Racial variability in the UDPglucuronosyltransferase 1 (UGT1A1) promoter: a balanced polymorphism for regulation of bilirubin metabolism? Proc Natl Acad Sci USA 1998;95:8170–8174. 72. Kadakol A, Ghosh SS, Sappal BS, et al. Genetic lesions of bilirubin uridine-diphosphoglucuronate glucuronosyltransferase (UGT1A1) causing Crigler–Najjar and Gilbert syndromes: correlation of genotype to phenotype. Hum Mutat 2000;16:297–306. 73. Bosma PJ, Chowdhury JR, Bakker C, et al. The genetic basis of the reduced expression of bilirubin UDPglucuronosyltransferase 1 in Gilbert’s syndrome. N Engl J Med 1995;333:1171–1175. 74. O’Connell MJ. Current status of adjuvant therapy for colorectal cancer. Oncology (Huntingt) 2004;18:751–755; discussion 755–758. 75. Cho LC, Choy H. Topoisomerase I inhibitors in the combinedmodality therapy of lung cancer. Oncology (Huntingt) 2004;18: 29–39. 76. Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000;343:905–914. 77. Rothenberg ML, Meropol NJ, Poplin EA, et al. Mortality associated with irinotecan plus bolus fluorouracil/leucovorin: summary findings of an independent panel. J Clin Oncol 2001;19:3801–3807. 78. Sargent DJ, Niedzwiecki D, O’Connell MJ, Schilsky RL. Recommendation for caution with irinotecan, fluorouracil, and leucovorin for colorectal cancer. N Engl J Med 2001;345:144–145; author reply 146. 79. Park DJ, Stoehlmacher J, Lenz HJ. Tailoring chemotherapy in advanced colorectal cancer. Curr Opin Pharmacol 2003;3:378– 385. 80. Iyer L, Hall D, Das S, et al. Phenotype–genotype correlation of in vitro SN-38 (active metabolite of irinotecan) and bilirubin
65
Section I. Pathophysiology of the Liver
81.
82. 83. 84.
85.
86. 87. 88.
89.
90.
91.
92.
66
glucuronidation in human liver tissue with UGT1A1 promoter polymorphism. Clin Pharmacol Ther 1999;65:576–582. Iyer L, Das S, Janisch L, et al. UGT1A1*28 polymorphism as a determinant of irinotecan disposition and toxicity. Pharmacogenomics J 2002;2:43–47. Faber KN, Muller M, Jansen PL. Drug transport proteins in the liver. Adv Drug Del Rev 2003;55:107–124. Koepsell H, Endou H. The SLC22 drug transporter family. Pflugers Arch 2004;447:666–676. Zamek-Gliszczynski M, Brouwer K. In vitro models for estimating hepatobiliary clearance. In: Borchardt R, Kerns E, Lipinski C, et al., eds. Biotechnology: Pharmaceutical aspects. Vol. 1. Pharmaceutical profiling in drug discovery for lead selection. Arlington: AAPS Press, 2004:259–292. Hediger MA, Romero MF, Peng JB, et al. The ABCs of solute carriers: physiological, pathological and therapeutic implications of human membrane transport proteins. Introduction. Pflugers Arch 2004;447:465–468. Gottesman MM, Pastan I, Ambudkar SV. P-glycoprotein and multidrug resistance. Curr Opin Genet Dev 1996;6:610–617. Holland IB, Cole SPC, Kuchler K, Higgins CF. ABC proteins: from bacteria to man. London: Academic Press, 2003. Hirano M, Maeda K, Shitara Y, Sugiyama Y. Contribution of OATP2 (OATP1B1) and OATP8 (OATP1B3) to the hepatic uptake of pitavastatin in humans. J Pharmacol Exp Ther 2004;311:139–146. Hagenbuch B, Scharschmidt BF, Meier PJ. Effect of antisense oligonucleotides on the expression of hepatocellular bile acid and organic anion uptake systems in Xenopus laevis oocytes. Biochem J 1996;316:901–904. Koepsell H, Schmitt BM, Gorboulev V. Organic cation transporters. Rev Physiol Biochem Pharmacol 2003;150: 36–90. Zhang L, Brett CM, Giacomini KM. Role of organic cation transporters in drug absorption and elimination. Annu Rev Pharmacol Toxicol 1998;38:431–460. Tirona RG, Leake BF, Merino G, Kim RB. Polymorphisms in OATP-C: identification of multiple allelic variants associated
93.
94.
95.
96. 97.
98.
99.
100.
101.
102.
with altered transport activity among European- and AfricanAmericans. J Biol Chem 2001;276:35669–35675. Shu Y, Leabman MK, Feng B, et al. Evolutionary conservation predicts function of variants of the human organic cation transporter, OCT1. Proc Natl Acad Sci USA 2003;100:5902–5907. Nishizato Y, Ieiri I, Suzuki H, et al. Polymorphisms of OATP-C (SLC21A6) and OAT3 (SLC22A8) genes: consequences for pravastatin pharmacokinetics. Clin Pharmacol Ther 2003;73:554–565. Niemi M, Schaeffeler E, Lang T, et al. High plasma pravastatin concentrations are associated with single nucleotide polymorphisms and haplotypes of organic anion transporting polypeptide-C (OATP-C, SLCO1B1). Pharmacogenetics 2004;14:429–440. Istvan E. Statin inhibition of HMG-CoA reductase: a 3dimensional view. Atheroscler 2003;4(Suppl):3–8. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–1389. Sacks FM, Pfeffer MA, Moye LA, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events Trial investigators. N Engl J Med 1996;335:1001–1009. Maitland-van der Zee AH, Klungel OH, Stricker BH, et al. Genetic polymorphisms: importance for response to HMG-CoA reductase inhibitors. Atherosclerosis 2002;163:213–222. Schmitz G, Drobnik W. Pharmacogenomics and pharmacogenetics of cholesterol-lowering therapy. Clin Chem Lab Med 2003;41:581–589. Dornbrook-Lavender KA, Pieper JA. Genetic polymorphisms in emerging cardiovascular risk factors and response to statin therapy. Cardiovasc Drugs Ther 2003;17:75–82. Chasman DI, Posada D, Subrahmanyan L, et al. Pharmacogenetic study of statin therapy and cholesterol reduction. JAMA 2004;291:2821–2827.
Section I: Pathophysiology of the Liver
5
MECHANISMS OF BILE SECRETION Peter L.M. Jansen and Albert K. Groen Abbreviations AA amino acids ABC ATP-binding cassette ASBT apical sodium dependent bile salt transporter BAAT bile acid CoA:aminoacid N-acyltransferase BRIC benign recurrent intrahepatic cholestasis BSEP bile salt export protein CAR constitutive androstane receptor C/EBP CCAAT-enhancer binding protein CYP cytochrome P450 CYP7a cholesterol 7a-hydroxylase DJS Dubin–Johnson syndrome DR direct repeats FHCA familial hypercholanemia FXR farnesoid X-receptor GSH reduced glutathione GSSG oxidized GSH
HNF I-BABP IR LXR MDR1 Mdr2 MDR3 MRP NHR NTCP Ntcp OATP Oatp
hepatocyte nuclear factor intestinal bile acid-binding protein inverted repeat liver X-receptor multidrug resistance protein 1 rodent phosphatidylcholine transporter human phosphatidylcholine transporter multidrug resistance-associated protein and its homologs nuclear hormone receptor human Na+/taurocholate co-transport polypeptide rat Na+/taurocholate co-transport polypeptide human organic anion-transporting protein rat organic anion-transporting protein
INTRODUCTION The generation of bile flow depends on the transepithelial movement of solutes and organic molecules. Bile is produced primarily in hepatocytes, and its composition is modified in the bile ducts. Bile is formed by a process of osmotic filtration in response to osmotic gradients created within the lumen of the bile canaliculus. This osmotic gradient is established by ongoing active secretion of solutes across the canalicular membrane of hepatocytes into the canalicular lumen. Water follows passively through the leaky pores of tight junctions. Bile secretion serves different important functions. First, it is one of the main mechanisms for the disposition of many endogenous and exogenous amphipathic compounds, including drugs, toxins, and waste products. Second, it supplies bile salts to the intestine, which is of crucial importance for the emulsification and subsequent digestion and absorption of dietary lipids. Since it became evident that bile salts are ligands for nuclear hormone receptors in both liver and gut, a third function can be assigned to bile: as a carrier of signaling molecules from liver to gut. The enterohepatic cycling of bile salts is a main determinant of bile flow, but also the secretion of bile salts, cholesterol, phospholipids, and glutathione contributes to the formation of bile. Bile salts are the predominant organic solutes in bile, and their vectorial secretion from blood into bile represents the major driving force for hepatic bile formation. Although bile is iso-osmotic in relation to plasma, bile salts are concentrated up to 1000-fold in bile,
OCT PC PE PFIC PPAR PS PXR RAR RE ROS RXR SHP-1 SLC SP1 TM TNF-a
organic cation transporter phosphatidylcholine phosphatidylethanolamine progressive familial intrahepatic cholestasis peroxisome proliferator-activated receptor phosphatidylserine pregnane X-receptor 9-cis retinoic acid receptor responsive element reactive oxygen species retinoid X-receptor small heterodimer partner solute carrier protein stimulating protein 1 transmembrane a-helix tumor necrosis factor-a
necessitating active transport by hepatocytes. After their secretion into the canaliculus, bile salts are prevented from regurgitation into the systemic circulation by hepatocyte tight junctions, the integrity of which is disturbed during bile duct obstruction. The size of the total bile salt pool in adult humans amounts to up to 50–60 mmol/kg body weight, corresponding to 3–4 g, and is largely stored in the gallbladder during the fasting state. Rats lack this reservoir function because of the absence of a gallbladder. The human bile salt pool circulates 6–10 times per 24 hours, resulting in a daily bile salt secretion of 20–40 g. Despite a high degree of intestinal bile salt conservation, about 0.5 g of bile salt is lost through fecal excretion. This loss is compensated for by de novo hepatic bile salt synthesis. The intrinsic link between intestinal bile salt absorption and hepatic synthesis has been found to be a complex system involving specific bile salt-binding nuclear receptors. Via this mechanism, bile salts can regulate their own enterohepatic circulation. Through interaction with the farnesoid X-receptor (FXR) and the pregnane X-receptor (PXR) bile salts regulate their own biosynthesis, hepatic uptake and secretion. FXR also regulates key steps in hepatic cholesterol, carbohydrate and lipid metabolism, and thus serves as a bridge between bile acids and a range of metabolic reactions. Uptake of bile salts from the sinusoidal blood and secretion across the canalicular membrane are the major determinants governing the rate of bile secretion. Disturbances of bile salt transport are important causes of acquired and genetic forms of cholestatic liver disease. In case of impaired bile salt secretion, the liver can generate a number of adaptations to detoxify or secrete bile salts via alternate
67
Section I. Pathophysiology of the Liver
pathways. When these fail or are overwhelmed, liver damage will ensue, with malnutrition secondary to reduced intestinal absorption of lipids and fat-soluble vitamins as a consequence.
BILE SECRETION UNIT The liver is specialized in the processing of albumin-bound compounds, and as such its function is complementary to that of the kidney. Anatomically the liver is well equipped for its function. Microscopically the liver looks like a sponge. The holes permit the passage of blood and the solid material consists of hepatocytes and a network of bile canaliculi. On a closer look the sponge has an ordered structure, with rows of hepatocytes radiating from portal areas towards a terminal hepatic vein. This relation between the portal triads and the terminal hepatic veins is the smallest anatomical unit in the liver, and is called the hepatic acinus.1 In the human liver these acini are not isolated units but are connected with each other in the periportal area.2 The portal vein carries metabolites from the intestine directly to the liver. In the liver the portal blood flows through the sinusoids from the portal triads towards the terminal hepatic vein. The fenestrated sinusoidal endothelium lacks a basal membrane and thus allows easy passage of molecules from the blood to the surface of the hepatocytes. Hepatocytes and endothelium are separated by the 10–15 mm wide space of Disse. In a three-dimensional view, single layers of hepatocytes form plates which are perfused on both sides by portal venous blood. A bile canalicular network is hidden within these plates. After immunohistochemical staining with antibodies directed against canalicular proteins, this network shows up as a chicken-wire pattern. Because of its three-dimensional structure the chicken wire, when seen in the plane of the microscope, seems incomplete. With antibodies against basolateral proteins one obtains a more regular honeycomb structure. As in the honeycomb, each hole in the chicken wire is a hepatocyte, indicating that each individual hexagonal hepatocyte is surrounded by a canaliculus. Within the plates of hepatocytes, the canaliculi form a cul-de-sac in the pericentral region. Near the portal triads they are connected with the bile ducts via the canals of Hering. Many canaliculi drain into one canal of Hering.3 Bile flows from the pericentral to the portal zone, opposite to the direction of the flow of blood. The fenestrated endothelium of the hepatic sinusoids allows the passage of small molecules, proteins, and large particles such as chylomicrons. Blood cells cannot pass. Thus, the sinusoidal endothelium acts as a dynamic biofilter:4 dynamic because the diameter of the fenestrae changes upon alterations in portal pressure and by pharmacological agents. In the postprandial state portal pressure rises, and this will cause an increase in the size of these fenestrae. Also, agents such as alcohol, nicotine and serotonin induce changes in the diameter of the fenestrae. In liver cirrhosis this regulation is disrupted, the sinusoids lose their fenestrations and acquire a basement membrane. This may contribute to liver dysfunction and portal hypertension.5 The space of Disse is continuous with the spaces between the hepatocytes. The diameter of white blood cells (WBC) is larger than the diameter of the sinusoids. Therefore, upon passage of WBC the space of Disse is temporarily obliterated when the endothelium is
68
pressed against the hepatocytes. During these periods the hepatocyte plasma membrane is directly in contact with the blood space. The space of Disse is continuous with the lymph vessels. Thus hepatic lymph is generated in the space of Disse. Tight junctions form a barrier between these intercellular spaces and the bile canalicular lumen. Tight junctions between hepatocytes are permeable for water and electrolytes and have a limited permeability for organic cations.6 They are impermeable for organic anions. Tight junctions are complex structures in which the transmembrane proteins occludin and claudin interact with similar proteins in neighboring cells and with the cytoplasmic tight junctional proteins ZO-1 and ZO-2.7 In cholestatic liver disease the tight junctional permeability alters and allows the passage of organic anions back from bile to the interstitial space of Disse.8 The many compounds for which the tight junctions form an impermeable barrier have to traverse the hepatocyte en route from blood to bile. Receptor-mediated endocytosis and pinocytosis play a role in the transcellular routeing of proteins and macromolecules. For small charged molecules the exact mechanism of vectorial transcellular transport is still obscure. Single-pass perfusion experiments with isolated perfused rat liver preparations showed that the paracellular permeation from blood to bile takes 2–3 minutes, whereas transcellular transport takes from 5 to 20 minutes.9 A relatively fast transcellular component is not inhibited by microtubule inhibitors, whereas a slower component is inhibited by these agents and therefore seems to be associated with intracellular vesicles, or with socalled ‘lipid rafts’.10 The canalicular domain is the hepatocyte plasma membrane section that surrounds the canaliculus. In fact, it surrounds half of the canaliculus, as the domains of two adjacent hepatocytes, linked together via tight junctions, form the complete surrounding membrane. A canalicular network in an entire liver plate is in contact with a large number of hepatocytes. Hepatocytes secrete bile salts, which are toxic detergents that have to be neutralized by cholesterol and phospholipids. Therefore, hepatocytes have to communicate with each other in order to tune the secretory activity of adjacent hepatocytes. If this did not occur the bile salts secreted by one hepatocyte would damage the neighboring cell. This communication occurs via the gap junctions. Gap junctions allow the passage of small signaling molecules, such as calcium and/or nucleotides, from one hepatocyte to the other. The canalicular membrane contains a number of proteins with a specialized transport function (Figure 5-1). These proteins belong mainly to the large ABC transporter superfamily.11,12 In fact, almost all compounds destined for biliary secretion are handled by these proteins. This raises the question as to how substrates are delivered from the cell interior to the canalicular transporter molecules. Two models have been proposed: a vacuum-cleaner model, in which compounds dissolved in the membrane are pumped from the outer leaflet into the bile, and a mechanism of transmembrane transfer wherein the substrate is picked up at the interior side and released at the cell exterior. Perhaps both models may coexist: the vacuumcleaner model for highly amphiphilic molecules and the transfer model for more hydrophilic compounds such as glutathione. The canalicular pumps are embedded within lipid microdomains of the canalicular membranes. The lipid composition of these microdomains is important for the function of the pumps, and
Chapter 5 MECHANISMS OF BILE SECRETION
Water and electrolytes OATP-C Bilirubin Bile acids Na+
MRP2 Bilirubin
ABCG5/G8 Cholesterol Phosphatidyl choline MDR3
Bile acids BSEP
NTCP
Enterohepatic cycle
HDL-cholesterol SR B1
Liver
CFTR Cl-
Bile ducts
MRP3 OST␣ ASBT
ASBT H O 2
AQP
Bile acids
derangements of intracellular lipid trafficking may lead to dysfunction of the surface pumps. The canalicular membrane represents about 15% of the total surface area of hepatocytes. It contains transport proteins that are able to pump the cholephilic compounds into bile against a 100-fold concentration gradient. This active ATP-dependent transport can be considered the principal driving force of bile flow. In terms of energy bile formation is a costly process: for example, the secretion of one molecule of unconjugated bilirubin costs four molecules of ATP equivalents: two molecules of UDP-glucuronic acid for conjugation and two molecules of ATP for canalicular transport. However, bile formation has at least a dual function: it rids the body of metabolic waste and it is important for the intestinal digestion of foodstuffs, in particular of energy-rich lipids. The inability to produce bile is associated with rapid weight loss. The energy it costs to produce bile seems well spent. The portal blood is rich in metabolites. Many of these are taken up from the blood in the first hepatocytes of the hepatic acinus. Studies with fluorescent or radiolabeled bile salts and fatty acids revealed a steep acinar gradient with a concentration that was high in the periportal hepatocytes and low in the pericentral hepatocytes. This indicates that bile salts and fatty acids are efficiently extracted in these first hepatocytes.13,14 Depending on the bile acid species, up to 98% is removed by the liver during one passage. Bile salts repress their own synthesis by inhibiting the first committed step of the classic neutral pathway, microsomal cholesterol 7a-hydroxylase or Cyp7A1, and the first step of the so-called acidic pathway, the mitochondrial sterol 27-hydroxylase Cyp8B1.15 Thus the periportal hepatocytes are intensively involved in the enterohepatic cycling of bile salts, whereas the hepatocytes in the pericentral zone are more active in de novo biosynthesis.16 When bile leaves the hepatic acinus it enters the bile ducts. These are the structures seen on light microscopy in the portal triads, where they are accompanied by one or two hepatic arteries and a
Figure 5-1. Human hepatic transporter proteins involved in bile formation. Transporter proteins located in the basolateral membrane are responsible for the uptake of bile salts (NTCP), bulky organic anions, uncharged compounds (OATPs) and cations (OATPs, OCT1). Transporter proteins located in the canalicular membrane are responsible for the biliary secretion of, for example, bile salts, phosphatidylcholine, cholesterol, bilirubin conjugates, and oxidized and reduced glutathione. These transporter proteins comprise the bile salt transporter BSEP, the phosphatidylcholine translocator MDR3, the anionic conjugate transporter MRP2 and the multidrug transporter MDR1 (not shown). The organic anion transporters MRP3 and MRP4 are present at very low levels in normal hepatocytes but are up-regulated during cholestasis. ABCG5/G8 are two half-transporters (half the molecular mass of regular ABC transporters) and together act as cholesterol and plant sterol transporters.
Intestine
branch of the portal vein. From here, bile flows via the intrahepatic and extrahepatic ducts to the intestine. Although the total volume of bile duct epithelial cells contributes no more than 3–5% to the total liver, bile ducts in normal human liver are estimated to be 1 km in length. Small and large bile duct epithelial cells have distinct morphological and functional features and differ in their proliferative capacity.17 Bile duct epithelial cells have a collection of transporter proteins on their apical and basolateral surfaces, indicating that the bile duct epithelium has both an absorptive and a secretory function. It is not surprising, therefore, that bile composition is considerably modified in the bile ducts. Here, bile becomes enriched in bicarbonate and chloride, whereas glucose and glutamate are reabsorbed. Also bile salts may to some extent be reabsorbed.18 Indirect proof for bile salt reabsorption is the presence of the apical sodiumdependent bile salt transporter ASBT (SLC10A2) and the observed uptake of fluorescent bile salts in the bile duct epithelium. In the gallbladder mucin is produced and added to bile. Thus what comes out in the gut through the ampulla of Vater is a rather viscous, mucin-containing yellow fluid composed of phospholipids, cholesterol and bile salts. The yellow color comes from bilirubin, which is present in millimolar quantities. Bile also contains amino acids, carnitine, and many other solutes. The final bile contains very little glutathione. This tripeptide is almost completely degraded and its components are largely reabsorbed in the bile ducts.19 Some drugs are highly concentrated in bile. An example is ceftriaxone, which can even precipitate out of solution and form gallstones, particularly in children.20 In the ileum bile salts are reabsorbed. ASBT mediates bile salt uptake into the ileal epithelium (Table 5-1). Human ASBT transports conjugated and unconjugated bile salts, having a higher affinity for dihydroxy than for trihydroxy bile salts. Some sodium-independent bile salt transport may be mediated by Oatp3 (Slc21a7), which is present in all small intestinal segments.21 After uptake the bile salt molecules move through the enterocyte to the
69
Section I. Pathophysiology of the Liver
Table 5-1. Human Hepatic Transporter Proteins Name
Gene
Chromosome
Localization
Transport function
NTCP ASBT OCT1 OAT-2 OATP OATP2 OATP8 OATP-B ABC1
SLC10A1 SLC10A2 SLC22A1 SLC22A7 SLC21A3 SLC21A6 SLC21A8 SLC21A9 ABCA1
14q24.1-24.2 13q33 6q26 6p21.2-21.1 12p12 12p 12p12 11q13 9q31
H-BL CH-A, E-A H-BL H-BL H-BL H-BL H-BL H-BL H-BL, E-BL
BS BS OC OA BS / OA / OC BS / OA / OC BS / OA / OC OA Cholesterol
MDR1 MDR3 BSEP MRP1 MRP2 MRP3 MRP4 MRP6
ABCB1 ABCB4 ABCB11 ABCC1 ABCC2 ABCC3 ABCC4 ABCC6
7q21 7q21 2q24 16p13.12-p13 10q23-q24 17q21.3 13q31 16p13.1
H-A, CH-A H-A H-A H-BL, CH-BL H-A; E-A H-BL, CH-BL H-BL H-BL
CFTR ABCG2
ABCC7 ABCG2
7q31.2 4q22-q23
CH-A H-A
ABCG5 ABCG8 FIC1 WND
ABCG5 ABCG8 ATP8B1 ATP7B
2p21 2p21 18q21-q22 13q14.3
H-A; E-A H-A; E-A H-A, CH-A; E-A H-INT
Drugs, chemotherapeutics PC BS OA, OA conjugates OA, OA conjugates OA conjugates BS sulfates Peptides, endothelin receptor antagonist BQ123 Chloride Chemotherapeutics, chlorophyl metabolites, protoporphyrin Plant sterols Plant sterols Aminophospholipid translocation Copper
Phenotype when defective Bile acid diarrhea
High-density lipoprotein deficiency, Tangier type PFIC3, ICP, intrahepatic gallstones PFIC2, BRIC 2 Dubin–Johnson syndrome
Pseudoxanthoma elasticum Cystic fibrosis Photosensitivity (mice) Sitosterolemia Sitosterolemia PFIC1, BRIC 1, ICP Wilson’s disease
H, hepatocytes; CH, cholangiocytes; E, enterocytes. A, apical; BL, basolateral; INT, intracellular. BS, bile salts; OA, organic anions; OC, organic cations; PC, phosphatidylcholine. PFIC, progressive familial intrahepatic cholestasis; ICP, intrahepatic cholestasis of pregnancy; MDR, multidrug resistance. The standard name in bold, the gene name assigned by the nomenclature committee (http://www.gene.ucl.ac.uk/nomenclature/genefamily/abc.html) is mentioned.
basolateral domain. The 14 kDa ileal bile acid-binding protein (IBABP) may play a role in this. The quest for proteins involved in basolateral efflux has been long. The most likely candidates that have been characterized recently are the heteromeric organic solute transporters Osta-Ostb.22 Reabsorption is very efficient, as only about 10% of the total biliary bile salts that enter the bowel each day escape reabsorption in the ileum. These enter the colon, where they become subject to bacterial metabolism that converts primary to secondary bile salts. Some of these, such as deoxycholate, become reabsorbed in the colon by an as yet undefined transport mechanism. Phospholipids are hydrolyzed in the intestine and the mono- and diphosphate esters are subsequently reabsorbed. Phospholipids in the intestinal lumen are important for the formation of chylomicrons. The formation of these lipoproteins is disturbed in Abcb4 (Mdr2) (-/-) knockout mice, which lack phospholipids in bile.23 Cholesterol reabsorption is 60–80%, depending on conditions and the expression of proteins, which may show species differences. Until recently cholesterol absorption in the intestine was considered to occur passively, only to some extent facilitated by proteins. However, in recent years it has become apparent that intestinal cholesterol absorption is a complex process involving separate counteracting transport systems. In drug therapy it must be realized that drugs may also participate in enterohepatic cycling. This adds considerably to their biological half-life. Ceftriaxone is one example, and because of its
70
enterohepatic cycling has to be administered only once daily. There is evidence that unconjugated bilirubin or bilirubin photoproducts also take part in the enterohepatic circulation.24 Oral bilirubin trapping agents, such as fresh calcium phosphate, interrupt this cycling and lower serum bilirubin levels in patients with disturbed hepatic bilirubin glucuronidation (Crigler–Najjar syndrome). Hepatocytes, bile ductuli and ducts together can be considered a ‘hepatic secretory unit’ with a certain analogy to the nephron. In both there is a primary solution produced by filtration, which in the liver occurs through the tight junctions, and active secretion by the hepatocytes. This ‘primary’ bile is modified in the bile ducts through reabsorption of unconjugated bile salts, glucose, glycine, glutamate, and by the secretion of water, chloride and bicarbonate. As in the nephron this ductal secretion is under hormonal and neural control.
HEPATIC TRANSPORT PROTEINS To secrete bile and to excrete metabolites of toxic substances, hepatocytes must transport bile salts, phospholipids and other solutes from blood to bile. Various basolateral transporters for organic solutes have been characterized. These transporters belong to the solute carrier superfamily and comprise the sodium-dependent transporter for the uptake of bile salts (NTCP; SLC10A), transporters for amphiphilic substrates such as members of the subfamilies of organic
Chapter 5 MECHANISMS OF BILE SECRETION
anion transporting polypeptide (OATPs; SLC21A), and organic cation transporters (OCTs; SLC22A) (Table 5-1).
BASOLATERAL TRANSPORT PROTEINS The Na+/taurocholate co-transporting polypeptide (NTCP; SLC10A1) represents the major bile salt uptake system of hepatocytes, localized exclusively in the basolateral membrane of hepatocytes (Table 5-1; Figure 5-1).25 NTCP preferentially mediates Na+-dependent transport of conjugated bile salts such as taurocholate, and this transport comprises the predominant if not the exclusive fraction in hepatic bile salt uptake.26 The human liver NTCP also transports conjugated bile salts, but human NTCP has a higher affinity for taurocholate than does rat NTCP (KM 6 mM compared to 25 mM).27 Rat NTCP has rather broad substrate specificity. In addition to bile salts, sulfated steroids, bromosulphophthalein (BSP) and thyroid hormones have also been shown to be transported by this protein. An important feature of the Na+-independent bile salt uptake pathway is its wide substrate preference. This includes conjugated and unconjugated bile salts, BSP, cardiac glycosides and other neutral steroids, linear and cyclic peptides, selected organic cations, and numerous drugs such as pravastatin. The organic anion transporting polypeptide 1 (Oatp1; Slc21a1) is localized in the basolateral membrane of hepatocytes and mediates Na+-independent uptake of a very broad spectrum of anionic and even cationic substrates. Oatp1 does not transport unconjugated bilirubin. Oatp2 (Slc21a5) is expressed in rat hepatocyte basolateral membranes. Oatp2 exhibits a 77% amino acid identity with Oatp1. Its spectrum of transport substrates is similar, but not identical to, that of Oatp1. An important difference between Oatp1 and Oatp2 was observed with respect to their acinar localization in the liver. Whereas Oatp1 is homogeneously expressed within the liver acinus, Oatp2 is predominantly expressed in perivenous hepatocytes, excluding the innermost one to two cell layers surrounding the central vein. Treatment of rats with phenobarbital resulted in a significant increase in Oatp2 expression and in the appearance of positive immunofluorescence signals even in the innermost layer of perivenous hepatocytes. Thus, Oatp2 behaves as an inducible transporter that may account for the increased organic anion transport after phenobarbital treatment. This is line with the fact that the constitutive androstane receptor CAR and the pregnane X-receptor PXR, members of the superfamily of nuclear hormone receptors, that are responsible for cytochrome P450 3A4 induction by chemicals, also induce Oatp2.28 Phenobarbital and pregnenolone16a-carbonitrile are the prototypical ligands for CAR and PXR, respectively.29 However, lithocholic acid also is a PXR ligand and can therefore induce PXR target genes.30 Because Oatp2 can also function as an effluxer, it may contribute to cellular protection against severe liver damage induced by toxic bile salts such as lithocholic acid. Oatp4 (Slc21a10) is a third member of the Oatp family. It is expressed mainly in liver and mediates Na+-independent uptake of bile salts in rat hepatocytes.31,32 Oatp4 accepts numerous organic anions as transport substrates. Additional members of the Slc21a family that are not expressed in hepatocytes and/or not involved in bile salt transport include the prostaglandin transporter (rPGT)
(Slc21a2), OAT-K1/2 (Slc21a4), and Oatp3 (Slc21a7). Oatp3 is mainly expressed in small intestine.21 In human liver, SLC21A6 (also called OATP-C, OATP2 or liverspecific transporter 1 LST-1), is expressed at the basolateral membrane of hepatocytes. It exhibits the highest (64%) amino acid identity with the Oatp4 of rat liver. Transport substrates of SLC21A6 include a number of organic anions, and this protein is probably the predominant Na+-independent bile salt uptake system of human liver. A second human Na+-independent bile salt uptake system is SLC21A3 (OATP-A; previously called OATP) (Figure 5-1; Table 5-1). Although SLC21A3 was originally isolated from human liver, it is mainly expressed in human cerebral endothelial cells. Its transport substrates include bile salts and other organic anions, as well as certain organic cations. This indicates that it can mediate charge-independent solute movement across membranes. SLC21A9 is most abundant in human liver, where it is localized at the basolateral membrane of hepatocytes. SLC21A9 (OATP-B), SLC21A6 (OATP-C, OATP2), and SLC21A8 (OATP8) mediate high-affinity uptake of BSP. SLC21A9 transports no bile salts. SLC21A9, SLC21A6, and SLC21A8 exhibit broad overlapping substrate specificities and account for the majority of clearance of sodium-independent bile salts, organic anions and drugs from the human liver.33 SLC21A8 uniquely transports digoxin and exhibits high transport activity for certain anionic cyclic peptides. Finally, bilirubin and its conjugates are taken up from the blood circulation into the liver by SLC21A6.34 In comparison to the high-affinity transport by SLC21A6, SLC21A8 transports BSP and monoglucuronosyl bilirubin with lower affinity. The clinical and pharmacological relevance of these uptake transporters is beginning to be appreciated. Polymorphisms of the genes encoding these proteins have an impact on drug clearance. This is of particular significance for drugs with a narrow therapeutic window, such as the chemotherapeutic agent irinotecan.35 Furthermore, drug–drug interactions may occur at the level of hepatic uptake proteins.36 A variety of small (type I) organic cations, including drugs, choline, and monoamine neurotransmitters, are translocated by OCT1 (SLC22A1).37 From the five SLC22A family members (OCT1-3, OCTN1, OCTN2) only OCT1 is relevant to the liver. The human OCT138 is 78% identical to rat Oct1 and has a comparable substrate specificity. However, in contrast to Oct1 expressed in kidney, liver, intestine and colon, human OCT1 is almost exclusively expressed in human liver. OCT transporters appear to have 12 transmembrane a-helices and contain four short motifs with five amino acid residues. Recent studies demonstrate that in rat liver the suggested organic cation uptake systems 1 (small cations) and 2 (bulky cations) correspond to Oct1 and Oatp2, respectively.39 Whereas OCT1 exhibits similar transport properties to rat Oct1, the rat-based type II organic cation transporter classification cannot be extended without modification from rat to human, because of the different specificities of human OATPs. With respect to the function of OCT1 in the liver, data have become available from knockout mice. Oct1(-/-) mice are viable, healthy, and fertile and display no obvious phenotypic abnormalities.40 In Oct1(-/-) mice the accumulation of small organic cations such as tetraethylammonium (TEA) or the neurotoxin 1-methyl-4-phenylpyridium in liver was lower than in wild-type mice, indicating an important role of this transporter in the uptake of organic cations into the liver.
71
Section I. Pathophysiology of the Liver
CANALICULAR TRANSPORT PROTEINS Most canalicular transport systems involved in bile formation belong to the 48 members of the ATP-binding cassette (ABC) transporter superfamily, one of the largest superfamilies of proteins in prokaryotes and eukaryotes.41 With respect to bile formation, four subclusters of this superfamily appear to be most important, the A, B, C, and D clusters (Table 5-1).
dependent. In fact, activation of the rat Mdr1b gene by TNF-a is a result of NF-kB signaling.55 Mdr1b up-regulation, at least in part, may provide antiapoptotic protection against oxidative-stress induced cell damage. PXR is the molecular target for certain steroids, such as pregnenolone 16a-carbonitrile (PCN), and for the toxic bile salt lithocholic acid. These induce CYP3A expression and thus protect the body from harmful chemicals or their own toxicity. PXR also regulates drug efflux by activating expression of the gene MDR1.56
MDR1 P-Glycoprotein (ABCB1) The multidrug resistance gene products, the human MDR1 (ABCB1), MDR3 (ABCB4) and their mouse and rat homologs Mdr1a, Mdr1b and Mdr2 (also called P-glycoproteins), were the first ABC transporters recognized in canalicular membranes of normal hepatocytes42 (Figure 5-1). Mdr1 and Mdr2 genes are both expressed in rodent liver; however, in normal rodent liver Mdr2 is the predominant gene product. Various physiological functions of MDR1/Mdr1 have been demonstrated or postulated, such as transport of exogenous and endogenous metabolites or toxins,43 steroid hormones,44 hydrophobic peptides,45 or amphiphilic cationic drugs.46 To establish the physiological function of Mdr1- and Mdr2-Pgps, a set of gene knockout mice were generated.47 These mutant mice did not express either functional Mdr1a (Mdr1a(-/-),48 Mdr1b (Mdr1b(-/-),47 or Mdr2 (mdr2(-/-)).49 Double-knockout mice were also produced (Mdr1a(-/-)/Mdr1b(-/-)).47 These mice were all fertile. The Mdr1-knockout mice exhibited an almost normal phenotype under laboratory conditions.47 From experiments with Mdr1a/1b(-/-) gene knockout mice no changes in bile composition became apparent.50 Disruption of both genes has no effect on the normal laboratory life of these mice, but renders them hypersensitive to drugs. MDR1 appears to be especially important in protecting the brain. In the gut, where MDR1 pumps from the enterocyte towards the lumen, MDR1 limits the uptake of hydrophobic drugs. By analogy, MDR1 may protect the hepatocyte against hydrophobic toxic drugs by pumping them into the bile. MDR1 polymorphisms have been reported to be relevant in cancer chemotherapy, in immunosuppressive therapy in organ transplant recipients, and for the natural clinical course of cancer patients.51 The regulation of MDR1 and its rodent homologs has been studied in detail. It is clear that MDR1 promoter activation is part of a general stress response resulting in cellular resistance. p53 is involved in the basal regulation of MDR1, rat Mdr1a, and Mdr1b.52 Whereas wild-type p53 represses Mdr1a expression, overexpression of mutant p53 results in markedly elevated levels of rat Mdr1a mRNA and protein levels. Similar regulation was reported for MDR1. A functional p53-binding site has been identified in the rat Mdr1b promoter.53 Wild-type p53 was shown to up-regulate Mdr1b promoter activity and to mediate the endogenous expression of rat Mdr1b. These results and other studies indicate that the two rodent Mdr1 genes are differentially regulated. The expression of Mdr1a in rat liver is, for example, not affected by endotoxin treatment and increases only slightly after bile duct ligation or partial hepatectomy.54 In contrast, Mdr1b expression is markedly enhanced during endotoxin and bile duct ligation-induced cholestasis, and even more in the remnant liver after partial hepatectomy.54 Upregulation of Mdr1b during liver regeneration, after partial hepatectomy or after endotoxin treatment is at least in part TNF-a
72
Phospholipid Transporter (ABCB4; MDR3) In the liver the human MDR3 and the rodent Mdr2 is expressed in canalicular membranes just as is Mdr157 (Figure 5-1). The function of Mdr2 became apparent after producing Mdr2(-/-) knockout mice with a complete absence of bile phospholipid.58 From these and other studies it is now well accepted that Mdr2 and its human counterpart MDR3 acts as a flippase, translocating phospholipids through the canalicular membrane.59 The current working hypothesis for the mechanism of phospholipid secretion is that MDR3/Mdr2 flips phosphatidylcholine from the inner to the outer leaflet of the lipid bilayer. These lipids are probably concentrated in microdomains in the exoplasmic hemileaflet of the canalicular membrane.60 Bile salts solubilize PC from these microdomains, either in the form of vesicles or as mixed bile salt/PC micelles. Cholesterol secretion is strongly reduced but can be increased in Mdr2 knockout mice by enriching the bile with taurocholate.61 In a physiological sense these mice are not cholestatic because the bile acid secretion is normal and bile flow elevated.58 However, histologically there are features as observed in cholestasis, such as bile ductule proliferation and feathery degeneration of hepatocytes.49 Biliary bile acid secretion in these mice is not accompanied by phospholipids. Therefore, the bile they produce is cytotoxic. Older mice with the gene disruption develop liver tumors.49 The human counterpart of the Mdr2 knockout mice are patients with progressive familial intrahepatic cholestasis, PFIC3 (see later). MDR3 mutations have been found in patients with intrahepatic gallstones and in a subgroup of patients with intrahepatic cholestasis of pregnancy.62–64 Although the histopathology of Mdr2 knockout mice resembles to a certain extent that seen in human primary biliary cirrhosis, no evidence has so far been obtained that MDR3 is involved in this disease.65 An imbalance between bile acid and phospholipid secretion after liver transplantation was correlated with a transient imbalance between BSEP and MDR3 expression, and this may provide an explanation for the vulnerability of the bile ducts in the immediate post-transplantation period.66 Expression of Mdr2 in rodent liver appears to be unaltered under most conditions of cellular stress.67 Mdr2 expression was not affected after endotoxin treatment,68 and was only slightly enhanced after partial hepatectomy.54 When mice were fed a diet supplemented with fibrates this increased Mdr2 mRNA and protein levels and increased PC secretion, suggesting the involvement of PPAR-a in Mdr2 gene expression.69 In mice fed a diet supplemented with the hydrophobic bile salt cholate Mdr2 mRNA levels were found to be induced, which was functionally reflected in a concomitant increase of the maximal PC secretion capacity.70 Feeding the (relatively) hydrophilic bile salt ursodeoxycholate did not influence the Mdr2 mRNA levels nor the maximal PC output capacity.70
Chapter 5 MECHANISMS OF BILE SECRETION
The Bile Salt Export Pump BSEP (ABCB11) The bile salt excretory pump (BSEP; ABCB11) is critical for ATPdependent transport of bile acids across the hepatocyte canalicular membrane and for the generation of bile acid-dependent bile secretion (Figure 5-1).71 Patients genetically lacking BSEP have a severe cholestatic liver disease characterized by high serum bile acid levels.72, 73 Bsep (-/-) mice are cholestatic in the sense that taurocholate accumulates in their plasma because its secretion into bile is strongly impaired.74 However, in contrast to patients, the mice excrete substantial amounts of tauromuricholate into bile, as well as tetrahydroxy bile salt metabolites. Apparently, Bsep is not the only bile salt transporting system in the canalicular membrane, as other systems appear capable of excreting these hydrophilic bile salts. This escape route prevents severe and progressive cholestasis, and as a consequence the animals have hardly any histopathological signs of liver injury. Because humans are not capable of converting bile salts into muricholate or tetrahydroxy bile salts to any significant extent, this escape route is not available. The regulation of rat Bsep has been studied under conditions of endotoxin treatment, bile duct ligation and ethinylestradiol-induced cholestasis.68,75 In these cholestatic and stress models Bsep mRNA and protein expression levels decrease only slightly compared to levels of the basolateral bile salt carriers Ntcp,76 Oatp1 and Oatp2, or the canalicular transporter Mrp2.54,75 Thus, Bsep may continue to secrete bile salts, although at impaired rates. Remarkably, after partial hepatectomy the mRNA level of Bsep is only slightly decreased and the proteins level of Bsep were unaffected in contrast to those of the bile salt uptake transporter Ntcp.54,77 This may explain why after partial hepatectomy the remnant liver is not cholestatic and not damaged by excess bile salts. Also in human liver disease BSEP expression is usually not much affected.65,78 The expression of BSEP is sensitive to the flux of bile acids through the hepatocyte. The BSEP promoter contains an IR-1 element that serves as a binding site for the farnesoid X-receptor, a nuclear receptor for bile acids.79 FXR activity requires heterodimerization with RXRa, and when bound by bile acids the complex effectively regulates the transcription of several genes involved in bile acid homeostasis. By this mechanism bile acids transcriptionally regulate the activity of BSEP, preventing increased hepatocellular levels of potential toxic bile salts (Figure 5-2). In addition to BSEP, FXR has a great number of target genes, some quite unrelated to bile salt metabolism. Therefore, the use of potent FXR ligands as drugs has to be studied with care, as they could have unpredictable side effects.80
The Multidrug Resistance Protein MRP2 (ABCC2) MRP2 is located in the canalicular membrane of hepatocytes as well as the apical membranes of enterocytes, renal tubular cells, the blood–brain barrier and seminiferous tubules (Figure 5-1). In the liver endogenous and xenobiotic lipophilic compounds are converted into more hydrophilic anionic conjugates with glutathione, glucuronate, or sulfate. These conjugates are transported across the canalicular membrane into bile by MRP2. MRP2 contributes to the bile formation by transporting glutathione, a major driving force for bile salt-independent bile flow. In Dubin–Johnson syndrome MRP2 is genetically lacking.81 These patients have a mild conjugated hyper-
bilirubinemia but are otherwise quite healthy. Drug side effects in these patients have rarely been reported. Rats lacking Mrp2 have a normal lifespan and a normal breeding capacity. These rats have been invaluable for the demonstration of canalicular transporter function for the disposition of drugs and endogenous metabolites.82 MRP2 gene expression is regulated by three nuclear receptors PXR, FXR, and CAR.83 The promoter regions of the human MRP2 and the rat Mrp2 genes have been isolated.84 Interestingly, in cholestasis Mrp2 is up-regulated in the kidney.85 This helps in eliminating organic anions, glucuronide and glutathione conjugates when hepatobiliary function is impaired.
The Basolateral Anionic Conjugate and Bile Salt Transporters MRP3 and MRP4 MRP3 (ABCC3) is a transporter protein that supports the basolateral export of organic anions, including glutathione and glucuronide conjugates as well as bile salts from hepatocytes.86 For bile salts its affinity is low, and its expression in normal liver is also low.87 Under control conditions MRP3 is expressed in the centrilobular hepatocytes, in bile duct epithelium and in the gallbladder.88 Basolateral MRP3 expression is up-regulated in livers of Dubin–Johnson patients and during cholestasis.65,89 In animal models Mrp3 is upregulated in Mrp2-deficient EHBR rats, in bilirubin UDPglucuronosyltransferase-deficient Gunn rats and in bile duct-ligated rats.86,90 These are models with conjugated hyperbilirubinemia, unconjugated hyperbilirubinemia and cholestasis. MRP3 is clearly the inducible counterpart of MRP2. What the natural agents might be that induce MRP3 is less clear. LPS and cytokines induce MRP3.91 MRP3 transcription is regulated by the drug-activated nuclear hormone receptors CAR and PXR.92 MRP4 (ABCC4) is an inducible basolateral transporter that cotransports reduced glutathione and the taurine and glycine conjugates of cholic acid.93 It is also a high-affinity transporter of sulfated bile salts and the sulfate conjugate of dihydroepiandrosterone.94 The drug-activated receptor CAR activates transcription of the Mrp4 gene and the dihydroepiandrosterone sulfotransferase2a1 gene. The latter is the main sulfo-conjugating enzyme in the liver, and as such prepares substrates for Mrp4-mediated transport. MRP4 is the main high-affinity bile salt overflow system, more so than MRP3, which is the dominant glucuronide- and glutathione-conjugate overflow system.
Other Hepatic ABC-Transporter Proteins The ABCA1 gene (also called ABC1) is mutated in Tangier patients.95 ABCA1 is the transporter that regulates plasma HDL levels.96,97 It promotes the efflux of cholesterol from various types of cell to apolipoprotein A1.98 The exact mechanism by which it transports cholesterol and phospholipid to lipid-poor apo A1 is still not clear. Most evidence points to a sequential transport of phospholipid and cholesterol. Hepatic ABCA1 plays a crucial role in HDL formation. After secretion from the hepatocyte, apo A1 is directly lipidated by ABCA1, a process that takes place in the space of Disse. Without this step, the apolipoprotein is rapidly degraded in the kidney. ABCA1, expressed in macrophages, may be important in prevention of foam cell formation but does not contribute significantly to plasma HDL levels.97 Abca1 knockout mice secrete
73
Section I. Pathophysiology of the Liver
Regulation
Cholesterol
CYP7A1 CYP 8B1
NTCP
BSEP
Bile acids Na+
Bile acids
Bile acids
FXR:RXR RAR:RXR LXR:RXR
BSEP
Figure 5-2. Hepatic gene regulation by bile salts and role of members of the nuclear hormone superfamily. A Bile salts are taken up by Ntcp and are secreted into bile by BSEP. Bile salts serve as ligands for FXR and PXR. These nuclear hormone receptors act as transcription factors for a number of genes, among which is SHP-1, a gene that encodes a small protein that interacts with LXR-mediated gene regulation, BSEP and MRP2. B SHP-1 interferes with the expression of the bile acid biosynthetic enzymes CYP7A1 and CYP8B and NTCP (as well as numerous other genes). Primary bile salts act as ligands for FXR. FXR activation causes up-regulation of BSEP and MRP2 gene expression.
NTCP CYP7A1 CYP8B1
A
Adaptation 1, down-regulation
Cholesteral CYP7A1 CYP 8B1
NTCP
4
4
BSEP
Bile acids Na+
Bile acids +
FXR:RXR FXR:RXR 3
SHP-1
1
2 BSEP SHP-1 NTCP, CYP7A1 CYP8B1
B
normal amounts of cholesterol in their bile,99 and from this one can conclude that HDL-derived cholesterol is not necessarily the origin of biliary cholesterol: alternative pathways must exist. ABCA1 gene expression is under the control of the nuclear hormone receptor PPAR-a and PPAR-g and of the oxysterol receptors LXRa and LXRb,100,101 indicating the tight regulation of ABCA1 by lipids and cholesterol metabolites. Several members of the G-cluster of ABC half-size transporters are expressed in the canalicular membrane of hepatocytes (Figure 5-1). ABCG2 (ABCP/MXR/BCRP) is highly expressed and trans-
74
ports a wide variety of large hydrophobic molecules, including cytotoxic compounds such as mitoxantrone, topotecan and methotrexate, as well as toxic compounds found in normal food, such as 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) or pheophorbide A.102 The protein is active when present as a homodimer.103 In contrast, ABCG5 and ABCG8 are functional as heterodimers. These transporters, defective in b-sitosterolemia,104 are considered to play an important role in the control of total body sterol homeostasis. Like ABCA1, they are regulated via a number of nuclear hormone receptors, including LXRa, LXRb and FXR.105
Chapter 5 MECHANISMS OF BILE SECRETION
Adaptation 2, up-regulation Liver
Cholesteral CYP7A1 CYP 8B1 NTCP Bile acids Na+
4
BSEP
4
Bile acids
Bile acids 1 2 CYP3A FXR:RXR FXR:RXR
Kidney 3
SHP-1
4
+ PXR
MRP2
MRP2
BSEP SHP-1
Figure 5-2, cont’d. C PXR is activated by drugs (rifampicin) and secondary bile salts (lithocholic acid). PXR activation causes up-regulation of OATP2, MRP2 and CYP3A4 and DHEA sulfotransferase expression. Apart from its function as a bilirubin conjugate transporter MRP2 transports bile salt conjugates. DHEA sulfotransferase is an enzyme that mediates the sulfation of bile salts. These are substrates for MRP4 which, upon its upregulation, acts as a high-affinity transporter of bile salt sulfoconjugates. As a consequence of this bile salt-dependent gene regulation intracellular levels of potential toxic bile salts are kept within limits.
SULT2A1
NTCP, CYP7A1 MRP4, MRP2 CYP3A, SULT2A1
MRP4 Bile acids
C
Abcg5/g8 double-knockout mice show 70–90% decreased cholesterol secretion in bile, indicating the crucial role of these transporters in biliary sterol secretion.105,106 Single-knockout mice for both proteins have also been constructed107,108 and share the phenotype with the double knockout, proving that the half-transporters need each other to display functional activity.106 Attempts to complement activity with other members of the ABCG family have been unsuccessful. The origin of the non-Abcg5/g8-dependent cholesterol secretion observed in knockout mice is not yet clear, but may be induced via direct micellization from the membrane by mixed bile salt/phospholipid micelles.109 In model membranes the flip-flop rates of cholesterol are high. It has therefore been speculated that Abcg5/g8 do not flip the sterol but act to lift the molecule somewhat up in the membrane, thereby decreasing the activation energy necessary for uptake. Whether such a mechanism is really involved remains to be seen. ABCG1 is also expressed in the liver, although under most conditions primarily in Kupffer cells.110 However, feeding mice a western-type diet up-regulated the protein, and it may play a role in intracelluar cholesterol trafficking. Adenoviral overexpression of the protein in the liver increased biliary cholesterol secretion.111 Despite intense research, the factors involved in intracellular transport of cholesterol and phospholipid remain enigmatic. Many proteins have been proposed to play a role, e.g. sterol carrier protein-2, Niemann–Pick C proteins 1 and 2, caveolin 1 and 2, and phosphatidylcholine transfer protein. Evidence for a direct role of any of these proteins in lipid trafficking to the canalicular membrane is lacking.
REGULATION OF BILE SECRETION Hepatocytes are strictly polarized cells. They absorb substrates from the blood and secrete metabolites into the bile. Bile flow depends
on the absorption of substrates and secretion of metabolites, in particular bile acids. As mentioned above, the supply of bile acids is variable. Absorption from the portal venous blood is almost complete, and in the fasting state occurs mainly in periportal hepatocytes. The main potential for regulation at the sinusoidal membrane is the involvement of more hepatocytes, downstream in the liver acinus, when the bile acid concentration in portal venous blood suddenly increases, as occurs during feeding. It is important to note that Ntcp, the bile acid uptake transporter, is evenly distributed along the hepatic acinus.25 Thus, perivenous hepatocytes which are commonly exposed to low concentrations of bile acids have a similar Ntcp expression to periportal hepatocytes, which are exposed to high bile acid concentrations. Bile acids are potentially cytotoxic and at high concentrations can induce apoptosis and necrosis.112 Binding to intracellular binding proteins, storage in the ER, and activation of NF-kB are possible defense mechanisms.113 Most important, however, is rapid canalicular secretion, which is in balance with sinusoidal absorption. Many high-affinity ligands of the nuclear hormone receptor (NHR) family of transcription factors are also substrates for ABC transporters (Figure 5-2). This relation is important for the physiological regulation of ABC transporter genes and other NHR-target genes in vivo. However, in liver disease this regulatory cross-talk may be disturbed because of the acute-phase response-coupled downregulation of NHRs and their target genes. Infection, inflammation, and trauma induce a wide array of metabolic changes in the liver that constitute the acute-phase response, mediated by cytokines, particularly TNF-a, IL-1b, and IL-6. For example, in fulminant hepatic failure serum levels of TNF-a and TNF receptors are significantly increased. In livers of patients with fulminant hepatic failure, infiltrating mononuclear cells express high amounts of TNFa and hepatocytes overexpress TNF receptor 1 (TNF-R1).114 Acutephase response is associated with a decrease in mRNAs coding for
75
Section I. Pathophysiology of the Liver
certain NHR proteins, such as RXR, LXR, PPAR-a and PPAR-g.115 Reduction of RXR levels, along with levels of other nuclear hormone receptors in the liver, could be a mechanism to coordinately downregulate the expression of a large number of genes, including ABC transporters, during the acute-phase response. Down-regulation of specific hepatic nuclear factors, such as HNF1 and HNF4, may play a key role in the regulation of certain negative acute-phase proteins. For example, a decrease in HNF1 is thought to be responsible for the reduced transcription of albumin and Ntcp. The acute-phase response also causes marked alterations in lipid metabolism in the liver. Many of the enzymes and transporters involved in these metabolic changes are known to be regulated by PPAR-a or LXRa. It is possible that during the acute-phase response the reduced availability of RXR protein, and possibly of NHRs, represents a mechanism to coordinately regulate these metabolic changes. In addition, the importance of RXRs for liver gene expression has been demonstrated.116 Biochemical parameters indicate that PPAR-a, CAR, PXR, LXR, and FXR-coupled metabolic pathways in the liver were compromised in the absence of RXRa. Thus, RXRa is integrated into a number of diverse physiological pathways as a common regulatory component of cholesterol, fatty acid, bile salt, steroid, and xenobiotic metabolism and homeostasis. FXR acts as a bile salt sensor: it needs to be activated by its natural ligands, the bile salts. FXR controls several key steps in bile salt metabolism: not only bile salt synthesis, but also transporters involved in the enterohepatic cycle. The liver plays a dominant role in the enterohepatic cycling of bile salts. In the liver bile acids are synthesized de novo from cholesterol via the so-called neutral and acidic biosynthetic pathways. CYP7A1 (cholesterol 7a-hydroxylase) is the gatekeeper of the neutral pathway. This enzyme and an enzyme more downstream in the neutral pathway, sterol 12a-hydroxylase, CYP8B1, are under the transcriptional control of FXR. In the regulation of these enzymes FXR acts indirectly through the action of at least two other transcription factors.117 FXR is a ligand-activated transcription factor. Chenodeoxycholic acid, cholic acid, deoxycholic acid and lithocholic acid bind and activate FXR. FXR complexes with the retinoid X-receptor, and this FXR–RXR heterodimer interacts with a highly conserved IR-1 motif (inverse repeat-1) in the promoter region of, for example, BSEP and SHP-1 (small heterodimer partner-1).79,118 SHP-1 suppresses the transcription of CYP7A1 and CYP8B1 by binding to a transcription factor known as liver receptor homolog 1.15,119 In addition, SHP-1 suppresses the expression of NTCP. The presumed mechanism is by interfering with RXR–RAR binding to the NTCP promotor.120 In contrast, FXR upregulates the expression of BSEP in the liver and the bile salt-binding protein in the ileum (iBABP). Studies in mice with a genetic disruption of FXR showed that the FXR response is particularly important in dealing with a bile salt load, such as occurs when feeding mice a high cholesterol- or cholate-containing diet. In FXR-null mice the expression of Ntcp, CYP7A1 and CYP8B1 fails to be down-regulated and the expression of Bsep, iBABP and SHP-1 is not enhanced, as occurs in wild-type mice under these conditions.121 In rats cholestasis is associated with a decreased expression of Ntcp.122 This is most probably caused by enhanced expression of SHP-1 through activation of FXR by retained bile salts. Also in humans, NTCP expression in cholestatic liver disease is decreased.123 Down-regulation of NTCP and CYP7A1 in chole-
76
static liver disease may be cytoprotective, reducing the entry and synthesis of bile acids when intracellular bile acid levels are already elevated. Canalicular Mrp2 is rapidly down-regulated in LPS- and bile duct ligation-induced cholestasis in rats, whereas Bsep expression is maintained.68,75,124 Also, during cholestasis in humans MRP2 is downregulated.78 Mrp3 and Mrp4 in the basolateral membrane is up-regulated under these conditions.86,88 Together these two proteins cover the entire spectrum of Mrp2 substrates, and so can fully compensate for the decreased canalicular Mrp2 activity. Mrp3 mediates the transport of non-bile acid glucuronides and glutathione conjugates, and Mrp4 that of bile salt sulfates. Mrp4 functions as a basolateral bile salt conjugate and glutathione co-transporter.93 Up-regulation of Mrp3 and Mrp4 has a cytoprotective function. Metabolites are cleared from the hepatocyte via basolateral membrane pumps when exit via the canalicular membrane is not possible. The Mrp3 and Mrp4 genes are controlled and activated by PXR and CAR.125,126 CAR activates not only Mrp4 but also the sulfotransferase Sult2a, which mediates the sulfation of bile salts, the high-affinity substrates of Mrp4.125
GENETIC DEFECTS OF BILE SECRETION The spectrum of diseases caused by defects of ABC transporter proteins is diverse and includes liver diseases: progressive familial intrahepatic cholestasis,127 benign recurrent intrahepatic cholestasis,128 intrahepatic cholestasis of pregnancy,129 cystic fibrosis,130 adrenoleukodystrophy,131 Dubin–Johnson syndrome;132,133 various eye disorders;134 disorders of cholesterol and carbohydrate metabolism; and connective tissue diseases.135–137 Progressive familial intrahepatic cholestasis (PFIC) constitutes a group of autosomal recessive diseases characterized by cholestasis starting in infancy. For a first differentiation of various PFIC subtypes measurement of the serum g-glutamyltransferase (GGT) activity is useful. Diseases associated with low bile salt concentration in bile have a low serum GGT activity. These are PFIC types 1 and 2 and BRIC types 1 and 2. These diseases have an intrahepatocellular blockade of bile salt secretion and could be called intrahepatocellular cholestasis rather than intrahepatic cholestasis. GGT in human liver is located mainly in the membranes lining the biliary tree. Elevation of serum GGT results from a detergent, membranolytic effect of bile salts on these membranes. Thus either a blockade of bile flow downstream of the location of GGT, or bile containing bile salts not antagonized by neutralizing phosphatidylcholine, causes GGT to be released in the circulation. Elevated serum gamma-GT occurs in various forms of intrahepatic and extrahepatic cholestasis. In PFIC type 3 GGT is elevated.
PROGRESSIVE FAMILIAL INTRAHEPATIC CHOLESTASIS TYPE 1 PFIC type 1 or Byler disease (Figure 5-3; Table 5-2) often begins with episodes of cholestasis, progressing to permanent cholestasis with fibrosis, cirrhosis and liver failure necessitating liver transplantation in the first two decades of life.138,139 Children with PFIC are small for their age and often have diarrhea. Rarely they also have
Chapter 5 MECHANISMS OF BILE SECRETION
MRP3 DJS Bilirubin
B
TJP2 FHCA MRP2
Bgluc
FIC 1 PFIC 1, BRIC 1 PS, PE
UGT1A1 PFIC 2, BRIC 2 BSEP T/G-BA
PC MDR3
FHCA BAAT Bile acids
BA MRP4
pancreatitis, and occasionally hearing loss.140 The larger bile ducts are anatomically normal and liver histology shows bland canalicular cholestasis without much bile duct proliferation, inflammation, fibrosis or cirrhosis.138 On electron microscopy there is a paucity of canalicular microvilli, a thickened pericanalicular network of microfilaments within the canaliculi, and coarse granular bile called ‘Byler bile’. Characteristically the serum GGT activity is not elevated, whereas parameters of cholestasis, such as alkaline phosphatase and serum primary bile acids (in particular chenodeoxycholic acid), are strongly increased. Serum cholesterol levels are usually normal. Patients belonging to the Byler kindred are descendants of Jacob and Nancy Byler, who in the late 18th century emigrated from Germany to the United States. Many patients outside the United States are unrelated to the Amish. The PFIC syndrome has been described in families in The Netherlands, Sweden, Greenland, and an Arab population.138 In Amish and non-Amish families the genetic defect could be mapped to the FIC1 locus on chromosome 18q21q22 encoding a P-type ATPase ATP8B1. The ATP8B1 protein is an aminophospholipid flipase in the apical membrane of enterocytes, pancreatic acinar cells, gastric pit epithelial cells, hepatocytes and cholangiocytes.141 ATP8B1 translocates phosphatidylserine and phosphatidylethanolamine from the outer to the inner layer of the plasma membrane.142 How dysfunction of this protein relates to cholestasis is unknown. One may speculate that its dysfunction causes alterations in membrane composition, with dysfunctioning transporters as a result. Knockout mice with a disruption of the ATP8B1 gene do not display a cholestatic phenotype and thus are quite different from the human disease model.143 A number of FIC1 mutations have been described, including the mutation causing Byler’s disease.139 In humans, FIC1 is highly expressed in pancreas, small intestine, urinary bladder, stomach and prostate. This may explain the increased frequency of diarrhea and
PFIC 3
Figure 5-3. Genetic cholestasis. PFIC type 1 and BRIC type 1 are caused by mutations of the FIC 1 gene. FIC 1 acts as an aminophospholipid translocator in the canalicular membrane of hepatocytes and the apical membrane of enterocytes. PS, phosphatidylserine; PE, phosphatidylethanol amine. PFIC type 2 and BRIC type 2 are caused by mutations of ABCB11/BSEP. PFIC type 3 is caused by mutations of ABCB4/MDR3. Mutations of the same gene are also responsible for intrahepatic gallstone formation and subgroups of intrahepatic cholestas is of pregnancy. ABCC2/MRP2 is mutated in Dubin–Johnson syndrome, a rare disease characterized by conjugated hyperbilirubinemia. B, bilirubin; Bgluc, bilirubin glucuronides; UGT1A1, bilirubin UDP glucuronosyltransferase. Familial hypercholanemia (FHCA) is caused by gene mutations affecting either BAAT, the bile acid conjugating enzyme (bile acid CoA:amino acid N-acyltransferase) or TJP2 (ZO-2), the tight junction protein-2. BA, bile acids; T/G-BA, bile acid taurine and glycine conjugates.
pancreatitis in these patients. The lack of catch-up growth after transplantation may be due to the persistent malabsorption.144 Dysfunction of FXR signaling has been reported in this disease.145 Whether this is a primary defect or secondary to bile acid malabsorption needs to be studied. FXR reduces lipogenesis, via repression of Srebp1c by SHP-1.146 After transplantation patients develop steatosis of the liver graft.144 The reported defective FXR function in PFIC type 1 patients may thus explain the steatosis of the transplanted liver. Some children with this disease respond to ursodeoxycholic acid therapy.147 In non-cirrhotic patients partial external biliary diversion by cholecystocutaneostomy relieves itching and improves cholestatic markers.148 The mechanism of this therapy is not well understood. It may be partially explained by a decrease in the overall bile salt pool. In some patients an improvement of liver morphology and a normalization of biliary bile acid composition was seen, suggesting improved bile acid secretion.149
BENIGN RECURRENT INTRAHEPATIC CHOLESTASIS TYPE 1 Recurrent familial intrahepatic cholestasis is a term coined by Tygstrup et al.150 This disease is also known as benign recurrent intrahepatic cholestasis (BRIC) or Summerskill syndrome151 (Table 5-2). Despite recurrent attacks of cholestasis there is usually no progression to chronic liver disease. During attacks the patients are severely jaundiced and have pruritus, steatorrhea and weight loss. In analogy to PFIC 1 the serum GGT is not elevated. Some patients also have renal stones, pancreatitis and diabetes.150 As in PFIC 1 the gene involved in recurrent familial intrahepatic cholestasis has been mapped to the FIC1 locus.128 This suggests that the two diseases are genetically related. However, the defect could not be traced to chromosome 18 mutations in all patients with BRIC.152
77
Section I. Pathophysiology of the Liver
Table 5-2. Genetic Diseases of Hepatic Transport Disease
Chromosome
Gene/function
Defect
Phenotype
Therapy
PFIC1 Progressive familial intrahepatic cholestasis type 1
18q21
FIC1 Aminophospholipid translocator
Hepatocellular cholestasis, pathogenesis unclear
Ursodeoxycholic acid, Partial external biliary diversion, Liver transplantation
BRIC 1 Benign recurrent intrahepatic cholestasis type 1 PFIC2 Progressive familial intrahepatic cholestasis type 2
18q21
FIC1 Aminophospholipid translocator
Ibid
First recurrent, later permanent and progressive cholestasis, no bile duct proliferation, normal GGT, diarrhea, pancreatitis, coarse granular bile on transmission EM Episodic cholestasis with pruritus, weight loss and steatorrhoea, normal GGT
2q24
ABCB11 (BSEP) Bile salt export pump
Hepatocellular cholestasis due to deficient canalicular bile salt transport
BRIC 2 Benign recurrent intrahepatic cholestasis type 2 PFIC3 Progressive familial intrahepatic cholestasis type 3
2q24
ABCB11 (BSEP) Bile salt export pump
Ibid
7q21
ABCB4 (PGY3, MDR 3), Phosphatidylcholine flippase
Intrahepatic cholestasis due to deficient canalicular phospholipid transport
ICP Intrahepatic cholestasis of pregnancy Intrahepatic gallstones Familial hypercholanemia
7q21
ABCB4 Phosphatidylcholine flippase
9q12-q139q22.3
TJP2/ZO-2, Tight junction protein BAAT, Bile acid CoA:amino acid N-acyltransferase 3b-D5-C27-hydroxysteroid oxidoreductase D4-3-oxosteroid-5b reductase 3b-hydroxy C27 steroid dehydrogenase/isomerase oxysterol 7alphahydroxylase (CYP7B1) ABCC2 (MRP2, cMOAT) Canalicular multispecific organic anion transporter
Bile acid biosynthesis defects
Dubin–Johnson syndrome
10q24
Cholestasis, jaundice less prominent, extensive bile duct proliferation and periportal fibrosis, elevated GGT Cholestasis in third trimester of pregnancy. High fecal loss. Elevated GGT
Partial external biliary diversion
Ursodeoxycholic acid, Liver transplantation
Ursodeoxycholic acid
Leaky tight junctions deficient bile salt conjugation
Elevated serum bile acids, pruritus, malabsorption
Abnormal bile salts inhibit bile salt transport
Intrahepatic cholestasis, neonatal giant cell hepatitis
Cholic acid
Deficient canalicular transport of bilirubin conjugates
Conjugated hyperbilirubinemia, increased urinary coproporphyrin isomer I, hepatic lysosomal pigment
Not needed
Ursodeoxycholic acid is of no benefit in BRIC.153 Case reports indicate that rifampicin may reduce the number of cholestatic episodes.154,155 In analogy to PFIC type 1, cholestasis may be improved and cholestatic episodes perhaps shortened by biliary drainage procedures. However, most reports regarding this therapy relate to PFIC type 1 rather than to BRIC.148,149
78
Neonatal hepatitis, progressive cholestasis, no bile duct proliferation, lobular and portal fibrosis, normal GGT, amorphous bile on transmission EM, BSEP protein absent Episodic cholestasis with pruritus, weight loss and steatorrhoea, normal GGT
Rifamipicin Ursodeoxycholic acid Fat soluble vitamine suppletion Partial external biliary diversion, Liver transplantation
PROGRESSIVE FAMILIAL INTRAHEPATIC CHOLESTASIS TYPE 2 Genetic studies revealed that the FIC1 locus is not involved in all patients with a PFIC type 1 phenotype and low serum GGT.138 Moreover, in a large number of non-Amish patients the disease was mapped to a locus on chromosome 2q24 which later proved to be
Chapter 5 MECHANISMS OF BILE SECRETION
the ABCB11 (BSEP) gene (see earlier and Table 5-2) (Figure 5-3).72,156,157 Antibodies directed against BSEP sequences enabled localization studies, and it became clear that this protein is not only liver specific but is located in the canalicular domain of the hepatocyte’s plasma membrane. Liver specimens of patients with PFIC type 2 stain negative for canalicular BSEP on immunohistochemistry using BSE antibodies.157 As in PFIC type 1, the serum GGT activity in these patients is not elevated and bile duct proliferation is absent. However, the disease differs from PFIC type 1 in several respects: PFIC2 frequently presents as non-specific giant cell hepatitis, which is undistinguishable from idiopathic neonatal giant cell hepatitis; patients are usually permanently jaundiced and the disease rapidly progresses to persistent and progressive cholestasis, requiring liver transplantation. The liver histology shows more inflammatory activity than in PFIC type 1, with giant cell transformation, and lobular and portal fibrosis.138 The bile of PFIC type 2 patients is amorphous or filamentous on transmission electron microscopy. This contrasts with the coarsely granular bile of PFIC type 1 patients. Extrahepatic manifestations are uncommon. In the majority of nonAmish patients, progressive familial intrahepatic cholestasis is type 2 rather than type 1. A particularly dreaded complication is cerebral or subdural hematoma at or shortly after birth as a result of vitamin K deficiency. Vitamin K must therefore be supplemented without delay. Bile acids are not completely absent in the bile of these patients. MRP2, the canalicular transporter of bilirubin, also transports glucuronidated or sulfated bile acids. This may also explain why these patients are jaundiced despite an intact bilirubin transporter: bilirubin transport may be inhibited by competition with bile acid conjugates. PFIC type 2 patients usually need to be transplanted in the first two decades. Living related donor transplantation should be considered cautiously, as parents may be carriers of the disease, which may become manifest after transplantation. Partial bile diversion may bring symptomatic relief of pruritus in these patients, cause amelioration of liver functions and induce catch-up growth.148 The majority of PFIC type 2 patients do not respond to ursodeoxycholic acid therapy: in fact, administration of ursodeoxycholic acid to some of these patients led to very high serum bile acid levels (>1 mmol/l) without any increase in biliary bile acid secretion.157
BENIGN RECURRENT INTRAHEPATIC CHOLESTASIS TYPE 2 Not all patients with BRIC have mutations of ATP8B1. In a subset of patients with episodic cholestasis, mutations of BSEP were found. The disease was called BRIC type 2. It appears that these patients are particularly prone to the development of cholelithiasis and less to pancreatitis. This distinguishes them from patients with BRIC type 1. Serum GGT levels are low in both diseases. Patients with BRIC type 1 can be completely asymptomatic between attacks of cholestasis, but whether this is also true for patients with BRIC type 2 needs to be studied.158
FAMILIAL HYPERCHOLANEMIA Familial hypercholanemia is characterized by elevated serum bile salt levels, severe pruritus and fat malabsorption.159 So far this
disease has been identified among Amish individuals. It was originally thought to result from a sinusoidal uptake defect. Recently it has been reported to be caused by mutations of one of two genes, one that encodes tight junction protein 2 (ZO-2) and BAAT, which encodes bile acid coenzyme A: amino acid N-acyltransferase. In these latter patients glycine and taurine bile acid conjugates cannot be formed.160
BILE ACID SYNTHESIS DEFECTS Defects of bile acid synthesis resemble PFIC type 2. Clayton et al.161 described a defect of 3b-D5-C27-hydroxysteroid oxidoreductase as a cause of giant cell hepatitis. Deficiency of D4-3-oxosteroid-5b reductase and 3b-hydroxy C27steroid dehydrogenase/isomerase and mutations of the oxysterol 7a-hydroxylase gene may also be causes of neonatal hepatitis and cholestasis.162–164 In these diseases toxic intermediates are formed which cause cholestasis by interaction with the hepatic bile acid transporter.165 Bile acid synthesis defects are called PFIC type 4 by some authors.
PROGRESSIVE FAMILIAL INTRAHEPATIC CHOLESTASIS TYPE 3 The third PFIC subtype, PFIC type 3, is quite different from the others. The serum GGT activity is markedly elevated in these patients and liver histology shows extensive bile duct proliferation, and portal and periportal fibrosis.64,166,167 Phenotypically PFIC type 3 resembles the Mdr2(-/-) mice. In humans with PFIC type 3, mutations of the ABCB4 (MDR3) gene are the underlying cause.129,167,168 Phosphatidylcholine, the main phospholipid in bile, is washed down from the canalicular membrane by bile acids (Figure 5-4A). In contrast to PFIC type 2, in PFIC type 3 bile acid transport proceeds unimpaired, but this occurs without phospholipids because of the MDR3 deficiency. This has major pathophysiologic consequences. In normal bile the inherent toxicity of bile acids is quenched by phosphatidylcholine. In the bile of PFIC 3 patients bile acid toxicity is unantagonized by phospholipids, and this causes damage to bile duct epithelium, with periportal inflammation and fibrosis and bile duct proliferation as a result (Figure 5-4B). In humans this is even more extreme than in Mdr2(-/-) mice, as human bile acids (e.g. chenodeoxycholic acid) are more toxic than those of the mouse – mainly the very hydrophilic muricholate. In patients with PFIC type 3 symptoms present somewhat later in life than in PFIC types 1 and 2, and liver failure also occurs at a later age. Jaundice may be less apparent, but pruritus is usually severe. Patients with a partial ABCB4 (MDR3) defect respond to ursodeoxycholic acid therapy.147 The majority of patients, however, have to be transplanted. Mutations of the ABCB4 gene on chromosome 7q21 are the underlying cause of the disease. Although PFIC3 is discussed as a cholestatic disease, in a strictly physiologic sense there is no cholestasis, as the bile flow is not impaired.58
INTRAHEPATIC CHOLESTASIS OF PREGNANCY Jacquemin et al.168 reported a high incidence of intrahepatic cholestasis of pregnancy (ICP) in families with PFIC type 3. This suggests that in persons carrying one mutated ABCB4 gene, cholestasis may occur during pregnancy. Mutations leading to single
79
Section I. Pathophysiology of the Liver
Cholesterol BSEP
ABC G5/G8
Bile acids
Bilirubin
MRP2 MDR3
Phosphatidyl choline
related to ATP8B1 serum GGT is not elevated. Ursodeoxycholic acid has been shown to be of benefit in patients with ICP, with a reduction in fetal loss.173,174
OTHER FORMS OF INTRAHEPATIC CHOLESTASIS More forms of intrahepatic cholestasis exist. Aageneas syndrome is a combination of severe progressive lymphedema and episodic intrahepatic cholestasis.175 The locus for this disease has been mapped to chromosome 15q.176
DUBIN–JOHNSON SYNDROME GGT
GGT GGT
GGT A
Cholesterol BSEP
Bilirubin
ABC G5/G8
BA
BA BA BA
Bile acids
Phosphatidyl choline MDR3
MRP2
GGT GGT GGT
GGT
REFERENCES
BA BA BA BA BA BA BA
GGT GGT GGT
GGT GGT
GGT
B Figure 5-4. Hepatocanalicular secretion of bile salts, phosphatidylcholine and cholesterol. A The toxicity of bile salts is neutralized in mixed micelles by the presence of phosphatidylcholine and cholesterol. B In progressive familial cholestasis type 3 bile salt secretion is unimpaired but the secretion of phosphatidylcholine is strongly reduced. Thus, bile salt toxicity is unantagonized by the lack of phosphatidylcholine, and this affects the cell integrity of the biliary epithelium. Bile duct epithelial cell proliferation, periportal inflammation and fibrosis and elevation of g-glutamyltransferase is a result of this.
amino acid substitutions of the MDR3 protein may cause intracellular traffic mutants, that is, the protein is synthesized but does not reach its destination in the canalicular membrane.169 One can hypothesize that in patients carrying these mutations the hormones in the third trimester of pregnancy impair the intracellular targeting that causes the disease to become clinically manifest. ICP has also been described in families with PFIC type 1.170 BSEP (ABCB11) polymorphisms seem to be of less importance for ICP.171 ICP not related to MDR3 or ATP8B1 has been reported in a Finnish group of patients.172 In contrast to ICP related to ABCB4, in ICP
80
Dubin–Johnson syndrome is described here not because it is an important cholestatic disease, but because it is caused by a mutation of ABCC2 encoding MRP2 (Table 5-2).133,177 Dubin–Johnson syndrome is characterized by conjugated hyperbilirubinemia without other serum enzyme abnormalities. Patients have a normal lifespan. A black or brownish lysosomal pigment in the hepatocytes is a characteristic histological feature. The excretion of urinary coproporphyrin isomer I is elevated in these patients. TR- and EHBR rats are animal models for this disease. These animals have a decreased hepatobiliary secretion of organic anions because of a mutation of the Abcc2 gene.81,178 Patients with Dubin–Johnson syndrome are homozygous carriers of ABCC2 gene mutations. Rapid degradation of mutated ABCC2 mRNA, or impaired MRP2 protein maturation and trafficking, may be the underlying cause of the disease.179 Dubin–Johnson syndrome has no influence on longevity. The disease needs no treatment.
1. Rappaport AM. The structural and functional unit in the human liver (liver acinus). Anat Rec 1958;130:673–689. 2. Lamers WH, Hilberts A, Furt E, et al. Hepatic enzymic zonation: a reevaluation of the concept of the liver acinus. Hepatology 1989;10:72–76. 3. Theise ND, Saxena R, Portmann BC, et al. The canals of Hering and hepatic stem cells in humans. Hepatology 1999;30:1425–1433. 4. Gatmaitan Z, Varticovski L, Ling L, et al. Studies on fenestral contraction in rat liver endothelial cells in culture. Am J Pathol 1996;148:2027–2041. 5. Bhunchet E, Fujieda K. Capillarization and venularization of hepatic sinusoids in porcine serum-induced rat liver fibrosis: a mechanism to maintain liver blood flow. Hepatology 1993;18:1450–1458. 6. Hardison WG, Dalle-Molle E, Gosink E, et al. Function of rat hepatocyte tight junctions: studies with bile acid infusions. Am J Physiol 1991;260:G167–G174. 7. Mitic LL, Van Itallie CM, Anderson JM. Molecular physiology and pathophysiology of tight junctions I. Tight junction structure and function: lessons from mutant animals and proteins. Am J Physiol Gastrointest Liver Physiol 2000;279:G250–G254. 8. Anderson JM. Leaky junctions and cholestasis: a tight correlation. Gastroenterology 1996;110:1662–1665. 9. Hayakawa T, Bruck R, Ng OC, Boyer JL. DBcAMP stimulates vesicle transport and HRP excretion in isolated perfused rat liver. Am J Physiol 1990;259:G727–G735. 10. Zegers MM, Zaal KJ, Hoekstra D. Functional involvement of proteins, interacting with sphingolipids, in sphingolipid transport to the canalicular membrane in the human hepatocytic cell line, HepG2? Hepatology 1998;27:1089–1097.
Chapter 5 MECHANISMS OF BILE SECRETION
11. Keppler D, Konig J. Hepatic secretion of conjugated drugs and endogenous substances. Semin Liver Dis 2000;20:265–272. 12. Kullak-Ublick GA, Stieger B, Hagenbuch B, Meier PJ. Hepatic transport of bile salts. Semin Liver Dis 2000;20:273–292. 13. Jones AL, Hradek GT, Renston RH, et al. Autoradiographic evidence for hepatic lobular concentration gradient of bile acid derivative. Am J Physiol 1980;238:G233–G237. 14. Groothuis GM, Hardonk MJ, Keulemans KP, et al. Autoradiographic and kinetic demonstration of acinar heterogeneity of taurocholate transport. Am J Physiol 1982;243:G455–G462. 15. Goodwin B, Jones SA, Price RR, et al. A regulatory cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol Cell 2000;6:517–526. 16. Ugele B, Kempen HJ, Kempen JM, et al. Heterogeneity of rat liver parenchyma in cholesterol 7 alpha-hydroxylase and bile acid synthesis. Biochem J 1991;276:73–77. 17. Marzioni M, Glaser SS, Francis H, et al. Functional heterogeneity of cholangiocytes. Semin Liver Dis 2002;22:227–240. 18. Gurantz D, Schteingart CD, Hagey LR, et al. Hypercholeresis induced by unconjugated bile acid infusion correlates with recovery in bile of unconjugated bile acids. Hepatology 1991;13:540–550. 19. Ballatori N, Jacob R, Barrett C, Boyer JL. Biliary catabolism of glutathione and differential reabsorption of its amino acid constituents. Am J Physiol 1988;254:G1–G7. 20. Park HZ, Lee SP, Schy AL. Ceftriaxone-associated gallbladder sludge. Identification of calcium–ceftriaxone salt as a major component of gallbladder precipitate. Gastroenterology 1991;100:1665–1670. 21. Walters HC, Craddock AL, Fusegawa H, et al. Expression, transport properties, and chromosomal location of organic anion transporter subtype 3. Am J Physiol Gastrointest Liver Physiol 2000;279:G1188–G1200. 22. Dawson PA, Hubbert M, Haywood J, et al. The heteromeric organic solute transporter alpha-beta, Ostalpha-Ostbeta, is an ileal basolateral bile acid transporter. J Biol Chem 2005;280: 6960-6968. 23. Voshol PJ, Minich DM, Havinga R, et al. Postprandial chylomicron formation and fat absorption in multidrug resistance gene 2 P-glycoprotein-deficient mice. Gastroenterology 2000;118:173–182. 24. Van der Veere CN, Jansen PL, Sinaasappel M, et al. Oral calcium phosphate: a new therapy for Crigler–Najjar disease? Gastroenterology 1997;112:455–462. 25. Stieger B, Hagenbuch B, Landmann L, et al. In situ localization of the hepatocytic Na+/Taurocholate co-transporting polypeptide in rat liver. Gastroenterology 1994;107:1781–1787. 26. Hagenbuch B, Scharschmidt BF, Meier PJ. Effect of antisense oligonucleotides on the expression of hepatocellular bile acid and organic anion uptake systems in Xenopus laevis oocytes. Biochem J 1996;316:901–904. 27. Hagenbuch B, Meier PJ. Sinusoidal (basolateral) bile salt uptake systems of hepatocytes. Semin Liver Dis 1996;16:129–136. 28. Guo GL, Staudinger J, Ogura K, Klaassen CD. Induction of rat organic anion transporting polypeptide 2 by pregnenolone16alpha-carbonitrile is via interaction with pregnane X receptor. Mol Pharmacol 2002;61:832–839. 29. Luo G, Guenthner T, Gan LS, Humphreys WG. CYP3A4 induction by xenobiotics: biochemistry, experimental methods and impact on drug discovery and development. Curr Drug Metab 2004;5:483–505. 30. Staudinger JL, Goodwin B, Jones SA, et al. The nuclear receptor PXR is a lithocholic acid sensor that protects against liver toxicity. Proc Natl Acad Sci USA 2001;98:3369–3374. 31. Cattori V, van Montfoort JE, Stieger B, et al. Localization of organic anion transporting polypeptide 4 (Oatp4) in rat liver and
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43. 44.
45.
46.
47. 48.
49.
50.
comparison of its substrate specificity with Oatp1, Oatp2 and Oatp3. Pflugers Arch 2001;443:188–195. Li N, Hartley DP, Cherrington NJ, Klaassen CD. Tissue expression, ontogeny, and inducibility of rat organic anion transporting polypeptide 4. J Pharmacol Exp Ther 2002;301:551–560. Kullak-Ublick GA, Ismair MG, Stieger B, et al. Organic aniontransporting polypeptide B (OATP-B) and its functional comparison with three other OATPs of human liver. Gastroenterology 2001;120:525–533. Cui Y, Konig J, Leier I, et al. Hepatic uptake of bilirubin and its conjugates by the human organic anion transporter SLC21A6. J Biol Chem 2001;276:9626–9630. Nozawa T, Minami H, Sugiura S, et al. Role of organic anion transporter OATP1B1 (OATP-C) in hepatic uptake of irinotecan and its active metabolite, 7-ethyl-10-hydroxycamptothecin: in vitro evidence and effect of single nucleotide polymorphisms. Drug Metab Dispos 2005;33:434–439. Shitara Y, Sato H, Sugiyama Y. Evaluation of drug–drug interaction in the hepatobiliary and renal transport of drugs. Annu Rev Pharmacol Toxicol 2005;45:689–723. Koepsell H. Polyspecific organic cation transporters: their functions and interactions with drugs. Trends Pharmacol Sci 2004;25:375–381. Zhang L, Dresser MJ, Gray AT, et al. Cloning and functional expression of a human liver organic cation transporter. Mol Pharmacol 1997;51:913–921. van Montfoort JE, Muller M, Groothuis GM, et al. Comparison of ‘type I’ and ‘type II’ organic cation transport by organic cation transporters and organic anion-transporting polypeptides. J Pharmacol Exp Ther 2001;298:110–115. Jonker JW, Wagenaar E, Mol CA, et al. Reduced hepatic uptake and intestinal excretion of organic cations in mice with a targeted disruption of the organic cation transporter 1 (Oct1 [Slc22a1]) gene. Mol Cell Biol 2001;21:5471–5477. Dean M, Rzhetsky A, Allikmets R. The human ATP-binding cassette (ABC) transporter superfamily. Genome Res 2001;11:1156–1166. Thiebaut F, Tsuruo T, Hamada H, et al. Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proc Natl Acad Sci USA 1987;84:7735–7738. Leveille-Webster CR, Arias IM. The biology of the Pglycoproteins. J Membrane Biol 1995;143:89–102. Schinkel AH, Wagenaar E, van Deemter L, et al. Absence of the mdr1a P-glycoprotein in mice affects tissue distribution and pharmacokinetics of dexamethasone, digoxin, and ciclosporin A. J Clin Invest 1995;96:1698–1705. Sharom FJ, Yu X, Lu P, et al. Interaction of the P-glycoprotein multidrug transporter (MDR1) with high affinity peptide chemosensitizers in isolated membranes, reconstituted systems, and intact cells. Biochem Pharmacol 1999;58:571–586. Muller M, Mayer R, Hero U, Keppler D. ATP-dependent transport of amphiphilic cations across the hepatocyte canalicular membrane mediated by mdr1 P-glycoprotein. FEBS Lett 1994;343:168–172. Borst P, Schinkel AH. Genetic dissection of the function of mammalian P-glycoproteins. Trends Genet 1997;13:217–222. Schinkel AH, Smit JJ, van Tellingen O, et al. Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in the blood–brain barrier and to increased sensitivity to drugs. Cell 1994;77:491–502. Mauad TH, van Nieuwkerk CM, Dingemans KP, et al. Mice with homozygous disruption of the mdr2 P-glycoprotein gene. A novel animal model for studies of nonsuppurative inflammatory cholangitis and hepatocarcinogenesis. Am J Pathol 1994;145: 1237–1245. Schinkel AH, Mayer U, Wagenaar E, et al. Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-
81
Section I. Pathophysiology of the Liver
51.
52.
53.
54.
55.
56.
57. 58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
82
transporting) P-glycoproteins. Proc Natl Acad Sci USA 1997;94:4028–4033. Eichelbaum M, Fromm MF, Schwab M. Clinical aspects of the MDR1 (ABCB1) gene polymorphism. Ther Drug Monit 2004;26:180–185. Thottassery JV, Zambetti GP, Arimori K, et al. p53-dependent regulation of MDR1 gene expression causes selective resistance to chemotherapeutic agents. Proc Natl Acad Sci USA 1997; 94:11037–11042. Zhou G, Kuo MT. Wild-type p53-mediated induction of rat mdr1b expression by the anticancer drug daunorubicin. J Biol Chem 1998;273:15387–15394. Vos TA, Ros JE, Havinga R, et al. Regulation of hepatic transport systems involved in bile secretion during liver regeneration in rats. Hepatology 1999;29:1833–1839. Ros JE, Schuetz JD, Geuken M, et al. Induction of Mdr1b expression by tumor necrosis factor-alpha in rat liver cells is independent of p53 but requires NF-kappaB signaling. Hepatology 2001;33:1425–1431. Synold TW, Dussault I, Forman BM. The orphan nuclear receptor SXR coordinately regulates drug metabolism and efflux. Nature Med 2001;7:584–590. Smit JJ, Schinkel AH, Mol CA, et al. Tissue distribution of the human MDR3 P-glycoprotein. Lab Invest 1994;71:638–649. Smit JJ, Schinkel AH, Oude Elferink RP, et al. Homozygous disruption of the murine mdr2 P-glycoprotein gene leads to a complete absence of phospholipid from bile and to liver disease. Cell 1993;75:451–462. Oude Elferink RP, Groen AK. Mechanisms of biliary lipid secretion and their role in lipid homeostasis. Semin Liver Dis 2000;20:293–305. Crawford JM, Mockel GM, Crawford AR, et al. Imaging biliary lipid secretion in the rat: ultrastructural evidence for vesiculation of the hepatocyte canalicular membrane. J Lipid Res 1995;36:2147–2163. Oude Elferink RP, Ottenhoff R, van Wijland M, et al. Uncoupling of biliary phospholipid and cholesterol secretion in mice with reduced expression of mdr2 P-glycoprotein. J Lipid Res 1996;37:1065–1075. Rosmorduc O, Hermelin B, Boelle PY, et al. ABCB4 gene mutation-associated cholelithiasis in adults. Gastroenterology 2003;125:452–459. Rosmorduc O, Hermelin B, Poupon R. MDR3 gene defect in adults with symptomatic intrahepatic and gallbladder cholesterol cholelithiasis. Gastroenterology 2001;120:1459–1467. Jacquemin E, De Vree JM, Cresteil D, et al. The wide spectrum of multidrug resistance 3 deficiency: from neonatal cholestasis to cirrhosis of adulthood. Gastroenterology 2001;120:1448–1458. Zollner G, Fickert P, Silbert D, et al. Adaptive changes in hepatobiliary transporter expression in primary biliary cirrhosis. J Hepatol 2003;38:717–727. Geuken E, Visser D, Kuipers F, et al. Rapid increase of bile salt secretion is associated with bile duct injury after human liver transplantation. J Hepatol 2004;41:1017–1025. Muller M. Transcriptional control of hepatocanalicular transporter gene expression. Semin Liver Dis 2000;20: 323–337. Vos TA, Hooiveld GJ, Koning H, et al. Up-regulation of the multidrug resistance genes, Mrp1 and Mdr1b, and downregulation of the organic anion transporter, Mrp2, and the bile salt transporter, Spgp, in endotoxemic rat liver. Hepatology 1998;28:1637–1644. Kok T, Wolters H, Bloks VW, et al. Induction of hepatic ABC transporter expression is part of the PPAR-a-mediated fasting response in the mouse. Gastroenterology 2003;124:160–171. Frijters CM, Ottenhoff R, van Wijland MJ, et al. Regulation of mdr2 P-glycoprotein expression by bile salts. Biochem J 1997;321:389–395.
71. Gerloff T, Stieger B, Hagenbuch B, et al. The sister of Pglycoprotein represents the canalicular bile salt export pump of mammalian liver. J Biol Chem 1998;273:10046–10050. 72. Strautnieks SS, Bull LN, Knisely AS, et al. A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis. Nature Genet 1998;20:233–238. 73. Jansen PL, Strautnieks SS, Jacquemin E, et al. Hepatocanalicular bile salt export pump deficiency in patients with progressive familial intrahepatic cholestasis. Gastroenterology 1999;117:1370–1379. 74. Wang R, Salem M, Yousef IM, et al. Targeted inactivation of sister of P-glycoprotein gene (spgp) in mice results in nonprogressive but persistent intrahepatic cholestasis. Proc Natl Acad Sci USA 2001;98:2011–2016. 75. Lee JM, Trauner M, Soroka CJ, et al. Expression of the bile salt export pump is maintained after chronic cholestasis in the rat. Gastroenterology 2000;118:163–172. 76. Trauner M, Arrese M, Lee H, et al. Endotoxin downregulates rat hepatic ntcp gene expression via decreased activity of critical transcription factors. J Clin Invest 1998;101:2092–2100. 77. Gerloff T, Geier A, Stieger B, et al. Differential expression of basolateral and canalicular organic anion transporters during regeneration of rat liver. Gastroenterology 1999;117:1408–1415. 78. Zollner G, Fickert P, Zenz R, et al. Hepatobiliary transporter expression in percutaneous liver biopsies of patients with cholestatic liver diseases. Hepatology 2001;33:633–646. 79. Ananthanarayanan M, Balasubramanian N, Makishima M, et al. Human bile salt export pump promoter is transactivated by the farnesoid X receptor/bile acid receptor. J Biol Chem 2001;276:28857–28865. 80. Claudel T, Sturm E, Kuipers F, Staels B. The farnesoid X receptor: a novel drug target? Expert Opin Investig Drugs 2004;13:1135–1148. 81. Paulusma CC, Bosma PJ, Zaman GJ, et al. Congenital jaundice in rats with a mutation in a multidrug resistance-associated protein gene. Science 1996;271:1126–1128. 82. Jansen PL, Peters WH, Lamers WH. Hereditary chronic conjugated hyperbilirubinemia in mutant rats caused by defective hepatic anion transport. Hepatology 1985;5:573–579. 83. Kast HR, Goodwin B, Tarr PT, et al. Regulation of multidrug resistance-associated protein 2 (ABCC2) by the nuclear receptors pregnane X receptor, farnesoid X-activated receptor, and constitutive androstane receptor. J Biol Chem 2002;277:2908–2915. 84. Tanaka T, Uchiumi T, Hinoshita E, et al. The human multidrug resistance protein 2 gene: functional characterization of the 5¢flanking region and expression in hepatic cells. Hepatology 1999;30:1507–1512. 85. Lee J, Azzaroli F, Wang L, et al. Adaptive regulation of bile salt transporters in kidney and liver in obstructive cholestasis in the rat. Gastroenterology 2001;121:1473–1484. 86. Ogawa K, Suzuki H, Hirohashi T, et al. Characterization of inducible nature of MRP3 in rat liver. Am J Physiol Gastrointest Liver Physiol 2000;278:G438–G446. 87. Donner MG, Keppler D. Up-regulation of basolateral multidrug resistance protein 3 (Mrp3) in cholestatic rat liver. Hepatology 2001;34:351–359. 88. Soroka CJ, Lee JM, Azzaroli F, Boyer JL. Cellular localization and up-regulation of multidrug resistance-associated protein 3 in hepatocytes and cholangiocytes during obstructive cholestasis in rat liver. Hepatology 2001;33:783–791. 89. Konig J, Rost D, Cui Y, Keppler D. Characterization of the human multidrug resistance protein isoform MRP3 localized to the basolateral hepatocyte membrane. Hepatology 1999;29:1156–1163. 90. Hirohashi T, Suzuki H, Ito K, et al. Hepatic expression of multidrug resistance-associated protein-like proteins maintained
Chapter 5 MECHANISMS OF BILE SECRETION
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109. 110.
in eisai hyperbilirubinemic rats. Mol Pharmacol 1998; 53:1068–1075. Cherrington NJ, Slitt AL, Li N, Klaassen CD. Lipopolysaccharide-mediated regulation of hepatic transporter mRNA levels in rats. Drug Metab Dispos 2004;32:734–741. Guo GL, Lambert G, Negishi M, et al. Complementary roles of farnesoid X receptor, pregnane X receptor, and constitutive androstane receptor in protection against bile acid toxicity. J Biol Chem 2003;278:45062–45071. Rius M, Nies AT, Hummel-Eisenbeiss J, et al. Co-transport of reduced glutathione with bile salts by MRP4 (ABCC4) localized to the basolateral hepatocyte membrane. Hepatology 2003; 38:374–384. Assem M, Schuetz EG, Leggas M, et al. Interactions between hepatic Mrp4 and Sult2a as revealed by the constitutive androstane receptor and Mrp4 knockout mice. J Biol Chem 2004;279:22250–22257. Bodzioch M, Orso E, Klucken J, et al. The gene encoding ATPbinding cassette transporter 1 is mutated in Tangier disease. Nature Genet 1999;22:347–351. Basso F, Freeman L, Knapper CL, et al. Role of the hepatic ABCA1 transporter in modulating intrahepatic cholesterol and plasma HDL cholesterol concentrations. J Lipid Res 2003;44:296–302. Brewer HB Jr, Remaley AT, Neufeld EB, et al. Regulation of plasma high-density lipoprotein levels by the ABCA1 transporter and the emerging role of high-density lipoprotein in the treatment of cardiovascular disease. Arterioscler Thromb Vasc Biol 2004;24:1755–1760. Oram JF. HDL apolipoproteins and ABCA1: partners in the removal of excess cellular cholesterol. Arterioscler Thromb Vasc Biol 2003;23:720–727. Groen AK, Bloks VW, Bandsma RH, et al. Hepatobiliary cholesterol transport is not impaired in Abca1-null mice lacking HDL. J Clin Invest 2001;108:843–850. Chawla A, Boisvert WA, Lee CH, et al. A PPAR gamma-LXRABCA1 pathway in macrophages is involved in cholesterol efflux and atherogenesis. Mol Cell 2001;7:161–171. Chinetti G, Lestavel S, Bocher V, et al. PPAR-alpha and PPARgamma activators induce cholesterol removal from human macrophage foam cells through stimulation of the ABCA1 pathway. Nature Med 2001;7:53–58. Jonker JW, Merino G, Musters S, et al. The breast cancer resistance protein BCRP (ABCG2) concentrates drugs and carcinogenic xenotoxins into milk. Nature Med 2005;11: 127–129. Schinkel AH, Jonker JW. Mammalian drug efflux transporters of the ATP binding cassette (ABC) family: an overview. Adv Drug Del Rev 2003;55:3–29. Hubacek JA, Berge KE, Cohen JC, Hobbs HH. Mutations in ATP-cassette binding proteins G5 (ABCG5) and G8 (ABCG8) causing sitosterolemia. Hum Mutat 2001;18:359–360. Yu L, Gupta S, Xu F, et al. Expression of ABCG5 and ABCG8 is required for regulation of biliary cholesterol secretion. J Biol Chem 2005;280:8742-8747. Graf GA, Cohen JC, Hobbs HH. Missense mutations in ABCG5 and ABCG8 disrupt heterodimerization and trafficking. J Biol Chem 2004;279:24881–24888. Klett EL, Lu K, Kosters A, et al. A mouse model of sitosterolemia: absence of Abcg8/sterolin-2 results in failure to secrete biliary cholesterol. BMC Med 2004;2:5. Plosch T, Bloks VW, Terasawa Y, et al. Sitosterolemia in ABCTransporter G5-deficient mice is aggravated on activation of the liver-X receptor. Gastroenterology 2004;126:290–300. Small DM. Role of ABC transporters in secretion of cholesterol from liver into bile. Proc Natl Acad Sci USA 2003;100:4–6. Hoekstra M, Kruijt JK, van Eck M, Van Berkel TJ. Specific gene expression of ATP-binding cassette transporters and nuclear
111. 112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127. 128.
129.
130.
hormone receptors in rat liver parenchymal, endothelial, and Kupffer cells. J Biol Chem 2003;278:25448–25453. Ito T. Physiological function of ABCG1. Drug News Perspect 2003;16:490–492. Higuchi H, Gores GJ. Bile acid regulation of hepatic physiology: IV. Bile acids and death receptors. Am J Physiol Gastrointest Liver Physiol 2003;284:G734–G738. Schoemaker MH, Gommans WM, de la Rosa LC, et al. Resistance of rat hepatocytes against bile acid-induced apoptosis in cholestatic liver injury is due to nuclear factor-kappa B activation. J Hepatol 2003;39:153–161. Streetz K, Leifeld L, Grundmann D, et al. Tumor necrosis factor alpha in the pathogenesis of human and murine fulminant hepatic failure. Gastroenterology 2000;119:446–460. Ghose R, Zimmerman TL, Thevananther S, Karpen SJ. Endotoxin leads to rapid subcellular re-localization of hepatic RXRalpha: A novel mechanism for reduced hepatic gene expression in inflammation. Nucl Recept 2004;2:4. Wan YJ, An D, Cai Y, et al. Hepatocyte-specific mutation establishes retinoid X receptor alpha as a heterodimeric integrator of multiple physiological processes in the liver. Mol Cell Biol 2000;20:4436–4444. Makishima M, Okamoto AY, Repa JJ, et al. Identification of a nuclear receptor for bile acids. Science 1999;284: 1362–1365. Shih DQ, Bussen M, Sehayek E, et al. Hepatocyte nuclear factor-1alpha is an essential regulator of bile acid and plasma cholesterol metabolism. Nature Genet 2001;27:375–382. Lu TT, Makishima M, Repa JJ, et al. Molecular basis for feedback regulation of bile acid synthesis by nuclear receptors. Mol Cell 2000;6:507–515. Denson LA, Sturm E, Echevarria W, et al. The orphan nuclear receptor, shp, mediates bile acid-induced inhibition of the rat bile acid transporter, ntcp. Gastroenterology 2001;121:140–147. Sinal CJ, Tohkin M, Miyata M, et al. Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis. Cell 2000;102:731–744. Gartung C, Ananthanarayanan M, Rahman MA, et al. Downregulation of expression and function of the rat liver Na+/bile acid co-transporter in extrahepatic cholestasis. Gastroenterology 1996;110:199–209. Zollner G, Fickert P, Zenz R, et al. Hepatobiliary transporter expression in percutaneous liver biopsies of patients with cholestatic liver diseases. Hepatology 2001;33:633–646. Trauner M, Arrese M, Soroka CJ, et al. The rat canalicular conjugate export pump (Mrp2) is down-regulated in intrahepatic and obstructive cholestasis. Gastroenterology 1997;113:255–264. Assem M, Schuetz EG, Leggas M, et al. Interactions between hepatic Mrp4 and Sult2a as revealed by the constitutive androstane receptor and Mrp4 knockout mice. J Biol Chem 2004;279:22250–22257. Schuetz EG, Strom S, Yasuda K, et al. Disrupted bile acid homeostasis reveals an unexpected interaction among nuclear hormone receptors, transporters, and cytochrome P450. J Biol Chem 2001;276:39411–39418. Thompson R, Jansen PL. Genetic defects in hepatocanalicular transport. Semin Liver Dis 2000;20:365–372. Bull LN, Juijn JA, Liao M, et al. Fine-resolution mapping by haplotype evaluation: the examples of PFIC1 and BRIC. Hum Genet 1999;104:241–248. De Vree JM, Jacquemin E, Sturm E, et al. Mutations in the MDR3 gene cause progressive familial intrahepatic cholestasis. Proc Natl Acad Sci USA 1998;95:282–287. Lamireau T, Monnereau S, Martin S, et al. Epidemiology of liver disease in cystic fibrosis: a longitudinal study. J Hepatol 2004;41:920–925.
83
Section I. Pathophysiology of the Liver
131. Mosser J, Lutz Y, Stoeckel ME, et al. The gene responsible for adrenoleukodystrophy encodes a peroxisomal membrane protein. Hum Mol Genet 1994;3:265–271. 132. Paulusma CC, Kool M, Bosma PJ, et al. A mutation in the human canalicular multispecific organic anion transporter gene causes the Dubin–Johnson syndrome. Hepatology 1997;25:1539–1542. 133. Kartenbeck J, Leuschner U, Mayer R, Keppler D. Absence of the canalicular isoform of the MRP gene-encoded conjugate export pump from the hepatocytes in Dubin–Johnson syndrome. Hepatology 1996;23:1061–1066. 134. Shroyer NF, Lewis RA, Allikmets R, et al. The rod photoreceptor ATP-binding cassette transporter gene, ABCR, and retinal disease: from monogenic to multifactorial. Vision Res 1999;39:2537–2544. 135. Brooks-Wilson A, Marcil M, Clee SM, et al. Mutations in ABC1 in Tangier disease and familial high-density lipoprotein deficiency. Nature Genet 1999;22:336–345. 136. Otonkoski T, Ammala C, Huopio H, et al. A point mutation inactivating the sulfonylurea receptor causes the severe form of persistent hyperinsulinemic hypoglycemia of infancy in Finland. Diabetes 1999;48:408–415. 137. Ringpfeil F, Lebwohl MG, Christiano AM, Uitto J. Pseudoxanthoma elasticum: mutations in the MRP6 gene encoding a transmembrane ATP-binding cassette (ABC) transporter. Proc Natl Acad Sci USA 2000;97:6001–6006. 138. Bull LN, Carlton VE, Stricker NL, et al. Genetic and morphological findings in progressive familial intrahepatic cholestasis (Byler disease [PFIC-1] and Byler syndrome): evidence for heterogeneity. Hepatology 1997;26:155–164. 139. Klomp LW, Vargas JC, van Mil SW, et al. Characterization of mutations in ATP8B1 associated with hereditary cholestasis. Hepatology 2004;40:27–38. 140. Oshima T, Ikeda K, Takasaka T. Sensorineural hearing loss associated with Byler disease. Tohoku J Exp Med 1999;187:83–88. 141. van Mil SW, van Oort MM, van den Berg IE, et al. Fic1 is expressed at apical membranes of different epithelial cells in the digestive tract and is induced in the small intestine during postnatal development of mice. Pediatr Res 2004;56:981–987. 142. Ding J, Wu Z, Crider BP, et al. Identification and functional expression of four isoforms of ATPase II, the putative aminophospholipid translocase. Effect of isoform variation on the ATPase activity and phospholipid specificity. J Biol Chem 2000;275:23378–23386. 143. Pawlikowska L, Groen A, Eppens EF, et al. A mouse genetic model for familial cholestasis caused by ATP8B1 mutations reveals perturbed bile salt homeostasis but no impairment in bile secretion. Hum Mol Genet 2004;13:881–892. 144. Lykavieris P, van Mil S, Cresteil D, et al. Progressive familial intrahepatic cholestasis type 1 and extrahepatic features: no catch-up of stature growth, exacerbation of diarrhea, and appearance of liver steatosis after liver transplantation. J Hepatol 2003;39:447–452. 145. Chen F, Ananthanarayanan M, Emre S, et al. Progressive familial intrahepatic cholestasis, type 1, is associated with decreased farnesoid X receptor activity. Gastroenterology 2004;126:756–764. 146. Watanabe M, Houten SM, Wang L, et al. Bile acids lower triglyceride levels via a pathway involving FXR, SHP, and SREBP-1c. J Clin Invest 2004;113:1408–1418. 147. Jacquemin E, Hermans D, Myara A, et al. Ursodeoxycholic acid therapy in pediatric patients with progressive familial intrahepatic cholestasis. Hepatology 1997;25:519–523. 148. Melter M, Rodeck B, Kardorff R, et al. Progressive familial intrahepatic cholestasis: partial biliary diversion normalizes serum lipids and improves growth in noncirrhotic patients. Am J Gastroenterol 2000;95:3522–3528.
84
149. Kurbegov AC, Setchell KD, Haas JE, et al. Biliary diversion for progressive familial intrahepatic cholestasis: improved liver morphology and bile acid profile. Gastroenterology 2003;125:1227–1234. 150. Tygstrup N, Steig BA, Juijn JA, et al. Recurrent familial intrahepatic cholestasis in the Faeroe Islands. Phenotypic heterogeneity but genetic homogeneity. Hepatology 1999;29:506–508. 151. Summerskill WH, Walshe JM. Benign recurrent intrahepatic ‘obstructive’ jaundice. Lancet 1959;2:686–690. 152. Floreani A, Molaro M, Mottes M, et al. Autosomal dominant benign recurrent intrahepatic cholestasis (BRIC) unlinked to 18q21 and 2q24. Am J Med Genet 2000;95:450–453. 153. Brenard R, Geubel AP, Benhamou JP. Benign recurrent intrahepatic cholestasis. A report of 26 cases. J Clin Gastroenterol 1989;11:546–551. 154. Cancado EL, Leitao RM, Carrilho FJ, Laudanna AA. Unexpected clinical remission of cholestasis after rifampicin therapy in patients with normal or slightly increased levels of gamma-glutamyl transpeptidase. Am J Gastroenterol 1998;93:1510–1517. 155. Balsells F, Wyllie R, Steffen R, Kay M. Benign recurrent intrahepatic cholestasis: improvement of pruritus and shortening of the symptomatic phase with rifampin therapy: a case report. Clin Pediatr (Phil) 1997;36:483–485. 156. Strautnieks SS, Kagalwalla AF, Tanner MS, et al. Identification of a locus for progressive familial intrahepatic cholestasis PFIC2 on chromosome 2q24. Am J Hum Genet 1997;61:630–633. 157. Jansen PL, Strautnieks SS, Jacquemin E, et al. Hepatocanalicular bile salt export pump deficiency in patients with progressive familial intrahepatic cholestasis. Gastroenterology 1999;117: 1370–1379. 158. van Mil SW, van der Woerd WL, van der BG, et al. Benign recurrent intrahepatic cholestasis type 2 is caused by mutations in ABCB11. Gastroenterology 2004;127:379–384. 159. Morton DH, Salen G, Batta AK, et al. Abnormal hepatic sinusoidal bile acid transport in an Amish kindred is not linked to FIC1 and is improved by ursodiol. Gastroenterology 2000;119:188–195. 160. Carlton VE, Harris BZ, Puffenberger EG, et al. Complex inheritance of familial hypercholanemia with associated mutations in TJP2 and BAAT. Nature Genet 2003;34:91–96. 161. Clayton PT, Leonard JV, Lawson AM, et al. Familial giant cell hepatitis associated with synthesis of 3 beta, 7 alpha-dihydroxyand 3 beta, 7 alpha, 12 alpha-trihydroxy-5-cholenoic acids. J Clin Invest 1987;79:1031–1038. 162. Jacquemin E, Setchell KD, O’Connell NC, et al. A new cause of progressive intrahepatic cholestasis: 3 beta-hydroxy-C27-steroid dehydrogenase/isomerase deficiency. J Pediatr 1994;125:379–384. 163. Setchell KD, Schwarz M, O’Connell NC, et al. Identification of a new inborn error in bile acid synthesis: mutation of the oxysterol 7alpha-hydroxylase gene causes severe neonatal liver disease. J Clin Invest 1998;102:1690–1703. 164. Shneider BL, Setchell KD, Whitington PF, et al. Delta 4-3oxosteroid 5 beta-reductase deficiency causing neonatal liver failure and hemochromatosis. J Pediatr 1994;124:234–238. 165. Stieger B, Zhang J, O’Neill B, et al. Differential interaction of bile acids from patients with inborn errors of bile acid synthesis with hepatocellular bile acid transporters. Eur J Biochem 1997;244:39–44. 166. De Vree JM, Jacquemin E, Sturm E, et al. Mutations in the MDR3 gene cause progressive familial intrahepatic cholestasis. Proc Natl Acad Sci USA 1998;95:282–287. 167. Deleuze JF, Jacquemin E, Dubuisson C, et al. Defect of multidrug-resistance 3 gene expression in a subtype of progressive familial intrahepatic cholestasis. Hepatology 1996;23:904–908.
Chapter 5 MECHANISMS OF BILE SECRETION
168. Jacquemin E, Cresteil D, Manouvrier S, et al. Heterozygous non-sense mutation of the MDR3 gene in familial intrahepatic cholestasis of pregnancy. Lancet 1999;353:210–211. 169. Dixon PH, Weerasekera N, Linton KJ, et al. Heterozygous MDR3 missense mutation associated with intrahepatic cholestasis of pregnancy: evidence for a defect in protein trafficking. Hum Mol Genet 2000;9:1209–1217. 170. de Pagter AG, Berge Henegouwen GP, Bokkel Huinink JA, Brandt KH. Familial benign recurrent intrahepatic cholestasis. Interrelation with intrahepatic cholestasis of pregnancy and from oral contraceptives? Gastroenterology 1976;71:202–207. 171. Pauli-Magnus C, Lang T, Meier Y, et al. Sequence analysis of bile salt export pump (ABCB11) and multidrug resistance pglycoprotein 3 (ABCB4, MDR3) in patients with intrahepatic cholestasis of pregnancy. Pharmacogenetics 2004;14:91–102. 172. Painter JN, Savander M, Ropponen A, et al. Sequence variation in the ATP8B1 gene and intrahepatic cholestasis of pregnancy. Eur J Hum Genet 2005; 13: 435–439. 173. Mazzella G, Nicola R, Francesco A, et al. Ursodeoxycholic acid administration in patients with cholestasis of pregnancy: effects on primary bile acids in babies and mothers. Hepatology 2001;33:504–508. 174. Palma J, Reyes H, Ribalta J, et al. Ursodeoxycholic acid in the treatment of cholestasis of pregnancy: a randomized, double-
175.
176.
177.
178.
179.
blind study controlled with placebo. J Hepatol 1997;27:1022–1028. Aagenaes O. Hereditary cholestasis with lymphoedema (Aagenaes syndrome, cholestasis-lymphoedema syndrome). New cases and follow-up from infancy to adult age. Scand J Gastroenterol 1998;33:335–345. Bull LN, Roche E, Song EJ, et al. Mapping of the locus for cholestasis–lymphedema syndrome (Aagenaes syndrome) to a 6.6-cM interval on chromosome 15q. Am J Hum Genet 2000;67:994–999. Wada M, Toh S, Taniguchi K, et al. Mutations in the canilicular multispecific organic anion transporter (cMOAT) gene, a novel ABC transporter, in patients with hyperbilirubinemia II/Dubin–Johnson syndrome. Hum Mol Genet 1998; 7:203–207. Buchler M, Konig J, Brom M, et al. cDNA cloning of the hepatocyte canalicular isoform of the multidrug resistance protein, cMrp, reveals a novel conjugate export pump deficient in hyperbilirubinemic mutant rats. J Biol Chem 1996;271:15091–15098. Keitel V, Nies AT, Brom M, et al. A common Dubin–Johnson syndrome mutation impairs protein maturation and transport activity of MRP2 (ABCC2). Am J Physiol Gastrointest Liver Physiol 2003;284:G165–G174.
85
Section I: Pathophysiology of the Liver
6
HEPATIC FIBROSIS AND CIRRHOSIS Don C. Rockey and Scott L. Friedman Abbreviations ALT alanine aminotransferase AST aspartate aminotransferase AUROC area under the receiver operator characteristic BMI body mass index CINC cytokine-induced neutrophil chemoattractant CTGF connective tissue growth factor DDR discoidin domain receptors ECM extracellular matrix EGF epidermal growth factor ELF European liver fibrosis ET-1 endothelin-1 FGF fibroblast growth factor FPI fibrosis probability index GGT g-glutamyl transferase
GnT-III HA HBV HCV HGF HIV HOMA-IR IL-10 JI LPS MCP-1 MEGX MELD MMPs MMP-2
N-acetylglucosaminyl transferase III hyaluronic acid hepatitis B virus hepatitis C hepatocyte growth factor human immunodeficiency virus insulin resistance by the homeostasis model assessment interleukin-10 jejuno-ileal lipopolysaccharide monocyte chemotactic protein-1 monoethylglycinexylidide model for end-stage liver disease matrix metalloproteinases matrix metalloproteinase 2
INTRODUCTION Hepatic fibrosis has emerged as a highly relevant aspect of liver biology because of the significant progress in uncovering its mechanisms, combined with a growing realization that effective antifibrotic therapies may soon alter the natural history of chronic liver disease. Thus, liver fibrosis can now be viewed as a clinical problem whose diagnosis and treatment will soon have rational, evidence-based approaches. This progress is very timely, as the continued ‘aging’ of the HCV-infected cohort and the growing prevalence of obesityrelated liver diseases are leading to precipitous increases in the prevalence of advanced liver disease.1 With these issues in mind, this chapter will review clinical aspects of hepatic fibrosis, including natural history, pathophysiologic mechanisms, current and future tools for diagnosis, and emerging antifibrotic strategies. In addition, several recent reviews highlight many of these aspects in greater detail.2–6 Hepatic fibrosis is the accumulation of extracellular matrix, or scar, in response to acute or chronic liver injury. Fibrogenesis represents a wound healing response to injury (Figure 6-1), and ultimately leads to cirrhosis. Cirrhosis is the end-stage consequence of fibrosis of the hepatic parenchyma, resulting in nodule formation that may lead to altered hepatic function and blood flow. Both fibrosis and cirrhosis are the consequences of a sustained wound-healing response to chronic liver injury from a range of causes, including viral, autoimmune, drug induced, cholestatic and metabolic diseases. The clinical manifestations of cirrhosis vary widely, from no symp-
MMP-9 NASH NGFR NO PDGF PELD PIIINP PPAR QTL ROC TGF-b1 TIMPs ULN VEGF
matrix metalloproteinase 9 non-alcoholic steatohepatitis nerve growth factor receptor nitric oxide platelet-derived growth factor pediatric end-stage liver disease propeptide of type III collagen peroxisomal proliferator-activated receptor quantitative trait loci receiver operating characteristic transforming growth factor beta 1 tissue inhibitors of metalloproteinases upper limit of normal vascular endothelial growth factor
toms at all to liver failure, and are determined by both the nature and severity of the underlying liver disease as well as the extent of hepatic fibrosis. Up to 40% of patients with cirrhosis are asymptomatic and may remain so for long periods, but progressive deterioration leading to death or liver transplantation is typical once complications (such as ascites, variceal hemorrhage or encephalopathy) develop. In such patients there is a 50% 5-year mortality, with approximately 70% of these deaths directly attributable to liver disease.7 In asymptomatic individuals cirrhosis may be first suggested during routine examination, although histologic analysis may be required to establish the diagnosis. Cirrhosis affects hundreds of millions of patients worldwide. The overall burden of liver disease in the United States – the vast majority of which is due to chronic disease with fibrosis – continues to expand, exacting an increasing economic and social cost.1 Indeed, in the US cirrhosis is the most common non-neoplastic cause of death among hepatobiliary and digestive diseases, accounting for approximately 30 000 deaths per year. In addition, 10 000 deaths are due to liver cancer, the majority of which arise in cirrhotic livers, consistent with a steadily rising mortality rate from hepatic cancer.8 Notably, hepatocellular carcinoma is the most rapidly increasing neoplasm in the US and western Europe.9 Initial studies of hepatic fibrosis focused on the composition of extracellular matrix in liver, and continued incremental progress in this area is still anticipated. However, attention has gradually shifted towards exploring the cellular basis of fibrosis and the cellular mediators that drive fibrosis progression and regression (see Pathophysiology, below). In general, the molecular composition of the scar
87
Section I. Pathophysiology of the Liver
Normal liver
Liver injury
Hepatocytes
Loss of Hepatocyte microvilli
Space of Disse
Quiescent stellate cell Kupffer cell
Activated stellate cell Endothelial cell
Hepatic sinusoid
Deposition of scar matrix
Loss of fenestrae
Kupffer cell activation
Figure 6-1. Hepatic liver cells and the hepatic sinusoid in normal and injured liver. On the left panel is shown the multiple key liver-specific cellular elements in the normal liver, including hepatocytes, endothelial cells, Kupffer cells, and stellate cells. Stellate cells are located within the subendothelial space of Disse (i.e. between the sinusoidal endothelium and hepatocytes). The figure emphasizes the close physical relationships between the various cellular elements in the liver. After liver injury, changes in numerous cells occur; for example, stellate and Kupffer cells become activated (see Figure 6-3), hepatocytes lose their microvilli, and endothelial cells lose their characteristic fenestrae. All of these features contribute to continued cell activation and injury, as well as dysfunction at the whole organ level.
tissue in cirrhosis is similar regardless of etiology, and resembles that of other parenchymal scarring (e.g. kidney), consisting of the extracellular matrix constituents, collagen types I and III (i.e. ‘fibrillar’ collagens), sulfated proteoglycans, and glycoproteins.10 However, some isoforms of extracellular matrix constituents, for example fibronectin11 and proteoglycans,12 may be relatively enriched during progressive injury. These scar constituents accumulate from a net increase in their deposition in liver and not simply from the collapse of existing stroma.
CLINICAL ASPECTS OF HEPATIC FIBROSIS NATURAL HISTORY AND RISK FACTORS Fibrosis leading to cirrhosis can accompany virtually any chronic liver disease that is characterized by the presence of architectural disruption and/or inflammation. The vast majority of patients with liver disease worldwide have chronic viral hepatitis, or steatohepatitis associated with either alcohol or obesity; other etiologies of liver disease include parasitic infestation (e.g. schistosomiasis), autoimmune attack on hepatocytes or biliary epithelium, neonatal liver disease, metabolic disorders including Wilson’s disease,
88
hemochromatosis and a variety of storage diseases, chronic inflammatory conditions (e.g. sarcoidosis), drug toxicity (e.g. methotrexate or hypervitaminosis A), and vascular derangements, either congenital or acquired. Of the many causes of chronic liver disease, our understanding of natural history of fibrosis is most complete in HCV, with some information about HBV and steatohepatitic diseases, including alcoholic liver disease and NASH. Information about fibrosis progression in other diseases is largely anecdotal, but the development of cirrhosis typically requires many years to decades. There are, however, some notable exceptions in which the development of cirrhosis can be greatly accelerated, possibly occurring within months rather than years: (1) neonatal liver disease – infants with biliary atresia may present at birth with severe fibrosis and marked parenchymal distortion; (2) HCV-infected patients after liver transplantation – a subset of patients who undergo liver transplantation for HCV cirrhosis may develop rapidly progressive cholestasis and recurrent cirrhosis within months, requiring retransplantation;13 (3) patients with HIV/HCV co-infection – these patients have relatively rapid fibrosis compared to those with HCV alone,14 especially if the HIV is untreated (see below); (4) severe delta hepatitis;15 and (5) some cases of drug-induced liver disease. These examples of ‘fulminant fibrosis’ probably reflect dysregulation of several pathways, including defective immunity, massive inflammation and necrosis, and/or
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
altered matrix resorption. Together, they highlight the highly dynamic nature of scar accumulation and degradation. Moreover, when matrix accumulation is unopposed because degradation is ineffective, more rapid fibrosis may ensue. Once cirrhosis and its complications develop the prognosis is predicted by widely used systems, including Child–Pugh, PELD16 and MELD,17 which are predictive independent of the etiology of liver disease.
Hepatitis C Virus The risk and natural history of fibrosis associated with HCV have been greatly clarified as a result of several large clinical studies incorporating standardized assessments of fibrosis that combine detailed historical and clinical information.18 The disease can run a remarkably variable course, from decades of viremia with little fibrosis to a rapid onset of cirrhosis within 10–15 years. It appears to be host factors rather than viral factors that correlate with fibrosis progression in HCV. The data supporting this conclusion include the following: (1) there is no relationship between viral load or genotype and severity of fibrosis even though these former factors affect the response to antiviral therapy; (2) human promoter polymorphisms (e.g. TGF-b1 and angiotensin) appear to correlate with fibrosis risk,19 with large-scale efforts currently under way to identify additional genetic markers of fibrosis risk;20 (3) host immune phenotype may be critical, as there is more rapid progression in immunosuppressed patients, whether due to HIV or to immunosuppressive drugs.14 In mice, a Th2 phenotype strongly correlates with fibrogenic potential,21 which has led to successful efforts to use quantitative trait loci (QTL) mapping to identify specific fibrosis risk genes in these animals.22 Other identified host risk factors for more rapid progression of HCV include: (1) older age at the time of infection; (2) concurrent liver disease due to HBV or alcohol (>50 g/day); it is uncertain, however, whether lesser amounts of alcohol intake are additive towards fibrosis progression: recent studies suggest that less than 50 g/day of alcohol results in a neglible increased risk of hepatic fibrosis;23 (3) male gender; (4) increased body mass index (BMI), associated with hepatic steatosis;24 (5) HIV infection or immunosuppression following liver transplantation. Because standard clinical indices cannot distinguish between minimal and even advanced fibrosis, knowledge about these risk factors and duration of infection can greatly inform clinical management. Thus, for chronic HCV, if the time of infection is known and a biopsy obtained at any time thereafter, the rate of progression per year based on either Ishak or METAVIR scoring can be estimated.25 Although initial analyses of this type suggested that fibrosis progression is truly linear, it is now increasingly clear that the progression rate accelerates as the disease advances,26 such that it takes less time to progress between Metavir stages 3 and 4, than from stage 1 to 2, for example. Assessment of fibrosis stage and rate of fibrosis progression can be valuable for at least three reasons: (1) the actual stage of fibrosis will indicate the likelihood of response to a-interferon or ainterferon/ribavirin, as the advanced stages of fibrosis (F3 or F4) generally have a lower response rate to antiviral therapy;27,28 (2) if little fibrosis progression has occurred over a long interval, then treatment with antiviral therapy may be deemed to be less urgent
and it may be safe to await more effective and/or better-tolerated therapy; (3) the approximate time to the development of cirrhosis can be estimated. This would not, however, indicate if or when clinical liver failure would occur, as the complications of liver disease may be delayed for up to a decade or more after the establishment of cirrhosis. As genetic risk markers that predict a rapid fibrosis progression rate are developed, this information, combined with the absolute stage of fibrosis, may enable more accurate identification of patients at risk for disease and thus in need of antifibrotic therapy.
Hepatitis B Virus Very few studies have assessed the progression rate of fibrosis in chronic HBV infection. In general, inflammatory activity, as influenced by viral factors, including e Ag status, that indicate active viral replication, correlates with fibrosis.29,30 Fibrosis progression has been correlated with HBV genotype in at least one study.31 In a subset of patients a rapidly progressive ‘fibrosing cholestatic hepatitis’ may occur,32 but there are neither definitive risk factors for this condition nor unique etiologic, cellular or molecular determinants identified. In addition, delta hepatitis superinfection or co-infection may greatly accelerate the risk of advanced fibrosis and cirrhosis.15 What is striking, however, is that virologic suppression in response to potent antiviral regimens can effect remarkable improvement not only in serum alanine aminotransferase (ALT) levels and histologic inflammation, but also in fibrosis.15,33–35 Indeed, dramatic resolution of cirrhosis in a 10-year follow-up has been reported in patients with delta hepatitis who were successfully treated with a-interferon.15
Alcoholic Liver Disease The clearest clinical determinant of fibrosis is continued alcohol abuse: patients with fibrosis who continue to drink are virtually assured of progression. In addition, two clinical features commonly seen in steatohepatitis, elevated BMI and serum glucose, also confer an increased risk of fibrosis in alcoholic liver disease.36 Pathologically, the presence of pericentral fibrosis (central hyaline sclerosis) carries a high risk of eventual panlobular cirrhosis, which is almost certain if alcohol intake continues.
Non-Alcoholic Steatohepatitis There is a critical need for better data about natural history, risk factors for fibrosis, and rate of fibrosis progression in NASH, issues now being addressed in several multicenter studies. Patients with only steatosis and no inflammation appear to have a benign course when followed for up to 19 years;37 however, it is unclear whether this lesion is completely distinct from steatohepatitis, or instead represents a precursor of NASH. It is instructive to remember that HCV fibrosis progression rates were underestimated shortly after the virus was first identified, as many patients had a relatively early fibrosis stage. With continued infection, however, a sizeable fraction eventually have progressed to more advanced stages. In a parallel situation, the obesity epidemic in the US and the developed world is only now being fully appreciated, and a threshold level of obesity may have only begun to confer a risk of liver disease that will become clinically significant in the next decade. In patients with sustained
89
Section I. Pathophysiology of the Liver
NASH spontaneous histologic improvement is very uncommon, but better longitudinal data are needed to understand the natural history of this disease; for example, data examining the evolution of NASH in patients undergoing bariatric surgery who have extensive weight loss and improvement in their metabolic syndrome are awaited. In three combined studies of 26 patients followed with sequential biopsies for up to 9 years, 27% had progression of fibrosis and 19% advanced to cirrhosis, but none had reversal of fibrosis.37 Of interest is the recurrence of NASH following liver transplantation in some patients with cryptogenic cirrhosis, implicating an underlying metabolic defect that may account for liver disease in both the native and the transplanted organs. Risk of fibrosis and rate of progression are critical issues that will influence risk stratification and patient selection for clinical trials, as progression to cirrhosis is the most important clinical consequence of NASH. Recently developed systems to grade and stage liver disease in NASH38 should allow for improved, prospective collection of standardized data that can further address these vital questions. In general, increasing obesity (BMI) >28 kg/m2) correlates with severity of fibrosis and risk of cirrhosis. Other risk factors include necroinflammatory activity with ALT >2¥ normal and/or AST/ALT >1, age, elevated triglycerides, insulin resistance and/or diabetes mellitus, and systemic hypertension.39 It is uncertain whether these features are comparable across the spectrum of disorders associated with NASH, including obesity with insulin-resistance, JI (jejuno-ileal) bypass, total parenteral nutrition and rapid weight loss, among others. Whether these factors represent surrogates for other risk factors (i.e. reduced antioxidant levels in older patients, increased renin–angiotensin activity in hypertensives) is unknown. Ratziu and colleagues40 have reported a clinicobio-logical score that combines age, BMI, triglycerides and ALT and which reportedly has 100% negative predictive value for excluding significant fibrosis.
REVERSIBILITY OF FIBROSIS AND CIRRHOSIS There is now clear evidence that fibrosis and even cirrhosis can be reversible. The feature common to all cases of cirrhosis improvement is the elimination of the underlying cause of liver disease, whether due to eradication of HBV,41 delta hepatitis15 or HCV,42 decompression of biliary obstruction in chronic pancreatitis,43 or to immunosuppressive treatment of autoimmune liver disease.44 Moreover, there is ample evidence of reversibility in animal models, which provide vital clues to underlying mechanisms.45 Earlier studies demonstrated that fibrosis improves with treatment of HCV,46 and even cirrhosis can regress following HCV eradication with a-interferon/ribavirin.42 Among a large cohort of patients successfully treated with this combination there were 150 with cirrhosis, half of whom had a reduction in their fibrosis score according to METAVIR staging, with several regressing by two or more stages. Because fibrosis in HCV typically progresses over three decades, one might anticipate an equally slow but steady regression of fibrosis following viral clearance.
90
It remains unclear what distinguishes those patients whose cirrhosis is reversible from those in whom it is fixed. Potential factors influencing reversibility probably include: (1) a prolonged period of established cirrhosis, which could reflect a longer period of crosslinking of collagen, rendering this collagen less sensitive to degradation by enzymes over time. Animal studies now clearly support this possibility;47 (2) total content of collagen and other scar molecules, which might lead to a large mass of scar that is physically inaccessible to degradative enzymes; (3) reduced expression of enzymes that degrade matrix, or sustained elevation of proteins that inhibit the function of these degradative enzymes, in particular elevated levels of tissue inhibitors of metalloproteinases (TIMPs), which block matrix proteases and also prevent apoptosis of activated stellate cells.48,49 All three scenarios highlight the dynamic process of collagen deposition and degradation.
PATHOPHYSIOLOGY OF HEPATIC FIBROSIS AND CIRRHOSIS EXTRACELLULAR MATRIX (ECM) IN THE NORMAL AND THE FIBROTIC LIVER Extracellular matrix refers to the array of macromolecules that comprise the scaffolding of normal and fibrotic liver. These macromolecules consist of three main families: collagens, glycoproteins and proteoglycans (see 10 for review). The number of collagens identified in liver is rapidly growing, and includes collagen XVIII, which is a precursor to the molecule angiostatin. Glycoproteins include fibronectin, laminin, merosin, tenascin, nidogen, and hyaluronic acid, among others. Proteoglycans include heparan, dermatan sulfates, chondroitin sulfates, perlecan dystroglycan syndecan, biglycan and decorin. There is tremendous heterogeneity of these matrix macromolecules with respect to their different isoforms, variable combinations within different tissue regions, and changes related to age. In normal liver the subendothelial space of Disse separates the epithelium (hepatocytes) from the sinusoidal endothelium. This space contains a basement membrane-like matrix which, unlike the typical basement membrane, is not electron dense. The hepatic basement membrane is composed of non-fibril-forming collagens, including types IV, VI and XIV, glycoproteins and proteoglycans. This normal subendothelial ECM is critical for maintaining the differentiated functions of resident liver cells, including hepatocytes, stellate cells and sinusoidal endothelium. In contrast to basement membrane-type matrix, in normal liver the so-called interstitial ECM is largely confined to the capsule, around large vessels, and in the portal areas. It is composed of fibrilforming collagens (e.g. types I and III) together with cellular (EDA) fibronectin, undulin, and other glycoconjugates. As the liver becomes fibrotic, the total content of collagens and non-collagenous components increases three- to fivefold, accompanied by a shift in the type of ECM in the subendothelial space from the normal low-density basement membrane-like matrix to interstitial-type matrix (see 10 for review). This ‘capillarization’ leads to the loss of hepatocyte microvilli and the disappearance of endothelial fenestration (Figure 6-1).
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
The outcome of fibrogenesis is the conversion of normal lowdensity basement membrane-like matrix to high-density interstitialtype matrix. A number of components are responsible for ECM remodeling (see 49,50 for reviews) (Figure 6.2). These include a family of zinc-dependent enzymes matrix metalloproteinases (MMPs),51 their inhibitors (tissue inhibitor of metalloproteinases, TIMP), and several converting enzymes (MT1-MMP, and stromelysin, for example). In human liver diseases there is down-regulation of MMP1 (interstitial collagenase, collagenase I) and up-regulation of MMP2 (gelatinase A) and MMP9 (gelatinase B). Based on the differing substrate specificities of these enzymes, the result is increased degradation of basement membrane collagen and decreased degradation of interstitial collagens. These activated MMPs are regulated in part by their tissue inhibitors, the so-called TIMPs. TIMP1 and TIMP2 are upregulated relative to MMP1 in progressive experimental liver fibrosis, which may explain the decreased degradation of interstitial-type matrix observed in experimental and human liver injury. In contrast, during the resolution of experimental liver injury TIMP-1 and TIMP-2 expression is decreased whereas collagenase expression is unchanged, resulting in a net increase in collagenase activity and increased resorption of scar matrix. Stellate cells are a key source of MMP-2 and stromelysin. They also express TIMP-1 and TIMP-2 mRNAs and produce TIMP-1 and MT1-MMP MMP-9, which is a type IV collagenase locally secreted by Kupffer cells, and may also be produced by stellate cells in
response to interleukin-1.52 The source of MMP-1, which plays a crucial role in degrading the excess interstitial matrix in advanced liver disease, is still uncertain.53 However, interstitial collagenase activity in liver may be attributable to either MT1-MMP or even MMP-2, although further studies are required.
ECM–CELL INTERACTIONS Changes in the microenvironment of the space of Disse result in phenotypic changes in all resident liver cells. Hepatic stellate cells are activated by the surrounding increase in interstitial matrix.54 Sinusoidal endothelial cells produce cellular fibronectin in very early liver injury, which also contributes to stellate cell activation. In addition, endothelial cells produce type IV collagen, proteoglycans and factors (e.g. urokinase-type plasminogen activator) that participate in the activation of latent cytokines such as TGF-b1. Activated Kupffer cells release cytokines and reactive oxygen intermediates that may stimulate stellate cells in a paracrine manner.55 Platelets are also an abundant source of cytokines upon injury, producing a rich array of important growth factors. Hepatocytes, the most abundant cells in the liver, generate lipid peroxides following injury that lead to stellate cell activation, a prerequisite for fibrogenesis (see below). The dynamic interactions between fibrogenic cells in liver and the ECM is an important determinant of fibrogenesis. The ECM is a reservoir for growth factors, for example platelet-derived growth factor (PDGF).10 Like all cytokines, PDGF signals by binding to
Activated stellate cell Early pathological degradation
Regression Apoptotic stellate cell Normal ECM
MT1-MMP MMP-2 TIMP-2
Kupffer cell MMP-1 MMP-13 Other MMPs Progression
TIMP-1 TIMP-2
Figure 6-2. Emerging Mechanisms of Early Pathologic Matrix Degradation, Fibrosis Progression & Fibrosis Resolution in Chronic Liver Disease. Activation of stellate cells (top left panel) is a key event in hepatic fibrosis, and is associated with pathologic matrix degradation due to increased production of membrane type matrix metalloproteinase 1 (MT1-MMP), matrix metalloproteinase-2 (MMP-2), and tissue inhibitors of metalloproteinases (TIMPs), leading to replacement by interstitial collagen or scar matrix. As fibrosis progresses (middle panel), sustained expression of TIMPs prevents matrix degradation and apoptosis of activated stellate cells. Regression of fibrosis (upper right panel) is associated with increased apoptosis of activated stellate cells. Apoptosis requires decreased expression of tissue inhibitor of metalloproteinase-1 (TIMP-1), yielding a net increase in protease activity. These events may occur coincident with production of matrix metalloproteinases, which could include MMP-1 (in humans) and/or MMP-13 (in rodents), although cellular sources of these enzymes (possibly including Kupffer cells), and clear evidence of their induction in vivo are still lacking. Validation of these events and further elucidation of mechanisms underlying fibrosis regression represent key challenges for future studies.147a
91
Section I. Pathophysiology of the Liver
membrane receptors. The PDGF receptor belongs to a receptor family known as receptor tyrosine kinases, which collectively are key transducers for many important cytokines, including hepatocyte growth factor (HGF), epidermal growth factor (EGF), vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF). Interestingly, a new subclass of receptor tyrosine kinases, socalled discoidin domain receptors (DDR), has been identified; this group of receptors signal in response to fibrillar collagens rather than peptide ligands.56 Indeed, stellate cell activation is accompanied by up-regulation of DDR2 receptors, and increased signaling is associated with altered MMP-2 expression.57,58 Intracellular signaling cascades downstream of receptor tyrosine kinases and other receptors are pervasive (see 59 for review). Integrins are another type of membrane receptor that transduce extracellular signals in liver. These are heterodimeric transmembrane proteins composed of an a and a b subunit whose ligands are matrix molecules rather than cytokines. Several integrins and their downstream effectors have been identified in stellate cells, including a1b1, a2b1, a5b1, avb1, avb3 and a6b4.6,60,61 Integrins may also complex with other receptor families in mediating cell motility and fibrogenesis, for example the tetraspanin family of receptors.62
HEPATIC STELLATE CELL ACTIVATION – THE COMMON PATHWAY LEADING TO HEPATIC FIBROSIS The identification of stellate cells as the key cellular source of extracellular matrix in liver has been a major advance. This distinct cell population, located in subendothelial space of Disse between hepatocytes and sinusoidal endothelial cells (Figure 6-1), represents onethird of the non-parenchymal population or about 15% of the total number of resident cells in normal liver.63 In normal liver they are the principal storage site for retinoids (vitamin A metabolites),
Inciting injury Recruitment of inflammatory cells T-cell NK cell Hepatocyte Stellate cell Kupffer cell
Expression of cytokines
Activated stellate cell
92
Stellate cell activation
which accounts for 40–70% of retinoids in the body. Most of the retinoids are in the form of retinyl esters and are confined to cytoplasmic droplets. Preferential expression of ECM genes in stellate cells has been confirmed in mechanistically distinct experiment models of injury. Recent studies have emphasized the heterogeneity of mesenchymal populations in the liver, with variable expression of neural,64 angiogenic,65 contractile,66 and even bone marrow-derived67 markers. Moreover, experimental genetic ‘marking’ of stellate cells by the expression of fluorescent proteins downstream of either fibrogenic or contractile gene promoters illustrates the plasticity of fibrogenic populations in vivo.68 In view of this capacity for ‘transdifferentiation’ between different mesenchymal cell lineages, and possibly even from epithelium,69 the key issue is whether fibrogenic cells express target molecules such as receptors or cytokines in sufficient concentrations in vivo to merit their targeting by diagnostic agents or antifibrotic compounds. Following liver injury of any etiology, stellate cells undergo a process known as ‘activation’, which is characterized by the transition of quiescent vitamin A-rich cells into proliferative, fibrogenic, and contractile myofibroblasts.54 Stellate cell activation is typically a result of complex interplay among ECM (Figure 6.2) and cellular (Figure 6.3) elements found in the local environment. It should be noted that activation most often occurs in the setting of hepatocellular injury and subsequent inflammation. Activation can be conceptually viewed as a two-stage process: initiation (also referred to as ‘preinflammatory’) and perpetuation54 (Figure 6-4). Initiation refers to early changes in gene expression and phenotype that render the cells responsive to other cytokines and stimuli, whereas perpetuation results from the effects of these stimuli on maintaining the activated phenotype and generating fibrosis. Initiation is largely due to paracrine stimulation, whereas perpetuation involves autocrine as well as paracrine loops.
Figure 6-3. Cellular response to wound healing. Most forms of liver injury result in hepatocyte injury followed by inflammation, which in turn leads to activation of hepatic stellate cells. Inflammatory effectors are multiple and include T cells, NK and NKT cells as well as Kupffer cells. These cells produce growth factors, cytokines, and chemokines that play an important role in stellate cell activation. Additionally, injury leads to disruption of the normal cellular environment, and also to stellate cell activation (right upper panel). Once activated, stellate cells themselves produce a variety of compounds, including growth factors, cytokines, chemokines, and vasoactive peptides. These substances have pleotropic effects in the local environment, including many which have autocrine effects on stellate cells themselves. One of the major results of stellate cell activation is extracellular matrix synthesis, as well as the production of matrix degrading enzymes.
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
PERPETUATION PROLIFERATION
CONTRACTILITY
INITIATION
ET-1 PDGF
FIBROGENESIS TGF-β1
RESOLUTION MMP-2 MATRIX DEGRADATION PDGF MCP-1 REVERSION?
PDGF MCP-1
CHEMOTAXIS APOPTOSIS LEUKOCYTE CHEMOTAXIS
RETINOID LOSS
Figure 6-4. Stellate cell activation. Stellate cell activation is a key pathogenic feature underlying liver fibrosis and cirrhosis. Multiple and varied stimuli contribute to the induction and maintenance of activation, including (but not limited to) cytokines, peptides, and the extracellular matrix itself. Key phenotypic features of activation include the production of extracellular matrix, loss of retinoids, proliferation, of up-regulation of smooth muscle proteins, secretion of peptides and cytokines (which have autocrine effects), and up-regulation of various cytokine and peptide receptors. Reprinted with permission from ref 54a.
Initiation Oxidant stress may be an early determinant of stellate cell activation. In hepatic injury, whether subclinical or overt, there is a perturbation of normal liver homeostasis, with extracellular release of either free radicals (i.e. ‘oxidant stress’), intracellular constituents, and/or cytokines and signaling molecules. Sources of these mediators may be circulating (i.e. endocrine), paracrine or autocrine. In particular, oxidant stress-mediated necrosis leading to stellate cell activation may underlie a variety of liver diseases, including hemachromatosis, alcoholic liver disease, viral hepatitis and nonalcoholic steatohepatitis (NASH).55,70,71 Liver injury is typically associated with infiltration of inflammatory cells, but even in their absence the liver contains sufficient resident macrophages (Kupffer cells) and natural killer cells (pit cells) to initiate local inflammation prior to the arrival of extrahepatic cells. In addition to oxidant stress, following early injury endothelial cells produce a splice variant of cellular fibronectin that is able to stimulate stellate activation.
Endothelial cells in early injury may also participate in the conversion of latent TGF-b1 to its active, profibrogenic form through the activation of plasmin. Whereas necrosis is considered a classic inflammatory and fibrogenic stimulus, recent findings also suggest that apoptosis may provoke a fibrogenic response in stellate cells. Apoptotic fragments released from hepatocytes are fibrogenic towards cultured stellate cells,72 and Fas-mediated hepatocyte apoptosis in vivo in experimental animals is also fibrogenic.73 Platelets in injured liver are a potent source of paracrine stimuli by generating multiple potentially important mediators, including PDGF, TGF-b1, and epidermal growth factor (EGF). Additionally, activated stellate cells have also been observed in primary and metastatic human tumors, as well as a murine model of metastatic melanoma to liver.74 In recent years, increasing interest has been focused on the molecular regulation of gene expression during early stellate cell activa-
93
Section I. Pathophysiology of the Liver
tion. There have been many advances in dissecting pathways of membrane and intracellular signaling and transcriptional gene regulation in activated hepatic stellate cells that are too numerous to detail here.75 A growing number of transcription factors may regulate stellate cell behavior, including peroxisomal proliferator-activated receptors (PPAR) a, b and g,76 retinoid receptors,77 NF-kB,78,79 Jun D,75 Krüppel-like factor 6 (previously called ‘Zf9’),80 Foxf1,81 and CRP282 among others.
Perpetuation After initiation, activated stellate cells undergo a series of phenotypic changes that collectively lead to the accumulation of ECM. These include proliferation; contractility; fibrogenesis; chemotaxis; matrix degradation; retinoid loss; and proinflammatory responses and cytokine release. The following sections detail the mechanisms underlying each of these events.
Proliferation An increase in the number of stellate cells has been documented after both human and experimental liver injury, in large part due to local proliferation. Following liver injury, many mitogenic factors as well as their cognate tyrosine kinase receptors are unregulated, primariliy through receptor tyrosine kinases.59 PDGF is the bestcharacterized and most potent mitogen towards stellate cells. Upregulation of PDGF receptor following liver injury enhances the responsiveness to autocrine PDGF, whose expression is also increased. The downstream signaling pathways involve ERK/MAP kinase, phosphoinositol 3 kinase (PI 3-kinase) and STAT-1 (signal transducers and activators of transcription) (see 3 for review). PDGF-induced proliferation correlates with increased intracellular Ca2+ and pH, raising the possibility that calcium channel blockers might modulate stellate cell mitogenesis or activation. Other stellate cell mitogens include endothelin-1 (ET-1),83,84 thrombin,85 FGF,86 and IGF,87,88 among others (see 89,90 for reviews). A recent study has documented increased sensitivity to ET-1 during activation,91 suggesting potentiation of autocrine/paracrine stimulation.
Contractility Contraction by stellate cells may be a major determinant of early and late increases in portal resistance during liver fibrosis. Activated stellate cells impede hepatic blood flow both by constricting individual sinusoids and by contracting the cirrhotic liver, as the collagenous bands typical of end-stage cirrhosis contain large numbers of activated stellate cells (see 66 for review). A key contractile stimulus towards stellate cells is ET-1.66 Other contractile agonists include arginine vasopressin, adrenomedullin, and eicosanoids.66 The regulation of stellate cell contraction is complex. The endothelium-derived relaxing factor nitric oxide (NO) appears to be an important relaxing factor in the sinusoid (although other factors, such as carbon monoxide, also play a role). The net contractile activity of stellate cells in vivo therefore reflects the relative strength of each of these opposing activities. Current evidence suggests that intrahepatic portal hypertension probably results from diminished NO (and/or other vasodilators) activity as well as increased stimulation by ET-1 (or other constrictors).66
94
The expression of smooth muscle a actin is increased during stellate cell activation. ET-1 and other vasoactive mediators increase their expression.83 Thus studies of contractile proteins in stellate cells may yield a therapeutic target for the treatment of intrahepatic portal hypertension.
Fibrogenesis Fibrogenesis is perhaps the key component of the stellate cell’s contribution to hepatic fibrosis. TGF-b1 is the most potent fibrogenic factor, with some fibrogenic activity documented for interleukin-1b, TNF, lipid peroxides, acetaldehyde, and others (see 2,6 for reviews). Because of its importance, TGF-b1 regulation has received considerable attention. TGF-b1 is up-regulated in experimental and human hepatic fibrosis. Although sources of this cytokine are many, autocrine expression is among the most important (see 2 for review). Several mechanisms underlie the increase in TGF-b1 expression by stellate cells during liver injury, including TGF-b transcriptional upregulation, activation of latent TGF-b1, increased TGF-b receptor expression, and up-regulation of TGF-b signaling components.92–96
Chemotaxis Stellate cells may accumulate both through proliferation and via directed migration into regions of injury, or chemotaxis. PDGF, the leukocyte chemoattractant MCP-1, and a growing family of chemokines have been identified as key stellate cell chemoattractants.97 In addition to tyrosine kinase receptors, new agents have been implicated in stellate cell migration, in particular tetraspanin receptors.3,62
Matrix Degradation A greater understanding of matrix degradation in liver is emerging. Quantitative and qualitative changes in the activity of MMPs and their inhibitors play a vital role in extracellular matrix remodeling in liver fibrogenesis (see ECM in the normal and fibrotic liver and Figure 6.2 above). As noted above, the net effect of changes in matrix degradation is the conversion of the low-density subendothelial matrix to one rich in interstitial collagens.
Retinoid Loss Stellate cell activation is accompanied by loss of their characteristic perinuclear retinoid (vitamin A) droplets. Although the intracellular form is largely retinyl esters, when retinol is exported from the cell during activation it is primarily as retinol, suggesting the possibility of intracellular hydrolysis of esters before being exported. Several nuclear retinoid receptors that bind intracellular retinoid ligands have been identified and their effects characterized in stellate cells.77,98
Proinflammatory Responses and Cytokine Release Hepatic stellate cells and sinusoidal endothelial cells have emerged as inflammatory effectors. Sinusoidal endothelial cells, normally fenestrated to allow rapid bidirectional transport of solutes between sinusoidal blood and parenchymal cells, may rapidly lose their fenestrations upon injury and express proinflammatory molecules, including ICAM-1, VEGF and adhesion molecules.74,99 Together with stellate cells, they activate angiogenic pathways in response to hypoxia associated with local injury or malignancy.74,97,100,101
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
Key inflammatory pathways converge on stellate cells, leading to fibrosis (see Figure 6-3). Thus, the cell type is a central mediator in inflammation, rather than just a passive target. Upon activation, they release chemokines97,102 and other leukocyte chemoattractants, proteinase-activated receptors,103 and up-regulate expression of key inflammatory receptors, including ICAM-1,104 chemokine receptors,105 and those mediating lipopolysaccharide (LPS) signaling, including Toll-like receptor 4.78 Stellate cells may also contribute to intrahepatic apoptosis of T lymphocytes.106 Remarkably, little attention has focused on the contribution of different lymphocyte subsets to hepatic fibrogenesis. Interest has increased recently, in part because of the observation that patients with HCV who are co-infected with HIV, as well as those who are immunosuppressed following liver transplantation, have accelerated fibrosis rates, implicating the immune system as a determinant of fibrogenesis. These observations have been supported by animal studies demonstrating that the immune phenotype regulates fibrogenesis independent of effects on injury, which in turn have led to efforts to map the genetic loci accounting for these differences.22 Most recently, CD8 lymphocytes have emerged as potential profibrogenic cells, based on their ability to induce early fibrogenesis following adoptive transfer to naïve SCID mice from animals with liver injury.107 Autocrine cytokines play vital roles in regulating the activation process of stellate cells. These cytokines include TGF-b1, PDGF, FGF, HGF, PAF, stem cell factor and ET-1, among others.59,90,97,108 Furthermore, stellate cells release neutrophil and monocyte chemoattractants, which can amplify inflammation in liver injury. These chemokines include colony-stimulating factor, monocyte chemotactic protein-1 (MCP-1), and cytokine-induced neutrophil chemoattractant (CINC).97 Anti-inflammatory cytokines produced by stellate cells have also been identified. Up-regulation of interleukin-10 (IL-10) occurs in early stellate cell activation. The anti-inflammatory effects of this cytokine are demonstrated by its ability to down-regulate TNF-a production from macrophages. Knockout mice lacking IL-10 have more severe hepatic fibrosis following CCl4 administration, and transgenic mice expressing IL-10 in liver have reduced fibrosis.107 Based on the consistent antifibrotic effect of IL-10 in experimental liver disease, a clinical trial was undertaken which failed to show an antifibrotic effect in patients with HCV infection, possibly because of marked increases in HCV replication109 (see Therapy of hepatic fibrosis, below).
DISEASE-SPECIFIC MECHANISMS REGULATING HEPATIC FIBROSIS – HCV AND NASH In addition to generic mechanisms of fibrogenesis common to all experimental and human liver disease, there has been progress in elucidating disease-specific mechanisms, in particular in hepatitis C (HCV) and NASH (non-alcoholic steatohepatitis). In HCV, stellate cells might be infectable by the virus because they express putative HCV receptors.104,110 Moreover, adenoviral transduction of HCV non-structural and core proteins induces stellate cell proliferation and the release of inflammatory signals.104 In HCV-infected liver chemokines and their receptors are up-regulated, stimulating lymphocyte recruitment.111 HCV proteins may also interact directly with sinusoidal endothelium.112
The increasing prevalence of obesity in the US and western Europe is associated with an alarming increase in NASH,39 leading to advanced fibrosis and cirrhosis. Leptin, a circulating adipogenic hormone that is proportionate to adipose mass in circulating blood, has been clearly linked to stellate cell fibrogenesis.113–115 Sources are likely to be both endocrine and autocrine, associated with enhanced signaling through the leptin receptor, which is up-regulated during stellate cell activation.113 Concurrently, down-regulation of adiponectin, a counterregulatory hormone, in obesity may amplify the fibrogenic activity of leptin. This possibility is supported by findings in mice lacking adiponectin, which have enhanced fibrosis following toxic liver injury.116
RESOLUTION OF LIVER FIBROSIS AND THE FATE OF ACTIVATED STELLATE CELLS During recovery from acute human and experimental liver injury the number of activated stellate cells decreases as tissue integrity is restored. Either reversion of stellate cell activation, or selective clearance of activated stellate cells by apoptosis, may explain the loss of activated cells in resolving liver injury. To date, evidence is strongest for stellate cell apoptosis in this setting. Apoptosis of stellate cells probably accounts for the decrease of activated stellate cells during resolution of hepatic fibrosis.49 Following injury, apoptosis may be inhibited by soluble factors and matrix components that are present during injury, whereas an apoptotic pathway otherwise represents a ‘default’ mode. Furthermore, cell death ligands, including TRAIL and fas, are expressed in liver injury, and activated stellate cells are more susceptible to TRAILmediated apoptosis.73,117,118 Another death receptor, nerve growth factor receptor (NGFR), is also expressed by activated stellate cells, and its stimulation with ligand drives apoptosis.119 Survival factors also regulate the net activity of stellate cell apoptosis. IGF-I promotes stellate cell survival via the PI3-K/c-Akt pathway and TNF-a has the same effect, but utilizes the NF-kB pathway.120,121 Molecules regulating matrix degradation appear closely linked to survival and apoptosis. Active MMP2 correlates closely with apoptosis, and in fact may be stimulated by it.122 Inhibition of MMP2 activity by TIMP-1 blocks apoptosis in response to a number of apoptotic stimuli.123 Interactions between stellate cells and the surrounding matrix also influence their propensity towards apoptosis, and this might partly explain the antiapoptotic activity of TIMP-1. Moreover, the fibrotic matrix may provide important survival signals to activated stellate cells. For example, animals expressing a mutant collagen I resistant to degradation have more sustained fibrosis and less stellate cell apoptosis following liver injury,48 and transgenic animals expressing TIMP-1 in liver have delayed resolution of fibrosis.124 Studies using gliotoxin,125 a fungal toxin that induces apoptosis in stellate cells, emphasize the role of this pathway in stellate cell removal during resolution of liver fibrosis. It is unknown whether an activated stellate cell can revert to a quiescent state in vivo, although it has been observed in culture. When stellate cells are grown on a basement membrane substratum (Matrigel) they remain quiescent, and plating of highly activated cells on this substratum down-regulates stellate cell activation.58,126
95
Section I. Pathophysiology of the Liver
METHODS TO MEASURE FIBROSIS OVERVIEW Measurement of fibrosis not only helps to stage the severity of disease, it allows serial determination of disease progression. The level of fibrosis may play an important role in clinical management and determine patients’ prognosis. For example, aggressive therapy is more appropriate in HCV-infected patients with advanced fibrosis. Further, the fibrosis progression rate is an important predictor of the time to develop cirrhosis.18 It is essential to measure fibrosis accurately, given the growing prospect of antifibrotic therapies and the need to track their efficacy. Moreover, with growing evidence that fibrosis is reversible, methods will need to assess both progression and regression accurately. For example, specific therapy leads to a reduction in fibrosis in a number of diseases, including autoimmune liver disease, hepatitis C, hepatitis B, and others.34,35,42,44,127–129 Percutaneous liver biopsy has traditionally been considered to be the gold standard test to assess liver fibrosis. However, a variety of non-invasive tests have been advanced as potential alternatives to biopsy. These include clinical signs, routine laboratory tests, quantitative assays of liver function, markers of extracellular matrix synthesis and/or degradation, and radiologic imaging studies. In addition to individual indicators of fibrosis, combination tests, and a number of models for predicting liver fibrosis have been developed. Individual and combination tests are discussed below. The ideal method to measure fibrosis would be simple, noninvasive, reproducible, inexpensive, accurate, and readily available. Unfortunately, none of the currently available approaches fulfills all of these criteria.
BEDSIDE DIAGNOSTIC TOOLS Clinical signs and symptoms of liver disease are frequently highlighted in assessing patients with liver disease, but these are of little value in detecting early, precirrhotic stages of liver fibrosis. In contrast, a number of clinical features can be utilized to assess whether cirrhosis with portal hypertension may be present. Signs of cirrhosis include spider angiomata, distension of abdominal wall veins, ascites, splenomegaly, muscle wasting, Dupuytren’s contractures (especially with ethanol-associated cirrhosis), gynecomastia and testicular atrophy in males, and palmar erythema. However, it is important to emphasize that even in patients with histologic cirrhosis, and in those with portal hypertension, these physical signs may not be present.
NON-INVASIVE MARKERS OF FIBROSIS Blood-Based Markers – Overview A wide variety of blood, serum, or plasma ‘markers’ for fibrosis have been proposed. There are several categories of marker or test. For example, some detect abnormalities in serum chemistries. Included in these types of test are aspartate aminotransferase (AST), alanine aminotransferase (ALT), g-glutamyl transferase (GGT), bilirubin, albumin, and a2-macroglobulin, among others. Moreover, some of these individual tests have been incorporated into simple and/or complex mathematical models or algorithms (see below).
96
Another major category of test includes those that are based specifically on the pathogenesis of fibrosis (see above). For example, proteins that are produced as a result of the fibrogenic process itself that have been studied as markers of fibrosis include procollagen I, fibronectin, tenascin, laminin, hyaluronic acid and others. Other markers have included cytokines (i.e. TGF-b1), connective tissue growth factor (CTGF), PDGF and others, matrix degrading enzymes (i.e. TIMP1), and others (Table 6-1). Finally, groups of tests, including those that utilize markers of fibrosis in combination with each other or in combination with other types of test, have been advanced in an attempt to detect and measure fibrosis. Ideally, a blood-based test should have both high sensitivity and high specificity. Many of the available tests have a high specificity (>95%) for advanced fibrosis. However, few (including algorithms) have great sensitivity to detect moderate levels of fibrosis. Moreover, a serum-based assay ideally should be linear over the full range of fibrosis, follow the natural history, and accurately reflect the effect of treatment.
Routine Laboratory Tests A number of studies have used routine laboratory tests in an attempt to determine whether a patient may have advanced liver disease, in particular to exclude or confirm portal hypertension and/or esophageal varices.130,131 Although tests such as the prothrombin time, albumin level, and portal vein diameter (measured by ultrasound) have all been associated with varices, studies have been remarkably consistent in their identification of the platelet count as
Table 6-1. Cytokines, Growth Factors, Peptides, Proteases, and other Components Important in Hepatic Fibrogenesis Cytokines
Growth factors
Peptides
Transforming growth factor-b Transforming growth factor-a Interleukin-1 Interleukin-4 *Interleukin-6 Interleukin-10 Interleukin-13 *Monocyte chemotactic factor
Transforming growth factor-b
Endothelin-1
Transforming growth factor-a
Norepinephrine
*Insulin-like growth factor (I, II) *Platelet-derived growth factor *Fibroblast growth factor Vascular endothelial growth factor Hepatocyte growth factor Connective tissue growth factor
Angiotensin II
Proteases and their inhibitors Matrix-metalloproteinase-1 (interstitial collagenase) Matrix-metalloproteinase-2 (gelatinase A) Matrix-metalloproteinase-3 (stromelysin-1) Matrix-metalloproteinase-7 (matrilysin) Matrix-metalloproteinase-8 Matrix-metalloproteinase-9 (gelatinase B) Matrix-metalloproteinase-10 (stromelysin-2) Tissue inhibitor of metalloproteinase-1
Miscellaneous Thrombospondin (1,2) Leptin Activin A *Thrombin Osteopontin
Agents may have direct effects on hepatic stellate cells, or indirect effects in the wounding environment. *Compounds whose effect is largely via stimulation of proliferation.
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
the best single predictor of esophageal varices. For example, in one study, cirrhotics without splenomegaly on physical examination and with a platelet count >88 000/mm3 had a risk of large esophageal varices of 7.2%, whereas the risk was 28% if the platelet count was less than 88 000/mm3.130 An AST/ALT ratio >1 has been proposed to indicate the presence of cirrhosis.132 In one study of patients with HCV, a ratio >1 had 100% specificity and positive predictive value for distinguishing cirrhotic from non-cirrhotic patients, with a 53.2% sensitivity and 80.7% negative predictive value.133 In addition, the ratio correlated positively with the stage of fibrosis, but not with the grade of activity or other biochemical indices. Of cirrhotic patients, 17% had no clinical or biochemical evidence of chronic liver disease except for an elevated AST/ALT ratio. In another study, the AST/ALT ratio had 81.3% sensitivity and 55.3% specificity in identifying cirrhotic patients who died within 1 year of follow-up.132,134 In a further attempt to develop non-invasive tools for the measurement of liver fibrosis, Forns and coworkers developed a model using data from HCV patients that included age, GGT, cholesterol, and platelet count.135 This model was developed with the intention to differentiate patients with significant fibrosis from those without. The sensitivity for detecting METAVIR F2–F4 fibrosis was 94%, and the presence of significant F2–F4 fibrosis could be excluded with high accuracy (negative predictive value of 96%).135 Likewise, Wai et al.136 constructed a simple model utilizing routine laboratory data (Table 6-2). The authors devised a novel index, termed the AST to platelet ratio index, or APRI, which is the AST level/upper limit of normal (ULN) divided by the platelet count (109/l) multiplied by 100. The sensitivity and specificity for fibrosis of the APRI value depended on the cut-offs used. Using an APRI value of 1.50, the positive and negative predictive values for significant fibrosis (Ishak score = 3) were 91% and 65%, respectively, whereas for cirrhosis and an APRI of 2.00, the positive and negative predictive values were 65% and 95%, respectively. Thus for a hypothetical patient, if the AST was 90 IU/l (and the ULN 45) and platelet count was 100 ¥ 109/l, then the APRI would be 2.00. This means that the patient has essentially a 90% chance of having significant fibrosis, and somewhat less likelihood of having cirrhosis. However, cirrhosis could not be excluded with certainty. Although the APRI is attractive because of its simplicity, it can neither definitively diagnose nor exclude cirrhosis, and it will not identify patients with early fibrosis. Other simple quantitative systems based on routine laboratory values have been developed. One early example was the ‘PGA index’, which combined prothrombin time, GGT and apolipoprotein A1 (Table 6-2); this test was examined in patients with alcoholic cirrhosis.137 The diagnostic accuracy of this index was later improved by the addition of a2-macroglobulin (and hence termed the ‘PGAA index’).138 The test characteristics of many of these indirect assays have been derived from datasets, but have not been validated on independent datasets. More complicated algorithms based on commonly available laboratory tests include the ‘Fibrotest,’ reported by the French MULTIVIRC group.25 This group used mathematical modeling to develop an algorithm including five different markers to predict fibrosis (the markers selected were a2-macroglobulin, haptoglobin, GGT, apolipoprotein A1, and total bilirubin). This index predicted a spe-
Table 6-2. Combined Panels of Blood Markers used to Detect Liver Fibrosis Panel
Components
References
AST/ALT § Forns APRI PGA index Fibrotest
AST/ALT Platelets, GGT, cholesterol AST, Platelets Platelets, GGT, apolipoprotein A GGT, haptoglobin, bilirubin, apolipoprotein A, a2-macroglobulin Hyaluronic acid, TIMP-1, a2-macroglobulin ECM proteins AST, cholesterol, HOMA-IR
132–134 135 136 137–138
Fibrospect *ELF FPI
25, 139–142 146 147
ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, g-glutamyl transpeptidase; APRI, AST to platelet ratio index; TIMP-1, tissue inhibitor of metalloproteinase 1; ECM, extracellular matrix; ELF, European liver fibrosis; FPI, fibrosis probability index; HOMA-IR, insulin resistance by the homeostasis model assessment. § Also includes age in the panel. *Components tested include collagen IV, collagen VI, amino terminal propeptide of type III collagen (PIIINP), matrix metalloproteinase 2 (MMP-2), matrix metalloproteinase 9 (MMP-9), tissue inhibitor of matrix metalloproteinase 1 (TIMP-1), tenascin, laminin, and hyaluronic acid (HA).
cific biopsy category in 46% of patients139 and has been validated in a number of hepatitis C patient cohorts, having been found to have an area under the receiver operator characteristic (AUROC) curve of 0.73–0.87.140 The addition of ALT to the marker panel allows for prediction of METAVIR necroinflammatory activity.140 The panel has also been examined in other liver disease cohorts.141,142 Limitations of this panel in fibrosis include false positive results due to increases in bilirubin or decreases in haptoglobin, for example from hemolysis secondary to ribavirin therapy. Likewise, false positive results may also occur in situations where there is hyperbilirubinemia, such as Gilbert’s disease and cholestasis. Acute inflammation may also affect the results of the test owing to changes in a2macroglobulin or increases in haptoglobin. Currently, it is unclear whether the ‘fibrotest’ assay meets sufficiently rigorous criteria, given a predictive value of only 46%, for routine clinical use.
Tests Using Extracellular Matrix/Fibrosis Markers Analyses of serum markers of extracellular matrix/fibrosis include many proteins important in fibrogenesis, ECM constituents (i.e. fibronectin, collagen I, collagen IV, collagen VI, amino terminal propeptide of type III collagen (PIIINP), tenascin, and hyaluronic acid, metalloproteinases (including many of those listed in Table 6-1), inhibitors of matrix metalloproteinases (i.e. TIMP-1, TIMP-2), and other proteins, peptides, and cytokines, as highlighted in Table 6-1. Although many tests have been studied individually, they are generally not sensitive for detection of fibrosis143,144 (see 145 for review).
Tests Using Combinations of Extracellular Matrix and/or Routine Markers
A combination test including hyaluronic acid, TIMP1, and a2macroglobulin was examined in a cohort of 294 patients with HCV infection and subsequently validated in a second cohort of 402 patients146 (‘Fibrospect’, Table 6-2). This had a combined AUROC of 0.831 for METAVIR F2–F4 fibrosis. The positive and negative predictive values were 74.3% and 75.8%, respectively, with an accuracy of 75%. This three-marker panel thus may help differentiate
97
Section I. Pathophysiology of the Liver
patients with HCV infection with moderate/severe fibrosis from those with no/mild fibrosis, although it was not possible to differentiate specific stages accurately. Another combination test was developed by the European Liver Fibrosis (ELF) Study Group.147 This group examined collagen IV, collagen VI, PIIINP, matrix metalloproteinase 2 (MMP-2), matrix metalloproteinase 9 (MMP-9), TIMP-1, tenascin, laminin, and hyaluronic acid (HA). The study was unique in that it examined patients with a wide variety of liver diseases, including those with chronic hepatitis C virus infection (n = 496), alcoholic liver disease (n = 64), non-alcoholic fatty liver disease (n = 61), chronic hepatitis B virus infection (n = 61), primary biliary cirrhosis or primary sclerosing cholangitis (n = 53), recurrent liver disease post orthotopic liver transplantation (n = 48), autoimmune hepatitis (n = 45), hemochromatosis (n = 32), cryptogenic cirrhosis (n = 19), both hepatitis B and C (n = 4), and other or no known diagnosis (n = 138); the cohort also had a wide distribution of fibrosis stages (Scheuer fibrosis stages were as follows: stage 0 = 24.6%; stage 1 = 35.5%; stage 2 = 13.4%; stage 3 = 14.9%; and stage 4 = 11.8%). An algorithm was developed that detected the upper third of fibrosis groups (Scheuer stages 2, 3, and 4) with a sensitivity of 90% and accurately detected the absence of fibrosis (Scheuer stages 0, 1), with a negative predictive value for this level of fibrosis of 92%. The AUC of a receiver operating characteristic (ROC) plot was 0.804. Interestingly, the addition of clinical chemistry tests including liver function tests, or hematological indices including platelet count and prothrombin time, did not improve test performance. The test appeared to be best in patients with hepatitis C, non-alcoholic fatty liver disease and alcoholic liver disease. The inclusion of patients with multiple etiologies of liver disease, although appealing, has the potential to limit the accuracy of these and other panels, as the characteristics of specific assays may be disease specific. Another model, including AST, cholesterol, and insulin resistance (as well as age and an estimate of past alcohol intake) in patients with HCV147a found that the sensitivity for detection of advanced fibrosis depended on the index value used. At a low probability index, the sensitivity for predicting significant fibrosis was high, but specificity was low, while at a high probability index, sensitivity for significant fibrosis was low, but specificity was high.
Proteomics With the recent explosion in proteomics, proteomic approaches have attempted to identify unique protein fingerprints in patients with liver disease. Various platforms are available, including those that measure protein expression, protein–protein interactions, or even enzymatic activity. The majority of approaches have used highthroughput technologies to identify novel protein expression patterns. For example, a recent study in 46 patients with chronic hepatitis B identified 30 proteomic features predictive of significant fibrosis (Ishak stage = 3) and cirrhosis. The AUROC for this analysis was 0.906 and 0.921, for advanced fibrosis and cirrhosis, respectively.148 Another study in 193 patients with chronic hepatitis C identified eight peaks that differentiated METAVIR fibrosis stages with an AUROC of 0.88; this was compared to an AUROC 0.81 for the Fibrotest.149 Another report in patients with HCV fibrosis identified several serum proteins to be differentially regulated.150 In
98
this study, patients with advanced fibrosis had elevated levels of a2macroglobulin, haptoglobin, and albumin, but apolipoprotein AI, apolipoprotein A-IV, complement C4, and serum retinol-binding protein were reduced. Another approach has included measurement of labeled Nglycans found in serum.151 The technique exploits the ability to analyze the desialylated total serum N-glycome on a DNA analyzer. The authors focused on cirrhosis (primarily ethanol induced), demonstrating unique patterns of serum N-glycans in those with cirrhosis compared to those with chronic liver disease alone. It was postulated that in cirrhotic livers characteristic N-glycans with a bisecting GlcNAc residue were prominent. In normal liver, the enzyme responsible for this modification, N-acetylglucosaminyl transferase III (GnT-III), is found only in non-parenchymal cells, but in regenerating liver (two-thirds partial hepatectomy) this enzyme is produced in hepatocytes. Thus, GnT-III expression is presumably a manifestation of hepatocellular regeneration, reflected by regenerative nodules. This approach was most sensitive for the detection of cirrhosis and was also able to exclude cirrhosis with great accuracy. When combined with the commercially available Fibrotest this test had 100% specificity and 75% sensitivity for diagnosing compensated cirrhosis.151
Summary of Blood-based Markers A key advantage of serum markers to detect fibrosis is their noninvasiveness. Additionally, it has been argued that serum markers overcome sampling problems associated with liver biopsy. However, these approaches have several drawbacks. First, most of the studies examining serum markers have been performed in cohorts of patients that have been biased toward advanced fibrosis/cirrhosis. A further problem is that the currently proposed serum marker algorithms use dichotomous rather than continuous scales. The dichotomous nature of these variables would be less problematic if there were clear clinical associations, for example if prognosis or treatment response were highly linked to stage 0–1 versus stages 2–4. In the absence of clinical correlates between dichotomous variables and outcomes, it remains important to diagnose the different stages of fibrosis accurately (0–4). Unfortunately, current tests and algorithms are unable to do this, and perhaps most importantly, the tests do not differentiate between intermediate levels of fibrosis. Thus, although assessments of fibrosis with approaches that use serum markers have great appeal, and indeed, in some areas the tests have begun to replace liver biopsy. Further investigation is required to optimize these tests.
Imaging Tests A wide variety of radiographic tests have been used to image patients with fibrosis/cirrhosis. Included in this group are ultrasound, CT, and MRI. In general, these tests are capable of detecting evidence of portal hypertension, thus they have the ability to detect advanced disease. As currently used in clinical practice, however, they are insensitive for the detection of moderate degrees of fibrosis. Transient elastography, which uses pulse-echo ultrasound acquisitions to measure liver stiffness and predict fibrosis stage, has gained interest as a method to quantify fibrosis as it appears that liver ‘stiffness’ may accompany the fibrogenic response.152 In a prospective multicenter study of 327 chronic HCV patients, the AUROCs for
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
METAVIR stage F2–F4 and cirrhosis were 0.79 and 0.97, respectively.153 In a separate study of 183 chronic HCV patients, transient elastography compared favorably with the Fibrotest and APRI (AUROC for F2–F4 = 0.83, 0.85 and 0.78, for transient transient elastography, Fibrotest and APRI, respectively).154 When transient elastography was combined with the Fibrotest, the predictive value for fibrosis stage F2–F4 was improved, with an AUROC of 0.88.154 Transient elastography (Fibroscan) reportedly offers good reproducibility with low inter- and intraobserver variability. The procedure is performed by obtaining multiple validated measurements in each patient, further reducing the potential for sampling errors. The depth of measurement from the skin surface is between 25 and 65 mm, raising the possibility that this technique may be difficult to use in obese patients or those with ascites. However, newer probes are being developed for obese patients, and further investigation is expected. Finally, it would theoretically be desirable to utilize advances in the molecular understanding of liver fibrosis to image the liver. For example, the number of activated stellate cells, which reflect fibrogenic activity, might be identified by tagging them with cell-specific markers.155 Alternatively, matrix or matrix turnover could be labeled using molecular tools. Although such approaches are appealing, they remain experimental at present.
in the other. Finally, in 10% of subjects, stage 0–2 disease was identified in one lobe and stage 3–4 fibrosis was found in the other. Similar variability was reported in another study in patients with fatty liver disease.161 There are several other limitations of liver biopsy. Quantification of fibrosis in biopsies is subject to significant interobserver variation. In chronic hepatitis C, for example, standardized grading systems, including Knodell, METAVIR, Scheuer or Ishak, are concordant in only 70–80% of samples. Specimen quality is very important, with smaller samples leading to an underestimation of disease severity.162 A recent study created digitized virtual image biopsy specimens of varying length from large liver sections, and revealed that 75% of 25-mm biopsy specimens were correctly classified using the METAVIR staging system, compared to only 65% for biopsies 15 mm long.163 Interestingly, a recent study noted that the experience of the pathologist may have more influence on interobserver agreement than specimen length.164 Another major problem with using liver biopsy or serum markers to quantify fibrosis is that all of the currently utilized grading systems use a simple linear numerical scoring approach, implying that they represent linear changes in fibrosis content. Such an inference is highly inaccurate, as METAVIR stage 4 fibrosis does not represent twice as much fibrosis as stage 2, but rather a 5–20-fold difference.
Tests of Liver Function A variety of bona fide liver function tests have been used to assess liver fibrosis and cirrhosis. Such tests generally measure advanced disease and several depend on perfusion, such as indocyanine green, sorbitol and galactose clearance tests, or tests such as the 13C– galactose breath test and the 13C–aminopyrine breath test that depend on the functional capacity of the liver.156–158 Another test, the MEGX test, which measures monoethylglycinexylidide (MEGX) formation after the administration of lidocaine, depends upon the activity hepatic cytochrome P450 3A4 isoenzyme (which catalyzes oxidative N-de-ethylation of lidocaine to MEGX.159 The MEGX test has a sensitivity and specificity in the 80% range for distinguishing chronic hepatitis from cirrhosis in comparison to standard liver tests.159 Unfortunately, although the MEGX test and other function tests may predict prognosis in cirrhotic patients, they are insensitive for quantifying fibrosis in patients with less advanced disease.156–158
Liver Biopsy Percutaneous liver biopsy has traditionally been considered to be the gold standard test to measure fibrosis. Although there is great experience with liver biopsy, this procedure is time consuming, inconvenient, uncomfortable, invasive, and makes both patients and physicians anxious. Further, liver biopsy can be associated with substantial sampling-error (see Chapter 12 for further details about liver biopsy). In a recent study in which 124 patients with chronic HCV infection underwent laparoscopy-guided biopsy of each the right and left hepatic lobes, 33.1% had a difference of at least one histologic stage (modified Scheuer system) between the two lobes.160 Furthermore, in 18 study subjects a stage consistent with cirrhosis was found in one lobe, whereas stage 3 fibrosis was reported
TREATMENT OF FIBROSIS Specific therapy for the treatment of liver fibrosis is attractive because the scarring response leads to many if not all of the complications of chronic liver disease, in particular impaired synthetic function, liver failure, and perhaps hepatocellular cancer. Fibrosis, particularly in its advanced stages, may also contribute to portal hypertension, by preventing blood flow through fibrotic nodules. Although attempts have been made previously to treat specifically the ‘fibrosis’ component of liver disease, these approaches have generally been unsuccessful. Thus, there remains a major unmet need for novel and effective antifibrotic therapy. Advances in elucidating the pathogenesis of fibrosis have led to renewed efforts in this area. Additionally, data indicating that fibrosis is reversible have helped fuel this effort (Figure 6.5). Although fibrosis is commonly accepted as the precursor to cirrhosis, it is not clear that mortality risk increases directly with the stage of fibrosis, until the patient actually becomes cirrhotic. Even with established cirrhosis, in a cohort of patients with chronic HCV infection Fattovich and colleagues demonstrated that complications of cirrhosis developed over prolonged periods, and only when complications occurred was mortality increased.7 The most effective ‘antifibrotic’ therapies are currently those that treat or remove the underlying stimulus to fibrogenesis (Table 6-3). In addition, preclinical and human clinical studies have highlighted a number of therapies that may abrogate fibrogenesis without affecting the underlying disease, by targeting specific steps in the fibrogenic response. Anti-inflammatory therapies have been based on the knowledge that inflammation drives the fibrogenic cascade. Other treatments have attempted to inhibit cellular injury or focused on stellate cell activation, whereas others have targeted collagen syn-
99
Section I. Pathophysiology of the Liver
Figure 6-5. Reversal of fibrosis. An example of reversal of advanced fibrosis (cirrhosis in this situation) is depicted. A liver biopsy prior to lamivudine treatment is shown (upper panel and left panel). After treatment with lamivudine, liver biopsy was repeated and reveals almost complete dissolution of fibrosis. Data similar to these have been published in autoimmune liver disease, hepatitis C, alcoholic hepatitis, hepatitis B, and others. (Reprinted with permission, Wanless, et al: Arch Pathol Lab Med 2001;124:1599–1607.)
Inset in A
B
A
C
Table 6-4. Diseases and Therapies in which there is Strong Evidence that Treatment Reduces Liver Fibrosis
Table 6-3. Approaches to Treat Liver Fibrosis Approach
Example
Disease
Therapy
References
Remove injurious agent Anti-inflammatory agents Antioxidants Cytoprotective agents Inhibit stellate cell activation Inhibit stellate cell activation phenotypes (fibrogenesis)
Eradication of HBV Corticosteroids in AIH PPC in alcoholic hepatitis Ursodeoxycholic acid Interferon-g Colchicine
Hepatitis B Hepatitis C Bile duct obstruction Autoimmune hepatitis Hemochromatosis Alcoholic hepatitis
Lamivudine Interferon-a* Surgical decompression Corticosteroids Iron depletion Corticosteroids
33–35,129 42 43 44 165,166 168,169
Note: some approaches have not been demonstrated to be successful. AIH, autoimmune hepatitis; PPC, polyenylphosphatidylcholine.
thesis and matrix deposition. The following section highlights human studies in these areas.
THERAPIES DIRECTED AT THE UNDERLYING DISEASE In many forms of liver disease treatment of the underlying inciting lesion leads to an improvement in fibrosis (Table 6-4). For example, eradication or inhibition of HBV33,34,129 or HCV replication42 leads to reversion of fibrosis, even in patients with histological cirrhosis. Fibrosis reverts in patients with hemochromatosis during iron depletion,165,166 after corticosteroid therapy in autoimmune hepatitis,165,166 and in patients with secondary biliary cirrhosis after decompression of bile duct obstruction.43 In a preliminary report in patients with non-alcoholic steatohepatitis (NASH) treated with the peroxisomal proliferator active receptor (PPAR)-g agonist rosiglitazone both steatosis and fibrosis were reduced.167
ANTI-INFLAMMATORY COMPOUNDS Many liver diseases, such as HCV disease, have an important inflammatory component. Inflammation in these disorders typically drives stellate cell activation and fibrogenesis, and it is these diseases in particular that have been studied in order to evaluate the efficacy of anti-inflammatory drugs.
Corticosteroids Classic examples of the benefits of steroids include autoimmune hepatitis and alcoholic hepatitis. In patients with autoimmune hep-
100
*or PEG-interferon-a, with or without ribavirin. MTX, methotrexate; PPAR, peroxisomal proliferator-activated receptor.
atitis who respond to medical treatment (prednisone or equivalent) advanced fibrosis and cirrhosis are reversible.44 Fibrosis may improve in patients with alcoholic liver disease who respond to corticosteroids.168,169 Thus, corticosteroids appear to have antifibrotic effects in patients with certain liver disorders.
Interleukin-10 (IL-10) Interleukin (IL)-10 has both anti-inflammatory and immunosuppressive effects. IL-10 has been shown to reduce the production of proinflammatory cytokines, such as TNF-a, IL-1, interferon-g, and IL-2 from T cells. These cytokines belong to the Th1 family. Endogenous IL-10 reduces the intrahepatic inflammatory response, shifts the cytokine milieu towards a Th2 predominance, and reduces fibrosis in several in vivo models of liver injury.170 It was hypothesized that in vivo administration of IL-10 in patients with hepatitis C virus infection may have an anti-inflammatory and hence an antifibrotic effect.109 Therefore, 30 patients with advanced HCV-mediated fibrosis who had failed standard interferon-a-based antiviral therapy were enrolled in a 12-month treatment trial of IL-10 given daily or thrice weekly subcutaneously. In 13 of 28 of these patients the hepatic inflammation score decreased by at least two points (Ishak score) and 11 of 28 had a reduction in fibrosis score (mean change from 5.0 ± 0.2 to 4.5 ± 0.3, p < 0.05). However, serum HCV RNA levels increased during therapy (mean HCV RNA at day 0: 12.3 ± 3.0 mEq/ml; and at 12 months: 38 mEq/ml; p < 0.05). Changes in liver histology and HCV RNA levels were accompanied by an apparent shift in toward a Th2-predominant lymphocyte phenotype, as had been originally hypothesized. Long-term therapy with
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
IL-10 decreased hepatic inflammatory activity and appeared to have an inhibitory overall effect on fibrosis. Thus, although IL-10 appears to reduce inflammation and fibrosis, it has not been pursued as an antifibrotic compound because of putative detrimental virologic effects.
Miscellaneous Anti-inflammatory Drugs A number of other anti-inflammatory approaches have gained attention as therapies for fibrosis. Because TNF-a drives inflammation in many diseases, and because TNF-a is up-regulated in liver diseases (such as alcoholic hepatitis), an anti-TNF-a compound should theoretically reduce inflammation and hence the stimulus for fibrosis.171–174 Preliminary analyses from a study of anti-TNF-a therapy suggest an improvement in inflammation which presumably precedes fibrosis in patients with alcoholic hepatitis,173 although there was little effect of anti-TNF-a on hepatic fibrosis. Such data, in addition to the favorable effects of the TNF-a inhibitor pentoxifylline on mortality, provide the rationale for future study in patients with alcoholic hepatitis. These approaches and others that broadly inhibit inflammation must be considered cautiously because of the concern about disruption of the immune system, with increased risk of infection. Penicillamine is a heavy metal chelating compound that has been proposed to have anti-inflammatory and thus antifibrogenic effects.175 However, this compound had no effect on fibrogenesis in patients with primary biliary cirrhosis.176,177 Metrothrexate is thought to have anti-inflammatory properties, but interestingly has typically been considered to be profibrogenic in the liver for patients receiving methotrexate for treatment of rheumatologic diseases178 (although it is noteworthy that the risk of fibrosis progression may be less prominent than typically believed178,179). Metrothrexate has been studied in patients with primary biliary cirrhosis. Although some investigators have reported highly favorable effects in this disease, including improvement of the disease and reversion of fibrosis,180 the majority of the data on methotrexate have either been negative181,182 or show that the drug’s effects have been marginal, either alone181 or in combination with colchicine.183 It is important to emphasize that if methotrexate is used to treat patients with primary biliary cirrhosis, this must be undertaken by an experienced hepatologist.
ANTIOXIDANT AGENTS Oxidative stress is thought to play an important role in injury, stellate cell activation, and the stimulation of extracellular matrix production, as discussed above. Thus, a wide variety of antioxidants have received attention as potential antifibrotics.
Polyenylphosphatidylcholine Polyenylphosphatidylcholine is a mixture of polyunsaturated phosphatidylcholines, extracted from soybeans. This compound has antioxidant properties and oxidant stress (see above) is thought to be important in the inflammatory and fibrogenic response to injury, particularly in alcoholic liver disease. As oxidative stress leads to lipid peroxidation, and lipid peroxidation is injurious at the level of the cell membrane, phosphatidylcholine has been proposed to be protective against injury to cell membranes, resulting in reduced cellular injury and fibrogenesis.184
A VA cooperative multicenter clinical trial examined the effect of polyenylphosphatidylcholine in 789 patients with alcoholic hepatitis who had a very high daily average alcohol intake (16 drinks/day).185 Subjects were randomized to either polyenylphosphatidylcholine or placebo for 2 years. The long period of treatment is noteworthy as it is likely that long periods of treatment will be required to effect changes in the liver fibrosis. Many subjects substantially reduced their ethanol consumption during the trial, which probably accounted for improvement in fibrosis in the control group, making it difficult to demonstrate an improvement in fibrosis in the polyenylphosphatidylcholine group. Thus, overall, polyenylphosphatidylcholine failed to lead to significant improvement in fibrosis.
Silymarin Silymarin is derived from the milk thistle Silybum marianum. This extract has been shown to reduce lipid peroxidation and inhibit fibrogenesis in rodent animal models,186,187 as well as in baboons.188 It has been tested in several carefully performed human clinical trials, although fibrosis was not used as an endpoint. The compound has been found to be safe, but reportedly has mixed effects.189,190 In one study examining silymarin in alcoholics189 mortality was reduced; in addition, patients with early stages of cirrhosis also appeared to benefit. However, in another study in alcoholics no survival benefit could be identified.190 In both of these trials, silymarin appeared to be safe. Thus, although the agent is safe and is commonly used by patients with fibrosing liver disease, there is limited evidence of its efficacy.
Other Antioxidants Antioxidants such as vitamin E have been examined in animal models191 as well as in humans.192–195 A vitamin E precursor, D-atocopherol (1200 IU/day for 8 weeks), was studied in six patients with hepatitis C virus infection who failed to respond to interferon therapy,192 and was found to inhibit stellate cell activation but did not affect fibrosis. A randomized controlled trial examined vitamin E in patients with mild to moderate alcoholic hepatitis and found that vitamin E reduced serum hyaluronic acid, but did not lead to a change in type III collagen.194 Combined antioxidant therapy, including vitamin E, had no effect on outcome in patients with severe alcoholic hepatitis, although fibrosis was not specifically addressed.195 Malotilate is another potential cytoprotective agent, perhaps acting via inhibition of cytochrome P450 2E1; in addition, this compound may have anti-inflammatory properties. In patients with primary biliary cirrhosis, although it was found to diminish plasma cell and lymphocytic infiltrate and piecemeal necrosis, it had no significant effect on fibrogenesis.196 Another agent used to antagonize oxidative stress is S-adenosylmethionine; this compound is important in the synthesis of the antioxidant glutathione. The enzyme (methionine adenosyltransferase) responsible for its synthesis is reduced in the injured liver;197 thus it has been hypothesized that if S-adenosylmethionine were replaced, then injury and fibrogenesis might be slowed. S-adenosylmethionine has been tested in a large randomized trial in patients with alcoholic cirrhosis.198 There was an improvement in overall mortality/need for liver transplantation in the treatment arm, espe-
101
Section I. Pathophysiology of the Liver
cially in patients with Child’s A/B cirrhosis, although histologic assessment of fibrosis was not specifically assessed.198 Propylthiouracil is an antithyroid drug that reacts with some of the oxidizing species derived from the respiratory burst and may thus be protective in alcoholic liver disease, a disease in which an increase in hepatic oxygen consumption may predispose the liver to ischemic injury. Thus, propylthiouracil has been tested in a number of randomized clinical trials in patients with alcoholic liver disease. Unfortunately, a systematic review and meta-analysis found that propylthiouracil had no benefit in fibrosis, or in any other outcome measured.199
CYTOPROTECTIVE AGENTS Ursodeoxycholic Acid Ursodeoxycholic acid binds to hepatocyte membranes, where it presumably stabilizes them and is thus cytoprotective. This cytoprotective action in turn theoretically reduces inflammation and may in turn have a beneficial effect on fibrogenesis.200 Neither experimental data nor human studies indicate a primary antifibrotic effect of ursodeoxycholic acid in the liver, but the compound has been examined extensively.201–209 Ursodeoxycholic acid has been studied in patients with cystic fibrosis, primary biliary injury (primary biliary cirrhosis, primary sclerosing cholangitis and progressive familial intrahepatic cholestasis), and miscellaneous liver diseases. Results with ursodeoxycholic acid in these conditions have been mixed. Both symptomatic and biochemical improvement has been observed in these diseases, in particular the biliary diseases, but data on histologic improvement (and survival) have not been consistent. For example, in a randomized controlled trial in patients with primary biliary cirrhosis, ursodeoxycholic acid led to reduced fibrosis in those with mild disease but had no effect on those with severe disease.202 In another study, survival was improved in patients treated with ursodeoxycholic acid but fibrosis was not improved.206 Further, in a histopathologic study of 54 patients with primary biliary cirrhosis and paired liver biopsies, 4 years of ursodeoxycholic acid therapy was associated with a significant decrease in the prevalence of florid interlobular bile duct lesions, lobular inflammation, and necrosis. Worsening of fibrosis was observed in 14 patients (the majority had only a onegrade progression in fibrosis score), whereas stabilization was noted in the 40 remaining patients.207 A recent combined analysis of the histologic effect of ursodeoxycholic acid on paired liver biopsies including a total of 367 patients (200 ursodeoxycholic acid and 167 placebo) revealed that subpopulations of patients with initial earlystage disease may benefit from therapy.208 Results of meta-analyses examining ursodeoxycholic acid have been mixed, and have largely reported that ursodeoxycholic acid is not effective in primary biliary cirrhosis.205 The aggregate data suggest that ursodeoxycholic acid may impede the progression of fibrosis in primary biliary cirrhosis via effects on (bile duct) inflammation, particularly if given early in the disease course. It should be emphasized that so far as we know, ursodeoxycholic acid is extremely safe. Thus, although it is also expensive, the available data justify its use as an antifibrotic in patients with primary biliary cirrhosis. Ursodeoxycholic acid has also been studied in children with progressive familial intrahepatic cholestasis,203 where it appeared to
102
improve fibrogenesis. Additionally, a small series indicated that seven of 10 patients with cystic fibrosis treated with ursodeoxycholic acid had a reduction in liver fibrosis.204 Although these effects are promising, it should be emphasized that the numbers of patients studied has been small. Finally, in a large randomized controlled trial of ursodeoxycholic acid in patients with non-alcoholic steatohepatitis over 2 years, including 107 subjects who had paired biopsy data, there was no improvement in fibrosis.209
MISCELLANEOUS Anabolic–androgenic steroids such as oxandrolone have been examined in randomized trials including patients with alcoholic liver disease, but have not been found to have significant effects on fibrosis (or other outcomes).210
Stellate Cell-Specific Compounds Interferon-g A wealth of data supports the antifibrotic potential of interferon-g. The interferons consist of a family of three major isoforms, a, b and g. These isoforms are unique, not only in structure but also in their biologic actions. Interferons-a and -b bind to the same receptor, whereas interferon-g binds to a different receptor. Interferon-a has more potent antiviral effects than does interferon-g, and interferong has been shown to specifically inhibit extracellular matrix synthesis in isolated cells, including stellate cells.211,212 Interferon-g potently inhibits multiple aspects of stellate cell activation,211,212 and appears to have antifibrotic effects in patients with pulmonary fibrosis.213 Such data have generated considerable enthusiasm about the use of interferon-g in patients with hepatic fibrosis, although there is theoretical concern about its use because it is proinflammatory, and moreover its overexpression in the liver leads to chronic hepatitis.214 None the less, it has now been tested in humans with fibrosing liver disease, and appears to be safe.215 Although this pilot study provides a firm foundation highlighting the potential use of interferon-g in patients, larger studies are needed to prove a therapeutic benefit.
COMPOUNDS THAT INHIBIT FIBROGENESIS Colchicine Colchicine is a plant alkaloid that inhibits polymerization of microtubules, a process that in turn is believed to be required for collagen secretion. Based on this concept, colchicine has been advanced as an antifibrotic agent. A sizeable body of literature indicates that colchicine has antifibrotic properties in experimental animal models.216 This work has led to a number of human clinical trials.217–220 A wide variety of liver diseases has been studied, including primary biliary cirrhosis, alcoholic cirrhosis, cryptogenic cirrhosis, and miscellaneous other liver diseases. In a double-blind randomized controlled trial examining colchicine in primary biliary cirrhosis improvements were noted in a number of biochemical markers, but the drug failed to reduce fibrosis.217 In an often-cited popularized double-blind randomized controlled trial of colchicine versus placebo in patients with various liver diseases, colchicine led to improved fibrosis as well as a dramatic improvement in survival.218 However, this study has been extended to clinical practice with great caution because of a variety of methodological concerns. First,
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
Table 6-5. New Potential Antifibrotic Targets in Humans Agent
Comments
Anti-TGF-b Anti-PDGF Interferon-g PPAR ligands
Blocks stellate cell fibrogenesis Blocks stellate cell proliferation Inhibits multiple features of stellate cell activation ? Stellate cell-specific effects
many patients were lost to follow-up, and in addition there was substantial unexplained excess mortality in the control group (unrelated to liver disease). In a recent large VA cooperative multicenter study involving 549 patients comparing colchicine (0.6 mg orally b.i.d.) to placebo in patients with alcoholic liver disease, there was no apparent effect of active treatment on survival. Histologic data that might have provided information on the anti-inflammatory effects of colchicine were not obtained.219 A meta-analysis including 1138 subjects found that colchicine had no effect on hepatic fibrosis or mortality.220 In summary, the data surrounding colchicine suggest that this compound is safe but likely to be ineffective.
FUTURE ANTIFIBROTICS Given the major effort in understanding the biology of hepatic fibrogenesis, it is not surprising that numerous pathways have been targeted as having therapeutic potential. Many compounds have been studied in experimental models and have been shown to have antifibrotic properties, including several with great potential in human liver disease (Table 6-5). Several important pathways merit discussion. One of the most important examples is the TGF-b pathway, as it plays a central role in the fibrogenic cascade. Several approaches to inhibit the action of TGF-g have been proposed and include the use of molecules such as decorin, the protein core component of proteoglycan, which binds and inactivates TGF-b,221 antibodies directed against TGF-b1, and soluble receptors which typically encode for sequences that bind active TGF-b and prevent it from binding to its cognate receptors. The concept has been well established experimentally; indeed, the effect of inhibition of TGF-b in animal models of liver injury and fibrogenesis has been striking.222,223 A limitation of approaches that target TGF-b is that the cytokine potently inhibits cellular proliferation, and inhibition of its effects in vivo could predispose to malignant transformation. Another critical pathway involves PDGF. PDGF is the most potent stellate cell mitogen known,3,224 and in addition stimulates stellate cell migration.225 A number of approaches have been used to inhibit the effect of PDGF. For example, kinase inhibitors that specifically inhibit PDGF signaling might be useful,226 as could those with more general effects on tyrosine kinase receptors. Additionally, stellate cells express angiotensin and endothelin receptors and their cognate ligands appear to be overproduced in the liver; further, stimulation of stellate cells with their respective ligands leads to stellate cell activation.66 Thus, inhibition of their binding may be clinically beneficial. Among others agents, compounds such as pirfenidone,227 peroxisomal proliferator-activated receptor (PPAR)-g ligands,76,228 and halofuginone229 appear to have direct effects on stellate cells and
thus could evolve into effective antifibrotic compounds. Many others have been highlighted (see 5 for review).
SUMMARY AND FUTURE DIRECTIONS FOR ANTIFIBROTIC THERAPY The explosion of information about the pathogenesis of fibrogenesis has spawned a field dedicated to antifibrotics focused on the activation of hepatic stellate cells. Stellate cell activation is characterized by a number of important features, including enhanced matrix synthesis and a prominent contractile phenotype, processes that each contribute to the dysfunction of the liver typically found in advanced disease. It should be emphasized that the control of activation is multifactorial, and thus several potential therapeutic interventions are possible. A further critical concept is that fibrosis, in particular the ECM component of fibrosis, is dynamic, and the accumulation of fibrosis may be inhibited. It is likely that fibrosis, including even advanced fibrosis, may be reversible under the appropriate conditions. Currently, effective therapy for hepatic fibrogenesis exists for several diseases in which the cause of the underlying disease is removed. In contrast, specific therapy directed only at the fibrotic lesion is not currently available; the most effective therapies will most likely be directed at the stellate cell. Additionally, approaches that regulate matrix remodeling (i.e. by enhancing matrix degradation or inhibiting factors that prevent matrix breakdown) will be attractive. Thus, multiple potential targets have been identified, and it is highly likely that candidates will emerge. The ideal antifibrotic compound will be specific, effective, safe, and inexpensive.
REFERENCES 1. Kim WR, Brown RS Jr, Terrault NA, El-Serag H. Burden of liver disease in the United States: summary of a workshop. Hepatology 2002;36:227–242. 2. Gressner AM, Weiskirchen R, Breitkopf K, Dooley S. Roles of TGF-beta in hepatic fibrosis. Front Biosci 2002;7:D793–807. 3. Pinzani M. PDGF and signal transduction in hepatic stellate cells. Front Biosci 2002;7:d1720–1726. 4. Friedman SL. Liver fibrosis – from bench to bedside. J Hepatol 2003;38(Suppl 1):S38–53. 5. Rockey DC. Antifibrotic therapy in chronic liver disease. Clin Gastroenterol Hepatol 2005;3:95–107. 6. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;115:209–218. 7. Fattovich G, Giustina G, Degos F, et al. Morbidity and mortality in compensated cirrhosis type C: a retrospective follow-up study of 384 patients. Gastroenterology 1997;112:463–472. 8. El-Serag HB. Hepatocellular carcinoma and hepatitis C in the United States. Hepatology 2002;36(Suppl 1):S74–83. 9. Bruix J, Boix L, Sala M, Llovet JM. Focus on hepatocellular cancer. Cancer Cell 2004;5:215–219. 10. Schuppan D, Ruehl M, Somasundaram R, Hahn EG. Matrix as modulator of stellate cell and hepatic fibrogenesis. Semin Liver Dis 2001;21:351–372. 11. Jarnagin WR, Rockey DC, Koteliansky VE, et al. Expression of variant fibronectins in wound healing: cellular source and
103
Section I. Pathophysiology of the Liver
12.
13.
14.
15.
16.
17.
18.
19.
20. 21.
22.
23.
24.
25.
26.
27. 28.
29.
30.
31.
32.
104
biological activity of the EIIIA segment in rat hepatic fibrogenesis. J Cell Biol 1994;127:2037–2048. Bedossa P, Ferlicot S, Paradis V, et al. Dystroglycan expression in hepatic stellate cells: role in liver fibrosis. Lab Invest 2002;82:1053–1061. Schiano TD, Kim-Schluger L, Gondolesi G, Miller CM. Adult living donor liver transplantation: the hepatologist’s perspective. Hepatology 2001;33:3–9. Benhamou Y, Di Martino V, Bochet M, et al. Factors affecting liver fibrosis in human immunodeficiency virus- and hepatitis C virus-coinfected patients: impact of protease inhibitor therapy. Hepatology 2001;34:283–287. Farci P, Roskams T, Chessa L, et al. Long-term benefit of interferon alpha therapy of chronic hepatitis D: regression of advanced hepatic fibrosis. Gastroenterology 2004;126:1740–1749. Wiesner RH, McDiarmid SV, Kamath PS, et al. MELD and PELD: application of survival models to liver allocation. Liver Transpl 2001;7:567–580. Kamath PS, Wiesner RH, Malinchoc M, et al. A model to predict survival in patients with end-stage liver disease. Hepatology 2001;33:464–470. Poynard T, Ratziu V, Charlotte F, et al. Rates and risk factors of liver fibrosis progression in patients with chronic hepatitis C. J Hepatol 2001;34:730–739. Powell EE, Edwards-Smith CJ, Hay JL, et al. Host genetic factors influence disease progression in chronic hepatitis C. Hepatology 2000;31:828–833. Huang H, et al. Hepatology 2004;40:230A. Shi Z, Wakil AE, Rockey DC. Strain-specific differences in mouse hepatic wound healing are mediated by divergent T helper cytokine responses. Proc Natl Acad Sci USA 1997;94:10663–10668. Hillebrandt S, Goos C, Matern S, Lammert F. Genome-wide analysis of hepatic fibrosis in inbred mice identifies the susceptibility locus Hfib1 on chromosome 15. Gastroenterology 2002;123:2041–2051. Monto A, Patel K, Bostrom A, et al. Risks of a range of alcohol intake on hepatitis C-related fibrosis. Hepatology 2004;39: 826–834. Patton HM, Patel K, Behling C, et al. The impact of steatosis on disease progression and early and sustained treatment response in chronic hepatitis C patients. J Hepatol 2004;40:484–490. Poynard T, Bedossa P, Opolon P. Natural history of liver fibrosis progression in patients with chronic hepatitis C. The OBSVIRC, METAVIR, CLINIVIR, and DOSVIRC groups. Lancet 1997;349:825–832. de Torres M, Poynard T. Risk factors for liver fibrosis progression in patients with chronic hepatitis C. Ann Hepatol 2003;2:5–11. Wright TL. Treatment of patients with hepatitis C and cirrhosis. Hepatology 2002;36(Suppl 1):S185–194. Arenas JI, Vargas HE. Hepatitis C virus antiviral therapy in patients with cirrhosis. Gastroenterol Clin North Am 2004;33:549–562, ix. Lindh M, Horal P, Dhillon AP, Norkrans G. Hepatitis B virus DNA levels, precore mutations, genotypes and histological activity in chronic hepatitis B. J Viral Hepatol 2000;7:258–267. Merican I, Guan R, Amarapuka D, et al. Chronic hepatitis B virus infection in Asian countries. J Gastroenterol Hepatol 2000;15:1356–1361. Kobayashi M, Arase Y, Ikeda K, et al. Clinical characteristics of patients infected with hepatitis B virus genotypes A, B, and C. J Gastroenterol 2002;37:35–39. Chen CH, Chen PJ, Chu JS, et al. Fibrosing cholestatic hepatitis in a hepatitis B surface antigen carrier after renal transplantation. Gastroenterology 1994;107:1514–1518.
33. Lai CL, Chien RN, Leung NW, et al. A one-year trial of lamivudine for chronic hepatitis B. Asia Hepatitis Lamivudine Study Group [see comments]. N Engl J Med 1998;339:61–68. 34. Hadziyannis SJ, Tassopoulos NC, Heathcote EJ, et al. Adefovir dipivoxil for the treatment of hepatitis B e antigen-negative chronic hepatitis B. N Engl J Med 2003;348:800–807. 35. Liaw YF, Sung JJ, Chow WC, et al. Lamivudine for patients with chronic hepatitis B and advanced liver disease. N Engl J Med 2004;351:1521–1531. 36. Raynard B, Balian A, Fallik D, et al. Risk factors of fibrosis in alcohol-induced liver disease. Hepatology 2002;35:635–638. 37. Falck-Ytter Y, Younossi ZM, Marchesini G, McCullough AJ. Clinical features and natural history of nonalcoholic steatosis syndromes. Semin Liver Dis 2001;21:17–26. 38. Brunt EM. Nonalcoholic steatohepatitis: definition and pathology. Semin Liver Dis 2001;21:3–16. 39. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med 2002;346:1221–1231. 40. Ratziu V, Giral P, Charlotte F, et al. Liver fibrosis in overweight patients. Gastroenterology 2000;118:1117–1123. 41. Kweon YO, Goodman ZD, Dienstag JL, et al. Decreasing fibrogenesis: an immunohistochemical study of paired liver biopsies following lamivudine therapy for chronic hepatitis B. J Hepatol 2001;35:749–755. 42. Poynard T, McHutchison J, Manns M, et al. Impact of pegylated interferon alfa-2b and ribavirin on liver fibrosis in patients with chronic hepatitis C. Gastroenterology 2002;122:1303–1313. 43. Hammel P, Couvelard A, O’Toole D, et al. Regression of liver fibrosis after biliary drainage in patients with chronic pancreatitis and stenosis of the common bile duct. N Engl J Med 2001;344:418–423. 44. Dufour JF, DeLellis R, Kaplan MM. Reversibility of hepatic fibrosis in autoimmune hepatitis. Ann Intern Med 1997;127:981–985. 45. Iredale JP, Benyon RC, Pickering J, et al. Mechanisms of spontaneous resolution of rat liver fibrosis. Hepatic stellate cell apoptosis and reduced hepatic expression of metalloproteinase inhibitors. J Clin Invest 1998;102:538–549. 46. Shiratori Y, Imazeki F, Moriyama M, et al. Histologic improvement of fibrosis in patients with hepatitis C who have sustained response to interferon therapy. Ann Intern Med 2000;132:517–524. 47. Issa R, Zhou X, Constandinou CM, et al. Spontaneous recovery from micronodular cirrhosis: Evidence for incomplete resolution associated with matrix cross-linking. Gastroenterology 2004;126:1795–1808. 48. Issa R, Williams E, Trim N, et al. Apoptosis of hepatic stellate cells: involvement in resolution of biliary fibrosis and regulation by soluble growth factors. Gut 2001;48:548–557. 49. Iredale JP. Stellate cell behavior during resolution of liver injury. Semin Liver Dis 2001;21:427–436. 50. Arthur MJ. Fibrogenesis II. Metalloproteinases and their inhibitors in liver fibrosis. Am J Physiol Gastrointest Liver Physiol 2000;279:G245–249. 51. Schaefer B, Rivas-Estilla AM, Meraz-Cruz N, et al. Reciprocal modulation of matrix metalloproteinase-13 and type I collagen genes in rat hepatic stellate cells. Am J Pathol 2003;162:1771–1780. 52. Han YP, Zhou L, Wang J, et al. Essential role of matrix metalloproteinases in interleukin-1-induced myofibroblastic activation of hepatic stellate cell in collagen. J Biol Chem 2004;279:4820–4828. 53. Benyon D, Arthur MJP. Extracellular matrix degradation and the role of stellate cells. Semin Liver Dis 2001;21:373–384. 54. Friedman SL. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury. J Biol Chem 2000;275:2247–2250.
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
54a. Friedman SL, Arthur MJD. Reversing hepatic fibrosis. Science and Medicine 2002;8:194–205. 55. Tsukamoto H. Redox regulation of cytokine expression in Kupffer cells. Antioxid Redox Signal 2002;4:741–748. 56. Vogel W, Gish GD, Alves F, Pawson T. The discoidin domain receptor tyrosine kinases are activated by collagen. Mol Cell 1997;1:13–23. 57. Ikeda K, Wang LH, Torres R, et al. Discoidin domain receptor 2 interacts with Src and Shc following its activation by type I collagen. J Biol Chem 2002;277:19206–19212. 58. Olaso E, Ikeda K, Eng FJ, et al. DDR2 receptor promotes MMP-2-mediated proliferation and invasion by hepatic stellate cells. J Clin Invest 2001;108:1369–1378. 59. Pinzani M, Marra F. Cytokine receptors and signaling in hepatic stellate cells. Semin Liver Dis 2001;21:397–416. 60. Racine-Samson L, Rockey DC, Bissell DM. The role of alpha1beta1 integrin in wound contraction. A quantitative analysis of liver myofibroblasts in vivo and in primary culture. J Biol Chem 1997;272:30911–30917. 61. Gao R, Brigstock DR. Connective tissue growth factor (CCN2) induces adhesion of rat activated hepatic stellate cells by binding of its C-terminal domain to integrin alpha(v)beta(3) and heparan sulfate proteoglycan. J Biol Chem 2004;279:8848–8855. 62. Mazzocca A, Carloni V, Sciammetta S, et al. Expression of transmembrane 4 superfamily (TM4SF) proteins and their role in hepatic stellate cell motility and wound healing migration. J Hepatol 2002;37:322–330. 63. Geerts A. History, heterogeneity, developmental biology, and functions of quiescent hepatic stellate cells. Semin Liver Dis 2001;21:311–335. 64. Cassiman D, Libbrecht L, Desmet V, et al. Hepatic stellate cell/myofibroblast subpopulations in fibrotic human and rat livers. J Hepatol 2002;36:200–209. 65. Corpechot C, Barbu V, Wendum D, et al. Hypoxia-induced VEGF and collagen I expressions are associated with angiogenesis and fibrogenesis in experimental cirrhosis. Hepatology 2002;35:1010–1021. 66. Rockey DC. Vascular mediators in the injured liver. Hepatology 2003;37:4–12. 67. Forbes SJ, Russo FP, Rey V, et al. A significant proportion of myofibroblasts are of bone marrow origin in human liver fibrosis. Gastroenterology 2004;126:955–963. 68. Magness ST, Bataller R, Yang L, Brenner DA. A dual reporter gene transgenic mouse demonstrates heterogeneity in hepatic fibrogenic cell populations. Hepatology 2004;40: 1151–1159. 69. Kalluri R, Neilson EG. Epithelial–mesenchymal transition and its implications for fibrosis. J Clin Invest 2003;112: 1776–1784. 70. Tuma DJ. Role of malondialdehyde–acetaldehyde adducts in liver injury. Free Radic Biol Med 2002;32:303–308. 71. Bataller R, Schwabe RF, Choi YH, et al. NADPH oxidase signal transduces angiotensin II in hepatic stellate cells and is critical in hepatic fibrosis. J Clin Invest 2003;112:1383–1394. 72. Canbay A, Taimr P, Torok N, et al. Apoptotic body engulfment by a human stellate cell line is profibrogenic. Lab Invest 2003;83:655–663. 73. Canbay A, Higuchi H, Bronk SF, et al. Fas enhances fibrogenesis in the bile duct ligated mouse: a link between apoptosis and fibrosis. Gastroenterology 2002;123:1323–1330. 74. Olaso E, Salado C, Egilegor E, et al. Proangiogenic role of tumor-activated hepatic stellate cells in experimental melanoma metastasis. Hepatology 2003;37:674–685. 75. Mann DA, Smart DE. Transcriptional regulation of hepatic stellate cell activation. Gut 2002;50:891–896. 76. Marra F, Efsen E, Romanelli RG, et al. Ligands of peroxisome proliferator-activated receptor gamma modulate profibrogenic
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89. 90. 91. 92.
93.
94.
and proinflammatory actions in hepatic stellate cells. Gastroenterology 2000;119:466–478. Hellemans K, Verbuyst P, Quartier E, et al. Differential modulation of rat hepatic stellate phenotype by natural and synthetic retinoids. Hepatology 2004;39:97–108. Paik YH, Schwabe RF, Bataller R, et al. Toll-like receptor 4 mediates inflammatory signaling by bacterial lipopolysaccharide in human hepatic stellate cells. Hepatology 2003;37:1043–1055. Rippe RA, Schrum LW, Stefanovic B, et al. NF-kappaB inhibits expression of the alpha1(I) collagen gene. DNA Cell Biol 1999;18:751–761. Ratziu V, Lalazar A, Wong L, et al. Zf9, a Kruppel-like transcription factor up-regulated in vivo during early hepatic fibrosis. Proc Natl Acad Sci USA 1998;95:9500–9505. Kalinichenko VV, Bhattacharyya D, Zhou Y, et al. Foxf1 +/– mice exhibit defective stellate cell activation and abnormal liver regeneration following CCl4 injury. Hepatology 2003;37:107–117. Weiskirchen R, Moser M, Weiskirchen S, et al. LIM-domain protein cysteine- and glycine-rich protein 2 (CRP2) is a novel marker of hepatic stellate cells and binding partner of the protein inhibitor of activated STAT1. Biochem J 2001;359(Pt 3):485–496. Rockey DC, Fouassier L, Chung JJ, et al. Cellular localization of endothelin-1 and increased production in liver injury in the rat: potential for autocrine and paracrine effects on stellate cells. Hepatology 1998;27:472–480. Pinzani M, Milani S, De Franco R, et al. Endothelin 1 is overexpressed in human cirrhotic liver and exerts multiple effects on activated hepatic stellate cells. Gastroenterology 1996;110:534–548. Marra F, Grandaliano G, Valente AJ, Abboud HE. Thrombin stimulates proliferation of liver fat-storing cells and expression of monocyte chemotactic protein-1: potential role in liver injury. Hepatology 1995;22:780–787. Rosenbaum J, Blazejewski S, Preaux AM, et al. Fibroblast growth factor 2 and transforming growth factor beta 1 interactions in human liver myofibroblasts. Gastroenterology 1995;109:1986–1996. Skrtic S, Wallenius V, Ekberg S, et al. Insulin-like growth factors stimulate expression of hepatocyte growth factor but not transforming growth factor beta1 in cultured hepatic stellate cells. Endocrinology 1997;138:4683–4689. Pinzani M, Abboud HE, Aron DC. Secretion of insulin-like growth factor-I and binding proteins by rat liver fat-storing cells: regulatory role of platelet-derived growth factor. Endocrinology 1990;127:2343–2349. Pinzani M, Marra F, Carloni V. Signal transduction in hepatic stellate cells. Liver 1998;18:2–13. Friedman SL. Cytokines and fibrogenesis. Semin Liver Dis 1999;19:129–140. Reinehr RM, Kubitz R, Peters-Regehr T, et al. Activation of rat hepatic stellate cells in culture is associated with increased sensitivity to endothelin 1. Hepatology 1998;28:1566–1577. Schnabl B, Kweon YO, Frederick JP, et al. The role of Smad3 in mediating mouse hepatic stellate cell activation. Hepatology 2001;34:89–100. Dooley S, Delvoux B, Lahme B, et al. Modulation of transforming growth factor beta response and signaling during transdifferentiation of rat hepatic stellate cells to myofibroblasts. Hepatology 2000;31:1094–1106. Liu C, Gaca MD, Swenson ES, et al. Smads 2 and 3 are differentially activated by transforming growth factor-beta (TGF-beta) in quiescent and activated hepatic stellate cells. Constitutive nuclear localization of Smads in activated cells is TGF-beta-independent. J Biol Chem 2003;278: 11721–11728.
105
Section I. Pathophysiology of the Liver
95. Wells RG. Fibrogenesis. V. TGF-beta signaling pathways. Am J Physiol Gastrointest Liver Physiol 2000;279:G845–850. 96. Wells RG, Kruglov E, Dranoff JA. Autocrine release of TGFbeta by portal fibroblasts regulates cell growth. FEBS Lett 2004;559:107–110. 97. Marra F. Chemokines in liver inflammation and fibrosis. Front Biosci 2002;7:d1899–914. 98. Ohata M, Yamauchi M, Takeda K, et al. RAR and RXR expression by Kupffer cells. Exp Mol Pathol 2000; 68:13–20. 99. LeCouter J, Moritz DR, Li B, et al. Angiogenesis-independent endothelial protection of liver: role of VEGFR-1. Science 2003;299:890–893. 100. Yoshiji H, Kuriyama S, Yoshii J, et al. Vascular endothelial growth factor and receptor interaction is a prerequisite for murine hepatic fibrogenesis. Gut 2003;52:1347–1354. 101. Ankoma-Sey V, Wang Y, Dai Z. Hypoxic stimulation of vascular endothelial growth factor expression in activated rat hepatic stellate cells. Hepatology 2000;31:141–148. 102. Schwabe RF, Bataller R, Brenner DA. Human hepatic stellate cells express CCR5 and RANTES to induce proliferation and migration. Am J Physiol Gastrointest Liver Physiol 2003;285:G949–958. 103. Gaca MD, Zhou X, Benyon RC. Regulation of hepatic stellate cell proliferation and collagen synthesis by proteinase-activated receptors. J Hepatol 2002;36:362–369. 104. Bataller R, Paik YH, Lindquist JN, et al. Hepatitis C virus core and nonstructural proteins induce fibrogenic effects in hepatic stellate cells. Gastroenterology 2004;126:529–540. 105. Efsen E, Grappone C, DeFranco RM, et al. Up-regulated expression of fractalkine and its receptor CX3CR1 during liver injury in humans. J Hepatol 2002;37:39–47. 106. Kobayashi S, Seki S, Kawada N, et al. Apoptosis of T cells in the hepatic fibrotic tissue of the rat: a possible inducing role of hepatic myofibroblast-like cells. Cell Tissue Res 2003;311:353–364. 107. Safadi R, Ohta M, Alvarez CE, et al. Immune stimulation of hepatic fibrogenesis by CD8 cells and attenuation by transgenic interleukin-10 from hepatocytes. Gastroenterology 2004;127: 870–882. 108. Gaca MD, Pickering JA, Arthur MJ, Benyon RC. Human and rat hepatic stellate cells produce stem cell factor: a possible mechanism for mast cell recruitment in liver fibrosis. J Hepatol 1999;30:850–858. 109. Nelson DR, Tu Z, Soldevila-Pico C, et al. Long-term interleukin 10 therapy in chronic hepatitis C patients has a proviral and anti-inflammatory effect. Hepatology 2003;38:859–868. 110. Mazzocca A, Sciammetta SC, Carloni V, et al. Binding of hepatitis C virus envelope protein E2 to CD81 up-regulates matrix metalloproteinase-2 in human hepatic stellate cells. J Biol Chem 2005;280:11329–11339. 111. Bonacchi A, Petrai I, Defranco RM, et al. The chemokine CCL21 modulates lymphocyte recruitment and fibrosis in chronic hepatitis C. Gastroenterology 2003;125:1060–1076. 112. Pohlmann S, Zhang J, Baribaud F, et al. Hepatitis C virus glycoproteins interact with DC-SIGN and DC-SIGNR. J Virol 2003;77:4070–4080. 113. Ikejima K, Takei Y, Honda H, et al. Leptin receptor-mediated signaling regulates hepatic fibrogenesis and remodeling of extracellular matrix in the rat. Gastroenterology 2002;122:1399–1410. 114. Leclercq IA, Farrell GC, Schriemer R, Robertson GR. Leptin is essential for the hepatic fibrogenic response to chronic liver injury. J Hepatol 2002;37:206–213. 115. Saxena NK, Saliba G, Floyd JJ, Anania FA. Leptin induces increased alpha2(I) collagen gene expression in cultured rat hepatic stellate cells. J Cell Biochem 2003;89:311–320.
106
116. Kamada Y, Tamura S, Kiso S, et al. Enhanced carbon tetrachloride-induced liver fibrosis in mice lacking adiponectin. Gastroenterology 2003;125:1796–1807. 117. Canbay A, Friedman S, Gores GJ. Apoptosis: the nexus of liver injury and fibrosis. Hepatology 2004;39:273–278. 118. Taimr P, Higuchi H, Kocova E, et al. Activated stellate cells express the TRAIL receptor-2/death receptor-5 and undergo TRAIL-mediated apoptosis. Hepatology 2003;37:87–95. 119. Trim N, Morgan S, Evans M, et al. Hepatic stellate cells express the low affinity nerve growth factor receptor p75 and undergo apoptosis in response to nerve growth factor stimulation. Am J Pathol 2000;156:1235–1243. 120. Lang A, Schoonhoven R, Tuvia S, et al. Nuclear factor kappaB in proliferation, activation, and apoptosis in rat hepatic stellate cells. J Hepatol 2000;33:49–58. 121. Reif S, Lang A, Lindquist JN, et al. The role of focal adhesion kinase-phosphatidylinositol 3-kinase-akt signaling in hepatic stellate cell proliferation and type I collagen expression. J Biol Chem 2003;278:8083–8090. 122. Preaux AM, D’Ortho MP, Bralet MP, et al. Apoptosis of human hepatic myofibroblasts promotes activation of matrix metalloproteinase-2. Hepatology 2002;36:615–622. 123. Li L, Tao J, Davaille J, et al. 15-deoxy-delta 12,14prostaglandin J2 induces apoptosis of human hepatic myofibroblasts. A pathway involving oxidative stress independently of peroxisome-proliferator-activated receptors. J Biol Chem 2001;276:38152–38158. 124. Yoshiji H, Kuriyama S, Yoshii J, et al. Tissue inhibitor of metalloproteinases-1 attenuates spontaneous liver fibrosis resolution in the transgenic mouse. Hepatology 2002;36:850–860. 125. Wright MC, Issa R, Smart DE, et al. Gliotoxin stimulates the apoptosis of human and rat hepatic stellate cells and enhances the resolution of liver fibrosis in rats. Gastroenterology 2001;121:685–698. 126. Sohara N, Znoyko I, Levy MT, et al. Reversal of activation of human myofibroblast-like cells by culture on a basement membrane-like substrate. J Hepatol 2002;37:214–221. 127. Lin SM, Sheen IS, Chien RN, et al. Long-term beneficial effect of interferon therapy in patients with chronic hepatitis B virus infection. Hepatology 1999;29:971–975. 128. Wanless IR, Nakashima E, Sherman M. Regression of human cirrhosis. Morphologic features and the genesis of incomplete septal cirrhosis. Arch Pathol Lab Med 2000;124:1599–1607. 129. Marcellin P, Chang TT, Lim SG, et al. Adefovir dipivoxil for the treatment of hepatitis B e antigen-positive chronic hepatitis B. N Engl J Med 2003;348:808–816. 130. Chalasani N, Imperiale TF, Ismail A, et al. Predictors of large esophageal varices in patients with cirrhosis. Am J Gastroenterol 1999;94:3285–3291. 131. Zaman A, Becker T, Lapidus J, Benner K. Risk factors for the presence of varices in cirrhotic patients without a history of variceal hemorrhage. Arch Intern Med 2001;161: 2564–2570. 132. Giannini E, Risso D, Botta F, et al. Validity and clinical utility of the aspartate aminotransferase–alanine aminotransferase ratio in assessing disease severity and prognosis in patients with hepatitis C virus-related chronic liver disease. Arch Intern Med 2003;163:218–224. 133. Sheth SG, Flamm SL, Gordon FD, Chopra S. AST/ALT ratio predicts cirrhosis in patients with chronic hepatitis C virus infection. Am J Gastroenterol 1998;93:44–48. 134. Giannini E, Testa R. Noninvasive diagnosis of fibrosis: the truth is rarely pure and never simple. Hepatology 2003;38:1312–1313; author reply 3. 135. Forns X, Ampurdanes S, Llovet JM, et al. Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model. Hepatology 2002;36:986–992.
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
136. Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003;38: 518–526. 137. Poynard T, Aubert A, Bedossa P, et al. A simple biological index for detection of alcoholic liver disease in drinkers. Gastroenterology 1991;100:1397–1402. 138. Naveau S, Poynard T, Benattar C, Bedossa P, Chaput JC. Alpha-2-macroglobulin and hepatic fibrosis. Diagnostic interest. Dig Dis Sci 1994;39:2426–2432. 139. Imbert-Bismut F, Ratziu V, Pieroni L, et al. Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study. Lancet 2001;357:1069–1075. 140. Poynard T, Imbert-Bismut F, Munteanu M, et al. Overview of the diagnostic value of biochemical markers of liver fibrosis (FibroTest, HCV FibroSure) and necrosis (ActiTest) in patients with chronic hepatitis C. Comp Hepatol 2004;3:8. 141. Myers RP, Tainturier MH, Ratziu V, et al. Prediction of liver histological lesions with biochemical markers in patients with chronic hepatitis B. J Hepatol 2003;39:222–230. 142. Naveau S, Raynard B, Ratziu V, et al. Biomarkers for the prediction of liver fibrosis in patients with chronic alcoholic liver disease. Clin Gastroenterol Hepatol 2005;3:167–174. 143. Colombo M, Annoni G, Donato MF, et al. Serum type III procollagen peptide in alcoholic liver disease and idiopathic hemochromatosis: its relationship to hepatic fibrosis, activity of the disease and iron overload. Hepatology 1985;5:475–479. 144. Tsutsumi M, Takase S, Urashima S, et al. Serum markers for hepatic fibrosis in alcoholic liver disease: which is the best marker, type III procollagen, type IV collagen, laminin, tissue inhibitor of metalloproteinase, or prolyl hydroxylase? Alcohol Clin Exp Res 1996;20:1512–1517. 145. Afdhal NH, Nunes D. Evaluation of liver fibrosis: a concise review. Am J Gastroenterol 2004;99:1160–1174. 146. Patel K, Gordon SC, Jacobson I, et al. Evaluation of a panel of non-invasive serum markers to differentiate mild from moderate-to-advanced liver fibrosis in chronic hepatitis C patients. J Hepatol 2004;41:935–942. 147. Rosenberg WM, Voelker M, Thiel R, et al. Serum markers detect the presence of liver fibrosis: a cohort study. Gastroenterology 2004;127:1704–1713. 147a. Sud A, Hui JM, Farrell GC, et al. Improved prediction of fibrosis in chronic hepatitis C using measures of insulin resistance in a probability index. Hepatology 2004;39:1239–1247. 148. Poon TC, Hui AY, Chan HL, et al. Prediction of liver fibrosis and cirrhosis in chronic hepatitis B infection by serum proteomic fingerprinting: a pilot study. Clin Chem 2005;51:328–335. 149. Bora R, Ratziu V, Bedossa P, et al. Diagnostic value of biochemical markers, fibrotest–actitest–HCV–fibrosure compared with serum protein profiling by seldi-TOF, for the diagnosis of bridging fibrosis in patients with chronic hepatitis C. Hepatology 2004;40:517A. 150. Dev A, White I, Symonds W, et al. Serum proteomic analysis in hepatitis C patients with and without fibrosis. J Hepatol 2005;42(suppl 2):117A. 151. Callewaert N, Van Vlierberghe H, Van Hecke A, et al. Noninvasive diagnosis of liver cirrhosis using DNA sequencerbased total serum protein glycomics. Nature Med 2004;10:429–434. 152. Wells RG. The role of matrix stiffness in hepatic stellate cell activation and liver fibrosis. J Clin Gastroenterol 2005;39:S158–161. 153. Ziol M, Handra-Luca A, Kettaneh A, et al. Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with chronic hepatitis C. Hepatology 2005;41:48– 54.
154. Castera L, Vergniol J, Foucher J, et al. Prospective comparison of transient elastography, Fibrotest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology 2005;128:343–350. 155. Beljaars L, Meijer DK, Poelstra K. Targeting hepatic stellate cells for cell-specific treatment of liver fibrosis. Front Biosci 2002;7:e214–222. 156. Adler M, Verset D, Bouhdid H, et al. Prognostic evaluation of patients with parenchymal cirrhosis. Proposal of a new simple score. J Hepatol 1997;26:642–649. 157. Giannini EG, Testa R. 13C-breath tests and liver fibrosis. Eur Rev Med Pharmacol Sci 2004;8:51–54. 158. Giannini EG, Fasoli A, Borro P, et al. 13C-galactose breath test and 13C-aminopyrine breath test for the study of liver function in chronic liver disease. Clin Gastroenterol Hepatol 2005;3: 279–285. 159. Testa R, Caglieris S, Risso D, et al. Monoethylglycinexylidide formation measurement as a hepatic function test to assess severity of chronic liver disease. Am J Gastroenterol 1997;92:2268–2273. 160. Regev A, Berho M, Jeffers LJ, et al. Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002;97:2614–2618. 161. Ratziu V, Charlotte F, Heurtier A, et al. Sampling variability of liver biopsy in nonalcoholic fatty liver disease. Gastroenterology 2005;128:1898–1906. 162. Colloredo G, Guido M, Sonzogni A, Leandro G. Impact of liver biopsy size on histological evaluation of chronic viral hepatitis: the smaller the sample, the milder the disease. J Hepatol 2003;39:239–244. 163. Bedossa P, Dargere D, Paradis V. Sampling variability of liver fibrosis in chronic hepatitis C. Hepatology 2003;38: 1449–1457. 164. Rousselet MC, Michalak S, Dupre F, et al. Sources of variability in histological scoring of chronic viral hepatitis. Hepatology 2005;41:257–264. 164a. Wanless IR, Nakashima E, Sherman M. Regression of human cirrhosis. Morphologic features and the genesis of incomplete septal cirrhosis. Arch Pathol Lab Med 2000;124:1599–1607. 165. Powell LW, Kerr JF. Reversal of ‘cirrhosis’ in idiopathic haemochromatosis following long-term intensive venesection therapy. Australas Ann Med 1970;19:54–57. 166. Blumberg RS, Chopra S, Ibrahim R, et al. Primary hepatocellular carcinoma in idiopathic hemochromatosis after reversal of cirrhosis. Gastroenterology 1988;95: 1399–1402. 167. Neuschwander-Tetri BA, Brunt EM, Wehmeier KR, et al. Improved nonalcoholic steatohepatitis after 48 weeks of treatment with the PPAR-gamma ligand rosiglitazone. Hepatology 2003;38:1008–1017. 168. Ramond MJ, Poynard T, Rueff B, et al. A randomized trial of prednisolone in patients with severe alcoholic hepatitis. N Engl J Med 1992;326:507–512. 169. Spahr L, Rubbia-Brandt L, Pugin J, et al. Rapid changes in alcoholic hepatitis histology under steroids: correlation with soluble intercellular adhesion molecule-1 in hepatic venous blood. J Hepatol 2001;35:582–589. 170. Thompson K, Maltby J, Fallowfield J, et al. Interleukin-10 expression and function in experimental murine liver inflammation and fibrosis [see comments]. Hepatology 1998;28:1597–1606. 171. Akriviadis E, Botla R, Briggs W, et al. Pentoxifylline improves short-term survival in severe acute alcoholic hepatitis: a doubleblind, placebo-controlled trial. Gastroenterology 2000;119:1637–1648. 172. Spahr L, Rubbia-Brandt L, Frossard JL, et al. Combination of steroids with infliximab or placebo in severe alcoholic hepatitis: a randomized controlled pilot study. J Hepatol 2002;37:448–455.
107
Section I. Pathophysiology of the Liver
173. Tilg H, Jalan R, Kaser A, et al. Anti-tumor necrosis factor-alpha monoclonal antibody therapy in severe alcoholic hepatitis. J Hepatol 2003;38:419–425. 174. Menon KV, Stadheim L, Kamath PS, et al. A pilot study of the safety and tolerability of etanercept in patients with alcoholic hepatitis. Am J Gastroenterol 2004;99:255–260. 175. Schaff Z, Lapis K, Szende B, et al. The effect of Dpenicillamine on CCl4-induced experimental liver cirrhosis. Exp Pathol 1991;43:111–120. 176. Bodenheimer HC Jr, Schaffner F, Sternlieb I, et al. A prospective clinical trial of D-penicillamine in the treatment of primary biliary cirrhosis. Hepatology 1985;5:1139–1142. 177. Dickson ER, Fleming TR, Wiesner RH, et al. Trial of penicillamine in advanced primary biliary cirrhosis. N Engl J Med 1985;312:1011–1015. 178. Aithal GP, Haugk B, Das S, et al. Monitoring methotrexateinduced hepatic fibrosis in patients with psoriasis: are serial liver biopsies justified? Aliment Pharmacol Ther 2004;19:391–399. 179. Te HS, Schiano TD, Kuan SF, et al. Hepatic effects of longterm methotrexate use in the treatment of inflammatory bowel disease. Am J Gastroenterol 2000;95:3150–3156. 180. Kaplan MM, DeLellis RA, Wolfe HJ. Sustained biochemical and histologic remission of primary biliary cirrhosis in response to medical treatment. Ann Intern Med 1997;126: 682–688. 181. Hendrickse MT, Rigney E, Giaffer MH, et al. Low-dose methotrexate is ineffective in primary biliary cirrhosis: longterm results of a placebo-controlled trial [see comments]. Gastroenterology 1999;117:400–407. 182. Bach N, Bodian C, Bodenheimer H, et al. Methotrexate therapy for primary biliary cirrhosis. Am J Gastroenterol 2003;98:187–193. 183. Kaplan MM, Cheng S, Price LL, Bonis PA. A randomized controlled trial of colchicine plus ursodiol versus methotrexate plus ursodiol in primary biliary cirrhosis: ten-year results. Hepatology 2004;39:915–923. 184. Aleynik SI, Leo MA, Ma X, et al. Polyenylphosphatidylcholine prevents carbon tetrachloride-induced lipid peroxidation while it attenuates liver fibrosis. J Hepatol 1997;27:554–561. 185. Lieber CS, Weiss DG, Groszmann R, et al. II. Veterans Affairs Cooperative Study of Polyenylphosphatidylcholine in Alcoholic Liver Disease. Alcohol Clin Exp Res 2003;27:1765–1772. 186. Boigk G, Stroedter L, Herbst H, et al. Silymarin retards collagen accumulation in early and advanced biliary fibrosis secondary to complete bile duct obliteration in rats. Hepatology 1997;26:643–649. 187. Jia JD, Bauer M, Cho JJ, et al. Antifibrotic effect of silymarin in rat secondary biliary fibrosis is mediated by downregulation of procollagen alpha1(I) and TIMP-1. J Hepatol 2001;35:392–398. 188. Lieber CS, Leo MA, Cao Q, et al. Silymarin retards the progression of alcohol-induced hepatic fibrosis in baboons. J Clin Gastroenterol 2003;37:336–339. 189. Ferenci P, Dragosics B, Dittrich H, et al. Randomized controlled trial of silymarin treatment in patients with cirrhosis of the liver. J Hepatol 1989;9:105–113. 190. Pares A, Planas R, Torres M, et al. Effects of silymarin in alcoholic patients with cirrhosis of the liver: results of a controlled, double-blind, randomized and multicenter trial [see comments]. J Hepatol 1998;28:615–621. 191. Brown KE, Poulos JE, Li L, et al. Effect of vitamin E supplementation on hepatic fibrogenesis in chronic dietary iron overload. Am J Physiol 1997;272:G116–123. 192. Houglum K, Venkataramani A, Lyche K, Chojkier M. A pilot study of the effects of d-alpha-tocopherol on hepatic stellate cell activation in chronic hepatitis C. Gastroenterology 1997;113:1069–1073.
108
193. Hasegawa T, Yoneda M, Nakamura K, et al. Plasma transforming growth factor-beta1 level and efficacy of alphatocopherol in patients with non-alcoholic steatohepatitis: a pilot study. Aliment Pharmacol Ther 2001;15:1667–1672. 194. Mezey E, Potter J, Rennie-Tankersley L, et al. A randomized placebo controlled trial of vitamin E in alcoholic hepatitis. Hepatology 2003;38:264A. 195. Stewart S, Prince M, Bassendine M, et al. A trial of antioxidant therapy alone or with corticosteroids in acute alcoholic hepatitis. J Hepatol 2002;36(S):16. 196. Anon. The results of a randomized double blind controlled trial evaluating malotilate in primary biliary cirrhosis. A European multicentre study group. J Hepatol 1993;17:227–235. 197. Lu SC, Tsukamoto H, Mato JM. Role of abnormal methionine metabolism in alcoholic liver injury. Alcohol 2002;27:155– 162. 198. Mato JM, Camara J, Fernandez de Paz J, et al. Sadenosylmethionine in alcoholic liver cirrhosis: a randomized, placebo-controlled, double-blind, multicenter clinical trial. J Hepatol 1999;30:1081–1089. 199. Rambaldi A, Gluud C. Meta-analysis of propylthiouracil for alcoholic liver disease – a Cochrane Hepato-Biliary Group Review. Liver 2001;21:398–404. 200. Nava-Ocampo AA, Suster S, Muriel P. Effect of colchiceine and ursodeoxycholic acid on hepatocyte and erythrocyte membranes and liver histology in experimentally induced carbon tetrachloride cirrhosis in rats. Eur J Clin Invest 1997;27:77–84. 201. Stiehl A. Ursodeoxycholic acid in the treatment of primary sclerosing cholangitis. Ann Med 1994;26:345–349. 202. Combes B, Carithers RL Jr, Maddrey WC, et al. A randomized, double-blind, placebo-controlled trial of ursodeoxycholic acid in primary biliary cirrhosis. Hepatology 1995;22:759–766. 203. Jacquemin E, Hermans D, Myara A, et al. Ursodeoxycholic acid therapy in pediatric patients with progressive familial intrahepatic cholestasis. Hepatology 1997;25:519–523. 204. Lindblad A, Glaumann H, Strandvik B. A two-year prospective study of the effect of ursodeoxycholic acid on urinary bile acid excretion and liver morphology in cystic fibrosis-associated liver disease. Hepatology 1998;27:166–174. 205. Goulis J, Leandro G, Burroughs AK. Randomised controlled trials of ursodeoxycholic-acid therapy for primary biliary cirrhosis: a meta-analysis. Lancet 1999;354:1053–1060. 206. Poupon RE, Bonnand AM, Chretien Y, Poupon R. Ten-year survival in ursodeoxycholic acid-treated patients with primary biliary cirrhosis. The UDCA-PBC Study Group. Hepatology 1999;29:1668–1671. 207. Degott C, Zafrani ES, Callard P, et al. Histopathological study of primary biliary cirrhosis and the effect of ursodeoxycholic acid treatment on histology progression. Hepatology 1999;29:1007–1012. 208. Poupon RE, Lindor KD, Pares A, et al. Combined analysis of the effect of treatment with ursodeoxycholic acid on histologic progression in primary biliary cirrhosis. J Hepatol 2003;39:12–16. 209. Lindor KD, Kowdley KV, Heathcote EJ, et al. Ursodeoxycholic acid for treatment of nonalcoholic steatohepatitis: results of a randomized trial. Hepatology 2004;39:770–778. 210. Rambaldi A, Iaquinto G, Gluud C. Anabolic–androgenic steroids for alcoholic liver disease: a Cochrane review. Am J Gastroenterol 2002;97:1674–1681. 211. Rockey DC, Maher JJ, Jarnagin WR, et al. Inhibition of rat hepatic lipocyte activation in culture by interferon-gamma. Hepatology 1992;16:776–784. 212. Rockey DC, Chung JJ. Interferon gamma inhibits lipocyte activation and extracellular matrix mRNA expression during experimental liver injury: implications for treatment of hepatic fibrosis. J Invest Med 1994;42:660–670.
Chapter 6 HEPATIC FIBROSIS AND CIRRHOSIS
213. Ziesche R, Hofbauer E, Wittman K, et al. A preliminary study of long-term treatment with interferon gamma-1b and low-dose prednisolone in patients with idiopathic pulmonary fibrosis. N Engl J Med 1999;341:1264–1269. 214. Toyonaga T, Hino O, Sugai S, et al. Chronic active hepatitis in transgenic mice expressing interferon-gamma in the liver. Proc Natl Acad Sci USA 1994;91:614–618. 215. Muir AJ, Sylvestre PB, Rockey DC. Interferon gamma-1b for the treatment of chronic hepatitis C infection. J Viral Hepatitis 2006; in press. 216. Rodriguez L, Cerbon-Ambriz J, Munoz ML. Effects of colchicine and colchiceine in a biochemical model of liver injury and fibrosis. Arch Med Res 1998;29:109–116. 217. Kaplan MM, Alling DW, Zimmerman HJ, et al. A prospective trial of colchicine for primary biliary cirrhosis. N Engl J Med 1986;315:1448–1454. 218. Kershenobich D, Vargas F, Garcia-Tsao G, et al. Colchicine in the treatment of cirrhosis of the liver. N Engl J Med 1988;318:1709–1713. 219. Morgan TR, Nemchausky B, Schiff ER, et al. Colchicine does not prolong life in patients with advanced alcoholic cirrhosis: results of a prospective, randomized, placebo-controlled trial. Gastroenterology 2002;122:641A. 220. Rambaldi A, Gluud C. Colchicine for alcoholic and nonalcoholic liver fibrosis or cirrhosis. Liver 2001;21:129–136. 221. Isaka Y, Brees DK, Ikegaya K, et al. Gene therapy by skeletal muscle expression of decorin prevents fibrotic disease in rat kidney. Nature Med 1996;2:418–423. 222. George J, Roulot D, Koteliansky VE, Bissell DM. In vivo inhibition of rat stellate cell activation by soluble TGF beta
223.
224.
225.
226.
227.
228.
229.
type II receptor: a potential new therapy for hepatic fibrosis. Proc Natl Acad Sci USA 1999;96:12719–12724. Yata Y, Gotwals P, Koteliansky V, Rockey DC. Dose-dependent inhibition of hepatic fibrosis in mice by a TGF-beta soluble receptor: implications for antifibrotic therapy. Hepatology 2002;35:1022–1030. Pinzani M, Milani S, Herbst H, et al. Expression of plateletderived growth factor and its receptors in normal human liver and during active hepatic fibrogenesis. Am J Pathol 1996;148:785–800. Ikeda K, Wakahara T, Wang YQ, et al. In vitro migratory potential of rat quiescent hepatic stellate cells and its augmentation by cell activation. Hepatology 1999;29:1760–1767. Kinnman N, Francoz C, Barbu V, et al. The myofibroblastic conversion of peribiliary fibrogenic cells distinct from hepatic stellate cells is stimulated by platelet-derived growth factor during liver fibrogenesis. Lab Invest 2003;83:163–173. Di Sario A, Bendia E, Svegliati Baroni G, et al. Effect of pirfenidone on rat hepatic stellate cell proliferation and collagen production. J Hepatol 2002;37:584–591. Miyahara T, Schrum L, Rippe R, et al. Peroxisome proliferatoractivated receptors and hepatic stellate cell activation. J Biol Chem 2000;275:35715–35722. Bruck R, Genina O, Aeed H, et al. Halofuginone to prevent and treat thioacetamide-induced liver fibrosis in rats. Hepatology 2001;33:379–386.
109
Section I: Pathophysiology of the Liver
7
REPLICATION OF HEPATITIS B VIRUS AND PATHOGENESIS OF DISEASE Angeline Bartholomeusz, Judy Chang, Stephen Locarnini, and Sharon Lewin Abbreviations ALT alanine aminotransaminase APC antigen-presenting cells BCP basal core promoter ccc covalently closed circular DC dendritic cells DHBV Duck hepatitis B virus EnhI enhancer I EnhII enhancer II ER endoplasmic reticulum FasL Fas ligand HBcAg hepatitis B core protein or antigen HBeAg hepatitis B e antigen HBsAg hepatitis B surface antigen HBSP hepatitis B splice protein HBx hepatitis Bx protein HBV hepatitis B virus HCC hepatocellular carcinoma
HLA HSC GAPD IFN LCMV LHBs lxRNA MHBs MHR MIP-1a NK NKT NLS nt PAMP pg pHSA
human leukocyte antigen hepatic stellate cells glyceraldehyde-3-phosphate dehydrogenase
interferon lymphocyte choriomeningitis virus large hepatitis B surface protein long-X RNA medium hepatitis B surface protein major hydrophilic region macrophage inflammatory protein 1a natural killer natural killer T nuclear localization signal nucleotide pathogen-associated molecular patterns pregenomic polymerized human serum albumin
INTRODUCTION Hepatitis B virus (HBV) is the prototype member of the family Hepadnaviridae, which also includes viruses that can infect higher primates such as chimpanzees and lower primates such as tupaia (tree shrews).1 Other hepadnaviruses can infect a range of other mammals, including woodchucks and ground-squirrels, and birds such as herons and ducks. HBV is an enveloped virus and contains a circular, partially double-stranded DNA genome. There are three types of virus-associated particle: the virion, and spherical and filamentous particles found in serum. Only the HBV virion is infectious, as the spherical particles and filaments do not contain the HBV genome and represent excess production of the viral envelope. The HBV virion is 42 nm in diameter and comprises an outer envelope that contains the three envelope proteins, all of which express the hepatitis B surface antigen (HBsAg). This envelope surrounds an inner nucleocapsid made up of the hepatitis B core protein or antigen (HBcAg) that packages the viral genome and associated polymerase. The spherical particles are 17–25 nm in diameter and can occur in large numbers up to 1013/ml. The filaments or tubular structures are approximately 20–22 nm in diameter. In spite of its small genome size, HBV is a complex virus. The HBV genome is only 3.2 kb in length, and so to compensate for this
PKR POL RANTES RNase H rt SHBs Th1 Th2 TLR TLR-2 TNF TNF-R1 TRAIL Tyr WMHBV
protein kinase activity polymerase regulated on activation, normal T expressed and secreted ribonuclease H reverse transcriptase small hepatitis B surface protein T-helper cell type 1 T-helper cell type 2 toll-like receptors Toll-like receptor 2 tumor necrosis factor TNF receptor 1 TNF-related apoptosis-inducing ligand tyrosine Woolly Monkey hepatitis B virus
limited coding potential HBV DNA is organized into a series of overlapping reading frames and co-terminal reading frames that encode for proteins which are either multifunctional or have very different functions despite sharing similar amino acid residues (Figure 7-1). Furthermore, HBV has also evolved unique strategies for genomic replication (Figure 7-2). It is the intricacies of this replication strategy and its impact on the hepatocyte and the host’s immune response that this chapter will unravel, as well as new systems that have been developed to facilitate research into HBV replication and pathogenesis.
MOLECULAR VIROLOGY NEW CELL LINES AND MODEL SYSTEMS THAT HAVE BEEN DEVELOPED TO INVESTIGATE HBV REPLICATION AND PATHOGENESIS A number of new cell lines and cloning strategies have been developed to deliver HBV into cells and to investigate HBV replication, virus–host interactions, pathogenesis and antiviral sensitivities. The major problem of studying HBV replication is the lack of a suitable cell-culture system for infecting cells in vitro. One of the recent advances has been the development of a cell line, HepaRG, by
111
Section I. Pathophysiology of the Liver
Pre-S2/S promoter
2.1 4 2.
kb
A RN
kb R
NA
Pre-S 2
S1 ePr
–S
d
S F-
n tra
OR
Pre-S1/S promoter
Figure 7-1. Diagrammatic representation of the HBV genome. The inner circle represents the virion genome DNA that is packaged within the viral particles in the cytoplasm of infected cells, and the dashes represent the region of the positive-sense DNA which is incompletely synthesized. The middle circle represents the four open reading frames corresponding to precore, core, HBx, polymerase, and the envelope proteins. The outer circle represents the HBV RNAs. The promoter and enhancer regions are indicated.
+ nd ra
DR
1
DR2 5'
5' AA A AA A AA A AA A
Basal core promoter
F
F-C OR
3.5 kb RNA
St
OR LPO
O RF -X Enhancer 1/X promoter
NA bR 0.7 k Enhancer 11
Gripon et al.2 that can be infected with HBV. The wider availability of this cell line should facilitate research into the early events of HBV replication, including virus entry and the hunt for the elusive cellular receptor for HBV. A number of delivery systems have been developed to transport HBV into cells. This includes the recombinant adenovirus3 and the recombinant baculovirus system.4 To study HBV replication and antiviral sensitivities from patient samples a number of groups have developed strategies for the PCR amplification of full genome length (or near full genome length) HBV from the patient and directly cloning the amplified product into vectors.5 These new patientspecific clones will enable mutants to be studied in the context of the entire authentic genetic framework, and thus is extremely useful for determining phenotypically the antiviral resistance profile for a patient at a given time. Other groups have developed particular plasmid vectors that facilitate the cloning and expression of fulllength HBV genomes amplified from the sera of patients6 that can then be used to enable efficient phenotypic analysis of patientderived virus in cell culture.7 These vectors will be useful in future studies of drug resistance, including surveillance and cross-resistance testing.8
HEPATITIS B VIRAL GENOME: MAJOR TRANSCRIPTS Transcription of the HBV genome results in the formation of the pregenomic (pg) RNA and the precore mRNA, which are approxi-
112
mately 3.5 kb and three subgenomic RNAs.9 The polymerase gene is the largest open reading frame and encodes for the multifunctional polymerase (POL) protein. The polymerase gene overlaps all six other genes, including the core gene that encodes for HBcAg and the precore gene that encodes for the hepatitis B e antigen (HBeAg), the three envelope genes PreS1, PreS2 and S that encode for the large, middle and small envelope proteins, respectively (LHBs, MHBs and SHBs), and the X gene encoding for the multifunctional X protein.
EARLY EVENTS: ATTACHMENT, PENETRATION AND ENTRY Current studies have demonstrated that LHBs are involved in virus entry and binding to hepatocytes.10 Monoclonal antibodies to the PreS1 region of LHBs (amino acids 20–23; Asp- Pro-Ala- Phe) prevented virus attachment.10 Interestingly, antisera to a conformational epitope within the S coding region also prevented attachment, whereas monoclonal antibodies directed to the PreS2 coding region did not totally prevent virus attachment. Earlier studies had determined that the N terminus region of the LHBs protein (codons 21–30) is critical for species specificity, as all hepadnaviruses are highly species and cell-type specific.11 However, using the HDVHBV system with Woolly Monkey hepatitis B virus (WMHBV), Barrera et al.12 found that replacement with codons 1–40 of HBV in the HDV-WMHBV was insufficient to enable infectivity of human
Chapter 7 REPLICATION OF HEPATTIS B VIRUS AND PATHOGENESIS OF DISEASE
Infectious virion S particles
Virus attachment Uncoating Nuclear transport 3'
5' 5'
2.1kb
Synthesis of ccc DNA
AAA
3' AAA
2.4kb
ccc DNA
S HBsAg
Generation of minichromosome
Pre-S1 Pre-S2
5'
3' 0.7kb
Minichromosome
AAA
HBx 5'
Precore mRNA 3.5kb
mRNA Transcription 3'
AAA 5' 3.5kb
Precore Protein synthesis Intracellular conversion pathway
Capsid protein (Core)
Pregenomic mRNA 3' AAA
Figure 7-2. Life cycle of HBV. After attachment, penetration and uncoating of the infectious virion, the viral nucleocapsid is released into the cytosol and transported to the nuclear pore. The viral genome is delivered into the nucleus, where it is converted into cccDNA and the viral minichromosome is generated. The minichromosome is the major transcriptional template from which all RNA transcripts are generated. The pregenomic RNA is translated in the cytosol to produce both the core and the polymerase protein. The polymerase and the pregenomic RNA are encapsidated. HBV DNA synthesis occurs within the nucleocapsid. The nucleocapsids can then either be transported back to the nucleus via the intracellular conversion pathway or be enveloped and secreted into the extracellular space as virions. The envelope proteins are translated at the rough endoplasmic reticulum, and in addition to the virions are also secreted as small particles and tubules containing only envelope proteins. The precore protein is synthesized from the precore mRNA, which is slightly larger than the pregenomic RNA. The precore protein is processed through the Golgi apparatus, where it undergoes multiple cleavages at both its N and its C terminus and is secreted as HBeAg.
Encapsidation with polymerase Capsid protein (Core)
POL
Hepadnavirus DNA synthesis
RNA (+)
Golgi apparatus Envelope and secretion
Mature genome
DNA (+)
DNA (–)
Free virions Secreted HBeAg
hepatocytes, suggesting that other regions of the PreS1 protein may be required for species specificity. The search for the HBV receptor using LHBs has uncovered a large number of potential candidates, including the receptors for immunoglobulin A, interleukin-6, transferrin and asialoglycoprotein, glyceraldehyde-3-phosphate dehydrogenase (GAPD), apolipoprotein H, and human liver annexin V. Unfortunately, none has been unequivocally identified as the major receptor for the specific binding of HBV.13 A number of cofactors as well as a primary receptor may be required for HBV attachment and penetration. MHBs are not necessary for infectivity, yet this region does interact with cellular proteins, which may possibly enhance infection uptake and uncoating.14 The PreS2 encoded region contains a binding site for polymerized human serum albumin (pHSA) and also the transferrin receptor that may facilitate attachment and penetration of HBV to target cells.14
S particles
TRANSPORT OF VIRAL GENOME TO THE NUCLEUS, UNCOATING AND FORMATION OF THE MAJOR TRANSCRIPTIONAL TEMPLATE The HBc protein of HBV encodes a nuclear localization signal (NLS) for the transport of the mature capsids containing the HBV genomes into the nucleus. The nuclear transport is mediated by the imporin pathway using nuclear transport receptors Impb/Impa. The phosphorylation of the C-terminal sequences on the HBc protein is linked with capsid maturation and exposure of the NLS signal. Rabe et al.15 have demonstrated that it is only these mature capsids that are able to move the capsid protein from the collection of nuclear proteins referred to as the ‘nuclear basket’ into the karyoplasms and uncoat their HBV DNA into the nucleus. Hepadnaviruses have an unusual process, referred to as the ‘intracellular conversion pathway’, in which newly synthesized viral
113
Section I. Pathophysiology of the Liver
nucleocapsids can be directed back into the nucleus to increase the pool of the transcriptional template, the covalently closed circular (ccc) DNA. In DHBV the PreS region is involved in the regulation of this intracellular conversion pathway. However, studies examining HBV envelope proteins have not demonstrated the same effect on regulation of the capsids into the nucleus, nor increased cccDNA pools.16 The next stage in HBV viral replication is the conversion of the viral genome into the minichromosome. The HBV viral genome is a partially double-stranded molecule. The HBV viral polymerase may mediate the repair of the ‘gap’ and, in association with host cellular DNA, repair enzymes that facilitate the conversion of viral genome into cccDNA. This conversion also requires the removal of the HBV polymerase protein and oligoribonucleotide and ligation of DNA. Kock et al.17 have demonstrated that the nucleos(t)ide analogs adefovir and lamivudine can inhibit the initial DNA repair process.17 The HBV cccDNA is chromatinized by cellular histone and nonhistone protein and converted into a minichromosome.18 The cccDNA is the major transcriptional template, thus the chromatinization of the HBV genome will affect the binding of transcription factors, thereby regulating transcription. The HBV core protein may influence the spacing of the nucleosome complex on the HBV DNA and hence also the binding of transcription factors.19
TRANSCRIPTION AND TRANSLATION OF THE HBV VIRAL PROTEINS Five promoters control the synthesis of the six viral transcripts of HBV. The HBV genome contains two enhancers, designated enhancer I (EnhI) and enhancer II (EnhII), both of which exhibit greater activity in cell lines of hepatic origin. Although the enhancers are located upstream of specific promoters, EnhI regulates all viral promoters and EnhII regulates the basal core promoter (BCP) as well as the transcription of the PreS2/S promoters. Doitsh et al.9 have proposed that HBV may have both early and late transcriptional events in which EnhI may regulate the expression of the early transcripts of X, and a long X-related transcript of 3.9 kb known as long-X RNA (lxRNA), whereas EnhII appears to regulate late gene transcription events. All RNA molecules are transcribed by the host cell RNA polymerase II using the cccDNA template, are capped and are polyadenylated. The pgRNA and the precore mRNA are longer than genomic length, and transcription is controlled by the BCP. The bifunctional pregenomic mRNA is utilized as the genomic template for reverse transcription of the viral negative-sense DNA and for translation of HBc, Ag and POL proteins, whereas the slightly longer precore mRNA encodes for only the precore protein, which is subsequently processed and secreted as HBeAg.
The Hepatitis B Core Protein (HBcAg) The HBc protein is translated in the cytosol, where it initially forms dimers, followed by multimerization of the dimers to form the nucleocapsid. The HBc protein has been crystallized and the nucleocapsid protein has been studied using cryoelectron microscopy.20 The multimerization of the HBc protein can occur independently of the encapsidization of the pgRNA–POL complex. The HBc protein possesses two distinct domains:14 the N-terminal domain (amino acid residues 1–144), which is involved in dimerization and multi-
114
merization of the nucleocapsid; and the C terminal domain, which includes a potential nuclear localization sequence and is arginine rich. This region is required for nucleic acid binding and in viral encapsidation. The negatively charged codons 113 and 117 were determined to be essential for pgRNA encapsidation.21 When expressed in bacteria the core protein can self-assemble into capsids. Two icosahedral shells of different sizes are observed. Particles with a T = 3 symmetry containing 90 homodimers of 32 nm, and particles with a T = 4 symmetry consisting of 120 homodimers of 36 nm.22 Deletion of an arginine-rich domain at the C terminus allows efficient expression of the protein in bacteria and favors the formation of T = 4 over T = 3 capsids. To date, the preference of T4 over T3 in infected patients has not been determined. Mutational analysis of core has been used to investigate potential interactions between the capsid and the viral envelope.23 The mutations that affected envelopment clustered around the base of the spike and to a small area at the capsid surface close to the pores in the capsid shell.
Hepatitis Be Antigen The precore mRNA is not used for reverse transcription and functions only in translation of the secreted HBeAg, which is one of two accessory proteins of the virus. The first 19 amino acids of the precore protein comprise a secretion signal that allows for the translocation of the precore protein into the lumen of the endoplasmic reticulum (ER). The signal sequence is cleaved off by a host cell signal peptidase and the protein is secreted through the ER and Golgi apparatus. A further modification of the C terminus results in the secretion of a heterogeneous population of 14–17 kDa proteins serologically defined as HBeAg. Alternatively, the precore protein also expresses a signal for transport into the nucleus. The role of HBeAg appears to be for the establishment of persistent infection in vivo,24 to serve an immunoregulatory role in natural infection, and to activate or tolerize T cells.25
Hepatitis B Polymerase (POL) The POL protein is translated from pregenomic RNA. POL contains four functional regions: (1) terminal protein (tp) used in priming HBV DNA synthesis; (2) the spacer region; (3) the reverse transcriptase (rt) that has RNA- and DNA-dependent DNA polymerase activities; and (4) the ribonuclease H (RNase H) that cleaves the RNA in the RNA–DNA hybrids during reverse transcription. The POL protein is covalently attached to the genome and is packaged within the nucleocapsid. The rt region encodes for seven domains G, F, and A to E, which are conserved in other polymerase proteins.26 The HBV POL has not been crystallized, but various homology models have been developed based on the structure of HIV.26 The antiviral nucleos(t)ide analogs that have been developed for the treatment of HBV are all targeted to the rt function of the POL. Resistance to lamivudine,27 adefovir28 and entecavir29 has now been detected. HBV DNA synthesis requires the encapsidation of the HBV POL protein (see below). However, in Duck hepatitis B virus (DHBV), and now recently in HBV-infected cells, a cytoplasmic form of HBV POL has been detected.30 This was detected at relatively lower amounts than DHBV POL, but had a shorter half-life. DHBV POL
Chapter 7 REPLICATION OF HEPATTIS B VIRUS AND PATHOGENESIS OF DISEASE
translation was detected at the same time or earlier than core, even though it is of the same template, and the ratio of POL relative to core dropped with time. Investigations of the translation of DHBV POL have determined that a shunting mechanism is used, and ribosomes are transferred from a donor region near the 5¢ end of the pgRNA to an acceptor site at or near the start AUG of POL. This shunting mechanism may be involved in the regulation of the amount of POL translated. Based on current understanding of HBV DNA replication, the cytoplasmic POL is unlikely to play a role in DNA replication. Mizukoshi et al.31 have identified 10 CD4 T cell epitopes in HBV POL. Cytoplasmic HBV POL may have a role in the adaptive or the innate immune regulation of HBV.
Envelope Proteins: Large (LHBs), Medium (MHBs) and Small Hepatitis B Surface Protein (SHBs) The 2.6 and 2.1 kb RNA transcripts encode the envelope proteins LHBs, MHBs and SHBs. The viral envelope, the small particles and filaments are synthesized and assembled at the ER membranes and then bud into its lumen. The SHBs is 226 amino acids long and is the most abundant protein in all three HBV-associated particles. The SHBs encodes a glycosylation site at codon 146 and both glycosylated and non-glycosylated forms are produced. SHBs contains a high number of cysteine residues that are cross-linked with each other, forming the major hydrophilic region (MHR) that is the major antigenic determinant of HBsAg, described as the ‘a’ determinant. The most-characterized vaccine escape mutant sG145R is located within this region. The MHBs containing the PreS2 domain is a minor component of the virion or HBs particle and consists of the S plus a 55 amino acid N-terminal extension. The MHBs protein is not required for infectivity or virus assembly; however, the glycosylated MHBs is required for virus secretion.32 An N-linked carbohydrate is also attached to Asn-4 of the PreS2 domain of the MHBs protein, which contains an additional modification of an O-linked glycosylation at Thr-37 in the PreS2 domain. The MHBs is considerably more immunogenic than SHBs, and PreS2-containing HBs particles generated from animal cell lines have been used in some countries as a prophylactic vaccine.33 LHBs is more prevalent than MHBs in virions and filaments, but less prevalent in the HBs spherical particles. LHBs contains a further 108 or 119 amino acids (depending on the subtype/genotype) compared to MHBs. LHBs is glycosylated and is modified at Gly-2 of the PreS1 domain by myristylation. The myristylation is essential for viral infection as well as assembly and release.
Hepatitis Bx Protein (HBx) The 1.1 kb transcript encodes for the X protein (HBx), which is 154 amino acids in length and is the second accessory protein of HBV. HBx is not required for in vitro virus replication but is required for the establishment of hepadnaviral infection in woodchucks.34 HBx is located in both the cytoplasm and the nucleus of the cell. The level of HBx expression can influence its cellular localization. It is predominantly nuclear when expressed in cells at very low levels, but becomes largely cytoplasmic as its expression level increases. A number of conflicting studies have been reported for the HBx
protein that are probably related to the expression levels of HBx, its ability to both be a substrate and an inhibitor of the proteasome complex, and its effect in the modulation of cytosolic calcium, which activates various signaling pathways involving Src kinases.1,35,36 HBx is a multifunctional viral protein. In the nucleus, HBx is a modest promiscuous trans-activator that can regulate transcription via direct interaction with different transcription factors. This regulation of viral and cellular genes affects viral replication and viral proliferation, directly or indirectly, and so, not surprisingly, HBV can influence apoptosis and cell cycle regulatory pathways. However, the trans-activation function of X was found to be reduced when proteasome inhibitors were investigated. This interaction of HBx with the 26S proteasome complex may also affect immune evasion by suppressing viral antigen presentation.37 HBx affect on cytosolic calcium may involve an interaction with or action upon the mitochondrial voltage-dependent anion channel,36,38 thereby functioning in the cytoplasm to activate various signaling pathways. In the nucleus, HBx can regulate transcription through a direct interaction with different transcription factors, and in some cases enhance their binding to specific transcriptional elements.39 Overall, whether the HBx activation of transcription or its effect on the cytoplasmic signaling pathways plays a significant role during natural infection with HBV remains an open question.
GENOMIC REPLICATION The process of HBV DNA synthesis is complex and involves not only reverse transcription but also a complicated process of three translocations of the polymerase and primer to complete the doublestranded genome synthesis. The major features of this complex synthesis are: (1) POL protein binds to the encapsidation signal (epsilon, or e) at the 5¢ end of the pgRNA; (2) using part of the bulge region of e as template, the POL protein synthesizes a complementary 3–4 nucleotide (nt) DNA primer that is covalently linked to a tyrosine (Tyr) residue of POL; (3) the covalent complex translocates to a complementary region in the 3¢-proximal DR-1; (4) the primer is extended to a complete negative (-)-strand DNA with concomitant degradation of the RNA template, except for some 15–18 nt at its 5¢ end; (5) this RNA, containing the 5¢ DR-1, is transferred to the complementary DR-2 on the newly made (-)-strand DNA, serving there as primer for the (+)-strand DNA; (6) (+)strand DNA is extended to the physical end of (-)-DNA; (7) using the short terminal redundancy (r), a template switch from the 5¢ to the 3¢ end of (-)-strand DNA occurs and the (+) strand is extended to form the typical relaxed circular (RC) DNA.40 HBV and the other members of the Hepadnaviridae replicate their DNA genome by reverse transcription of a pgRNA template within the subviral core particle. Mature nucleocapids and virions contain the HBV 3.2 kb RC DNA genome with the POL covalently attached to the 5¢ end of the (-)-strand DNA.
ASSEMBLY AND RELEASE OF HBV VIRION The assembly of nucleocapsids containing RC DNA occurs in the cytosol, and these are then selectively enveloped prior to exiting the cell. The LHBs, MHBs and SHBs have different but interrelated functions during viral assembly, related to their transmembrane topologies.41
115
Section I. Pathophysiology of the Liver
The topologies of the SHBs and MHBs are similar and are determined by at least two signal sequences, resulting in a short luminal exposed N-terminal sequence, two transmembrane regions separated by a 55 amino acid cytosolic loop, and a luminal 70 amino acid domain containing the major epitope of the protein and a glycosylation site. The PreS2 region of MHBs is located in the ER lumen. LHBs protein biosynthesis is different, with the entire PreS domain initially remaining in the cytosol following translation. The PreS1 region of LHBs within the virion has a dual topology in which half are on the exterior surface of the virion and half are on the internal surface.41 A specific region within PreS1 at amino acids 70–94 binds to a heat-shock protein that controls the topology of this protein.42 Deletion of this translocation control region results in PreS with a uniform topology. The dual topology of the PreS1 region is required for virus infection and interaction with cellular receptors, as well for as the envelopment of the replicating core particles. The envelope proteins can assemble into either spherical or filamentous subviral lipoprotein particles as well as being incorporated into the outer shell of virions. The L and S proteins are essential for virion formation.43 Studies to determine which regions of the envelope proteins are important for virion formation versus subparticle formation have been performed.44,45 One region was located in LHBs between amino acids 103 and 124. The other region was located in SHBs between amino acids 35 and 46. Although the latter sequence is also present in the C-terminal part of L, the mutations affected virion morphogenesis mainly in the context of S. The MHBs protein is not required for assembly.
DEFECTIVE HBV PARTICLES: ROLE OF SPLICING In addition to the unspliced major HBV RNA transcripts described above, single- or double-spliced 2.2 kb RNAs have been detected in HBV-DNA-transfected hepatoma cells46 and in infected human livers.47 Sequencing of the single-spliced 2.2 kb HBV RNAs typically reveals a deletion from the last codon of the core gene to the middle of the S gene, that creates a new open reading frame, known as the hepatitis B splice protein (HBSP), which includes truncated S and POL proteins.48 The in vivo expression of HBSP is associated with viral replication and, more importantly, liver fibrosis.49 As well as pathogenesis, this alternative replicative strategy may be a mechanism of viral persistence,50 and further studies are clearly indicated to determine its role in viral replication and its effect on the host.
IMMUNOPATHOGENESIS OF HBV HBV infection results in an initial hepatitis that may or may not be symptomatic. Viral clearance depends on the age and immune status of the individual. Most infections in immunocompetent adults are self-limiting. Persistent or chronic infection is more likely to occur following perinatal transmission (from mother to child) or after horizontal transmission to children or immunocompromised adults. The immune determinants of successful clearance of HBV are not fully understood, but depend on both the cellular and the humoral immune responses. At the same time, however, liver inflammation and disease are also believed to be largely immune mediated. Therefore, a complex interaction exists between HBV and the host in both
116
the initial clearance of HBV and the long-term pathogenesis of HBV disease. As discussed above, the study of the replication and pathogenesis of HBV has been limited owing to the lack of available animal models and in vitro cell lines that support HBV infection. HBV can infect chimpanzees, who only acquire a self-limiting acute hepatitis. Other animal models include infection of ducks with Duck hepatitis B and woodchucks with Woodchuck hepatitis B virus. More recent developments in transgenic mouse models have allowed a better understanding of the relative contributions of the different arms of the immune system to HBV, in particular the contribution of the early innate immune response.
ACUTE HBV INFECTION From studies of acute HBV infection in chimpanzees it is clear that HBV DNA peaks and declines prior to the onset of symptoms of acute hepatitis or the induction of a T cell-mediated response. Although no immunological data are available about early intrahepatic events in human infection, animal data support the concept that clearance of HBV DNA is largely mediated by antiviral cytokines produced by cells of the innate and adaptive immune response. In particular, interferon (IFN)-g, tumor necrosis factor (TNF)-a and IFN-a/b are believed to trigger several pathways leading to the inhibition of viral replication without the direct destruction of infected cells. Clinical hepatitis is observed following the decline of HBV DNA levels and is associated with the influx of inflammatory cells, including both HBV-specific and non-specific T cells.51,52 In particular, the appearance of CD8+ T cells that mediate cytolytic activity against HBV-infected cells coincides with an increase in alanine aminotransaminase (ALT) detected in serum.51 Following clearance of the virus and a reduction in ALT, HBV surface antibodies are detected. HBV-specific antibodies, together with HBVspecific memory T cells, incur protective immunity against future infections.
The Innate Immune Response The initial response to HBV is believed to be mediated by nonspecific mechanisms that can be activated in a very short time, ranging from minutes to hours. Among these mechanisms, killing of virus-infected cells without human leukocyte antigen (HLA) restriction or apparent specificity of viral antigens is believed to occur via natural killer (NK) cells, natural killer T (NKT) cells and Kupffer cells (liver-residing macrophages).52–54 NKT cells depend on nonclassic major histocompatibility complex class I-like CD1 molecules for their development, and mainly recognize glycolipids presented by CD1 (recently reviewed in 55). Following infection with HBV, it is believed that hepatocytes, which have low expression of human leukocyte antigen (HLA) class I, release IFN-a and IFN-b. Initial recognition of HBV infection may be mediated by toll-like receptors (TLR) following the detection of pathogen-associated molecular patterns (PAMP).56 These PAMPs probably include viral envelope glycoproteins, double-stranded or single-stranded RNA, and viral accessory proteins such as HBeAg and HBx. Recent studies have shown that several TLRs can recognize viral components and be important mediators of innate immune responses to various viral infections.57 In hepatitis C, the core and
Chapter 7 REPLICATION OF HEPATTIS B VIRUS AND PATHOGENESIS OF DISEASE
NS-3 proteins trigger Toll-like receptor 2 (TLR-2)-mediated pathways and inflammatory activation in vitro.58 In the peripheral blood monocytes of individuals chronically infected with HBV there is a substantial down-regulation of the TLR-2 receptor which is reversed during effective antiviral therapy.56 This latter finding indicates that viruses such as HBV have evolved powerful and effective mechanisms to suppress or block the innate immune response in order to establish productive replication and/or persistence. Further studies in this exciting area are clearly indicated. In HBV transgenic mice the production of IFN-a/b is associated with a 10-fold reduction of viral capsids containing HBV pregenomic RNA and the activation of double-stranded dependent protein kinase activity (PKR), which inhibits HBV protein synthesis.59 In addition to this, IFN-a/b recruits and mediates the activities of antigen-presenting cells (APC), in particular Kupffer cells and dendritic cells (DC). These APCs in turn produce interleukin-18 (IL18) and chemokine CCL3, which induces NK and NKT cell activity (Figure 7-3).60 NKT cells in HBV transgenic mice directly inhibit HBV replication via IFN-g.53,54,60 Injection of a-galactosylceramide, a ligand of CD1d which is used to stimulate NKT cells, resulted in the secretion of cytokines IFN-g and IFN-a/b and subsequent control and inhibition of HBV replication.53 Suppression of HBV was still detected even in T cell-depleted mice, suggesting that IFN-g production is not dependent on CD4+ and CD8+ T cell activity.53 The activity of NK and NKT cells is likely to be an important anti-HBV response that precedes the up-regulation of HLA class I expression on hepatocytes. Up-regulation of HLA class I expression is critical for the presentation and recognition of foreign antigen by T cells in the adaptive arm of the immune response.54 Therefore, the innate immune system probably controls a substantial burden of HBV replication in the early stages of infection prior to the detection of any hepatic inflammatory cell infiltrates or associated liver damage. Kupffer cells play a major role in mediating both early innate and adaptive immune responses. The activation of Kupffer cells via other viral infections, such as malaria, can also lead to enough cytokine production to effectively control and clear HBV. Following infection
of HBV transgenic mice with a liver-specific malaria strain, Kupffer cells produced cytokines that led to the reduction of both the malaria infection and the chronic HBV infection.61 Furthermore, Kupffer cells coordinate the recruitment and maturation of HBVspecific T cells via synthesis of several cytokines and chemokines, including IFN-g, CXCL9, and CXCL10.62 However, complete eradication and control of viral infection cannot be accomplished by the innate immune system alone: the adaptive immune response is needed for total clearance and protection from further HBV infection.
The Adaptive Immune Response Antigen-Presenting Cells (APC) APCs, namely Kupffer cells and in particular DCs, are important for the presentation and maturation of HBV-specific T cells, the main effectors of HBV clearance. APCs present foreign antigen to CD4+ and CD8+ T cells and produce cytokines, IL-12 and TNF-a, which induce IFN-g production and proliferation of CD8+ T cells. IL-12 also induces CD4+ T-cell differentiation into the T-helper cell type 1 (Th1) subset (Figure 7-4).60
CD4+ T Cells
Activated HBV-specific Th1 CD4+ T cells are multispecific, although strong responses against peptides c50–69, found in both HBcAg and HBeAg, are observed following resolution of acute HBV infection, regardless of the HLA phenotype of the infected individual. In acute HBV infection HBV-specific CD4+ T cells can be detected at the time of elevated HBV DNA (before the peak of liver damage) and persist long after recovery. Following the maturation of CD4+ T cells, communication between these Th1 cells and CD8+ T cells is needed for the activation of CD8+ T cell activity. This communication is also mediated by DCs, which are stimulated by the antigenspecific Th1 cells. The altered DCs can then present foreign antigens
Antigen presentation to both CD4+ and CD8+ T-cells DC
HBV
Hepatocyte
CD8+ T-cell
Kupffer cells
Cytolytic activity and production of IFN- and TFN-
IFN-
TLR IFN-
DC IFN- Hepatocyte
NK cells
IL-18 CCL3
NKT cells
Antigen presentation
CD4 differentiates into two subsets CD4+ T-cell
IL-12
IL-2 IFN- TNF- IFN- Th1 TNF- CD4
Hepatocyte
IL-10 IL-4 Adaptive immune responses
Cytokine and chemokine production Figure 7-3. Innate immune response to HBV. A number of these steps are hypothetical and/or derived from animal model studies, with confirmation required from clinical investigation.
Th2 CD4
IL-10 IL-4
B cell
HBeAb HBcAb HBsAb
Figure 7-4. The adaptive immune response to HBV. The complexity of the dynamic cellular interactions is explained in detail in the text.
117
Section I. Pathophysiology of the Liver
to the CD8+ T cells and also induce CD8+ T cell maturation into HBV-specific cytolytic T cells.
CD8+ T Cells
Mature CD8+ T cells are the main effector cells involved in HBV clearance. This was clearly shown by depletion of CD8+ T cells following acute HBV infection in chimpanzees.63 Depletion of CD8+ T cells led to persistence of HBV infection and demonstrated the importance of both cytolytic and non-cytolytic activity of HBVspecific CD8+ T cells.63 In humans, not all those who recover from acute HBV infection have elevated ALT levels or clinical symptoms, suggesting that non-lytic mechanisms such as those induced by IFNg and TNF-a are used to clear acute HBV infection.51 IFN-g is produced mainly by HBV-specific CD8+ T cells, but can also be produced by NK, NKT cells and HBV-specific Th1 CD4+ T cells.53,64 TNF-a and IFN-g clear HBV via several mechanisms, including destabilization of the viral capsid via the NF-kB pathway, degradation of viral proteins via nitric oxide and proteosome activity, and post-transcriptional degradation of HBV RNA.65–68 The importance of these cytokines was confirmed in studies of HBV transgenic mice, where administration of anti-IFN-g and anti-TNF-a antibodies obliterated the ability of CD8+ T cells to clear HBV RNA intermediates and nucleocapsid protein (HBcAg). In acute HBV infection the HBV-specific CD8+ T-cell response is polyclonal and multispecific to most HBV proteins, meaning that there are several receptors present for each epitope and that several epitopes are recognized by a single CD8+ T cell. These two factors respectively increase recognition of the target epitope and reduce viral ‘escape’ via mutation. Therefore, the presence of functional HBV-specific CD8+ T cells, which are maintained by Th1 CD4+ T cells and IL-12, may be more important than the quantity.69 These HBV-specific CD8+ responses have been assessed mainly in HLAA2-positive individuals in whom CD8+ T-cell responses most frequently target core (c18–27), envelope/surface (s183–191, s250–258, s335–343) and polymerase (p455–463, p575–583) epitopes with increasing frequency.70
B Cells The humoral response is also critical to long-term clearance of HBV and protection from infection with HBV. In individuals who recover from acute HBV infection, activated T-helper cell type 2 (Th2) CD4+ T cells induce B-cell production of anti-HBs, anti-HBc and anti-HBe. Anti-HBs antibodies are synthesized early in infection but are not detectable because they are complexed with the excess of envelope antigens produced during virus replication. Anti-HBs is important in providing protective immunity against subsequent HBV infections, and is the basis of protection in vaccinated individuals. The pathogenetic role of antibody to non-envelope protein remains controversial. It is generally accepted that anti-HBc does not have virus-neutralizing activity, although protection of chimpanzees against HBV infection by passive administration of antiHBc/anti-HBe antibodies has been observed, suggesting a possible but undefined role for anti-HBc.
OCCULT HBV INFECTION Both HBV-specific humoral and cellular responses persist after recovery from acute HBV infection. However, in some individuals
118
who recover from acute HBV infection the long-lasting HBVspecific CD8+ T cell response is believed to be a combination of persistence of HBV-specific memory T cells as well as constant restimulation of CD8+ and CD4+ T cells by low amounts of HBV DNA. Persistent HBV at low levels may be considered an occult infection. This long-lasting presence of HBV DNA may be due to the stable nature of cccDNA. Evidence has also shown that HBV infection of immunologically privileged sites, although rare, does occur and may also contribute to occult HBV infection. Therefore, in the setting of immunodeficiency, such as advanced HIV infection or chemotherapy, relapse of HBV infection may occur even if the individual has previously cleared the virus and is anti-HBs positive.
CHRONIC HBV INFECTION The mechanism of HBV persistence is not fully understood but is probably multifactorial, including HBV-specific immune suppression, persistence of stable forms of HBV such as cccDNA, and/or infection of immunologically privileged sites.
HBV-Specific Immune Suppression In individuals with chronic HBV infection the HBV-specific CD4+ and CD8+ T-cell response is significantly diminished.70,71 In particular, in HBeAg-positive chronic carriers the core epitope (in region c18–27) specific CD8+ T cells are almost undetectable and have diminished ability to produce IFN-g. Some HBV-specific CD80+ T cells in persistent HBV infection have been described as ‘partially tolerant’, being unable to bind specific tetramers, to produce IFNg, to lyze, and to expand following stimulation.72 Circulating HBVspecific CD8+ T cells from individuals with chronic HBV infection and a high ALT and high HBV viral load also have decreased proliferative capacity compared to circulating HBV-specific CD8+ T cells from individuals with low ALT and viral loads.70 The reduction in HBV-specific T cells and reduction in IFN-g production is consistent with the responses observed in animal models of other persistent viral infections, such as lymphocyte choriomeningitis virus (LCMV), where there is consecutive elimination of TNF-a and IFNg-producing CD8+ T cells.73 HBV-specific CD8+ T cells are, however, found in the liver, where they may cause an inflammatory response but are ineffective in clearing HBV infection.70 Using HLA-A2-restricted epitopes and tetramers to evaluate HBV-specific T-cell responses, a comparison of CD8+ T-cell responses between HBV chronic carriers with and without an abnormal ALT showed no difference in the absolute quantity of HBV-specific CD8+ T cells in the liver. Individuals with chronic HBV infection and an elevated serum ALT had a larger infiltrate of non-specific CD8+ T cells than patients with a normal ALT.70 Therefore, control of HBV replication may occur by HBV-specific cytotoxic T lymphocytes without causing hepatocyte destruction. Indeed, the presence of these cells is associated with inhibition of viral replication in the absence of liver damage. A generalized CD4+ T-cell hyporesponsiveness in individuals with chronic HBV infection has also been demonstrated. This may be a consequence of impaired function of HBV-infected DCs, which have reduced IFN-g, TNF-a and IL-12 production.74 A reduced overall responsiveness of CD4+ T cells may also contribute to the
Chapter 7 REPLICATION OF HEPATTIS B VIRUS AND PATHOGENESIS OF DISEASE
lack of production of neutralizing antibodies to HBV in chronically infected individuals, as demonstrated in other chronic viral infections, such as that of LCMV.75 The effect of HBV infection on DC function is not well understood, but recent in vitro analysis of monocyte-derived DCs from individuals chronically infected with HBV showed that there is a reduction in the ability of DCs to prime T cells.74 Although there is no difference in absolute numbers of myeloid or plasmacytoid dendritic cells in individuals with HBV infection compared to uninfected controls, significant functional defects were recently demonstrated in both subsets of DCs.76 The mechanism of how HBV impairs dendritic cell function is not currently understood.
Viral Factors Leading to HBV Persistence HBx Protein HBx can modify several cellular pathways, including NF-kB, and this may subsequently affect immune response and antigen presentation.77 HBx protein trans-activates HLA I molecules on hepatocytes, which are normally expressed at low levels. Although an increase in HLA I expression on hepatocytes may assist recognition of foreign antigen by T cells, in the case of HBV it could also recruit T cells to the liver and hence lead to liver damage. Furthermore, in chronic HBV carriers CD8+ T cells have a reduced capacity to downregulate HBx post-transcriptional mRNAs. Therefore, HBx may contribute to the ongoing expression of HLA I on hepatocytes and the recruitment to the liver of inflammatory cells that lack effective antiviral activity.
HBe Antigen (HBeAg) HBeAg acts both as an immunogen and a tolerogen. HBeAg (precore protein) and HBcAg (core protein) share an overlapping reading frame. HBeAg has a leading peptide sequence and a different conformational structure.25 Although HBcAg is needed for the formation of viral capsid and replication, HBeAg is not necessary for infection or replication of HBV.25 However, HBeAg has been conserved in all hepadnaviruses and is believed to be important in the persistence of HBV infection. HBeAg may play a role in immune regulation by depletion of HBeAg- and HBcAg-specific Th1 CD4+ T cells via Fas-mediated apoptosis. This is thought to occur in the setting of perinatal transmission, where HBeAg crosses the placenta from the infected mother and establishes tolerance to HBV in vitro. The imbalance of Th1/Th2 responses leads to the production of anti-inflammatory cytokines such as IL-4 and IL-10, and promotes suppression of HBeAg/HBcAg-specific CD8+ T cell responses and Th1 effector cells. This ‘suppressive’ cytokine profile is reversed when patients chronically infected with HBV seroconvert from HBeAg to anti-HBe, leading to an increase in IL-12 and IFN-g (a Th1 cytokine profile) that may enhance CD8+ T cell function. In keeping with this hypothesis, perinatal transmission from a mother who has persistent HBV infection with an HBeAg-negative mutant (such as HBV with a mutation in the basal core promoter or precore stop codon) is more likely to lead to an acute rather than a persistent HBV infection in the infant. Although infection with an HBeAgnegative mutant is associated with a reduced likelihood of chronic HBV infection, there are conflicting reports demonstrating that HBeAg-negative HBV chronic infection can lead to worse liver
injury and a poorer long-term prognosis,78 but this is always in the setting of previous HBeAg-positive chronic hepatitis B.24 Infection with HBeAg-negative strains has been associated with cases of severe acute hepatitis or fulminant hepatitis.79 The increase in liver injury may be due to the lack of the Th2 ‘skew’ observed in HBeAgpositive HBV chronic infections, which would lead to increased Th1 CD8+ T cell activation.
Other Mechanisms Other possible mechanisms leading to HBV persistence include the replication of HBV in ‘privileged’ locations, such as extrahepatic sites, but why this should occur in some individuals and not others is unclear. Alternatively, induction of Fas-L on hepatocytes could delete HBV-specific CTL more efficiently. Virus-specific CTL might be inactivated if antigen is presented in the absence of co-stimulatory signals in the liver, a mechanism that would specifically allow for T-cell tolerance.80 Finally, viral mutations that abrogate or antagonize antigen recognition of HBV have been reported. Therefore, selection for CTL escape may occur, either during acute infection or once persistent infection is already established.
MECHANISMS OF HBV DISEASE Abnormal Liver Function or ‘Flares’ Levels of HBV viremia in persistent infection are substantially lower than in primary infection, although they vary from person to person. Higher titers of HBV DNA are often indicated by the continued presence of HBeAg. With the passage of time there is a decrease in HBV DNA titer and a tendency for HBeAg to disappear along with the development of anti-HBe antibodies. Seroconversion from HBeAg to anti HBe occurs at a rate of 5–10% per year.81 Often, the disappearance of HBeAg is preceded or accompanied by a transient rise in ALT, known as a flare. A significant reduction of HBV DNA may accompany seroconversion to anti-HBe antibodies.82 These acute exacerbations can be mild with no detectable symptoms, or lead to severe disease, including hepatic decompensation and failure. It is believed that hepatic flares are an attempt by the immune response to clear the virus, and flares occur more commonly in HBeAg-positive individuals.25 However, hepatic flares are also common in the phase of HBeAg-negative chronic HBV.24 In HBeAgpositive HBV, it has been suggested that these responses are triggered by the increasing concentrations of serum HBeAg and intracellular HBcAg. High antigen concentrations would be required for the induction of a T-cell response because of the low avidity of HBcAg and HBeAg-specific T cells in patients with chronic HBV infection.83 Hepatitis flares can also be followed by a significant rise in IL-12 levels that can precede or occur simultaneously with HBeAg seroconversion. The immunopathogenesis of hepatic flares in HBeAg-negative chronic HBV is unknown. Liver biopsies taken during a hepatic flare show infiltration of the lobules with both HBV-specific and non-specific T cells. The recruitment of non-specific inflammatory cells, including macrophages, T cells and neutrophils, may allow for control of HBV viral load before an increase in ALT is detected.51,52 It is believed that IFN-g, needed for recruitment of HBV-specific T cells and also for non-cytopathic clearance of HBV, is also responsible for increasing the susceptiblity of hepatocytes to TNF-a-induced apoptosis and mediating
119
Section I. Pathophysiology of the Liver
macrophage recruitment of more necroinflammatory cells. IFN-ginducible chemokines, including macrophage inflammatory protein 1a (MIP-1a), MIP-1b and RANTES (Regulated on Activation, Normal T Expressed and Secreted), are up-regulated and together with CXCL9 and CXCL10 bind chemokine co-receptor CCR5, which activates lymphocytes regardless of their specificity. Interestingly, in transgenic mice the administration of antibodies against CXCL9 and CXCL10 can lead to reduced intrahepatic inflammation without a reduction in the non-cytopathic effects of IFN-g.62
Chronic Liver Disease, Cirrhosis and Fibrosis Repeated flares or continuous recruitment of inflammatory cells to the liver ultimately results in fibrosis, cirrhosis, and eventually hepatocellular carcinoma (HCC).84,85 Inflammatory cells recruited to the liver cause cell death and hence liver injury through several different mechanisms. Apoptosis of hepatocytes is largely induced by CD8+ T cells via several pathways, including TNF-a.86 Although TNF-a has a non-cytopathic antiviral effect, it can also cause apoptosis via TNF receptor 1 (TNF-R1) and TNF-related apoptosisinducing ligand (TRAIL). In vitro transfection of Hep G2 and Chang cell lines leads to an increase in the susceptibility of hepatocyte cell lines to apoptosis via TRAIL. HBx protein mediates TNF-a-induced apoptosis by increasing the expression of TRAIL receptors on hepatocytes by up to fourfold.87 During HBV infection Fas expression is also up-regulated on hepatocytes.88 Interaction between Fas ligand (FasL; expressed on CD8+ T cells) and Fas (expressed on hepatocytes) also leads to cell apoptosis. Fas-mediated cell death is believed to also play an important role in the severe acute hepatitis observed in fulminant hepatitis where expression of FasL is elevated.89 Finally, the perforin/granzyme pathway has been shown to also mediate hepatocyte apoptosis in HBV infection. Measurement of perforin and granzyme mRNA shows a correlation with histologic activity index and an increase in serum ALT.88 The combined effects of these various apoptosis-inducing mechanisms results in long-term liver damage. Hepatic stellate cells (HSC) are perisinusoidal mesenchymal cells that are activated during liver injury through controlled pathways.84 Activated HSC release extracellular matrix that contains mainly type I collagen, leading to fibrosis. Following liver injury HSC can also phagocytose residual dead cells, and this further induces the release of type I collagen.85 HSCs are themselves protected against apoptosis. The establishment of fibrosis in individuals chronically infected with HBV is largely non-reversible. Individuals with chronic HBV infection have a 200-fold increased risk in HCC compared to HBV-uninfected individuals. HBx protein is thought to have an important role in the pathogenesis of HCC. HBx protein has a high frequency of overexpression in HCC tissue.90,91 HBx protein alters signal transduction pathways and is therefore thought to induce HCC by altering oncogene expression, or by sensitizing HBV-infected cells to other carcinogens.92 Analysis of HBx sequences in HCC show an association with HCC and truncation of the C-terminus of the HBx protein.93 This truncated HBx protein is less effective in inhibiting cellular proliferation, and at the same time suppresses full-length HBx protein activity, allowing for increased proliferation of cells and the development of HCC.94 Likewise, mutations in envelope proteins of HBV have also been shown to be overexpressed in tissue from patients with HCC. Deletions in
120
the PreS1 and PreS2 regions of surface protein lead to the accumulation of truncated surface antigens in the endoplasmic reticulum. Long-term accumulation of the truncated surface antigens leads to oxidative stress, which can cause mutations in hepatocyte DNA and promote HCC formation.95
ANTI-HBV THERAPY AND VIRAL CLEARANCE A reduction in HBV viral load in natural infection appears to precede the detection of HBV-specific T-cell responses in both self-limiting acute infections and flares associated with HBeAg seroconversion in chronic HBV infection. Therefore, a reduction of HBV viral load by antiviral chemotherapy could potentially reduce ‘anergy’ associated with persistent HBV infection. Treatment of HBV-infected patients includes the use of IFN-a and reverse transcriptase inhibitors such as nucleoside analogs (lamivudine; 2¢-deoxy-3¢-thiacytidine; 3TC), and nucleotide analogs (tenofovir [9-(R)-(2-phosphonylmethoxypropyl)adenine] and adefovir [9-(2-phosphonylmethoxyethyl)adenine]). Interferon-a was the first drug to be licensed to treat chronic HBV infection and induce HBeAg and HBsAg seroconversion. Following interferon-a treatment, roughly 35–40% of individuals have a reduction in HBV viral load, and in 8–10% of cases HBsAg seroconversion and viral clearance occurs. Unlike reverse transcriptase inhibitors, IFN-a is capable of mediating immunomodulatory effects that lead to HBV suppression by the immune response. Examination of IFN-a efficacy in perinatally acquired HBV infection has demonstrated mixed results, with some studies showing reduced efficacy and others showing similar seroconversion rates as in Caucasian patients who did not acquire the disease through perinatal transmission. More recently the use of pegylated forms of interferon have shown promising results, with HBeAg seroconversion rates of ~29–37% following 48 weeks of therapy.96 In HBeAg-negative chronic HBV the use of peginterferon-a2a, either alone or in combination with lamivudine, leads to better virological and immunological outcome than the use of lamivudine alone.97 A number of nucleoside analogs have been developed for the treatment of HBV. Lamivudine is most commonly used and has been associated with an increase in HBV-specific CD4+ and CD8+ T-cell responses in both HBeAg-positive and -negative individuals.31,98,99 Lamivudine can successfully suppress HBV viral load, and an increase in functional HBV-specific CD4+ and CD8+ T cells has been detected within the first few months of treatment.99 These HBVspecific T cells can proliferate and produce IFN-g in response to HBV antigen stimulation. Detection of HBV-specific T cells was not associated with HBeAg seroconversion or viral clearance, and appears to be transient.99 Long-term lamivudine therapy (more than 100 weeks) results in HBeAg seroconversion in 27% of individuals. Withdrawal of lamivudine, even in individuals who have HBeAg seroconverted, can lead to a relapse of HBV DNA replication. Interestingly, in individuals infected with both HIV and HBV who are receiving HBV-active therapy, such as lamivudine or tenofovir, HBVspecific CD8+ T cells are detected, but HBV-specific CD4+ T cells are not reconstituted.71 Therefore, it currently appears that antiviral therapy such as lamivudine or adefovir will not cure HBV infection but will provide long-term suppression of HBV replication until drug resistance emerges. The role of a therapeutic vaccine for the treatment of HBV infection remains an area to be further explored.
Chapter 7 REPLICATION OF HEPATTIS B VIRUS AND PATHOGENESIS OF DISEASE
REFERENCES 1. Starkman SE, MacDonald DM, Lewis JC, et al. Geographic and species association of hepatitis B virus genotypes in non-human primates. Virology 2003;314:381–393. 2. Gripon P, Rumin S, Urban S, et al. Infection of a human hepatoma cell line by hepatitis B virus. Proc Natl Acad Sci USA 2002;99:15655–15660. 3. Ren S, Nassal M. Hepatitis B virus (HBV) virion and covalently closed circular DNA formation in primary tupaia hepatocytes and human hepatoma cell lines upon HBV genome transduction with replication-defective adenovirus vectors. J Virol 2001;75: 1104–1116. 4. Delaney WE IV, Isom HC. Hepatitis B virus replication in human HepG2 cells mediated by hepatitis B virus recombinant baculovirus. Hepatology 1998;28:1134–1146. 5. Durantel D, Carrouee-Durantel S, Werle-Lapostolle B, et al. A new strategy for studying in vitro the drug susceptibility of clinical isolates of human hepatitis B virus. Hepatology 2004;40:855–864. 6. Gunther S, Li BC, Miska S, et al. A novel method for efficient amplification of whole hepatitis B virus genomes permits rapid functional analysis and reveals deletion mutants in immunosuppressed patients. J Virol 1995;69:5437–5444. 7. Yang H, Westland C, Xiong S, Delaney WE. In vitro antiviral susceptibility of full-length clinical hepatitis B virus isolates cloned with a novel expression vector. Antiviral Res 2004;61:27–36. 8. Delaney WEt, Yang H, Miller MD, et al. Combinations of adefovir with nucleoside analogs produce additive antiviral effects against hepatitis B virus in vitro. Antimicrob Agents Chemother 2004;48:3702–3710. 9. Doitsh G, Shaul Y. A long HBV transcript encoding pX is inefficiently exported from the nucleus. Virology 2003;309:339–349. 10. Glebe D, Aliakbari M, Krass P, et al. Pre-s1 antigen-dependent infection of Tupaia hepatocyte cultures with human hepatitis B virus. J Virol 2003;77:9511–21. 11. Chouteau P, Le Seyec J, Cannie I, Nassal M, Guguen-Guillouzo C, Gripon P. A short N-proximal region in the large envelope protein harbors a determinant that contributes to the species specificity of human hepatitis B virus. J Virol 2001;75:11565–11572. 12. Barrera A, Guerra B, Lee H, Lanford RE. Analysis of host range phenotypes of primate hepadnaviruses by in vitro infections of hepatitis D virus pseudotypes. J Virol 2004;78:5233–5243. 13. De Falco S, Ruvoletto M, Verdoliva A, et al. Cloning and expression of a novel hepatitis B virus-binding protein from HepG2 cells. J Biol Chem 2001;276:36613–36623. 14. Kann M, Gerlich W. Hepadnaviridae: Structure and molecular virology. In: Zuckerman A, Thomas H, eds. Viral hepatitis. London: Churchill Livingstone, 1998: 77–105. 15. Rabe B, Vlachou A, Pante N, et al. Nuclear import of hepatitis B virus capsids and release of the viral genome. Proc Natl Acad Sci USA 2003;100:9849–9854. 16. Sprinzl MF, Oberwinkler H, Schaller H, Protzer U. Transfer of hepatitis B virus genome by adenovirus vectors into cultured cells and mice: crossing the species barrier. J Virol 2001;75:5108–5118. 17. Kock J, Baumert TF, Delaney WEt, et al. Inhibitory effect of adefovir and lamivudine on the initiation of hepatitis B virus infection in primary tupaia hepatocytes. Hepatology 2003;38:1410–1418. 18. Newbold JE, Xin H, Tencza M, et al. The covalently closed duplex form of the hepadnavirus genome exists in situ as a heterogeneous population of viral minochromosomes. J Virol 1995;69:3350–3357.
19. Bock CT, Schwinn S, Locarnini S, et al. Structural organization of the hepatitis B virus minichromosome. J Mol Biol 2001;307:183–196. 20. Wynne SA, Crowther RA, Leslie AG. The crystal structure of the human hepatitis B virus capsid. Mol Cell 1999;3:771–780. 21. Lee SM, Park SG, Park E, et al. The 113th and 117th charged amino acids in the 5th alpha-helix of the HBV core protein are necessary for pgRNA encapsidation. Virus Genes 2003;27:227–235. 22. Zlotnick A, Cheng N, Conway JF, et al. Dimorphism of hepatitis B virus capsids is strongly influenced by the C-terminus of the capsid protein. Biochemistry 1996;35:7412–7421. 23. Ponsel D, Bruss V. Mapping of amino acid side chains on the surface of hepatitis B virus capsids required for envelopment and virion formation. J Virol 2003;77:416–422. 24. Hadziyannis SJ, Vassilopoulos D. Hepatitis B e antigen-negative chronic hepatitis B. Hepatology 2001;34:617–624. 25. Milich D, Liang TJ. Exploring the biological basis of hepatitis B e antigen in hepatitis B virus infection. Hepatology 2003;38:1075–1086. 26. Bartholomeusz A, Tehan BG, Chalmers DK. Comparisons of the HBV and HIV polymerase, and antiviral resistance mutations. Antivir Ther 2004;9:149–160. 27. Allen MI, Deslauriers M, Andrews CW, et al. Identification and characterization of mutations in hepatitis B virus resistant to lamivudine. Lamivudine Clinical Investigation Group. Hepatology 1998;27:1670–1677. 28. Angus P, Vaughan R, Xiong S, et al. Resistance to adefovir dipivoxil therapy associated with development of a novel mutation in the HBV polymerase. Gastroenterology 2003;125:292–297. 29. Tenney DJ, Levine SM, Rose RE, et al. Clinical emergence of entecavir-resistant hepatitis B virus requires additional substitutions in virus already resistant to Lamivudine. Antimicrob Agents Chemother 2004;48:3498–3507. 30. Cao F, Tavis JE. Detection and characterization of cytoplasmic hepatitis B virus reverse transcriptase. J Gen Virol 2004;85:3353–3360. 31. Mizukoshi E, Sidney J, Livingston B, et al. Cellular immune responses to the hepatitis B virus polymerase. J Immunol 2004;173:5863–5871. 32. Schmitt S, Glebe D, Alving K, et al. Analysis of the pre-S2 N- and O-linked glycans of the M surface protein from human hepatitis B virus. J Biol Chem 1999;274:11945–11957. 33. Tron F, Degos F, Brechot C, et al. Randomized dose range study of a recombinant hepatitis B vaccine produced in mammalian cells and containing the S and PreS2 sequences. J Infect Dis 1989;160:199–204. 34. Zoulim F, Saputelli J, Seeger C. Woodchuck hepatitis virus X protein is required for viral infection in vivo. J Virol 1994;68:2026–2030. 35. Zhang Z, Protzer U, Hu Z, et al. Inhibition of cellular proteasome activities enhances hepadnavirus replication in an HBX-dependent manner. J Virol 2004;78:4566–4572. 36. Rahmani Z, Huh KW, Lasher R, Siddiqui A. Hepatitis B virus X protein colocalizes to mitochondria with a human voltagedependent anion channel, HVDAC3, and alters its transmembrane potential. J Virol 2000;74:2840–2846. 37. Huang J, Kwong J, Sun E-Y, Liang T. Proteasome complex as a potential cellular target of hepatitis H virus X protein. J Virol 1996;70:5582–5591. 38. Bouchard MJ, Wang LH, Schneider RJ. Calcium signaling by HBx protein in hepatitis B virus DNA replication. Science 2001;294:2376–2378. 39. Bouchard MJ, Schneider RJ. The enigmatic X gene of hepatitis B virus. J Virol 2004;78:12725–12734. 40. Nassal M. Hepatitis B virus replication: novel roles for virus–host interactions. Intervirology 1999;42:100–116.
121
Section I. Pathophysiology of the Liver
41. Bruss V, Thomssen R. Mapping a region of the large envelope protein required for hepatitis B virion maturation. J Virol 1994;68:1643–1650. 42. Loffler-Mary H, Werr M, Prange R. Sequence-specific repression of cotranslational translocation of the hepatitis B virus envelope proteins coincides with binding of heat shock protein Hsc70. Virology 1997;235:144–152. 43. Bruss V, Ganem D. The role of envelope proteins in hepatitis B virus assembly. Proc Natl Acad Sci USA 1991;88:1059–1063. 44. Bruss V. A short linear sequence in the pre-S domain of the large hepatitis B virus envelope protein required for virion formation. J Virol 1997;71:9350–9357. 45. Loffler-Mary H, Dumortier J, Klentsch-Zimmer C, Prange R. Hepatitis B virus assembly is sensitive to changes in the cytosolic S loop of the envelope proteins. Virology 2000;270: 358–367. 46. Chen PJ, Chen CR, Sung JL, Chen DS. Identification of a doubly spliced viral transcript joining the separated domains for putative protease and reverse transcriptase of hepatitis B virus. J Virol 1989;63:4165–4171. 47. Su TS, Lui WY, Lin LH, et al. Analysis of hepatitis B virus transcripts in infected human livers. Hepatology 1989;9:180–185. 48. Soussan P, Garreau F, Zylberberg H, et al. In vivo expression of a new hepatitis B virus protein encoded by a spliced RNA. J Clin Invest 2000;105:55–60. 49. Soussan P, Tuveri R, Nalpas B, et al. The expression of hepatitis B spliced protein (HBSP) encoded by a spliced hepatitis B virus RNA is associated with viral replication and liver fibrosis. J Hepatol 2003;38:343–348. 50. Rosmorduc O, Petit MA, Pol S, et al. In vivo and in vitro expression of defective hepatitis B virus particles generated by spliced hepatitis B virus RNA. Hepatology 1995;22:10–19. 51. Webster GJ, Reignat S, Maini MK, et al. Incubation phase of acute hepatitis B in man: dynamic of cellular immune mechanisms. Hepatology 2000;32:1117–1124. 52. Tang TJ, Kwekkeboom J, Laman JD, et al. The role of intrahepatic immune effector cells in inflammatory liver injury and viral control during chronic hepatitis B infection. J Viral Hepatol 2003;10:159–167. 53. Kakimi K, Guidotti LG, Koezuka Y, Chisari FV. Natural killer T cell activation inhibits hepatitis B virus replication in vivo. J Exp Med 2000;192:921–930. 54. Kimura K, Kakimi K, Wieland S, et al. Interleukin-18 inhibits hepatitis B virus replication in the livers of transgenic mice. J Virol 2002;76:10702–10707. 55. Raulet DH. Interplay of natural killer cells and their receptors with the adaptive immune response. Nature Immunol 2004;5:996–1002. 56. Visvanathan K, Skinner N, Locarnini S, et al. Impaired toll-like receptor expression in chronic hepatitis B. Gut 2003;52:130. 57. Boehme KW, Compton T. Innate sensing of viruses by toll-like receptors. J Virol 2004;78:7867–7873. 58. Dolganiuc A, Oak S, Kodys K, et al. Hepatitis C core and nonstructural 3 proteins trigger toll-like receptor 2-mediated pathways and inflammatory activation. Gastroenterology 2004;127:1513–1524. 59. Wieland SF, Guidotti LG, Chisari FV. Intrahepatic induction of alpha/beta interferon eliminates viral RNA-containing capsids in hepatitis B virus transgenic mice. J Virol 2000;74:4165–4173. 60. Kimura K, Kakimi K, Wieland S, et al. Activated intrahepatic antigen-presenting cells inhibit hepatitis B virus replication in the liver of transgenic mice. J Immunol 2002;169:5188–5195. 61. Pasquetto V, Guidotti LG, Kakimi K, et al. Host–virus interactions during malaria infection in hepatitis B virus transgenic mice. J Exp Med 2000;192:529–536. 62. Kakimi K, Lane TE, Chisari FV, Guidotti LG. Cutting edge: Inhibition of hepatitis B virus replication by activated NK T cells does not require inflammatory cell recruitment to the liver. J Immunol 2001;167:6701–6705.
122
63. Thimme R, Wieland S, Steiger C, et al. CD8(+) T cells mediate viral clearance and disease pathogenesis during acute hepatitis B virus infection. J Virol 2003;77:68–76. 64. Szkaradkiewicz A, Jopek A, Wysocki J, et al. HBcAg-specific cytokine production by CD4 T lymphocytes of children with acute and chronic hepatitis B. Virus Res 2003;97:127–133. 65. Biermer M, Puro R, Schneider RJ. Tumor necrosis factor alpha inhibition of hepatitis B virus replication involves disruption of capsid integrity through activation of NF-kappaB. J Virol 2003;77:4033–4042. 66. Guidotti LG, McClary H, Loudis JM, Chisari FV. Nitric oxide inhibits hepatitis B virus replication in the livers of transgenic mice. J Exp Med 2000;191:1247–1252. 67. Robek MD, Wieland SF, Chisari FV. Inhibition of hepatitis B virus replication by interferon requires proteasome activity. J Virol 2002;76:3570–3574. 68. Guidotti LG, Morris A, Mendez H, et al. Interferon-regulated pathways that control hepatitis B virus replication in transgenic mice. J Virol 2002;76:2617–2621. 69. Maini MK, Reignat S, Boni C, et al. T cell receptor usage of virus-specific CD8 cells and recognition of viral mutations during acute and persistent hepatitis B virus infection. Eur J Immunol 2000;30:3067–3078. 70. Maini M, Boni C, Lee C, et al. The role of virus-specific CD8(+) cells in liver damage and viral control during persistent hepatitis B virus infection. J Exp Med 2000;191:1269. 71. Chang J, Wightman F, Bartholomeusz A, et al. Reduced hepatitis B virus (UBV)-specific CD4+ T cell responses in human immunodeficiency virus type 1-HBV coinfected individuals receiving HBV-active antiretroviral therapy. J Virol 2005;79:3038–3051. 72. Reignat S, Webster GJ, Brown D, et al. Escaping high viral load exhaustion: CD8 cells with altered tetramer binding in chronic hepatitis B virus infection. J Exp Med 2002;195:1089–1101. 73. Wherry EJ, Blattman JN, Murali-Krishna K, et al. Viral persistence alters CD8 T cell immunodominance and tissue distribution and results in distinct stages of functional impairment. J Virol 2003;77:4911–4927. 74. Beckebaum S, Cicinnati V, Zhang X, et al. Hepatitis B virusinduced defect of monocyte-derived dendritic cells leads to impaired T helper type 1 response in vitro: mechanisms for viral immune escape. Immunology 2003;109:487–495. 75. Ciurea A, Hunziker L, Klenerman P, et al. Impairment of CD4(+) T cell responses during chronic virus infection prevents neutralizing antibody responses against virus escape mutants. J Exp Med 2001;193:297–305. 76. van der Molen RG, Sprengers D, Binda RS, et al. Functional impairment of myeloid and plasmacytoid dendritic cells of patients with chronic hepatitis B. Hepatology 2004;40:738–746. 77. Murakami S. Hepatitis B virus X protein: a multifunctional viral regulator. J Gastroenterol 2001;36:651–660. 78. Funk ML, Rosenberg DM, Lok AS. World-wide epidemiology of HBeAg-negative chronic hepatitis B and associated precore and core promoter variants. J Viral Hepatol 2002;9:52–61. 79. Bartholomeusz A, Locarnini S. Hepatitis B virus mutants and fulminant hepatitis B: fitness plus phenotype. Hepatology 2001;34:432–435. 80. Limmer A, Ohl J, Kurts C, et al. Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T cell tolerance. Nature Med 2000;6:1348–1354. 81. Ribeiro RM, Lo A, Perelson AS. Dynamics of hepatitis B virus infection. Microbes Infect 2002;4:829–835. 82. Tedder RS, Ijaz S, Gilbert N, et al. Evidence for a dynamic host–parasite relationship in e-negative hepatitis B carriers. J Med Virol 2002;68:505–512. 83. Chen M, Sallberg M, Thung SN, et al. Nondeletional T cell receptor transgenic mice: model for the CD4(+) T cell
Chapter 7 REPLICATION OF HEPATTIS B VIRUS AND PATHOGENESIS OF DISEASE
84.
85.
86.
87.
88.
89.
90.
91.
repertoire in chronic hepatitis B virus infection. J Virol 2000;74:7587–7599. Friedman SL. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury. J Biol Chem 2000;275:2247–2250. Canbay A, Taimr P, Torok N, et al. Apoptotic body engulfment by a human stellate cell line is profibrogenic. Lab Invest 2003;83:655–663. Jo M, Kim TH, Seol DW, et al. Apoptosis induced in normal human hepatocytes by tumor necrosis factor-related apoptosisinducing ligand. Nature Med 2000;6:564–567. Janssen HL, Higuchi H, Abdulkarim A, Gores GJ. Hepatitis B virus enhances tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) cytotoxicity by increasing TRAIL-R1/death receptor 4 expression. J Hepatol 2003;39:414–420. Lee JY, Chae DW, Kim SM, et al. Expression of FasL and perforin/granzyme B mRNA in chronic hepatitis B virus infection. J Viral Hepatol 2004;11:130–135. Rivero M, Crespo J, Fabrega E, et al. Apoptosis mediated by the Fas system in fulminant hepatitis by hepatitis B virus. J Viral Hepatol 2002;9:107–113. Wang XZ, Chen XC, Chen YX, et al. Overexpression of HBxAg in hepatocellular carcinoma and its relationship with Fas/FasL system. World J Gastroenterol 2003;9:2671–2675. Hwang GY, Lin CY, Huang LM, et al. Detection of the hepatitis B virus X protein (HBx) antigen and anti-HBx antibodies in cases of human hepatocellular carcinoma. J Clin Microbiol 2003;41:5598–5603.
92. Tralhao JG, Roudier J, Morosan S, et al. Paracrine in vivo inhibitory effects of hepatitis B virus X protein (HBx) on liver cell proliferation: an alternative mechanism of HBx-related pathogenesis. Proc Natl Acad Sci USA 2002;99:6991–6996. 93. Iavarone M, Trabut JB, Delpuech O, et al. Characterisation of hepatitis B virus X protein mutants in tumour and non-tumour liver cells using laser capture microdissection. J Hepatol 2003;39:253–261. 94. Tu H, Bonura C, Giannini C, et al. Biological impact of natural COOH-terminal deletions of hepatitis B virus X protein in hepatocellular carcinoma tissues. Cancer Res 2001;61:7803–7810. 95. Hsieh YH, Su IJ, Wang HC, et al. Pre-S mutant surface antigens in chronic hepatitis B virus infection induce oxidative stress and DNA damage. Carcinogenesis 2004;25:2023–2032. 96. Cooksley WG, Piratvisuth T, Lee SD, et al. Peginterferon alpha-2a (40 kDa): an advance in the treatment of hepatitis B e antigen-positive chronic hepatitis B. J Viral Hepatol 2003;10:298–305. 97. Marcellin P, Lau GK, Bonino F, et al. Peginterferon alfa-2a alone, lamivudine alone, and the two in combination in patients with HBeAg-negative chronic hepatitis B. N Engl J Med 2004;351:1206–1217. 98. Malacarne F, Webster GJ, Reignat S, et al. Tracking the source of the hepatitis B virus-specific CD8 T cells during lamivudine treatment. J Infect Dis 2003;187:679–682. 99. Boni C, Penna A, Bertoletti A, et al. Transient restoration of antiviral T cell responses induced by lamivudine therapy in chronic hepatitis B. J Hepatol 2003;39:595–605.
123
Section I: Pathophysiology of the Liver
REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
8
Darius Moradpour and Charles M. Rice Abbreviations ARF alternative reading frame ARFP alternative reading frame protein BVDV bovine viral diarrhea virus cDNA complementary DNA CRE cis-acting replication element CsCl cesium chloride DC-SIGN dendritic cell-specific intercellular adhesion molecule-3-grabbing integrin ER endoplasmic reticulum F frameshift protein GBV GB virus HCC hepatocellular carcinoma
HCV HCVDB HGV hVAP-A HVR IRES ISDR LDL LDLR L-SIGN
hepatitis C virus Hepatitis C Virus DataBase hepatitis G virus human vesicle-associated membrane protein-associated protein A hypervariable region internal ribosome entry site interferon sensitivity-determining region low-density lipoprotein low-density lipoprotein receptor liver/lymph node-specific intercellular adhesion molecule-3-grabbing integrin
NCR NK NS PEG-IFN-a RdRp SR-BI TBE VLDL VSV YFV
non-coding region natural killer non-structural pegylated interferon-a RNA-dependent RNA polymerase scavenger receptor class B type I tickborne encephalitis very low-density lipoprotein vesicular stomatitis virus yellow fever virus
INTRODUCTION
TAXONOMY
Hepatitis C virus (HCV) infection is a major cause of chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma (HCC) worldwide.1 A protective vaccine is not available and therapeutic options are limited. Current standard therapy, pegylated interferon-a (PEGIFN-a) combined with ribavirin, results in a sustained virologic response in 20–80% of patients, depending on the HCV genotype and other factors.2–5 However, in clinical practice many patients do not qualify for or do not tolerate IFN-based therapy.6 As a consequence, the number of patients presenting with long-term sequelae of chronic hepatitis C, including HCC, is expected to further increase over the next 20–30 years.7 Thus, there is an urgent need to develop more effective and better-tolerated therapies for chronic hepatitis C. A detailed understanding of the molecular virology of hepatitis C underpins these efforts. HCV was identified in 1989 as the most common etiologic agent of post-transfusion and sporadic non-A, non-B hepatitis by the use of recombinant DNA technology.8 Investigation of the viral life cycle has been limited by the low viral titers found in the sera and livers of infected individuals and the lack of efficient cell culture systems or small animal models permissive for HCV. Nevertheless, considerable progress has been made using heterologous expression systems,9,10 functional cDNA clones,11 the replicon system,12,13 functional HCV pseudoparticles,14,15 and most recently, recombinant infections HCV produced in vitro181–183 (see 16–19 for recent reviews). These and other milestones in HCV research are listed in Table 8-1.
HCV has been classified in the Hepacivirus genus within the family Flaviviridae, which includes the classic flaviviruses, such as yellow fever (YFV) and dengue viruses, the animal pestiviruses, such as bovine viral diarrhea virus (BVDV), and the as yet unassigned GB viruses A (GBV-A), GBV-B and GBV-C20 (Figure 8-1). GBV-C was also designated hepatitis G virus (HGV). However, it was subsequently found that GBV-C/HGV is not a common agent of viral hepatitis and its pathogenic relevance, if any, remains unknown. An important feature of HCV is its high genetic variability.21 HCV isolates fall into three major categories, depending on the degree of sequence divergence: genotypes, subtypes, and isolates. There are six major genotypes (also called ‘clades’) that differ in their nucleotide sequence by 30–35%. Within an HCV genotype, several subtypes (designated a, b, c etc.) can be defined that differ in their nucleotide sequence by 20–25%. The term quasispecies refers to the genetic heterogeneity of the population of HCV genomes coexisting in an infected individual. The genetic variability of HCV may have important implications for the pathogenesis, natural course and prevention of hepatitis C. The E1 and E2 glycoprotein regions are particularly variable, whereas the core and some of the non-structural protein sequences are more conserved. The highest degree of sequence conservation is found in the 5¢ and 3¢ non-coding regions (NCR). In the United States and western Europe, genotypes 1a and 1b are the most frequent, followed by genotypes 2 and 3. In Europe, genotype 3 is distributed widely among injection drug users. In
125
Section I. Pathophysiology of the Liver
Pestivirus BVDV
CSFV
GBV-B GBV-C Flavivirus
3a
YFV HCV
1b 2a 1a
JEV
DENV
Hepacivirus
Identification of HCV Polyprotein processing Three-dimensional structure of the NS3 serine protease Infectious clone of HCV Replicon system HCV pseudoparticles Proof-of-concept clinical studies of an HCV protease inhibitor Recombinant infectious HCV
southern and eastern Europe, genotype 1b is most frequent. In Japan, China, and Taiwan genotypes 1b and 2 are predominant. Genotype 4 is found primarily in Egypt, North and Central Africa, and the Middle East. It is typically associated with past medical treatment, e.g. parenteral treatment for schistosomiasis. Genotype 5 is found commonly only in South Africa and genotype 6 is found among intravenous drug users in Hong Kong, Vietnam and, more recently, Australia. Patients infected with genotype 1 have a poorer response to IFNa therapy than those infected with genotype 2 or 3. However, the clinical significance of HCV genotypes with respect to the natural history of hepatitis C is controversial. Close to 20 000 HCV sequences, including nearly 200 full-length genomes, have so far been deposited in generic databanks such as GenBank, the EMBL Nucleotide Sequence Database or the DNA Data Bank of Japan (DDBJ). A number of sequence databases are dedicated specifically to HCV, including the Hepatitis C Virus DataBase (HCVDB) of the French Réseau National Hépatites (http://hepatitis.ibcp.fr), the Los Alamos Hepatitis C Virus Databases (http://hcv.lanl.gov), and the Japanese Hepatitis Virus Database (http://s2as02.genes.nig.ac.jp). These offer a number of specialized features as well as useful links for HCV sequence analysis, structure predictions, CD4+ and CD8+ T-cell epitope compilations etc.
GENETIC ORGANIZATION HCV contains a 9.6 kb positive-strand RNA genome composed of a 5¢ NCR, a long open reading frame encoding a polyprotein precursor of about 3000 amino acids, and a 3¢ NCR (Figure 8-2).
126
GBV-A
2b
Table 8-1. Milestones in HCV Research 1989 1993 1996 1997 1999 2003 2003 2005
Figure 8-1. Simplified phylogenetic tree of the Flaviviridae family. The Flaviviridae comprise the genera Flavivirus, Pestivirus and Hepacivirus as well as the as yet unassigned GB viruses A (GBV-A), GBVB and GBV-C. Only few examples of flavi- and pestiviruses, as well as few HCV genotypes and subtypes, are shown. YFV, yellow fever virus; JEV, Japanese encephalitis virus; DENV, Dengue virus; BVDV, bovine viral diarrhea virus; CSFV, classical swine fever virus.
It took 8 years from the discovery of HCV to establish the first infectious cDNA clone,11 because in the absence of a robust tissue culture system the only read-out for infectivity was the direct inoculation of in vitro transcribed, synthetic RNA into the liver of a chimpanzee. In addition, owing to the variation present in the quasispecies and errors introduced by PCR (polymerase chain reaction) amplification, construction of infectious cDNA clones required the preparation of a consensus sequence from a number of clones. Functional cDNA clones now exist for genotypes 1a,11,22–24 1b,25 and 2a.26 Genetic studies using infectious clones have shown the essential nature of the HCV enzymes, the conserved elements of the 3¢ NCR and the difficult-to-study proteins such as p7.27,28
THE 5¢ AND 3¢ NON-CODING REGIONS The 5¢ NCR is highly conserved among different HCV isolates and contains an internal ribosome entry site (IRES) essential for capindependent translation of the viral RNA.29,30 Because the vast majority of cellular mRNAs are translated by a cap-dependent mechanism, the HCV IRES represents an attractive antiviral target. The 5¢ NCR contains four highly ordered domains, numbered I–IV. Domain I is not required for IRES activity, but is essential for HCV RNA replication.31 Domains II and III include two large stemloops. Subdomain IIIf forms a pseudoknot with domain IV, which contains the translation initiation codon. Domains II, III and IV, together with the first 24–40 nucleotides of the core coding region, constitute the IRES. The key element is domain III, which permits direct binding of the 40 S ribosomal subunit to subdomains IIIa, IIIc, IIId and IIIe, as well as of eukaryotic translation initiation factor 3 (eIF3) to subdomain IIIb. The three-dimensional structure of the HCV IRES bound to the 40 S ribosomal subunit was resolved at 20 Å resolution by cryoelectron microscopy.32 Strikingly, it was found that IRES binding induces a significant conformational change in the 40 S subunit, indicating that the HCV IRES dynamically modulates the host translational machinery. In addition, high-resolution structural information is now available for critical elements of the IRES, including stem loops II, IIIb, IIId and IIIe, as well as the IIIabc four-way junction, facilitating the design of small molecule inhibitors of HCV translation initiation.33–36 A current model of HCV translation initiation includes the formation of a binary complex between the IRES and the 40 S ribosomal subunit, followed by assembly of a 48 S-like complex at the
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
5´ NCR
5B-SL3 5B-SL3.2
III
b II a
c e
d
(U)n
9.6 kb
f
I
3´ NCR
IV
IRES-mediated translation
C
E1
E2
NS2
NS3
A NS4 B
A
NS5
B
Polyprotein processing 192
1 C Core
E1
384
747 810 E2
* * * * * * * * Envelope glycoproteins
p7
NS2 Protease
1658 1712
1027 NS3
A
NS4B
Serine Helicase Serine Membr. web protease protease cofactor
3011
2421
1973 NS5A
NS5B
?
RNA-dependent RNA polymerase
Figure 8-2. Genetic organization and polyprotein processing of HCV. The 9.6 kb positive-strand RNA genome is schematically depicted at the top. Simplified RNA secondary structures in the 5¢ and 3¢ non-coding regions (NCRs) as well as in the NS5B stem-loop 3 cis-acting replication element (5B-SL3) are shown. Internal ribosome entry site (IRES)-mediated translation yields a polyprotein precursor of about 3000 amino acids that is processed into the mature structural and non-structural proteins. Amino acid positions are shown above each protein (HCV H strain; genotype 1a; GenBank accession number AF009606). Solid diamonds denote cleavage sites of the HCV polyprotein precursor by the endoplasmic reticulum signal peptidase. The open diamond indicates further C-terminal processing of the core protein by signal peptide peptidase. Arrows indicate cleavages by the HCV NS2-3 and NS3 proteases. Asterisks in the E1 and E2 region indicate glycosylation of the envelope proteins. Note that polyprotein processing, illustrated here as a separate step for simplicity, occurs both co- and posttranslationally.
AUG initiation codon upon association of eIF3 and ternary complex (eIF2•Met-tRNAiMet•GTP) and, in a rate limiting step, GTPdependent association of the 60 S subunit to form the 80 S complex.37 The 3¢ NCR is composed of a short variable region, a poly(U/UC) tract with an average length of 80 nucleotides, and an almost invariant 98 nucleotide RNA element, designated the X-tail.27,38–43
CIS-ACTING REPLICATION ELEMENTS Promoter elements regulating the replication of positive-strand viral RNAs, called cis-acting replication elements (CRE), are often found at or near the 5¢ and 3¢ termini of genome RNA (or its complement). Although only the 5¢-terminal 125 bases including domain I are absolutely required, downstream sequences that include the entire 5¢ NCR dramatically enhance the efficiency of HCV RNA replication.31,44 The conserved elements in the 3¢ NCR, including a minimal poly(U) tract of about 25 bases, are also essential for replication both in cell culture41–43 and in vivo.27,40 Besides the 5¢ and 3¢ NCRs, conserved RNA structures have been predicted within the HCV open reading frame.45,46 A new CRE was recently confirmed in the sequence encoding the C-terminal domain of non-structural protein 5B (NS5B).47 An essential stem-loop, designated 5B-SL3.2, was identified within a larger cruciform RNA
element designated 5B-SL3 (Figure 8-2). More recently, it was shown that the upper loop of 5B-SL3.2 is engaged in a kissing interaction with a stem-loop in the X-tail, suggesting that a pseudoknot structure is formed at the 3¢ end of the HCV genome that is essential for RNA replication.48
POLYPROTEIN PROCESSING IRES-mediated translation of the HCV open reading frame yields a polyprotein precursor that is co- and post-translationally processed by cellular and viral proteases into the mature structural and nonstructural proteins (Figure 8-2). The structural proteins include the core protein and the envelope glycoproteins E1 and E2. These are released from the polyprotein precursor by the endoplasmic reticulum (ER) signal peptidase. The structural proteins are separated from the non-structural proteins by the p7 polypeptide. The nonstructural proteins include the NS2-3 protease and the NS3 serine protease, an RNA helicase/NTPase located in the C-terminal twothirds of NS3, the NS4A polypeptide, the NS4B and NS5A proteins, and the NS5B RNA-dependent RNA polymerase (RdRp). The NS2-3 protease cleaves at the NS2/NS3 site, whereas the NS3 serine protease is responsible for processing of the downstream nonstructural proteins (Figure 8-2). These are cleaved in a preferential
127
Section I. Pathophysiology of the Liver
order, as shown in heterologous expression systems and in HuH-7 cells harboring HCV replicons (see below).10,49 The first cleavage occurs co-translationally and liberates NS3 from the remainder of the polyprotein. Subsequent processing events can be mediated in trans, with rapid processing at the NS5A/NS5B site. The resulting NS4A-5A precursor is cleaved first between NS4A and NS4B, resulting in a relatively stable NS4B-5A intermediate, and subsequently between NS4B and NS5A.
STRUCTURAL PROTEINS CORE The first structural protein encoded by the HCV open reading frame is the core protein, which presumably forms the viral nucleocapsid. During translation of the HCV polyprotein, the nascent polypeptide is targeted to the ER membrane for translocation of the E1 ectodomain into the ER lumen, a process mediated by an internal signal sequence located between the core and E1 sequences. Cleavage of the signal sequence by signal peptidase yields an immature 191-amino-acid core protein, which contains the E1 signal peptide at its C terminus. This signal peptide is further processed by signal peptide peptidase (SPP), yielding the mature 21 kDa core protein of approximately 179 amino acids.50 The N-terminal hydrophilic domain of core contains a high proportion of basic amino acid residues and has been implicated both in RNA binding and homo-oligomerization. When expressed in mammalian cells, core is found on membranes of the ER, in seemingly ER-derived membranous webs (see below), and on the surface of lipid droplets.51–54 It is at present unclear whether the association with lipid droplets, which is mediated by the central, relatively hydrophobic domain of core and was detected in different heterologous expression systems, in transgenic mice and in liver specimens from HCV-infected chimpanzees, plays a role during viral replication or virion morphogenesis. It has been speculated that the interaction of core with lipid droplets may affect lipid metabolism, which in turn may contribute to the development of liver steatosis. The observation that certain HCV core-transgenic mice develop steatosis and HCC has lent further support to this hypothesis.55,56 A small proportion of the core protein may also be found in the nucleus. Little is known about the assembly of core into nucleocapsids. In vitro studies with recombinant HCV core proteins demonstrated that the N-terminal 124 amino acid residues are sufficient for the assembly of nucleocapsid-like structures, and that the presence of structured RNA is required for this process.57 However, under these experimental conditions RNA encapsidation is not specific, and the signals and processes that mediate RNA packaging and nucleocapsid assembly during HCV replication are unknown. More recently, assembly of nucleocapsid-like particles has been observed in cellfree translation systems.58 These capsids sediment at 100S and have a buoyant density of 1.28 g/ml on CsCl gradients. Intriguingly, the core protein has been reported to interact with a variety of cellular proteins and to influence numerous host cell functions, including apoptosis, cell cycle control, gene expression and many others.59,60 However, the relevance of these observations, derived mainly from heterologous overexpression experiments, for the natural course and pathogenesis of hepatitis C is currently unknown.
128
ARFP/F PROTEIN An alternative reading frame (ARF) was recently identified in the HCV core region which, as a result of a -2/+1 ribosomal frameshift, has the potential to encode a protein of up to 160 amino acids, designated ARFP (alternative reading frame protein) or F (frameshift) protein46,61,62 (Figure 8-3). Expression of the ARFP/F protein of HCV genotype 1a in vitro or in mammalian cells yields a 17 kDa protein. Amino acid sequencing indicated that the frameshift probably occurs at or near codon 11 of the core protein sequence.61 However, multiple frameshifting events have recently been reported in this region, and a 1.5 kDa protein could also be produced by -1/+2 frameshifting.63 In addition, the frameshift position seems to be genotype dependent, as a +1 frameshift at codon 42 was recently reported for genotype 1b.64 Detection of antibodies65 and T cells66 specific for the ARFP/F protein in patients with hepatitis C suggests that this protein is expressed during HCV infection. However, given that the ARF is not present in subgenomic HCV replicons, the ARFP/F protein is not required for HCV RNA replication in vitro. Recent studies incorporating multiple stop codons into the ARF have shown that ARFP/F protein expression is not absolutely required for replication in vivo.67 Rather, these results suggest that the ARF may harbor additional conserved RNA elements that are required for translation and/or replication of full-length HCV genome RNAs. Thus, the functions, if any, of the ARFP/F protein in the life cycle and pathogenesis of HCV remain to be elucidated.68
ENVELOPE GLYCOPROTEINS The envelope proteins E1 and E2 are extensively glycosylated and have an apparent molecular weight of 30–35 and 70–72 kDa, respectively. They form a non-covalent complex, which is believed to represent the building block for the viral envelope.69 E2 is believed to make contact with the cellular receptor(s) for HCV, whereas E1 has been predicted to possess fusion activity.
5´ NCR
C
E1
ARF
ARFP/F Figure 8-3. The alternative reading frame (ARF) in the HCV core coding region and the ARF protein (ARFP)/F protein. The 5¢ non-coding region (5¢ NCR) of the HCV genome contains extensive secondary structures forming an internal ribosome entry site. The main open reading frame of the HCV genome with the core protein (C) and the N-terminal portion of envelope glycoprotein 1 (E1) are depicted in gray. The ARF is illustrated in red. The putative ARFP/F protein is shown at the bottom. The frameshift from the core reading frame into the ARF occurs at or near codon 11 of the core coding sequence.
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
E1 and E2 are type I transmembrane glycoproteins. Interestingly, the transmembrane domains, located at their C termini, are involved in heterodimerization and have ER retention properties. Each of these transmembrane domains is composed of two stretches of hydrophobic amino acids separated by a short polar segment. The second hydrophobic stretch acts as an internal signal peptide for the downstream protein. Before signal sequence cleavage, the E1 and E2 transmembrane domains have been proposed to adopt a hairpin structure at the translocon. After cleavage, the signal sequence is reoriented towards the cytosol, resulting in a single transmembrane passage.70 The ectodomains of E1 and E2 contain numerous highly conserved cysteine residues that may form four and nine intramolecular disulfide bonds, respectively. In addition, E1 and E2 contain up to five and 11 glycosylation sites, respectively. Thus, HCV glycoprotein maturation and folding is a complex process that involves the ER chaperone machinery and depends on disulfide bond formation as well as glycosylation. A model for E2 based on the structure of the envelope protein from tick-borne encephalitis virus (TBE; a member of the flavivirus genus)71 was proposed.72 According to this model, E2 forms an elongated and flat head-to-tail homodimer. The fact that the envelope protein of Semliki Forest virus,73 a more distantly related alphavirus, has a similar structure to the envelope proteins of TBE71 and dengue virus,74 suggests that HCV may have a similar surface architecture. However, virtually nothing is known about the actual structure of the HCV E1–E2 complex, and the processes that mediate viral attachment, entry, and fusion have only recently become amenable to systematic study (see below). As discussed above, the genes encoding the envelope glycoproteins E1 and E2 are particularly variable. A hypervariable region (HVR) of approximately 28 amino acids in the N-terminal domain of E2 has been termed HVR1.75,76 The HVR1 amino acid sequence differs by up to 80% among HCV isolates. Interestingly, despite high variability at the sequence level, the structure of this domain was found to be quite conserved.77 HVR1 appears to contain a neutralization epitope78 and variability, therefore, may be driven by antibody selection. Of note, the HVR1 could be deleted from infectious HCV cDNA clones without abrogating infectivity, although the mutant virus replicated poorly and compensating changes in E1 and E2 were selected upon passaging.79 These observations suggest a functional role of this domain, probably in virus entry into the host cell. A second hypervariable region, HVR2, has been described at amino acid positions 91–97 of genotype 1 E2 protein.
p7 p7 is a 63-amino-acid polypeptide that is often incompletely cleaved from E2. It has two transmembrane domains connected by a short hydrophilic segment which forms a cytoplasmic loop, and the N and C termini are oriented toward the ER lumen.80 Both transmembrane passages have been predicted to form a-helices, and the C-terminal transmembrane segment has been shown to function as an internal signal peptide. p7 of the related pestivirus BVDV is essential for the production of infectious progeny, but not for RNA replication.81 Similarly, HCV p7 is not required for HCV RNA replication because it is not present
in subgenomic replicons. However, it is essential for virus infectivity in vivo, as shown by genetic studies using infectious HCV cDNA clones.28 p7 has recently been reported to form hexamers and to possess ion channel activity.82,83 These properties suggest that p7 belongs to the viroporin family, could have an important role in viral particle maturation and release, and may represent an attractive target for antiviral intervention.
VIRION STRUCTURE Although exciting progress has recently been made with respect to related flavi-84–86 and alphaviral virion structures,73 HCV has not so far been conclusively visualized and its structure remains unknown. By analogy to these related viruses, it can be assumed that the core protein and the envelope glycoproteins E1 and E2 are the principal structural components of the virion. E1 and E2 are presumably anchored to a host cell-derived double-layer lipid envelope that surrounds a nucleocapsid composed of multiple copies of the core protein and encapsidating the genomic RNA. The basic biophysical properties of the HCV particle were revealed early on by experiments performed in chimpanzees. Infectivity was abolished by treatment with lipid solvents, indicating that the viral particle is enveloped.87 A rough estimate of virion size was obtained by filtration studies demonstrating that the particle is able to pass through 50 nm pore filters.88 Based on subsequent electron microscopy studies, HCV particles are believed to have a diameter of 40–70 nm.89,90 HCV circulates in various forms in the infected host, including virions bound to low-density (LDL) and very lowdensity lipoproteins (VLDL),91 which appear to represent the infectious fraction, virions bound to immunoglobulins, and free virions. In addition, non-enveloped nucleocapsids harboring HCV RNA have been described.92
NON-STRUCTURAL PROTEINS NS2-3 PROTEASE Cleavage of the polyprotein precursor at the NS2/NS3 junction is accomplished by a protease encoded by NS2 and the N-terminal one-third of NS3.93,94 NS2 is dispensable for replication of subgenomic replicons in vitro (see below) and is thus not essential for the formation of a functional replication complex. However, the NS2-3 protease activity is essential for the replication of full-length HCV genomes in vivo. Site-directed mutagenesis has shown that amino acids His 143 (i.e. His 952 of the HCV polyprotein), Glu 163 (i.e. Glu 972) and Cys 184 (i.e. Cys 993) are essential for catalytic activity.93,94 The importance of these amino acid residues is consistent with a thiol protease catalytic mechanism, but NS2-3 activity is stimulated by zinc or certain other divalent metal ions. Interestingly, the cellular chaperone Hsp90 was found to be essential for the activation of the NS2-3 protease.95 The membrane topology of NS2 has been controversial. It has been proposed that NS2 is a transmembrane protein with at least one and up to four transmembrane segments in the N-terminal domain.96,97 There are also data which suggest that the C terminus of NS2, after autocatalysis in the cytosol, may relocate to the ER lumen.97 Recombinant proteins lacking the N-terminal membrane domain of NS2 have been found to retain cleavage activity,
129
Section I. Pathophysiology of the Liver
allowing further characterization of this unique enzymatic activity.98,99 Indeed, recent efforts have led to the determination of a high-resolution structure for the NS2 protease domain, which reveals a dimeric structure with a thiol protease-like active site.100
NS3-4A COMPLEX A distinct serine protease located in the N-terminal one-third of NS3 is responsible for the downstream cleavage events in the nonstructural region9,101,102 (reviewed in 103). In addition, an RNA helicase/NTPase domain is found in the C-terminal two-thirds of NS3.104 The NS4A polypeptide functions as a cofactor for the NS3 serine protease and is incorporated as an integral component into the enzyme core. Complex formation occurs via a tight interaction of the 22 N-terminal residues of NS3 with 12 amino acid residues in the center of NS4A. NS3 by itself has no membrane anchor. The N-terminal domain of NS4A is strongly predicted to form a transmembrane a-helix responsible for membrane anchorage of the NS34A complex.105 The crystal structures of the serine protease106–108 and RNA helicase domains of NS3109,110 as well as the entire NS3 protein111 (Figure 8-4) have been elucidated. These enzymes are essential for viral replication and have emerged as prime targets for the design of specific inhibitors as antiviral agents112,113 (see below). The catalytic triad of the NS3 serine protease is formed by His 57 (i.e. His 1083 of the HCV polyprotein), Asp 81 (i.e. Asp 1107)
and Ser 139 (i.e. Ser 1165). Crucial determinants of substrate specificity include an acidic amino acid residue at the P6 position, a P1 cysteine (trans cleavage sites) or threonine (cis cleavage site between NS3 and NS4A), and an amino acid residue with a small side chain, i.e. alanine or serine, at the P1¢ position. A consensus cleavage sequence, therefore, would read D/E-X-X-X-X-C/T | S/AX-X-X. The three-dimensional structure of full-length NS3 revealed that a C-terminal b strand of the helicase domain lies within the active site of the serine protease domain, where it is expected to be located during the cis cleavage that separates NS3 from NS4A. This results in autoinhibition that is released upon trans substrate binding.111 Helicases catalyze the unwinding of doubled-stranded RNA or DNA into single-stranded nucleic acids. The energy required for this process is generated by hydrolysis of NTPs by an associated NTPase activity. Thus, the NS3 helicase couples unwinding of RNA regions with extensive secondary structures with NTP hydrolysis. The NS3 helicase is a member of the so-called helicase superfamily 2. These are also called DEXH/D helicases, according to a characteristic signature sequence in one of the essential enzyme motifs. It was recently shown that NS3 unwinds RNA through a highly coordinated cycle of fast ripping and local pausing that occurs with regular spacing along the duplex substrate, suggesting that nucleic acid motors can function in a manner analogous to cytoskeletal motor proteins.114
Figure 8-4. Three-dimensional structure of the HCV NS3-4A complex. The structure was determined using an engineered single-chain molecule consisting of NS3 and 14 NS4A residues known to activate the serine protease linked to the N terminus of NS3. The two bbarrels in the serine protease domain are shown in magenta and red, the helicase subdomains are shown in green, light blue and dark blue, and the central NS4A domain interacting with NS3 is shown in olive. Residues of the serine protease catalytic triad (His 57, Asp 81 and Ser 139) are shown in ball-and-stick representation, and the protease structural zinc ion is shown as a white sphere. The red spheres represent a phosphate molecule located at the NTP-binding site of the helicase. Note the interaction between the C terminus and the protease active site region. (Reproduced from111, with permission.)
130
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
NS4B NS4B, a 27 kDa integral membrane protein, is the least characterized HCV protein.115 It is predicted to be a polytopic membrane protein with a cytoplasmic N-terminal region followed, depending on the prediction, by four or six transmembrane segments and a Cterminal region in the cytosol.115,116 It has been shown experimentally that the bulk of the protein is cytosolically oriented.115 Introduction of glycosylation acceptor sites at various positions of NS4B recently confirmed the presence of a predicted ER luminal loop around amino acid position 161.117 Surprisingly, the N terminus of NS4B was found to be translocated into the ER lumen at least partially, presumably by a post-translational mechanism.117 The NS4B proteins of HCV, pesti- and flaviviruses are similar in size, amino acid composition, and hydrophobic properties. No function, however, has yet been ascribed to NS4B in any of these related viruses. More recently, it was found by electron microscopy that expression of HCV NS4B induces the formation of a seemingly ERderived specific cellular membrane alteration, designated the membranous web, that harbors the viral replication complex53,118 (see below). Thus, a function of NS4B may be to induce the specific membrane alteration that serves as a scaffold for the HCV replication complex.
NS5A NS5A is a phosphoprotein of unknown structure and function. It is found in a basally phosphorylated form of 56 kDa and in a hyperphosphorylated form of 58 kDa. NS5A of HCV and BVDV, as well as NS5 of YFV, are phosphorylated by as yet unidentified serine/threonine kinases, suggesting that these proteins share a common function related to their phosphorylation state.119 Basal phosphorylation requires domains in the center and C terminus of
1
Domain I
C57 C59 C39 C80
Membrane anchor domain
NS5A. The centrally located serine residues 225, 229 and 232 (i.e. Ser 2197, Ser 2201 and Ser 2204 of the HCV polyprotein) are important for NS5A hyperphosphorylation (Figure 8-5). However, it is unknown whether these serine residues are actually phosphorylated or whether they affect phosphorylation indirectly. The only phosphoacceptor sites that have been mapped experimentally are serine residues 222 (i.e. Ser 2194 of the polyprotein)120 and 349 (i.e. Ser 2321) (genotype 1a HCV H strain).121 The cellular kinase(s) responsible for NS5A phosphorylation appear to belong to the so-called CMCG group of serine/threonine kinases to which casein kinase II, cyclin-dependent kinases and mitogen-activated protein kinases belong. However, the cellular kinase(s) responsible for NS5A phosphorylation remains elusive. Interestingly, adaptive mutations have been found to cluster in the central region of NS5A in the context of selectable subgenomic HCV replicons, suggesting that NS5A is involved – either directly or by interaction with cellular proteins and pathways – in the viral replication process. This observation, together with the modulation of NS5A hyperphosphorylation by nonstructural proteins 3, 4A and 4B,122,123 strongly supports the notion of NS5A being an essential component of the HCV replication complex. An N-terminal amphipathic a-helix mediates membrane association of NS5A.124–126 This helix exhibits a hydrophobic, tryptophanrich side embedded in the cytosolic leaflet of the membrane bilayer, while the polar, charged side is exposed to the cytosol. Thus, NS5A is a monotopic protein with an in-plane amphipathic a-helix as membrane anchor. Structure–function analyses demonstrated that this helix displays fully conserved polar residues at the membrane surface, which define a unique platform probably involved in specific protein–protein interactions essential for the formation of a functional HCV replication complex.126 Comparative sequence analyses and limited proteolysis of recombinant NS5A protein have
213
250
Domain II
Δ 235-282 237
276 ISDR
237
356
Putative NLS
Adaptive changes
222
342
302
Domain III
447
GFP insertions
354-362 384 418 349 Δ 399-441 (1a) 384-407
V3
PKR interaction domain Figure 8-5. Overview of the HCV NS5A protein. NS5A is drawn to scale as a box. Amino acid positions relate to the HCV Con1 sequence (genotype 1b; GenBank accession number AJ238799; add 1972 amino acids to obtain positions relative to the HCV polyprotein). The N-terminal amphipathic a-helix that mediates membrane association of NS5A, the region where cell culture-adaptive changes have been found to cluster in the replicon system, the so-called interferon sensitivity determining region (ISDR), the double-strand RNA-activated protein kinase (PKR) interaction domain, the putative nuclear localization signal (NLS), and variable region 3 (V3) are highlighted. The domain organization recently proposed by Tellinghuisen et al.127 is also shown. Cysteine residues 39, 57, 59 and 80, denoted by blue lines, coordinate one zinc atom per NS5A protein. Deletions identified in the replicon system are shown in light green.13,178,163 GFP insertion sites tolerated in the replicon system are highlighted by green lines.180 Mapped phosphoacceptor sites for genotype 1b (amino acid position 222)120 and the genotype 1a HCV H isolate (amino acid position 349)121 are highlighted in red. Dashed red lines denote serine residues that affect NS5A phosphorylation.
131
Section I. Pathophysiology of the Liver
recently led to a proposed domain organization of NS5A (Figure 85).127 The relatively highly conserved domain I immediately following the membrane-anchoring a-helix has been shown to contain four absolutely conserved cysteine residues that coordinate one zinc atom per NS5A protein.127 Mutation of these residues abolishes HCV RNA replication, indicating that the zinc is essential for NS5A structure and/or function. Thus, NS5A is a zinc metalloprotein. The structure of the NS5A domain I has recently been solved, revealing a completely novel protein fold, a new zinc coordination motif, and a rare cytoplasmic disulfide bond. The structure also defines surface properties that may be involved in NS5A dimer formation and NS5A interaction with viral and cellular proteins, membranes and RNA128 (Figure 8-6). HCV NS5A has attracted considerable interest because of its potential role in modulating the IFN response. Studies performed in Japan first described a correlation between mutations within a discrete region of NS5A, termed interferon sensitivity-determining region (ISDR) (Figure 8-5), and a favorable response to IFN-a therapy.129 These studies demonstrated that strains closely matching the prototype HCV genotype 1b (HCV-J) ISDR sequence correlated with IFN resistance. These findings were largely confirmed in Japan, but not in Europe and North America. The reasons for this discrepancy are not understood, but may involve differences in both doses and regimens of IFN treatment and the low prevalence of ‘mutant type’ HCV genotype 1b isolates in western countries. Even if a meta-analysis of published data seems to confirm an association of specific ISDR sequences with the IFN response,130 this remains a controversial issue that has thus far not translated into clinically
applicable predictors. The same is true for other regions of NS5A that have been associated with the response to IFN therapy, such as a variable region in the C-terminal domain of NS5A termed V3 (Figure 8-5). Interestingly, however, an interaction with and repression of the catalytic activity of PKR by NS5A has been found by biochemical, transfection, and yeast functional analyses.131 Mutations within the ISDR that were observed in clinically IFN-sensitive genotype 1b strains disrupted the ability of NS5A to interact with and repress PKR activity, supporting the notion that NS5A mediates HCV resistance to IFN through down-regulation of PKR.132 However, these findings are controversial, and numerous additional potential functions have recently been attributed to NS5A (reviewed in 60,133,134). However, similar to the core protein, only very few of these postulated interactions and functional properties have been validated in a meaningful context involving active HCV RNA replication or HCV infection in vivo.
NS5B HCV replication proceeds via synthesis of a complementary negative-strand RNA using the genome as a template and the subsequent synthesis of genomic positive-strand RNA from this template. The key enzyme responsible for both of these steps is the NS5B RdRp. This essential viral enzyme has been extensively characterized at the biochemical135–138 and the structural level139–142 and has emerged as a major target for antiviral intervention. The HCV NS5B protein contains motifs shared by all RdRps, including the hallmark GDD sequence within motif C, and, based on the similarity of the enzyme structure with the shape of a right hand, possesses the classic fingers,
ER lumen
Membrane anchor
Membrane anchor
Cytoplasm
II III
II III
Figure 8-6. Structure of HCV NS5A domain I. The dimeric form of NS5A domain I (amino acids 36–198) modeled in relation to the membrane at the site of RNA replication.128 The cytoplasmic and luminal leaflets of the membrane are indicated. The two NS5A domain I monomers are colored in blue and green. Also shown in blue and green are the N-terminal amphipathic membrane anchors of NS5A lying flat in the plane of the membrane.126 Zinc atoms coordinated by domain I are shown as red spheres. The hypothetical locations of domains II and III of NS5A are indicated by schematic spheres. This orientation of domain I places the largely basic surface towards the phospholipid head groups of the membrane and positions the large ‘claw’ or groove of the NS5A dimer away from the membrane, where it may interact with RNA. This figure was reproduced from an illustration kindly provided by Dr Timothy L. Tellinghuisen, Rockefeller University, New York, USA.
132
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
UTP
C-ter tP
N-ter
palm and thumb subdomains. A special feature of the HCV RdRp is that extensive interactions between the fingers and thumb subdomains result in a completely encircled active site139–142 (Figure 87). This feature is shared by other RdRps, including those of the bacteriophage F6 and of BVDV.143,144 As with poliovirus RdRp,145–147 oligomerization of HCV NS5B has recently been reported to be important for cooperative RNA synthesis activity.148,149 Membrane association of the HCV RdRp is mediated by the Cterminal 21 aa residues, which are dispensable for polymerase activity in vitro. Membrane targeting occurs via a post-translational mechanism and results in integral membrane association of NS5B.150 These features, namely post-translational membrane targeting via a hydrophobic C-terminal insertion sequence; integral membrane association; and cytosolic orientation of the functional protein domain, define the HCV RdRp as a member of the so-called tailanchored proteins. The HCV RdRp insertion sequence crosses the membrane bilayer as a transmembrane segment,151 is essential for HCV RNA polymerase in cells, and is likely to possess additional functions apart from its membrane anchor function.152
Figure 8-7. Crystal structure of the catalytic domain of the HCV RNA-dependent RNA polymerase (140; PDB accession code 1GX6). Ribbon diagram of the NS5A-D55 protein complexed with UTP and Mn2+. a-Helices are colored blue, b-strands red, and connecting loops silver. The bound nucleotide and the side chains of the catalytic aspartic acids (Asp 220 and Asp 318) in the center of the structure are represented as ball-and-stick colored black and cyan, respectively. Mn2+ ions are shown as green spheres. Also labeled is the triphosphate (tP) moiety of a nucleotide bound to the ‘priming’ site. This figure was reproduced from an illustration kindly provided by Dr Felix A. Rey, Laboratoire de Virologie Moléculaire et Structurale, UMR 2472 CNRS/UMR 1157 INRA, Gif-sur-Yvette, France.
Table 8-2. In Vitro and in Vivo Models to Study HCV In vitro models In vitro transcription–translation Transient cellular expression systems Stably transfected cell lines (constitutive/inducible expression) Infection of primary hepatocytes and established cell lines Retroviral pseudoparticles displaying functional HCV glycoproteins Replicons (subgenomic/full-length; selectable/transient) Chimeric viruses (e.g. poliovirus—HCV) Related viruses (e.g. BVDV) In vivo models Transgenic mice Immunodeficient mice/hepatocellular reconstitution models Chimpanzee (Pan troglodytes) Tree shrew (Tupaia belangeri chinensis)? Related viruses (e.g. GBV-B in tamarins)
Given the lack of a robust cell culture system allowing natural infection, replication, and release of viral progeny, various in vitro and in vivo models have been used to study HCV (Table 8-2) (reviewed in 153).
results. In their present format some of these systems may be useful for neutralization assays, but not for a systematic investigation of the viral life cycle.154,155 An alternative approach involves the generation of cell lines constitutively or inducibly expressing viral sequences from chromosomally integrated cDNA.156 Moreover, viable chimeras of certain positive-strand RNA viruses, such as polio and Sindbis virus, with HCV genetic elements, such as the IRES157 or NS3,158 have been constructed and may facilitate the screening of selected antiviral compounds.
IN VITRO MODELS
The Replicon System
Infection of primary hepatocytes and established cell lines in vitro yielded only low-level replication and often poorly reproducible
The replicon system has revolutionized the investigation of HCV RNA replication.12 The prototype subgenomic replicon was a
MODEL SYSTEMS
133
Section I. Pathophysiology of the Liver
5¢ NCR
3¢ NCR
EMCV IRES
(U)n NS3
NeoR
A NS4 B
A
NS5
B
In vitro transcription
5¢ NCR
3¢ NCR
EMCV IRES NeoR
Figure 8-8. Prototype subgenomic HCV replicon.12 RNA is transcribed in vitro from a plasmid containing the HCV IRES followed by a neomycin resistance cassette, a second heterologous IRES from encephalomyocarditis virus (EMCV IRES), the HCV non-structural region (NS3 to NS5B), and the HCV 3’ NCR. RNA is subsequently transfected into HuH7 human hepatoma cells, followed by selection with G418 of clones harboring autonomously replicating subgenomic HCV RNA.
(U)n NS3
A NS4 B
A
NS5
B
Transfection
HuH-7 cells
Selection
Clones carrying HCV replicons
bicistronic RNA where the structural region of HCV was replaced by the neomycin phosphotransferase gene and translation of the non-structural proteins 3–5B was driven by a second, heterologous IRES from encephalomyocarditis virus (Figure 8-8). Using this approach it became possible, for the first time, to study efficient and genuine HCV RNA replication in HuH-7 human hepatoma cells in vitro. Interestingly, certain amino acid substitutions, i.e. cell cultureadaptive changes, can increase the efficiency of replication initiation by more than 105-fold.13,159 Adaptive changes cluster in certain regions, such as the center of NS5A (Figure 8-4), the C-terminal part of the NS3 serine protease and the N-terminal part of the NS3 RNA helicase domains, as well as two positions in NS4B.160 In NS5A these changes often affect serine residues required for hyperphosphorylation, suggesting that hyperphosphorylation of NS5A reduces HCV RNA replication.161–163 According to one model, hyperphosphorylation of NS5A reduces interaction with the human vesicleassociated membrane protein-associated protein A (hVAP-A).161 However, adaptive changes are likely to increase HCV RNA replication by additional, as yet unidentified mechanisms. In this context, it is interesting to note that there is an inverse correlation between mutations that permit efficient replication of HCV RNA in HuH-7 cells in vitro and productive replication in chimpanzees in vivo after intrahepatic inoculation.164 The replicon system has allowed genetic dissection of HCV RNA elements and proteins, provided material for biochemical and ultrastructural characterization of the viral replication complex, and facilitated drug discovery efforts (Table 8-3). Moreover, the replicon system has been exploited for analysis of the effect of cytokines on
134
Table 8-3. Lessons from the Replicon System Adaptive changes Cell-cycle dependence of HCV RNA replication Inverse correlation between replicon activity in vitro and HCV replication/particle formation in vivo Evaluation of new antiviral agents and antiviral resistance Effect of cytokines on HCV RNA replication Identification and characterization of the HCV replication complex Host factors Identification of essential RNA elements Requirements for infectious HCV particle formation
HCV RNA replication165,166 and the study of other aspects of the interaction between HCV and the host cell.167–169a Since the original reports of functional genotype 1b replicons, replicons for genotype 1a170 and 2a171, as well as derivatives expressing easily quantifiable marker enzymes in a separate cistron, have been made to facilitate genetic studies as well as drug screening and evaluation.172–174 In addition, full-length replicons and HCV genomes efficiently replicating in tissue culture have been developed,175–177 and the spectrum of permissive host cells has been expanded.178,179 Finally, replicons have been established that allow tracking of functional HCV replication complexes in living cells.180 One puzzling (and disappointing) observation was that full-length genome RNAs with adaptive mutations were incapable of producing infectious virus. This led investigators to favor the idea that
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
HuH-7 and other HCV permissive cell lines lacked some factor(s) necessary for particle formation and release. However, this does not appear to be the case, given recent results with a genotype 2a isolate from Japan, JFH-1. JFH-1 was isolated from a patient with acute fulminant hepatitis C, and JFH-1 subgenomic replicons can replicate efficiently in HuH-7 cells without adaptive mutations.171 Remarkably, full-length JFH-1 produces infectious virus.181,182 Chimeras consisting of the 5¢ and 3¢ NTRs and replicase region of JFH-1 (NS3-5B) and the C-NS2 region of other isolates will also sometimes produce cell culture infectious particles.183 One possibility, which seems to be growing in popularity, is that cell cultureadaptive mutations that promote efficient RNA replication may be deleterious in vivo because they compromise particle assembly and release. With these advances, the late (assembly and egress) and early (entry) events in HCV infection can now be studied. These steps can also be explored as possible new therapeutic targets.
IN VIVO MODELS The restricted host range of HCV has hampered the development of a suitable small animal model of viral replication and pathogenesis. Apart from a single report on the transmission of HCV to tree shrews (Tupaia belangeri chinensis),184 the chimpanzee (Pan troglodytes) is the only animal known to be susceptible to HCV infection.185 Indeed, the chimpanzee was essential in the early characterization of the agent of non-A, non-B hepatitis, and has allowed the determination of important aspects of HCV replication, pathogenesis and prevention. In this context, it would not have been possible to demonstrate the functionality of infectious clones of HCV without chimpanzees.11 In addition, recent studies in chimpanzees have provided new insight into the host immune response against hepatitis C,186–188 and the chimpanzee remains the only faithful model to test the immunogenicity and efficacy of vaccine candidates.189–191 However, ethical and financial restrictions limit the use of primates to highly selected experimental questions. Expression of HCV proteins in transgenic mice provided some insights into the pathogenesis of HCV-induced liver disease.55,56 However, expression of HCV proteins from chromosomally integrated cDNA does not appropriately reflect the viral life cycle, and studies on viral entry and replication are hardly conceivable in this system. GBV-B, the closest relative of HCV within the family Flaviviridae can be transmitted to tamarins (Saguinus sp.) and may represent a valuable surrogate model for HCV. Remarkably, GBV-B can be cultured in tamarin hepatocytes in vitro.192 In addition, infectious cDNA clones193 and replicons194 have been established for GBV-B. However, in tamarins GBV-B typically leads to self-limited infection without viral persistence unless animals are immunosuppressed or molecular clones are used.195 GBV-B can also be propagated in common marmosets (Callithrix jacchus), which are easier to breed in captivity, are smaller, and are already regularly used for drug metabolism, pharmacokinetic and toxicology studies.196 Progress in the development of a small animal model of HCV replication was achieved with the successful HCV infection of immunodeficient mice reconstituted with human hepatocytes.197 The properties of two different mouse strains, the Alb-uPA-transgenic and the immunodeficient SCID mouse, were combined to develop a model system that allows orthotopic engraftment of
human hepatocytes (Figure 8-9). Expression of the murine urokinase-type plasminogen activator under the transcriptional control of the albumin promoter (Alb-uPA) programs murine hepatocyte death, providing a suitable microenvironment for the engraftment and expansion of transplanted human hepatocytes. In homozygous animals, reconstitution with human hepatocytes was reported to reach >50% of the liver cell mass. In consequence, Alb-uPA homozygous animals were characterized by persistent high human albumin production. Inoculation with serum from patients with hepatitis C resulted in persistent HCV viremia in about 75% of mice with high-level human hepatocyte engraftment. HCV RNA could be detected by PCR for up to 35 weeks, with titers ranging from 3 ¥ 104 to 3 ¥ 106 copies/ml. These viral titers are similar to those found in infected humans. Moreover, an approximately 3-log rise in viral titers after inoculation, detection of viral negative-strand RNA in the liver, and the ability to serially passage the virus through several generations of animals provided convincing evidence for active replication and production of infectious viral progeny in this system. However, the handling of these fragile animals affected by major bleeding disorders and severe immunodeficiency (approximately 35% mortality in newborns) presents a non-trivial challenge and requires special expertise. Moreover, access to fresh human hepatocytes is limited for many investigators.
REPLICATION CYCLE The life cycle of HCV includes 1) binding to an as yet unidentified cell surface receptor and internalization into the host cell; 2) cytoplasmic release and uncoating of the viral RNA genome; 3) IRESmediated translation and polyprotein processing by cellular and viral proteases; 4) RNA replication; 5) packaging and assembly; and 6) virion maturation and release from the host cell (Figure 8-10).
RECEPTOR CANDIDATES CD81, a tetraspanin molecule found on the surface of many cell types, including hepatocytes,198 the low-density lipoprotein receptor (LDLR)199 and scavenger receptor class B type I (SR-BI)200 have, among others, been proposed as HCV receptors or components of a receptor complex. Both CD81 and SR-BI bind E2 and are currently viewed as necessary, but not sufficient, for HCV entry.14,15,201–203 Expression of CD81 in CD81-negative liver-derived cell lines confers susceptibility to HCV pseudoparticles (see below), and blocking antibodies against CD81 or SR-BI, recombinant CD81, or siRNA-mediated down-regulation of CD81 expression reduces infectivity. However, additional, as yet unidentified hepatocyte-specific factors are required for HCV entry. The LDLR has been attractive as a candidate receptor because infectious HCV has been reported to be associated with LDL or VLDL (see above). At present, however, it is unclear whether interaction of HCV with the LDLR can lead to productive infection. Cell lines without LDLR are still infectable with HCV pseudoparticles. HCV E2 also binds to DC-SIGN (dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin) and L-SIGN (liver/lymph node-specific intercellular adhesion molecule-3grabbing integrin). The latter is a calcium-dependent lectin expressed
135
Section I. Pathophysiology of the Liver
Figure 8-9. A small animal model of HCV replication.197
SCID
Alb-uPA
Immunodeficient Alb-uPA transgenic mouse
Newborns with liver cell destruction
Human hepatocytes
Partial repopulation with human hepatocytes
Mouse with ‘chimeric’ liver
(+) RNA 1
3
5
3 3 5A
ER
2
C
p7
E1
2
5B
E2 4A 4B (-) RNA 4
3 5
6 (+) RNA
5 5
136
MW 3
Figure 8-10. Life cycle of HCV. 1, Virus binding and internalization; 2, cytoplasmic release and uncoating; 3, IRESmediated translation and polyprotein processing; 4, RNA replication; 5, packaging and assembly; 6, virion maturation and release. The topology of HCV structural and nonstructural proteins at the endoplasmic reticulum (ER) membrane is shown schematically. HCV RNA replication occurs in a specific membrane alteration, the membranous web (MW). Note that IRES-mediated translation and polyprotein processing, as well as membranous web formation and RNA replication, illustrated here as separate steps for simplicity, may occur in a tightly coupled fashion.
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
on liver sinusoidal endothelial cells that may facilitate the infection process by trapping the virus for subsequent interaction with the receptor.204–207 Identification and validation of HCV receptor candidates has been limited by the paucity of systems for analysis of the early steps of the viral life cycle. Given the lack of native HCV particles and efficient cell culture systems, various alternatives have been explored to study the early steps of HCV infection. Soluble C-terminal truncated versions of HCV envelope glycoprotein E2,198,200,205,208,209 liposomes reconstituted with HCV E1 and E2,210 and virus-like particles expressed in insect cells211,212 have been used to study HCV glycoprotein interactions with the cell surface. The production of viruslike particles has also been described in mammalian cells.213 However, it is unclear how virus-like particles produced in inset or mammalian cells will compare to authentic HCV virions. Pseudotyped vesicular stomatitis virus (VSV) or influenza virus particles have been reported incorporating chimeric E1 and/or E2 glycoproteins whose C-terminal transmembrane domains were modified to allow transport to the cell surface.208,214–216 However, such modifications may interfere with the multiple and complex roles of the E1 and E2 transmembrane domains69 and may perturb the conformation and functions of E1–E2 complexes. Therefore, the use of such pseudotypes as a tool to study HCV assembly and entry remains controversial.216 Against this background, the recent establishment of infectious retroviral pseudotypes displaying functional HCV glycoproteins as a robust model system for the study of viral entry represents a major breakthrough14,15 (Figure 8-11). HCV pseudoparticle infectivity is restricted primarily to human hepatocytes and hepatocyte-derived cell lines, and entry is pH dependent. Thus, HCV entry likely involves transit through an endosomal low-pH compartment and fusion with the endosomal membrane. The structural basis for low pH-induced membrane fusion has recently been elucidated for the dengue, TBE and Semliki Forest viruses.217–219 The envelope proteins of these related flavi- and alphaviruses possess an internal fusion peptide that is exposed during low pH-mediated domain rearrangement and trimerization of the protein. The scaffolds of these so-called class II fusion proteins are remarkably similar, suggesting that all members of the Flaviviridae, including HCV, could behave similarly.
CMV
MLV Gag-Pol Plasmid 2
CMV
E1
ψ
E2
Plasmid 1
CMV
GFP
Plasmid 3
293T cells
ψ
CMV
E1
GFP
E2
HCV pseudoparticle Figure 8-11. Generation of infectious HCV pseudoparticles. Cotransfection of 293T human embryo kidney cells with plasmids allowing expression of (1) unmodified HCV E1-E2 glycoproteins, (2) retroviral core proteins, and (3) a packaging-competent green fluorescent protein (GFP) expression construct leads to secretion into the supernatant of pseudoparticles bearing HCV envelope glycoproteins instead of the retroviral envelope protein on their surface. CMV, cytomegalovirus promoter; y, retroviral packaging sequence.
REPLICATION COMPLEX The formation of a membrane-associated replication complex, composed of viral proteins, replicating RNA, and altered cellular membranes, is a hallmark of all positive-strand RNA viruses investigated thus far (see 220,221 for reviews). Depending on the virus, replication may occur on altered membranes derived from the ER,222–226 Golgi apparatus,227–229 mitochondria230 or even lysosomes.231 The role of membranes in viral RNA synthesis is not well understood. It may include (i) the physical support and organization of the RNA replication complex;147 (ii) the compartmentalization and local concentration of viral products;232 (iii) tethering of the viral RNA during unwinding;220 (iv) provision of lipid constituents important for replication;233,234 and (v) protection of the viral RNA from double-strand RNA-mediated host defenses or RNA interference. In the case of HCV, protein–protein interactions among HCV non-structural proteins have been described235,236 and determinants
for membrane association of the HCV proteins have been mapped. The membrane association of HCV proteins is schematically illustrated in Figure 8-12. For a more comprehensive review on the interactions of HCV proteins, including the structural proteins, with host cell membranes, see references.237,238 A specific membrane alteration, designated the membranous web, was recently identified as the site of RNA replication in HuH-7 cells harboring subgenomic HCV replicons118 (Figure 8-13). Formation of the membranous web could be induced by NS4B alone (see above), and it was very similar to the ‘sponge-like inclusions’ previously found by electron microscopy in the liver of HCV-infected chimpanzees. The membranous web was often found closely associated with the rough ER. Based on this observation, together with earlier studies demonstrating the co-localization of individually expressed HCV proteins with membranes of the ER,105,115,124,150 and data
137
Section I. Pathophysiology of the Liver
Figure 8-12. Membrane association of HCV proteins. Note that the topologies of NS2, NS4A and NS4B are currently under investigation and are illustrated only schematically. A recent study indicated that the C terminus of NS2 may be localized in the ER lumen, resulting in four transmembrane domains.97 Also, it was recently reported that the N terminus of NS4B can be at least partially translocated into the ER lumen.117
Cytosol NS5B NS3 p7
C
NS2
CC
NS5A C C
N
N
NS4B C
N N
N
C
NS4A
E1 E2
ER lumen
ER N
ER
N
M M A
B
Figure 8-13. HCV replication complex.118 (A) Low-power overview of a HuH-7 cell harboring a subgenomic HCV replicon. A distinct membrane alteration, named membranous web (arrows), is found in the juxtanuclear region. Note the circumscript nature of this specific membrane alteration and the otherwise unaltered cellular organelles. Bar, 1 mm. (B) Higher magnification of a membranous web (arrows) composed of small vesicles embedded in a membrane matrix. Note the close association of the membranous web with the rough endoplasmic reticulum. Bar, 500 nm. The membranous web harbors all HCV non-structural proteins and nascent viral RNA in HuH-7 cells harboring subgenomic replicons, and therefore represents the HCV RNA replication complex. N, nucleus; ER, endoplasmic reticulum; M, mitochondria.
indicating that HCV RNA replication takes place in a compartment that sustains endoglycosidase H-sensitive glycosylation,151 it is currently believed that the membranous web is derived from membranes of the ER. Ongoing studies are aimed at isolating and further characterizing this complex and at defining the viral and cellular processes involved in the formation of the membranous web. Recent studies demonstrate a complex interaction between HCV RNA replication and the cellular lipid metabolism, presumably via the trafficking and association of viral and host proteins with intracellular membranes. In this context, it was found, for example, that geranylgeranylation of one or more host proteins is required for HCV RNA replication.239,240 Such observations suggest that pharmacologic manipulation of lipid metabolism may have therapeutic potential in hepatitis C.
138
EVOLVING THERAPEUTIC STRATEGIES In principle, each of the steps of the HCV life cycle illustrated in Figure 8-12 represents a target for antiviral intervention.241–243 Specific inhibitors of the biochemically and structurally wellcharacterized NS3 serine protease, as well as the RNA helicase/ NTPase and the NS5B RdRp, are currently being developed as antiviral agents, and the first candidates are already in early-phase clinical trials.112,113 Already at this early stage it becomes evident that the genetic variability of HCV represents a major challenge to the clinical development of specific enzyme inhibitors and that, similar
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
to HIV infection, combination therapy will be necessary for therapeutic success.244,245 In addition to these more classic pharmacological approaches, gene therapeutic strategies aimed at inhibiting HCV replication and gene expression are currently being explored in various experimental systems. These include, among others, antisense oligodeoxynucleotides, ribozymes, and small interfering RNAs. Moreover, based on the concept that a quantitatively and qualitatively insufficient CD4+ and CD8+ T-cell response may contribute to viral persistence, immunotherapeutic strategies aimed at enhancing the cellular immune response against HCV are currently being investigated. Apart from more efficient therapeutic strategies, the development and implementation of preventive measures is of paramount importance. The development of an effective recombinant vaccine has been hampered by the high genetic variability of HCV and the lack of a suitable cell culture infection system and small animal model.189–191 It has been shown, however, that vaccination with recombinant envelope proteins expressed in mammalian cells can protect chimpanzees from primary infection with a homologous virus isolate.246 The correlates of and requirements for a broader protection and a potentially neutralizing immune response still need to be defined, however. The potential of DNA vaccination to induce a humoral and cellular immune response is particularly interesting in this regard. Alternative currently pursued strategies include peptide and protein vaccines, dendritic cell-based vaccines, and virus-like particles. Although sterilizing immunity will probably be difficult to achieve, the aim of inducing a state of immunity that prevents the development of chronic infection appears more realistic. Along these lines, and in contrast to earlier more pessimistic views citing lack of protective immunity,247–249 more recent observations indicate that chimpanzees that clear infection do exhibit protective – albeit not sterilizing – immunity upon rechallenge,250,251 i.e. they show an attenuated course with rapid control of the rechallenge inoculum. In addition, studies in intravenous drug users have shown that there is some protective immunity in hepatitis C.252 This study showed that the risk of developing HCV viremia was lower for intravenous drug users who had successfully cleared a previous HCV infection than for those who had no evidence of previous HCV infection. In any case, it is likely that induction of both a humoral and a cellular immune response will be required for an effective HCV vaccine. Such a vaccine might also be useful therapeutically.253
PATHOGENESIS HCV infection is a highly dynamic process with a viral half-life of only a few hours and an average daily virion production and clearance of up to more than 1012.254 This high replicative activity, together with the lack of a proofreading function of the viral RdRp, provides the basis for the genetic variability of HCV. In addition, these findings are similar to the dynamics of HIV infection and provide, as discussed above, a rationale for the development and implementation of combination antiviral therapies. The mechanisms responsible for liver injury in acute and chronic hepatitis C are poorly understood.255–257 In acute HCV infection, liver cell damage coincides with the development of the host immune response and not with infection and viral replication. In
addition, persistent viral replication often occurs without evidence of liver cell damage, suggesting that HCV is not directly cytopathic. The immune response against HCV therefore plays a central role in the HCV pathogenesis of hepatitis C. HCV-specific major histocompatibility complex (MHC) class IIrestricted CD4+ helper T-cell258,259 and MHC class I-restricted CD8+ cytotoxic T-lymphocyte (CTL) responses260,261 have been identified in patients with acute and chronic HCV infection. CTL-mediated lysis of virus-infected host cells may lead to clearance of the virus or, if incomplete, to viral persistence and eventually chronic hepatitis. Based on these observations and parallels in other viral diseases, viral persistence and immunologically mediated liver cell injury are important mechanisms leading to chronic hepatitis C.255 Patients who clear HCV infection have a more vigorous CD4+258,259 and CD8+ T-cell response early on.261 The role of specific CD4+ and CD8+ T-cell responses in control of HCV infection was elegantly illustrated by in vivo depletion studies in chimpanzees.187,188 Despite the presence of an immune response, however, HCV is rarely eliminated. Thus, HCV may overwhelm, not induce, or evade antiviral immune responses. Perhaps the simplest explanation is quantitative, based on the kinetics of infection relative to the induction of a CTL response during the early phase of infection. According to this model, viral persistence would be predicted if the replication rate of the virus exceeded the kinetics of the immune response. Indeed, HCV reaches high serum titers within 1 week of infection, whereas adaptive cellular immune responses are delayed by at least 1 and humoral immune responses by at least 2 months.262–264 The rate of increase of the viral titer slows only several weeks after infection, and HCV RNA titers decline after approximately 8–12 weeks, when the serum ALT levels peak.264 Studies performed in chimpanzees showed that the appearance of adaptive cellular immune responses and the induction of type II interferon, i.e. interferon-g, coincides with the decrease in HCV RNA titers.186,264–265 Interferon-g may have a direct antiviral effect, as it efficiently inhibits the replication of HCV replicons.166 However, the effector functions of HCV-specific T cells appear to be reduced.266 Most patients develop chronic infection with relatively stable viral titers, about 2–3 logs lower than in the acute phase. Only a small proportion of patients recover and test negative for HCV RNA using standard assays. Whether HCV is completely cleared after recovery, or whether trace amounts of virus persist, similar to hepatitis B virus, is debated.267 HCV-specific antibodies may disappear completely 10–20 years after recovery.268 Microarray analyses of serial liver biopsy samples in experimentally infected chimpanzees revealed that HCV induces the intrahepatic expression of many genes, including type I interferon, i.e. interferon-a and -b, responses.265 However, even if HCV RNA replication in vitro is efficiently inhibited by type I interferons,166 HCV seems to be resistant to these responses and frequently succeeds in establishing chronic hepatitis. As discussed above, HCV may have evolved numerous mechanisms to counteract the innate immune response, including interference with the interferon system at the induction,167–169a signaling269–271 and effector levels.131,132,272 In addition, HCV may interfere with natural killer (NK) cell functions.273,274 In this context, a recent large immunogenetic study revealed an association between a NK cell receptor (KIR2DL3
139
Section I. Pathophysiology of the Liver
allele)-HLA compound genotype and HCV clearance and clinical recovery, pointing toward a role of NK cells in early HCV infection.275 However, to be consistent with the repeated observation that the CTL response is less vigorous in chronically infected patients than it is during acute, self-limited infection, additional mechanisms must be involved. These may include the induction of peripheral tolerance or exhaustion of the T-cell response, infection of immunologically privileged sites, inhibition of antigen presentation, downregulation of viral gene expression, and viral mutations that abrogate, anergize or antagonize antigen recognition by virus-specific T cells.255 An impairment of dendritic cell function has been proposed,276 but this is controversial.277 There is some evidence that privileged sites may play a role, as HCV may infect extrahepatic cells and tissues. As mentioned above, the role of viral escape mutations and the quasispecies nature of HCV as a cause of viral persistence has attracted considerable interest. In this context, HCV escape to antibodies278,279 and T cells280–283 has been demonstrated both in humans and chimpanzees. The role of the humoral immune response in the natural course and pathogenesis of hepatitis C is not well understood. Recent studies using HCV pseudoparticles (see above) validated earlier studies demonstrating neutralizing of antibodies.284,285 However, the highest antibody titers are found in patients with chronic hepatitis C, and the role of antibodies capable of neutralizing a minor fraction of the HCV population is unknown.
5.
6.
7.
8.
9.
10.
11.
12.
13. 14.
15.
CONCLUSIONS AND PERSPECTIVES The development of powerful model systems has allowed us to systematically dissect important steps of the HCV life cycle. These efforts have translated into the identification of novel antiviral targets and the development of new therapeutic strategies, some of which are already in early-phase clinical evaluation. Much work remains to be done with respect to virion structure, the early and late steps of the HCV life cycle, the mechanism and regulation of RNA replication, and the pathogenesis of HCV-induced liver disease. Ultimately, a detailed understanding of the viral life cycle should result in innovative therapeutic and preventive strategies for one of the most common causes of chronic hepatitis, liver cirrhosis and HCC worldwide.
16.
17. 18.
19. 20.
21. 22.
REFERENCES 1. National Institutes of Health Consensus Development Conference Statement: Management of Hepatitis C: 2002. Hepatology 2002;36:S2–S20. 2. Manns MP, McHutchison JG, Gordon SC, et al. and the International Hepatitis Interventional Therapy Group. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 2001;358:958–965. 3. Fried M, Shiffman M, Reddy K, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 2002;347:975–982. 4. Hadziyannis SJ, Sette H Jr, Morgan TR, et al. Peginterferonalpha2a and ribavirin combination therapy in chronic hepatitis
140
23.
24.
25.
26.
C: a randomized study of treatment duration and ribavirin dose. Ann Intern Med 2004;140:346–355. Muir AJ, Bornstein JD, Killenberg PG. Peginterferon alfa-2b and ribavirin for the treatment of chronic hepatitis C in blacks and non-Hispanic whites. N Engl J Med 2004;350:2265–2271. Falck-Ytter Y, Kale H, Mullen KD, et al. Surprisingly small effect of antiviral treatment in patients with hepatitis C. Ann Intern Med 2002;136:288–292. Davis GL, Albright JE, Cook SF, Rosenberg DM. Projecting future complications of chronic hepatitis C in the United States. Liver Transpl 2003;9:331–338. Choo Q-L, Kuo G, Weiner AJ, et al. Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome. Science 1989;244:359–362. Grakoui A, McCourt DW, Wychowski C, et al. Characterization of the hepatitis C virus-encoded serine proteinase: determination of proteinase-dependent polyprotein cleavage sites. J Virol 1993;67:2832–2843. Bartenschlager R, Ahlborn-Laake L, Yasargil K, Mous J, Jacobsen H. Kinetic and structural analyses of hepatitis C virus polyprotein processing. J Virol 1994;68:5045–5055. Kolykhalov AA, Agapov EV, Blight KJ, et al. Transmission of hepatitis C by intrahepatic inoculation with transcribed RNA. Science 1997;277:570–574. Lohmann V, Körner F, Koch J-O, et al. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 1999;285:110–113. Blight KJ, Kolykhalov AA, Rice CM. Efficient initiation of HCV RNA replication in cell culture. Science 2000;290:1972–1974. Bartosch B, Dubuisson J, Cosset FL. Infectious hepatitis C virus pseudo-particles containing functional E1–E2 envelope protein complexes. J Exp Med 2003;197:633–642. Hsu M, Zhang J, Flint M, et al. Hepatitis C virus glycoproteins mediate pH-dependent cell entry of pseudotyped retroviral particles. Proc Natl Acad Sci USA 2003;100:7271–7276. Bartenschlager R, Frese M, Pietschmann T. Novel insights into hepatitis C virus replication and persistence. Adv Virus Res 2004;63:71–180. Penin F, Dubuisson J, Rey FA, et al. Structural biology of hepatitis C virus. Hepatology 2004;39:5–19. Lindenbach BD, Rice CM. Unravelling hepatitis C virus replication from genome to function. Nature 2005;436:933–938. Seeger C. Salient molecular features of hepatitis C virus revealed. Trends Microbiol 2005;13:528–534. van Regenmortel MHV, Fauquet CM, Bishop DHL, et al. (eds) Virus taxonomy. The VIIth Report of the International Committee on Taxonomy of Viruses. San Diego: Academic Press, 2000. Simmonds P. Genetic diversity and evolution of hepatitis C virus – 15 years on. J Gen Virol 2004;85:3173–3188. Yanagi M, Purcell RH, Emerson SU, Bukh J. Transcripts from a single full-length cDNA clone of hepatitis C virus are infectious when directly transfected into the liver of a chimpanzee. Proc Natl Acad Sci USA 1997;94:8738–8743. Hong Z, Beaudet-Miller M, Lanford RE, et al. Generation of transmissible hepatitis C virions from a molecular clone in chimpanzees. Virology 1999;256:36–44. Lanford RE, Lee H, Chavez D, et al. Infectious cDNA clone of the hepatitis C virus genotype 1 prototype sequence. J Gen Virol 2001;82:1291–1297. Beard MR, Abell G, Honda M, et al. An infectious molecular clone of a Japanese genotype 1b hepatitis C virus. Hepatology 1999;30:316–324. Yanagi M, Purcell RH, Emerson SU, Bukh J. Hepatitis C virus: An infectious molecular clone of a second major genotype (2a) and lack of viability of intertypic 1a and 2a chimeras. Virology 1999;262:250–263.
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
27. Kolykhalov AA, Mihalik K, Feinstone SM, Rice CM. Hepatitis C virus-encoded enzymatic activities and conserved RNA elements in the 3¢ nontranslated region are essential for virus replication in vivo. J Virol 2000;74:2046–2051. 28. Sakai A, Claire MS, Faulk K, et al. The p7 polypeptide of hepatitis C virus is critical for infectivity and contains functionally important genotype-specific sequences. Proc Natl Acad Sci USA 2003;100:11646–11651. 29. Rijnbrand RC, Lemon SM. Internal ribosome entry sitemediated translation in hepatitis C virus replication. Curr Topics Microbiol Immunol 2000;242:85–116. 30. Sarnow P. Viral internal ribosome entry site elements: novel ribosome–RNA complexes and roles in viral pathogenesis. J Virol 2003;77:2801–2806. 31. Friebe P, Lohmann V, Krieger N, Bartenschlager R. Sequences in the 5¢ nontranslated region of hepatitis C virus required for RNA replication. J Virol 2001;75:12047–12057. 32. Spahn CM, Kieft JS, Grassucci RA, et al. Hepatitis C virus IRES RNA-induced changes in the conformation of the 40S ribosomal subunit. Science 2001;291:1959–1962. 33. Lukavsky PJ, Otto GA, Lancaster AM, et al. Structures of two RNA domains essential for hepatitis C virus internal ribosome entry site function. Nature Struct Biol 2000;7:1105–1110. 34. Collier AJ, Gallego J, Klinck R, et al. A conserved RNA structure within the HCV IRES eIF3-binding site. Nature Struct Biol 2002;9:375–380. 35. Kieft JS, Zhou K, Grech A, et al. Crystal structure of an RNA tertiary domain essential to HCV IRES-mediated translation initiation. Nature Struct Biol 2002;9:370–374. 36. Lukavsky PJ, Kim I, Otto GA, Puglisi JD. Structure of HCV IRES domain II determined by NMR. Nature Struct Biol 2003;10:1033–1038. 37. Otto GA, Puglisi JD. The pathway of HCV IRES-mediated translation initiation. Cell 2004;119:369–380. 38. Kolykhalov AA, Feinstone SM, Rice CM. Identification of a highly conserved sequence element at the 3¢ terminus of hepatitis C virus genome RNA. J Virol 1996;70:3363–3371. 39. Tanaka T, Kato N, Cho M-J, et al. Structure of the 3¢ terminus of the hepatitis C virus genome. J Virol 1996;70: 3307–3312. 40. Yanagi M, St Claire M, Emerson SU, et al. In vivo analysis of the 3¢ untranslated region of the hepatitis C virus after in vitro mutagenesis of an infectious cDNA clone. Proc Natl Acad Sci USA 1999;96:2291–2295. 41. Friebe P, Bartenschlager R. Genetic analysis of sequences in the 3¢ nontranslated region of hepatitis C virus that are important for RNA replication. J Virol 2002;76:5326–5338. 42. Yi M, Lemon SM. 3¢ nontranslated RNA signals required for replication of hepatitis C virus RNA. J Virol 2003;77:3557–3568. 43. Yi M, Lemon SM. Structure–function analysis of the 3¢ stemloop of hepatitis C virus genomic RNA and its role in viral RNA replication. RNA 2003;9:331–345. 44. Kim YK, Kim CS, Lee SH, Jang SK. Domains I and II in the 5¢ nontranslated region of the HCV genome are required for RNA replication. Biochem Biophys Res Commun 2002;290:105–112. 45. Smith DB, Simmonds P. Characteristics of nucleotide substitution in the hepatitis C virus genome: constraints on sequence change in coding regions at both ends of the genome. J Mol Evol 1997;45:238–246. 46. Walewski JL, Keller TR, Stump DD, Branch AD. Evidence for a new hepatitis C virus antigen encoded in an overlapping reading frame. RNA 2001;7:710–721. 47. You S, Stump DD, Branch AD, Rice CM. A cis-acting replication element in the sequence encoding the NS5B RNAdependent RNA polymerase is required for hepatitis C virus RNA replication. J Virol 2004;78:1352–1366.
48. Friebe P, Boudet J, Simorre J-P, Bartenschlager R. A kissing loop interaction in the 3¢ end of the hepatitis C virus genome essential for RNA replication. J Virol 2005;79: 380–392. 49. Pietschmann T, Lohmann V, Rutter G, et al. Characterization of cell lines carrying self-replicating hepatitis C virus RNAs. J Virol 2001;75:1252–1264. 50. McLauchlan J, Lemberg MK, Hope G, Martoglio B. Intramembrane proteolysis promotes trafficking of hepatitis C virus core protein to lipid droplets. EMBO J 2002;21:3980–3988. 51. Moradpour D, Englert C, Wakita T, Wands JR. Characterization of cell lines allowing tightly regulated expression of hepatitis C virus core protein. Virology 1996;222:51–63. 52. Barba G, Harper F, Harada T, et al. Hepatitis C virus core protein shows a cytoplasmic localization and associates to cellular lipid storage droplets. Proc Natl Acad Sci USA 1997;94:1200–1205. 53. Egger D, Wölk B, Gosert R, et al. Expression of hepatitis C virus proteins induces distinct membrane alterations including a candidate viral replication complex. J Virol 2002;76:5974–5984. 54. Hope RG, Murphy DJ, McLauchlan J. The domains required to direct core proteins of hepatitis C virus and GB virus-B to lipid droplets share common features with plant oleosin proteins. J Biol Chem 2002;277:4261–4270. 55. Moriya K, Fujie H, Shintani Y, et al. The core protein of hepatitis C virus induces hepatocellular carcinoma in transgenic mice. Nature Med 1998;4:1065–1067. 56. Lerat H, Honda M, Beard MR, et al. Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus. Gastroenterology 2002;122:352–365. 57. Kunkel M, Lorinczi M, Rijnbrand R, et al. Self-assembly of nucleocapsid-like particles from recombinant hepatitis C virus core protein. J Virol 2001;75:2119–2129. 58. Klein KC, Polyak SJ, Lingappa JR. Unique features of hepatitis C virus capsid formation revealed by de novo cell-free assembly. J Virol 2004;78:9257–9269. 59. McLauchlan J. Properties of the hepatitis C virus core protein: a structural protein that modulates cellular processes. J Viral Hepat 2000;7:2–14. 60. Tellinghuisen TL, Rice CM. Interaction between hepatitis C virus proteins and host cell factors. Curr Opin Microbiol 2002;5:419–427. 61. Xu Z, Choi J, Yen TS, et al. Synthesis of a novel hepatitis C virus protein by ribosomal frameshift. EMBO J 2001;20:3840–3848. 62. Varaklioti A, Vassilaki N, Georgopoulou U, Mavromara P. Alternate translation occurs within the core coding region of the hepatitis C viral genome. J Biol Chem 2002;277: 17713–17721. 63. Choi J, Xu Z, Ou JH. Triple decoding of hepatitis C virus RNA by programmed translational frameshifting. Mol Cell Biol 2003;23:1489–1497. 64. Boulant S, Becchi M, Penin F, Lavergne JP. Unusual multiple recoding events leading to alternative forms of hepatitis C virus core protein from genotype 1b. J Biol Chem 2003;278:45785–45792. 65. Komurian-Pradel F, Rajoharison A, Berland JL, et al. Antigenic relevance of F protein in chronic hepatitis C virus infection. Hepatology 2004;40:900–909. 66. Bain C, Parroche P, Lavergne JP, et al. Memory T-cell-mediated immune responses specific to an alternative core protein in hepatitis C virus infection. J Virol 2004;78:10460–10469. 67. McMullan LK, Grakoui A, Puig M, et al. Evidence for a functional RNA element in the hepatitis C virus alternative reading frame. In preparation.
141
Section I. Pathophysiology of the Liver
68. Branch AD, Stump DD, Gutierrez JA, et al. The hepatitis C virus alternate reading frame (ARF) and its family of novel products: the alternate reading frame protein/F-protein, the double-frameshift protein, and others. Semin Liver Dis 2005;25:105–117. 69. Op De Beeck A, Cocquerel L, Dubuisson J. Biogenesis of hepatitis C virus envelope glycoproteins. J Gen Virol 2001;82:2589–2595. 70. Cocquerel L, Op de Beeck A, Lambot M, et al. Topologic changes in the transmembrane domains of hepatitis C virus envelope glycoproteins. EMBO J 2002;21:2893–2902. 71. Rey FA, Heinz FX, Mandl C, et al. The envelope glycoprotein from tick-borne encephalitis virus at 2 Å resolution. Nature 1995;375:291–298. 72. Yagnik AT, Lahm A, Meola A, et al. A model for the hepatitis C virus envelope glycoprotein E2. Proteins 2000;40:355–366. 73. Lescar J, Roussel A, Wien MW, et al. The fusion glycoprotein shell of Semliki Forest virus: an icosahedral assembly primed for fusogenic activation at endosomal pH. Cell 2001;105:137–148. 74. Modis Y, Ogata S, Clements D, Harrison SC. A ligand-binding pocket in the dengue virus envelope glycoprotein. Proc Natl Acad Sci USA 2003;100:6986–6991. 75. Hijikata M, Kato N, Ootsuyama Y, et al. Hypervariable regions in the putative glycoprotein of hepatitis C virus. Biochem Biophys Res Commun 1991;175:220–228. 76. Weiner AJ, Brauer MJ, Rosenblatt J, et al. Variable and hypervariable domains are found in the regions of HCV corresponding to the flavivirus envelope and NS1 proteins and the pestivirus envelope glycoproteins. Virology 1991;180:842–848. 77. Penin F, Combet C, Germanidis G, et al. Conservation of the conformation and positive charges of hepatitis C virus E2 envelope glycoprotein hypervariable region 1 points to a role in cell attachment. J Virol 2001;75:5703–5710. 78. Farci P, Shimoda A, Wong D, et al. Prevention of hepatitis C virus infection in chimpanzees by hyperimmune serum against the hypervariable region 1 of the envelope 2 protein. Proc Natl Acad Sci USA 1996;93:15394–15399. 79. Forns X, Thimme R, Govindarajan S, et al. Hepatitis C virus lacking the hypervariable region 1 of the second envelope protein is infectious and causes acute resolving or persistent infection in chimpanzees. Proc Natl Acad Sci USA 2000;97:13318–13323. 80. Carrère-Kremer S, Montpellier-Pala C, Cocquerel L, et al. Subcellular localization and topology of the p7 polypeptide of hepatitis C virus. J Virol 2002;76:3720–3730. 81. Harada T, Tautz N, Thiel HJ. E2-p7 region of the bovine viral diarrhea virus polyprotein: processing and functional studies. J Virol 2000;74:9498–9506. 82. Griffin SD, Beales LP, Clarke DS, et al. The p7 protein of hepatitis C virus forms an ion channel that is blocked by the antiviral drug amantadine. FEBS Lett 2003;535:34–38. 83. Pavlovic D, Neville DC, Argaud O, et al. The hepatitis C virus p7 protein forms an ion channel that is inhibited by long-alkylchain iminosugar derivatives. Proc Natl Acad Sci USA 2003;100:6104–6108. 84. Ferlenghi I, Clarke M, Ruttan T, et al. Molecular organization of a recombinant subviral particle from tick-borne encephalitis virus. Mol Cell 2001;7:593–602. 85. Kuhn RJ, Zhang W, Rossmann MG, et al. Structure of dengue virus: implications for flavivirus organization, maturation, and fusion. Cell 2002;108:717–725. 86. Mukhopadhyay S, Kim BS, Chipman PR, et al. Structure of West Nile virus. Science 2003;302:248. 87. Feinstone SM, Mihalik KB, Kamimura T, et al. Inactivation of hepatitis B virus and non-A, non-B hepatitis by chloroform. Infect Immun 1983;41:816–821.
142
88. He LF, Alling D, Popkin T, et al. Determining the size of nonA, non-B hepatitis virus by filtration. J Infect Dis 1987;156:636–640. 89. Kaito M, Watanabe S, Tsukiyama-Kohara K, et al. Hepatitis C virus particle detected by immunoelectron microscopy study. J Gen Virol 1994;75:1755–1760. 90. Shimizu YK, Feinstone SM, Kohara M, et al. Hepatitis C virus: detection of intracellular virus particles by electron microscopy. Hepatology 1996;23:205–209. 91. André P, Komurian-Pradel F, Deforges S, et al. Characterization of low- and very-low-density hepatitis C virus RNA-containing particles. J Virol 2002;76:6919–6928. 92. Maillard P, Krawczynski K, Nitkiewicz J, et al. Nonenveloped nucleocapsids of hepatitis C virus in the serum of infected patients. J Virol 2001;75:8240–8250. 93. Grakoui A, McCourt DW, Wychowski C, et al. A second hepatitis C virus-encoded proteinase. Proc Natl Acad Sci USA 1993;90:10583–10587. 94. Hijikata M, Mizushima H, Akagi T, et al. Two distinct proteinase activities required for the processing of a putative nonstructural precursor protein of hepatitis C virus. J Virol 1993;67: 4665–4675. 95. Waxman L, Whitney M, Pollok BA, et al. Host cell factor requirement for hepatitis C virus enzyme maturation. Proc Natl Acad Sci USA 2001;98:13931–13935. 96. Santolini E, Pacini L, Fipaldini C, et al. The NS2 protein of hepatitis C virus is a transmembrane polypeptide. J Virol 1995;69:7461–7471. 97. Yamaga AK, Ou JH. Membrane topology of the hepatitis C virus NS2 protein. J Biol Chem 2002;277:33228–33234. 98. Pallaoro M, Lahm A, Biasiol G, et al. Characterization of the hepatitis C virus NS2/3 processing reaction by using a purified precursor protein. J Virol 2001;75:9939–9946. 99. Thibeault D, Maurice R, Pilote L, et al. In vitro characterization of a purified NS2/3 protease variant of hepatitis C virus. J Biol Chem 2001;276:46678–46684. 100. Lorenz I, Marcotrigiano J, Rice CM. Structure of the catalytic domain of the hepatitis C virus NS2–3 protease. In preparation. 101. Bartenschlager R, Ahlborn-Laake L, Mous J, Jacobsen H. Nonstructural protein 3 of the hepatitis C virus encodes a serine-type proteinase required for cleavage at the NS3/4 and NS4/5 junctions. J Virol 1993;67:3835–3844. 102. Tomei L, Failla C, Santolini E, et al. NS3 is a serine protease required for processing of hepatitis C virus polyprotein. J Virol 1993;67:4017–4026. 103. De Francesco R, Steinkühler C. Structure and function of the hepatitis C virus NS3-NS4A serine protease. Curr Topics Microbiol Immunol 2000;242:149–169. 104. Kwong AD, Kim JL, Lin C. Structure and function of hepatitis C virus NS3 helicase. Curr Topics Microbiol Immunol 2000;242:171–196. 105. Wölk B, Sansonno D, Kräusslich H-G, et al. Subcellular localization, stability and trans-cleavage competence of the hepatitis C virus NS3-NS4A complex expressed in tetracyclineregulated cell lines. J Virol 2000;74:2293–2304. 106. Kim JL, Morgenstern KA, Lin C, et al. Crystal structure of the hepatitis C virus NS3 protease domain complexed with a synthetic NS4A cofactor peptide. Cell 1996;87: 343–355. 107. Love RA, Parge HE, Wickersham JA, et al. The crystal structure of hepatitis C virus NS3 proteinase reveals a trypsinlike fold and a structural zinc binding site. Cell 1996;87:331–342. 108. Yan Y, Li Y, Munshi S, et al. Complex of NS3 protease and NS4A peptide of BK strain of hepatitis C virus: a 2.2 Å resolution structure in a hexagonal crystal form. Protein Sci 1998;7:837–847.
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
109. Yao N, Hesson T, Cable M, et al. Structure of the hepatitis C virus RNA helicase domain. Nature Struct Biol 1997;4:463–467. 110. Kim JL, Morgenstern KA, Griffith JP, et al. Hepatitis virus NS3 RNA helicase domain with a bound oligonucleotide: the crystal structure provides insights into the mode of unwinding. Structure Fold Des 1998;6:89–100. 111. Yao N, Reichert P, Taremi SS, et al. Molecular views of viral polyprotein processing revealed by the crystal structure of the hepatitis C virus bifunctional protease–helicase. Structure Fold Des 1999;7:1353–1363. 112. Lamarre D, Anderson PC, Bailey M, et al. An NS3 protease inhibitor with antiviral effects in humans infected with hepatitis C virus. Nature 2003;426:186–189. 113. Hinrichsen H, Benhamou Y, Wedemeyer H, et al. Short-term antiviral efficacy of BILN 2061, a hepatitis C virus serine protease inhibitor, in hepatitis C genotype 1 patients. Gastroenterology 2004;127:1347–1355. 114. Serebrov V, Pyle AM. Periodic cycles of RNA unwinding and pausing by hepatitis C virus NS3 helicase. Nature 2004;430:476–480. 115. Hügle T, Fehrmann F, Bieck E, et al. The hepatitis C virus nonstructural protein 4B is an integral endoplasmic reticulum membrane protein. Virology 2001;284:70–81. 116. Qu L, McMullan LK, Rice CM. Isolation and characterization of noncytopathic pestivirus mutants reveals a role for nonstructural protein NS4B in viral cytopathogenicity. J Virol 2001;75:10651–10662. 117. Lundin M, Monne M, Widell A, et al. Topology of the membrane-associated hepatitis C virus protein NS4B. J Virol 2003;77:5428–5438. 118. Gosert R, Egger D, Lohmann V, et al. Identification of the hepatitis C virus RNA replication complex in Huh-7 cells harboring subgenomic replicons. J Virol 2003;77:5487–5492. 119. Reed KE, Gorbalenya AE, Rice CM. The NS5A/NS5 proteins of viruses from three genera of the family flaviviridae are phosphorylated by associated serine/threonine kinases. J Virol 1998;72:6199–6206. 120. Katze MG, Kwieciszewski B, Goodlett DR, et al. Ser(2194) is a highly conserved major phosphorylation site of the hepatitis C virus nonstructural protein NS5A. Virology 2000;278:501–513. 121. Reed K, Rice CM. Identification of the major phosphorylation site of the hepatitis C virus H strain NS5A protein as serine 2321. J Biol Chem 1999;274:28011–28018. 122. Koch JO, Bartenschlager R. Modulation of hepatitis C virus NS5A hyperphosphorylation by nonstructural proteins NS3, NS4A, and NS4B. J Virol 1999;73:7138–7146. 123. Neddermann P, Clementi A, De Francesco R. Hyperphosphorylation of the hepatitis C virus NS5A protein requires an active NS3 protease, NS4A, NS4B, and NS5A encoded on the same polyprotein. J Virol 1999;73: 9984–9991. 124. Brass V, Bieck E, Montserret R, et al. An aminoterminal amphipathic alpha-helix mediates membrane association of the hepatitis C virus nonstructural protein 5A. J Biol Chem 2002;277:8130–8139. 125. Elazar M, Cheong KH, Liu P, et al. Amphipathic helixdependent localization of NS5A mediates hepatitis C virus RNA replication. J Virol 2003;77:6055–6061. 126. Penin F, Brass V, Appel N, et al. Structure and function of the membrane anchor domain of hepatitis C virus nonstructural protein 5A. J Biol Chem 2004;279:40835–40843. 127. Tellinghuisen TL, Marcotrigiano J, Gorbalenya AE, Rice CM. The NS5A protein of hepatitis C virus is a zinc metalloprotein. J Biol Chem 2004;279:48576–48587. 128. Tellinghuisen TL, Marcotrigiano J, Rice CM. Structure of the zinc-binding domain of an essential replicase component of
129.
130.
131.
132.
133.
134. 135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
hepatitis C virus reveals a novel fold. Nature 2005;435:375–379. Enomoto N, Sakuma I, Asahina Y, et al. Mutations in the nonstructural protein 5A gene and response to interferon in patients with chronic hepatitis C virus 1b infection. N Engl J Med 1996;334:77–81. Witherell GW, Beineke P. Statistical analysis of combined substitutions in nonstructural 5A region of hepatitis C virus and interferon response. J Med Virol 2001;63:8–16. Gale M Jr, Korth MJ, Tang NM, et al. Evidence that hepatitis C virus resistance to interferon is mediated through repression of the PKR protein kinase by the nonstructural 5A protein. Virology 1997;230:217–227. Gale M Jr, Blakely CM, Kwieciszewski B, et al. Control of PKR protein kinase by hepatitis C virus nonstructural 5A protein: molecular mechanisms of kinase regulation. Mol Cell Biol 1998;18:5208–5218. Tan S-L, Katze MG. How hepatitis C virus counteracts the interferon response: the jury is still out on NS5A. Virology 2001;284:1–12. Macdonald A, Harris M. Hepatitis C virus NS5A: tales of a promiscuous protein. J Gen Virol 2004;85:2485–2502. Behrens S-E, Tomei L, De Francesco R. Identification and properties of the RNA-dependent RNA polymerase of hepatitis C virus. EMBO J 1996;15:12–22. Lohmann V, Körner F, Herian U, Bartenschlager R. Biochemical properties of hepatitis C virus NS5B RNAdependent RNA polymerase and identification of amino acid sequence motifs essential for enzymatic activity. J Virol 1997;71:8416–8428. Yamashita T, Kaneko S, Shirota Y, et al. RNA-dependent RNA polymerase activity of the soluble recombinant hepatitis C virus NS5B protein truncated at the C-terminal region. J Biol Chem 1998;273:15479–15486. Ferrari E, Wright-Minogue J, Fang JW, et al. Characterization of soluble hepatitis C virus RNA-dependent RNA polymerase expressed in Escherichia coli. J Virol 1999;73:1649–1654. Ago H, Adachi T, Yoshida A, et al. Crystal structure of the RNA-dependent RNA polymerase of hepatitis C virus. Structure Fold Des 1999;7:1417–1426. Bressanelli S, Tomei L, Roussel A, et al. Crystal structure of the RNA-dependent RNA polymerase of hepatitis C virus. Proc Natl Acad Sci USA 1999;96:13034–13039. Lesburg CA, Cable MB, Ferrari E, et al. Crystal structure of the RNA-dependent RNA polymerase from hepatitis C virus reveals a fully encircled active site. Nature Struct Biol 1999;6:937–943. Bressanelli S, Tomei L, Rey FA, De Francesco R. Structural analysis of the hepatitis C virus RNA polymerase in complex with ribonucleotides. J Virol 2002;76:3482–3492. Butcher SJ, Grimes JM, Makeyev EV, et al. A mechanism for initiating RNA-dependent RNA polymerization. Nature 2001;410:235–240. Choi KH, Groarke JM, Young DC, et al. The structure of the RNA-dependent RNA polymerase from bovine viral diarrhea virus establishes the role of GTP in de novo initiation. Proc Natl Acad Sci USA 2004;101:4425–4430. Pata JD, Schultz SC, Kirkegaard K. Functional oligomerization of poliovirus RNA-dependent RNA polymerase. RNA 1995;1:466–477. Hobson SD, Rosenblum ES, Richards OC, et al. Oligomeric structures of poliovirus polymerase are important for function. EMBO J 2001;20:1153–1163. Lyle JM, Bullitt E, Bienz K, Kirkegaard K. Visualization and functional analysis of RNA-dependent RNA polymerase lattices. Science 2002;296:2218–2222. Qin W, Luo H, Nomura T, et al. Oligomeric interaction of hepatitis C virus NS5B is critical for catalytic activity of RNA-
143
Section I. Pathophysiology of the Liver
149.
150.
151.
152.
153. 154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
144
dependent RNA polymerase. J Biol Chem 2002;277:2132–2137. Wang QM, Hockman MA, Staschke K, et al. Oligomerization and cooperative RNA synthesis activity of hepatitis C virus RNA-dependent RNA polymerase. J Virol 2002;76: 3865–3872. Schmidt-Mende J, Bieck E, Hügle T, et al. Determinants for membrane association of the hepatitis C virus RNAdependent RNA polymerase. J Biol Chem 2001;276: 44052–44063. Ivashkina N, Wölk B, Lohmann V, et al. The hepatitis C virus RNA-dependent RNA polymerase membrane insertion sequence is a transmembrane segment. J Virol 2002;76:13088–13093. Moradpour D, Brass V, Bieck E, et al. Membrane association of the RNA-dependent RNA polymerase is essential for hepatitis C virus RNA replication. J Virol 2004;78:13278–13284. Lanford RE, Bigger C. Advances in model systems for hepatitis C virus research. Virology 2002;293:1–9. Shimizu YK, Igarashi H, Kiyohara T, et al. Hyperimmune serum against a synthetic peptide corresponding to the hypervariable region 1 of hepatitis C virus can prevent viral infection in cell cultures. Virology 1996;223:409–412. Castet V, Fournier C, Soulier A, et al. Alpha interferon inhibits hepatitis C virus replication in primary human hepatocytes infected in vitro. J Virol 2002;76:8189–8199. Moradpour D, Heim MH, Cerny A, et al. Cell lines that allow regulated expression of HCV proteins: principles and applications. In: Schinazi RF, Sommadossi J-P, Rice CM, eds. Frontiers in viral hepatitis. Amsterdam: Elsevier Science, 2003: 175–186. Zhao WD, Wimmer E, Lahser FC. Poliovirus/hepatitis C virus (internal ribosomal entry site-core) chimeric viruses: improved growth properties through modification of proteolytic cleavage site and requirement for core RNA sequences but not for corerelated polypeptides. J Virol 1999;73:1546–1554. Filocamo G, Pacini L, Nardi C, et al. Selection of functional variants of the NS3-NS4A protease of hepatitis C virus by using chimeric Sindbis viruses. J Virol 1999;73:561–575. Lohmann V, Körner F, Dobierzewska A, Bartenschlager R. Mutations in hepatitis C virus RNAs conferring cell culture adaptation. J Virol 2001;75:1437–1449. Lohmann V, Hoffmann S, Herian U, et al. Viral and cellular determinants of hepatitis C virus RNA replication in cell culture. J Virol 2003;77:3007–3019. Evans MJ, Rice CM, Goff SP. Phosphorylation of hepatitis C virus nonstructural protein 5A modulates its protein interactions and viral RNA replication. Proc Natl Acad Sci USA 2004;101:13038–13043. Neddermann P, Quintavalle M, Di Pietro C, et al. Reduction of hepatitis C virus NS5A hyperphosphorylation by selective inhibition of cellular kinases activates viral RNA replication in cell culture. J Virol 2004;78:13306–13314. Appel N, Pietschmann T, Bartenschlager R. Mutational analysis of hepatitis C virus nonstructural protein 5A: potential role of differential phosphorylation in RNA replication and identification of a genetically flexible domain. J Virol 2005;79:3187–3194. Bukh J, Pietschmann T, Lohmann V, et al. Mutations that permit efficient replication of hepatitis C virus RNA in Huh-7 cells prevent productive replication in chimpanzees. Proc Natl Acad Sci USA 2002;99:14416–14421. Frese M, Pietschmann T, Moradpour D, et al. Interferon-a inhibits hepatitis C virus subgenomic RNA replication by an MxA-independent pathway. J Gen Virol 2001;82:723–733. Frese M, Schwärzle V, Barth K, et al. Interferon-g inhibits replication of subgenomic and genomic hepatitis C virus RNAs. Hepatology 2002;35:694–703.
167. Foy E, Li K, Wang C, et al. Regulation of interferon regulatory factor-3 by the hepatitis C virus serine protease. Science 2003;300:1145–1148. 168. Foy E, Li K, Sumpter R Jr, et al. Control of antiviral defenses through hepatitis C virus disruption of retinoic acid-inducible gene-I signaling. Proc Natl Acad Sci USA 2005;102:2986–2991. 169. Li K, Foy E, Ferreon JC, et al. Immune evasion by hepatitis C virus NS3/4A protease-mediated cleavage of the Toll-like receptor 3 adaptor protein TRIF. Proc Natl Acad Sci USA 2005;102:2992–2997. 169a. Meylan E, Curran J, Hofmann K, et al. Cardif is an adaptor protein in the RIG-1 antiviral pathway and is targeted by hepatitis C virus. Nature 2005;437:1167–1172. 170. Blight KJ, McKeating JA, Marcotrigiano J, Rice CM. Efficient replication of hepatitis C virus genotype 1a RNAs in cell culture. J Virol 2003;77:3181–3190. 171. Kato T, Date T, Miyamoto M, et al. Efficient replication of the genotype 2a hepatitis C virus subgenomic replicon. Gastroenterology 2003;125:1808–1817. 172. Krieger N, Lohmann V, Bartenschlager R. Enhancement of hepatitis C virus RNA replication by cell culture-adaptive mutations. J Virol 2001;75:4614–4624. 173. Yi M, Bodola F, Lemon SM. Subgenomic hepatitis C virus replicons inducing expression of a secreted enzymatic reporter protein. Virology 2002;304:197–210. 174. Murray EM, Grobler JA, Markel EJ, et al. Persistent replication of hepatitis C virus replicons expressing the beta-lactamase reporter in subpopulations of highly permissive Huh7 cells. J Virol 2003;77:2928–2935. 175. Blight KJ, McKeating JA, Rice CM. Highly permissive cell lines for subgenomic and genomic hepatitis C virus RNA replication. J Virol 2002;76:13001–13014. 176. Ikeda M, Yi M, Li K, Lemon SM. Selectable subgenomic and genome-length dicistronic RNAs derived from an infectious molecular clone of the HCV-N strain of hepatitis C virus replicate efficiently in cultured Huh7 cells. J Virol 2002;76:2997–3006. 177. Pietschmann T, Lohmann V, Kaul A, et al. Persistent and transient replication of full-length hepatitis C virus genomes in cell culture. J Virol 2002;76:4008–4021. 178. Zhu Q, Guo JT, Seeger C. Replication of hepatitis C virus subgenomes in nonhepatic epithelial and mouse hepatoma cells. J Virol 2003;77:9204–9210. 179. Kato T, Date T, Miyamoto M, et al. Nonhepatic cell lines HeLa and 293 support efficient replication of the hepatitis C virus genotype 2a subgenomic replicon. J Virol 2005;79:592–596. 180. Moradpour D, Evans MJ, Gosert R, et al. Insertion of green fluorescent protein into nonstructural protein 5A allows direct visualization of functional hepatitis C virus replication complexes. J Virol 2004;78:7400–7409. 181. Wakita T, Pietschmann T, Kato T, et al. Production of infectious hepatitis C virus in tissue culture from a cloned viral genome. Nat Med 2005;11:791–796. 182. Zhong J, Gastaminza P, Cheng G, et al. Robust hepatitis C virus infection in vitro. Proc Natl Acad Sci USA 2005;102:9294–9299. 183. Lindenbach BD, Evans MJ, Syder AJ, et al. Complete replication of hepatitis C virus in cell culture. Science 2005;309:623–626. 184. Xie ZC, Riezu-Boj JI, Lasarte JJ, et al. Transmission of hepatitis C virus infection to tree shrews. Virology 1998;244:513–520. 185. Bukh J. A critical role for the chimpanzee model in the study of hepatitis C. Hepatology 2004;39:1469–1475. 186. Cooper S, Erickson AL, Adams EJ, et al. Analysis of a successful immune response against hepatitis C virus. Immunity 1999;10:439–449.
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
187. Grakoui A, Shoukry NH, Woollard DJ, et al. HCV persistence and immune evasion in the absence of memory T cell help. Science 2003;302:659–662. 188. Shoukry NH, Grakoui A, Houghton M, et al. Memory CD8+ T cells are required for protection from persistent hepatitis C virus infection. J Exp Med 2003;197:1645–1655. 189. Houghton M. Strategies and prospects for vaccination against the hepatitis C viruses. Curr Topics Microbiol Immunol 2000;242:327–339. 190. Forns X, Bukh J, Purcell RH. The challenge of developing a vaccine against hepatitis C virus. J Hepatol 2002;37: 684–695. 191. Inchauspé G, Feinstone S. Development of a hepatitis C virus vaccine. Clin Liver Dis 2003;7:243–259, xi. 192. Beames B, Chavez D, Guerra B, et al. Development of a primary tamarin hepatocyte culture system for GB virus-B: a surrogate model for hepatitis C virus. J Virol 2000;74:11764–11772. 193. Bukh J, Apgar CL, Yanagi M. Toward a surrogate model for hepatitis C virus: an infectious molecular clone of the GB virus-B hepatitis agent. Virology 1999;262:470–478. 194. De Tomassi A, Pizzuti M, Graziani R, et al. Cell clones selected from the Huh7 human hepatoma cell line support efficient replication of a subgenomic GB virus B replicon. J Virol 2002;76:7736–7746. 195. Martin A, Bodola F, Sangar DV, et al. Chronic hepatitis associated with GB virus B persistence in a tamarin after intrahepatic inoculation of synthetic viral RNA. Proc Natl Acad Sci USA 2003;100:9962–9967. 196. Bright H, Carroll AR, Watts PA, Fenton RJ. Development of a GB virus B marmoset model and its validation with a novel series of hepatitis C virus NS3 protease inhibitors. J Virol 2004;78:2062–2071. 197. Mercer DF, Schiller DE, Elliott JF, et al. Hepatitis C virus replication in mice with chimeric human livers. Nature Med 2001;7:927–933. 198. Pileri P, Uematsu Y, Campagnoli S, et al. Binding of hepatitis C virus to CD81. Science 1998;282:938–941. 199. Agnello V, Abel G, Elfahal M, et al. Hepatitis C virus and other flaviviridae viruses enter cells via low density lipoprotein receptor. Proc Natl Acad Sci USA 1999;96:12766–12771. 200. Scarselli E, Ansuini H, Cerino R, et al. The human scavenger receptor class B type I is a novel candidate receptor for the hepatitis C virus. EMBO J 2002;21:5017–5025. 201. Bartosch B, Vitelli A, Granier C, et al. Cell entry of hepatitis C virus requires a set of co-receptors that include the CD81 tetraspanin and the SR-B1 scavenger receptor. J Biol Chem 2003;278:41624–41630. 202. Cormier EG, Tsamis F, Kajumo F, et al. CD81 is an entry coreceptor for hepatitis C virus. Proc Natl Acad Sci USA 2004;101:7270–7274. 203. Zhang J, Randall G, Higginbottom A, et al. CD81 is required for hepatitis C virus glycoprotein-mediated viral infection. J Virol 2004;78:1448–1455. 204. Gardner JP, Durso RJ, Arrigale RR, et al. L-SIGN (CD 209L) is a liver-specific capture receptor for hepatitis C virus. Proc Natl Acad Sci USA 2003;100:4498–4503. 205. Pöhlmann S, Zhang J, Baribaud F, et al. Hepatitis C virus glycoproteins interact with DC-SIGN and DC-SIGNR. J Virol 2003;77:4070–4080. 206. Cormier EG, Durso RJ, Tsamis F, et al. L-SIGN (CD209L) and DC-SIGN (CD209) mediate transinfection of liver cells by hepatitis C virus. Proc Natl Acad Sci USA 2004;101:14067–14072. 207. Lozach PY, Amara A, Bartosch B, et al. C-type lectins L-SIGN and DC-SIGN capture and transmit infectious hepatitis C virus pseudotype particles. J Biol Chem 2004;279:32035–32045.
208. Flint M, Thomas JM, Maidens CM, et al. Functional analysis of cell surface-expressed hepatitis C virus E2 glycoprotein. J Virol 1999;73:6782–6790. 209. Lozach PY, Lortat-Jacob H, De Lacroix De Lavalette A, et al. DC-SIGN and L-SIGN are high affinity binding receptors for hepatitis C virus glycoprotein E2. J Biol Chem 2003;278:20358–20366. 210. Lambot M, Fretier S, Op De Beeck A, et al. Reconstitution of hepatitis C virus envelope glycoproteins into liposomes as a surrogate model to study virus attachment. J Biol Chem 2002;277:20625–20630. 211. Triyatni M, Saunier B, Maruvada P, et al. Interaction of hepatitis C virus-like particles and cells: a model system for studying viral binding and entry. J Virol 2002;76:9335–9344. 212. Wellnitz S, Klumpp B, Barth H, et al. Binding of hepatitis C virus-like particles derived from infectious clone H77C to defined human cell lines. J Virol 2002;76:1181–1193. 213. Blanchard E, Brand D, Trassard S, et al. Hepatitis C virus-like particle morphogenesis. J Virol 2002;76:4073–4079. 214. Lagging LM, Meyer K, Owens RJ, Ray R. Functional role of hepatitis C virus chimeric glycoproteins in the infectivity of pseudotyped virus. J Virol 1998;72:3539–3546. 215. Matsuura Y, Tani H, Suzuki K, et al. Characterization of pseudotype VSV possessing HCV envelope proteins. Virology 2001;286:263–275. 216. Buonocore L, Blight KJ, Rice CM, Rose JK. Characterization of vesicular stomatitis virus recombinants that express and incorporate high levels of hepatitis C virus glycoproteins. J Virol 2002;76:6865–6872. 217. Bressanelli S, Stiasny K, Allison SL, et al. Structure of a flavivirus envelope glycoprotein in its low-pH-induced membrane fusion conformation. EMBO J 2004;23:728–738. 218. Gibbons DL, Vaney MC, Roussel A, et al. Conformational change and protein–protein interactions of the fusion protein of Semliki Forest virus. Nature 2004;427:320–325. 219. Modis Y, Ogata S, Clements D, Harrison SC. Structure of the dengue virus envelope protein after membrane fusion. Nature 2004;427:313–319. 220. Egger D, Gosert R, Bienz K. Role of cellular structures in viral RNA replication. In: Semler B, Wimmer E, eds. Molecular biology of picornaviruses. Washington DC: ASM Press, 2002: 247–253. 221. Ahlquist P, Noueiry AO, Lee WM, et al. Host factors in positive-strand RNA virus genome replication. J Virol 2003;77:8181–8186. 222. Restrepo-Hartwig MA, Ahlquist P. Brome mosaic virus helicaseand polymerase-like proteins colocalize on the endoplasmic reticulum at sites of viral RNA synthesis. J Virol 1996;70:8908–8916. 223. Schaad MC, Jensen PE, Carrington JC. Formation of plant RNA virus replication complexes on membranes: role of an endoplasmic reticulum-targeted viral protein. EMBO J 1997;16:4049–4059. 224. van der Meer Y, van Tol H, Locker JK, Snijder EJ. ORF1aencoded replicase subunits are involved in the membrane association of the arterivirus replication complex. J Virol 1998;72:6689–6698. 225. Rust RC, Landmann L, Gosert R, et al. Cellular COPII proteins are involved in production of the vesicles that form the poliovirus replication complex. J Virol 2001;75: 9808–9818. 226. Ritzenthaler C, Laporte C, Gaire F, et al. Grapevine fanleaf virus replication occurs on endoplasmic reticulum-derived membranes. J Virol 2002;76:8808–8819. 227. Mackenzie JM, Jones MK, Westaway EG. Markers for transGolgi membranes and the intermediate compartment localize to induced membranes with distinct replication functions in flavivirus-infected cells. J Virol 1999;73:9555–9567.
145
Section I. Pathophysiology of the Liver
228. Gazina EV, Mackenzie JM, Gorrell RJ, Anderson DA. Differential requirements for COPI coats in formation of replication complexes among three genera of Picornaviridae. J Virol 2002;76:11113–11122. 229. Krogerus C, Egger D, Samuilova O, et al. The replication complex of human parechovirus 1. J Virol 2003;77:8512–8523. 230. Miller DJ, Schwartz MD, Ahlquist P. Flock house virus RNA replicates on outer mitochondrial membranes in Drosophila cells. J Virol 2001;75:11664–11676. 231. Kujala P, Ikäheimonen A, Ehsani N, et al. Biogenesis of the Semliki Forest virus RNA replication complex. J Virol 2001;75:3873–3884. 232. Schwartz M, Chen J, Janda M, et al. A positive-strand RNA virus replication complex parallels form and function of retrovirus capsids. Mol Cell 2002;9:505–514. 233. Wu S-X, Ahlquist P, Kaesberg P. Active complete in vitro replication of nodavirus RNA requires glycerophospholipid. Proc Natl Acad Sci USA 1992;89:11136–11140. 234. Ahola T, Lampio A, Auvinen P, Kääriäinen L. Semliki Forest virus mRNA capping enzyme requires association with anionic membrane phospholipids for activity. EMBO J 1999;18:3164–3172. 235. Shirota Y, Luo H, Qin W, et al. Hepatitis C virus NS5A binds RNA-dependent RNA polymerase NS5B and modulates RNAdependent RNA polymerase activity. J Biol Chem 2002;277:11149–11155. 236. Dimitrova M, Imbert I, Kieny MP, Schuster C. Protein–protein interactions between hepatitis C virus nonstructural proteins. J Virol 2003;77:5401–5414. 237. Dubuisson J, Penin F, Moradpour D. Interaction of hepatitis C virus proteins with host cell membranes and lipids. Trends Cell Biol 2002;12:517–523. 238. Moradpour D, Gosert R, Egger D, et al. Membrane association of hepatitis C virus nonstructural proteins and identification of the membrane alteration that harbors the viral replication complex. Antiviral Res 2003;60:103–109. 239. Ye J, Wang C, Sumpter R Jr, et al. Disruption of hepatitis C virus RNA replication through inhibition of host protein geranylgeranylation. Proc Natl Acad Sci USA 2003;100:15865–15870. 240. Kapadia SB, Chisari FV. Hepatitis C virus RNA replication is regulated by host geranylgeranylation and fatty acids. Proc Natl Acad Sci USA 2005;102:2561–2566. 241. Brass V, Blum HE, Moradpour D. Recent developments in target identification against hepatitis C virus. Expert Opin Ther Targets 2004;8:295–307. 242. Pawlotsky JM, McHutchison JG. Hepatitis C. Development of new drugs and clinical trials: promises and pitfalls. Hepatology 2004;39:554–567. 243. De Francesco R and Migliaccio G. Challenges and successes in developing new therapies for hepatitis C. Nature 2005;436: 953–960. 244. Trozzi C, Bartholomew L, Ceccacci A, et al. In vitro selection and characterization of hepatitis C virus serine protease variants resistant to an active-site peptide inhibitor. J Virol 2003;77:3669–3679. 245. Reiser M, Hinrichsen H, Benhamou Y, et al. Antiviral efficacy of NS3-serine protease inhibitor BILN-2061 in patients with chronic genotype 2 and 3 hepatitis C. Hepatology 2005;41:832–835. 246. Choo Q-L, Kuo G, Ralston R, et al. Vaccination of chimpanzees against infection by the hepatitis C virus. Proc Natl Acad Sci USA 1994;91:1294–1298. 247. Farci P, Alter HJ, Govindarajan S, et al. Lack of protective immunity against reinfection with hepatitis C virus. Science 1992;258:135–140. 248. Prince AM, Brotman B, Huima T, et al. Immunity in hepatitis C infection. J Infect Dis 1992;165:438–443.
146
249. Lai ME, Mazzoleni AP, Argiolu F, et al. Hepatitis C virus in multiple episodes of acute hepatitis in polytransfused thalassaemic children. Lancet 1994;343:388–390. 250. Bassett SE, Guerra B, Brasky K, et al. Protective immune response to hepatitis C virus in chimpanzees rechallenged following clearance of primary infection. Hepatology 2001;33:1479–1487. 251. Major ME, Mihalik K, Puig M, et al. Previously infected and recovered chimpanzees exhibit rapid responses that control hepatitis C virus replication upon rechallenge. J Virol 2002;76:6586–6595. 252. Mehta SH, Cox A, Hoover DR, et al. Protection against persistence of hepatitis C. Lancet 2002;359:1478–1483. 253. Nevens F, Roskams T, Van Vlierberghe H, et al. A pilot study of therapeutic vaccination with envelope protein E1 in 35 patients with chronic hepatitis C. Hepatology 2003;38:1290–1296. 254. Neumann AU, Lam NP, Dahari H, et al. Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferon-alpha therapy. Science 1998;282:103–107. 255. Cerny A, Chisari FV. Pathogenesis of chronic hepatitis C: immunological features of hepatic injury and viral persistence. Hepatology 1999;30:595–601. 256. Shoukry NH, Cawthon AG, Walker CM. Cell-mediated immunity and the outcome of hepatitis C virus infection. Annu Rev Microbiol 2004;58:391–424. 257. Rehermann B, Nascimbeni M. Immunology of hepatitis B virus and hepatitis C virus infection. Nature Rev Immunol 2005;5:215–229. 258. Diepolder HM, Zachoval R, Hoffmann RM, et al. Possible mechanism involving T lymphocyte response to non structural protein 3 in viral clearance in acute hepatitis C virus infection. Lancet 1995;346:1006–1007. 259. Missale G, Bertoni R, Lamonaca V, et al. Different clinical behaviors of acute hepatitis C virus infection are associated with different vigor of the anti-viral cell-mediated immune response. J Clin Invest 1996;98:706–714. 260. Cerny A, McHutchinson JG, Pasquinelli C, et al. Cytotoxic T lymphocyte response to hepatitis C virus-derived peptides containing the HLA A2.1 binding motif. J Clin Invest 1995;95:521–530. 261. Lechner F, Wong DK, Dunbar PR, et al. Analysis of successful immune responses in persons infected with hepatitis C virus. J Exp Med 2000;191:1499–1512. 262. Thimme R, Oldach D, Chang KM, et al. Determinants of viral clearance and persistence during acute hepatitis C virus infection. J Exp Med 2001;194:1395–1406. 263. Thimme R, Bukh J, Spangenberg HC, et al. Viral and immunological determinants of hepatitis C virus clearance, persistence, and disease. Proc Natl Acad Sci USA 2002;99:15661–15668. 264. Major ME, Dahari H, Mihalik K, et al. Hepatitis C virus kinetics and host responses associated with disease and outcome of infection in chimpanzees. Hepatology 2004;39:1709–1720. 265. Su AI, Pezacki JP, Wodicka L, et al. Genomic analysis of the host response to hepatitis C virus infection. Proc Natl Acad Sci USA 2002;99:15669–15674. 266. Appay V, Dunbar PR, Callan M, et al. Memory CD8+ T cells vary in differentiation phenotype in different persistent virus infections. Nature Med 2002;8:379–385. 267. Pham TN, MacParland SA, Mulrooney PM, et al. Hepatitis C virus persistence after spontaneous or treatment-induced resolution of hepatitis C. J Virol 2004;78:5867–5874. 268. Takaki A, Wiese M, Maertens G, et al. Cellular immune responses persist and humoral responses decrease two decades after recovery from a single-source outbreak of hepatitis C. Nature Med 2000;6:578–582.
Chapter 8 REPLICATION AND PATHOGENESIS OF HEPATITIS C VIRUS
269. Heim MH, Moradpour D, Blum HE. Expression of hepatitis C virus proteins inhibits signal transduction through the Jak-STAT pathway. J Virol 1999;73:8469–8475. 270. Blindenbacher A, Duong FH, Hunziker L, et al. Expression of hepatitis c virus proteins inhibits interferon alpha signaling in the liver of transgenic mice. Gastroenterology 2003;124:1465–1475. 271. Duong FH, Filipowicz M, Tripodi M, et al. Hepatitis C virus inhibits interferon signaling through up-regulation of protein phosphatase 2A. Gastroenterology 2004;126:263–277. 272. Taylor DR, Shi ST, Romano PR, et al. Inhibition of the interferon-inducible protein kinase PKR by HCV E2 protein. Science 1999;285:107–110. 273. Crotta S, Stilla A, Wack A, et al. Inhibition of natural killer cells through engagement of CD81 by the major hepatitis C virus envelope protein. J Exp Med 2002;195: 35–41. 274. Tseng CT, Klimpel GR. Binding of the hepatitis C virus envelope protein E2 to CD81 inhibits natural killer cell functions. J Exp Med 2002;195:43–49. 275. Khakoo SI, Thio CL, Martin MP, et al. HLA and NK cell inhibitory receptor genes in resolving hepatitis C virus infection. Science 2004;305:872–874. 276. Bain C, Fatmi A, Zoulim F, et al. Impaired allostimulatory function of dendritic cells in chronic hepatitis C infection. Gastroenterology 2001;120:512–524. 277. Longman RS, Talal AH, Jacobson IM, et al. Presence of functional dendritic cells in patients chronically infected with hepatitis C virus. Blood 2004;103:1026–1029.
278. Shimizu YK, Hijikata M, Iwamoto A, et al. Neutralizing antibodies against hepatitis C virus and the emergence of neutralization escape mutant viruses. J Virol 1994;68:1494–1500. 279. Farci P, Shimoda A, Coiana A, et al. The outcome of acute hepatitis C predicted by the evolution of the viral quasispecies. Science 2000;288:339–344. 280. Chang K-M, Rehermann B, McHutchison JG, et al. Immunological significance of cytotoxic T lymphocyte epitope variants in patients chronically infected by the hepatitis C virus. J Clin Invest 1997;100:2376–2385. 281. Tsai SL, Chen YM, Chen MH, et al. Hepatitis C virus variants circumventing cytotoxic T lymphocyte activity as a mechanism of chronicity. Gastroenterology 1998;115: 954–965. 282. Erickson AL, Kimura Y, Igarashi S, et al. The outcome of hepatitis C virus infection is predicted by escape mutations in epitopes targeted by cytotoxic T lymphocytes. Immunity 2001;15:883–895. 283. Seifert U, Liermann H, Racanelli V, et al. Hepatitis C virus mutation affects proteasomal epitope processing. J Clin Invest 2004;114:250–259. 284. Logvinoff C, Major ME, Oldach D, et al. Neutralizing antibody response during acute and chronic hepatitis C virus infection. Proc Natl Acad Sci USA 2004;101:10149–10154. 285. Yu MY, Bartosch B, Zhang P, et al. Neutralizing antibodies to hepatitis C virus (HCV) in immune globulins derived from anti-HCV-positive plasma. Proc Natl Acad Sci USA 2004;101:7705–7710.
147
Section I: Pathophysiology of the Liver
9
THE LIVER AND THE IMMUNE SYSTEM Percy A. Knolle Abbreviations APC Antigen-presenting cell IFN-g Interferon-g LSEC Liver sinusoidal endothelial cell MHC Major histocompatibility complex
NKT PGE2 TGF-b
Natural killer T lymphocytes Prostaglandin E2 Transforming growth factor-b
LIVER AND IMMUNE SYSTEM The connotation of the liver with the immune system usually comprises a number of observations, such as immune-mediated development of hepatitis during persistent viral infection, or autoimmunity on one side and the development of immune tolerance towards antigens delivered to the liver on the other. Whereas during immune-mediated damage the liver is believed to serve as a target for the immune response, the induction of immune tolerance in the liver is rather considered to be an active process attributing active immune-regulatory potential to the liver. This chapter will attempt to shed light on the molecular and cellular immune mechanisms active in the liver with respect to the induction of immunity as well as of immune tolerance.
FUNCTIONS OF THE IMMUNE SYSTEM The immune system is a remarkable defense mechanism to protect the organism against constant challenge by potentially pathogenic microorganisms. Genetically determined and acquired immunodeficiencies clearly illustrate the central role of the immune response in the survival of the organism. Thus, it is generally accepted that interaction with microorganisms has shaped the immune system in evolutionary terms. Although other roles for the immune system, such as the elimination of tumor cells, have been debated it has become clear that immune defense against pathogens must avoid attack against host antigens in order to prevent autoimmunity. However, immune tolerance towards autoantigens is an active process that requires constant ‘education’ of lymphocytes. To accomplish these complex features the immune system consists of highly specialized cell populations. In general, immune responses are shaped by the interaction between these cell populations in specific anatomic compartments. The immune system can generally be categorized into two parts: the innate immune response, which comprises cells such as NK cells, macrophages, granulocytes, dendritic cells, and soluble molecules such as complement, which display fast and antigen non-specific
TLR TNF VAP-1
Toll-like receptor Tumor necrosis factor Vascular adhesion protein 1
effector functions. The adaptive immune response is comprised of cells with a clonally restricted antigen-specific receptor, such as T and B cells. Activation of cells of the adaptive immune system requires a complex and cognate interaction with antigen-presenting cells. One of the most eminent features of the immune system is its dynamic. Being of bone marrow origin, immune cells continuously circulate through the body in order to retrieve information (antigen) and/or to exert effector function. Antigen-presenting cells, most importantly dendritic cells, patrol peripheral tissues and collect antigens. They migrate to secondary lymphoid tissue (lymph nodes) where, in a unique microenvironment, MHC-restricted presentation of antigen to naïve CD4 and CD8 T cells takes place. Depending on their activation status dendritic cells will then either induce tolerance in T cells or stimulate T-cell immunity. In recent years it has become increasingly clear that many different dendritic cell subpopulations exist.1 These subpopulations display different functional capacities, which further adds to the complex reaction pattern of the immune system. Dendritic cells have been found that continuously sample antigens from mucosal surfaces. Contrary to the belief that epithelial cells in the mucosa constitute a tight barrier that prevents (commensal) bacteria from entering the body, dendritic cells breach this barrier by forming transepithelial dendrites and continuously sample antigen and bacteria directly from the gut lumen or other mucosal surfaces.2,3 Moreover, the dendritic cell collecting antigen may not be identical to the dendritic cell that finally presents the antigen to T cells. It has become clear that there is a ‘division of labor’ among different dendritic cell subtypes. For instance, Langerhans’ cells reside in the skin, continuously collect antigen, and upon appropriate stimulation migrate into regional lymph nodes. However, it is not Langerhans’ cells that present skinderived antigen to T cells but a lymph node-resident CD8+ dendritic cell population, indicating that Langerhans’ cells transfer antigen to this apparently more specialized cell population.4 Finally, antigens may gain access directly to the organism. Following oral ingestion a rapid dissemination of antigen via the bloodstream is observed that leads to systemic activation of the immune system towards gut-derived antigens.5 Antigens may even gain access directly to lymph nodes using a conduit system that delivers them into certain anatomic compartments in the lymph node, where
149
Section I. Pathophysiology of the Liver
interaction with specialized dendritic cells occurs.6 All these results strongly support the notion that different dendritic cell subpopulations exist in different anatomic compartments and are shaped in their function by the unique microenvironment in order to achieve local and systemic immune surveillance. Immune tolerance in the organism is achieved via many different mechanisms. Clonal elimination of autoreactive T cells occurs early during T-cell ontogeny in the thymus. However, not all autoreactive T cells are eliminated here, and further control in the periphery is required. Many different mechanisms have been described as functioning in the maintenance of peripheral immune tolerance, such as clonal deletion, anergy and regulation/deviation. Central to all these mechanisms are antigen-presenting cells. Under immature steadystate conditions dendritic cells induce T-cell tolerance.7 It is difficult to study the mechanisms involved in tolerance induction by dendritic cells, because isolation and culture conditions modify the functional phenotype of the cells. Their capacity to induce T-cell tolerance in vivo under non-activating conditions has been clearly demonstrated.8 But what induces the switch from a tolerogenic to an immunogenic dendritic cell? Activation of the dendritic cell is the key to understanding their ability to trigger immunity. This activation can be achieved by a number of different molecular events, the most important probably being activation via CD40 ligation and via Toll-like receptors (TLR). Conserved microbial patterns are recognized by these pattern recognition receptors and initiate strong activation of the dendritic cell. This interaction appears to license the dendritic cells to educate T cells for subsequent immune effector function.9 Furthermore, CD40 ligation on dendritic cells or help from CD4 T cells seems to be critical for the induction of strong immunity by dendritic cells.10,11 Upon priming of naïve T cells in lymphatic tissue, activated antigen-specific T cells undergo clonal expansion and will eventually exit lymphoid tissue for recirculation via the bloodstream. T cells that have undergone appropriate activation display a migratory pattern different from that of naïve T cells, and will patrol peripheral tissues where they exert an effector function after antigenspecific stimulation. Given the ubiquitous distribution of immune cells, one would assume that immune reactions occur in a similar fashion regardless of anatomic localization. However, a wealth of experimental data rather supports the contrary: immune responses are strongly influenced by the local microenvironment.
FUNCTIONAL HEPATIC ANATOMY The liver is optimally structured to function as a metabolic organ, i.e. in the clearance of blood from macromolecules and the release of metabolic products from hepatocytes into the bloodstream. Blood from the gastrointestinal tract rich in nutrients and in microbial degradation products enters the liver via the portal vein, which drains after extensive ramifications into the so-called portal field, which is comprised of one portal venous vessel, one arterial vessel and a bile duct surrounded by connective tissue. Portovenous and arterial blood both drain into the hepatic sinusoids, which form a three-dimensional meshwork of vessels generating a mixed arteriovenous perfusion of the liver. This generates a microenvironment with a low oxygen pressure, and metabolically active hepatocytes
150
have adapted to this unique situation. It is assumed that the hepatic microenvironment is further characterized by the presence of gutderived molecules, for example microbial degradation products such as endotoxin.12
HEPATIC CELL POPULATIONS: IMMUNE PHENOTYPE AND FUNCTION SINUSOIDAL CELL POPULATIONS Although hepatic sinusoidal cell populations (Kupffer cells, LSEC and stellate cells) contribute only to 6.3% of total liver volume they represent approximately 40% of the total number of hepatic cells, 26% of total membrane surface (mainly LSEC), 58% of total endocytotic vesicles (mainly LSEC) and 43% of total lysosomal volume (mainly Kupffer cells and LSEC) (Table 9-1).13
Kupffer Cells The hepatic macrophage population is named after the scientist von Kupffer. Kupffer cells are located predominantly in the periportal area.14 They are of bone marrow origin, as has been shown by the detection of recipient-derived macrophages in hepatic allografts.15 The lifespan of Kupffer cells appears to be more than 3 months.16 Under certain conditions proliferation of Kupffer cells is observed and can account for bone marrow-independent amplification of Kupffer cells.17 Depletion of Kupffer cells can be achieved through the application of gadolinium chloride or liposome-encapsulated chlodronate.18 Kupffer cell depletion does not by itself lead to liver damage, but affects certain hepatic immune functions (see below). Kupffer cells have three complex functions: (1) phagocytosis of particulate matter and uptake of macromolecules; (2) presentation of antigen; and (3) the release of soluble mediators. Together with the liver sinusoidal endothelial cells they form the reticuloendothelial cell system of the liver. They are efficient in phagocytosis of particulate matter and uptake of macromolecules via receptormediated endocytosis. Elimination of bacterial degradation products such as LPS is to a large extent achieved by Kupffer cells.19 Kupffer cells further contribute to the elimination of tumor cells,20 apoptotic cellular material21 and bacterial degradation products.22 As Kupffer cells are situated mainly in the periportal area, clearance of blood is achieved soon after entry into the hepatic microcirculation.23 The localization of Kupffer cells correlates with their function.24 Periportal Kupffer cells display high phagocytic capacity but
Table 9-1. Hepatic Cell Populations Hepatic cell population
Percent of liver volume*
Percent of liver cells
Kupffer cells LSEC Stellate cells Pit cells Hepatocytes
2.1 2.8 1.4 n.d. 78
15 19 5–8 n.d. 60
* Sinusoidal lumen 10.6%, space of Dissé 4.9%. (Adapted from 13)
Chapter 9 THE LIVER AND THE IMMUNE SYSTEM
Table 9-2. Soluble Mediators Released by Kupffer Cells Molecule
Function
Prostanoids IL-1 IL-1RA IL-6 IL-10 IL-12 IL-18 TNF-a TGF-b1 and TGF-b2 ROI NO
Modulation of immune function Inflammation Blockade of IL-1 activity Inflammation Anti-inflammatory activity Induction of IFN-g, immunity Induction of IFN-g, immunity Inflammation Fibrosis, anti-inflammatory activity Effector function, inflammation Effector function, vasorelaxation
Table 9-3. Phenotype of Kupffer Cells Reference 42 43 44 37, 45 36 46 47 48 49, 50
Molecule
Function
CD54 (ICAM) Fc-receptor
Cell adhesion Uptake of pathogens coated with antibodies Receptors for endotoxin Cell adhesion Co-stimulation Co-stimulation Co-stimulation Antigen presentation
CD14, TLR4, MD2 CD11a / CD11b CD40 CD80 CD86 MHC II
Reference 53 27 54 27, 55 56 56 57 56 57 27
51 52
low expression levels of MHC class II molecules and low release of mediators. In contrast, Kupffer cells located in the perivenous area show lower phagocytic capacity but express higher levels of MHC class II molecules.25 It has been demonstrated that Kupffer cells function as antigen-presenting cells for CD4 T cells by MHC class II restricted antigen presentation.26–29 Kupffer cells are endowed with a large number of receptors (see Table 9-3) that allow them to function as sentinel cells. In response to activation Kupffer cells release soluble mediators such as IL-1 and IL-6 that trigger hepatocellular expression of acute-phase proteins.30 Further mediators released by Kupffer cells include reactive oxygen species, eicosanoids, cytokines, chemokines, proteinases, nitric oxide (NO) and hemoxygenase (Table 9-2). However, the unique hepatic microenvironment, which is rich in bacterial degradation products and gut-derived antigens, appears to shape the immune function of Kupffer cells. Although Kupffer cells resemble macrophage populations in other anatomic sites there is a clear difference between them with respect to the release of soluble mediators after contact with pathogenic microorganisms.31 As already mentioned, Kupffer cells contribute to the clearance of bloodborne endotoxin. Interestingly, different sets of receptors are involved in the clearance and sensing of endotoxin by Kupffer cells. Scavenger receptors are functional in binding and endocytosis of endotoxin, whereas expression of the pattern recognition receptors CD14 and TLR4, in combination with the adaptor protein MD2, contributes to endotoxin-triggered Kupffer cell activation.22 In contrast to macrophages obtained from the peritoneal cavity, CD14 is not required for Kupffer cell activation by endotoxin.32 As portal venous blood contains bacterial degradation products,12 it seems plausible that protective mechanisms have evolved to prevent the inadvertent activation of immune reactions while conserving scavenger activity. In this respect, Kupffer cell activity towards endotoxin seems to be restricted. The high hepatic arginase activity that results in low arginin concentrations locally in the liver appears to limit Kupffer cell reactivity towards endotoxin.33 Furthermore, levels of reactive oxygen intermediates released by Kupffer cells, in contrast to macrophages derived from other locations, are rather low.34 In addition, Kupffer cells develop a refractory state after repetitive stimulation with endotoxin. After a first contact with endotoxin, the release of soluble mediators such as TNF-a by Kupffer cells is dramatically decreased.35 At the same time scavenger
activity, as determined by increased phagocytosis, is increased.35 These results demonstrate that local populations of innate immune cells have adapted their function to physiological needs. Further mechanisms may operate to restrict Kupffer cell reactivity to endotoxin. Kupffer cells release IL-10 after contact with endotoxin.36 IL10 is known to have anti-inflammatory effects, and the expression of IL-10 in Kupffer cells indeed controls reactivity to subsequent stimulations with endotoxin.37 The expression of IL-10 thus can be considered to function as a negative autoregulatory feedback loop. It is interesting to note that activation of Kupffer cells is not only achieved by direct contact with microbial products but may occur via the sympathetic nervous system, resulting in fast release of IL10.38 There is an increasing wealth of data suggesting a contribution of the autonomic nervous system to the control of immune responses in the liver. Adrenergic innervation appears to downregulate inflammatory immune responses in the liver, whereas peptidergic innervation aggravates immune-mediated liver injury.39 Whereas inadvertent reactivity of Kupffer cells to inflammatory stimuli appears to be controlled at multiple levels, their effector function is critically linked to activation via TLR, because their inability to react to TLR-4 ligands leads to failure to eliminate Gramnegative bacteria from the liver.40 However, Kupffer cells alone do not achieve elimination of bacteria, but require cooperation from neutrophils for efficient elimination of pathogens (see below). Kupffer cells may further engage in amplification of cell-mediated immune responses in the liver leading to organ pathology. Such a role has been described for ischemia–reperfusion injury, alcoholinduced liver injury and neutrophil-induced liver injury. Central to the deleterious function of Kupffer cells is the release of TNF-a, which exerts deleterious affects in the liver via TNF receptor 2 expressed on parenchymal cells.41 Clearly, the expression and release of proinflammatory mediators in the liver must be precisely controlled to avoid unnecessary damage secondary to immune activation.
NKT Cells The liver harbors a large number of lymphocytes that share the characteristics of T cells and NK cells. Characteristically, NKT cells express a T-cell receptor and a prototypical NK cell marker of the C type II lectin superfamily, i.e. NK1.1. Most NKT cells express an invariant T-cell receptor, Va14/Ja281 in the mouse and Va24/JaQ in humans, together with a skewed repertoire of TCRb chains, Vb8.2 in the mouse and Vb11 in humans.58 However, NK1.1+
151
Section I. Pathophysiology of the Liver
Table 9-4. Different Populations of NKT Cell
Repertoire Vb8.2 Vb diverse Vb diverse Phenotype NK receptor NK1.1 NK1.1+/Restriction And others MHC II Reactivity
I
II
Va14–Va18 Va3.2–Ja9/Va8Vb8 Va diverse
Va diverse
III
IV
CD4+ or CD8+
CD4+ or DN NK1.1 DX5
CD4+ or DN NK1.1+/-
CD8+, CD4+ or DN DX5+/-
CD1d MHC I
CD1d
MHC I
a-GalCer
Not determined
Self-agonist
Not determined
DN, double negative. (Adapted from58)
TCR+ cells are heterogeneous and can be classified according to their surface phenotype and their requirement for antigen-specific stimulation (Table 9-4). The classic NKT cells are CD1d-restricted but their natural ligands have not so far been identified. Most classic NKT cells respond to CD1d-restricted presentation of a-GalCer with fast release of soluble mediators. Most NKT cells develop in the thymus: only certain subpopulations appear to originate from other sites.59 Although the natural ligands for NKT cells are not known and their skewed expression of TCR genes imply a rather narrow ligand specificity, other ligands have been detected.60 The distinct properties of glycosphingolipid antigen recognition by classic NKT cells further suggests high ligand specificity.61 In contrast to conventional T cells, NKT cells are more numerous in the liver than in other organs. Expression of CD54 and CD11a is required for homing of NKT cells to the liver.62,63 For survival and expansion of NKT cells in the liver one single cytokine, IL-15, is most important. Interestingly, migration or retention of NKT cells to other organs is independent of CD11a.63 The hepatic cell population most important for recruitment of NKT cells is NK cells.64 In turn, Kupffer cells are operative in the recruitment of NK cells65 and dendritic cells66 to the liver. The molecular mechanisms underlying recruitment of monocytes and hepatic differentiation of monocytes into organ-resident Kupffer cells are still not entirely clear. After stimulation in vivo NKT cells undergo a wave of proliferation in the liver, spleen and bone marrow, accompanied by sustained and fast release of cytokines. Depending on the subtype, NKT cells may release IFN-g, TNF-a and/or IL-4. It has been assumed that release of these mediators shifts T helper immune responses into either the Th1 or the Th2 direction.58 Moreover, a contribution of NKT cells in many experimental disease models has been described, which suggests a role in the elimination of tumors, control of infection, mediation of autoimmunity, maintenance of tolerance and others.58 It is likely that the diverse subpopulations of NKT cells and the organ microenvironment influencing NKT cell function all contribute to modulation of local immune responses and are thus responsible for the observation of diverse functional phenotypes of NKT cell.
152
Table 9-5. Expression of Receptors Involved in Receptor-Mediated Endocytosis Receptors expressed on LSEC
Ligands
Hyaluronan/ scavenger receptor Mannose receptor
Hyaluronan, oxidized LDL, advanced glycation end products, aminoterminal propeptide of type I and III collagen Lysosomal enzymes, tissue plasminogen acivator, carboxyterminal propeptides, mannose-containing structures Collagen a chain
76–79
Immunoglobulins
82, 83
Collagen a-chain receptor Fc-receptor (FcgR, FcRn)
Reference
80
81
Liver Sinusoidal Endothelial Cells (LSEC) LSEC form a thin but continuous layer between leukocytes passing the liver in the bloodstream from hepatocytes.67 In contrast to endothelial cells in other organs, they do not express tight junctions and are not separated from parenchymal tissue by a basement membrane. The space between hepatocytes and LSEC is called the space of Dissé, which contains abundant extracellular matrix produced by LSEC and is populated by the stellate cells that span the LSEC and control sinusoidal blood flow by contraction, leading to a reduction in sinusoidal diameter.68 LSEC possess enormous endocytotic capacity. Uptake of macromolecules from the blood is achieved mainly by receptor-mediated endocytosis and not pinocytosis or macropinocytosis in these cells. Via receptor-mediated binding LSEC immobilize particulate material but are unable to internalize molecules larger than 200 nm.69 The receptors active in endocytotic activity of LSEC and their main ligands are summarized in Table 95. The wide ligand range of these receptors ensures the effectiveness of LSEC scavenger function.70 This scavenger function is not observed in endothelial cells from other vascular beds and emphasizes the notion that unique local cell populations define liver
Chapter 9 THE LIVER AND THE IMMUNE SYSTEM
function.70 Scavenged molecules are quickly degraded by LSEC or are transported across the cell to neighboring hepatocytes in a transcytotic fashion.71,72 The molecular mechanisms determining lysosomal degradation or transcytosis in LSEC have not yet been identified. The scavenger function of LSEC may further be operative in the hepatotropism observed for certain viruses. Uptake of bloodborne virus by LSEC has been demonstrated for duck hepatitis B virus73 and for hepatitis C virus.74 LSEC express receptors for these viruses. In analogy to what has been observed for human immunodeficiency virus (HIV) binding to dendritic cells through DC-SIGN,75 binding of hepatotropic viruses to receptors on scavenger LSEC may target the virus to the liver and eventually lead to infection of hepatocytes in trans. LSEC have pores – so-called fenestrae – approximately 100– 150 nm in size67 which can be dynamically regulated by the actin cytoskeleton upon contact with substances such as alcohol or nicotine.84 Blood cells passing through the narrow hepatic sinusoids exert a ‘sinusoidal massage’, causing improved exchange of fluid between the sinusoidal lumen and the space of Dissé.67 Flexible macromolecules larger than 100 nm in diameter or rigid macromolecules larger than 12 nm are excluded from access to the space of Dissé via diffusion through the fenestrae, resulting in a ‘sieve’ function of LSEC.67 Larger molecules, such as chylomicrons exceeding 100 nm in size, must be metabolized by membrane-associated lipase85 before they can pass through fenestrae.86 Interestingly, the infection rate of hepatocytes by bloodborne adenovirus has been demonstrated to depend critically on the diameter of LSEC fenestrae.87 Alternatively, molecules may gain access to hepatocytes through receptormediated uptake by LSEC and subsequent transcytosis (see also above).88 LSEC constitutively express the molecules necessary to establish interaction with passenger leukocytes. Expression levels of CD54 and CD106 are linked to bacterial colonization of the gut, supporting the notion that bacterial degradation products derived from the gastrointestinal tract shape the hepatic microenvironment. LSEC further express two molecules that support the adhesion of leukocytes, i.e. VAP-1 and L-SIGN.89,90 These molecules are expressed in the liver on LSEC and in lymphatic tissue. MHC class I and low levels of MHC class II molecules are constitutively expressed, together with low levels of co-stimulatory molecules (CD80, CD86 and CD40).57
Dendritic Cells Dendritic cells are the prototypic antigen-presenting cells of the immune system. Depending on their maturation status (either immature, semi-mature or mature), they are considered to shape the immune response in the direction of either immune tolerance or immunity.1,91,92 Dendritic cells are found in the liver under normal steady-state conditions. They are found primarily in the portal tract, but are present in lower numbers in the periportal and perivenous areas.93 As the liver is a strongly vascularized organ, immune surveillance here by dendritic cells is different from immune surveillance in other sites such as the skin or the gut. Dendritic cells arrive in the liver via the bloodstream, and as a consequence of the slow blood flow within hepatic sinusoids have the opportunity to interact with sinusoidal lining cells. Within the liver dendritic cells translocate from hepatic sinusoids to the lymph, and
finally accumulate in draining celiac lymph nodes.94 The transition of dendritic cells from the bloodstream into adherent cells within the hepatic sinusoid appears to be accompanied by changes in their ultrastructural characteristics, which is suggestive of a maturation step.95 It is unclear whether changes in dendritic cell ultrastructure are paralleled by functional maturation. However, after transition from the hepatic sinusoid dendritic cells lose their phagocytic capacity. The preferential accumulation of circulating dendritic cells in the liver implies that these cells will engage in antigen collection within the liver. Indeed, after translocation from the hepatic sinusoid into the space of Dissé dendritic cells are observed to engage in close physical contact with hepatocytes and stellate cells.95 Although the total number of dendritic cells in liver is high compared to that of other parenchymal organs, their relative density in the liver is much lower.96 However, a fast turnover of dendritic cells may compensate for lower numbers. As already mentioned, constant recruitment of dendritic cells from the marginating blood pool may result in different migration kinetics of dendritic cells in the tissue. The composition of dendritic cell subtypes in the liver appears to differ from those in other organs such as the spleen.97 The liver bears more plasmacytoid dendritic cells, characterized by CD8a B220+ expression, than the spleen. However, it seems difficult to attribute functional capacity to dendritic cells purely on the basis of their phenotypic characteristics. It has become clear that subpopulations of dendritic cells are shaped by external factors, and changes in the pattern of surface molecule expression may in fact reflect local influences rather than the recruitment of lineage-dependent subpopulations of cells. A number of investigations have been performed that yielded extensive information on the phenotype of hepatic dendritic cells.97–99 A common result is that hepatic dendritic cells are rather immature compared to dendritic cells from other organs. In addition to the different distribution of dendritic cell subpopulations, hepatic dendritic cells show distinct immune-regulatory features.98 Dendritic cells isolated from liver show a reduced capacity to prime naïve allogeneic T cells compared to dendritic cells isolated from bone marrow. T cells primed by hepatic dendritic cells do not develop strong cytotoxic effector function.100 Furthermore, adoptive transfer of these in vitro-propagated hepatic dendritic cells leads to increased expression of IL-10 in lymphatic tissue.100 Similar results were obtained by a number of other investigators.97,99,101 Using a new technique to obtain dendritic cells from human liver, it was possible to investigate the immune function of human hepatic dendritic cells. These cells were less effective in T-cell stimulation than were dendritic cells isolated from the skin. Moreover, hepatic dendritic cells expressed significant amounts of IL-10, a cytokine known to mediate potent anti-inflammatory action.102 T cells primed by hepatic dendritic cells release IL-10 and IL-4 but no IFNg, which is suggestive of their ability to induce regulatory T cells. Because dendritic cells derived from other organs show different functional features it is most likely that dendritic cells entering the liver were modified by the microenvironment. Increasing the numbers of dendritic cells and NK cells by injection of flt3-ligand leads to modification of intrahepatic immune regulation. Liver transplants are typically well tolerated in rodent models of allotransplantation. However, if livers from flt3-ligand-treated animals were transplanted increased transplant rejection was observed, which
153
Section I. Pathophysiology of the Liver
suggests that flt3-ligand-induced changes in hepatic cell populations tip the balance from a tolerogenic to an immunogenic milieu.103 Another important feature of hepatic dendritic cells is their low expression of TLR4. Stimulation of hepatic dendritic cells with low concentrations of endotoxin does not lead to strong activation and induction of a mature phenotype, but rather results in a reduced capacity to induce T-cell priming and proliferation compared to dendritic cells isolated from spleen.104 Given the continuous presence of bacterial degradation products in portal venous blood, it seems important to prevent inadvertent immune activation locally in the liver. Dendritic cells in diseased liver are mainly found in the portal tract. Hepatic expression of the chemokine CCL21 stimulates increased recruitment of CCR7+ immune cells to the portal tract, which resembles tertiary lymphoid tissue.105 CCL21 recruits naïve T cells and dendritic cells, which both bear CCR7+, and thus promotes strong local interaction between these cell populations,106 similar to the role of CCR7 for recruitment of immune cells to lymphatic tissue.107 So far, little is known about the ability of migratory hepatic dendritic cells to stimulate T cells in the draining lymph node. It is interesting to note that the composition of subpopulations of dendritic cells in the hepatic draining lymph node does not reflect the subpopulation composition in the liver.108 In particular, dendritic cells in the draining hepatic lymph node had an activated mature phenotype, and only few cells had the surface phenotype characteristic of plasmacytoid dendritic cells.108
Stellate Cells Stellate cells are located in the space of Dissé between LSEC and hepatocytes. Stellate cells store 80% of retinoids in the body as retinyl palmitate within cytoplasmic vesicles. Following hepatic injury, stellate cells transdifferentiate into cells similar to myofibroblasts, with a profibrogenic phenotype depositing large amounts of extracellular matrix.109 They are further involved in regulating intrahepatic vascular resistance in response to a number of vasoactive substances.110 Although stellate cells may not have direct contact with passenger leukocytes in the sinusoid, they release a number of mediators that are important for local immune control. Several chemokines, such as MCP-1, MIP-2 and IL-8, are produced by stellate cells upon activation by proinflammatory cytokines.111 However, under resting conditions few if any chemokines are expressed by these cells, suggesting that they contribute only in inflammatory situations to the recruitment of leukocytes from the blood into the hepatic parenchyma. In addition to their important scaffolding function, stellate cells have the capacity to take up antigens by fluid-phase endocytosis, receptor-mediated endocytosis and phagocytosis.112 They further express low levels of MHC class I and II molecules and co-stimulatory molecules (CD80 and CD40), which are further increased following incubation with proinflammatory mediators (Table 9-6). Whereas resting stellate cells fail to engage in cognate interaction with naïve T cells, cytokine-stimulated stellate cells support the proliferation of CD4 T cells in a mixed lymphocyte reaction.112 In conclusion, stellate cells do not function as full antigen-presenting cells but can support ongoing inflammatory reactions by inducing T-cell stimulation and proliferation. Similar to Kupffer cells and LSEC, stellate cells express TLR4 and are thus responsive to stimulation by endotoxin, leading to
154
Table 9-6. Immune Phenotype of Stellate Cells Surface molecule
Expressed in resting/activated stellate cells
Reference
CD54 (ICAM-1) CD106 (VCAM-1) CD40 CD80 MHC class I MHC class I NCAM
-/+ -/+ (+)/+ -/+ +/++ (+)/+ +/++
113, 116 113, 117 112 112 112 112 118
the release of chemokines.113 However, upon activation stellate cells also release potent anti-inflammatory mediators. Stellate cells treated in vitro with TNF-a or endotoxin or isolated from animals after bile duct ligation demonstrated a prominent increase in IL-10 release.114 Moreover, stellate cell activation results in the expression of transforming growth factor b (TGF-b), one of the most potent anti-inflammatory cytokines.49 Latent TGF-b-binding protein, expressed by trans-differentiating stellate cells, serves as a matrix to anchor latent TGF-b in hepatic tissue. The release of bioactive TGF-b then requires proteolytic cleavage of the binding protein.115
HEPATOCYTES Hepatocytes constitute the vast majority of hepatic cells. They are situated behind a physical barrier constituted by LSEC and stellate cells. However, a large body of evidence suggests that hepatocytes have direct access to molecules contained in portal venous blood as well as to passenger leukocytes in the hepatic sinusoid. In normal liver hepatocytes express few MHC I molecules and are negative for MHCI II. Other molecules relevant for interaction with T cells are either not expressed or expressed at very low levels, e.g. CD54, or co-stimulatory molecules such as CD80 and CD86. The main function of hepatocytes is metabolism. Consequently, enormous amounts of molecules are generated within hepatocytes and are subsequently released into serum or bile. Degradation of waste or toxic products also occurs in hepatocytes and gives rise to modifications in the structure of protein antigens released by hepatocytes. Furthermore, hepatocytes metabolize nutrients extracted from portal venous blood. Given the huge metabolic function of hepatocytes, it seems plausible that mechanisms have evolved to protect these cells from inadvertent immune attack.
Characteristics of Blood Flow and Leukocyte–Liver Cell Interaction in the Liver The liver holds a unique position with regard to the blood circulation. It receives venous blood draining from almost the entire gastrointestinal tract via the portal vein, and from the systemic circulation via the hepatic artery. More than 2000 l of blood stream daily through the human liver. Even with an average circulation time of 1 hour, peripheral blood leukocytes pass through the liver more than 12 times per day. These simple facts clearly demonstrate that the liver is a ‘meeting-point’ for antigens and leukocytes circulating in the blood. Hepatic sinusoids are narrow channels with an average diameter ranging from 7 to 12 mm. Leukocytes, having a mean
Chapter 9 THE LIVER AND THE IMMUNE SYSTEM
diameter of 10–12 mm, have to force their way through the sinusoidal meshwork in the liver.119 Moreover, blood flow in the liver is slow because of low-pressure perfusion. Interaction between passenger cells flowing through hepatic sinusoids and the sinusoidal cell populations is facilitated by these physical conditions.120 Expression of selectins by endothelial cells is typically required to engage in interaction with leukocytes in the bloodstream. Hepatic sinusoidal cell populations express little if any CD62E (E-selectin),121 which is instrumental in the recruitment of leukocytes to endothelial cells. Despite the absence of CD62E, recruitment of leukocytes to sinusoidal cells can be observed even under inflammatory conditions.122 LSEC rather express other molecules that mediate interaction with passenger leukocytes, notably L-SIGN90 and VAP-1.89 Blood flow in the liver is intermittent as a result of interaction of passenger leukocytes with sinusoidal cell populations.123 Kupffer cells have been reported to patrol the sinusoids at low speed and to temporarily block sinusoidal perfusion.124 This further facilitates the interaction between passenger leukocytes and sinusoidal cell populations. Given the large cumulative surface of LSEC and the high numbers of Kupffer cells, interaction of passenger leukocytes with these cell populations is most likely. Contact with other hepatic cell populations, such as stellate cells, hepatocytes or dendritic cells, is not excluded but is likely to occur at much lower frequencies.
IMMUNE FUNCTIONS OF THE LIVER Immune Tolerance in the Liver Among the many functions of the liver, clearance of the blood from macromolecules and its metabolization are important for the understanding of the liver as an immune-regulatory organ. Nutrients have to be extracted from portal venous blood and further used for hepatocellular metabolism, but at the same time the liver must eliminate toxic waste products and proinflammatory agents, such as endotoxin or other bacterial degradation products derived by translocation from the gut, from blood without eliciting an immune response to all these antigens. Induction of immune tolerance in the liver was reported in 1967 by Cantor et al. and in 1969 by Calne et al.,125,126 and since then by many other groups. Three main points demonstrate the ability of the liver to induce antigen-specific immune tolerance. 1. Liver transplants are accepted by the recipient’s immune system despite MHC discrepancy and even in the absence of immune suppression.125,126 2. Simultaneous transplantation of the liver and another organ from the same donor leads to increased graft acceptance of the cotransplanted organ. Further transplants from another donor led to graft rejection, demonstrating antigen-specific induction of immune tolerance by the transplanted liver.127 3. Drainage of a transplant directly into the portal vein, or the direct application of donor cells into the portal vein, led to increased acceptance of the graft.128–131 Although for a long time antigen-specific induction of immune tolerance in the liver was observed in the context of organ transplantation, it is clear from the physiological function of the liver that immune tolerance needs equally to be established for circulating antigens. Although the mechanisms involved in tolerance induction
do not necessarily need to be different for antigens produced or taken up by hepatic cell populations, it is evident that immune tolerance towards circulating antigens requires the participation of cells conveying this specific information, i.e. antigen-presenting cells.
The Hepatic Microenvironment Constitutive exposure to gut-derived bacterial degradation products in portal venous blood contributes to the unique hepatic microenvironment.12,132 Endotoxin not only induces the release of proinflammatory mediators from hepatic sinusoidal cell populations, but at the same time leads to the expression of a number of potent anti-inflammatory immunosuppressive mediators such as IL-10,36 TGF-b49 and certain prostanoids, such as PGE2.133,134 This creates a local environment that suppresses rather than stimulates immune responses in the liver. IL-10, TGF-b and PGE2 are known to ‘educate’ antigen-presenting cells such as dendritic cells and to induce a tolerogenic phenotype. Dendritic cells exposed to these mediators fail to induce immunity, but rather support the induction of tolerant T cells.135 It is difficult to separate the influence of specific hepatic antigen-presenting cell populations and the hepatic microenvironment on the induction of immune tolerance from each other, as the microenvironment certainly has a role in shaping the functional phenotype of hepatic antigen-presenting cell populations.
Induction of T-Cell Tolerance by AntigenPresenting Cells in the Liver Different hepatic antigen-presenting cell populations contribute to antigen-specific induction of tolerance in T cells.136 Although the relevance of each of the antigen-presenting cell populations has been described separately, more than one such population may be involved in the induction of tolerance in the liver. In fact, a number of determinants, such as the origin and quantity of antigen, may strengthen or diminish the contribution of individual antigenpresenting cell populations to tolerance induction (see below).
Hepatocytes Hepatocytes can serve as antigen-presenting cells and induce stimulation of naïve CD8 T cells in a transgenic mouse setting, where hepatocytes express a transgenic MHC I molecule that is recognized together with an endogenous peptide. Although antigen-specific stimulation of T cells by hepatocytes even in the absence of co-stimulatory molecules CD80/CD86 is rather effective during the first 3 days, T cells undergo apoptosis at later time points. Thus, clonal elimination of T cells is involved in the mediation of hepatic T-cell tolerance. As interaction with passenger T cells and antigen-presenting hepatocytes occurs even in the absence of local inflammation in the transgenic mouse model described above, hepatocytes may continuously contribute to shaping of the immune response.137 As hepatocytes do not have the capacity to cross-present exogenous antigens on MHC I to CD8 T cells, tolerance induction is limited to proteins expressed by hepatocytes but does not extend to antigens entering the body.
Hepatic Dendritic Cells As described above, immature hepatic dendritic cells contribute to the induction of T-cell tolerance. Hepatic dendritic cells may be
155
Section I. Pathophysiology of the Liver
directly involved in the induction of T-cell tolerance by virtue of their tolerogenic phenotype, which may result from different subpopulations of dendritic cells being present in the liver or from the influence of the local microenvironment.97–99,101,102,104,138 On the other hand, dendritic cells may indirectly contribute to hepatic immune tolerance via the release of soluble mediators or the induction of regulatory T cells. Release of type I IFN from dendritic cells attenuates liver injury139 and at the same time promotes the induction of regulatory T cells.140 Activated CD8 T cells are found to undergo apoptosis in the liver.141 The trapping of CD8 T cells may occur antigen specifically or through CD54/CD106-dependent mechanisms.142–144 Using bone marrow chimeric animals it was possible to demonstrate that antigen-specific recruitment of CD8 T cells was achieved by nonmyeloid organ-resident cells, whereas elimination of CD8 T cells occurred through a bone marrow-dependent cell population.142
LSEC Similar to dendritic cells, LSEC have the capacity to prime CD4+ T cells, i.e. stimulation of cytokine release from naïve CD4+ T cells that have not previously encountered their specific antigen.145 Whereas dendritic cells require maturation and signals from the highly specialized lymphatic microenvironment in order to function as potent APC for naive CD4+ T cells,146 LSEC do not require maturation or migration into lymphatic tissue in order to gain APC function. This function of LSEC as sessile, organ-specific and constitutively active antigen-presenting cells is not shared by endothelial cells from other organs. Microvascular endothelial cells from the skin or the gut are unable to act as antigen-presenting cells for naive CD4+ T cells unless stimulated by proinflammatory cytokines such as IFN-g.147–149 In contrast to antigen presentation by dendritic cells, however, CD4+ T cells stimulated by antigenpresenting LSEC fail to differentiate into effector Th1 CD4+ T cells, but rather gain an immune-regulatory phenotype.145 CD4+ T cells primed by LSEC release large amounts of IL-4 and IL-10 following triggering by the T-cell receptor,145 which efficiently downregulates ongoing T cell-mediated immune responses (P. Knolle, unpublished results). Thus, antigen presentation by LSEC to naive CD4+ T cells may rather down-regulate Th1-type cell-mediated immune responses and at the same time stimulate Th2-type immune responses, leading to increased production of antibodies. Indeed, ineffective cell-mediated immune responses despite the presence of an efficient antibody response are observed during persistent infection of the liver with non-cytopathic viruses.150 Endothelial cells from other sites equally fail to lead to the development of fully differentiated effector Th1 CD4+ T cells.149,151 It is important to note that these cells lack the capacity to actively engage in immune modulation, as either endothelial cells or T cells have to be prestimulated in order to observe functional interaction, thus requiring other cell populations that drive the developing immune response. Together with the observation that intraportal injection of antigen leads to the development of T cells that release IL-4 and IL-10 upon restimulation,152 it can be assumed that LSEC rather than endothelial cells in other organs are involved in the induction of tolerance to intraportally applied antigens. Cytotoxic CD8+ T cells are of crucial importance for a successful immune response to infection with intracellular pathogens and
156
against the development of cancer cells. Presentation of antigen on MHC class I molecules to CD8+ T cells was believed to be restricted to those antigens synthesized de novo within the same cell. Although this allows for immune surveillance of parenchymal cells by CD8+ T cells, it is difficult to envisage how professional antigen-presenting cells, not infected by the pathogenic microorganism or not transformed into neoplastic cells, could in the first place induce a protective and efficient CD8+ T cell-mediated immune response. Thus, presentation of exogenous antigens on MHC class I molecules (termed cross-presentation) is obviously required. Initially identified by Bevan et al.,153 it was recently demonstrated that cross-presentation occurs in bone marrowderived antigen-presenting cells such as dendritic cells and macrophages, and in some instances in B cells.154,155 Cross-presentation by dendritic cells was shown to be necessary in order to mount an efficient CD8+ T cell-mediated immune response against virus infection, although not all virus infections appear to require cross-presentation by myeloid APC for the induction of immunity.156 It is therefore surprising to find that LSEC can efficiently crosspresent exogenous antigens on MHC class I molecules to CD8+ T cells.157 Cross-presentation by LSEC is characterized by a number of features: efficient uptake of antigen by receptor-mediated endocytosis; shuttling of antigen from endosome to cytosol for proteasomal degradation; TAP-dependent loading of processed peptides on de novo synthesized MHC class I molecules in the ER; and transport to the cell surface.157 LSEC require only 60–120 minutes to complete cross-presentation and to express peptide-loaded MHC class I molecules on the surface. Minute amounts of antigen, i.e. in the low nM range, are sufficient for cross-presentation by LSEC, suggesting an important role of cross-presenting LSEC in the hepatic immune response.157 LSEC not only cross-present antigen to armed effector CD8+ T cells, but have in fact the capacity to stimulate naïve CD8+ T cells.157 Following an encounter with cross-presenting LSEC in vitro, naïve CD8+ T cells release cytokines and begin proliferating. However, antigen-specific restimulation of these T cells revealed that they lost the ability to express effector cytokines such as IL-2 and IFN-g and that they lost their cytotoxic activity.157 In vivo it has been demonstrated that LSEC cross-present antigen to naive CD8+ T cells outside the lymphatic system. So far, stimulation of naïve T cells was believed to occur exclusively in the highly specialized lymphatic microenvironment . Following stimulation by cross-presenting LSEC, naïve CD8+ T cells start to proliferate locally in the liver. However, the outcome of cross-presentation by LSEC in vivo is the induction of systemic immune tolerance. Similar to CD8+ T cells stimulated by crosspresenting LSEC in vitro, CD8+ T cells in vivo lose the capacity to express effector cytokines and to exert cytotoxic activity against their specific target antigens once stimulated by cross-presenting LSEC.157 Deletion of antigen-specific CD8+ T cells occurs to some extent but is not the main mechanism of immune tolerance induced by LSEC. Mice rendered tolerant by LSEC cross-presenting a model antigen fail to develop an immune response against a tumor carrying this model antigen, which constitutes the prime target of the immune response in non-tolerant littermates, leading to immunity and tumor rejection in control animals.157
Chapter 9 THE LIVER AND THE IMMUNE SYSTEM
Concept of Organ-resident Antigen-Presenting Cells in the Liver LSEC represent a new type of organ-resident APC which is organ specific. In order to establish organ-specific control of immune responses, local presentation of antigen by resident APC has a number of advantages. 1. Dendritic cells take up antigen in the peripheral organs and, after appropriate stimuli, migrate to draining lymph nodes. During this journey they undergo functional maturation, which renders them potent APC once they arrived in the highly specialized and structured microenvironment of lymphatic organs. In contrast, LSEC simultaneously perform all the salient functions of an APC, i.e. uptake, processing and presentation of antigen, without the requirement for maturation. This ensures that antigen presentation of bloodborne antigens by LSEC occurs within a short time frame. 2. Although LSEC preclude access of bloodborne antigen-specific T cells to hepatocytes presenting the cognate antigen in the absence of local inflammation,158 it has been shown that armed effector cells can gain access to hepatocytes159 once LSEC can present the cognate antigen. Depending on the presence of sufficient numbers of armed effector T cells, antigen presentation by LSEC then apparently allows for immune surveillance of the liver. 3. Continuous culture of T cells or professional APC such as dendritic cells with immune suppressive mediators such as IL-10 or TGF-b in vitro gives rise to APC that induce T-cell tolerance rather than immunity.160,161 Situated in the hepatic sinusoid, sessile LSEC are continuously exposed to the unique hepatic microenvironment, which is especially rich in immune suppressive mediators. Incorporation of signals from an organ-specific microenvironment is clearly more prominent in sessile LSEC than in conventional APC that stay only for short periods in peripheral organs before migrating into lymphatic tissue. The unique hepatic microenvironment may thus have a considerable influence on the way immune responses are modulated by sessile LSEC. 4. Systemic distribution of antigen leads to the development of immune tolerance.162,163 Given the dual function of LSEC – fast and efficient presentation of bloodborne antigens and the induction of immune tolerance – the timing and distribution of an antigen appear to critically determine the outcome of the ensuing immune response. As dendritic cells require time for migration, maturation and induction of T-cell immunity in the lymphatic system,146 tolerance induction by LSEC can occur in a much shorter time frame. Immune tolerance ensues if antigen is first presented in the liver.157,164 Given the ever-changing nature of antigens released from metabolizing hepatocytes, tolerance induction by LSEC appears a useful mechanism to prevent immune attack against innocuous antigens released from hepatocytes. However, it is possible that LSEC contribute to the persistence of viral infection in hepatocytes, as abundant viral proteins are released from infected hepatocytes and can be taken up and presented by LSEC to T cells. Local presentation of antigen by LSEC may thus constitute a mechanism to balance the immune response in the liver and protect hepatocytes from immune-mediated damage.
LSEC are ideally positioned in the hepatic sinusoid to scavenge bloodborne antigens and to present these antigens to passenger T cells. Given the large volume of blood – containing both T cells and antigens – passing daily through the liver and the large cumulative surface of LSEC, the liver sinusoid appears to be a perfect ‘meeting point’ where immune responses towards bloodborne antigens can be shaped.
Implication of the Liver in Immune Surveillance The liver may not only serve as a target for immune responses generated in lymphatic tissue but may actively contribute to the modulation of immune responses. Such an active role in the shaping of local and systemic immune responses involves the induction of both immune tolerance137,157 and immunity.165 Antigens and pathogens are not contained entirely by local immune cell populations in the skin or the gut. Following oral ingestion of antigens, such antigens can be found in the systemic circulation within minutes, and within a few hours antigen-specific systemic activation of T cells is detected.5 Within minutes after application to the skin, antigens can be found in the liver, where they lead to the induction of cytokine release from local hepatic cell populations.165 The fact that antigens as well as pathogens overcome physical barriers and the hurdles of local immune surveillance operating at external body surfaces, together with their rapid systemic dissemination, necessitates a role for the liver in the containment of immune responses towards these antigens. The clearance function of the liver, which is mainly achieved by Kupffer cells and LSEC, is important to limit the systemic dissemination of pathogens and antigens. The largest source of environmental antigens and commensal microorganisms is the gut, and subsequently portal venous blood draining from the gastrointestinal tract into the liver. In the liver, the immune system has to discriminate between nutrient or innocuous antigens and (potentially) pathogenic microorganisms. As hepatic antigen-presenting cells, i.e. Kupffer cells, dendritic cells and LSEC, take up antigens and present them to the immune system, it is unlikely that maintenance of a tolerant state towards bloodborne antigens is simply achieved by ignorance of the immune system towards these antigens. It seems rather that T cells with specificity for antigens presented locally in the liver are actively tolerized by hepatic antigen-presenting cells. Hepatic dendritic cells show considerable plasticity, as they are activated upon encounter with microorganisms and subsequently induce pathogen-specific T-cell immunity.166 In contrast, LSEC do not undergo maturation upon stimulation with proinflammatory mediators,157 which suggests that antigen presentation by these cells always leads to the induction of T-cell tolerance. Hepatic dendritic cells and LSEC may thus constitute a functional framework of local antigen-presenting cells, where both cell populations induce tolerance towards soluble bloodborne antigens in the absence of inflammatory signals, but only hepatic dendritic cells upon sensing of ‘danger’ will contribute to the induction of specific T-cell immunity. The absence of functional plasticity in LSEC may operate to protect hepatocytes from inadvertent immune responses by constantly inducing T-cell tolerance at the sinusoidal level. Achieving immunity in the liver involves several cell populations. Bacteria are not eliminated simply by phagocytic uptake through Kupffer cells, but require the recruitment of neutrophils to Kupffer cells.167 Furthermore, hepatic dendritic cells interact with NKT cells
157
Section I. Pathophysiology of the Liver
in order to elicit sustained release of mediators promoting the development of specific T-cell immunity.168 Complex signaling involving chemokines and cytokines between sinusoidal cell populations and passenger leukocytes, leading to hepatic recruitment of immune cells, is operative in protective immune responses against viruses infecting the liver.169,170 Furthermore, CD1d-restricted NKT cells recognize bacterial antigens on antigen-presenting cells that have ingested infectious microorganisms and release mediators promoting pathogen-specific immune responses.171 But even IL-12 released from pathogen-activated dendritic cells is sufficient to activate NKT cells in the vicinity for the sustained release of soluble mediators.172 CD1d-restricted NKT cells recognize lipid autoantigens,173 and in a so-far uncharacterized fashion may contribute to local hepatic immune tolerance. Skewing of the microenvironment towards inflammation by local release of mediators such as IL-12 from dendritic cells or Kupffer cells may represent a pivotal point in deciding whether immune tolerance is locally maintained or whether immunity is induced.172 On the other hand, interaction of dendritic cells with distinct CD1 reactive T cells leads to maturation of dendritic cells, and may therefore contribute to early polarization of the immune response into immunity or immune tolerance.173 The continuous presence of a large population of lymphocytes in the liver may thus ensure efficient immune control, but at the same time lymphocytes must be actively kept in a tolerant state during physiological situations. In this respect it is worth mentioning that immune tolerance in T cells is induced through molecular mechanisms, which suggests a temporary nature to immune tolerance.174 Immune tolerance in the liver must then be regarded as a labile equilibrium requiring constant induction of T-cell tolerance through local antigen-presenting cell populations. Failure to deliver these tolerogenic signals is assumed to result in breakdown of hepatic tolerance and local induction of immunity. Antigen presentation by hepatic antigen-presenting cells not only has consequences for local control of immune responses in the liver, but even influences systemic immunity. Induction of antigen-specific T-cell tolerance by LSEC leads to outgrowth of tumor cells expressing the antigen initially presented by LSEC to T cells.157 Adoptive transfer of hepatic dendritic cells equally modulates systemic immune responses, rather favoring the Th2 type of response.100,104 However, it is not only the type and the functional status of the antigen-presenting cell itself that determines the outcome of immune responses. The decision whether to mount immunity or tolerance towards an antigen appears to depend on the anatomic location where the antigen is first encountered by the immune system. First encounter with the antigen in the liver leads to the induction of specific T-cell tolerance, whereas first encounter in lymphatic tissue gives rise to strong immunity.164 If antigen is sequentially encountered first in lymphatic tissue and then in the liver, T cells are not tolerized but mediate strong immunity, leading to hepatic tissue damage.164 It is intriguing to speculate that such mechanisms may be operative in the immunopathogenesis of persistent liver infection with viruses that show a strong hepatotropism and therefore fail to be presented to the immune system first in lymphatic tissue. However, so far there are no definitive experimental data supporting the notion that local hepatic immune control is responsible for deviating immune responses towards infectious microorganisms or hematogenously metastasizing tumor cells.
158
The liver appears not only to promote tolerance towards soluble antigens, but under certain conditions even promotes local immunity at peripheral sites. Using a model of skin contact sensitivity it was demonstrated that the liver is involved in mediating the recruitment of antigen-specific T cells to the site of initial antigen exposure, i.e. the skin. Minutes after application to the skin, antigen is found to stimulate NKT cells in the liver to release IL-4, which in turn stimulates peritoneal B1 B cells to express IgM antibodies.165 Circulating IgM antibodies form complexes with the antigen at the site of initial antigen application, and trigger local complement activation and the release of vasoactive mediators that finally lead to Tcell recruitment.175 This example nicely illustrates the extraordinary position of the liver in scavenging antigens from the systemic circulation, and its function as a sentinel organ to detect the presence of foreign antigens. Central to this complex sentinel function is the concomitant presence of scavenger activity and sufficient numbers of lymphocytes that can engage in cognate interaction with local scavenger cells. In addition to the immune-regulatory function of the liver towards bloodborne soluble antigens, and thereby control of systemic immune responses, immune tolerance needs to be achieved towards antigens endogenously expressed by hepatic parenchymal cells. In contrast to the conventional immunologic view that initiation of immune responses is restricted to lymphatic tissue, a large body of experimental evidence suggests that hepatic cell populations induce immune tolerance towards endogenously produced antigens. Hepatocytes are sufficient in presenting antigen to T cells and eliciting T-cell tolerance through clonal elimination.137,164 Biliary epithelial cells function as antigen-presenting cells and modulate T-cell responses towards endogenous antigens.176 Such a regulatory function of biliary epithelial cells and hepatocytes suggests that immune responses towards tissue antigens are further modified once T cells have gained access to the tissue. Indeed, other studies have clearly demonstrated that parenchymal cells induce tolerance under steady-state circumstances that allow tissue access to lymphocytes.158,177–179 In conclusion, the liver not only serves as a target for the immune response during infection with microorganisms, but appears to be constantly involved in the modulation of local as well as systemic immune responses towards bloodborne antigens. Central to systemic modulation of immune responses is the scavenger activity of hepatic antigen-presenting cell populations and their unique functional immune phenotype: antigen-presenting organ-resident LSEC and hepatic dendritic cells mediate the induction of T-cell tolerance rather than immunity. Parenchymal liver cells are equally involved in attenuating immune responses towards tissue-specific antigens, thus preventing autoimmune reactivity, but these cells do not engage in the induction of tolerance towards soluble antigens as they lack the ability to present soluble exogenous antigens to T cells. The liver further contributes to the induction of immunity. Phagocytic hepatic cell populations eliminate pathogens in cooperation with neutrophils. Following activation by pathogenic microorganisms hepatic dendritic cells are activated, and in this activated functional state sensitively induce pathogen-specific immunity. Moreover, the large population of unconventional T cells in the liver contributes to early signaling and sustained release of mediators following specific recognition of antigens scavenged by the various antigen-presenting cell
Chapter 9 THE LIVER AND THE IMMUNE SYSTEM
populations of the liver, and thus contribute to the decision as to whether to mount immunity or immune tolerance towards bloodborne antigens.
REFERENCES 1. Shortman K, Liu YJ. Mouse and human dendritic cell subtypes. Nature Rev Immunol 2002;2:151. 2. Rescigno M, Urbano M, Valzasina B, et al. Dendritic cells express tight junction proteins and penetrate gut epithelial monolayers to sample bacteria. Nature Immunol 2001;2:361. 3. Niess JH, Brand S, Gu X, et al. CX3CR1-mediated dendritic cell access to the intestinal lumen and bacterial clearance. Science 2005;307:254. 4. Allan RS, Smith CM, Belz GT, et al. Epidermal viral immunity induced by CD8alpha+ dendritic cells but not by Langerhans cells. Science 2003;301:1925. 5. Gutgemann I, Fahrer AM, Altman JD, et al. Induction of rapid T cell activation and tolerance by systemic presentation of an orally administered antigen. Immunity 1998;8:667. 6. Sixt M, Kanazawa N, Selg M, et al. The conduit system transports soluble antigens from the afferent lymph to resident dendritic cells in the T cell area of the lymph node. Immunity 2005;22:19. 7. Steinman RM, Hawiger D, Liu K, et al. Dendritic cell function in vivo during the steady state: a role in peripheral tolerance. Ann NY Acad Sci 2003;987:15. 8. Probst HC, McCoy K, Okazaki T, et al. Resting dendritic cells induce peripheral CD8(+) T cell tolerance through PD-1 and CTLA-4. Nature Immunol 2005;6:280–286. 9. Hoebe K, Janssen E, Beutler B, et al. The interface between innate and adaptive immunity. Nature Immunol 2004;5:971. 10. Albert ML, Jegathesan M, Darnell RB. Dendritic cell maturation is required for the cross-tolerization of CD8+ T cells. Nature Immunol 2001;2:1010. 11. Janssen EM, Lemmens EE, Wolfe T, et al. CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes. Nature 2003;421:852. 12. Lumsden AB, Henderson JM, Kutner MH. Endotoxin levels measured by a chromogenic assay in portal, hepatic and peripheral venous blood in patients with cirrhosis. Hepatology 1988;8:232. 13. Blouin A, Bolender RP, Weibel ER. Distribution of organelles and membranes between hepatocytes and nonhepatocytes in the rat liver parenchyma. A stereological study. J Cell Biol 1977;72:441. 14. Wake K, Decker K, Kirn A, et al. Cell biology and kinetics of Kupffer cells in the liver. Int Rev Cytol 1989;118:173. 15. Gale RP, Sparkes RS, Golde DW. Bone marrow origin of hepatic macrophages (Kupffer cells) in humans. Science 1978;201:937. 16. Bouwens L, Baekeland M, De Zanger R, et al. Quantitation, tissue distribution and proliferation kinetics of Kupffer cells in normal rat liver. Hepatology 1986;6:718. 17. Yamada M, Naito M, Takahashi K. Kupffer cell proliferation and glucan-induced granuloma formation in mice depleted of blood monocytes by strontium-89. J Leukocyte Biol 1990;47:195. 18. Yamamoto T, Naito M, Moriyama H, et al. Repopulation of murine Kupffer cells after intravenous administration of liposome-encapsulated dichloromethylene diphosphonate. Am J Pathol 1996;149:1271. 19. van Oosten M, van Amersfoort ES, van Berkel TJ, et al. Scavenger receptor-like receptors for the binding of lipopolysaccharide and lipoteichoic acid to liver endothelial and Kupffer cells. J Endotoxin Res 2001;7:381. 20. Roh MS, Wang L, Oyedeji C, et al. Human Kupffer cells are cytotoxic against human colon adenocarcinoma. Surgery 1990;108:400.
21. Falasca L, Bergamini A, Serafino A, et al. Human Kupffer cell recognition and phagocytosis of apoptotic peripheral blood lymphocytes. Exp Cell Res 1996;224:152. 22. van Oosten M, van de Bilt E, van Berkel TJ, et al. New scavenger receptor-like receptors for the binding of lipopolysaccharide to liver endothelial and Kupffer cells. Infect Immun 1998;66:5107. 23. Bykov I, Ylipaasto P, Eerola L, et al. Phagocytosis and LPSstimulated production of cytokines and prostaglandin E2 is different in Kupffer cells isolated from the periportal or perivenous liver region. Scand J Gastroenterol 2003;38:1256. 24. Sleyster EC, Knook DL. Relation between localization and function of rat liver Kupffer cells. Lab Invest 1982;47:484. 25. Itoh Y, Okanoue T, Morimoto M, et al. Functional heterogeneity of rat liver macrophages: interleukin-1 secretion and Ia antigen expression in contrast with phagocytic activity. Liver 1992;12:26. 26. Rogoff TM, Lipsky PE. Antigen presentation by isolated guinea pig Kupffer cells. J Immunol 1980;124:1740. 27. Richman LK, Klingenstein RJ, Richman JA, et al. The murine Kupffer cell. I. Characterization of the cell serving accessory function in antigen-specific T cell proliferation. J Immunol 1979;123:2602. 28. Rubinstein D, Roska AK, Lipsky PE. Liver sinusoidal lining cells express class II major histocompatibility antigens but are poor stimulators of fresh allogeneic T lymphocytes. J Immunol 1986;137:1803. 29. Rubinstein D, Roska AK, Lipsky PE. Antigen presentation by liver sinusoidal lining cells after antigen exposure in vivo. J Immunol 1987;138:1377. 30. Winwood PJ, Arthur MJ. Kupffer cells: their activation and role in animal models of liver injury and human liver disease. Semin Liver Dis 1993;13:50. 31. Laskin DL, Weinberger B, Laskin JD. Functional heterogeneity in liver and lung macrophages. J Leukocyte Biol 2001;70:163. 32. Lichtman SN, Wang J, Lemasters JJ. LPS receptor CD14 participates in release of TNF-alpha in RAW 264.7 and peritoneal cells but not in Kupffer cells. Am J Physiol 1998;275:G39. 33. Callery MP, Mangino MJ, Flye MW. A biologic basis for limited Kupffer cell reactivity to portal-derived endotoxin. Surgery 1991;110:221. 34. Lepay DA, Nathan CF, Steinman RM, et al. Murine Kupffer cells. Mononuclear phagocytes deficient in the generation of reactive oxygen intermediates. J Exp Med 1985;161:1079. 35. Hafenrichter DG, Roland CR, Mangino MJ, et al. The Kupffer cell in endotoxin tolerance: mechanisms of protection against lethal endotoxemia. Shock 1994;2:251. 36. Knolle P, Schlaak J, Uhrig A, et al. Human Kupffer cells secrete IL-10 in response to lipopolysaccharide (LPS) challenge. J Hepatol 1995;22:226. 37. Knolle PA, Loser E, Protzer U, et al. Regulation of endotoxininduced IL-6 production in liver sinusoidal endothelial cells and Kupffer cells by IL-10. Clin Exp Immunol 199;107:555. 38. van der Poll T, Coyle SM, Barbosa K, et al. Epinephrine inhibits tumor necrosis factor-alpha and potentiates interleukin 10 production during human endotoxemia. J Clin Invest 1996;97:713. 39. Neuhuber WL, Tiegs G. Innervation of immune cells: evidence for neuroimmunomodulation in the liver. Nature Rec 2004;280A:884. 40. Vazquez-Torres A, Vallance BA, Bergman MA, et al. Toll-like receptor 4 dependence of innate and adaptive immunity to Salmonella: importance of the Kupffer cell network. J Immunol 2004;172:6202. 41. Schumann J, Muhlen K, Kiemer AK, et al. Parenchymal, but not leukocyte, TNF receptor 2 mediates T cell-dependent hepatitis in mice. J Immunol 2003;170:2129.
159
Section I. Pathophysiology of the Liver
42. Decker K. Biologically active products of stimulated liver macrophages (Kupffer cells). Eur J Biochem 1990;192:245. 43. Shirahama M, Ishibashi H, Tsuchiya Y, et al. Kinetics and parameters of the induction of interleukin 1 secretion by rat Kupffer cells. J Clin Lab Immunol 1988;27:127. 44. Shirahama M, Ishibashi H, Tsuchiya Y, et al. Kupffer cells may autoregulate interleukin 1 production by producing interleukin 1 inhibitor and prostaglandin E2. Scand J Immunol 1988;28:719. 45. Busam KJ, Bauer TM, Bauer J, et al. Interleukin-6 release by rat liver macrophages. J Hepatol 1990;11:367. 46. Takahashi M, Ogasawara K, Takeda K, et al. LPS induces NK1.1+ alpha beta T cells with potent cytotoxicity in the liver of mice via production of IL-12 from Kupffer cells. J Immunol 1996;156:2436. 47. Okamura H, Tsutsi H, Komatsu T, et al. Cloning of a new cytokine that induces IFN-gamma production by T cells. Nature 1995;378:88. 48. Karck U, Peters T, Decker K. The release of tumor necrosis factor from endotoxin-stimulated rat Kupffer cells is regulated by prostaglandin E2 and dexamethasone. J Hepatol 1988;7:352. 49. Bissell DM, Wang SS, Jarnagin WR, et al. Cell-specific expression of transforming growth factor-beta in rat liver. Evidence for autocrine regulation of hepatocyte proliferation. J Clin Invest 1995;96:447. 50. Meyer DH, Bachem MG, Gressner AM. Modulation of hepatic lipocyte proteoglycan synthesis and proliferation by Kupffer cellderived transforming growth factors type beta 1 and type alpha. Biochem Biophys Res Commun 1990;171:1122. 51. Bautista AP, Spitzer JJ. Superoxide anion generation by in situ perfused rat liver: effect of in vivo endotoxin. Am J Physiol 1990;259:G907. 52. Ahmad N, Chen LC, Gordon MA, et al. Regulation of cyclooxygenase-2 by nitric oxide in activated hepatic macrophages during acute endotoxemia. J Leukocyte Biol 2002;71:1005. 53. van Oosten M, van de Bilt E, de Vries HE, et al. Vascular adhesion molecule-1 and intercellular adhesion molecule-1 expression on rat liver cells after lipopolysaccharide administration in vivo. Hepatology 1995;22:1538. 54. Matsuura K, Ishida T, Setoguchi M, et al. Upregulation of mouse CD14 expression in Kupffer cells by lipopolysaccharide. J Exp Med 1994;179:1671. 55. Myers KJ, Eppihimer MJ, Hall L, et al. Interleukin-12-induced adhesion molecule expression in murine liver. Am J Pathol 1998;152:457. 56. Burgio VL, Ballardini G, Artini M, et al. Expression of costimulatory molecules by Kupffer cells in chronic hepatitis of hepatitis C virus etiology. Hepatology 1998;27:1600. 57. Lohse AW, Knolle PA, Bilo K, et al. Antigen-presenting function and B7 expression of murine sinusoidal endothelial cells and Kupffer cells. Gastroenterology 1996;110:1175. 58. Kronenberg M, Gapin L. The unconventional lifestyle of NKT cells. Nature Rev Immunol 2002;2:557. 59. Emoto M, Kaufmann SH. Liver NKT cells: an account of heterogeneity. Trends Immunol 2003;24:364. 60. Wu D, Xing GW, Poles MA, et al. Bacterial glycolipids and analogs as antigens for CD1d-restricted NKT cells. Proc Natl Acad Sci USA 2005;102:1351. 61. Sidobre S, Hammond KJ, Benazet-Sidobre L, et al. The T cell antigen receptor expressed by Valpha14i NKT cells has a unique mode of glycosphingolipid antigen recognition. Proc Natl Acad Sci USA 2004;101:12254. 62. Ohteki T, Maki C, Koyasu S, et al. Cutting edge: LFA-1 is required for liver NK1.1+TCR alpha beta+ cell development: evidence that liver NK1.1+TCR alpha beta+ cells originate from multiple pathways. J Immunol 1999;162:3753. 63. Emoto M, Mittrucker HW, Schmits R, et al. Critical role of leukocyte function-associated antigen-1 in liver accumulation of CD4+NKT cells. J Immunol 1999;162:5094.
160
64. Emoto M, Miyamoto M, Namba K, et al. Participation of leukocyte function-associated antigen-1 and NK cells in the homing of thymic CD8+NKT cells to the liver. Eur J Immunol 2000;30:3049. 65. Vanderkerken K, Bouwens L, Van Rooijen N, et al. The role of Kupffer cells in the differentiation process of hepatic natural killer cells. Hepatology 1995;22:283. 66. Matsuno K, Nomiyama H, Yoneyama H, et al. Kupffer cellmediated recruitment of dendritic cells to the liver crucial for a host defense. Dev Immunol 2002;9:143. 67. Wisse E, De Zanger RB, Charels K, et al. The liver sieve: considerations concerning the structure and function of endothelial fenestrae, the sinusoidal wall and the space of Disse. Hepatology 1985;5:683. 68. Oda M, Han JY, Yokomori H. Local regulators of hepatic sinusoidal microcirculation: recent advances. Clin Hemorheol Microcirc 2000;23:85. 69. Steffan AM, Gendrault JL, McCuskey RS, et al. Phagocytosis, an unrecognized property of murine endothelial liver cells. Hepatology 1986;6:830. 70. Smedsrod B. Clearance function of scavenger endothelial cells. Comp Hepatol 2004;3:S22. 71. Tavassoli M, Kishimoto T, Soda R, et al. Liver endothelium mediates the uptake of iron–transferrin complex by hepatocytes. Exp Cell Res 1986;165:369. 72. Tavassoli M, Kishimoto T, Kataoka M. Liver endothelium mediates the hepatocyte’s uptake of ceruloplasmin. J Cell Biol 1986;102:1298. 73. Breiner K, Schaller H, Knolle P. Endothelial cell-mediated uptake of a hepatitis B virus—a new concept of liver-targeting of hepatotropic micro-organisms. Hepatology 2001;34:803–808. 74. Cormier EG, Durso RJ, Tsamis F, et al. L-SIGN (CD209L) and DC-SIGN (CD209) mediate transinfection of liver cells by hepatitis C virus. Proc Natl Acad Sci USA 2004;101:14067. 75. Geijtenbeek TB, Kwon DS, Torensma R, et al. DC-SIGN, a dendritic cell-specific HIV-1-binding protein that enhances trans-infection of T cells. Cell 2000;100:587. 76. Smedsrod B, Melkko J, Araki N, et al. Advanced glycation end products are eliminated by scavenger-receptor-mediated endocytosis in hepatic sinusoidal Kupffer and endothelial cells. Biochem J 1997;322:567. 77. Smedsrod B, Pertoft H, Eriksson S, et al. Studies in vitro on the uptake and degradation of sodium hyaluronate in rat liver endothelial cells. Biochem J 1984;223:617. 78. Politz O, Gratchev A, McCourt PA, et al. Stabilin-1 and -2 constitute a novel family of fasciclin-like hyaluronan receptor homologues. Biochem J 2002;362:155. 79. Van Berkel TJ, De Rijke YB, Kruijt JK. Different fate in vivo of oxidatively modified low density lipoprotein and acetylated low density lipoprotein in rats. Recognition by various scavenger receptors on Kupffer and endothelial liver cells. J Biol Chem 1991;266:2282. 80. Smedsrod B, Pertoft H, Gustafson S, et al. Scavenger functions of the liver endothelial cell. Biochem J 1990;266:313. 81. Melkko J, Hellevik T, Risteli L, et al. Clearance of NH2terminal propeptides of types I and III procollagen is a physiological function of the scavenger receptor in liver endothelial cells. J Exp Med 1994;179:405. 82. Skogh T, Blomhoff R, Eskild W, et al. Hepatic uptake of circulating IgG immune complexes. Immunology 1985;55:585. 83. Johansson AG, Sundqvist T, Skogh T. IgG immune complex binding to and activation of liver cells. An in vitro study with IgG immune complexes, Kupffer cells, sinusoidal endothelial cells and hepatocytes. Int Arch Allergy Immunol 2000;121:329. 84. Fraser R, Clark SA, Day WA, et al. Nicotine decreases the porosity of the rat liver sieve: a possible mechanism for hypercholesterolaemia. Br J Exp Pathol 1988;69:345.
Chapter 9 THE LIVER AND THE IMMUNE SYSTEM
85. Sanan DA, Fan J, Bensadoun A, et al. Hepatic lipase is abundant on both hepatocyte and endothelial cell surfaces in the liver. J Lipid Res 1997;38:1002. 86. Fraser R, Dobbs BR, Rogers GW. Lipoproteins and the liver sieve: the role of the fenestrated sinusoidal endothelium in lipoprotein metabolism, atherosclerosis, and cirrhosis. Hepatology 1995;21:863. 87. Lievens J, Snoeys J, Vekemans K, et al. The size of sinusoidal fenestrae is a critical determinant of hepatocyte transduction after adenoviral gene transfer. Gene Ther 2004;11:1523. 88. Kempka G, Kolb-Bachofen V. Binding, uptake, and transcytosis of ligands for mannose-specific receptors in rat liver: an electron microscopic study. Exp Cell Res 1988;176:38. 89. McNab G, Reeves JL, Salmi M, et al. Vascular adhesion protein 1 mediates binding of T cells to human hepatic endothelium. Gastroenterology 1996;110:522. 90. Bashirova AA, Geijtenbeek TB, van Duijnhoven GC, et al. A dendritic cell-specific intercellular adhesion molecule 3-grabbing nonintegrin (DC-SIGN)-related protein is highly expressed on human liver sinusoidal endothelial cells and promotes HIV-1 infection. J Exp Med 2001;193:671. 91. Steinman RM, Hawiger D, Nussenzweig MC. Tolerogenic dendritic cells. Annu Rev Immunol 2003;21:685. 92. Lutz MB, Schuler G. Immature, semi-mature and fully mature dendritic cells: which signals induce tolerance or immunity? Trends Immunol 2002;23:445. 93. Matsuno K, Ezaki T, Kudo S, et al. A life stage of particle-laden rat dendritic cells in vivo: their terminal division, active phagocytosis, and translocation from the liver to the draining lymph [see comments]. J Exp Med 1996;183:1865. 94. Kudo S, Matsuno K, Ezaki T, et al. A novel migration pathway for rat dendritic cells from the blood: hepatic sinusoids–lymph translocation. J Exp Med 1997;185:777. 95. Sato T, Yamamoto H, Sasaki C, et al. Maturation of rat dendritic cells during intrahepatic translocation evaluated using monoclonal antibodies and electron microscopy. Cell Tissue Res 1998;294:503. 96. Steptoe RJ, Patel RK, Subbotin VM, et al. Comparative analysis of dendritic cell density and total number in commonly transplanted organs: morphometric estimation in normal mice. Transplant Immunol 2000;8:49. 97. Pillarisetty VG, Shah AB, Miller G, et al. Liver dendritic cells are less immunogenic than spleen dendritic cells because of differences in subtype composition. J Immunol 2004;172:1009. 98. Lau AH, Thomson AW. Dendritic cells and immune regulation in the liver. Gut 2003;52:307. 99. Jomantaite I, Dikopoulos N, Kroger A, et al. Hepatic dendritic cell subsets in the mouse. Eur J Immunol 2004;34:355. 100. Khanna A, Morelli AE, Zhong C, et al. Effects of liver-derived dendritic cell progenitors on Th1- and Th2-like cytokine responses in vitro and in vivo. J Immunol 2000;164:1346. 101. Lu L, Bonham CA, Liang X, et al. Liver-derived DEC205+B220+CD19- dendritic cells regulate T cell responses. J Immunol 2001;166:7042. 102. Goddard S, Youster J, Morgan E, et al. Interleukin-10 secretion differentiates dendritic cells from human liver and skin. Am J Pathol 2004;164:511. 103. Qian S, Lu L, Fu F, et al. Donor pretreatment with Flt-3 ligand augments antidonor cytotoxic T lymphocyte, natural killer, and lymphokine-activated killer cell activities within liver allografts and alters the pattern of intragraft apoptotic activity. Transplantation 1998;65:1590. 104. De Creus A, Abe M, Lau AH, et al. Low TLR4 expression by liver dendritic cells correlates with reduced capacity to activate allogeneic T cells in response to endotoxin. J Immunol 2005;174:2037. 105. Grant AJ, Goddard S, Ahmed-Choudhury J, et al. Hepatic expression of secondary lymphoid chemokine (CCL21)
106. 107.
108.
109. 110.
111. 112.
113.
114.
115.
116.
117.
118.
119. 120. 121.
122.
123. 124.
125. 126. 127.
promotes the development of portal-associated lymphoid tissue in chronic inflammatory liver disease. Am J Pathol 2002;160:1445. Cyster JG. Chemokines and the homing of dendritic cells to the T cell areas of lymphoid organs. J Exp Med 1999;189:447. Forster R, Schubel A, Breitfeld D, et al. CCR7 coordinates the primary immune response by establishing functional microenvironments in secondary lymphoid organs. Cell 1999;99:23. Tanis W, Mancham S, Binda R, et al. Human hepatic lymph nodes contain normal numbers of mature myeloid dendritic cells but few plasmacytoid dendritic cells. Clin Immunol 2004;110:81. Friedman SL. Hepatic stellate cells. Prog Liver Dis 1996;14:101. Reynaert H, Thompson MG, Thomas T, et al. Hepatic stellate cells: role in microcirculation and pathophysiology of portal hypertension. Gut 2002;50:571. Maher JJ. Interactions between hepatic stellate cells and the immune system. Semin Liver Dis 2001;21:417. Vinas O, Bataller R, Sancho-Bru P, et al. Human hepatic stellate cells show features of antigen-presenting cells and stimulate lymphocyte proliferation. Hepatology 2003;38:919. Paik YH, Schwabe RF, Bataller R, et al. Toll-like receptor 4 mediates inflammatory signaling by bacterial lipopolysaccharide in human hepatic stellate cells. Hepatology 2003;37:1043. Wang SC, Ohata M, Schrum L, et al. Expression of interleukin10 by in vitro and in vivo activated hepatic stellate cells. J Biol Chem 1998;273:302. Breitkopf K, Lahme B, Tag CG, et al. Expression and matrix deposition of latent transforming growth factor beta binding proteins in normal and fibrotic rat liver and transdifferentiating hepatic stellate cells in culture. Hepatology 2001;33:387. Hellerbrand S, Wang C, Tsukamoto H, et al. Expression of intracellular adhesion molecule 1 by activated hepatic stellate cells. Hepatology 1996;24:670. Knittel T, Dinter C, Kobold D, et al. Expression and regulation of cell adhesion molecules by hepatic stellate cells (HSC) of rat liver: involvement of HSC in recruitment of inflammatory cells during hepatic tissue repair. Am J Pathol 1999;154:153. Knittel T, Aurisch S, Neubauer K, et al. Cell-type-specific expression of neural cell adhesion molecule (N-CAM) in Ito cells of rat liver. Up-regulation during in vitro activation and in hepatic tissue repair. Am J Pathol 1996;149:449. Knolle PA, Gerken G. Local control of the immune response in the liver. Immunol Rev 2000;174:21. McCuskey RS, Reilly FD. Hepatic microvasculature: dynamic structure and its regulation. Semin Liver Dis 1993;13:1. Essani NA, McGuire GM, Manning AM, et al. Endotoxininduced activation of the nuclear transcription factor kappa B and expression of E-selectin messenger RNA in hepatocytes, Kupffer cells, and endothelial cells in vivo. J Immunol 1996;156:2956. Wong J, Johnston B, Lee SS, et al. A minimal role for selectins in the recruitment of leukocytes into the inflamed liver microvasculature. J Clin Invest 1997;99:2782. MacPhee PJ, Schmidt EE, Groom AC. Intermittence of blood flow in liver sinusoids, studied by high-resolution in vivo microscopy. Am J Physiol 1995;269:G692. MacPhee PJ, Schmidt EE, Groom AC. Evidence for Kupffer cell migration along liver sinusoids, from high-resolution in vivo microscopy. Am J Physiol 1992;263:G17. Calne RY. Induction of immunological tolerance by porcine liver allografts. Nature 1969;223:472. Cantor H, Dumont A. Hepatic suppression of sensitization to antigen absorbed into the portal system. Nature 1967;215:744. Dahmen U, Qian S, Rao AS, et al. Split tolerance induced by orthotopic liver transplantation in mice. Transplantation 1994;58:1.
161
Section I. Pathophysiology of the Liver
128. Barker CF, Corriere JN Jr. Canine renal homotransplantation with venous drainage via the portal vein. Ann Surg 1967;165:279. 129. Boeckx W, Sobis H, Lacquet A, et al. Prolongation of allogeneic heart graft survival in the rat after implantation on portal vein. Transplantation 1975;19:145. 130. Gorczynski RM, Chan Z, Chung S, et al. Prolongation of rat small bowel or renal allograft survival by pretransplant transfusion and/or by varying the route of allograft venous drainage. Transplantation 1994;58:816. 131. May AG, Bauer S, Leddy JP, et al. Survival of allografts after hepatic portal venous administration of specific transplantation antigen. Ann Surg 1969;170:824. 132. Jacob AI, Goldberg PK, Bloom N, et al. Endotoxin and bacteria in portal blood. Gastroenterology 1977;72:1268. 133. Rieder H, Ramadori G, Allmann KH, et al. Prostanoid release of cultured liver sinusoidal endothelial cells in response to endotoxin and tumor necrosis factor. Comparison with umbilical vein endothelial cells. J Hepatol 1990;11:359. 134. Kuiper J, Zijlstra FJ, Kamps JA, et al. Identification of prostaglandin D2 as the major eicosanoid from liver endothelial and Kupffer cells. Biochim Biophys Acta 1988;959:143. 135. Steinbrink K, Wolfl M, Jonuleit H, et al. Induction of tolerance by IL-10-treated dendritic cells. J Immunol 1997;59:4772. 136. Crispe IN. Hepatic T cells and liver tolerance. Nature Rev Immunol 2003;3:51–62. 137. Bertolino P, Bowen DG, McCaughan GW, et al. Antigen-specific primary activation of CD8+ T cells within the liver. J Immunol 2001;166:5430. 138. Gorczynski L, Chen Z, Hu J, et al. Evidence that an OX-2positive cell can inhibit the stimulation of type 1 cytokine production by bone marrow-derived B7-1 (and B7-2)-positive dendritic cells. J Immunol 1999;162:774. 139. Trobonjaca Z, Kroger A, Stober D, et al. Activating immunity in the liver. II, IFN-beta attenuates NK cell-dependent liver injury triggered by liver NKT cell activation. J Immunol 2002;168:3763. 140. Dikopoulos N, Bertoletti A, Kroger A, et al. Type II,FN negatively regulates CD8+ T cell responses through IL-10producing CD4+ T regulatory 1 cells. J Immunol 2005; 174:99. 141. Huang L, Soldevila G, Leeker M, et al. The liver eliminates T cells undergoing antigen-triggered apoptosis in vivo. Immunity 1994;1:741. 142. Mehal WZ, Azzaroli F, Crispe IN. Antigen presentation by liver cells controls intrahepatic T cell trapping, whereas bone marrowderived cells preferentially promote intrahepatic T cell apoptosis. J Immunol 2001;167:667. 143. Mehal WZ, Juedes AE, Crispe IN. Selective retention of activated CD8+ T cells by the normal liver. J Immunol 1999;163:3202. 144. John B, Crispe IN. Passive and active mechanisms trap activated CD8+ T cells in the liver. J Immunol 2004;172:5222. 145. Knolle PA, Schmitt E, Jin S, et al. Induction of cytokine production in naïve CD4(+) T cells by antigen-presenting murine liver sinusoidal endothelial cells but failure to induce differentiation toward Th1 cells. Gastroenterology 1999;116:1428. 146. Cella M, Sallusto F, Lanzavecchia A. Origin, maturation and antigen presenting function of dendritic cells. Curr Opin Immunol 1997;9:10. 147. Cunningham AC, Zhang JG, Moy JV, et al. A comparison of the antigen-presenting capabilities of class IIM,HC-expressing human lung epithelial and endothelial cells. Immunology 1997;91:458. 148. Haraldsen G, Sollid LM, Bakke O, et al. Major histocompatibility complex class II-dependent antigen presentation by human intestinal endothelial cells. Gastroenterology 1998;114:649.
162
149. Marelli-Berg FM, Hargreaves RE, Carmichael P, et al. Major histocompatibility complex class II-expressing endothelial cells induce allospecific nonresponsiveness in naïve T cells. J Exp Med 1996;183:1603. 150. Rehermann B. Immunopathogenesis of viral hepatitis. Baillières Clin Gastroenterol 1996;10:483. 151. Ma W, Pober JS. Human endothelial cells effectively costimulate cytokine production by, but not differentiation of, naïve CD4+ T cells. J Immunol 1998;161:2158. 152. Gorczynski RM. Adoptive transfer of unresponsiveness to allogeneic skin grafts with hepatic gamma delta + T cells. Immunology 1994;81:27. 153. Bevan MJ. Interaction antigens detected by cytotoxic T cells with the major histocompatibility complex as modifier. Nature 1975;256:419. 154. Heath WR, Kurts C, Miller JF, et al. Cross-tolerance: a pathway for inducing tolerance to peripheral tissue antigens. J Exp Med 1998;187:1549. 155. Kurts C. Cross-presentation: inducing CD8 T cell immunity and tolerance. J Mol Med 2000;78:326. 156. Sigal LJ, Crotty S, Andino R, et al. Cytotoxic T-cell immunity to virus-infected non-haematopoietic cells requires presentation of exogenous antigen. Nature 1999;398:77. 157. Limmer A, Ohl J, Kurts C, et al. Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T-cell tolerance. Nature Med 2000;6:1348. 158. Limmer A, Sacher T, Alferink J, et al. Failure to induce organspecific autoimmunity by breaking of tolerance: importance of the microenvironment. Eur J Immunol 1998;28:2395. 159. Ando K, Guidotti LG, Cerny A, et al. CTL access to tissue antigen is restricted in vivo. J Immunol 1994;153:482. 160. Groux H, O’Garra A, Bigler M, et al. AC,D4+ T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 1997;389:737. 161. Jonuleit H, Schmitt E, Steinbrink K, et al. Dendritic cells as a tool to induce anergic and regulatory T cells. Trends Immunol 2001;22:394. 162. Jacobs MJ, van den Hoek AE, van de Putte LB, et al. Anergy of antigen-specific T lymphocytes is a potent mechanism of intravenously induced tolerance. Immunology 1994;82:294. 163. Liblau RS, Tisch R, Shokat K, et al. Intravenous injection of soluble antigen induces thymic and peripheral T-cell apoptosis. Proc Natl Acad Sci USA 1996;93:3031. 164. Bowen DG, Zen M, Holz L, et al. The site of primary T cell activation is a determinant of the balance between intrahepatic tolerance and immunity. J Clin Invest 2004;114:701. 165. Campos RA, Szczepanik M, Itakura A, et al. Cutaneous immunization rapidly activates liver invariant Valpha14 NKT cells stimulating B-1 B cells to initiate T cell recruitment for elicitation of contact sensitivity. J Exp Med 2003;198:1785. 166. Johansson C, Wick MJ. Liver dendritic cells present bacterial antigens and produce cytokines upon Salmonella encounter. J Immunol 2004;172:2496. 167. Gregory SH, Wing EJ. Neutrophil-Kupffer cell interaction: a critical component of host defenses to systemic bacterial infections. J Leukocyte Biol 2002;72:239. 168. Trobonjaca Z, Leithauser F, Moller P, et al. Activating immunity in the liver. I. Liver dendritic cells (but not hepatocytes) are potent activators of IFN-gamma release by liver NKT cells. J Immunol 2001;167:1413. 169. Salazar-Mather TP, Hamilton TA, Biron CA. A chemokine-tocytokine-to-chemokine cascade critical in antiviral defense. J Clin Invest 2000;105:985. 170. Salazar-Mather TP, Lewis CA, Biron CA. Type I interferons regulate inflammatory cell trafficking and macrophage inflammatory protein 1alpha delivery to the liver. J Clin Invest 2002;110:321.
Chapter 9 THE LIVER AND THE IMMUNE SYSTEM
171. Fischer K, Scotet E, Niemeyer M, et al. Mycobacterial phosphatidylinositol mannoside is a natural antigen for CD1drestricted T cells. Proc Natl Acad Sci USA 2004;101:10685. 172. Brigl M, Bry L, Kent SC, et al. Mechanism of CD1d-restricted natural killer T cell activation during microbial infection. Nature Immunol 2003;4:1230. 173. Vincent MS, Leslie DS, Gumperz JE, et al. CD1-dependent dendritic cell instruction. Nature Immunol 2002;3:1163. 174. Mueller DL. E3 ubiquitin ligases as T cell anergy factors. Nature Immunol 2004;5:883. 175. Askenase PW, Szczepanik M, Itakura A, et al. Extravascular Tcell recruitment requires initiation begun by Valpha14+ NKT cells and B-1 B cells. Trends Immunol 2004;25:441.
176. Kamihira T, Shimoda S, Nakamura M, et al. Biliary epithelial cells regulate autoreactive T cells: implications for biliary-specific diseases. Hepatology 2005;41:151. 177. Tafuri A, Alferink J, Moller P, et al. T cell awareness of paternal alloantigens during pregnancy. Science 1995;270:630. 178. Alferink J, Tafuri A, Vestweber D, et al. Control of neonatal tolerance to tissue antigens by peripheral T cell trafficking. Science 1998;282:1338. 179. Arnold B. Parenchymal cells in immune and tolerance induction. Immunol Lett 2003;89:225.
163
Section I: Pathophysiology of the Liver
10
MOLECULAR PATHOGENESIS OF HEPATOCELLULAR CARCINOMA Jack R. Wands and Darius Moradpour Abbreviations AFB1 aflatoxin B1 FZD7 Frizzled-7 GBV-A GB viruses A Grb2 growth factor receptor-bound protein 2 HBV hepatitis B virus HCC hepatocellular carcinoma HCV hepatitis C virus hTERT human telomerase reverse transcriptase iNOS inducible nitric oxide synthase
IP3R1/ inositol 1,4,5-triphosphate receptor types IP3R2 1 and 2 IRAK2 IL-1R-associated kinase 2 LOH loss of heterozygosity MHBst truncated middle hepatitis B surface antigen NTRK2 neurotropic tyrosine receptor kinase 2 p42MAPK1 p42 mitogen-activated protein kinase 1 PI3K phosphoinositide 3-kinase PKC protein kinase C
INTRODUCTION Hepatocellular carcinoma (HCC) is one of the most common malignant tumors worldwide.1,2 The incidence ranges from fewer than 10 cases per 100 000 population in North America and western Europe to 50–150 cases per 100 000 population in parts of Africa and Asia, where HCC is responsible for a large proportion of cancer deaths. A rise in the incidence of and mortality from HCC, most likely reflecting the increased prevalence of hepatitis C virus (HCV) infection, has recently been observed in most industrialized countries.3,4 The major etiologies of HCC are now well defined and include chronic viral hepatitis B, C and D, toxins and drugs (e.g. alcohol, aflatoxins, anabolic steroids), metabolic liver diseases (e.g. hereditary hemochromatosis, a1-antitrypsin deficiency) and, to an as yet less clearly defined extent, non-alcoholic fatty liver disease. On a global scale, chronic hepatitis B virus (HBV) and HCV infection account for well over 80% of HCCs. Hepatocarcinogenesis is a multistep process involving different genetic alterations that ultimately lead to malignant transformation of the hepatocyte.5–7 However, although significant progress has been made in recognizing the sequence of events involved in other forms of cancer, notably colorectal cancer and certain hematopoietic malignancies, the molecular contribution of the multiple factors and their interactions in hepatocarcinogenesis is still poorly understood. In fact, a picture of HCCs as genetically very heterogenous tumors is emerging. This is not unexpected, given the heterogeneity of etiologic factors implicated in HCC development, the complexity of hepatocyte functions, and the late stage at which HCCs are usually detected and analyzed. As shown in Figure 10-1, malignant transformation of hepatocytes may occur regardless of the
PKR RAR-b SERCA1 SH3 SITA TRUP VEGF WHV
RNA-activated protein kinase retinoic acid receptor-b sarco/endoplasmic reticulum calcium ATPase src homology 3 a-2,3-sialyltransferase thyroid hormone uncoupling protein vascular endothelial growth factor woodchuck hepatitis virus
etiologic agent through a pathway of increased liver cell turnover, induced by chronic liver injury and regeneration in a context of inflammation and oxidative DNA damage. This may result in genetic alterations, such as the activation of cellular oncogenes; the inactivation of tumor suppressor genes, possibly in cooperation with genomic instability, including DNA mismatch repair defects and impaired chromosomal segregation; overexpression of growth and angiogenic factors; and telomerase activation. Chronic viral hepatitis, alcohol, metabolic liver diseases such as hemochromatosis and a1-antitrypsin deficiency, as well as nonalcoholic fatty liver disease, may act predominantly through this pathway of chronic liver injury, regeneration, and cirrhosis. Accordingly, the major clinical risk factor for HCC development is liver cirrhosis, as 70–90% of HCCs develop in a cirrhotic liver. The HCC risk in patients with liver cirrhosis depends on the activity, duration and etiology of the underlying liver disease. It is particularly high in cirrhosis due to chronic viral hepatitis and hemochromatosis, followed, in descending order, by alcoholic cirrhosis, autoimmune hepatitis and primary biliary cirrhosis, and it is low in Wilson’s disease. Coexistence of etiologies, e.g. HBV and HCV co-infection, HBV infection and aflatoxin B1 (AFB1),8 or HCV infection and alcohol, increases the relative risk of HCC development.9 In general, HCCs are two to four times more frequent in males than in females, and the incidence increases with age. On the other hand, there is evidence that HBV – and possibly also HCV – may under certain circumstances play an additional direct role in the molecular pathogenesis of HCC. Moreover, well defined mutations in the p53 tumor suppressor gene induced by aflatoxin exposure are a prime example of how environmental factors contribute to tumor development. Finally, advances in our understanding of the molecular genetics of HCC has led to the identification
165
Section I. Pathophysiology of the Liver
of signal transduction pathways that are activated during hepatic transformation.
HEPATITIS B VIRUS An estimated 300–400 million individuals worldwide are chronically infected with HBV. Prevalence rates range from 0.1–1% of the general population in North America and western Europe to up to 20% in Southeast Asia and parts of Africa. Epidemiologic studies have convincingly shown that HCC is closely associated with chronic HBV infection.10 The incidence of HCC in chronically HBVinfected individuals is approximately 100 times higher than in the uninfected population, and the lifetime HCC risk of males infected at birth is estimated to approach 40%. Importantly, a study from Taiwan demonstrated a decline in the incidence of HCC in children
HCV HBV
Growth factor activation
Alcohol Chronic injury inflammation
Metabolic disorders
Regeneration
Environmental factors (AFB)
Genetic and/ or epigenetic alterations
Hepatocellular carcinoma Figure 10-1. Factors involved in the pathogenesis of hepatocellular carcinoma.
Primary 1 2 3 kb 23 9.8 6.6
166
1
Focus 2
3
1
Nude 2
3
after the implementation of a universal hepatitis B vaccination program.11 A common molecular mechanism for HBV-induced hepatocarcinogenesis has not yet been discovered.12 Most cases of HCC occur after many years of chronic hepatitis, which could provide the mitogenic and mutagenic environment to precipitate random genetic alterations and lead to the development of HCC. In this context, using an elegant transgenic mouse model, Nakamoto et al.13 showed that chronic immune-mediated liver cell injury is sufficient to cause HCC. Inhibition of cytotoxic T lymphocyte-induced apoptosis and chronic inflammation by neutralization of Fas ligand prevented HCC development in this model.14 On the other hand, several lines of evidence support a more direct oncogenic contribution of HBV to HCC development, as summarized below.
HBV DNA INTEGRATION HBV and retroviruses share a replication strategy that includes the reverse transcription of an RNA intermediate. However, in contrast to proviral integration into the host cell genome, HBV DNA integration is not part of the viral life cycle, but rather occurs as an epiphenomenon of HBV replication. Hepadnaviral DNA integration does not preserve the viral genome sequence, thereby rendering it impossible for the viral integrant to function as a template for virus replication. Integrated HBV sequences have been found in the majority of HCCs that develop in patients with chronic HBV infection (Figure 10-2). Expectations that there might be a common target sequence in the cellular DNA led investigators to examine the viral junction and flanking cellular DNA sequences of many different HBV integration sites. These studies have shown that integration was random within human chromosomes. HBV may therefore act as a non-selective insertional mutagenic agent. In addition, secondary chromosomal rearrangements associated with HBV DNA integration, such as duplications, translocations, and deletions, suggest that a major oncogenic effect of HBV DNA integration may be an increased genomic instability. Only a few examples of HBV DNA integration within or near known cellular genes have been documented. In this regard, characterization of a single HBV DNA integration in an early HCC revealed insertion into the retinoic acid receptor-b (RAR-b) gene,
Liver 1 2
3
Figure 10-2. Southern blot analysis of DNA isolated from a HCC (Primary), a cell line derived from the same tumor (Focus), and a Focus tumor grown in a nude mouse (Nude). DNA was digested with three different restriction enzymes, electrophoresed through an agarose gel, transferred to a nylon membrane, and probed with 32P-labeled HBV DNA. Note the stable integration pattern in the primary tumor, the derived cell line, and the nude mouse tumor. No HBV DNA integration was detectable in the non-tumorous liver (Liver).
Chapter 10 MOLECULAR PATHOGENESIS OF HEPATOCELLULAR CARCINOMA
resulting in overexpression of a truncated RAR-b with altered functions.15–17 Investigation of another HBV DNA integration site led to the identification of the human cyclin A gene.18 This integration resulted in an HBV PreS2/S-cyclin A fusion protein with increased stability. Constitutive expression of this stabilized cyclin A protein may have led to or contributed to increased cell proliferation. In another HCC, HBV DNA was found integrated into the gene encoding SERCA1 (sarco/endoplasmic reticulum calcium ATPase), which plays a pivotal role in regulating intracellular calcium levels as a second messenger for cell proliferation and death signals.19,20 Viral integration has also been associated with mutations in key regulatory genes, including neurotropic tyrosine receptor kinase 2 (NTRK2), IL-1R-associated kinase 2 (IRAK2), p42 mitogen-activated protein kinase 1 (p42MAPK1), inositol 1,4,5-triphosphate receptor types 1 and 2 (IP3R1/IP3R2), a-2,3-sialyltransferase (SITA), thyroid hormone uncoupling protein (TRUP), and human telomerase reverse transcriptase (hTERT).21 These genes are involved in cell signaling, proliferation, differentiation and survival. Based on sequence analyses of these and other viral cellular DNA integration sites, it appears that HBV insertional mutagenesis provides both specific and non-specific molecular mechanisms that contribute to hepatocarcinogenesis.
TRANS-ACTIVATION OF CELLULAR GENES BY HBV Mammalian hepadnaviruses contain a gene that can function as a transcriptional trans-activator. It is called the X gene, as its role in the viral life cycle is unknown. The X gene product, referred to as HBx, is a 154 amino acid protein that is essential for productive HBV infection in vivo. It can function as a trans-activator of various cellular genes associated with growth control. This observation has led to the hypothesis that HBx may be involved in the development of HBV-associated HCC.22 A plethora of additional functions, e.g. interference with cell cycle control, DNA mismatch repair and apoptosis, as well as numerous apparently unrelated cellular interaction partners, have been reported for HBx, primarily in heterologous overexpression systems.23 However, the function of HBx in the HBV life cycle and its role, in HCC development remains elusive. The failure of HBx to bind directly to any defined DNA sequences suggests that trans-activation does not involve a known direct DNA sequence-specific interaction. The biologic role of HBx may rather be mediated through an effect on cellular transcription factors. For example, it has been reported that HBx may trans-activate via a signal transduction pathway comprising 1,2-diacylglycerol, protein kinase C (PKC), and the transcription factors AP-1 (Jun-Fos), AP-2, and NF-kB.24 Alternatively, it was postulated that HBx may promote hepatocarcinogenesis through activation of the Ras-RafMAP kinase pathway. A role of HBx in HBV-associated hepatocarcinogenesis was further supported by the observation that certain X gene-transgenic mice develop HCC.25,26 However, other investigators have not found tumors in HBx-transgenic mice. This discrepancy may be explained by differences in the level and duration of X gene expression and the genetic background of the mouse strains used in these studies. It is of interest to note that the avian hepadnaviruses lack an X open reading frame. The fact that HCCs do not arise in the context
of chronic duck or heron hepatitis B virus infection supports an oncogenic role of HBx in tumor development. Studies suggested that HBx may, like gene products of other DNA tumor viruses, interact with p53 and thereby interfere with its functions.27 The p53 tumor suppressor gene modulates gene transcription and controls apoptosis, cell cycle progression, and DNA repair. HBx may block p53-mediated apoptosis and provide a clonal selective advantage to hepatocytes.28 However, it is currently unclear whether such an interaction occurs at HBx levels expressed in hepatocytes during natural HBV infection. Thus, further studies are needed to clarify this intriguing issue. Another HBV gene product that has been reported to possess trans-activational properties is a truncated form of the PreS2/S gene, referred to as MHBst (truncated middle hepatitis B surface antigen).22,29,30 The fact that a structural viral protein gains regulatory functions following truncation is unusual. However, truncated PreS2/S sequences have frequently been found in HBV DNA integration sites in HCC.31
CHRONIC HEPADNAVIRUS INFECTION AND HEPATOCELLULAR CARCINOMA HBV is a member of the Hepadnaviridae family. These hepatotropic DNA viruses share a similar virion structure, genetic organization and replication strategy. Although the evidence for a role of HBV in the pathogenesis of human HCC is convincing, the association of woodchuck hepatitis virus (WHV) with HCC development is even stronger.32 HCCs occur almost invariably in chronically WHVinfected woodchucks.33,34 In this context, it is of considerable interest that HCC develops more frequently also in woodchucks serologically recovered from WHV infection. WHV DNA was detected in a substantial number of such tumors by Southern blot analysis.35 This is reminiscent of the increased incidence of HCC found in HBsAg-negative individuals with serological evidence of past HBV infection.36 Investigation of hepadnaviral integration sites to identify cellular genes involved in HCC development was particularly rewarding in the case of HCCs associated with chronic WHV infection. Activation of myc family oncogenes, presumably resulting from cis- and trans-acting effects of integrated WHV regulatory elements, was found in the majority of these tumors.37,38 Analysis of WHV DNA integration sites in woodchuck HCCs has led to the identification of a second intron-less N-myc gene.39 WHV DNA integration, either upstream of N-myc2 or in the 3¢ non-coding region of N-myc2, was observed in a total of 27 of 66 (41%) woodchuck HCCs investigated in three studies.39–41 Interestingly, in a significant proportion of tumors where N-myc2 expression was up-regulated in the absence of a nearby viral integration, WHV DNA integration was found in a common chromosomal site located about 200 kb downstream of N-myc2, suggesting long-range protooncogene activation by the WHV enhancer.42 There is accumulating evidence that the level of replication induced by naturally occurring mutations in the precore and core promoter regions increases the risk of HCC.43–45 Thus, a high viral replication phenotype places the infected liver at greater risk for transformation. Finally, with the development of diagnostic techniques sensitive enough to detect very low levels (<100 copies per
167
Section I. Pathophysiology of the Liver
ml) of HBV DNA, it has become increasingly apparent that many individuals with HCV are also co-infected with low-level HBV. In this setting, HBV maintains its oncogenic properties46 and evidence is accumulating that occult HBV infection may be associated with chronic hepatitis and cirrhosis of heretofore unknown etiology.47
HEPATITIS C VIRUS HCV infection is a leading cause of chronic hepatitis, liver cirrhosis, and HCC worldwide. In many industrialized countries it is now the leading cause of HCC. HCV has been classified in the Hepacivirus genus within the Flaviviridae family of viruses, which includes the classic flaviviruses, such as yellow fever virus, the animal pestiviruses, and the as yet unassigned GB viruses A (GBV-A), GBVB and GBV-C. HCV contains a 9.6 kb positive-strand RNA genome that encodes a polyprotein precursor of about 3000 amino acids.48,49 The HCV polyprotein precursor is co- and post-translationally processed by cellular and viral proteases to yield the mature structural and non-structural proteins. The structural proteins include the core protein and the envelope glycoproteins E1 and E2. The nonstructural proteins include the NS2-3 protease and the NS3 serine protease, an RNA helicase located in the C-terminal two-thirds of NS3, the NS4A polypeptide, the NS4B and NS5A proteins, and the NS5B RNA-dependent RNA polymerase. Similar to HBV, HCC associated with HCV infection evolves after many years of chronic infection and is generally preceded by the development of cirrhosis (Figure 10-3).50 HCV infection
Blood transfusion
CPH
CAH
Cirrhosis
HCC
therefore offers a paradigm for the role of chronic liver injury followed by regeneration, cirrhosis, and the development of HCC, as shown in Figure 10-1. Clinical and experimental evidence raised the possibility that HCV might operate also through other pathways in promoting malignant transformation of hepatocytes.51 HCCs have in some cases been found in chronically HCV-infected patients without liver cirrhosis.52 Different viral proteins, notably core,53,54 NS3 and NS5A,55–57 have been implicated in HCC development.58–60 Certain transgenic mice harboring HCV sequences, notably the core gene, have been reported to develop HCC.53,61 The core protein has been reported to interact with a variety of cellular proteins and to influence various host cell functions.58 These include, among others, enhancement or inhibition of apoptosis, repression of p53 as well as p21WAF1 promoter activity, and immunosuppression. NS5A has been reported to interact with the interferon-induced double-stranded RNA-activated protein kinase (PKR) and to inhibit its catalytic activity. In addition, disruption of PKR-dependent translational control and apoptotic programs by NS5A has been suggested to confer oncogenic potential to HCV.55 Furthermore, a study has provided evidence for an src homology 3 (SH3) domain-dependent interaction of NS5A with growth factor receptor-bound protein 2 (Grb2) adaptor protein which could interfere with mitogenic signal transduction pathways.56 Finally, NS5A was proposed to bind directly to the SH3 domain of the p85 regulatory subunit of phosphoinositide 3-kinase (PI3K), thereby enhancing the catalytic p110 subunit and resulting in increased phosphorylation as well as activity of Akt and, finally, promotion of an antiapoptotic survival response.57 However, most of these studies were performed in heterologous overexpression systems, and their relevance to natural course of HCV infection and the pathogenesis of HCC needs to be further investigated. Another possible pathway for tumorigenesis is associated with oxidative stress caused by the increased production of free radicals as a result of viral protein expression and chronic inflammation. In this regard, both HCV core62 and NS5A proteins63 have been reported to induce the production of reactive oxygen species. In addition, an upregulation of inducible nitric oxide synthase (iNOS) has been shown to occur in the HCV-infected liver, which may cause mutations by oxidation, deamination, or strand breaks in the cellular DNA.64,65 Oxidative damage is also increased with inflammatory infiltration in non-tumorous liver tissue adjacent to HCC.66
METABOLIC LIVER DISEASES
–70 –60 –50 –40 –30 –20
–15
–10
–5
0
Years
Figure 10-3. The role of chronic HCV infection in the sequence of clinical events leading to the development of hepatocellular carcinoma (HCC). CPH, chronic persistent hepatitis; CAH, chronic active hepatitis. (Adapted from Kiyosawa K et al. Interrelationship of blood transfusion, non-A, non-B hepatitis and hepatocellular carcinoma: analysis by detection of antibody to HCV. Hepatology 1990;12:671–675, with permission.)
168
The importance of the pathway of chronic liver cell injury, regeneration, and cirrhosis in the pathogenesis of HCC is best illustrated by the emergence of HCC in association with metabolic diseases of the liver. Metabolic diseases such as acute intermittent and variegated porphyria, hypercitrullinemia and hereditary fructose intolerance have no associated cirrhosis, and HCC is rare in these conditions. In contrast, there are metabolic liver diseases associated with cirrhosis, such as hereditary hemochromatosis and a1-antitrypsin deficiency. In this setting, a greatly increased risk of HCC has been observed. Particularly noteworthy in this regard is hemochromatosis, where the risk of HCC is extraordinarily high
Chapter 10 MOLECULAR PATHOGENESIS OF HEPATOCELLULAR CARCINOMA
TUMOR SUPPRESSOR GENES Several investigations have demonstrated chromosomal allelic losses in HCC tissue, suggesting the deletion or alteration of tumor suppressor genes, which may play a role in the development and progression of HCC. Studies of paired HCC and non-tumorous liver samples have revealed relatively frequent loss of heterozygosity (LOH) on, among others, chromosomes 1, 2q, 4, 5q, 6q, 8, 9, 10q, 11p, 13q, 14q, 16, 17 and 22q,5–7 suggesting that these sites may harbor tumor suppressor genes involved in the pathogenesis of HCC. More recently, large genome-wide scans using microsatellite markers, which can be typed by PCR, or comparative genomic hybridization have confirmed and extended these observations. At present, however, only few tumor suppressor genes located in these deleted regions have been clearly involved in a significant subset of HCC. The retinoblastoma protein p16INK4A and cyclin D are involved in regulation of the G1 phase of the cell cycle. LOH at the RB1 locus has quite frequently been observed in HCC, and mutations of the RB1 gene were found in about 15% of these tumors. Mutations of the p16INK4A gene or inactivation of the gene by methylation (a so-called epigenetic modification) have been found as well.68 Thus, even if RB1, p16INK4A, and cyclin D are mutated individually in only 10–20% of HCCs, their involvement in control of the same cell cycle checkpoint implies that, when combined, these mutations will lead to a loss of growth control in more than 30% of HCCs.5 A protein consisting of six ankyrin repeats, termed gankyrin, was found to be overexpressed in HCC.69 Gankyrin was found to accelerate the degradation of the retinoblastoma protein and was identical to or interacted with a subunit of the 26S proteasome. It was therefore speculated that gankyrin overexpression contributes to hepatocarcinogenesis by destabilizing the retinoblastoma protein. The adenomatous polyposis coli gene product APC and b-catenin are central molecules in a signal transduction pathway mediated by Wnt/wingless ligands (see below). Somatic APC gene mutations appear to be very rare in HCC. Activating mutations of b-catenin were reported in a subset of HCCs.70,71 It has been noted that b-catenin mutations in HCC correlate with a low rate of LOH, suggesting that a mechanism of hepatocarcinogenesis involving b-catenin mutations overrides the need for multiple genetic events and loss of tumor suppressor genes in the multistep process of hepatocarcinogenesis, and may lead more directly to the malignant phenotype. Interestingly, mutations of Axin1, another factor in the Wnt signaling pathway, have been found in an out proportion of HCCs with b-catenin accumulation in the absence of mutation of the bcatenin gene.72 Taken together, deregulated expression of b-catenin, resulting from APC defects, b-catenin gene mutations, and/or Wnt signaling pathway alterations, appears to play an important role in HCC (see below).
THE P53 TUMOR SUPPRESSOR GENE AND AFLATOXINS The p53 gene is mutated in about 30% of HCCs. The frequency and type of p53 mutations are different, depending on geographic location and suspected etiology. A GÆT mutation at the third base position of codon 249 of the p53 gene, leading to an arginine to serine substitution (R249S), was found in a significant number of HCCs in patients from southern Africa and the Qidong area of China (Figure 10-4).73,74 It was suggested that this ‘hot spot’ mutation was associated with high AFB1 intake in food, and may have contributed to the high incidence of HCC in these areas.75 Indeed, when one compares the frequency of the p53 R249S mutation in more than 500 HCCs analyzed, there are mainly two regions, namely southern China and southern Africa, in particular Mozambique, where clustering of this specific mutation (up to 50–70% of HCCs) has been found.5 These are two areas of the highest AFB1 exposure in the world. In other regions where aflatoxin levels in food are low or undetectable no such mutational specificity of the p53 gene has been detected (<4% of HCCs). Aflatoxins are mycotoxins produced by the molds Aspergillus flavus and Aspergillus parasiticus and contaminate inappropriately
249 Mutation Hae III W-T
GAGGCC
M
GAGTCC
NO
YES 110 bp
75 bp 35 bp
EXON 7 Ple I W-T M
GAGGCC
NO 98 bp
GAGTCC EXON 7
12 bp
YES 80 bp 18 bp 12 bp
A
249 Mutation screening PLC/PRF/5 T8 T9 T12 T13 T15 T17 T19 T21 T25 T27 T31 T33 T35 T37 T39 T41 T43 T45 T49 T51 T53 T55
once liver cirrhosis is established.67 Indeed, there are only a few reported cases of HCC occurring in precirrhotic hemochromatosis. A notable exception is Wilson’s disease, where the risk of HCC is low even in the setting of cirrhosis, suggesting that copper may play a cytoprotective role.
110 bp75 bp35 bpB Figure 10-4. Identification of the GÆT mutation at the third base of codon 249 of the p53 gene by restriction enzyme analysis of PCR-amplified exon 7 sequences with HaeIII. A Digestion of the wild-type p53 gene with HaeIII results in two fragments of 75 bp and 35 bp respectively in length. In the presence of a mutation, the HaeIII recognition site at codon 249 is lost and only one 110 bp fragment is detected after HaeIII digestion. B In this example, four tumors (15, 27, 37, and 55) carry the specific GÆT mutation at codon 249. (Courtesy of Dr. Mehmet Ozturk.)
169
Section I. Pathophysiology of the Liver
processed and stored foods, such as corn, peanuts, and rice.76 There is a linear relationship between the content of AFB1 in the diet and the risk of HCC. The use of experimental models and specific biomarkers for aflatoxin exposure, such as urinary metabolites or aflatoxin adducts, has validated these epidemiologic observations. Importantly, dietary AFB1 exposure and coexisting HBV infection are associated with even higher rates of HCC development within a given population, and therefore the actions of these two carcinogenic agents may be synergistic.8,77 Polymorphisms of enzymes involved in the biotransformation of environmental toxins such as AFB1, benz(a)pyrene and other polycyclic aromatic hydrocarbons in tobacco smoke and others may underlie a genetic susceptibility to HCC development.78
DNA MISMATCH REPAIR Defective DNA mismatch repair can lead to the accumulation of mutations and microsatellite instability in the cellular genome and thus increase the chance of malignant transformation. DNA mismatch repair gene defects appear to play a role in a subset of HCCs,79 although the frequency of such defects is debated.80
TELOMERASE ACTIVATION The progressive shortening of chromosome ends, or telomeres, accompanies normal cell division, contributes to cellular aging, and serves as a control mechanism against unregulated cellular proliferation. A remarkable correlation has been found between certain types of cancer and the expression of telomerase, a ribonucleoprotein enzyme that prevents the shortening of telomeres. This enzyme restores telomere length and allows malignant cells to escape from the senescence process. Indeed, expression of telomerase may be a common pathway during oncogenesis. In this regard, telomerase activation has been found in a high proportion of HCCs.81 Normal liver has no telomerase activity, whereas cirrhotic liver and low-grade dysplastic nodules display mild increases in telomerase activity. The activity of telomerase in high-grade dysplastic nodules and HCCs is significantly increased. In contrast to the data on telomerase activation, a number of studies revealed shorter telomeres in HCC than in adjacent non-cancerous liver tissue, and even more pronounced in comparison to liver tissue from normal controls. A progressive shortening of telomeres through progression from normal liver tissue, cirrhosis, regenerative nodules, low-grade to high-grade dysplastic nodules and HCC has been observed. In addition, telomere shortening in HCC correlated with the aneuploidy, a marker of chromosomal instability.82 Thus, recent evidence suggests that telomere dysfunction, leading to chromosomal instability, may be operative during the early stages of hepatocarcinogenesis, whereas telomerase activation occurs during HCC progression. Telomeres and telomerase may therefore have a dual role in hepatocarcinogenesis. Telomere shortening induces chromosomal instability and the initiation of HCC, whereas tumor progression requires telomerase activation and stabilization of telomeres.83
170
ANGIOGENESIS Angiogenesis may be particularly important in hepatocarcinogenesis, as HCC is often a highly vascular tumor.84 Overexpression of vascular endothelial growth factor (VEGF), one of the most important angiogenic factors, has been documented in HCC, and additional factors, such as angiopoietins and Tie-2,85,86 seem to be involved as well. Interfering with these factors may offer new possibilities of targeted therapeutic intervention against HCC.87
SIGNAL TRANSDUCTION PATHWAYS AND VIRAL PROTEIN INTERACTIONS Investigators have identified activation of the insulin/IGF-1/IRS-1 signal transduction cascade in the majority of HCCs.88 Indeed, IRS1 overexpression is associated with increased HCC tumor size and downstream activation of the MAPK cascade, leading to cell proliferation (Figure 10-5). The IRS-1 protein is the main intracellular substrate for receptor tyrosine kinase activity of the insulin/IGF-1 receptor, emitting downstream signals through interaction with SH2 domain-containing molecules at specific motifs located in the C-terminal region. Such protein–protein interactions lead to Ras/Raf/MAPKK/MAPK cascade activation, which is involved in cell growth and survival. There is recent experimental evidence to suggest that HBV and HCV proteins, namely HBx and NS5A, interact with Ras and Grb2 and modulate the growth factor signal transduction cascade. These investigations provide a link between viral and cellular proteins, emphasizing the potential importance of the IRS-1 pathway in the pathogenesis of HCC and the increased risk associated with chronic HBV and HCV infection. The Wnt/Frizzled signaling network influences diverse biological processes, ranging from cell fate determination to cell motility and proliferation, as outlined above and as shown in Figure 10-6. Previous studies have emphasized that at least two major genetic pathways may be involved during the emergence of HCC in human disease, as well as in rodent models that involve Wnt signaling. One pathway is characterized by an intact Wnt signaling cascade in the setting of a high mutator phenotype and chromosomal instability due to viral infection. The second is characterized by disruption of the Wnt signaling cascade, but the process is not limited to activating b-catenin mutations as there is nuclear accumulation in the setting of a wild-type gene. This pathway is associated with a low rate of LOH.89–91 Thus, b-catenin activation may be a very important early event in hepatocarcinogenesis. However, it is estimated that in 35–80% of HCCs the aberrant b-catenin cellular accumulation is not associated with a mutation affecting the b-catenin, Axin1 or APC genes of the Wnt/b-catenin signal transduction cascade. The expression and/or functional role of Frizzled gene family members, which are the cellular receptors for Wnt ligands, has recently been explored during human hepatocarcinogenesis.92 Realtime RT-PCR measurement of Frizzled-7 (FZD7) receptor expression in human HCC tumors and adjacent uninvolved liver revealed high levels, as shown in Figure 10-6. There was a functional
Chapter 10 MOLECULAR PATHOGENESIS OF HEPATOCELLULAR CARCINOMA
Insulln/IGF-1
S-S
P
S-S P P
IRS
PIP3
Shc
P
Gab1
p85 p3k p110
P
Grb2 Crk Nck
IRS-1
P P
PDK1 P
SOS
P
PTEN P
P
Ras
AKt P
Cell Survival
P GSK3-
P
P
P
P Bad
P Caspase9
SOS
MEK1/2 14-3-3
Caspase9
Cell survival
P
Bad
Apoptosis
ERK
P
Mitogenesis
A
C-1 N T
C-4 N T
IRS-1 T/N
3.0
C-7 N T
2.5
C-8 N T
C-10 N T
2.0
C-13 N T
1.5
C-15 N T
C-18 N T
1.0
C-22 N T
0.5 2
B
C
4
6
8
10
Tumor size (cm)
Figure 10-5. The role of the insulin/IGF-1 signal transduction pathway in promoting cell survival and mitogenesis during hepatocarcinogenesis. A Cartoon exhibiting the various downstream signal transduction molecules that stimulate cell proliferation and survival. Note that IRS-1 plays a key role in sending both survival and mitogenic signals to hepatocytes during the transformation process. B Western blot analysis of normal (N) tumor (T) pairs of tissue illustrating the overexpression of IRS-1 in tumors compared to adjacent uninvolved liver. C Correlation between tumor size and the protein levels of IRS-1 in 22 HCCs. Note that there is a correlation between tumor size and the level of IRS-1 expression.
171
Section I. Pathophysiology of the Liver
Wnt
Frizzled Dsh
Dsh -catenin
A Axin P GSK3 C
-catenin
A Axin P GSK3 C
P
-catenin
-catenin
-catenin
-catenin
P
Proteosome Cell migration and proliferation A
90
Huh7 180 FZD7 per 109 copies of 18SrRNA
Pearent of total migrated cells
80 70 60 50
Focus Hep3
40 30 20
HepG2
10
140 120 100 80 60 40 20 0
0 0
200
400
600
800
1000
Normal livers (n=4)
1200
FZD7 per 109 copies of 18SrRNA B
160
C
pT
T
South Africa paired samples (n=14)
pT
T
Taiwan paired samples (n=16)
Figure 10-6. The role of the cononical Wnt/b-catenin pathway in the pathogenesis of HCC. A Cartoon illustrating the major signaling molecules and the general features of pathway activation. The left side shows that when the Frizzled receptors are not engaged with the Wnt ligand there is complex formation, followed by phosphorylation of b-catenin and subsequent degradation of the molecule. However, when the Wnt pathway is activated by binding of the Wnt ligand to the Frizzled receptor, phosphorylation of b-catenin is inhibited. It accumulates in the cytoplasm, where it translocates and, in combination with Tcf and Lef transcription factors, up-regulates the genes controlling cell migration and proliferation. B Frizzled-7 (FZD7) receptor levels as measured by real-time RT-PCR correlate with a phenotype of enhanced cell migration and invasion. C Overexpression of FZD7 receptors in HCCs (T) compared to peritumorous areas (pT) in 30 tumors from South Africa and Taiwan. Note the high-level expression in tumor compared to the peritumoral area. FZD7 levels are also strikingly increased in peritumorous areas versus normal liver, suggesting that up-regulation of FZD7 receptor expression is an early genetic change in the pathogenesis of HCC.
172
Chapter 10 MOLECULAR PATHOGENESIS OF HEPATOCELLULAR CARCINOMA
association of FZD7 receptor expression with enhanced cellular motility and invasiveness. These observations suggest that the Wnt/b-catenin signal transduction pathway is much more commonly involved in the molecular pathogenesis of HCC than previously recognized, as FZD7 overexpression was found to be common in over 90% of HBV-related tumors, occurred early in the disease process, stabilized wild-type b-catenin levels, and contributed to enhanced tumor cell migration.92
SUMMARY AND PERSPECTIVES HCC is one of the most common malignant tumors worldwide. The major etiologies and risk factors for HCC development are well defined, and some of the multiple steps involved in hepatocarcinogenesis have been elucidated in recent years. However, no clear picture of how and in what sequence these factors interact at the molecular level has yet emerged. Hepatocyte transformation occurs in the setting of chronic liver injury, regeneration, and cirrhosis. Increased cell turnover in this context of inflammation and oxidative DNA damage may result in genetic alterations, such as the activation of cellular oncogenes; the inactivation of tumor suppressor genes, possibly in cooperation with genomic instability, including DNA mismatch repair defects and impaired chromosomal segregation; overexpression of growth and angiogenic factors; and telomerase activation. New technologies, including gene expression profiling93–95 and proteomic analysis,96 may eventually lead not only to a better understanding of the cellular events involved in hepatocyte transformation, but in all likelihood also to improved preventive measures and innovative therapies for one of the most devastating human malignancies in the world today.
REFERENCES 1. Okuda K. Hepatocellular carcinoma. J Hepatol 2000;32(Suppl 1):225–237. 2. Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet 2003;362:1907–1917. 3. Bosch FX, Ribes J, Diaz M, Cleries R. Primary liver cancer: worldwide incidence and trends. Gastroenterology 2004;127:S5–S16. 4. El-Serag HB. Hepatocellular carcinoma: recent trends in the United States. Gastroenterology 2004;127:S27–S34. 5. Ozturk M. Genetic aspects of hepatocellular carcinogenesis. Semin Liver Dis 1999;19:235–242. 6. Thorgeirsson SS, Grisham JW. Molecular pathogenesis of human hepatocellular carcinoma. Nature Genet 2002;31:339–346. 7. Suriawinata A, Xu R. An update on the molecular genetics of hepatocellular carcinoma. Semin Liver Dis 2004;24:77–88. 8. Ming L, Thorgeirsson SS, Gail MH, et al. Dominant role of hepatitis B virus and cofactor role of aflatoxin in hepatocarcinogenesis in Qidong, China. Hepatology 2002;36:1214–1220. 9. Fattovich G, Stroffolini T, Zagni I, Donato F. Hepatocellular carcinoma in cirrhosis: incidence and risk factors. Gastroenterology 2004;127:S35–S50. 10. Beasley RP, Hwang L-Y, Lin C-C, Chien C-S. Hepatocellular carcinoma and hepatitis B virus – a prospective study of 22707 men in Taiwan. Lancet 1981;ii:1129–1133. 11. Chang M-H, Chen C-J, Lai M-S, et al., for the Taiwan Childhood Hepatoma Study Group. Universal hepatitis B vaccination in
12.
13.
14.
15.
16.
17. 18.
19.
20.
21.
22.
23. 24.
25.
26.
27.
28.
29.
30.
31.
Taiwan and the incidence of hepatocellular carcinoma in children. N Engl J Med 1997;336:1855–1859. Bréchot C. Pathogenesis of hepatitis B virus-related hepatocellular carcinoma: old and new paradigms. Gastroenterology 2004;127:S56–S61. Nakamoto Y, Guidotti LC, Kuhlen CV, et al. Immune pathogenesis of hepatocellular carcinoma. J Exp Med 1998;188:341–350. Nakamoto Y, Kaneko S, Fan H, et al. Prevention of hepatocellular carcinoma development associated with chronic hepatitis by antifas ligand antibody therapy. J Exp Med 2002;196:1105–1111. Dejean A, Bougueleret L, Grzeschik K-H, Tiollais P. Hepatitis B virus DNA integration in a sequence homologous to v-erb-A and steroid receptor genes in a hepatocellular carcinoma. Nature 1986;322:70–72. de Thé H, Marchio A, Tiollais P, Dejean A. A novel steroid thyroid hormone receptor-related gene inappropriately expressed in human hepatocellular carcinoma. Nature 1987;330:667–670. Brand N, Petkovich M, Krust A, et al. Identification of a second human retinoic acid receptor. Nature 1988;332:850–853. Wang J, Chenivesse X, Henglein B, Bréchot C. Hepatitis B virus integration in a cyclin A gene in a hepatocellular carcinoma. Nature 1990;343:555–557. Chami M, Gozuacik D, Saigo K, et al. Hepatitis B virus-related insertional mutagenesis implicates SERCA1 gene in the control of apoptosis. Oncogene 2000;19:2877–2886. Chami M, Gozuacik D, Lagorce D, et al. SERCA1 truncated proteins unable to pump calcium reduce the endoplasmic reticulum calcium concentration and induce apoptosis. J Cell Biol 2001;153:1301–1314. Paterlini-Bréchot P, Saigo K, Murakami Y, et al. Hepatitis B virusrelated insertional mutagenesis occurs frequently in human liver cancers and recurrently targets human telomerase gene. Oncogene 2003;22:3911–3916. Caselmann WH, Koshy R. Transactivators of HBV, signal transduction and tumorigenesis. In: Caselmann WH, Koshy R, eds. Hepatitis B virus. London: Imperial College Press, 1998: 91–98. Bouchard MJ, Schneider RJ. The enigmatic X gene of hepatitis B virus. J Virol 2004;78:12725–12734. Kekulé AS, Lauer U, Weiss L, et al. Hepatitis B virus transactivator HBx uses a tumour promoter signalling pathway. Nature 1993;361:742–745. Kim C-M, Koike R, Saito I, et al. HBx gene of hepatitis B virus induces liver cancer in transgenic mice. Nature 1991;351:317–320. Koike K, Moriya K, Iino S, et al. High-level expression of hepatitis B virus HBx gene and hepatocarcinogenesis in transgenic mice. Hepatology 1994;19:810–819. Wang XW, Forrester K, Yeh H, et al. Hepatitis B virus X protein inhibits p53 sequence-specific DNA binding, transcriptional activity, and association with transcription factor ERCC3. Proc Natl Acad Sci USA 1994;91:2230–2234. Huo TI, Wang XW, Forgues M, et al. Hepatitis B virus X mutants derived from human hepatocellular carcinoma retain the ability to abrogate p53-induced apoptosis. Oncogene 2001;20: 3620–3628. Caselmann WH, Meyer M, Kekulé AS, et al. A trans-activator function is generated by integration of hepatitis B virus preS/S sequences in human hepatocellular carcinoma DNA. Proc Natl Acad Sci USA 1990;87:2970–2974. Kekulé AS, Lauer U, Meyer M, et al. The preS2/S region of integrated hepatitis B virus DNA encodes a transcriptional transactivator. Nature 1990;343:457–461. Schlüter V, Meyer M, Hofschneider PH, et al. Integrated hepatitis B virus X and 3¢ truncated preS/S sequences derived from human hepatomas encode functionally active transactivators. Oncogene 1994;9:3335–3344.
173
Section I. Pathophysiology of the Liver
32. Tennant BC, Toshkov IA, Peek SF, et al. Hepatocellular carcinoma in the woodchuck model of hepatitis B virus infection. Gastroenterology 2004;127:S283–293. 33. Popper H, Shih J-K, Gerin JL, et al. Woodchuck hepatitis and hepatocellular carcinoma: correlation of histologic with virologic observations. Hepatology 1981;1:91–98. 34. Radaeva S, Li Y, Hacker HJ, et al. Hepadnaviral hepatocarcinogenesis: in situ visualization of viral antigens, cytoplasmic compartmentation, enzymic patterns, and cellular proliferation in preneoplastic hepatocellular lineages in woodchucks. J Hepatol 2000;33:580–600. 35. Korba BE, Wells FV, Baldwin B, et al. Hepatocellular carcinoma in woodchuck hepatitis virus-infected woodchucks: presence of viral DNA in tumor tissue from chronic carriers and animals serologically recovered from acute infection. Hepatology 1989;9:461–470. 36. Yuen MF, Wong DK, Sablon E, et al. HBsAg seroclearance in chronic hepatitis B in the Chinese: virological, histological, and clinical aspects. Hepatology 2004;39:1694–1701. 37. Möröy T, Marchio A, Etiemble J, et al. Rearrangement and enhanced expression of c-myc in hepatocellular carcinoma of hepatitis virus infected woodchucks. Nature 1986;324:276–279. 38. Hsu T-Y, Möröy T, Etiemble T, et al. Activation of c-myc by woodchuck hepatitis virus insertion in hepatocellular carcinoma. Cell 1988;55:627–635. 39. Fourel G, Trépo C, Bougueleret L, et al. Frequent activation of N-myc genes by hepadnavirus insertion in woodchuck liver tumors. Nature 1990;347:294–298. 40. Wei Y, Fourel G, Ponzetto A, et al. Hepadnavirus integration: mechanisms of activation of the N-myc 2 retrotransposon in woodchuck liver tumors. J Virol 1992;66:5265–5276. 41. Hansen LJ, Tennant BC, Seeger C, Ganem D. Differential activation of myc gene family members in hepatic carcinogenesis by closely related hepatitis B viruses. Mol Cell Biol 1993;13:659–667. 42. Fourel G, Couturier J, Wei Y, et al. Evidence for long-range oncogene activation by hepadnavirus insertion. EMBO J 1994;13:2526–2534. 43. Baptista M, Kramvis A, Kew MC. High prevalence of 1762(T) 1764(A) mutations in the basic core promoter of hepatitis B virus isolated from black Africans with hepatocellular carcinoma compared with asymptomatic carriers. Hepatology 1999;29:946–953. 44. Kao JH, Chen PJ, Lai MY, Chen DS. Basal core promoter mutations of hepatitis B virus increase the risk of hepatocellular carcinoma in hepatitis B carriers. Gastroenterology 2003;124:327–334. 45. Kuang SY, Jackson PE, Wang JB, et al. Specific mutations of hepatitis B virus in plasma predict liver cancer development. Proc Natl Acad Sci USA 2004;101:3575–3580. 46. Pollicino T, Squadrito G, Cerenzia G, et al. Hepatitis B virus maintains its pro-oncogenic properties in the case of occult HBV infection. Gastroenterology 2004;126:102–110. 47. Wands JR. Prevention of hepatocellular carcinoma. N Engl J Med 2004;351:1567–1570. 48. Moradpour D, Blum HE. A primer on the molecular virology of hepatitis C. Liver Int 2004;24:519–525. 49. Penin F, Dubuisson J, Rey FA, et al. Structural biology of hepatitis C virus. Hepatology 2004;39:5–19. 50. Kiyosawa K, Sodeyama T, Tanaka E, et al. Interrelationship of blood transfusion, non-A, non-B hepatitis and hepatocellular carcinoma: analysis by detection of antibody to hepatitis C virus. Hepatology 1990;12:671–675. 51. Liang TJ, Heller T. Pathogenesis of hepatitis C-associated hepatocellular carcinoma. Gastroenterology 2004;127: S62–S71. 52. De Mitri MS, Poussin K, Baccarini P, et al. HCV-associated liver cancer without cirrhosis. Lancet 1995;345:413–415.
174
53. Moriya K, Fujie H, Shintani Y, et al. The core protein of hepatitis C virus induces hepatocellular carcinoma in transgenic mice. Nature Med 1998;4:1065–1067. 54. Yoshida T, Hanada T, Tokuhisa T, et al. Activation of STAT3 by the hepatitis C virus core protein leads to cellular transformation. J Exp Med 2002;196:641–653. 55. Gale M Jr, Kwieciszewski B, Dossett M, et al. Antiapoptotic and oncogenic potentials of hepatitis C virus are linked to interferon resistance by viral repression of the PKR protein kinase. J Virol 1999;73:6506–6516. 56. Tan SL, Nakao H, He Y, et al. NS5A, a nonstructural protein of hepatitis C virus, binds growth factor receptor-bound protein 2 adaptor protein in a Src homology 3 domain/ligand-dependent manner and perturbs mitogenic signaling. Proc Natl Acad Sci USA 1999;96:5533–5538. 57. Street A, Macdonald A, Crowder K, Harris M. The hepatitis C virus NS5A protein activates a phosphoinositide 3-kinasedependent survival signaling cascade. J Biol Chem 2004;279:12232–12241. 58. McLauchlan J. Properties of the hepatitis C virus core protein: a structural protein that modulates cellular processes. J Viral Hepatol 2000;7:2–14. 59. Tellinghuisen TL, Rice CM. Interaction between hepatitis C virus proteins and host cell factors. Curr Opin Microbiol 2002;5:419–427. 60. Macdonald A, Harris M. Hepatitis C virus NS5A: tales of a promiscuous protein. J Gen Virol 2004;85:2485–2502. 61. Lerat H, Honda M, Beard MR, et al. Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus. Gastroenterology 2002;122:352–365. 62. Okuda M, Li K, Beard MR, et al. Mitochondrial injury, oxidative stress, and antioxidant gene expression are induced by hepatitis C virus core protein. Gastroenterology 2002;122:366–375. 63. Gong G, Waris G, Tanveer R, Siddiqui A. Human hepatitis C virus NS5A protein alters intracellular calcium levels, induces oxidative stress, and activates STAT-3 and NF-kappa B. Proc Natl Acad Sci USA 2001;98:9599–9604. 64. Machida K, Cheng KT, Sung VM, et al. Hepatitis C virus infection activates the immunologic (type II) isoform of nitric oxide synthase and thereby enhances DNA damage and mutations of cellular genes. J Virol 2004;78:8835–8843. 65. Machida K, Cheng KT, Sung VM, et al. Hepatitis C virus induces a mutator phenotype: enhanced mutations of immunoglobulin and protooncogenes. Proc Natl Acad Sci USA 2004;101:4262–4267. 66. Jungst C, Cheng B, Gehrke R, et al. Oxidative damage is increased in human liver tissue adjacent to hepatocellular carcinoma. Hepatology 2004;39:1663–1672. 67. Kowdley KV. Iron, hemochromatosis, and hepatocellular carcinoma. Gastroenterology 2004;127:S79–S86. 68. Liew CT, Li HM, Lo KW, et al. High frequency of p16INK4A gene alterations in hepatocellular carcinoma. Oncogene 1999;18:789–795. 69. Higashitsuji H, Itoh K, Nagao T, et al. Reduced stability of retinoblastoma protein by gankyrin, an oncogenic ankyrin-repeat protein overexpressed in hepatomas. Nature Med 2000;6: 96–99. 70. de La Coste A, Romagnolo B, Billuart P, et al. Somatic mutations of the beta-catenin gene are frequent in mouse and human hepatocellular carcinomas. Proc Natl Acad Sci USA 1998;95: 8847–8851. 71. Miyoshi Y, Iwao K, Nagasawa Y, et al. Activation of the beta-catenin gene in primary hepatocellular carcinomas by somatic alterations involving exon 3. Cancer Res 1998;58: 2524–2527. 72. Satoh S, Daigo Y, Furukawa Y, et al. AXIN1 mutations in hepatocellular carcinomas, and growth suppression in cancer cells
Chapter 10 MOLECULAR PATHOGENESIS OF HEPATOCELLULAR CARCINOMA
73.
74.
75.
76.
77. 78.
79.
80.
81.
82.
83.
84. 85.
by virus-mediated transfer of AXIN1. Nature Genet 2000;24:245–250. Bressac B, Kew M, Wands J, Ozturk M. Selective G to T mutations of p53 gene in hepatocellular carcinoma from southern Africa. Nature 1991;350:429–431. Hsu IC, Metcalf RA, Sun T, et al. Mutational hotspot in the p53 gene in human hepatocellular carcinomas. Nature 1991;350:427–428. Aguilar F, Hussain SP, Cerutti P. Aflatoxin B1 induces the transversion of G Æ T in codon 249 of the p53 tumor suppressor gene in human hepatocytes. Proc Natl Acad Sci USA 1993;90:8586–8590. Smela ME, Currier SS, Bailey EA, Essigmann JM. The chemistry and biology of aflatoxin B1: from mutational spectrometry to carcinogenesis. Carcinogenesis 2001;22:535–545. Henry SH, Bosch FX, Troxell TC, Bolger PM. Reducing liver cancer – global control of aflatoxin. Science 1999;286:2453–2454. Chen CJ, Chen DS. Interaction of hepatitis B virus, chemical carcinogen, and genetic susceptibility: multistage hepatocarcinogenesis with multifactorial etiology. Hepatology 2002;36:1046–1049. Macdonald GA, Greenson JK, Saito K, et al. Microsatellite instability and loss of heterozygosity at DNA mismatch repair gene loci occurs during hepatic carcinogenesis. Hepatology 1998;28:90–97. Piao Z, Kim H, Malkhosyan S, Park C. Frequent chromosomal instability but no microsatellite instability in hepatocellular carcinoma. Int J Oncol 2000;17:507–512. Kojima H, Yokosuka O, Imazeki F, et al. Telomerase activity and telomere length in hepatocellular carcinoma and chronic liver disease. Gastroenterology 1997;112:493–500. Plentz RR, Caselitz M, Bleck JS, et al. Hepatocellular telomere shortening correlates with chromosomal instability and the development of human hepatoma. Hepatology 2004;40:80–86. Satyanarayana A, Manns MP, Rudolph KL. Telomeres and telomerase: a dual role in hepatocarcinogenesis. Hepatology 2004;40:276–283. Semela D, Dufour JF. Angiogenesis and hepatocellular carcinoma. J Hepatol 2004;41:864–880. Tanaka S, Mori M, Sakamoto Y, et al. Biologic significance of angiopoietin-2 expression in human hepatocellular carcinoma. J Clin Invest 1999;103:341–345.
86. Mitsuhashi N, Shimizu H, Ohtsuka M, et al. Angiopoietins and Tie-2 expression in angiogenesis and proliferation of human hepatocellular carcinoma. Hepatology 2003;37:1105–1113. 87. Yoshiji H, Kuriyama S, Yoshii J, et al. Halting the interaction between vascular endothelial growth factor and its receptors attenuates liver carcinogenesis in mice. Hepatology 2004;39:1517–1524. 88. Ito T, Sasaki Y, Wands JR. Overexpression of human insulin receptor substrate 1 induces cellular transformation with activation of mitogen-activated protein kinases. Mol Cell Biol 1996;16:943–951. 89. Legoix P, Bluteau O, Bayer J, et al. Beta-catenin mutations in hepatocellular carcinoma correlate with a low rate of loss of heterozygosity. Oncogene 1999;18:4044–4046. 90. Laurent-Puig P, Legoix P, Bluteau O, et al. Genetic alterations associated with hepatocellular carcinomas define distinct pathways of hepatocarcinogenesis. Gastroenterology 2001;120:1763–1773. 91. Calvisi DF, Factor VM, Ladu S, et al. Disruption of beta-catenin pathway or genomic instability define two distinct categories of liver cancer in transgenic mice. Gastroenterology 2004;126:1374–1386. 92. Merle P, de la Monte S, Kim M, et al. Functional consequences of frizzled-7 receptor overexpression in human hepatocellular carcinoma. Gastroenterology 2004;127:1110–1122. 93. Iizuka N, Oka M, Yamada-Okabe H, et al. Oligonucleotide microarray for prediction of early intrahepatic recurrence of hepatocellular carcinoma after curative resection. Lancet 2003;361:923–929. 94. Ye QH, Qin LX, Forgues M, et al. Predicting hepatitis B viruspositive metastatic hepatocellular carcinomas using gene expression profiling and supervised machine learning. Nature Med 2003;9:416–423. 95. Lee JS, Thorgeirsson SS. Genome-scale profiling of gene expression in hepatocellular carcinoma: classification, survival prediction, and identification of therapeutic targets. Gastroenterology 2004;127:S51–S55. 96. Chignard N, Beretta L. Proteomics for hepatocellular carcinoma marker discovery. Gastroenterology 2004;127:S120–S125.
175
Section I: Pathophysiology of the Liver
STEM CELLS AND HEPATOCYTE TRANSPLANTATION
11
Sanjeev Gupta Abbreviations AFP a-fetoprotein FAH fumaryl acetoacetate hydrolase hESC human embryonic stem cells LDLR low density lipoprotein receptor
LEC MAPC NAR
long–evans cinnamon multipotent adult progenitor cells nagase analbuminemic rat
INTRODUCTION Identification of the hepatic stem cell will advance hepatology in many ways, including by facilitating insights into the development, regeneration, and repair of the liver. From an applied perspective, identification of cells amenable to cryostorage, manipulation and expansion in vitro, as well as survival in vivo following transplantation, will be highly appropriate. The common belief is that stem cells are capable of self-renewal, while producing progeny that may generate an entire animal (totipotency), all lineages (pluripotency), multiple lineages (multipotency), two lineages (bipotency), or even single lineages, through cells with extensive proliferation ability that are often designated as transit-amplifying, lineage-restricted, or facultative progenitor cells. Most experts agree that stem/progenitor cells have a role to play in the liver.1,2 However, the hepatic stem cell has neither been isolated nor fully defined, although under specific circumstances activation of candidate liver stem cells has been identified. As discussed elsewhere, the liver is composed of multiple cell types which arise from separate germ layers, for example epithelial cells, including hepatocytes, are endodermal in origin, whereas liver sinusoidal endothelial cells, the next largest liver cell compartment, as well as Kupffer cells, are of mesodermal origin, and hepatic stellate cells exhibit features of the neuroectoderm. This embryological diversity results in unique, albeit incompletely defined, cell–cell interactions during both health and disease.3 In contrast with the embryonic or the fetal liver, the normal adult liver is essentially a stable organ without significant cell cycling, and consequently it has generally been difficult to identify or to classify hepatic stem/progenitor cells in the conventional sense of cell lineages, as for the hematopoietic system, for example. Nonetheless, some non-parenchymal liver epithelial cell subsets, such as ‘oval cells,’ which were originally identified in the oncogenetically perturbed rat liver, have been assigned progenitor cell properties.2 The molecular basis of this lineage relationship is not well understood. This situation is not unique to the liver and applies to other solid organs, e.g. the pancreas.4 On the other hand, recent studies showed that under specific circumstances adult hepatocytes possess an indefinite stem cell-like replication potential in vivo.5 Moreover, additional cell types may generate liver cells (Figure 11-1).
OLT T3 VEGF
orthotopic liver transplantation triiodothyronine vascular endothelial growth factor
However, supplies of donor human livers are limited, which has hampered the wider application of orthotopic liver transplantation (OLT), and only a declining number of livers are available for cell isolation.6 Therefore, alternative sources are necessary if cell therapy is to be advanced. If stem cells are to meet this challenge, it will be helpful to obtain unlimited supplies of undifferentiated stem/ progenitor cells that could be maintained and expanded without the requirement for feeder cells or biological products from other species; to have the ability to genetically manipulate stem/ progenitor cells without deleterious karyotypic perturbations or malignant transformation; to efficiently and reproducibly differentiate stem/progenitor cells along desired lineages either before or after cell transplantation; and to identify stem/progenitor cells with the capacity to repopulate healthy or diseased organs without undergoing rejection. In parallel, it will be helpful to develop clinical protocols for transplanting cells in optimal ways, including the identification of strategies to promote cell engraftment and proliferation, as well as to demonstrate the overall mass, function, and eventual fate of transplanted stem/progenitor cells.
CANDIDATE STEM/PROGENITOR CELL POPULATIONS FOR LIVERDIRECTED CELL THERAPY In principle, allogeneic stem/progenitor cells can be isolated from extrahepatic sources, including embryonic, fetal, and adult organs, as well as from the liver.1,2 Similarly, autologous cells can be harvested from the bone marrow or peripheral blood, and hepatocytes can even be isolated from surgically resected portions of the liver.7 In 1998, human embryonic stem cells (hESC) were first isolated from discarded embryos by dissociating the inner cell mass of blastocysts.8 These cells constituted a unique example of totipotent or pluripotent cells. However, only a few hESC lines are available at present. The current policy of the United States restricts federally funded research to hESC lines that were created prior to 9 August 2001 (to review the evolution of these policies and qualified hESC cell lines, visit: http://stemcells.nih.gov/index.asp). Other countries, including South Korea and the UK, have less restrictive policies. Most experts agree that large numbers of new hESC lines are
177
Section I. Pathophysiology of the Liver
Circulating stem / progenitor cells
Hepatocytes
Endothelial cells Kupffer cells Stellate cells Oval cells
Ductal cells
Figure 11-1. Various sources of stem/progenitor cells in the adult liver. The adult liver is replicatively quiescent, but following injury to hepatocytes (e.g. liver resection) or bile ducts (e.g. biliary obstruction), these cell compartments can selfrenew themselves, without recruiting stem/ progenitor cells from elsewhere. However, in the presence of overwhelming parenchymal injury resident facultative progenitor cells – ‘oval cells’ – may appear, and these cells can replenish either hepatocytes or bile duct cells, as necessary. Evidence has also been provided for the capacity of hepatocytes to produce bile duct cells under certain circumstances. Circulating stem/progenitor cells do not contribute to the replenishment of parenchymal liver cells under physiological circumstances, although liver sinusoidal endothelial cells, hepatic stellate cells, and Kupffer cells may be derived from stem/progenitor cells in the blood circulation. Bone marrow-derived cells do not generate hepatic oval cells and may generate occasional hepatocytes, although the role of cell fusion in this process is controversial.
Circulating stem / progenitor cells
needed to study the biological potential of hESC. For instance, hESC lines differ from one another for unknown reasons that alter requirements for their growth, maintenance, and differentiation.9 For clinical applications, it will be appropriate to develop large banks of hESC lines spanning the genetic diversity of human populations, although, of course, social and ethical reservations about deriving hESC have not been fully resolved. A major source of the excitement in hESC research concerns generating indefinite supplies of autologous cells from individual donors, as recently demonstrated by the application of somatic cell nuclear transfer to generate hESC lines from people with specific medical conditions.10 The use of autologous cells will obviously help avoid rejection after cell transplantation, which is a major clinical goal. Similarly, the availability of genetically defined hESC lines will help define the basic biology of hESC, including critical issues such as what imparts ‘stemness’ to hESC; what regulates the silencing and activation of specific genes and gene networks in hESC compared with other more differentiated cells; what underlies the capacity of hESC to replicate indefinitely without senescence or other deleterious perturbations; and what mechanisms could be utilized to induce differentiation in hESC along desired lineages. Progress has already been made in understanding the potential of hESC, although much more needs to be accomplished. For instance, transplantation of undifferentiated hESC results in teratomas or teratocarcinomas, essentially by default, although this intrinsic teratogenic potential can be circumvented by the intermediary step of ‘embryoid body’ formation before cell dissociation.11,12 Unique, albeit incompletely characterized, hepatic-like hESC subpopulations have been selected through the a-fetoprotein (AFP) promoter trap,
178
expression of the endoderm-specifying Foxa (HNF-3b) transcription factor, or chemicals, e.g. butyrate.13–16 Also, specific soluble signals have been utilized to circumvent the requirement of embryoid body formation or other manipulations to generate hepatic-like daughter cells.16,17 During these manipulations, early genome-wide transcription profiling of undifferentiated and differentiated hESC has begun to identify changes in gene expression patterns under different states of hESC, including the onset of fetal hepatic enriched genes after the formation of embryoid bodies.16 Through such manipulations it has been possible to activate hepatic lineage-specific gene expression in hESC, including switching between AFP and albumin expression, as expected during fetal to adult hepatic transitions, although these changes have largely been at the mRNA, rather than the protein level, suggesting limited or incomplete cell differentiation.18,19 Nonetheless, hESC have been shown capable of generating suitable liver cell types after transplantation into appropriate animal models.20 The role of specific genetic pathways in determining whether hESC will remain undifferentiated or not has begun to be explored. In recent studies, specific cell signaling pathways, e.g. the wnt pathway, have been found to play key roles in promoting and maintaining hESC in an undifferentiated state.21 The principles driving differentiation in hESC along some lineages are better understood,22 whereas regulation of hESC differentiation along endodermal lineages, e.g. hepatic and pancreatic b-cell lineages, is poorly defined. Another restriction at present concerns expanding the number of differentiated cells derived from hESC because cell proliferation declines rapidly in differentiating cells. Moreover, scaling-up of hESC cultures for clinical applications will be an expensive propo-
Chapter 11 STEM CELLS AND HEPATOCYTE TRANSPLANTATION
sition owing to the cost of necessary hormones, sera, and additives. Also, it will be essential to eliminate or substitute animal products and feeder cells for using hESC in people. Therefore, considerable work lies ahead before hESC or their derivatives will be ready for clinical testing. Cells isolated from fetal tissues are an alternative source of stem cells or stem/progenitor cells. For instance, instead of being completely undifferentiated, progenitor cells are committed in some way along specific lineages while retaining remarkable proliferation capacities and the ability to produce mature cells. Stem cells can be isolated from the embryonic gonadal ridges.11 Also, unique stem/progenitor cells can be isolated from fetal organs after further liver development.23,24 For instance, the fetal human liver develops rapidly during the first trimester, such that the lobular architecture begins to be acquired within 7–8 weeks of gestation, and bile is formed by 12 weeks. During this period, the liver plate structure is not fully developed and the liver simultaneously produces hematopoietic cells and parenchymal epithelial cells. Within 6–7 weeks of gestation the parenchyma of the fetal human liver contains epithelial cells with evidence for cell maturation along hepatic and biliary lineages, as well as the existence of cells with a mixed phenotype.25 Most of these fetal parenchymal epithelial cells are in an active state of proliferation. Moreover, these cells express markers of multiple lineages, including those expressed in hepatocytes (AFP, albumin, a1-antitrypsin, a1-microglobulin, glucose6-phosphatase, glycogen etc.), bile duct cells (CK-19, g-glutamyltranspeptidase, dipeptidyl peptidase IV), liver sinusoidal endothelial cells (von Willebrand factor), mesenchymal cells (desmin, vimentin, a-smooth muscle actin), and even undifferentiated hESC (e.g. Oct-4, nanog) (23,24 and unpublished data). Such multilineage gene expression capacity and ongoing replication in fetal stem/progenitor cells is of much interest for defining their cell therapy potential. Recent studies established that isolated fetal human liver cells are successfully cryopreserved with over 90% viability.23 After thawing, fetal liver cells retain the capacity to replicate extensively in culture conditions and can be expanded enormously, such that cells isolated from a single midterm donor fetal human liver could potentially be utilized for multiple recipients, although it is necessary to define this potential in vivo. Moreover, fetal human cells have been amenable to genetic reconstitution using viral vectors, including adenoviral and lentivirus vectors.23,26 Despite their capacity for significant proliferation, fetal liver cells show a different behavior from hESC and undergo attrition of telomere length during cell culture, which is associated with a gradual decline in cell proliferation.27 However, it has been possible to restore the proliferation potential of fetal human liver cells with genetic reconstitution of telomerase activity. Moreover, fetal cells are amenable to additional genetic manipulation.23,26 Transplantation studies of fetal liver cells isolated from rat or mouse verified that fetal cells mature rapidly, produce and secrete albumin, and proliferate extensively in the liver of syngeneic recipients, where cell rejection is not an issue.28–31 Similarly, fetal human liver cells engraft and proliferate in the livers of immunodeficient animals.23 Also, fetal human liver cells have been transplanted into people with acute liver failure, although the precise therapeutic value of such therapy has been unclear, as only a few such patients
have been studied.32,33 Moreover, candidate fetal liver stem/progenitor cell populations can be conveniently isolated by the display of unique cell surface markers, and cells exhibit major histocompatibility antigens poorly, which should result in an attenuated host immune response.34–36 During hepatic development, cell–cell signals, including those of embryonic endothelial cells, play a significant role in lineage differentiation.3 The nature of these signals has been best defined for the mouse liver, and knowledge of the regulatory molecules and signals in humans is limited at present. Nonetheless, rapidly proliferating parenchymal cells in the fetal human liver demonstrate unique profiles of gene expression, as partly described above, which is obviously distinct from gene expression in mature liver cells. These findings raise the possibility of stem/progenitor cells from the fetal liver being manipulated to express additional phenotypes. For instance, when Pdx-1, a homeobox regulator of pancreatic b-cell development and function, was expressed in fetal human liver cells, these cells were altered with the expression of insulin and additional pancreatic genes.37 Therefore, programs aimed at harvesting, storing, and using fetal liver cells for cell therapy should be appropriate. The use of fetal tissues will not interfere with the supply of donor livers needed for OLT programs. However, because of moral reservations some people favor the use of adult stem cells instead of embryonic or fetal cells. By and large, stem/progenitor cells are neither readily visible nor necessarily recruited in most adult organs during the course of normal events, including the liver.1,2 Therefore, undifferentiated stem cells are thought to be rare in adult life, although in some organs, typically the bone marrow and gastrointestinal tract, facultative progenitor cells participate in cell replenishment. On the other hand, stem/progenitor cells do appear to reside in the adult liver, as shown by many studies involving specific perturbations, such as oncogenic manipulations, toxic injury, chronic hepatitis or cell transplantation analysis. The so-called hepatic oval cells, which constitute a well-studied type of facultative stem/progenitor cell, were originally described by Farber in rats subjected to carcinogenic treatments.2 Subsequently, oval cells were identified to emanate from the canal of Hering, which is a specialized peribiliary structure in the liver.38 Oval cells exhibit unique properties, including hybrid isoenzyme profiles, markers of both hepatic and biliary lineages, as well as the capacity to establish long-term cell cultures. Indeed, oval cell lines established from the rat liver have been studied in detail, including in cell transplantation assays, to establish their differentiation potential. These studies demonstrated that oval cells can produce mature hepatocytes as well as unrelated lineages, e.g. cardiomyocytes.39 However, despite their capacity to replicate extensively in cell culture, oval cell lines have not shown the capacity to repopulate the liver, whereas primary oval cells isolated from the mouse liver or even the mouse pancreas can repopulate the liver, including in the fumaryl acetoacetate hydrolase (FAH) mutant mouse model of tyrosinemia type 1, the Long–Evans Cinnamon (LEC) rat model of Wilson’s disease, and the Nagase analbuminemic rat (NAR).40,41 The evidence to date suggests that hepatic oval cells constitute an organ-specific stem cell niche, and that these cells do not arise from extrahepatic reservoirs of stem cells, such as from the bone marrow.41,42 This raises the possibility that oval cells from the human liver might well be suitable for cell therapy applications.
179
Section I. Pathophysiology of the Liver
Of course, the isolation of hepatic oval cells will require adult human livers, which are in short supply, although the normal adult liver contains only rare oval cells – perhaps less than 0.001% – whereas diseased livers, e.g. livers explanted at OLT, which are normally discarded, might be a more suitable source. It is well established that mature hepatocytes can restore the liver in many situations, including after partial hepatectomy or other types of limited hepatic injury.1,2 The voluminous literature concerning the partial hepatectomy model provided extensive insights into the capacity of hepatocytes to regenerate the liver over the past 70 years. More recently, in studies with adult mouse hepatocytes, using a transplantation system involving FAH mice burdened with progressive liver injury, it was found that transplanted hepatocytes possessed an indefinite replication potential, with the capacity to divide >90 times following transplantation across multiple generations of animals.5 In contrast, primary adult hepatocytes undergo profound alterations and exhibit very limited proliferation capacity in vitro. This raises fundamental issues as to how one defines the ‘stemness’ of cells and the necessity of applying appropriate cellspecific models to elicit stem cell-like capacity. Despite the obvious liver-repopulating capacity of adult hepatocytes, there is little doubt that existing sources of donor livers are inadequate. Whether innovative solutions to this problem, e.g. by utilizing live-related donors, isolating cells from liver resections, or using xenogeneic donors, will be effective, is being studied. Extrahepatic sources of cells, e.g. cells derived from the bone marrow, umbilical cord blood, placenta, or amniotic fluid, which recently showed unimagined differentiation potential, including the capacity to generate liver cells, have generated much interest. They offer the possibility of harvesting autologous cells (from bone marrow or peripheral blood) or preserving autologous cells for use in the future (from amniotic fluid, umbilical cord blood, placenta). The amniotic fluid has been shown to contain epithelial cells in animals as well as humans.43,44 Surprisingly, a large proportion of amniotic epithelial cells demonstrate liver gene expression, including albumin, and a1-antitrypsin. These cells can be expanded in culture following growth factor stimulation and are amenable to gene transfer in vitro. Moreover, transplantation of amniotic epithelial cells in animals has been shown to result in the integration of transplanted cells in the liver parenchyma, long-term cell survival, and secretion of hepatic proteins. Significant numbers of amniotic epithelial cells can be harvested from individual donors, up to 2 ¥ 108 cells. Also, it appears that these cells show poor expression of class I and class II histocompatibility antigens. Similarly, the human placenta is thought to represent a major additional source of stem cells, and the potential of multipotent cells resident in the placenta is being defined.45 The idea that circulating cells could contribute to organ repair during physiological processes gained currency following early demonstrations of the differentiation potential of hematopoietic stem cells into hepatocytes.46–48 Subsequently, specific subsets of bone marrow-derived mouse hematopoietic stem cells, particularly those with expression of c-kit and sca-1 antigen, low-level expression of Thy-1 antigen, and absence of lineage markers, as well as bone marrow-derived mesenchymal cells, which were designated multipotent adult progenitor cells (MAPC), were capable of generating multiple cell lineages.49,50 A large body of literature rapidly
180
accumulated, verifying that such cells can produce mature hepatocytes both in vitro and in vivo. Analysis of the kinetics by which bone marrow-derived stem cells generated hepatocytes in the mouse liver indicated that 7 weeks would elapse after bone marrow transplantation before donor-derived hepatocytes first appeared.51 Even a single bone marrow-derived mouse hematopoietic stem cell was sufficient for recapitulating multiple lineages additional to the liver.52 Similarly, transplantation of MAPC into the inner cell mass of mouse blastocysts led to the appearance of mature cells in all three germ layers,53 verifying their multipotentiality. The umbilical cord blood offers many advantages for use as a source of stem cells, including its well established hematopoietic and mesenchymal potential, which has resulted in routine clinical applications. In addition, cells derived from umbilical cord blood have shown the capacity to produce hepatocytes in intact animals,54–58 including after intrablastocyst or intrafetal injections.58 However, the overall frequency with which umbilical cord blood cells produced hepatocytes was small, similar to that of bone marrowderived hematopoietic stem cells. On the other hand, umbilical cord blood may contain unique populations of epithelial-like cells, which can be expanded in vitro with soluble growth factors.54 These promising results need to be verified and strengthened, especially as umbilical cord blood cells are routinely harvested and banked for clinical use. Highly convincing demonstrations of the hepatic differentiation potential of bone marrow-derived cells were provided by studies in the FAH mutant mouse, where transplantation of healthy bone marrow-derived cells resulted in liver repopulation and correction of metabolic abnormalities.49 Similar studies of therapeutic efficacy have not yet been performed with other types of extrahepatic stem cell. One difficulty concerns the general lack of suitable animal models, where proliferation in human cells can be convincingly demonstrated. The ability of extrahepatic stem cells to form hepatocytes is astounding and has far-reaching consequences. However, problems have been encountered. First, it is now clear that conversion of bone marrow-derived cells into hepatocytes is an infrequent process.51,59–61 For instance, bone marrow-derived hematopoietic cells were found to generate only rare hepatocytes in FAH mice with liver injury, estimated to range from 1 per 104 to 106 liver cells,51 and perhaps none to fewer than 10 cells in the entire liver of healthy mice.59–61 Similarly, although FAH mice showed significant liver repopulation and therapeutic correction, this resulted from only 50–500 ‘repopulation events,’ i.e. the number of hepatocytes emanating from donor hematopoietic stem cells.51 Secondly, during the generation of mature hepatocytes, bone marrow-derived hematopoietic stem cells, as well as umbilical cord blood cells, have been shown to fuse with native cells, a phenomenon that appears to be particularly peculiar to the myelomonocytic stem cell fraction.57,62,63 However, this observation has become controversial, because careful studies have shown that such cell fusion may not be invariant.64,65 It is unclear whether the stem cell subsets used or the different nature of liver injury in various recipients produced unique conditions for cell fusion after transplantation of bone marrow or umbilical cord blood-derived hematopoietic stem cells. Nevertheless, one consequence of cell fusion is genetic instability, e.g. aneuploidy,63 which could lead to oncogenetic transformation in fused cells. On the other hand, it
Chapter 11 STEM CELLS AND HEPATOCYTE TRANSPLANTATION
should be noteworthy that transplantation of mature hepatocytes does not result in cell fusion in the liver.63 Whether other types of stem cell, e.g. MAPC, generate hepatocytes via cell fusion is unknown. Third, hematopoietic stem cells can additionally – and possibly more efficiently – produce non-parenchymal liver cells, including Kupffer cells, liver sinusoidal endothelial cells, and hepatic stellate cells,66–68 which could be mistaken for other cell types. Another critical issue concerns the use of cells expanded in culture – in their native state or following genetic manipulation, e.g. insertion of oncogenes or removal of cell cycle suppressor genes. An example of the former is provided by MAPC, where cumbersome and incompletely understood cell culture requirements are necessary for expanding cells in culture. During this process, cells could potentially accumulate undesirable genetic lesions. Similarly, clinical use of cells immortalized with the simian virus 40 T antigen will probably raise issues of safety, although methods are available to genetically excise the oncogene after cell expansion is completed.69 The removal of cell cycle suppressor genes, e.g. p27kip or p19ARF, leads to accelerated cell proliferation, although the safety of this approach needs to be defined.70,71 This concise overview of the applied state of the art of stem cell biology should indicate that liver-directed cell therapy in the short term will most effectively utilize hepatocytes from adult human
livers or stem/progenitor cells from fetal human livers, although cells from additional sources will probably become available at some time in the future.
CLINICAL TARGETS OF LIVER CELL THERAPY Several considerations should drive cell therapy applications in various clinical disorders. A large number of genetic and acquired conditions are thought to be amenable to liver-directed cell therapy, based on a variety of studies in various animal models (Table 11-1). In general, animal models capable of mimicking acute liver failure in humans are lacking for the study of cell therapy. The therapeutic considerations extend to the replacement of deficient proteins in monogenetic disorders, including lack of circulating proteins such as coagulation factors, without organ injury, loss of function through genetic mutations, e.g. abnormal low density lipoprotein receptor (LDLR) producing familial hypercholesterolemia with premature atherosclerosis and coronary artery disease but not liver damage, as well as abnormal intracellular proteins, e.g. the copper transporter ATP7B, which produces copper toxicosis and damage to the liver, brain etc. in Wilson’s disease. Acquired disorders represent the other
Table 11-1. Useful Animal Models for Cell Therapy and Other Studies* Animal strain and disease model designation
Donor cells
Applications
Healthy wild-type, syngeneic or congeneic, or allogeneic mice, rats and larger animals, dogs, pigs, non-human primates, and fetal sheep Alb-uPA, Alb-HSV-TK transgenic or AdMadtreated mice, rats, rabbits, pigs or monkeys treated with chemicals or other agents to induce acute or chronic liver injury Alb-uPA-Rag-2 mice, Alb-uPA-NOD-SCID mice, Tupaia DPPIV-deficient mutant F344 rats and mice
Genetically modified transgenic reporter containing cells, sex-mismatched cells
Eizai hyperbilirubinemic rats FAH mutant mice
Healthy hepatocytes Healthy or genetically altered mouse hepatocytes, various stem/progenitor cells Healthy hepatocytes
Analysis of transplanted cell biology, including cell engraftment, proliferation, tolerance and function Studies of transplanted cell biology; cell therapy for acute liver failure, chronic liver failure, and complications of cirrhosis—hepatic encephalopathy etc. Hepadnavirus or HCV replication and pathogenesis, antiviral drug testing Biology of transplanted hepatocytes and stem/progenitor cells Restoration of biliary transport abnormality Studies of stem cell biology and cell and gene therapy for hereditary tyrosinemia, type I Cell and gene therapy model for Crigler–Najjar syndrome, type 1 Cell and gene therapies for hemophilia A
Gunn rats
Healthy hepatocytes or stem/progenitor cells
Woodchuck, human, tupaia hepatocytes Healthy rat or mouse liver or stem/progenitor cells
Hemophilia A knockout mice, hemophilia dogs Histidinemia mice Long–Evans Cinnamon rats, atp7b null mice, toxic milk mice Mdr2 knockout mice
Healthy hepatocytes, endothelial cells or stem/ progenitor cells Healthy hepatocytes Healthy hepatocytes
Nagase analbuminemic rats ODSod/od mutant rats Purebred dalmatian dogs
Healthy hepatocytes Healthy hepatocytes Healthy hepatocytes
Spfash mice
Healthy hepatocytes
Watanabe heritable hyperlipidemia rabbits, ApoE knockout mice
Normal hepatocytes
Healthy hepatocytes
Restoration of amino acid metabolism Cell and gene therapies for Wilson disease Cell therapy model for progressive familial intrahepatic cholestasis Cell and gene therapies for hypoalbuminemia Restoration of ascorbate synthesis Cell transplantation for correcting purine metabolism and urate handling Alleviation of ornithine transcarbamylase deficiency Cell and gene therapies for hypercholesterolemia
*Alb-uPA, urokinase-type plasminogen activator gene driven by the albumin promoter; rag-2, recombination activation gene-2; FAH, fumarylacetoacetate hydrolase; mdr, multidrug resistance; DPPIV, dipeptidyl-peptidase IV; ODS, osteogenic disorder Shionogi rats, Spfash, sparse fur mice. (Modified from Gupta S, Rogler CE. Am J Physiol Gastrointest Liver Physiol 1999;40:G1097–G1102)
181
Section I. Pathophysiology of the Liver
end of the spectrum, including acute or chronic liver failure, as well as other forms of liver damage. In acute liver failure the hope is that cell transplantation will offer critical metabolic support in the short term and promote regeneration of the native liver in the long term. The former should help in bridging patients to OLT, whereas the latter might help avoid OLT altogether, especially if cell transplantation could be coupled with bioartificial liver-assist devices to prolong survival. In chronic liver failure, the immediate hope is that by providing additional synthetic or metabolic liver support cell therapy will ameliorate complications such as coagulopathy, hypoproteinemia, and encephalopathy, and thereby improve the quality of life or prolong survival. Whether cell transplantation could help in regression of hepatic fibrosis and portal hypertension or portosystemic shunting in cirrhosis is less well defined. At the conceptual level, these applications of cell therapy raise many issues concerning the route and site of cell transplantation, the number of cells needed for therapeutic effect, the rapidity with which transplanted cells will provide deficient function, particularly in acute liver failure, and the intrinsic mechanisms that would govern whether transplanted cells overcome microenvironmental or systemic barriers to their engraftment, proliferation, and function. Many of these issues have been addressed in experimental studies. For instance, cells can be transplanted into the liver, most conveniently through catheters into the hepatic artery and less simply into the portal venous system, by approaches involving percutaneous transhepatic, intrasplenic, mesenteric veins or portal vein branches.72 Alternatively, cells can be injected into a variety of extrahepatic sites, particularly the spleen or peritoneal cavity (see 73 for a more detailed discussion). Besides the technical ease of cell administration, one needs to consider the risk of potential complications and the fate of transplanted cells. The data indicate that injection of cells into arterial circulations, including hepatic, splenic, or pulmonary arteries, is ineffective because cells are entrapped in high-flow high-pressure capillaries with embolic complications, including tissue infarction, as well as rapid destruction of transplanted cells.74,75 On the other hand, injection of cells into the sinusoids of the spleen or liver offers better sanctuaries, and transplanted cells can then survive throughout the life of the recipients.76 The use of the spleen as an extrahepatic reservoir of cells has interested many investigators because transplanted cells can be tracked more readily, although the capacity of splenic sinusoids for cells is relatively limited. The peritoneal cavity offers much larger space to accommodate cells, although in view of their anchorage dependence and need for cell–cell interactions hepatocyte survival in this location requires extracellular matrix components or non-parenchymal liver cells.77 Survival of transplanted cells is most effectively achieved when cells are injected into the portal venous system, which distributes cells in hepatic sinusoids throughout the liver. Injection of cells directly into the liver parenchyma leads to limited cell distribution, and transplanted cells could translocate into pulmonary capillaries if the needle accidentally enters a hepatic vein. Injection of hepatocytes into the portal venous system leads to transient, albeit significant, portal hypertension, because transplanted cells occlude portal vein radicles and sinusoids with perturbation of the hepatic microcirculation.78 The larger size of hepatocytes (20–40 mm) compared to that
182
of hepatic sinusoids (3–6 mm) leads to their retention in hepatic sinusoids, with translocation of occasional cells into the pulmonary capillaries in the healthy liver or the liver with an acute injury. In the cirrhotic liver, portosystemic shunting can lead to extensive cell translocations into pulmonary capillaries. However, hepatic microcirculatory alterations regress within several hours after cell transplantation, and hepatocytes entering the pulmonary capillaries do not cause lasting cardiovascular hemodynamic perturbations because they are rapidly destroyed – starting within minutes – in the pulmonary capillaries.74 These considerations, and further issues discussed below, are relevant for developing optimal strategies for the clinical application of cell therapy. To date, over 50 patients with various clinical disorders have been treated with cell transplantation, including autologous hepatocytes from people with cirrhosis or familial hypercholesterolemia, or allogeneic cells on one or more occasions (see 6 and 79 for detailed discussions). In the earliest studies, approximately 2 ¥ 107 to 6 ¥ 108 autologous hepatocytes were transplanted via the spleen, splenic artery, or portal vein in 10 patients with cirrhosis and ascites, without any apparent benefits, although one patient, in whom transplanted cells were detected in the spleen using an iminodiacetic acid radiotracer, improved and returned to work.80 Several patients with advanced liver disease or cirrhosis have been treated in the US with injection of hepatocytes into the splenic artery or the spleen.81 Cell transplantation may have been beneficial in some of these patients, although the data are difficult to interpret, partly because it is not possible to establish correlations between the number of transplanted cells surviving and disease outcomes. In a small series from India of seven patients with acute liver failure and grade III–IV hepatic encephalopathy, 6 ¥ 107 fetal human hepatocytes per kilogram body weight were transplanted intraperitoneally.32 Four of these patients died within 13–48 hours after transplantation, whereas three (43%) recovered. However, the fate of transplanted cells was not established and no further follow-up studies have been reported. More recently, these investigators treated another patient with acute liver failure caused by fatty liver of pregnancy and ascribed complete recovery to the intraperitoneal injection of 3 ¥ 108 fetal human hepatocytes.33 In the US some 20 patients (age range 4 months to 69 years) with acute liver failure have been treated with approximately 3 ¥ 107 to 4 ¥ 1010 fresh or frozen adult human hepatocytes via the spleen or another route.81,82 In some patients intrapulmonary shunting of transplanted cells was observed, along with pulmonary infiltrates and transient hypoxia.82 After cell transplantation, 11 of these 20 patients died (55%), seven subsequently received OLT (35%), and two (10%) recovered completely.79 In some patients the presence of transplanted cells was verified histologically with in situ hybridization to identify sex-mismatched donor cells.82 These early studies seem promising, although disease in patients with acute liver failure is highly variable, which indicates that analysis of larger numbers will be necessary, along with some measure of the transplanted cell mass, to fully define the benefits of cell therapy. People with metabolic deficiency states constitute the most effective examples of cell therapy because the function of transplanted cells can be verified in these conditions by specific assays. Several patients with genetic conditions have undergone hepatocyte transplantation (Table 11-2). In six patients with familial hypercholes-
Chapter 11 STEM CELLS AND HEPATOCYTE TRANSPLANTATION
Table 11-2. Genetic Disorders Amenable to Liver Cell Therapy Disorders already treated in humans a1-Antitrypsin deficiency with decompensated liver disease Familial hypercholesterolemia Crigler–Najjar syndrome, type 1 Ornithine transcarbamylase deficiency Infantile Refsum disease Glycogen storage disease, type 1a Coagulation factor VII deficiency Additional candidate disorders Additional hyperammonemia syndromes Other glycogen storage disorders Congenital hyperbilirubinemia syndromes, including progressive familial intrahepatic cholestasis Hemophilia A Apolipoprotein E deficiency Oxalosis Protoporphyrias Wilson’s disease
terolemia, 1–3 ¥ 109 autologous hepatocytes modified with a retrovirus vector to express LDLR were injected into the portal venous system, following the establishment of critical principles in a rabbit model of the disease.83,84 These patients showed presence of LDLRexpressing transplanted cells in the liver, as well as decreases in serum cholesterol levels in four of them. However, this therapeutic effect was limited, presumably owing to inferior gene transfer and poor survival of transplanted cells in the recipients. In a well studied child with Crigler–Najjar syndrome type 1, intraportal transplantation of 7.5 ¥ 109 adult human hepatocytes resulted in significant decreases in serum bilirubin levels, and the appearance of detectable hepatic UGT1A1 activity, along with conjugated bilirubin in bile, indicating the presence of functioning transplanted cells in the liver.85 However, this patient eventually required an auxiliary liver transplant because the therapeutic benefit of cell transplantation was not sustained beyond several months.79 Similarly, only transient improvements were observed in an infant with severe ornithine transcarbamylase deficiency following intraportal infusion of adult human hepatocytes, and OLT was eventually required.86 In glycogen storage disease type 1a and severe fasting hypoglycemia, intraportal transplantation of 2 ¥ 109 cells produced significant longterm improvement.87 A 4-year-old girl with infantile Refsum’s disease was treated with 2 ¥ 109 fresh and cryopreserved adult hepatocytes over several sessions, and this decreased serum levels of total bile acids, as well as of abnormal dihydroxycoprostanoic acid.88 Survival of transplanted hepatocytes was verified by post-transplant liver biopsy and analysis of sex-mismatched chromosomal sequences. On the other hand, cell transplantation in two brothers with severe coagulation factor VII deficiency was less effective, and these patients were eventually treated with OLT.89 This clinical experience offers multiple insights. First, cells can be safely transplanted in people ranging from those with acute liver failure to those with advanced cirrhosis, from infants to the elderly, and in people with genetic as well as acquired disorders. Second, some ways have been developed to identify transplanted cells in recipients, such as the use of sex-mismatched chromosomal markers, and more recently the use of short tandem repeat sequences.90 Novel
technologies are being developed to permit the identification of transplanted cells through non-invasive imaging modalities, e.g. by introducing genes capable of producing signals that can be imaged.91 Similarly, cells or cell surrogates have been labeled with radioisotopes for short-term tracking in recipients.92,93 These methods are useful for studying the biodistribution of transplanted cells. The ability to verify the presence and abundance of transplanted cells will help develop correlations with therapeutic outcomes in patients, which is most necessary. Third, suitable immunosuppressive regimens have been identified,87 although this area needs to be investigated further, as mechanisms regulating the rejection of allogeneic hepatocytes appear to be different from those involved in the rejection of solid organs.94,95 Fourth, further insights are needed in transplanted cell engraftment and proliferation to improve the therapeutic results obtained so far (see below). Finally, careful and systematic studies in the future, perhaps in multiple centers using identical clinical protocols, including cell transplantation routes, dose of cells, as well as specific cell preparations, will probably be necessary to further establish the potential of cell therapy.
THE BIOLOGICAL BASIS OF TRANSPLANTED CELL ENGRAFTMENT, PROLIFERATION, AND REGULATION OF LIVER REPOPULATION WITH TRANSPLANTED CELLS Work over the past decade and a half has generated novel insights into how transplanted cells engraft, function, and proliferate in the liver. These insights will obviously be critical for advancing clinical applications, for studying the biology of stem/progenitor cells, and for developing chimeric animal models containing human cells to facilitate pharmaceutical development and toxicology testing of novel compounds. To understand how cells engraft in the liver, working models have been developed that take into account the unique organization of the hepatic circulation, the potential for interactions between transplanted cells and various hepatic cell types in the recipient liver, and the need for reorganization of the liver plate structure during the insinuation and integration of transplanted cells in the hepatic parenchyma (Figure 11-2). The first of several critical steps in cell engraftment concerns the deposition of transplanted cells in hepatic sinusoids. This culminates in a blood flow-dependent process of ‘cell embolization’, where the advancement of transplanted cells in sinusoids is determined by the difference in the sizes of cells and sinusoids.78,96 It is advantageous for transplanted cells to reach distal sinusoids in the liver lobule because cells left behind in portal vein radicles cannot enter the liver parenchyma and are largely cleared by phagocytes. On the other hand, transplanted hepatocytes can recognize specific adhesion molecules and extracellular matrix receptors, such as fibronectin receptors, on liver sinusoidal endothelial cells. In the meantime, entrapment of transplanted cells in liver sinusoids rapidly results in the activation of microcirculatory perturbations and the onset
183
Section I. Pathophysiology of the Liver
B A
C
Figure 11-2. Illustration of transplanted cell engraftment and proliferation in the liver. Shown are the key features of transplanted cell biology as demonstrated by studies using DPPIV-deficient F344 recipient rats. (A) Integration of a transplanted cell (asterisk), with the development of dot-like hybrid gap junctions (arrowheads) that join the transplanted cell with native cells in the liver parenchyma. (B) Transplanted cells forming a network of hybrid bile canaliculi with contributions from DPPIV-positive domains (red color, arrowheads) and ATPase-positive domains from native hepatocytes (brown color, arrows). (C) Expanding clusters of transplanted cells in a rat pretreated with retrorsine and partial hepatectomy. Note that some clusters of transplanted cells are far smaller than others. Also, proliferation in transplanted cells leads to progressive enlargement of confluent areas containing transplanted cells. Native hepatocytes exhibit megalocytosis and polyploid nuclei as evidence for liver injury induced by retrorsine and partial hepatectomy. Transplanted cells were identified in the liver with DPPIV histochemistry, seen with a red reaction product. Gap junctions were visualized by immunostaining for connexin using a diaminobenzidine substrate of peroxidase. Bile canaliculi in native cells were visualized by ATPase histochemical staining. A and B, methylgreen counterstain; C, hematoxylin counterstain.
184
Chapter 11 STEM CELLS AND HEPATOCYTE TRANSPLANTATION
of hepatic ischemia in affected regions.96,97 For instance, video microscopy analysis in live animals demonstrated that sinusoidal blood flow may cease immediately after cell transplantation, although blood flow begins to be restored after several minutes to hours, although with periods of variable ischemia–reperfusion and evidence for some injury across the recipient liver.97 This appears to be an inevitable component of cell transplantation in the liver and is accompanied by multiple perturbations, including activation of adjacent Kupffer cells, hepatic stellate cells, endothelial cells and native hepatocytes.98–100 Ischemic damage to the hepatic endothelium would help transplanted cells enter the space of Disse and then the liver parenchyma.101 On the other hand, ischemia–reperfusion is a potent stimulus for Kupffer cell activation, which results in the release of several inflammatory cytokines, including TNF-a, interleukin-6 etc. Kupffer cells can play an important role in regulating the engraftment of transplanted cells.99 For instance, activation of Kupffer cells and the generation of pro-oxidant stress in liver sinusoidal endothelial cells can help permeabilize the hepatic endothelial barrier, which would facilitate the entry of transplanted cells into the space of Disse.101 On the other hand, activated Kupffer cells phagocytose significant fractions of transplanted cells,99 which impairs overall transplanted cell engraftment, such that less than 20% of the transplanted cells eventually survive in the liver.98 The process of cell entry in the liver is facilitated by additional factors. For instance, activated hepatic stellate cells and hepatocytes release soluble factors, including vascular endothelial growth factor (VEGF), matrix-type metalloproteinases, and other molecules that help permeabilize endothelial cells in the vicinity of transplanted cells.98 VEGF is released within 6–8 hours after cell transplantation from both transplanted cells and native hepatocytes. This occurs before the demonstrated requirement of 16–20 hours for transplanted cells to penetrate through the endothelial barrier. During isolation of liver cells, plasma membrane structures, including tight junctions, gap junctions, and bile canaliculi, must be severed, but must subsequently be restored. Therefore, the next process in transplanted cell engraftment concerns the restoration of cell polarity. Finally, integration of transplanted cells in the liver parenchyma must be completed. Analysis of these processes has shown that gap junctions and bile canaliculi are reconstituted over 3–7 days after cell transplantation, when transplanted cells and native hepatocytes begin to exhibit hybrid plasma membrane structures. This is a critical process, because transplanted cells then become indistinguishable from native hepatocytes and can begin to manifest normal functions, including bile excretion.102,103 Identification of this series of critical events following cell transplantation has provided ways to design specific manipulations aimed at enhancing transplanted cell engraftment. For instance, initial deposition of transplanted cells in the liver lobule can be improved by the use of hepatic sinusoidal dilators, e.g. nitroglycerine, phentolamine, or sodium nitroprusside.96,104 This protects the liver from ischemia and microcirculatory disruption, helps accelerate the deposition of cells from the spleen into the liver, and promotes the entry of transplanted cells into distal liver sinusoids. Similarly, Kupffer cell activation can be prevented by the use of pharmacological approaches.99 Prior depletion of Kupffer cells significantly improves transplanted cell engraftment as well as the kinetics of liver repopulation. This effect can even be mediated by the inhibition of TNF-
a activity by clinically effective drugs such as etanercept or infliximab (unpublished observations), which will have clinical applications. Another significant mechanism that has been identified for improving transplanted cell engraftment concerns damage to the hepatic endothelium. In early studies, the use of cyclophosphamide was shown to be effective for disrupting endothelial cells, which accelerated the entry of transplanted cells into the space of Disse and improved cell engraftment.101 Subsequently, additional mechanisms have been identified to induce endothelial damage, e.g. doxorubicin, which will be suitable for clinical use, and monocrotaline, a pyrrolizidine alkaloid, which will be useful for animal studies. Additional mechanisms are currently being investigated to improve transplanted cell engraftment, e.g. the use of specific receptors to improve binding of transplanted cells to liver sinusoidal endothelial cells. Also, intrinsic differences in the properties of cells, e.g. the display of specific proteins on the cell surface, or the greater capacity of cell subsets to proliferate, could be helpful in promoting cell engraftment or proliferation.105–107 Moreover, genetic manipulation to introduce protective genes has been effective in some circumstances.108,109 A major advantage of superior transplanted cell engraftment concerns acceleration of the kinetics of liver repopulation, which will obviously be helpful for clinical applications. It should be noted that transplanted hepatocytes are able to engraft and proliferate in the liver despite the presence of significant fibrosis or acute liver injury,110,111, although proliferation in transplanted cells is delayed in the acutely injured liver by several days, owing to the additional time required to complete cell engraftment.110 When cell engraftment is completed, transplanted cells can survive lifelong in syngeneic experimental animals, where cellular rejection is not an issue.76,102 Throughout that period transplanted cells exhibit normal hepatic functions and respond to mitogenic stimuli in a physiological fashion. Moreover, in the normal liver transplanted cells do not proliferate. These findings have practical implications concerning the creation of suitable therapeutic masses of transplanted cells. For instance, transplantation of 1 ¥ 107 cells in an adult rat or 1 ¥ 1010 cells in an adult person is considered to be the equivalent of not more than 1–2% of total liver cells. As only a fraction of this cell mass would engraft and survive long term, it is essential to find ways to augment the transplanted cell mass. This could potentially be accomplished by transplanting cells repeatedly, which has been found to be safe and effective, with reconstitution of 5–7% of the liver mass following transplantation of cells on three separate occasions in rats.112 On the other hand, induction of proliferation in transplanted cells represents an alternative approach that will be particularly appropriate for using stem/progenitor cells in small numbers. A variety of animal studies indicated that the creation of instability in the liver parenchyma, such that native hepatocytes are at a survival or proliferation disadvantage compared to transplanted cells, promotes liver repopulation (see 113 for a review). Several mouse models verified this principle, including alb-uPA mice, FAH mice, alb-HSV-TK mice, and AdMad mice, where toxic transgenes serve to deplete native hepatocytes, such that transplanted healthy cells can proliferate extensively and repopulate large areas of the liver. Similarly, when transgenic cells expressing the antiapoptotic human Bcl-2 gene
185
Section I. Pathophysiology of the Liver
are transplanted into mice susceptible to Fas ligand-mediated apoptosis, followed by administration of the mouse-specific J0-2 antibody, transplanted cells can proliferate significantly.114 The principle of selective hepatic ablation also applies to chemical toxins, such as carbon tetrachloride, which induces proliferation in transplanted cells located in periportal areas of the liver lobule, away from perivenous areas susceptible to carbon tetrachloride injury.115 Moreover, the chemical retrorsine, which is another DNAbinding pyrrolizidine alkaloid, similar to monocrotaline, was effective in inhibiting proliferation in hepatocytes, particularly when combined with additional manipulations, such as two-thirds partial hepatectomy, repeated administration of the thyroid hormone triiodothyronine (T3), as well as carbon tetrachloride.116,117 In these situations, proliferation in transplanted cells was markedly accelerated and the liver was repopulated virtually completely with transplanted cells. These manipulations were effective because genotoxic damage in retrorsine-treated hepatocytes produced susceptibility to additional oxidative injury produced by partial hepatectomy, T3, or carbon tetrachloride, leading to progressive repopulation with trans-
HSC activation
planted cells over several weeks. In contrast, when cells were not transplanted in retrorsine-treated animals, liver repopulation was completed by the activation of endogenous liver cell populations.118 The use of retrorsine or monocrotaline has been effective in developing further insights into the biology of transplanted cells. However, these chemicals are not suited to clinical applications in view of their potential for oncogenicity. On the other hand, recent studies using genotoxic damage with hepatic radiation has shown efficacy in inducing transplanted cell proliferation.119 In initial studies, radiation was found to induce extensive liver injury in animals subjected to partial hepatectomy, which again suggested the role of cumulative genotoxic damage in this process. However, additional studies demonstrated that the role of partial hepatectomy in this process was to activate oxidative DNA damage, thus adding ischemia–reperfusion injury to the radiation injury model.120,121 Hepatic ischemia–reperfusion is a potent source of oxidative stress and has been used to treat liver cancer in people. This combination of ischemia–reperfusion and radiation has been highly effective in promoting transplanted cell proliferation in the
HSC activation
ECM and membrane reconstitution
Endothelial loss Ischemia reperfusion
Liver sinusoid
VEGF, etc.
Liver cell plate Kupffer cell activation Early hepatic ischemia and cell activators (minutes to hours)
Endothelial cell layer and space of Disse Endothelial disruption, cell entry and integration in liver (hours to days)
Adhesion molecule Cell adhesion and other early events (minutes to hours)
Figure 11-3. A working model outlining critical steps during transplanted cell engraftment in the liver. Immediately after the arrival of transplanted cells in liver sinusoids (arrow on the right) cells become entrapped in distal sinusoids (shown on extreme left), which temporarily blocks sinusoidal blood flow, activates portal hypertension, and perturbs the hepatic microcirculation with the onset of ischemia–reperfusion. This promptly activates Kupffer cells and hepatic stellate cells, which release multiple cytokines and chemokines, including mediators of inflammation. Simultaneously, transplanted cells adhere to the hepatic endothelium through specific cell surface molecules and receptors (shown on the right-hand side). Activation of hepatic endothelial cells results from multiple mechanisms, including oxidative damage, permeabilization of cells through the release of VEGF, various matrix metalloproteinases, and other molecules. Sixteen to 20 hours after cell transplantation, hepatocytes begin to enter the space of Disse (middle portion of the figure) and insinuate themselves between native hepatocytes in periportal areas. Cells entrapped in portal vein radicles or sinusoids are largely cleared by 24–48 hours. Subsequently, transplanted cells begin to become integrated in the liver parenchyma, with remodeling of the liver plate structure, which is facilitated by the coordinated release of various matrix metalloproteinases, as well as tissue inhibitors of matrix metalloproteinases, e.g. TIMP-1. Completion of plasma membrane reconstitution results in recovery of cell polarity and the formation of conjoint structures, such as gap junctions and bile canaliculi, which regain functional integrity 3–7 days after cell transplantation.
186
Chapter 11 STEM CELLS AND HEPATOCYTE TRANSPLANTATION
– Mitogenic self-stimulus – Loss of cell cycle suppressors
of cell growth B Induced regulation of
transplanted cell growth
– Necrosis / apoptosis in native cells – Healthy transplanted cells – Combination of engraftment controls and proliferation induction
Loss of native cells
G
Liver repopulation (%)
A Intrinsic control
140 120
50 Gy RT + IRP
100
30 Gy RT + IRP
80 60 40 Projected doubling
20 0 Proliferation in transplanted cells
C
D
E
0
1 2 3 Time after cell transplantation (months)
F
Figure 11-4. Mechanisms regulating liver repopulation with cells. As discussed in the text, transplanted cell proliferation can be induced by regulating intrinsic cell cycling properties (A) or by manipulating microenvironmental conditions (B), which is more effective. Panels C–F show transplanted cells (red color) in the liver of syngeneic dipeptidyl peptidase IV (DPPIV)-deficient F344 rats 3 months after cell transplantation. (C) A control animal without any treatment before cell transplantation with occasional transplanted cells (arrow). (D) A rat treated with liver ischemia–reperfusion before cell transplantation. (E) A rat treated with 50 Gy liver irradiation, showing some transplanted cell proliferation. (F) Treatment with ischemia–reperfusion plus radiation led to extensive liver repopulation. Tissues were subjected to DPPIV histochemistry with hematoxylin counterstaining. (G) The chart shows liver repopulation kinetics with projected doublings of transplanted cells shown. The shaded area at the top shows liver repopulation after 50 Gy irradiation and ischemia–reperfusion, and the hatched area towards the bottom shows liver repopulation after 30 Gy irradiation and ischemia–reperfusion. The findings suggested nine doublings in the transplanted cell mass owing to the asynchronous loss of native hepatocytes over 12 weeks (in part from Malhi H et al. Cell transplantation after oxidative hepatic preconditioning with radiation and ischemia–reperfusion leads to extensive liver repopulation. Proc Natl Acad Sci USA 2002;99:13114–13119).
rat liver.121 The process of liver repopulation is gradual, suggesting an asynchronous loss of native hepatocytes, and is completed in the rat over several weeks, during which transplanted hepatocytes are required to undergo an estimated nine population doublings (Figures 11-3 and 11-4). Therefore, such a manipulation should be clinically applicable, although studies in larger animals will be necessary to further verify the safety and efficacy of radiation-based approaches. One scenario will concern conformal targeting of the liver with radiation, followed by transcatheter occlusion of a portal vein branch and the administration of cells through the same catheter after another day or two. In this way, it should be possible to develop further clinical protocols for treating patients with cell therapy using the induction of an optimal transplanted cell mass.
FURTHER CONSIDERATIONS FOR DEVELOPING CELL THERAPY APPLICATIONS Transplantation of human cells into immunodeficient mice offers ways to develop highly useful additional animal models, including
for addressing the basic biology of stem cell populations, characterization and assessment of the viability of human cell preparations, as well as the development of chimeric mice for various purposes.122 For instance, such mice will be most valuable for use as models of hepatitis B or C viruses, including for testing drugs or biological modifiers.123–125 However, it is difficult to repopulate the mouse liver with human cells, for unclear reasons, although many laboratories are actively addressing this issue. It appears that despite the survival of transplanted human cells in the mouse liver, such cells show relatively limited proliferation. Whether the differences between the survival and proliferation of human versus animal cells in immunodeficient mice are due to species-specific fundamental requirements of specific cell types is unclear, but may extend to specific hepatotrophic factors or other requirements.126 More than one liver cell type may need to be reconstituted.127 On the other hand, it is appropriate to contemplate whether cell therapy could be applied to large groups of patients instead of small numbers of people with less common or rare conditions. For instance, as virus hepatitis constitutes the single largest liver problem, affecting some 500 million worldwide, a large number of who will develop chronic liver failure, it is appropriate to consider whether cell therapy could be applied to them. This would require
187
Section I. Pathophysiology of the Liver
cells that resist virus replication. Recently, genetic strategies have been developed to interrupt hepatitis B or C replication,128,129 which could be utilized to transduce cells and begin approaching cell therapy in people following OLT and disease recurrence.
REFERENCES 1. Fausto N. Liver regeneration and repair: hepatocytes, progenitor cells, and stem cells. Hepatology 2004;39:1477–1487. 2. Alison MR. Characterization of the differentiation capacity of rat-derived hepatic stem cells. Semin Liver Dis 2003;23:325–336. 3. Zhao R, Duncan SA. Embryonic development of the liver. Hepatology 2005;41:956–967. 4. Habener J. A perspective on pancreatic stem/progenitor cells. Pediatr Diabetes 2004;5(Suppl 2):29–37. 5. Overturf K, al-Dhalimy M, Ou CN, Finegold M, Grompe M. Serial transplantation reveals the stem-cell-like regenerative potential of adult mouse hepatocytes. Am J Pathol 1997;151:1273–1280. 6. Horslen SP, Fox IJ. Hepatocyte transplantation. Transplantation 2004;77:1481–1486. 7. Haghighi KS, Woon WW, Akhter J, et al. A new source of hepatocytes for transplantation. Transplant Proc 2004;36:2466–2468. 8. Thomson JA, Itskovitz-Eldor J, Shapiro SS, et al. Embryonic stem cell lines derived from human blastocysts. Science 1998;282:1145–1147. 9. Lee JB, Kim JM, Kim SJ, et al. Comparative characteristics of three human embryonic stem cell lines. Mol Cell 2005;19:31–38. 10. Hwang WS, Roh SI, Lee BC, et al. Patient-specific embryonic stem cells derived from human SCNT blastocysts. Science 2005;308:1777–1783. 11. Shamblott MJ, Axelman J, Littlefield JW, et al. Human embryonic germ cell derivatives express a broad range of developmentally distinct markers and proliferate extensively in vitro. Proc Natl Acad Sci USA 2001;98:113–118. 12. Chinzei R, Tanaka Y, Shimizu-Saito K, et al. Embryoid-body cells derived from a mouse embryonic stem cell line show differentiation into functional hepatocytes. Hepatology 2002;36:22–29. 13. Yin Y, Lim YK, Salto-Tellez M, et al. AFP(+), ESC-derived cells engraft and differentiate into hepatocytes in vivo. Stem Cells 2002;20:338–346. 14. Ishizaka S, Shiroi A, Kanda S, et al. Development of hepatocytes from ES cells after transfection with the HNF3beta gene. FASEB J 2002;16:1444–1446. 15. Rambhatla L, Chiu CP, Kundu P, Peng Y, Carpenter MK. Generation of hepatocyte-like cells from human embryonic stem cells. Cell Transplant 2003;12:1–11. 16. Lavon N, Yanuka O, Benvenisty N. Differentiation and isolation of hepatic-like cells from human embryonic stem cells. Differentiation 2004;72:230–238. 17. Teratani T, Yamamoto H, Aoyagi K, et al. Direct hepatic fate specification from mouse embryonic stem cells. Hepatology 2005; 41:836–846. 18. Ku HT, Zhang N, Kubo A, et al. Committing embryonic stem cells to early endocrine pancreas in vitro. Stem Cells 2004;22:1205–1217. 19. Kania G, Blyszczuk P, Jochheim A, Ott M, Wobus AM. Generation of glycogen- and albumin-producing hepatocyte-like cells from embryonic stem cells. Biol Chem 2004;385:943–953. 20. Shirahashi H, Wu J, Yamamoto N, et al. Differentiation of human and mouse embryonic stem cells along a hepatocyte lineage. Cell Transplant 2004;13:197–211.
188
21. Sato N, Meijer L, Skaltsounis L, Greengard P, Brivanlou AH. Maintenance of pluripotency in human and mouse embryonic stem cells through activation of Wnt signaling by a pharmacological GSK-3-specific inhibitor. Nature Med 2004;10:55–63. 22. Golan-Mashiach M, Dazard JE, Gerecht-Nir S, et al. Design principle of gene expression used by human stem cells: implication for pluripotency. FASEB J 2005;19:147–149. 23. Malhi H, Irani AN, Gagandeep S, Gupta S. Isolation of human progenitor liver epithelial cells with extensive replication capacity and differentiation into mature hepatocytes. J Cell Sci 2002;115:2679–2688. 24. Lazaro CA, Croager EJ, Mitchell C, et al. Establishment, characterization, and long-term maintenance of cultures of human fetal hepatocytes. Hepatology 2003;38:1095–1106. 25. Badve S, Logdberg L, Sokhi R, et al. An antigen reacting with das-1 monoclonal antibody is ontogenically regulated in diverse organs including liver and indicates sharing of developmental mechanisms among cell lineages. Pathobiology 2000;68:76–86. 26. Zahler MH, Irani A, Malhi H, et al. The application of a lentiviral vector for gene transfer in fetal human hepatocytes. J Gene Med 2000;2:186–193. 27. Wege H, Le HT, Chui MS, et al. Telomerase reconstitution immortalizes human fetal hepatocytes without disrupting their differentiation potential. Gastroenterology 2003;124:432–444. 28. Dabeva MD, Petkov PM, Sandhu J, et al. Proliferation and differentiation of fetal liver epithelial progenitor cells after transplantation into adult rat liver. Am J Pathol 2000;156:2017–2031. 29. Sierra E, Maganto P, Codesal J, et al. Liver gene expression and increase in albumin synthesis by fetal hepatocytes transplanted into analbuminemic rats. Life Sci 2000;67:2417–2432. 30. Sandhu, JS, Petkov PM, Dabeva MD, Shafritz DA. Stem cell properties and repopulation of the rat liver by fetal liver epithelial progenitor cells. Am J Pathol 2001;159:1323– 1334. 31. Cantz T, Zuckerman DM, Burda MR, et al. Quantitative gene expression analysis reveals transition of fetal liver progenitor cells to mature hepatocytes after transplantation in uPA/RAG-2 mice. Am J Pathol 2003;162:37–45. 32. Habibullah CM. Hepatocyte transplantation: Need for liver cell bank. Trop Gastroenterol 1992;13:129–131. 33. Khan AA, Habeeb A, Parveen N, et al. Peritoneal transplantation of human fetal hepatocytes for the treatment of acute fatty liver of pregnancy: a case report. Trop Gastroenterol 2004;25:141–143. 34. Kubota H, Reid LM. Clonogenic hepatoblasts, common precursors for hepatocytic and biliary lineages, are lacking classical major histocompatibility complex class I antigen. Proc Natl Acad Sci USA 2000;97:12132–12137. 35. Suzuki A, Zheng YW, Fukao K, Nakauchi H, Taniguchi H. Liver repopulation by c-Met-positive stem/progenitor cells isolated from the developing rat liver. Hepatogastroenterology 2004;51:423–426. 36. Nierhoff D, Ogawa A, Oertel M, Chen YQ, Shafritz DA. Purification and characterization of mouse fetal liver epithelial cells with high in vivo repopulation capacity. Hepatology 2005;42:130–139. 37. Zalzman M, Gupta S, Giri RK, et al. Reversal of hyperglycemia in mice using human expandable insulin-producing cells differentiated from fetal liver progenitor cells. Proc Natl Acad Sci USA 2003; 100:7253–7258. 38. Saxena R, Theise N. Canals of Hering: recent insights and current knowledge. Semin Liver Dis 2004;24:43–48. 39. Muller-Borer BJ, Cascio WE, Anderson PA, et al. Adult-derived liver stem cells acquire a cardiomyocyte structural and functional phenotype ex vivo. Am J Pathol 2004;165: 135–145.
Chapter 11 STEM CELLS AND HEPATOCYTE TRANSPLANTATION
40. Yasui O, Miura N, Terada K, et al. Isolation of oval cells from Long–Evans Cinnamon rats and their transformation into hepatocytes in vivo in the rat liver. Hepatology 1997;25:329–334. 41. Wang X, Foster M, Al-Dhalimy M, et al. The origin and liver repopulating capacity of murine oval cells. Proc Natl Acad Sci USA 2003;100(Suppl 1):11881–11888. 42. Menthena A, Deb N, Oertel M, et al. Bone marrow progenitors are not the source of expanding oval cells in injured liver. Stem Cells 2004;22:1049–1061. 43. Terada S, Matsuura K, Enosawa S, et al. Inducing proliferation of human amniotic epithelial (HAE) cells for cell therapy. Cell Transplant 2000;9:701–704. 44. Takashima S, Ise H, Zhao P, Akaike T, Nikaido T. Human amniotic epithelial cells possess hepatocyte-like characteristics and functions. Cell Struct Funct 2004;29:73–84. 45. Yen BL, Huang HI, Chien CC, et al. Isolation of multipotent cells from human term placenta. Stem Cells 2005;23:3–9. 46. Petersen BE, Bowen WC, Patrene KD, et al. Bone marrow as a potential source of hepatic oval cells. Science 1999;284:1168–1170. 47. Theise ND, Badve S, Saxena R, et al. Derivation of hepatocytes from bone marrow cells in mice after radiation-induced myeloablation. Hepatology 2000;31:235–240. 48. Alison MR, Poulsom R, Jeffery R, et al. Hepatocytes from nonhepatic adult stem cells. Nature 2000;406:257. 49. Lagasse E, Connors H, Al Dhalimy M, et al. Purified hematopoietic stem cells can differentiate into hepatocytes in vivo. Nature Med 2000; 6:1229–1234. 50. Jiang Y, Jahagirdar BN, Reinhardt RL, et al. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 2002;418:41–49. 51. Wang X, Montini E, Al-Dhalimy M, et al. Kinetics of liver repopulation after bone marrow transplantation. Am J Pathol 2002;161:565–574. 52. Krause DS, Theise ND, Collector MI, et al. Multi-organ, multi-lineage engraftment by a single bone marrow-derived stem cell. Cell 2001;105:369–377. 53. Keene CD, Ortiz-Gonzalez XR, Jiang Y, et al. Neural differentiation and incorporation of bone marrow-derived multipotent adult progenitor cells after single cell transplantation into blastocyst stage mouse embryos. Cell Transplant 2003;12:201–213. 54. Kakinuma S, Tanaka Y, Chinzei R, et al. Human umbilical cord blood as a source of transplantable hepatic progenitor cells. Stem Cells 2003;21:217–227. 55. Wang X, Ge S, McNamara G, et al. Albumin-expressing hepatocyte-like cells develop in the livers of immunedeficient mice that received transplants of highly purified human hematopoietic stem cells. Blood 2003;101:4201– 4208 56. Newsome PN, Johannessen I, Boyle S, et al. Human cord bloodderived cells can differentiate into hepatocytes in the mouse liver with no evidence of cellular fusion. Gastroenterology 2003;124:1891–1900. 57. Tanabe Y, Tajima F, Nakamura Y, et al. Analyses to clarify rich fractions in hepatic progenitor cells from human umbilical cord blood and cell fusion. Biochem Biophys Res Commun 2004;324:711–718. 58. Turrini P, Monego G, Gonzalez J, et al. Human hepatocytes in mice receiving pre-immune injection with human cord blood cells. Biochem Biophys Res Commun 2005;326:66–73. 59. Wagers AJ, Sherwood RI, Christensen JL, Weissman IL. Little evidence for developmental plasticity of adult hematopoietic stem cells. Science 2002;297:2256–2259. 60. Kanazawa Y, Verma IM. Little evidence of bone marrow-derived hepatocytes in the replacement of injured liver. Proc Natl Acad Sci USA 2003;100(Suppl 1):11850–11853.
61. Cantz T, Sharma AD, Jochheim-Richter A, et al. Reevaluation of bone marrow-derived cells as a source for hepatocyte regeneration. Cell Transplant 2004;13:659–666. 62. Vassilopoulos G, Wang PR, Russell DW. Transplanted bone marrow regenerates liver by cell fusion. Nature 2003;422:901–904. 63. Willenbring H, Bailey AS, Foster M, et al. Myelomonocytic cells are sufficient for therapeutic cell fusion in liver. Nature Med 2004;10:744–748. 64. Jang YY, Collector MI, Baylin SB, Diehl AM, Sharkis SJ. Hematopoietic stem cells convert into liver cells within days without fusion. Nature Cell Biol 2004;6:532–539. 65. Kajstura J, Rota M, Whang B, et al. Bone marrow cells differentiate in cardiac cell lineages after infarction independently of cell fusion. Circ Res 2005;96:127–137. 66. Aldeguer X, Debonera F, Shaked A, et al. Interleukin-6 from intrahepatic cells of bone marrow origin is required for normal murine liver regeneration. Hepatology 2002;35:40–48. 67. Bailey AS, Jiang S, Afentoulis M, et al. Transplanted adult hematopoietic stems cells differentiate into functional endothelial cells. Blood 2004;103:13–19. 68. Baba S, Fujii H, Hirose T, et al. Commitment of bone marrow cells to hepatic stellate cells in mouse. J Hepatol 2004;40:255–260. 69. Kobayashi N, Fujiwara T, Westerman KA, et al. Prevention of acute liver failure in rats with reversibly immortalized human hepatocytes. Science 2000;287:1258–1262. 70. Yuan RH, Ogawa A, Ogawa E, et al. p27Kip1 inactivation provides a proliferative advantage to transplanted hepatocytes in DPPIV/Rag2 double knockout mice after repeated host liver injury. Cell Transpl 2003;12:907–919. 71. Mikula M, Fuchs E, Huber H, et al. Immortalized p19ARF null hepatocytes restore liver injury and generate hepatic progenitors after transplantation. Hepatology 2004;39:628–634. 72. Darwish AA, Sokal E, Stephenne X, et al. Permanent access to the portal system for cellular transplantation using an implantable port device. Liver Transpl 2004;10:1213–1215. 73. Gupta S, Rajvanshi P, Bhargava KK, Kerr A. Hepatocyte transplantation: progress toward liver repopulation. Prog Liv Dis 1996;14:199–222. 74. Rajvanshi P, Fabrega A, Bhargava KK, et al. Rapid clearance of hepatocytes from pulmonary capillaries in rats allows development of surrogates for testing safety of liver repopulation. J Hepatol 1999;30:299–310. 75. Nagata H, Ito M, Shirota C, et al. Route of hepatocyte delivery affects hepatocyte engraftment in the spleen. Transplantation 2003;76:732–734. 76. Sokhi RP, Rajvanshi P, Gupta S. Transplanted reporter cells help in defining onset of hepatocyte proliferation during the life of F344 rats. Am J Physiol Gastrointest Liver Physiol 2000; 279: G631–G640. 77. Ogawa K, Ochoa ER, Borenstein J, Tanaka K, Vacanti JP. The generation of functionally differentiated, three-dimensional hepatic tissue from two-dimensional sheets of progenitor small hepatocytes and nonparenchymal cells. Transplantation 2004;77:1783–1789. 78. Muraca M, Neri D, Parenti A, et al. Intraportal hepatocyte transplantation in the pig: hemodynamic and histopathological study. Transplantation 2002;73:890–896. 79. Fox IJ, Roy Chowdhury J. Hepatocyte transplantation. J Hepatol 2004;40:878–886. 80. Mito M, Kusano M. Hepatocyte transplantation in man. Cell Transplant 1993;2:65–74. 81. Strom SC, Fisher RA, Thompson MT, et al. Hepatocyte transplantation as a bridge to orthotopic liver transplantation in terminal liver failure. Transplantation 1997;63:559–569. 82. Bilir BM, Guinette D, Karrer F, et al. Hepatocyte transplantation in acute liver failure. Liver Transpl 2000;6:32–40.
189
Section I. Pathophysiology of the Liver
83. Grossman M, Rader DJ, Muller DW, et al. A pilot study of ex vivo gene therapy for homozygous familial hypercholesterolaemia. Nature Med 1995;1:1148–1154. 84. Grossman M, Rader DJ, Muller DWM, et al. A pilot study of ex vivo gene therapy for homozygous familiar hypercholesterolaemia. Nature Med 1995;1:1148–1154. 85. Fox IJ, Chowdhury JR, Kaufman SS, et al. Treatment of the Crigler–Najjar syndrome type I with hepatocyte transplantation. N Engl J Med 1998;338:1422–1426. 86. Horslen SP, McCowan TC, Goertzen TC, et al. Isolated hepatocyte transplantation in an infant with a severe urea cycle disorder. Pediatrics 2003;111:1262–1267. 87. Muraca M, Gerunda G, Neri D, et al. Hepatocyte transplantation as a treatment for glycogen storage disease type 1a. Lancet 2002;359:317–318. 88. Sokal EM, Smets F, Bourgois A, et al. Hepatocyte transplantation in a 4-year-old girl with peroxisomal biogenesis disease: technique, safety, and metabolic follow-up. Transplantation 2003;76:735–738. 89. Dhawan A, Mitry RR, Hughes RD, et al. Hepatocyte transplantation for inherited factor VII deficiency. Transplantation 2004;78:1812–1814. 90. Mas VR, Maluf DG, Thompson M, Ferreira-Gonzalez A, Fisher RA. Engraftment measurement in human liver tissue after liver cell transplantation by short tandem repeats analysis. Cell Transplant 2004;13:231–236. 91. Penuelas I, Mazzolini G, Boan JF, et al. Positron emission tomography imaging of adenoviral-mediated transgene expression in liver cancer patients. Gastroenterology 2005;128:1787–1795. 92. Bohnen NI, Charron M, Reyes J, et al. Use of indium-111labeled hepatocytes to determine the biodistribution of transplanted hepatocytes through portal vein infusion. Clin Nucl Med 2000;25:447–450. 93. Schneider A, Attaran M, Gratz KF, et al. Intraportal infusion of 99mtechnetium-macro-aggregrated albumin particles and hepatocytes in rabbits: assessment of shunting and portal hemodynamic changes. Transplantation 2003;75:296–302. 94. Bumgardner GL, Gao D, Li J, Bickerstaff A, Orosz CG. MHCidentical heart and hepatocyte allografts evoke opposite immune responses within the same host. Transplantation 2002;74:855–864. 95. Reddy B, Gupta S, Chuzhin Y, et al. The effect of CD28/B7 blockade on alloreactive T and B cells after liver cell transplantation. Transplantation 2001;71:801–811. 96. Slehria S, Rajvanshi P, Ito Y, et al. Hepatic sinusoidal vasodilators improve transplanted cell engraftment and ameliorate microcirculatory perturbations in the liver. Hepatology 2002;35:1320–1328. 97. Gupta S, Rajvanshi P, Malhi H, et al. Cell transplantation causes loss of gap junctions and activates GGT expression permanently in host liver. Am J Physiol Gastrointest Liver Physiol 2000; 279: G815–G826. 98. Gupta S, Rajvanshi P, Sokhi RP, et al. Entry and integration of transplanted hepatocytes in liver plates occur by disruption of hepatic sinusoidal endothelium. Hepatology 1999;29:509–519. 99. Joseph B, Malhi H, Bhargava KK, et al. Kupffer cells participate in early clearance of syngeneic hepatocytes transplanted in the rat liver. Gastroenterology 2002;123:1677–1685. 100. Wilhelm A, Leister I, Sabandal P, et al. Acute impairment of hepatic microcirculation and recruitment of nonparenchymal cells by intrasplenic hepatocyte transplantation. J Pediatr Surg 2004;39:1214–1219. 101. Malhi H, Annamaneni P, Slehria S, et al. Cyclophosphamide disrupts hepatic sinusoidal endothelium and improves transplanted cell engraftment in rat liver. Hepatology 2002;36:112–121. 102. Gupta S, Rajvanshi P, Sokhi R, et al. Position-specific gene expression in the liver lobule is directed by the
190
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
microenvironment and not by the previous cell differentiation state. J Biol Chem 1999;274:2157–2165. Gupta S, Rajvanshi P, Lee C-D. Integration of transplanted hepatocytes in host liver plates demonstrated with dipeptidyl peptidase IV deficient rats. Proc Natl Acad Sci USA 1995;92:5860–5864. Sigot V, Mediavilla MG, Furno G, Rodriguez JV, Guibert EE. A simple and effective method to improve intrasplenic rat hepatocyte transplantation. Cell Transplant 2004;13:775–781. Katayama S, Tateno C, Asahara T, Yoshizato K. Size-dependent in vivo growth potential of adult rat hepatocytes. Am J Pathol 2001;158:97–105. Ise H, Nikaido T, Negishi N, et al. Effective hepatocyte transplantation using rat hepatocytes with low asialoglycoprotein receptor expression. Am J Pathol 2004;165:501–510. Walldorf J, Aurich H, Cai H, et al. Expanding hepatocytes in vitro before cell transplantation: donor age-dependent proliferative capacity of cultured human hepatocytes. Scand J Gastroenterol 2004;39:584–593. Held PK, Olivares EC, Aguilar CP, et al. In vivo correction of murine hereditary tyrosinemia type i by varphiC31 integrasemediated gene delivery. Mol Ther 2005;11:399–408. Wang J, Li W, Min J, Ou Q, Chen J. Fas siRNA reduces apoptotic cell death of allogeneic-transplanted hepatocytes in mouse spleen. Transplant Proc 2003;35:1594–1595. Gupta S, Rajvanshi P, Irani AN, Palestro CJ, Bhargava KK. Integration and proliferation of transplanted cells in hepatic parenchyma following D-galactosamine-induced acute injury in F344 rats. J Pathol 2000;190:203–210. Gagandeep S, Rajvanshi P, Sokhi R, et al. Transplanted hepatocytes engraft, survive and proliferate in the liver of rats with carbon tetrachloride-induced cirrhosis. J Pathol 2000;191:78–85. Rajvanshi P, Kerr A, Bhargava KK, Burk RD, Gupta S. Efficacy and safety of repeated hepatocyte transplantation for significant liver repopulation in rodents. Gastroenterology 1996;111:1092–1102. Gupta S, Inada M, Joseph B, Kumaran V, Benten D. Emerging insights into liver-directed cell therapy for genetic and acquired disorders. Transplant Immunol 2004;12:289–302. Chen SJ, Tazelaar J, Wilson JM. Selective repopulation of normal mouse liver by hepatocytes transduced in vivo with recombinant adeno-associated virus. Hum Gene Ther 2001;12:45–50. Gupta S, Rajvanshi P, Aragona E, et al. Transplanted hepatocytes proliferate differently after CCl4 treatment and hepatocyte growth factor infusion. Am J Physiol 1996;276:G629–638. Oren R, Dabeva MD, Karnezis AN, et al. Role of thyroid hormone in stimulating liver repopulation in the rat by transplanted hepatocytes. Hepatology 1999;30:903–913. Guo D, Fu T, Nelson JA, Superina RA, Soriano HE. Liver repopulation after cell transplantation in mice treated with retrorsine and carbon tetrachloride. Transplantation 2002;73:1818–1824. Gordon GJ, Coleman WB, Grisham JW. Temporal analysis of hepatocyte differentiation by small hepatocyte-like progenitor cells during liver regeneration in retrorsine-exposed rats. Am J Pathol 2000;157:771–786. Guha C, Sharma A, Gupta S, et al. Amelioration of radiationinduced liver damage in partially hepatectomized rats by hepatocyte transplantation. Cancer Res 1999;59:5871–5874. Gorla GR, Malhi H, Gupta S. Polyploidy associated with oxidative DNA injury attenuates proliferative potential of cells. J Cell Sci 2001;114:2943–2951. Malhi H, Gorla GR, Irani AN, Annamaneni P, Gupta S. Cell transplantation after oxidative hepatic preconditioning with radiation and ischemia–reperfusion leads to extensive liver repopulation. Proc Natl Acad Sci USA 2002;99:13114–13119.
Chapter 11 STEM CELLS AND HEPATOCYTE TRANSPLANTATION
122. Allain JE, Dagher I, Mahieu-Caputo D, et al. Immortalization of a primate bipotent epithelial liver stem cell. Proc Natl Acad Sci USA 2002;99:3639–3644. 123. Dandri M, Burda MR, Török E, et al. Repopulation of mouse liver with human hepatocytes and in vivo infection with hepatitis B virus. Hepatology 2001;33:981–988. 124. Mercer DF, Schiller DE, Elliott JF, et al. Hepatitis C virus replication in mice with chimeric human livers. Nature Med 2001;7:927–933. 125. Dandri M, Burda MR, Zuckerman DM, et al. Chronic infection with hepatitis B viruses and antiviral drug evaluation in uPA mice after liver repopulation with tupaia hepatocytes. J Hepatol 2005;42:54–60. 126. Ohashi K, Marion PL, Nakai H, et al. Sustained survival of human hepatocytes in mice: A model for in vivo infection with
human hepatitis B and hepatitis delta viruses. Nature Med 2000;6:327–331. 127. Benten D, Follenzi A, Bhargava KK, et al. Hepatic targeting of transplanted liver sinusoidal endothelial cells in intact mice. Hepatology 2005;42:140–148. 128. Uprichard SL, Boyd B, Althage A, Chisari FV. Clearance of hepatitis B virus from the liver of transgenic mice by short hairpin RNAs. Proc Natl Acad Sci USA 2005;102:773–778. 129. Wang Q, Contag CH, Ilves H, Johnston BH, Kaspar RL. Small hairpin RNAs efficiently inhibit hepatitis C IRES-mediated gene expression in human tissue culture cells and a mouse model. Mol Ther 2005;(Epub ahead of print).
191
Section II: Approach to the Patient with Liver Disease
LIVER BIOPSY TECHNIQUES Christian P. Strassburg and Michael P. Manns Abbreviations CT computed tomography
INR
12
international normalized ratios
INTRODUCTION The analysis of a liver specimen is an integral part of the clinical and laboratory work-up of any chronic liver disease, despite the fact that indications and the techniques employed have seen considerable changes since it was first performed.1 The first liver biopsy obtained by aspiration was performed by Ehrlich in 1883 to assess hepatic glycogen content in a diabetic patient, and 12 years later by Lucatello to analyze a tropical abscess of the liver. Its first application for the diagnosis of cirrhotic liver disease in humans and rats was published in a series by Schüpfer in France in 1907, and the diagnostic potential was expanded by Bingel in Germany in 1923. Over the subsequent 50 years the technique of obtaining liver biopsy samples has been modified regarding the approach, the needle type, and the combination with other diagnostic modalities, such as ultrasound, computed tomography (CT), angiography, and laparoscopy. Since the publication of a ‘1-second needle biopsy of the liver’ by Menghini in 19582 the technique of hepatic needle biopsy has seen a broad introduction into clinical non-operative medicine and is performed by experienced practitioners and hepatologists on a daily basis in hepatology centers.1 Although the etiology of most chronic liver diseases can be ascertained by modern biochemical, serological, immunological, and molecular biological tests, histological evaluation remains firmly integrated in the management of chronic hepatic disease.3 Not only cases of undefined liver diseases are subjected to histological analysis, but most importantly the determination of inflammatory activity (grading) and degree of fibrosis/cirrhosis (staging) is relevant for the prognosis of the patient and for the indication for cost-intensive as well as potentially side effectprone therapies, such as the administration of interferon-a in chronic hepatitis C virus infection. The increasing number of liver transplant patients within the hepatological spectrum requires regular, safe, and qualified biopsies and their assessment. In addition, the management of infectious diseases allows a fast and sensitive discovery of mycobacteria or viruses in hepatic tissues, i.e. in HIV-infected patients, and in patients with granulomatous diseases. The determination of copper and iron content in hepatic tissue can be achieved in biopsies from patients with hereditary storage diseases such as hemochromatosis and Wilson’s disease, but also in a1antitrypsin deficiency, amyloidosis, unclear space-occupying lesions, and suspected drug toxicity, important clues to the etiology and for the further management are reached.4–7 In view of these considerations liver biopsy techniques are of considerable importance. The technique of obtaining a liver biopsy is not only a question of how technically to perform such a biopsy, but also to consider and appre-
ciate the specific risk of side effects as well as the probability of obtaining information that will answer clinical questions and lead to a modification or initiation of a therapeutic approach.
DEFINITIONS OF A LIVER BIOPSY From a technical point of view liver tissue can be obtained either by cutting or by aspiration. Regarding the route of penetration, a liver biopsy can proceed transcutaneously via an intercostal or a subcostal route, as well as through a transvenous approach, usually via the jugular vein and then proceeding anatomically toward the hepatic veins. Ultrasound or CT guidance, as well as visual guidance during laparoscopy, is an option.
TRANSCUTANEOUS LIVER BIOPSY Transcutaneous liver biopsy is performed in a prone patient with the right arm elevated behind the head. The extent of the liver is examined by percussion. In our center every transcutaneous liver biopsy is preceded by an abdominal ultrasound, which confirms the absence of complicating factors including dilated bile ducts, venous collaterals, abnormal vascular findings such as hemangiomas, echinococcal cysts, Chilaiditi syndrome, very small cirrhotic livers, and the localization of the gallbladder, in addition to an accessible mass of liver tissue in the envisioned path of penetration. However, conclusive evidence that ultrasound reduces morbidity and mortality is controversial in the literature.8–11 Usually, an intercostal puncture site in the midaxillar line pointed at the xiphoid process is chosen. In very large livers biopsy can be performed subcostally. Complications appear to favor a subcostal route (4.1%) to a transthoracic route (2.7%),12 which is most often not possible in patients with small cirrhotic livers (Figure 12-1). After local infiltrating anaesthesia of the thoracic wall and the liver capsule a skin incision is made by scalpel. Depending on the type of needle used, one of the two following procedures is then employed. 1. Using a Menghini (aspiration) device, the needle, which is attached to a syringe optionally containing 10 ml sterile normal saline, is advanced up to the pleural surface and flushed with 1–2 ml of saline. The patient is asked to breath-hold in complete expiration, suction is applied by retracting the plunger, and the needle is quickly advanced and withdrawn from the liver tissue. The specimen is visualized and deposited into the appropriate specimen container (Figures 12-2 and 12-3).
195
Section II. Approach to the Patient with Liver Disease
Lung Sternum
Subcostal Intercostal
Ribs
Diaphragm Cirrhotic liver Enlarged liver
Figure 12-2. Typical set-up of materials required for an aspiration transcutaneous liver biopsy. The syringe is already connected to an aspiration needle (Menghini) biopsy set. The other syringe is used for local anesthesia.
Cut
Aspirate
Figure 12-1. The approach to transcutaneous liver biopsy. A transcutaneous biopsy of the liver is performed in complete expiration to limit accidental pulmonary injury. As the liver is frequently cirrhotic and therefore small, an intercostal approach is necessary. However, in large livers a subcostal route is preferred as a safer route. The schematic illustrates the two principal types of needle used: cutting needles and aspiration needles, for which several models are commercially available and which should be chosen according to the patient’s risk profile and the clinical indication for the biopsy (see text).
increased risk of hemorrhage and other risks of biopsy. Irrespective of whether the subcostal or the transthoracic approach is used, one study found an increase in complications when more than three passes were performed.12,13 However, the risk of hemorrhage in transcutaneous liver biopsy also depends on other factors, which include age and the presence of malignant tumors,13 as well as renal impairment.14,15 Two passes during aspiration needle biopsy have been shown to increase diagnostic quality and minor complications, compared to three passes.16
Experience 2. Using a Tru-cut needle, the tissue is obtained by cutting of a specimen lodged into a niche in the obturator needle by a second cylindrical needle sliding over it. To this end the needle is advanced into the liver, the sliding mechanism is triggered manually or automatically (‘biopsy gun’), and the needle is then withdrawn from the liver. The specimen is recovered from the obturator needle and placed in an appropriate specimen container. Overall, the Tru-cut needle remains in the liver for a longer time, increasing the possibility of patient movement and visceral injury, but it has been shown to produce superior tissue specimens. In a variation of the Tru-cut needle technique a plugged biopsy can be performed. In this case the needle is withdrawn from the cutting sheath, which remains in the liver with the patient still holding his or her breath. The needle is replaced by a plastic cannula, which is used to embolize the puncture canal with gelatin. This procedure can be considered as an alternative to transvenous liver biopsy in selected patients (discussed below).
TECHNIQUE AND RISK ASSESSMENT Number of Passes When a small or fragmented tissue cylinder is obtained the biopsy is repeated by a pass through the same incision. The number of passes to obtain a representative biopsy must be weighed against the
196
As with every invasive measure, complications are associated with the experience of the operator. The frequency of complications was 3.2% in physicians with a history of less than 20 biopsies, compared to 1.1% when over 100 had been performed,17 and showed that no differences were seen between gastroenterologists and general practitioners.
Tru-cut versus Aspiration Needle In general, the accepted mortality rate from liver biopsy is between 0.1 and 0.01%. However, as most studies rely on retrospective data these numbers have been found to vary considerably.13,17,18 Estimating this number is difficult for another reason. Patients subjected to liver biopsies usually suffer from advanced liver disease or malignancy, with a high mortality rate which is not dependent on a liver biopsy. The overall mortality in one study was shown to be 19% within 3 months after liver biopsy,17 but differences were also reported for the type of needle used. The overall complication rate in one series was 0.35% for Tru-cut and 0.1% for Menghini needles.17 This study identified a higher incidence of hemorrhage, pneumothorax, biliary leakage, and peritonitis with Tru-cut needles, but puncture of other internal organs and sepsis occurred more often with Menghini needles. In contrast, a study comparing Jamshidi suction needles with Tru-cut Vim Silverman needles was unable to demonstrate such technique-related differences.12,13 Whether needle diameter is a factor predisposing to hemorrhage is contro-
Chapter 12 LIVER BIOPSY TECHNIQUES
B
A
D
C
E
F
G
Figure 12-3. Procedure of transcutaneous aspiration liver biopsy using a Menghini needle. Ultrasound is used to visualize the path of the biopsy needle (A), which is routinely performed in our center to minimize complications, and is considered controversial as conclusive data suggesting a reduction of complications are not available. While the patient is holding her/his breath the skin, the thoracic wall down to the liver capsule is locally anesthesized (B) prior to performing a small scalpel incision (C). The needle is positioned (D) in the incision canal, advanced to the liver capsule while the patient is holding their breath, suction is applied with the connected syringe, and the needle rapidly advanced into the liver tissue. It is then withdrawn with the plunger still arrested in the suction position (E). The tissue cylinder is deposited in an appropriate specimen container (F).
197
Section II. Approach to the Patient with Liver Disease
versial. Whereas in humans no differences were observed when a 1.6 mm or a 1.9 mm Menghini needle was used,9 experiments on anesthetized pigs showed more bleeding with the 2.1 mm than with the 1.6 mm needle, as well as 1.2 versus 1.6 mm needles.15 Considerations regarding gauge as well as between Menghini and Tru-cut needles should bear in mind that smaller needles and smaller specimens may increase the number of passes,16 or necessitate repuncture with an increase of bleeding risk in order to obtain representative tissue.
Bacteremia In normal and cirrhotic livers biopsies have been shown to lead to bacteremia.20–22 In patients with biliary enteric anastomoses the occurrence of septic complications following liver biopsy has been controversial.23–26 At present there are insufficient data to recommend routine antibiotic prophylaxis. Antibiotics should, however, be administered to patients suffering from valvular heart disease, a documented history of septic complications following liver biopsy, or in suspected cholangitis.
TRANSVENOUS LIVER BIOPSY PROCEDURE Because hemorrhage is the single most feared complication of transcutaneous liver biopsy, and because patients suffering from severe coagulation disorders due to hepatic and other diseases are frequent candidates for histological assessment, transvenous liver biopsy was developed, described by Dotter in 1964.27 The biopsy is obtained via a usually right-sided puncture of the internal jugular vein, and rarely through a transfemoral route. Technically it is a modified version of the aspiration needle biopsy procedure. However, transvenous liver biopsy requires a technically more complex setting, which includes fluoroscopy, the transvenous catheter and needle kit, cardiac monitoring, and the application of contrast medium. The internal jugular vein is cannulated and a sheath inserted following the technique of Seldinger. A catheter is then guided through the right atrium into the inferior vena cava. After loading with the transvenous biopsy needle it is advanced into one of the hepatic veins, which is visualized by the injection of contrast medium. The patient is asked to stop breathing and the needle is rapidly advanced 1–2 cm beyond the tip of the venous catheter, and suction is applied by a connected syringe during liver passage. Liver tissue can subsequently be recovered from the needle.
TECHNIQUE AND RISK ASSESSMENT Complications and Alternatives Transvenous liver biopsy is a second-line procedure limited to patients with significant coagulation disorders in whom liver histology is likely to alter therapeutic management. In patients with coagulation disorders liver biopsy coupled with embolization of the puncture canal (plugged biopsy) is an alternative and has been shown to be a safe procedure.28–30 Direct comparison of transvenous and embolization liver biopsy showed that both methods were equally successful, with low complication rates. Transcutaneous embolization biopsy led to larger liver specimens, as would be expected with
198
a cutting biopsy technique, but was also associated with a 3.5% rate of bleeding in one study.31 It would therefore appear prudent to employ embolization liver biopsy in patients with coagulation abnormalities in the absence of tension ascites whenever a transjugular approach is impossible or has previously failed.32
Technical Considerations The main disadvantage of transvenous liver biopsy is the considerable effort required. Whereas a standard transcutaneous liver biopsy can be performed in 15 minutes, including ultrasound visualization, a transvenous procedure requires a complex set-up and takes 45 minutes. There is a theoretical risk of arrhythmia and of contrast material-related reactions, in addition to the use of X-rays. The cost of the catheter/needle kit and the longer physician presence is high. This profile clearly limits its application to selected patients.
LAPAROSCOPIC LIVER BIOPSY Laparoscopic liver biopsy is not a biopsy technique per se but an alternative way to obtain a liver biopsy during visual assessment of the peritoneum and the abdominal organs.33 Laparoscopic liver biopsy has attracted increasing attention since minimally invasive mini laparoscopy was established for routine clinical practice.34 Mini laparoscopy is distinguished from standard or midi laparoscopy by the use of optical instruments with a diameter <2 mm and the use of a sheathed Veress needle that allows for the application of a pneumoperitoneum and the abdominal introduction of the optical instrument through a single puncture. The abdomen is punctured at the so-called Kalk position, 2 cm left and cranially of the umbilicus, and usually 3–4 l of N2O are introduced to produce a sufficient pneumoperitoneum. An instrumentation sheath is subsequently positioned under direct vision, typically in the lateral upper right quadrant within reach of the liver. After a macroscopic assessment of the cranial and inferior liver surface, including the edges, the gallbladder, the hepatic ligament, and the peritoneum, a liver biopsy can be performed either by the aspiration or by the Tru-cut technique by advancing the respective biopsy device through the instrumentation sheath (Figures 12-4 and 12-5). This is preceded by macroscopically identifying a specific area of interest on the liver surface, thereby increasing the likelihood of a representative biopsy specimen relative to the original indication for the biopsy. The site of puncture can be monitored, and in case of prolonged hemorrhage monopolar electric coagulation can be employed. The presence of ascites is not a limiting factor, as ascitic fluid can be evacuated during laparoscopy prior to the liver biopsy procedure.
TECHNIQUE AND RISK ASSESSMENT Mini laparoscopy is less invasive than standard laparoscopy but has the disadvantage of a 0° optical instrument of 1.9–2.0 mm diameter with decreased brightness and therefore a reduced overview of the abdominal organs. A number of studies have suggested laparoscopy to be the gold standard for the diagnosis of liver cirrhosis.33–35 The macroscopic diagnosis of nodular cirrhosis and increased tissue consistency by direct palpation is relatively simple
Chapter 12 LIVER BIOPSY TECHNIQUES
Figure 12-4. Typical set-up of materials required for a mini laparoscopy and biopsy. Apart from the Veress needle and optical instruments an instrumentation sheath is required which is used to introduce the biopsy needle into the abdomen under direct vision and to obtain the liver biopsy sample. This photograph shows an automatic biopsy gun for a Tru-cut biopsy needle.
and leads to an increase of sensitivity compared to ‘blind’ biopsy or ultrasound visualization and biopsy. A recent study using Menghini aspiration needles has found the sensitivity to detect cirrhosis to be 96.4% for standard and 91.9% for mini laparoscopy, compared to 68% for histology without laparoscopy.34 Compared to transcutaneous liver biopsy the rate of complications appears not be increased for mini laparoscopy.36 An important issue is the availability of this approach to patients with coagulation disorders. Mini laparoscopy has been shown to be feasible in patients with prolonged international normalized ratios (INR) >1.5, thrombocytopenia <50/nl, and even in cases of von Willebrand’s disease or hemophilia without significant hemorrhage.37 The risk assessment includes the risks of the laparoscopy itself, i.e. sedation, accidental vascular injection of N2O, injury to the viscera, and hemorrhage of abdominal wall vessels. During 63 845 standard laparoscopies with 48 766 liver biopsies38, 39 the reported rate of hemorrhage was 0.09%, overall complications 2.5%, and the mortality rate 0.03%.
ULTRASOUND-GUIDED FINENEEDLE ASPIRATION Ultrasound-guided fine-needle aspiration is extensively used to obtain histological or cytological information regarding focal hepatic lesions. It is not routinely employed to determine the grade of liver inflammation or the stage of hepatic fibrosis.40 The indications for fine-needle aspiration cytology of diffuse hepatic diseases are controversial. The hepatic lesion is visualized by ultrasound and a path for the needle aspiration is plotted to avoid intersecting vessels. Usually, an ultrasound array with an integrated needle guidance slot is employed. A needle with a diameter of <1 mm is advanced into the lesion while the patient holds their breath, suction is applied with a syringe connected to the needle, and after three to five passes the needle is withdrawn. The aspirated material is spread on a glass microscopy coverslip, dried, and forwarded to the cytologist. Ultra-
199
Section II. Approach to the Patient with Liver Disease
A
B
C
Figure 12-5. Liver biopsy by mini laparoscopy. (A) Application of the pneumoperitoneum after puncture with a sheathed 2 mm Veres needle. The typical insertion point left and cranial of the umbilicus is visible (B). After generation of the pneumoperitoneum the 2 mm optical instrument is inserted. Image shows the situation prior to the insertion of the instrument. Images (C) show the liver surface and a visually guided cutting needle positioned on the superior surface of the right lobe. Direct visualization allows for the biopsy to proceed within an area of interest, thereby increasing the likelihood of obtaining a representative tissue sample. In the case of bleeding complications the same approach can be used to apply monopolar electric coagulation for hemostasis.
200
Chapter 12 LIVER BIOPSY TECHNIQUES
sound-guided biopsy can also be performed with cutting needles. In general, a patient with cirrhosis and a higher likelihood of bleeding would be a candidate for fine-needle aspiration rather than cutting needle biopsy. The biopsy may also be performed under CT guidance.
TECHNIQUE AND RISK ASSESSMENT Hemorrhage Hemorrhage with fine-needle aspiration using <1 mm needles is rare. The mortality rate has been estimated to range between 0.006 and 0.1%.41 An analysis of 2091 biopsies using non-cutting needles reported the lowest complication rate.42
general risk assessment of the patient and the observation of contraindications.
INFORMED CONSENT Informed consent should always be obtained in writing on the day before the planned biopsy. Care should be taken to provide information in the candidate’s native language or through an interpreter.
COMPLIANCE Most techniques (except for laparoscopy) require a patient who is cooperative and able to hold their breath for the required length of time. In a frightened patient medication with midazolam can be considered, but should be weighed against the ability to cooperate.
Specificity A number of studies have documented that the specificity of fineneedle aspiration cytology is accurate, with sensitivities and specificities near 100%.43–45 Regarding biopsy size, it was suggested that both fine-needle and cutting needle aspiration both resulted in a diagnostic accuracy of 78%, but when the two were combined accuracy rose to 88%.46 According to these studies, fine-needle aspiration cytology is a safe and sensitive diagnostic procedure.
Dissemination of Malignant Cells One of the most controversial issues surrounding the biopsy of space-occupying lesions is the potential seeding of cancer cells following fine-needle aspiration. This is a problem, because the diagnostic role of this technique is most often the assessment of suspect hepatic lesions. Seeding rates of biopsies obtained from all abdominal organs have been estimated to rage between 0.0003% and 0.009%.47,48 However, in a retrospective study by Takamori49 the rate in hepatocellular carcinomas was 5.1%, which questions the value of needle aspiration in hepatocellular carcinoma patients. When polymerase chain reaction for the detection of a1-antitrypsin mRNA was employed, tumor cell dissemination after fine-needle aspiration was suggested.50 The evaluation of hepatic masses without fine-needle biopsy in one study showed that standard imaging procedures give a diagnostic accuracy of 99.6% and a sensitivity of 98.6% for hepatocellular carcinoma, and similarly for liver metastases.51 However, the issue of tumor cell seeding and its clinical relevance remains controversial,52–54 and differs according to the lesion to be biopsied.55,56 Based on these considerations, the biopsy of hepatocellular carcinomas or other suspected malignant lesions of the liver should be weighed cautiously against the risk of dissemination, the therapeutic consequences of the biopsy, and the availability of other imaging tests with sufficient information. Fineneedle aspiration should be considered in patients with unclear hepatic lesions in whom the side effects of therapy outweigh the consequences of seeding. In HCC a lesion for cytology typically measures <2 cm and lacks the typical features of HCC in other imaging tests, including negative serum AFP.
THE CANDIDATE FOR LIVER BIOPSY The technical aspects of a successful and therapeutically or diagnostically relevant liver biopsy can only be viewed in light of a
MECHANICAL BILIARY OBSTRUCTION This is a relative contraindication. Biliary peritonitis, sepsis, and pain are possible consequences. If the benefit of the biopsy outweighs these complications a biopsy can be performed; a transvenous approach should be considered.
CHOLANGITIS Bacterial cholangitis is a relative contraindication and may actually provide useful bacteriological information in some patients. In view of the risk of disseminating pathogens during the biopsy antibiotic treatment should be considered in these cases, although data to support a general recommentation are lacking.
HEMOSTASIS A normal INR is desired for liver biopsies, but is frequently not present in patients requiring a liver biopsy. The fact that 90% of hemorrhages have been found in patients with an INR <1.3 reflects the fact that although abnormal INR values predispose to bleeding, a normal INR does not represent an absolute reassurance that hemorrhage will not occur (Fig. 12-6).17 Coagulation studies should be performed 24 hours prior to the biopsy and should result in an INR <1.4. If this is not the case other strategies, such as transvenous biopsy or mini laparoscopy, should be considered, including a potential coagulation factor substitution to improve the coagulopathy.29,37,57 Platelet count is also a matter of controversy. In some patients, such as hemodialysis patients or individuals with renal impairment, as well as those with hematological disorders or cholestasis, platelet function may be compromised despite normal numbers.58 Evidence suggests that a liver biopsy can be safely performed when platelets are above 60 000/mm3.13 The use of aspirin and other non-steroidal anti-inflammatory drugs 1 week prior to biopsy is controversial, but data confirming an increased risk in these patients are currently lacking.
ASCITES There are no randomized controlled studies to suggest that ascites is a strict contraindication for liver biopsy, although in many publications a potentially higher rate of postinterventional bleeding is assumed. Experience with CT-guided liver biopsy has not demonstrated an increase in complications in patients with ascites.29,59,60
201
Section II. Approach to the Patient with Liver Disease
9.
10. 11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
Figure 12-6. CT image of an intrahepatic hemorrhage after aspiration liver biopsy in a patient with HCV infection and normal coagulation as well as blood count parameters. (A) Subcapsular and intrahepatic hematoma 24 hours after puncture. (B) Longitndinal rendering demonstrates the extent of the hematoma into the right lobe of the liver.
REFERENCES 1. Sheela H, Seela S, Caldwell C, et al. Liver biopsy: evolving role in the new millennium. J Clin Gastroenterol 2005;39:603– 610. 2. Menghini G. One-second needle biopsy of the liver. Gastroenterology 1958;35:190–199. 3. Strassburg CP, Obermayer-Straub P, Manns MP. Autoimmunity in liver diseases. Clin Rev Allergy Immunol 2000;18:127–139. 4. Guido M, Rugge M. Liver biopsy sampling in chronic viral hepatitis. Semin Liver Dis 2004;24:89–97. 5. Li MK, Crawford JM. The pathology of cholestasis. Semin Liver Dis 2004;24:21–42. 6. Brunt EM. Nonalcoholic steatohepatitis. Semin Liver Dis 2004;24:3–20. 7. Jevon GP, Dimmick JE. Histopathologic approach to metabolic liver disease: Part 1. Pediatr Dev Pathol 1998;1:179–199. 8. Caturelli E, Biasini E, Bartolucci F, et al. Diagnosis of hepatocellular carcinoma complicating liver cirrhosis: utility of
202
21.
22. 23.
24.
25.
26.
27. 28.
29.
30.
repeat ultrasound-guided biopsy after unsuccessful first sampling. Cardiovasc Intervent Radiol 2002;25:295–299. Caturelli E, Giacobbe A, Facciorusso D, et al. Percutaneous biopsy in diffuse liver disease: increasing diagnostic yield and decreasing complication rate by routine ultrasound assessment of puncture site. Am J Gastroenterol 1996;91:1318–1321. Stotland BR, Lichtenstein GR. Liver biopsy complications and routine ultrasound. Am J Gastroenterol 1996;91:1295–1296. Vautier G, Scott B, Jenkins D. Liver biopsy: blind or guided? Br Med J 1994;309:1455–1456. Perrault J, McGill DB, Ott BJ, et al. Liver biopsy: complications in 1000 inpatients and outpatients. Gastroenterology 1978;74:103–106. McGill DB, Rakela J, Zinsmeister AR, et al. A 21-year experience with major hemorrhage after percutaneous liver biopsy. Gastroenterology 1990;99:1396–1400. Ahmad A, Hasan F, Abdeen S, et al. Transjugular liver biopsy in patients with end-stage renal disease. J Vasc Interv Radiol 2004;15:257–260. Gazelle GS, Haaga JR, Rowland DY. Effect of needle gauge, level of anticoagulation, and target organ on bleeding associated with aspiration biopsy. Work in progress. Radiology 1992;183:509–513. Maharaj B, Bhoora IG. Complications associated with percutaneous needle biopsy of the liver when one, two or three specimens are taken. Postgrad Med J 1992;68:964–967. Gilmore IT, Burroughs A, Murray-Lyon IM, et al. Indications, methods, and outcomes of percutaneous liver biopsy in England and Wales: an audit by the British Society of Gastroenterology and the Royal College of Physicians of London. Gut 1995;36:437–441. Piccinino F, Sagnelli E, Pasquale G, et al. Complications following percutaneous liver biopsy. A multicentre retrospective study on 68,276 biopsies. J Hepatol 1986;2:165–173. Forssell PL, Bonkowsky HL, Anderson PB, et al. Intrahepatic hematoma after aspiration liver biopsy. A prospective randomized trial using two different needles. Dig Dis Sci 1981;26:631–635. Larson AM, Chan GC, Wartelle CF, et al. Infection complicating percutaneous liver biopsy in liver transplant recipients. Hepatology 1997;26:1406–1409. Le Frock JL, Ellis CA, Turchik JB, et al. Transient bacteremia associated with percutaneous liver biopsy. J Infect Dis 1975;131(Suppl):S104–107. McCloskey RV, Gold M, Weser E. Bacteremia after liver biopsy. Arch Intern Med 1973;132:213–215. Ben-Ari Z, Neville L, Rolles K, et al. Liver biopsy in liver transplantation: no additional risk of infections in patients with choledochojejunostomy. J Hepatol 1996;24:324–327. Bubak ME, Porayko MK, Krom RA, et al. Complications of liver biopsy in liver transplant patients: increased sepsis associated with choledochojejunostomy. Hepatology 1991;14:1063–1065. de Diego Lorenzo A, Romero M, Duran F, et al. Bacteremia following liver biopsy in transplant recipients with roux-en-Y choledochojejunostomy. Rev Esp Enferm Dig 1997;89:289–295. Galati JS, Monsour HP, Donovan JP, et al. The nature of complications following liver biopsy in transplant patients with Roux-en-Y choledochojejunostomy. Hepatology 1994;20:651–653. Dotter CT. Catheter biopsy. Experimental technic for transvenous liver biopsy. Radiology 1964;82:312–314. Albeniz Arbizu E, Lopez San Roman A, Garcia Gonzalez M, et al. Fibrin-glue sealed liver biopsy in patients with a liver transplantation or in liver transplantation waiting list: preliminary results. Transplant Proc 2003;35:1911–1912. Kamphuisen PW, Wiersma TG, Mulder CJ, et al. Pluggedpercutaneous liver biopsy in patients with impaired coagulation and ascites. Pathophysiol Haemost Thromb 2002;32:190–193. Tobin MV, Gilmore IT. Liver biopsy with plugged needle track. Lancet 1984;2:694.
Chapter 12 LIVER BIOPSY TECHNIQUES
31. Sawyerr AM, McCormick PA, Tennyson GS, et al. A comparison of transjugular and plugged-percutaneous liver biopsy in patients with impaired coagulation. J Hepatol 1993;17:81–85. 32. Jackson JE, Adam A, Allison DJ. Transjugular and plugged liver biopsies. Baillières Clin Gastroenterol 1992;6:245–258. 33. Jalan R, Harrison DJ, Dillon JF, et al. Laparoscopy and histology in the diagnosis of chronic liver disease. QJ Med 1995;88:559–564. 34. Weickert U, Siegel E, Schilling D, et al. [The diagnosis of liver cirrhosis: a comparative evaluation of standard laparoscopy, minilaparoscopy and histology]. Z Gastroenterol 2005;43:17–21. 35. Nord HJ. Biopsy diagnosis of cirrhosis: blind percutaneous versus guided direct vision techniques – a review. Gastrointest Endosc 1982;28:102–104. 36. Helmreich-Becker I, Meyer zum Buschenfelde KH, Lohse AW. Safety and feasibility of a new minimally invasive diagnostic laparoscopy technique. Endoscopy 1998;30:756–762. 37. Denzer U, Helmreich-Becker I, Galle PR, et al. Liver assessment and biopsy in patients with marked coagulopathy: value of minilaparoscopy and control of bleeding. Am J Gastroenterol 2003;98:893–900. 38. Bruhl W. [Incidents and complications in laparoscopy and directed liver puncture. Result of a survey]. Dtsch Med Wschr 1966;91:2297–2299. 39. Nord HJ. Complications of laparoscopy. Endoscopy 1992;24:693–700. 40. Wee A. Fine-needle aspiration biopsy of the liver: algorithmic approach and current issues in the diagnosis of hepatocellular carcinoma. Cytojournal 2005;2:7. 41. Pitman MB. Fine-needle aspiration biopsy of the liver. Principal diagnostic challenges. Clin Lab Med 1998;18:483–506, vi. 42. Buscarini L, Fornari F, Bolondi L, et al. Ultrasound-guided fineneedle biopsy of focal liver lesions: techniques, diagnostic accuracy and complications. A retrospective study on 2091 biopsies. J Hepatol 1990;11:344–348. 43. Fornari F, Civardi G, Cavanna L, et al. Ultrasonically guided fineneedle aspiration biopsy: a highly diagnostic procedure for hepatic tumors. Am J Gastroenterol 1990;85:1009–1013. 44. Hertz G, Reddy VB, Green L, et al. Fine-needle aspiration biopsy of the liver: a multicenter study of 602 radiologically guided FNA. Diagn Cytopathol 2000;23:326–328. 45. Jain D. Diagnosis of hepatocellular carcinoma: fine-needle aspiration cytology or needle core biopsy. J Clin Gastroenterol 2002;35(Suppl 2):S101–108. 46. Franca AV, Valerio HM, Trevisan M, et al. Fine-needle aspiration biopsy for improving the diagnostic accuracy of cut needle biopsy of focal liver lesions. Acta Cytol 2003;47:332–336.
47. Cedrone A, Rapaccini GL, Pompili M, et al. Neoplastic seeding complicating percutaneous ethanol injection for treatment of hepatocellular carcinoma. Radiology 1992;183:787–788. 48. Smith EH. Complications of percutaneous abdominal fine-needle biopsy. Review. Radiology 1991;178:253–258. 49. Takamori R, Wong LL, Dang C, et al. Needle-tract implantation from hepatocellular cancer: is needle biopsy of the liver always necessary? Liver Transplant 2000;6:67–72. 50. Louha M, Nicolet J, Zylberberg H, et al. Liver resection and needle liver biopsy cause hematogenous dissemination of liver cells. Hepatology 1999;29:879–882. 51. Torzilli G, Minagawa M, Takayama T, et al. Accurate preoperative evaluation of liver mass lesions without fine-needle biopsy. Hepatology 1999;30:889–893. 52. Caturelli E, Ghittoni G, Roselli P, et al. Fine needle biopsy of focal liver lesions: the hepatologist’s point of view. Liver Transplant 2004;10(Suppl 1):S26–29. 53. Ng KK, Poon RT, Lo CM, et al. Impact of preoperative fineneedle aspiration cytologic examination on clinical outcome in patients with hepatocellular carcinoma in a tertiary referral center. Arch Surg 2004;139:193–200. 54. Torzilli G, Olivari N, Del Fabbro D, et al. Indication and contraindication for hepatic resection for liver tumors without fine-needle biopsy: validation and extension of an Eastern approach in a Western community hospital. Liver Transplant 2004;10(Suppl 1):S30–33. 55. Shah JN, Fraker D, Guerry D, et al. Melanoma seeding of an EUS-guided fine needle track. Gastrointest Endosc 2004;59:923–924. 56. de Sio I, Castellano L, Calandra M, et al. Subcutaneous needletract seeding after fine-needle aspiration biopsy of pancreatic liver metastasis. Eur J Ultrasound 2002;15:65–68. 57. Shin JL, Teitel J, Swain MG, et al. A Canadian multicenter retrospective study evaluating transjugular liver biopsy in patients with congenital bleeding disorders and hepatitis C: is it safe and useful? Am J Hematol 2005;78:85–93. 58. Pihusch R, Rank A, Gohring P, et al. Platelet function rather than plasmatic coagulation explains hypercoagulable state in cholestatic liver disease. J Hepatol 2002;37:548–555. 59. Little AF, Ferris JV, Dodd GD 3rd, et al. Image-guided percutaneous hepatic biopsy: effect of ascites on the complication rate. Radiology 1996;199:79–83. 60. Murphy FB, Barefield KP, Steinberg HV, et al. CT- or sonographyguided biopsy of the liver in the presence of ascites: frequency of complications. AJR Am J Roentgenol 1988;151:485–486.
203
Section II: Approach to the Patient with Liver Disease
13
LIVER BIOPSY HISTOPATHOLOGY Jay H. Lefkowitch
Abbreviations AAT a1-antitrypsin AFLD fatty liver disease ASH alcoholic steatohepatitis H&E hematoxylin and eosin
NAFLD NASH PAS
non-alcoholic fatty liver disease non-alcoholic steatohepatitis periodic acid–Schiff
GENERAL PRINCIPLES ROLE OF LIVER BIOPSY Liver biopsy has been an important component in the evaluation of patients with hepatobiliary disease since Menghini first popularized the percutaneous liver biopsy technique in the late 1950s.1 Other techniques for obtaining liver biopsy specimens have emerged since then and are discussed in Chapter 12. Although most of the descriptions in this chapter pertain to needle biopsy specimens, they are also usually applicable to larger specimens taken from explanted or postmortem livers, a few caveats aside. Examination of liver biopsy specimens under the microscope confers the unique diagnostic advantage of directly visualizing the morphological changes present in the liver. The pathologist interpreting these changes may then address a number of important clinicopathological issues (Table 13-1). Among these are several that in recent years have grown to constitute a considerable portion of the pathologist’s workload, including grading and staging chronic hepatitis, documentation of steatosis and its complications (particularly important because of the widespread prevalence of obesity and diabetes as causes of abnormal liver function tests2,3), and liver transplantation pathology. Frozen and formalin-fixed biopsy specimens also serve as important archival resources for immunohistochemical, molecular and genetic investigations, including studies that utilize gene chip technology.4
SPECIAL STAINS AND IMMUNOHISTOCHEMISTRY The standard hematoxylin and eosin (H&E)-stained section is the mainstay of routine liver biopsy interpretation. However, the application of a routine panel of special stains (Table 13-2) provides helpful, and in some instances critical, diagnostic information (Figure 13-1). For example, the staging of chronic hepatitis is reliant on determining the presence or absence, and degree, of fibrous scar deposition in the liver. An H&E stain alone is often insufficient for
PBC PiMM PSC
Primary biliary cirrhosis protease inhibitor MM primary sclerosing cholangitis
this evaluation, or raises architectural questions that must be more fully determined by reviewing connective tissue stains, such as the trichrome and reticulin methods. Hepatic iron and copper overload conditions are other examples where special stains are mandated. There are also liver diseases which may only first come to light because a special stain was applied, such as the patient with chronic hepatitis of unknown etiology whose biopsy demonstrates a1-antitrypsin globules when a diastase–PAS (periodic acid–Schiff) stain is performed. Immunohistochemical stains also have a substantial impact in the evaluation of liver tumors and a variety of nonneoplastic conditions (Table 13-2). The indications for special stains or immunohistochemistry are discussed in the later sections on specific liver diseases.
PRACTICAL GUIDELINES TO INTERPRETATION The general microscopic approach to liver biopsy interpretation begins with an assessment of the lobular architecture (Figure 13-2). Although other microanatomic units such as the acinus5 have been described,6 Kiernan’s lobule, a hexagonal unit of liver parenchyma with a central vein as its hub and peripherally located portal tracts (Figure 13-3), remains the most popular construct in use. Evaluation of the integrity of lobular architecture is contingent on obtaining an adequate specimen, and this must also be taken into account by the pathologist. Most pathologists favor a 1.5 cm needle biopsy core as being an adequate length for being representative, and a 2.5 cm long core as particularly optimal.7 The data from small biopsy samples must therefore be viewed cautiously, particularly in the diagnosis of mass lesions and conditions with known regional tissue variations (e.g. chronic hepatitis). In certain instances consensus panels have codified the criteria for an optimal specimen, as in the post-liver transplantation biopsy, which should contain a minimum of five portal tracts for review in order to grade acute rejection.8 The initial observations under the microscope can therefore be framed based on the status of the lobular architecture: is it preserved, disturbed, or distorted? Disturbed lobular architecture
205
Section II. Approach to the Patient with Liver Disease
comprises a range of findings, such as the parenchymal disarray in acute hepatitis produced by a combination of hepatocellular damage and infiltrating inflammatory cells, or the expansion of periportal regions by edema, neutrophils and ductular reaction in biliary tract obstruction. Distorted lobular architecture is usually used to describe chronic liver diseases characterized by fibrosis, nodularity or cirrhosis, in which the relationships of vascular structures, central veins and portal tracts are altered. Once an initial impression of the state of the lobular architecture has been obtained, routine liver biopsy examination usually benefits from a regular review of specific structural points and pertinent negatives, as outlined below.
ZONAL CHANGES The predilection of certain liver diseases for lobular (or acinar) zones can assist in establishing the appropriate diagnosis. On low magnification, the centrilobular (perivenular), midzonal and periportal regions (acinar zones 3, 2 and 1, respectively) should be examined for alterations. Centrilobular hepatocytes receive the endflow of perfusion from hepatic arteriolar and portal vein sources within the portal tracts, and are therefore most susceptible to ischemia and hypotension. These hepatocytes also contain the major hepatic drug metabolizing enzymes, so that the effects of acetaminophen (Figure 13-4) and carbon tetrachloride (predictable hepatotoxins) and alcohol are first evident in perivenular regions. As the prime area of venous outflow from the liver to the hepatic veins, the centrilobular region is also prone to congestion and sinusoidal dilatation in conditions of hepatic venous outflow obstruction, including Budd–Chiari syndrome and right heart failure of any cause. Pathologic changes in periportal regions, in contrast, are usually related to a different set of hepatic disorders, such as iron overload within periportal hepatocytes in hereditary hemochromatosis, acute and chronic biliary tract diseases (ductular reaction, pseudoxanthomatous change, copper accumulation), chronic hepatitis and hepatocellular regenerative hyperplasia. The structural and functional heterogeneity of the liver lobule is therefore extremely useful to bear in mind in liver biopsy interpretation.
Table 13-1. Issues Addressed by Liver Biopsy Cause of abnormal liver function tests Cause of hepatomegaly Diagnosis of mass lesions Cause of fever of unknown origin Grading and staging of chronic hepatitis Cause of portal hypertension Liver dysfunction after liver transplantation Fatty liver and its complications Jaundice of uncertain origin Documentation of cirrhosis and its type
A
B
C
D
Figure 13-1. Use of special stains in liver histopathology. Steatohepatitis with cirrhosis. A Hematoxylin and eosin stain shows marked fat and fibrosis (pink) subdividing the parenchyma into cirrhotic nodules. B Periodic acid–Schiff (PAS) contrasts the glycogen-containing (purple) nodules of fatty hepatocytes with the surrounding pale fibrous septa. C Reticulin stain emphasizes the distorted architecture, with extensive central-to-central and central-to-portal bridging fibrosis surrounding nodular parenchyma. D Masson trichrome stain shows diffuse fibrosis in blue.
206
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Table 13-2. Special Stains and Immunohistochemistry in Liver Histopathology Stain
Diagnostic information
Hematoxylin and eosin Trichrome Reticulin
Routine interpretation
Iron PAS* DPAS**
Orcein/Victoria blue
Fibrosis, cirrhosis Fibrosis, cirrhosis Regenerative hyperplasia (thickened liver cell plates) Parenchymal collapse (e.g. confluent necrosis) Hepatocellular carcinoma (paucireticulin pattern) Hemosiderosis Hereditary hemochromatosis Glycogen content (e.g. glycogen storage disease; depletion from damaged centrilobular regions) a1-antitrypsin globules Phagocytic debris in macrophages after necrosis Basement membrane (e.g. around bile ducts) Hepatitis B surface antigen Copper-binding protein Elastic tissue fibers (e.g. indigenous or in scars)
Immunohistochemical stains Anti-HBs HBV surface antigen in hepatocytes Anti-HBc HBV core antigen in hepatocytes (i.e. viral replication) Hep par 1 Positive in hepatocellular carcinoma and non-tumor (‘hepatocyte’) liver pCEA*** Apical/canalicular staining in hepatocellular carcinoma Cytokeratin 7 Negative staining in hepatocellular carcinoma and 20 pair Cytokeratin 7 Bile duct epithelium or intermediate hepatobiliary or 19 cells Ductular reaction Cytokeratins 8 Mallory bodies and 18 Ubiquitin Mallory bodies Smooth muscle Activated stellate cells (i.e. hepatic myofibroblasts) actin (SMA) CD 31 and 34 Endothelium (e.g. vascular tumors)
P
P
C
P
P
P Figure 13-3. Normal pig lobule. The lobule was first described in Kiernan’s pig dissections. Note the overall structural similarities to human liver, with central vein (C) and peripheral portal tracts (P). However, fibrous septa (blue) divide the lobular units into their hexagonal pattern. (Masson trichrome.)
* Periodic acid–Schiff. ** Diastase–PAS. *** Polyclonal carcinoembryonic antigen.
Figure 13-4. Zonal necrosis due to acetaminophen. Centrilobular (zone 3) necrosis due to acetaminophen is seen in the middle of this field. The zonal nature of this injury is related to high microsomal enzyme content in this region. (Hematoxylin and eosin.) P
P
BILE DUCT INTEGRITY C
P P
P
Figure 13-2. Normal lobule. The central vein (C) is the hub of this microanatomic unit, with liver cell cords (plates) radiating outward toward peripherally located portal tracts (P). (Hematoxylin and eosin.)
The portal tracts should routinely be assessed for the integrity of bile ducts, as a number of important conditions result in their damage or disappearance. In principle, portal tracts should show one bile duct for every hepatic artery, although the number of bile duct profiles within portal tracts varies even within normal specimens.9 The absence of bile ducts from 50% or more of portal tracts represents ductopenia (sometimes referred to as the vanishing bile duct syndrome10). Primary biliary cirrhosis, primary sclerosing cholangitis, chronic liver allograft rejection and drug hepatotoxicity are the major conditions in which bile duct damage and/or ductopenia may be found in adults. In pediatric liver biopsies, syndromatic paucity of intrahepatic bile ducts (Alagille syndrome11) and non-syndromatic paucity (sometimes associated with metabolic disease, such as a1antitrypsin deficiency, bile salt transport disorders such as Byler
207
Section II. Approach to the Patient with Liver Disease
Figure 13-5. Lipofuscin pigment. Yellow lipofuscin granules (arrows) are present in centrilobular hepatocytes. (Hematoxylin and eosin.)
Figure 13-6. Kupffer cell hemosiderosis. Clumped, granular hemosiderin is seen throughout the sinusoids in Kupffer cells, a feature seen after transfusions, hemolysis and hemodialysis. (Prussian blue iron stain.)
Table 13-3. Hepatic Pigments Type
Quality
Location
Lipofuscin
Yellow-brown, fine granules Brown, glassy, refractile
Centrilobular hepatocytes (pericanalicular/membranous) Kupffer cells Hepatocytes (periportal-tocentrilobular diminishing gradient) Bile ducts (hemochromatosis) Centrilobular hepatocyte cytoplasm and/or bile canaliculi Kupffer cells (centrilobular predominance) Panlobular with perivenular accentuation
Hemosiderin
Bile
Brown, dull
Ceroid
Tan
Dubin–Johnson Coarse brown
disease,12 or perinatal infection by cytomegalovirus) are additional diagnostic considerations.
Figure 13-7. Acute hepatitis. There is diffuse inflammation within the lobular parenchyma and the portal tract (P). A central vein is at upper left. Note the scattered apoptotic bodies (arrows). (Hematoxylin and eosin.)
PIGMENTS A variety of pigments accumulate in the liver, many of which appear brown on routine H&E staining. The location and quality of the pigment is helpful in the differential diagnosis (Table 13-3). For example, refractile brown pigment in periportal hepatocytes is virtually pathognomonic of hemosiderin, whereas perivenular pigment in hepatocytes is likely to be either lipofuscin (Figure 13-5) or bile (the latter when canalicular bile plugs are also identified). Intrasinusoidal brown pigment in Kupffer cells is usually hemosiderin (Figure 13-6), the result of hemolysis, transfusions or hemodialysis. Prussian blue iron stain is used to confirm the presence of hemosiderin, and Hall’s bile stain can be used to confirm the presence of bile.
PERTINENT NEGATIVES A number of commonly found abnormalities should routinely be excluded when examining liver tissue under the microscope. This list of ‘pertinent negatives’ includes the presence of fat, ground-glass inclusions, cholestasis, hemosiderin, a1-antitrypsin globules in peri-
208
portal hepatocytes (seen on diastase–PAS stain, or sometimes on H&E stain as eosinophilic globules) and liver cell dysplasia. Surveillance for these lesions is a helpful safeguard against inadvertently missing important diagnoses such as fatty liver disease, the chronic hepatitis B virus carrier state, a1-antitrypsin deficiency and various iron overload conditions.
HISTOPATHOLOGY OF LIVER DISEASES ACUTE HEPATITIS The histopathologic changes in acute hepatitis due to viruses and drugs have broad similarities, including lobular disarray, inflammation involving portal tracts and lobules, and hepatocellular degeneration in the form of ballooning and/or apoptosis (Figures 13-7 and 13-8). Because acute viral hepatitis is an immune-mediated disease, the portal tracts and sinusoids are infiltrated by lymphocytes (pre-
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-8. Acute hepatitis. Parenchymal changes include ballooning degeneration (long arrow), an apoptotic body (short arrow), lymphocytes and ceroid-filled Kupffer cells (double arrows). (Hematoxylin and eosin.)
Figure 13-9. Bridging hepatic necrosis on reticulin stain. Collapse of the reticulin framework between the central vein (at center) and the portal area at upper left has resulted from confluent necrosis (arrows). Note that in this hepatitis there is also intralobular reticulin condensation. (Gordon and Sweets reticulin stain.)
dominantly T cells). Within the lobule, lymphocytes are often seen in close approximation to hepatocytes, many of which show ballooning or apoptosis (Figure 13-8). Individual apoptotic bodies are evident within sinusoidal spaces. Enlarged ceroid-laden Kupffer cells are also frequently prominent within sinusoids. They appear tan on H&E stain and are positive on diastase–PAS staining. The lobular changes show a predilection for centrilobular regions, where reticulin staining may disclose condensed fibers (collapse of the reticulin framework) where hepatocyte necrosis is most pronounced. The term ‘spotty necrosis’ is used to refer to the panlobular changes of typical acute hepatitis. Confluence of regions of spotty necrosis results in progressively worse forms of hepatitis, including bridging hepatic necrosis (Figure 13-9) (extending between portal tracts, or between central veins and portal tracts), multilobular necrosis, and massive hepatic necrosis (the morphologic counterpart of fulminant hepatitis). Bridging necrosis and other severe lesions deserve special note because of their negative prognostic implications.13 Multilobu-
Figure 13-10. Massive hepatic necrosis. The bulk of the liver parenchyma in the lower right has undergone necrosis. Neocholangioles (proliferating bile ductules, at arrows) are seen emerging from the edge of the portal tract. (Hematoxylin and eosin.)
lar and massive hepatic necrosis often trigger a prominent proliferation of bile ductular structures in periportal regions (Figure 13-10); this ‘ductular reaction’ reflects the activity of progenitor or stem cells located in the periportal regions.14,15 Determination of etiology is often not feasible based on histopathology alone. Therefore, serologic studies to exclude hepatitis viruses and a comprehensive review of medications, particularly for drugs or alternative/herbal agents,16 should be undertaken. There may be certain helpful morphologic features, however. Among the hepatitis viruses, cholestasis may be unusually prominent with acute hepatitis A17 or E18 infections. Steatosis is commonly found in hepatitis C,19 particularly in patients with genotype 3 infection.20 The first presentation of autoimmune hepatitis may occasionally cause changes indistinguishable from those of acute viral hepatitis, so serologic evaluation for autoantibodies may be necessary in the appropriate clinical setting. This is especially important in the variant form of autoimmune hepatitis characterized by centrilobular necrosis and inflammation21 (Figure 13-11). In considering drug-induced acute hepatitis, a varied spectrum of histopathology may be seen.22 As indicated above, the prototypic changes of acute hepatitis are more commonly seen with agents responsible for idiosyncratic hepatotoxicity, such as isoniazid or nitrofurantoin, whereas predictable hepatotoxins (e.g. acetaminophen, carbon tetrachloride) often produce zonal liver injury affecting centrilobular regions (see Figure 13-4). Drugs can also produce selective injury of specific structures, such as paraquat-induced bile duct destruction23 or pyrrolizidine alkaloid-related endothelial damage24 in veno-occlusive disease (sinusoidal obstruction syndrome).25 Drug hepatotoxicity should always be considered if atypical histopathologic features such as steatosis, granulomas or bile duct damage are present in conjunction with the previously described changes of acute hepatitis.
CHRONIC HEPATITIS Chronic hepatitis is a disease process of multifactorial etiology (Table 13-4) defined as inflammation of the liver continuing without
209
Section II. Approach to the Patient with Liver Disease
Table 13-4. Etiology and Related Histopathology in Chronic Hepatitis Cause of chronic hepatitis
Histopathologic lesion
Chronic hepatitis B Chronic hepatitis C
Ground-glass hepatocytes Lymphoid aggregates/follicles in portal tracts Steatosis Bile duct damage Interface hepatitis Plasma cells Regenerative liver cell rosettes Diastase–PAS-positive globules in periportal hepatocytes Copper and copper-binding protein* Mallory bodies (periportal)
Autoimmune hepatitis
Alpha-1-antitrypsin deficiency Wilson’s disease
BD
*Absence of staining for copper or copper-binding protein does not preclude the diagnosis of Wilson’s disease.
Figure 13-12. Lymphoid aggregate in chronic hepatitis C. The portal tract shown contains a diffuse lymphoplasmacytic infiltrate and a dense, rounded aggregate of lymphocytes (between arrows) above the bile duct (BD). Lymphoid aggregates and follicles are a common diagnostic finding in chronic hepatitis C. (Hematoxylin and eosin.)
C
Figure 13-11. Centrilobular necrosis with inflammation (variant form of autoimmune hepatitis). Near the central vein (C) there is disruption of the liver plates by lymphocytes and plasma cells, with destruction of hepatocytes. This less common histologic form of autoimmune hepatitis may show only perivenular lesions, as shown here, or is sometimes accompanied by the more typical interface hepatitis of autoimmune hepatitis (see Figure 13-18). (Hematoxylin and eosin.)
improvement for 6 months or longer.26 The etiology is often known at the time liver biopsy is obtained, but morphologic assessment also provides useful information about causation (Table 13-4). For example, in chronic hepatitis C portal tract lymphoid aggregates are a common finding (Figure 13-12). Ground-glass cytoplasmic inclusions are seen in chronic hepatitis B (Figure 13-13) and may be stained with orcein and Victoria blue (Figure 13-14) as well as with specific immunohistochemical stains (Figure 13-15). The immunohistochemical demonstration of hepatitis B virus core antigen is important evidence of active viral replication (Figures 13-16 and 1317), particularly when there is abundant staining of nuclei and/or cytoplasm of hepatocytes. For autoimmune chronic hepatitis, histologic points in the International Autoimmune Hepatitis Scoring System27 are given for biopsies showing interface hepatitis, significant numbers of plasma cells among the portal and periportal infiltrates,28 and regenerative liver cell rosettes (Figure 13-18).
210
Figure 13-13. Ground-glass hepatocytes. The pink, homogeneous intracellular inclusions (arrows) represent hepatitis B surface antigen. They are often separated from the liver cell membrane by a halo (double arrow). (Hematoxylin and eosin.)
Diastase–PAS stain and methods for identifying copper and copperbinding protein are important in establishing the diagnosis of a1antitrypsin deficiency and Wilson’s disease, respectively (Figures 13-19 and 13-20). More importantly, liver biopsy provides a ‘snapshot’ of the necroinflammatory activity and fibrosis (if any) present at a given point in a dynamic disease process. Pathological scoring of the grade and stage of these respective components can be accomplished using one of a number of scoring systems.29 These provide both descriptive (i.e. mild, moderate, marked) and semiquantitative assessments (e.g. 0–4) that can be used in clinical management and in drug trials. The concept of scoring chronic hepatitis was first developed formally by Knodell and colleagues.30 The Scheuer,31 Ishak32 and METAVIR33 systems offer variations on the basic theme of grading and staging, and are now widely used. The Scheuer system34 (Table
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-14. Orcein stain of hepatitis B surface antigen. Note the numerous positive inclusions of HBsAg staining brown-black within hepatocytes (arrows). (Orcein stain.)
Figure 13-17. ‘Sanded’ nuclei. The light-blue ‘sanded’ appearance in many hepatocyte nuclei is due to the presence of intranuclear hepatitis B core antigen in this biopsy from a patient with chronic hepatitis B virus infection. (Hematoxylin and eosin.)
R
Figure 13-15. Hepatitis B surface antigen. Cytoplasmic hepatitis B surface antigen stains dark brown with the immunoperoxidase method. (Specific immunoperoxidase stain).
Figure 13-16. Hepatitis B core antigen. Extensive core antigen expression is present within hepatocyte nuclei, staining dark brown. Note that many cells also show positive cytoplasmic staining. (Specific immunoperoxidase stain).
Figure 13-18. Autoimmune hepatitis. The edge of an inflamed portal tract (at right) is shown here. Extension of lymphocytes and numerous plasma cells into the limiting plate region of hepatocytes constitutes interface hepatitis. A regenerative liver-cell rosette (R) is trapped between inflammatory cells. (Hematoxylin and eosin.)
13-5) illustrates a simple, readily practicable method of scoring. Selection of a given system is usually determined institutionally or is stipulated in protocols used in drug investigative studies. Grading of necroinflammation takes into account the intensity and distribution of inflammation, as well as liver cell damage such as apoptosis. Inflammation in chronic hepatitis may be present within portal tracts, in periportal regions (interface hepatitis), within the lobular parenchyma, or combinations of these. Staging takes into account the degree of fibrosis (or cirrhosis) present. Fibrosis evolves through the progressive deposition of scar tissue at the edges of portal tracts, the formation of bridging fibrous septa between portal tracts, and, in the most advanced cases, circumscription of regenerative nodules by diffuse fibrosis to form cirrhosis. Examples of grading and staging are shown in Figures 13-21 to 13-23 and 13-24 to13-25, respectively.
211
Section II. Approach to the Patient with Liver Disease
Figure 13-19. a1-Antitrypsin (AAT) deficiency. The purple globules of varied size within periportal hepatocytes represent abnormal AAT sequestered within endoplasmic reticulum cisternae of hepatocytes. The diastase–PAS staining method highlights non-glycogen glycoproteins such as AAT. (Periodic acid–Schiff stain with diastase digestion.)
Figure 13-22. Chronic hepatitis with marked activity (grade 4). The entire circumference of the portal tract shows interface hepatitis. (Hemotoxylin and eosin.)
Table 13-5. A Simple Scoring System for Chronic Hepatitis* Grade Portal inflammation and interface hepatitis 0 Absent or minimal 1 Portal inflammation only 2 Mild or localized interface hepatitis 3 Moderate or more extensive interface hepatitis 4 Severe and widespread interface hepatitis Lobular activity 0 None 1 Inflammatory cells but no hepatocellular damage 2 Focal necrosis or apoptosis 3 Severe hepatocellular damage 4 Damage includes bridging confluent necrosis
Figure 13-20. Copper-binding protein. The dark black-brown granules stained with the orcein method are intralysosomal metallothionein protein to which copper is bound in chronic biliary tract diseases. (Orcein stain.)
Stage 0 1 2 3 4
No fibrosis Fibrosis confined to portal tracts Periportal or portal–portal septa, but intact vascular relationships Fibrosis with distorted structure but no obvious cirrhosis Probable or definite cirrhosis
*Modified from Scheuer (Ref. 34)
BILIARY DISEASE AND CHOLESTASIS
Figure 13-21. Chronic hepatitis with minimal activity (grade 1). The portal tract at center shows a predominantly lymphocytic infiltrate without interface hepatitis. The lobular parenchyma is quiescent. (Hematoxylin and eosin.)
212
The presence of morphologic cholestasis on liver biopsy (Figure 1326) may be due to a variety of disorders affecting the liver itself (e.g. hepatitis, bile salt transporter mutations, sepsis) or to diseases specifically affecting the large or small bile ducts (Table 13-6). Pathologic differentiation between these two categories is essential, as therapeutic management differs substantially, with supportive or pharmacotherapeutic therapy in the former and endoscopic, radiologic or surgical intervention in instances of large bile duct obstruction. The pathology of cholestatic liver disease is discussed below in the context of large bile duct obstructive diseases, intrahepatic bile duct diseases and cholestatic parenchymal or extrahepatic diseases. Visible bile (or bilirubinostasis) is most often seen within hepato-
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-23. Chronic hepatitis with moderate to marked lobular activity (grade 3/4). Numerous foci of lobular inflammation with liver-cell dropout are present. (Hematoxylin and eosin.)
Figure 13-25. Chronic hepatitis with architectural distortion due to moderate fibrosis and nodular regeneration (stage 3). (Reticulin.)
Figure 13-26. Cholestasis. Bile is seen in canaliculi (single arrow), hepatocytes (double arrows), and sinusoidal Kupffer cells. (Hematoxylin and eosin.) Figure 13-24. Chronic hepatitis with mild fibrosis (stage 2). The portal tract shows an irregular, stellate appearance with mild fibrosis at its edges. (Masson trichrome.) Table 13-6. Conditions Associated with Cholestasis
cyte cytoplasm and/or bile canaliculi, with the greatest prominence in centrilobular regions. Bile leakage across bile canaliculi into sinusoids may result in Kupffer cell cholestasis. In adults with sepsis, bile may be present as inspissated concretions within periportal bile ductular structures35 (bile ductular cholestasis) (Figure 13-27). This unusual pattern can also be seen in pediatric liver biopsies in cases of extrahepatic biliary atresia.
LARGE BILE DUCT OBSTRUCTIVE DISEASES Mechanical obstruction of the large bile ducts has many causes, from gallstone impaction within the common bile duct to primary sclerosing cholangitis or cholangiocarcinoma. In addition to parenchymal cholestasis, a triad of portal changes develops within days of obstruction, including portal tract edema, neutrophil infiltrates and bile ductular structures36 (Figure 13-28). The currently preferred term for this reaction pattern is ‘the ductular reaction’37 (replacing
Large bile duct obstructive diseases Gallstone choledocholithiasis Biliary stricture Pancreatic carcinoma Chronic pancreatisis Primary sclerosing cholangitis Intrahepatic bile duct diseases Primary biliary cirrhosis Drug hepatotoxicity with bile duct injury Chronic liver allograft rejection Graft-versus-host disease Paucity of intrahepatic bile ducts Parenchymal or extrahepatic diseases Viral and drug hepatitis Sepsis Bile salt transport protein mutations Hodgkin’s and non-Hodgkin’s lymphoma
213
Section II. Approach to the Patient with Liver Disease
P
Figure 13-27. Bile ductular cholestasis (‘cholangitis lenta’). The presence of periportal bile ductular structures with inspissated bile is seen in sepsis due to Gram-negative and other organisms. (Hematoxylin and eosin.)
Figure 13-28. Portal tract changes in large bile duct obstruction. The portal tract is expanded by edema, inflammation, and a prominent ductular reaction (short arrows). Note the native bile duct (large arrow). (Hematoxylin and eosin.)
the term ‘bile ductular proliferation’). The genesis of the ductular structures is believed to be from periportal progenitor or stem cells.14 The ductular reaction can be highlighted by immunohistochemical staining for cytokeratins 7 or 19 (Figure 13-29). Neutrophil infiltrates in the edematous portal connective tissue, in close proximity to or within the ductular structures, reflect an active cytokine milieu.38 Other lesions considered pathognomonic of large duct obstruction include bile extravasates (Figure 13-30), infarcts (Figure 13-31) and lakes (Figure 13-32). The aforementioned lesions do not histologically differentiate the anatomic site of obstruction. Ascending infection in the setting of large bile duct obstruction (acute cholangitis) results in extensive portal neutrophil infiltrates within the connective tissue and within bile ducts and ductular structures (Figure 13-33). In primary sclerosing cholangitis (PSC), large bile ducts are commonly involved and diagnostic liver biopsy samples may show characteristic periductal ‘onion-skin’ fibrosis39 (Figure 13-34). Bile ducts
214
Figure 13-29. Cytokeratin 7 immunostain of the ductular reaction. A portal tract (P) is shown, including the native bile duct (at arrow) and brown-staining ductular structures comprising the ductular reaction at the edges of the tract. (Specific immunoperoxidase stain for cytokeratin 7.)
Figure 13-30. Bile extravasate. Bile has ruptured from the portal area into the parenchyma after large duct obstruction. (Hematoxylin and eosin.)
may eventually be replaced entirely by rounded fibrous scars (fibroobliterative cholangitis) (Figure 13-35). However, these diagnostic lesions are not always present and the more generic features of large duct obstruction are seen, or changes of chronic liver disease resembling chronic hepatitis, including interface hepatitis and portal tract lymphoid aggregate and follicle formation are found. In chronic large bile duct obstruction there is progressive portal and periportal fibrosis with the formation of bridging fibrous septa linking portal tracts. Parenchymal regeneration also occurs in tandem in the form of thickened liver-cell plates in periportal and periseptal regions. Prolonged retention of bile salts in periportal hepatocytes results in pale, bloated cells exhibiting pseudoxanthomatous change (cholate stasis) (Figure 13-36). These cells also show progressive accumulation of copper, with positive stains for copper (rhodanine, rubeanic acid) and copper-binding protein (orcein, Victoria blue) (Figure 13-37). Oxidative stress produced by
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-31. Bile infarct. A pale zone of faintly bile-tinged hepatocytes is present in this case of large duct obstruction. (Hematoxylin and eosin.)
Figure 13-34. Primary sclerosing cholangitis. There is concentric, periductal fibrosis (arrows). (Hematoxylin and eosin.)
Figure 13-32. Bile lake. A collection of parenchymal bile is seen in a case of large duct obstruction. (Hematoxylin and eosin.) Figure 13-35. Fibro-obliterative cholangitis. This histologic section was taken from an explanted liver in a patient with primary sclerosing cholangitis. A rounded scar has replaced the bile duct to the left of the hepatic artery branch. Other portal tracts in this specimen showed the more characteristic periductal ‘onion-skin’ fibrosis. (Masson trichrome.)
Figure 13-33. Acute cholangitis. Collections of neutrophils are present within bile ducts and ductular structures and in the portal connective tissue. (Hematoxylin and eosin.)
copper retention is one of several stimuli for the development of Mallory bodies,40 which are therefore also a feature of chronic biliary obstruction. Eventually, biliary cirrhosis develops, with thick fibrous septa linking portal tracts to one another, the ductular reaction still prominent (Figure 13-38). Studies of explanted livers in patients with PSC may show perihilar collections of bile and lipid material, with the formation of multinucleated giant cells and collections of lymphocytes and numerous plasma cells, termed xanthogranulomatous cholangiopathy41 (Figure 13-39). Bile duct loss (ductopenia) and fibro-obliterative lesions at this stage are useful markers of preceding primary sclerosing cholangitis. It should be noted that, possibly with the exception of fibro-obliterative scars, the majority of morphologic changes seen in the progressive stages of primary sclerosing cholangitis may also be seen in other chronic forms of large bile duct obstruction (i.e. secondary sclerosing cholangitis). The
215
Section II. Approach to the Patient with Liver Disease
Figure 13-36. Cholate stasis (pseudoxanthomatous change). The periportal hepatocytes show a foamy, vacuolated appearance owing to the chronic detergent effects of bile salt stasis in this area. This is a major feature of late biliary obstructive diseases. (Hematoxylin, phloxine, and saffron.)
Figure 13-38. Biliary cirrhosis. Late biliary obstructive diseases result in a cirrhosis that has a ‘geographic’ pattern of nodules or islands of liver parenchyma surrounded by broad fibrous septa interconnecting portal tracts. Note the continued prominence of the ductular reaction within the fibrosis. (Hematoxylin and eosin.)
GC
Figure 13-37. Victoria blue stain. This versatile method stains hepatitis B surface antigen, elastic tissue fibers and copper-binding protein. The latter is seen here as granular navy blue pigment in periportal hepatocytes. Compare to the orcein method shown in Figure 13-20.
Figure 13-39. Xanthogranulomatous cholangiopathy. In primary sclerosing cholangitis the rupture of biliary material into the hilar region results in collections of bile (upper left) surrounding macrophages (xanthoma cells with lipid), neutrophils and frequently prominent plasma cells, and even multinucleated foreign body giant cells (GC) with cholesterol. (Hematoxylin and eosin.)
corollary of this is that radiologic demonstration of PSC is therefore essential to its diagnosis.
damage proceeds. Granulomas form in a small percentage of cases, in conjunction with epithelial damage and other chronic inflammatory cell infiltrates. Stages 2–4 of PBC consist of the ductular reaction, portal and periportal fibrosis, and cirrhosis, respectively. It should be stressed that morphologic cholestasis in PBC is a late phenomenon in stages 3 and 4, when sufficient bile ducts have been destroyed throughout the liver and the presence of portal and periportal fibrosis presents a mechanical blockade for bile secretion. Other immune-mediated forms of intrahepatic bile duct injury, such as acute liver transplant rejection and graft-versus-host disease, bear similar pathologic features to the stage 1 lesion. Ultimately in late PBC, the disappearance of most intrahepatic bile ducts is marked
INTRAHEPATIC BILE DUCT DISEASES Primary biliary cirrhosis (PBC) is the prototype of immunemediated damage to intrahepatic bile ducts.39 The florid bile duct lesion of stage 1 PBC (Figure 13-40) demonstrates mononuclear cells (lymphocytes and plasma cells) with frequently prominent eosinophil infiltrates surrounding and infiltrating the epithelium of interlobular bile ducts, with resultant epithelial loss and attenuation, vacuolization, stratification and hyperplasia. The basement membrane (seen on diastase–PAS stain) becomes fragmented as immune
216
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-40. Bile duct damage in stage 1 primary biliary cirrhosis. The interlobular bile duct is surrounded by lymphocytes and plasma cells that transgress the basement membrane around the duct (arrow). (Hematoxylin and eosin.)
Figure 13-41. Large droplet (macrovesicular) steatosis. Large droplets of lipid are seen within hepatocytes. Causes include alcohol ingestion, obesity, and diabetes. (Hematoxylin and eosin.)
by the presence of lymphoid aggregates or follicles (lymphoid ‘tombstones’ of the ducts) and parenchymal features of chronic cholestasis described above for large bile duct obstructive diseases. Late PBC, therefore, is one of several conditions often termed ‘vanishing bile duct diseases’.10 Other intrahepatic bile duct diseases included under this rubric include drug-induced ductopenia (e.g. amoxicillin– clavulanic acid42), idiopathic adulthood ductopenia,43 chronic liver allograft rejection44 and sarcoidosis-related bile duct damage.45 In pediatric liver disease, syndromatic paucity of intrahepatic bile ducts (Alagille syndrome,46 non-syndromatic paucity), Byler disease (progressive intrahepatic cholestasis type 1) with bile duct paucity and postviral or metabolic disease-related bile duct paucity (e.g. post cytomegalovirus hepatitis and a1-antitrypsin deficiency, respectively) are included in this category.
CHOLESTATIC PARENCHYMAL OR EXTRAHEPATIC DISEASES Morphologic cholestasis on liver biopsy in the absence of diagnostic portal tract or bile duct changes occurs in several settings, including viral or drug hepatitis, sepsis, and specific bile transport disorders characterized by bile secretory defects at the level of the bile canaliculus.47 As with acute large bile duct obstruction, cholestasis is most prominent in centrilobular regions. In the case of hepatitis, the parenchymal changes of inflammation accompanied by liver cell damage clarify the cause of the cholestasis. In the absence of such changes, however, a spectrum of extrahepatic conditions may have to be considered in the differential diagnosis, including far-reaching cytokine-mediated effects such as those seen, for example, in Hodgkin’s and non-Hodgkin’s lymphoma-related cholestasis.48
FATTY LIVER DISEASES Abnormalities in lipid and lipoprotein metabolism result in steatosis49 (fatty liver), visible on light microscopy as the accumulation of empty vacuoles (largely triglyceride50) within hepatocytes. The most common type, large droplet (macrovesicular) steatosis (Figure 13-
Figure 13-42. Small droplet (microvesicular) steatosis. The pericentral hepatocytes show predominantly small droplets of lipid surrounding central nuclei. This type of fat is seen in Reye’s syndrome, during the third trimester of pregnancy, in urea cycle defects, and in toxicity due to nucleoside analogs. (Hematoxylin and eosin.)
41), preferentially affects centrilobular hepatocytes and is usually related to ethanol use, obesity, diabetes and/or the metabolic syndrome51 or corticosteroids. Hepatocyte nuclei in large droplet steatosis are compressed to the cellular periphery. This contrasts with small droplet (microvesicular) steatosis (Figure 13-42), in which small intracellular lipid vacuoles accumulate within hepatocytes surrounding a centrally placed nucleus. Microvesicular steatosis connotes more serious liver injury, with deficits in mitochondrial b-oxidation of fatty acids, as seen with liver injury due to nucleoside analogs,52 urea cycle defects53 and, classically, in acute fatty liver of pregnancy.54 Fatty liver diseases are currently classified broadly as alcoholic fatty liver disease (AFLD) and non-alcoholic fatty liver disease (NAFLD). Both share pathogenetic and pathologic features, including the potential progression to steatohepatitis and cirrhosis.55 The
217
Section II. Approach to the Patient with Liver Disease
Figure 13-43. Steatohepatitis. The constellation of changes in this condition includes steatosis, liver cell ballooning, Mallory bodies (at arrows) and inflammation. Note the satellitosis of neutrophils about the Mallory body-containing liver cells. (Hematoxylin and eosin.)
pathology of alcoholic liver disease was elucidated during the last century56 and has served as the basis for current work on NAFLD.3 Prolonged accumulation of macrovesicular fat in hepatocytes produces oxidative stress (by lipid peroxidation, cytokine-mediated effects and other mechanisms57). The evolution of fatty liver to steatohepatitis is marked by the development of a constellation of morphologic changes, including hepatocyte ballooning, inflammation, perivenular and perisinusoidal stellate cell-mediated fibrosis and Mallory body formation (Figure 13-43). Differentiation of alcoholic from non-alcoholic steatohepatitis (ASH vs NASH) on the basis of histology alone is difficult.58 As ASH and NASH progress, there is ensuing bridging fibrosis linking central veins to one another, central veins to portal tracts and, ultimately, cirrhosis (Figure 1344). A scoring system for NASH developed by Brunt and colleagues59 provides grading and staging scores for the degree of necroinflammatory changes and fibrosis, respectively, from 0 to 4.
VASCULAR DISEASES These conditions can be classified into two groups: conditions affecting the extrahepatic inflow vessels (hepatic artery and portal vein) and the outflow system (hepatic veins, inferior vena cava and heart), and conditions affecting the intrahepatic vessels (portal arterioles, veins and lymphatics; sinusoids; and centrilobular or efferent veins). The following discussion considers the more clinically common conditions affecting extrahepatic inflow and outflow system first, followed by the intrahepatic vascular system.
EXTRAHEPATIC INFLOW AND OUTFLOW VASCULAR DISORDERS Hepatic Artery Ligation or thrombosis of the hepatic artery is a major cause of hepatic infarction. The affected liver parenchyma typically shows contiguous lobules in which hepatocytes have undergone coagulative necrosis, with the disappearance of nuclei (Figure 13-45). Preserved parenchyma usually shows discrete centrilobular to midzonal necrosis with viable rims of periportal hepatocytes. Infarction may also
218
Figure 13-44. Steatohepatitis and developing cirrhosis. The centrilobular region (single arrow) is enclosed in fibrous tissue and has been linked to a portal area (open arrow) surrounding an early nodule (n). (Hematoxylin and eosin.)
P
Figure 13-45. Hepatic infarct. The entire right side of this field shows infarcted liver, with coagulative necrosis of hepatocytes, absence of liver cell nuclei, and sinusoidal congestion. This patient had undergone liver transplantation 10 days earlier and developed hepatic artery thrombosis. Note an early infiltrate of acute rejection in the portal tract at left (P) and preservation of periportal parenchyma. (Hematoxylin and eosin.)
develop following injection of chemotherapeutic agents (free, or in gel form) for ablating liver tumors. Bile duct necrosis followed by biliary stricture, obstruction and a sclerosing cholangitis-like picture has been described after 5-fluorouracil arterial injection for metastatic colon carcinoma,60 or may occur after gel chemoembolization (Figure 13-46) or due to thrombotic or other interruption of hepatic artery perfusion (Figure 13-46). Hepatic artery thrombosis is also a recognized complication following liver transplantation.61 Mediumsized branches of the hepatic artery may be involved by polyarteritis nodosa, with resultant ischemic damage, and approximately onethird of these patients have underlying hepatitis B virus infection.62 Because hepatic artery perfusion is dependent on left ventricular function, any serious compromise in cardiac ejection fraction
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
C
Figure 13-46. Bile duct necrosis. The brown-staining structure at the center is the remnant of a necrotic large bile duct, with complete necrosis of the duct epithelium, which is densely infiltrated by bile. This patient had sustained surgical trauma to the hepatic artery. (Hematoxylin and eosin.)
Figure 13-47. Ischemic hepatitis. This postmortem specimen shows a central vein (C) and surrounding congested parenchyma from a patient who died in cardiac failure. Note the collections of neutrophils (arrows) in areas of hepatocyte dropout. (Hematoxylin and eosin.)
(particularly congestive heart failure) is likely to result in centrilobular ischemia with coagulative necrosis. Perivenular neutrophil infiltrates surrounding necrotic hepatocytes (ischemic hepatitis) (Figure 13-47) may develop within a day or two if perfusion improves or is supported by pressor agents.63 The hospitalized patient with multisystem disease and episodes of hypotensive or septic shock is a likely candidate for such centrilobular insults. Such changes are particularly common in postmortem liver material. It should also be noted in this context that if cirrhosis is present, cirrhotic nodules are served primarily by arterial perfusion (as portal venous inflow is obstructed), and decreased hepatic perfusion due to cardiac failure, ruptured varices or septic shock, for example, will produce necrosis involving the centers of or entire cirrhotic nodules.
Portal Vein The major condition affecting the portal venous system is cirrhosis with portal hypertension (see Cirrhosis). Pylephlebitis refers to inflammation affecting the major portal vein and/or its branches (Figure 13-48) and is seen in association with intra-abdominal suppurative processes such as perforated ulcers, or appendiceal rupture in appendicitis. The most common cause of portal vein thrombosis is cirrhosis.64
Hepatic Veins and Inferior Vena Cava Budd–Chiari syndrome65–68 refers to obstruction of venous outflow from the liver owing to thrombosis or obstruction of the major hepatic veins and/or inferior vena cava. Prothrombotic disorders and oral contraceptive use are the most common etiologic factors. The liver shows centrilobular sinusoidal dilatation and congestion (Figure 13-49). Prolonged venous outflow obstruction may result in perisinusoidal fibrosis in centrilobular regions, with bridging fibrosis to central veins or to portal tracts, and eventually cirrhosis. Venous outflow obstruction combined with regions of arterial hyperperfusion may stimulate regenerative hyperplasia, evident as nodular
Figure 13-48. Pylephlebitis. The portal vein branch outlined by the arrows contains neutrophils and suppurative debris. This ascending portal vein inflammatory lesion develops in individuals with intra-abdominal suppurative processes such as perforated ulcers and ruptured acute appendicitis. (Hematoxylin and eosin.)
regenerative hyperplasia (to be distinguished from cirrhosis, in which regenerative nodules are surrounded diffusely by cirrhosis), or as solitary or multiple nodules resembling focal nodular hyperplasia66 (see Tumors and tumor-like lesions). Congential webs of the inferior vena cava are a rare cause of Budd–Chiari syndrome.
RIGHT VENTRICULAR FAILURE, COR PULMONALE AND CONSTRICTIVE PERICARDITIS As with lesions affecting the hepatic veins and inferior vena cava, right heart failure, cor pulmonale and constrictive pericarditis effectively produce hepatic venous outflow obstruction, with resultant
219
Section II. Approach to the Patient with Liver Disease
Figure 13-49. Centrilobular congestion and sinusoidal dilatation. Conditions of impaired hepatic venous outflow, as in heart failure and Budd-Chiari syndrome, produce such a pattern. Note the intact portal tracts and periportal parenchyma (arrows). (Hematoxylin and eosin.)
sinusoidal dilatation and congestion in centrilobular regions. The centrilobular liver cell plates are thin, compressed and atrophic, and may be surrounded by perisinusoidal fibrosis, often referred to as cardiac sclerosis.
INTRAHEPATIC VASCULAR DISORDERS Central Veins The classic lesion involving central veins is veno-occlusive disease (VOD), in which fibrous obliteration of central veins follows endothelial damage. Inciting agents include pyrrolizidine alkaloids24 in ‘bush tea’, azathioprine,69 and certain chemotherapeutic agents. Fibrosis within and surrounding central veins may also be a component of steatohepatitis70 and hepatic venous outflow obstruction. Endotheliitis of central veins (central venulitis) may accompany portal vein endotheliitis in acute liver transplant rejection, and when accompanied by congestion and perivenular necrosis may presage chronic rejection.71,72 Patchy central venulitis is also a feature of viral (Figure 13-50) or drug-induced hepatitis.73 Centrilobular necrosis with lymphoplasmacytic inflammation can also be seen as a histologic variant of autoimmune hepatitis21 (see Figure 13-11), alone or in combination with the more characteristic portal and periportal plasma cell-rich interface hepatitis seen in most cases.
Sinusoids Sinusoidal dilatation is a common pathologic finding in liver biopsies, most often seen in conjunction with variable degrees of congestion in cardiac failure and other causes of venous outflow obstruction (Figure 13-49). It may also be present when a mass lesion is nearby, though not sampled in the biopsy specimen,74 or may be due to oral contraceptives75 (particularly if periportal or midzonal). The most extreme example of sinusoidal congestion is known as peliosis hepatis (blood lakes), in which patulous, broad regions of lobular parenchyma are filled with blood and the endothelial lining is often incomplete. This is seen in association with neoplastic diseases, systemic diseases such as AIDS,76 drugs, and in the setting of liver and kidney transplantation.77,78
220
Figure 13-50. Endotheliitis. In this example there is mononuclear cell inflammation of the central vein. This patient had viral hepatitis. Endotheliitis (lymphocyte attachment to class II histocompatibility antigens on endothelium) is also an important pathologic feature seen in acute liver transplant rejection. (Hematoxylin and eosin.)
Portal Veins Portal vein endotheliitis is an important aspect of the triad of findings in acute liver transplant rejection.8 Effector immune cells (primarily lymphocytes, with admixed eosinophils, neutrophils or plasma cells) are attached to the endothelium within vessel lumina and also infiltrate in the subendothelial position, with resultant lifting off of the endothelial cells. Thrombosis of portal vein branches within the liver may be found in cases of nodular regenerative hyperplasia in which the hyperplastic response develops as a result of a combination of parenchymal extinction and residual arterial perfusion.79
Hepatic Arterioles Abnormal numbers of irregular vessels (arterioles, veins, lymphatics) within portal tracts, fibrous tissue, or in proximity to sinusoids are the major feature seen in hereditary hemorrhagic telangiectasia.80 Muscular hypertrophy of arterioles is commonly found in older individuals with systemic hypertension.
METABOLIC AND METAL STORAGE DISEASES Documentation of hepatic involvement in metabolic and metal storage diseases utilizes light microscopy of liver sections, special histochemical and/or immunohistochemical stains, and other possible modalities, such as tissue quantification of the storage product (e.g. iron or copper quantification for hereditary hemochromatosis or Wilson’s disease, respectively), transmission electron microscopy (e.g. sphingomyelin electron-dense whorled lysosomal inclusions in Niemann–Pick disease), and even DNA analysis from archived paraffin-embedded tissue (e.g. single-strand conformational polymorphism analysis for a1-antitrypsin phenotype). If metabolic disease is clinically suspected, particularly in pediatric cases, the appropriate techniques that will optimize establishing the diagnosis should be considered prior to obtaining the liver biopsy. For example, if transmission electron microscopy may be required (as in inherited lysosomal storage disorders), several 1 mm cubes of liver
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-51. Niemann–Pick sphingolipidosis. The sinusoids are replete with enlarged, foamy Kupffer cells with sphingomyelin (arrows). Note that the hepatocytes also contain vacuolated material consistent with sphingomyelin. (Hematoxylin and eosin.)
Figure 13-52. Glycogen nuclei. Glycogen nuclei (‘empty’) inclusions in hepatocyte nuclei are often seen in diabetes but may be present in pediatric liver biopsies, in Wilson’s disease, and in livers with metastatic tumor. (Hematoxylin and eosin.)
tissue should be fixed in cold glutaraldehyde. A small portion of the specimen may be snap frozen for specific enzyme analysis or other studies. The paraffin block of liver tissue can be sent for quantification of iron or copper. The range of hepatic pathology in metabolic disorders is broad and is discussed in several reviews.81 Lysosomal storage disorders often involve hepatic macrophages (Kupffer cells and portal macrophages) with or without hepatocyte involvement (Figure 1351). Several types of glycogen storage disease show alterations in hepatocyte cytoplasm owing to the accumulated glycogen, and may also result in fibrosis or cirrhosis. Glycogenated hepatocyte nuclei (‘glycogen nuclei’) are prominent in periportal regions in diabetes (Figure 13-52). They are also common in pediatric liver biopsies and are a fairly constant feature in Wilson’s disease. The remainder of this section will focus on a1-antitrypsin (AAT) deficiency, hereditary hemochromatosis and Wilson’s disease, and the histopathologic approach to their diagnosis. AAT deficiency shows protean manifestations at various ages, from giant cell hepatitis or paucity of intrahepatic bile ducts with cholestasis in the neonate to portal fibrosis in adolescents and chronic hepatitis in adults.82 These varied tissue expressions of gene mutations in AAT alleles (the most important being the Z gene, which encodes a substitution of lysine for glutamic acid at position 342) develop because of accumulated diastase–PAS-positive inclusions of polymerized abnormal AAT within cisternae of the endoplasmic reticulum (see Figure 13-19). Immunohistochemical staining for AAT is used to confirm the identity of the intracellular globules. AAT deficiency has therefore recently been considered, along with Alzheimer’s and Parkinson’s disease, a ‘serpinopathy’ (a conformational disorder due to abnormal folding and degradation of serine proteases83). The normal AAT phenotype is PiMM (protease inhibitor MM), with the M or wild-type gene encoding normal AAT structure and serum levels. In homozygous PiZZ disease markedly reduced serum AAT levels are associated with liver disease and panacinar emphysema. Various heterozygous forms of AAT
deficiency exist, including PiMZ and PiSZ, and are also associated with hepatic pathology. The diagnosis should be suspected when diastase–PAS-positive globules of various sizes are found in periportal hepatocytes. In neonates under 13 weeks of age the diagnostic inclusions may be poorly developed, and phenotyping should therefore be performed to exclude the diagnosis. In adults, liver biopsy may show chronic hepatitis, large cell dysplasia, cirrhosis, or even hepatocellular carcinoma.84 Hereditary hemochromatosis is a primary iron overload disorder85 affecting the liver, and the majority of cases with the classic phenotype of multiorgan iron storage in parenchymal cells are associated with homozygous mutations of the HFE gene (C282Y/C282Y). Non-HFE mutations (e.g. genes for transferrin receptor-2, ferroportin-1) are uncommon causes of primary iron overload that in many instances closely resemble classic HFE-related hemochromatosis. Histologic studies of the liver in hereditary hemochromatosis show extensive hepatocellular deposits of hemosiderin with a periportal-to-central diminishing gradient86 (Figures 13-53 and 1354). Kupffer cell hemosiderin is typically absent or minimal. Following several decades of parenchymal iron loading, a ‘hollyleaf ’ pattern of portal and periportal fibrosis develops, followed by cirrhosis87 if the condition remains undiagnosed and untreated. During the period of fibrogenesis, bile ducts and portal macrophages demonstrate increasing amounts of hemosiderin. Once the cirrhotic stage is reached, the risk of hepatocellular carcinoma is markedly increased.88 Secondary iron overload of cirrhotic livers due to causes other than hereditary hemochromatosis (e.g. cirrhosis following alcohol use, chronic hepatitis C) may occasionally resemble the genetic disease histologically,89 and a combined assessment of gene test results, clinical features, liver tissue quantification and determination of the hepatic iron index90 may be necessary to fully distinguish one from the other. It should be noted, however, that in routine liver biopsy practice Kupffer cell hemosiderosis (see Figure 13-6) is encountered far more commonly than the marked hepatocellular hemosiderosis seen in hereditary hemochromatosis. Such causes as hemolysis,
221
Section II. Approach to the Patient with Liver Disease
Figure 13-53. Hereditary hemochromatosis. Massive brown hemosiderin pigment is present in bile ducts (long arrow), portal macrophages (short arrows), and hepatocytes (open arrows). (Hematoxylin and eosin.)
Figure 13-55. Gross morphology of cirrhosis. Micronodular cirrhosis (bottom) and macronodular cirrhosis (top).
Figure 13-54. Early hemochromatosis. There is moderate to marked liver cell hemosiderosis, grade 2–3/4. The periportal parenchyma (the portal tract is at the arrow) marks the region of heaviest iron deposition. A decreasing gradient is seen from periportal to pericentral regions. (Prussian blue iron stain.)
Figure 13-56. Micronodular cirrhosis. The liver biopsy specimen has fragmented and shows circumscribed nodules the size of lobules or less. (Reticulin stain.)
transfusions and hemodialysis comprise the major differential diagnosis of Kupffer cell hemosiderosis in these cases. Wilson’s disease, the primary copper overload disease affecting the liver and nervous system, is due to mutations in the ATP7B gene whose protein product is located in the trans-Golgi site and mediates vesicular transport of copper to the bile canaliculus.91 The accumulation of copper within the liver results in diverse pathologic features, with early steatosis and glycogenated hepatocyte nuclei (Figure 13-52) and distinctive mitochondrial changes (size variations, abnormal cristae, vacuoles) on electron microscopy.92 Chronic hepatitis, periportal Mallory bodies and cirrhosis are seen later.93 Massive hepatic necrosis may develop, superimposed on an underlying cirrhosis.94 Stains for copper (rhodanine, rubeanic acid) and copper-binding protein (orcein, Victoria blue) may be applied to routine paraffin-embedded liver tissue, but the absence of staining does not preclude the diagnosis. Tissue quantification of fresh or
222
paraffin-embedded liver specimens is often the most useful test for confirmation of the diagnosis, and should demonstrate liver copper in excess of 250 mg/g dry weight.95 Hepatic copper overload also develops in individuals with chronic cholestasis and biliary tract obstruction (see Biliary tract disease and cholestasis), in which case stainable copper and copper-binding protein are prominent in periportal hepatocytes. Quantitative liver copper levels are elevated, but in a lesser range than in Wilson’s disease.
CIRRHOSIS The morphologic diagnosis of cirrhosis is based on demonstrating diffuse fibrosis surrounding architecturally abnormal regenerative nodules. On gross examination cirrhosis can be subdivided into micronodular (nodules <3 mm in diameter) (Figure 13-55), macronodular (nodules >3 mm in diameter; Figure 13-56) and mixed micro-macronodular types. The morphologic features of cirrhosis correlated with etiology were reported in the classic 1988
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-57. Macronodular cirrhosis. Compare the nodule size and irregularity of the septa with those in Figure 13-56. (Reticulin stain.)
Figure 13-58. Incomplete septal cirrhosis. This form of cirrhosis, a variant of macronodular cirrhosis, shows nodules intersected by incomplete fibrous septa. (Reticulin stain.)
publication in the Bulletin of the World Health Organization.96 On microscopy, nodules the size of lobules or smaller are consistent with micronodular cirrhosis and are usually fairly uniform in size (Figure 13-56). In contrast, variability of nodule size (usually exceeding that of normal lobules) and the presence of thin, fibrous septa are seen in macronodular cirrhosis (Figure 13-57). Incomplete septal cirrhosis has been considered a variant of macronodular cirrhosis in which the fibrous septa surrounding nodules show regions of incomplete formation (Figure 13-58). An alternative hypothesis, however, is that this microscopic form represents a type of regressed cirrhosis in which fibrosis has been partially removed.97 The possibility of true reversal of cirrhosis has been subject to controversy, particularly the difficulty of reversing the abnormal regenerative hyperplasia and the altered vascular relationships established therein.98 The pathologic characterization of cirrhosis should take into account the etiology, activity and possible complications. Evidence of the possible underlying disease processes that lead to cirrhosis (i.e. steatohepatitis, chronic hepatitis, biliary tract diseases, hepatic
Figure 13-59. Active cirrhosis. The portal tracts are markedly inflamed and there is ongoing erosion of the interface of the parenchyma and fibrosis. (Hematoxylin and eosin.)
venous outflow conditions) should be sought microscopically. For example, a preponderance of portal lymphoplasmacytic infiltrates with interface hepatitis and portal tract-based fibrous septa is consistent with chronic hepatitis as the underlying pathogenesis. Histologic features such as ground-glass inclusions (i.e. chronic hepatitis B), portal lymphoid aggregates or follicles (chronic hepatitis C), or excessive numbers of plasma cells (autoimmune hepatitis) are helpful in further delineating the cause. In contrast, if steatosis is present the pathologist should review the status of the centrilobular regions and central veins for current evidence of steatohepatitis and a primary axis of centrilobular fibrosis. The presence of spared (uninvolved) portal tracts within cirrhotic nodules (‘reversed lobulation’) is a useful finding in conditions such as steatohepatitis or chronic hepatic venous outflow obstruction, which exert their major pathophysiologic effects on perivenular regions. In chronic biliary tract disease with cirrhosis the ductular reaction usually remains an important feature in cirrhosis, and the additional presence of cholestasis and periportal changes such as pseudoxanthomatous change, Mallory bodies, and copper-binding protein on orcein or Victoria blue staining helps confirm this impression. Absence or paucity of interlobular bile ducts also needs to be excluded in the differential diagnosis of cirrhosis with biliary features. The activity of cirrhosis is assessed microscopically based on the degree of current necroinflammation as well as the quality of the fibrosis (Figure 13-59). For cirrhosis following chronic hepatitis, the degree of ongoing interface hepatitis and/or intraparenchymal necroinflammation forms the basis to establish the histological activity (Figure 13-60) (see Table 13-5). The degree of ongoing steatohepatitis in cirrhosis can also be assessed using a scoring system.59 The evolution of biliary cirrhosis may vary considerably in a given histologic section: well formed cirrhotic nodules may coexist adjacent to areas where there is only mild periportal fibrosis. The inference from this is that small needle biopsies in chronic biliary tract disease must be interpreted cautiously with regard to the presence or absence of established cirrhosis. Special stains are usually helpful in evaluating the nature of fibrous scars as cirrhosis evolves. For example, old fibrous septa typically contain elastic fibers that can be
223
Section II. Approach to the Patient with Liver Disease
Figure 13-60. Active cirrhosis. Ongoing interface hepatitis (arrows) is present, with resultant irregularity of the limiting plate region. The infiltrate includes plasma cells and lymphocytes. (Hematoxylin and eosin.)
Figure 13-62. Liver cell dysplasia, large-cell type. In this nodule from a cirrhotic liver the hepatocytes at left show enlargement with nuclear atypia, features characteristic of the large-cell type of dysplasia. (Hematoxylin and eosin.)
Figure 13-61. Inactive cirrhosis. There is virtually no inflammatory activity, either within the fibrous septa or at the parenchymal–septal interface. (Hematoxylin and eosin.)
Figure 13-63. Hepatic granulomas. There are multiple parenchymal granulomas (arrows) composed of plump, epithelioid macrophages near the central vein, which is at right. The cause was leishmaniasis (kala-azar). (Hematoxylin and eosin.)
demonstrated on Victoria blue staining (in contrast to necrosis or early fibrosis). Collagen stains, such as the Masson trichrome method, also provide a helpful index of the age of the fibrosis (Figure 13-1d). Inactive cirrhosis is usually characterized by well formed regenerative nodules and fibrous septa, with little inflammation (Figure 13-61). Among the important complications to be noted in cirrhotic specimens are liver cell dysplasia and hepatocellular carcinoma. Large-cell and small-cell types of liver cell dysplasia are described. Large-cell dysplasia (large-cell change) (Figure 13-62) shows enlarged hepatocytes with atypical nuclei. This has long been associated with an increased risk of hepatocellular carcinoma.99 However, more recently small-cell dysplasia (small-cell change) has been considered the more likely directly preneoplastic condition.100 Foci of small-cell dysplasia show small hepatocytes with atypical nuclei. A further complication of cirrhosis often present in explant liver specimens and postmortem liver is necrosis at the centers of
224
nodules (centronodular necrosis) if there has been a recent period of decreased hepatic perfusion, as in hypotensive or septic shock.
HEPATIC GRANULOMAS The granuloma (Figure 13-63) as an immunologic response is caused by numerous factors.101 The common denominator is usually ineffective clearance of a foreign antigen, for example mycobacteria, fungus, parasite, or immunologically generated antigenic material following a drug reaction. Hepatic granulomas are a common finding on liver biopsy, identified in up to 10% of liver specimens. The most common causes are sarcoidosis, primary biliary cirrhosis, tuberculosis, drugs, and schistosomiasis102,103 (see Chapter 39). The major role of the pathologist in evaluating hepatic granulomas is in establishing obvious causes by routine histopathologic observations (e.g. ova of schistosomes, damage to bile ducts) and through the use of special stains (e.g. mycobacteria, fungi). Accord-
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
F
F
Figure 13-64. Sarcoidosis. Three non-caseating granulomas are present in this portal tract (arrows) and show adjacent hyaline fibrosis (F). (Hematoxylin and eosin.)
ingly, the pathologic work-up of granulomas necessitates the use of a series of special stains, including acid-fast, silver, and PAS stains. The reader is referred to several comprehensive reviews103–106 and to Chapter 39 for additional details. The microscopic location of the granuloma(s) and related changes is also helpful, but does not provide the etiology in all cases. In sarcoidosis, non-necrotizing (non-caseating) granulomas tend to cluster in and around portal tracts, often producing dense hyaline fibrosis (Figure 13-64). In some cases sarcoidosis also elicits substantial portal, periportal and lobular lymphoplasmacytic inflammation, suggesting the diagnosis of concomitant hepatitis,107 or results in bile duct damage or destruction, resembling primary biliary cirrhosis.45 In a recent 10-year survey of hepatic granulomas, hepatitis C infection was noted as the cause in nearly 10% of cases,108 although their pathogenesis in this setting is unknown. Simon and Woolf109 described an idiopathic febrile illness with hepatic granulomas that was steroid responsive, also sometimes effectively treated with methotrexate.110
Figure 13-65. Neonatal giant cell hepatitis. This field shows several markedly enlarged, multinucleated giant hepatocytes. The brown pigment within the cells is bile. (Hematoxylin and eosin.)
Figure 13-66. Extrahepatic biliary atresia. The fibrotic portal tract shown here contains a prominent ductular reaction. Many of the ductular structures contain bile. (Hematoxylin and eosin.)
Pediatric Liver Disease Neonatal and later pediatric liver diseases present a distinctive set of pathologic challenges in liver biopsy interpretation. Among the common problems encountered are the evaluation of neonatal jaundice, establishing the diagnosis of extrahepatic biliary atresia, demonstrating the presence of a storage disorder, grading and staging chronic hepatitis, and the diagnosis of liver tumors. The increasing prevalence of obesity in childhood has also focused attention on fatty liver disease. Liver biopsy in the neonate with jaundice is used to differentiate neonatal hepatitis from other causes, including extrahepatic biliary atresia, metabolic disorders (such as a1-antitrypsin deficiency) and paucity of intrahepatic bile duct disorders (syndromatic and non-syndromatic). In neonatal hepatitis liver biopsy shows prominent giant, multinucleated hepatocytes (so-called ‘giant cell hepatitis’) (Figure 13-65) as well as variable portal and lobular inflammation with
hepatocyte degeneration. It should be noted that giant cell transformation of hepatocytes is a stereotypic change induced by cholestasis of varying etiologies, including biliary atresia (though they are far less numerous, or absent, in the latter). The major diagnostic biopsy feature of biliary atresia is the portal and periportal ductular reaction (Figure 13-66). Inspissated bile is frequently present within the proliferated portal and periportal ductular structures. Depending on the age at which biopsy is obtained, variable portal and periportal fibrosis is present, with biopsies obtained at 7 weeks or later often showing a developing biliary cirrhosis. In Alagille syndrome (syndromatic paucity of intrahepatic bile ducts) there is cholestasis with conspicuous loss of interlobular bile ducts from portal tracts (Figure 13-67). This condition results from mutation of the Jagged-1 gene
225
Section II. Approach to the Patient with Liver Disease
Figure 13-67. Alagille syndrome (syndromatic paucity of intrahepatic bile ducts). The diagnostic portal tract shown here contains a prominent hepatic arteriole but no corresponding bile duct. (Hematoxylin and eosin.)
and comprises a syndrome of anomalies including pulmonic stenosis, abnormal facies, butterfly vertebrae, and posterior embryotoxin in the optic fundus. Patients with non-syndromatic paucity of intrahepatic bile ducts show bile duct paucity on liver biopsy but do not have the spectrum of anomalies seen in Alagille syndrome. Metabolic storage disorders were discussed in an earlier section. Among the major pediatric liver tumors to be considered briefly here are hepatoblastoma, mesenchymal hamartoma and infantile hemangioendothelioma. Hepatoblastoma is a malignant tumor that develops in children under 5, with most presenting in the first 2 years of life. These tumors can be broadly subdivided into epithelial and mixed epithelial–mesenchymal types. The epithelial hepatoblastoma shows either the ‘fetal’ growth pattern (Figure 13-68) (sheets of hepatocytes with granular or clear cytoplasm with adjacent intrasinusoidal extramedullary hematopoiesis), the ‘embryonal’ pattern (tubules containing epithelium with high nucleus:cytoplasmic ratios and mitotic figures), or both. Mesenchymal components that may be present include osteoid and primitive cartilage. Mesenchymal hamartoma is a multicystic tumor composed of islands of hepatocytes, proliferated biliary structures (thought to be derived from the embryonic bile duct plate), loose mesenchyme and lymphatic channels (Figure 13-69). It typically develops under the age of 2. Infantile hemangioendothelioma is usually diagnosed in the first 6 months of life when hepatomegaly, mass, cardiac failure (due to arteriovenous shunting in the tumor) or failure to thrive are present. This multinodular tumor is composed of endothelium-lined vascular channels which often surround intrahepatic bile ducts (Figure 13-70). The malignant form of this tumor (formerly designated type 2 infantile hemangioendothelioma) is now considered to represent angiosarcoma.111
Tumors and Tumor-like Lesions Tumors of the liver may arise from indigenous cells of the liver or their progenitors, including hepatocytes (adenoma, hepatocellular
226
Figure 13-68. Hepatoblastoma, fetal epithelial type. This pediatric liver tumor grows in several patterns. The fetal epithelial type grows in sheets of light and dark hepatocytes and frequently contains foci of extramedullary hematopoiesis (cluster of dark cells at center). (Hematoxylin and eosin.)
Figure 13-69. Mesenchymal hamartoma. This tumor is composed of hepatocytes (top left and right), abnormal bile duct structures (center) and loose stroma with lymphatics (bottom). (Hematoxylin and eosin.)
carcinoma, hepatoblastoma), bile duct epithelium (adenoma, cystadenoma/cystadenocarcinoma, cholangiocarcinoma), endothelial cells (hemangioma, angiosarcoma, Kaposi’s sarcoma, epithelioid hemangioendothelioma), nerves (neurofibroma, malignant schwannoma), mixed elements (mesenchymal hamartoma), or various mesenchymal tissues (lipoma, fibroma, leiomyoma, and their malignant counterparts).112,113 In addition, nodular non-cirrhotic conditions are occasionally encountered, including focal nodular hyperplasia,114 partial nodular transformation,115,116 and nodular regenerative hyperplasia (nodular transformation).117–119 Tumors are considered in greater detail in Chapters 60–62. A number of general observations can be made about the diagnosis of tumors or tumor-like conditions encountered in routine
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-70. Infantile hemangioendothelioma. This pediatric tumor arises from a proliferation of vascular endothelium to create numerous slit-like spaces, often trapping indigenous bile ducts (at center). (Hematoxylin and eosin.)
Figure 13-72. Bile duct adenoma. These nodular masses are usually located at the liver capsule or just beneath. They are composed of medium-sized bile duct structures embedded in fibrous tissue. Clusters of lymphocytes (right of center) are often present at their interface with adjacent normal liver parenchyma. (Hematoxylin and eosin.)
Figure 13-71. Microhamartoma (von Meyenburg complex or bile duct malformation). This portal-based structure (at arrow) consists of irregularly dilated bile ductular structures (embryonic ductal plate malformations) embedded in fibrous tissue. (Hematoxylin and eosin.)
Figure 13-73. Focal nodular hyperplasia (gross resection specimen). This nodular, non-cirrhotic lesion shows a characteristic central, stellate scar.
practice. Solitary, small primary liver tumors found at autopsy or laparotomy are likely to be either hemangiomas,120 focal nodular hyperplasia, or microhamartomas (bile duct malformations or von Meyenburg complexes121). The microhamartoma is a white surface nodule several millimeters in diameter and must be distinguished from metastatic tumor; frozen section of the former shows a portal or periportal collection of abnormal, focally dilated bile ducts embedded in fibrous tissue (Figure 13-71), representing embryologic maldevelopment of the bile duct plate.122–125 These are to be distinguished from true bile duct adenomas,121,126,127 which also appear as small white surface nodules. These are composed of medium-sized regular bile duct structures embedded in fibrous tissue, often with collections of lymphocytes aggregated at their periphery (Figure 13-72). Focal nodular hyperplasia (FNH) is a
multilobulated lesion characterized by a central, stellate scar which subdivides cirrhosis-like nodular units of liver parenchyma114,128 (Figure 13-73). It is considered a hyperplastic, regenerative parenchymal lesion arising from hyperperfusion by an arterial malformation within the central scar. Multiple FNH may infrequently be present.129 A rare nodular lesion of the hilum which can produce portal hypertension is partial nodular transformation,115,116 in which the parenchyma has regionally undergone regenerative hyperplasia. Nodular regenerative hyperplasia,117–119 on the other hand, is a diffuse condition in which there are many parenchymal nodules showing regenerative hyperplasia but little or no attendant fibrosis (Figures 13-74–13-76). A liver cell adenoma is a well circumscribed or encapsulated tan homogeneous mass composed of thickened sheets of benign hepatocytes, venules and arterioles, but without bile ducts or true portal tracts130–133 (Figures 13-77 and 13-78). Oral
227
Section II. Approach to the Patient with Liver Disease
Figure 13-74. Nodular regenerative hyperplasia (gross postmortem specimen). Note the small parenchymal nodules without associated fibrosis.
Figure 13-76. Nodular regenerative hyperplasia. The junction of a nodule with normal liver (at arrows) shows the transition from thickened liver cell plates (above) to dilated sinusoids and one-cell-thick plates (below). (Hematoxylin and eosin.)
Figure 13-75. Nodular regenerative hyperplasia. There is parenchymal nodularity (n) with no intervening fibrosis. The separation of nodules is indicated by sinusoidal dilatation and compression of liver cell plates. (Hematoxylin and eosin.)
Figure 13-77. Liver cell adenoma (gross resection specimen). The tumor is pale tan, homogeneous, and well circumscribed from the rest of the liver.
contraceptives134 and mutations of the gene for hepatocyte nuclear factor-1a131(particularly in patients with adenomatosis or multiple adenomas) are important pathogenetic factors. The major malignant liver tumor (apart from metastasis) seen in practice is hepatocellular carcinoma (HCC). Liver biopsy specimens containing HCC are fragmented (Figure 13-79) and paler than normal liver tissue. Reticulin stain is useful because in contrast to primary carcinomas of the lung or gastrointestinal tract, these epithelial neoplasms show poor formation of reticulin. In a minority of cases bile production can be demonstrated histologically and is therefore diagnostic. An underlying cirrhosis is present in the majority of cases of HCC, except in the fibrolamellar variant (Figure 13-80), where this does not obtain.135–137
Liver Transplantation Liver biopsy is fundamental to the management of liver transplant recipients, particularly in monitoring rejection, post-transplant
228
Figure 13-78. Liver cell adenoma. The tumor is composed of thickened sheets of benign hepatocytes and blood vessels, without bile ducts. (Hematoxylin and eosin.)
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-79. Hepatocellular carcinoma. The typical microtrabecular hepatocellular carcinoma shows fragmentation on low magnification. (Hematoxylin and eosin.)
Figure 13-81. Preservation injury. In this example of preservation (cold ischemia/warm reperfusion) injury following liver transplantation, hepatocytes around the central vein (V) are ballooned. The nearby portal tracts show mild infiltrates of acute rejection. (Hematoxylin and eosin.)
CV
Figure 13-80. Fibrolamellar hepatocellular carcinoma. In this special variant of hepatocellular carcinoma there are plump, eosinophilic hepatocytes in islands surrounded by fibrous lamellae (arrows). (Hematoxylin and eosin.)
complications such as vascular thrombosis, infection, and sepsis, and recurrence of the original disease. More than one pathologic process may be evident in a given biopsy specimen. In the early post-transplant period (days to several weeks), functional cholestasis and centrilobular liver cell ballooning are manifestations of the cold ischemia/warm reperfusion damage seen in preservation injury (Figure 13-81). If the donor liver is steatotic and preservation injury results in necrosis, rupture of lipid vacuoles from the necrotic cells into perivenular sinusoids may result in lipopeliosis (Figure 13-82). Among rejection lesions, hyperacute rejection occurs only rarely and is associated with microvascular damage with periportal edema and necrosis in the first several days,138 and later hemorrhagic infarction of the liver. The most common type, acute rejection, is seen within the first few weeks after transplantation, or at any later time when effective immunosuppression is lowered, and is characterized by the histologic triad of portal inflammation, bile
Figure 13-82. Lipopeliosis. The sinusoids near the central vein (CV) contain lipid that has ruptured from necrotic hepatocytes following preservation injury. Note that many of the lipid vacuoles are engulfed by collections of Kupffer cells (arrows). (Hematoxylin and eosin.)
duct damage, and endotheliitis8 (the latter not present in all cases). The portal infiltrate consists of lymphocytes and plasma cells, with scattered neutrophils and often prominent eosinophils139 (Figure 1383). In endotheliitis (Figure 13-84) there is infiltration by inflammatory cells of the walls of veins in subendothelial and/or endothelial sites. Portal vein branches and, less frequently, central veins and/or sinusoids are involved. Chronic rejection is associated with progressive loss of interlobular bile ducts with or without obliterative arteriopathy.44 The progressive loss of intrahepatic bile ducts (vanishing bile duct syndrome) results in empty-seeming portal tracts with sparse lymphocytic infiltration and absent ducts (Figure 13-85). Vascular lesions of rejection arteriopathy are best recognized in sections of hilar vessels that demonstrate a characteristic accumulation of subintimal foam cells (Figure 13-86). The presence of arteriopathy can be inferred when perivenular hepatocyte ballooning, necrosis and fibrosis are seen.
229
Section II. Approach to the Patient with Liver Disease
Figure 13-83. Acute rejection following liver transplantation. The portal tract is infiltrated by lymphocytes and scattered neutrophils (seen within and around the bile duct, at top, which shows an irregular contour and variable nuclear morphology). (Hematoxylin and eosin.)
Figure 13-84. Endotheliitis in acute rejection following liver transplantation. A portal vein branch has numerous lymphocytes adhering to its endothelium. (Hematoxylin and eosin.)
Figure 13-85. Chronic rejection following liver transplantation. The portal tract shows sparse collections of lymphocytes but no bile duct (‘ductopenia’). (Hematoxylin and eosin.)
230
Figure 13-86. Rejection arteriopathy following liver transplantation. This hepatic artery branch was sampled from the liver hilum. The lumen is markedly narrowed by subintimal foam cells. (Hematoxylin and eosin.)
Figure 13-87. Recurrent hepatitis C virus infection following liver transplantation. Numerous apoptotic bodies (arrows) within the lobular parenchyma are one of the earliest signs of recurrent infection. (Hematoxylin and eosin.)
Recurrence of the original disease in the transplanted liver occurs in many of the major conditions for which transplantation is indicated, including hepatitis B and C, primary biliary cirrhosis, primary sclerosing cholangitis and autoimmune hepatitis.140 Recurrence may be diagnosed by observing characteristic histologic changes such as ground-glass cells in hepatitis B, lymphoid follicles in hepatitis C,141 and granulomatous bile duct destruction in primary biliary cirrhosis.142,143 In the first months of recurrent hepatitis C there are numerous apoptotic bodies144 (Figure 13-87), but light microscopy is not as sensitive as polymerase chain reaction for detection.145 The unusual pattern of fibrosing cholestatic hepatitis (FCH) may be seen in a subset of patients infected with either hepatitis B or C who have undergone liver, cardiac or renal transplantation.146–149 This lesion consists of parenchymal cholestasis accompanied by portal fibrosis with proliferating bile ductules associated with neutrophil infiltrate (Figure 13-88), clearly differing from either the changes of chronic viral hepatitis or typical acute rejection.
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
Figure 13-88. Fibrosing cholestatic hepatitis. This biopsy came from a hepatitis C-positive renal transplant recipient. There is parenchymal cholestasis (arrows) associated with portal fibrosis and a ductular reaction. (Hematoxylin and eosin.)
Figure 13-89. Cytomegalovirus infection following liver transplantation. A microabscess has formed and a hepatocyte at the edge of the microabscess shows an intranuclear viral inclusion (arrow). (Hematoxylin and eosin.)
Among the several possible intercurrent post-transplantation problems that may be seen in liver biopsies, cytomegalovirus infection is often suspected when liver dysfunction occurs in the appropriate time frame150 (within 1–3 months after transplantation) and the appropriate recipient (negative for antibodies to cytomegalovirus). The characteristic parenchymal microabscess (Figure 13-89) is virtually pathognomonic, even when intranuclear or intracytoplasmic viral inclusions are absent. Immunohistochemical staining for the virus is helpful to confirm the diagnosis.151 A small number of patients develop de novo autoimmune hepatitis following transplantation, with characteristic portal lymphoplasmacytic inflammation and interface hepatitis.152
REFERENCES 1. Menghini G. One-second needle biopsy of the liver. Gastroenterology 1958;35:190–199.
2. Clark JM, Diehl AM. Defining nonalcoholic fatty liver disease: implications for epidemiologic studies. Gastroenterology 2003;124:248–250. 3. Zafrani ES. Non-alcoholic fatty liver disease: an emerging pathological spectrum. Virchow’s Arch 2004;444:3–12. 4. Shackel NA, Gorrell MD, McCaughan GW. Gene array analysis and the liver. Hepatology 2002;36:1313–1325. 5. Rappaport AM. The microcirculatory acinar concept of normal and pathologic hepatic structure. Beitr Pathol 1976;157:215–243. 6. Reuben A. Now you see it, now you don’t. Hepatology 2003;38:781–784. 7. Demetris AJ, Ruppert K. Pathologist’s perspective on liver needle biopsy size? J Hepatol 2003;39:275–277. 8. International Panel. Banff schema for grading liver allograft rejection: an international consensus document. Hepatology 1997;25:658–663. 9. Crawford AR, Lin X-Z, Crawford JM. The normal adult human liver biopsy: a quantitative reference standard. Hepatology 1998;28:323–331. 10. Sherlock S. The syndrome of disappearing intrahepatic bile ducts. Lancet 1987;2:493–496. 11. Emerick KM, Rand EB, Goldmuntz E, et al. Features of Alagille syndrome in 92 patients: frequency and relation to prognosis. Hepatology 1999;29:822–829. 12. Bull LN, Carolton VEH, Stricker NL, et al. Genetic and morphological findings in progressive familial intrahepatic cholestasis (Byler Disease [PFIC-1] and Byler syndrome): evidence for heterogeneity. Hepatology 1997;26:155–164. 13. Boyer JL, Klatskin G. Pattern of necrosis in acute viral hepatitis. Prognostic value of bridging (subacute hepatic necrosis). N Engl J Med 1970;283:1063–1071. 14. Roskams T. Progenitor cell involvement in cirrhotic human liver diseases: from controversy to consensus. J Hepatol 2003;39: 431–434. 15. Fausto N. Liver regeneration and repair: hepatocytes, progenitor cells, and stem cells. Hepatology 2004;39:1477–1487. 16. Fogden E, Neuberger J. Alternative medicines and the liver. Liver Int 2003;23:213–220. 17. Teixeira MR Jr, Weller IVD, Murray AM, et al. The pathology of hepatitis A in man. Liver 1982;2:53–60. 18. Harrison TJ. Hepatitis E virus – an update. Liver 1999;19:171–176. 19. El-Refaie A, Savage K, Bhattacharya S, et al. HCV-associated hepatocellular carcinoma without cirrhosis. J Hepatol 1996;24:277–285. 20. Kumar D, Farrell GC, Fung C, George J. Hepatitis C virus genotype 3 is cytopathic to hepatocytes: reversal of hepatic steatosis after sustained therapeutic response. Hepatology 2002;36:1266–1272. 21. Misdraji J, Thiim M, Graeme-Cook FM. Autoimmune hepatitis with centrilobular necrosis. Am J Surg Pathol 2004;28:471–478. 22. Goodman ZD. Drug hepatotoxicity. Clin Liver Dis 2002;6:381–398. 23. Mullick FG, Ishak KG, Mahabir R, et al. Hepatic injury associated with paraquat toxicity in humans. Liver 1981;1:209–221. 24. Chojkier M. Hepatic sinusoidal-obstruction syndrome: toxicity of pyrrolizidine alkaloids. J Hepatol 2003;39:437–446. 25. DeLeve LD, Shulman HM, McDonald GB. Toxic injury to hepatic sinusoids: sinusoidal obstruction syndrome (venoocclusive disease). Semin Liver Dis 2002;22:27–42. 26. Fogarty International Center Criteria Committee. Diseases of the liver and biliary tract. Standardization of nomenclature, diagnostic criteria, and diagnostic methodology. Washington DC: US Government Printing Office, 1976: 9–11. 27. Alvarez F, Berg PA, Biandin FB, et al. International Autoimmune Hepatitis Group report: review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999;31:929–938.
231
Section II. Approach to the Patient with Liver Disease
28. Czaja AJ, Carpenter HA. Sensitivity, specificity, and predictability of biopsy interpretations in chronic hepatitis. Gastroenterology 1993;105:1824–1832. 29. Brunt EM. Grading and staging the histopathological lesions of chronic hepatitis: the Knodell histology activity index and beyond. Hepatology 2000;31:241–246. 30. Knodell RG, Ishak KG, Black WC, et al. Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology 1981;1:431–435. 31. Scheuer PJ. Assessment of liver biopsies in chronic hepatitis: how is it best done? J Hepatol 2003;38:240–242. 32. Ishak K, Baptista A, Bianchi L, et al. Histological grading and staging of chronic hepatitis. J Hepatol 1995;22:696–699. 33. Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. Hepatology 1996;24:289–293. 34. Scheuer PJ. Classification of chronic viral hepatitis: a need for reassessment. J Hepatol 1991;13:372–374. 35. Lefkowitch JH. Bile ductular cholestasis: an ominous histopathologic sign related to sepsis and ‘cholangitis lenta.’ Hum Pathol 1982;13:19–24. 36. Christoffersen P, Poulsen H. Histological changes in human liver biopsies following extrahepatic biliary obstruction. Acta Pathol Microbiol Scand 1970;212:150–157. 37. Roskams T, Desmet V. Ductular reaction and its diagnostic significance. Semin Diagn Pathol 1998;15:259–269. 38. Crawford JM, Boyer JL. Clinicopathology conferences: inflammation-induced cholestasis. Hepatology 1998;28:253–260. 39. Scheuer PJ. Pathologic features and evolution of primary biliary cirrhosis and primary sclerosing cholangitis. Mayo Clin Proc 1998;73:179–183. 40. Müller T, Langner C, Fuchsbichler A, et al. Immunohistochemical analysis of Mallory bodies in Wilsonian and non-Wilsonian hepatic copper toxicosis. Hepatology 2004;39:963–969. 41. Keaveny AP, Gordon FD, Goldar-Najafi A, et al. Native liver xanthogranulomatous cholangiopathy in primary sclerosing cholangitis: impact on posttransplant outcome. Liver Transplant 2004;10:115–122. 42. O’Donohue J, Oien KA, Donaldson P, et al. Co-amoxiclav jaundice: clinical and histological features and HLA class II association. Gut 2000;47:717–720. 43. Ludwig J. Idiopathic adulthood ductopenia: an update. Mayo Clin Proc 1998;73:285–291. 44. International Panel. Update of the international Banff schema for liver allograft rejection: working recommendations for the histopathologic staging and reporting of chronic rejection. Hepatology 2000;31:792–799. 45. Rudzki C, Ishak KG, Zimmerman HJ. Chronic intrahepatic cholestasis of sarcoidosis. Am J Med 1975;59:373–387. 46. Li L, Krantz ID, Deng Y, et al. Alagille syndrome is caused by mutations in human Jagged 1, which encodes a ligand for Notch 1. Nature Genet 1997;16:243–251. 47. Jansen PLM, Sturm E. Genetic cholestasis, causes and consequences for hepatobiliary transport. Liver Int 2003;23:315–322. 48. Turkish A, Levy J, Kato M, et al. Pancreatitis and paraneoplastic cholestasis as presenting manifestations of pancreatic lymphoma: evaluation, management, and clinical course in a child. J Pediatr Gastroenterol Nutr 2004;39:552–556. 49. Hoyumpa AM, Greene HL, Dunn GD, et al. Fatty liver: biochemical and clinical considerations. Dig Dis 1975;20: 1142–1170. 50. Ishak KG, Zimmerman HJ, Ray MB. Alcoholic liver disease: pathologic, pathogenetic and clinical aspects. Alcohol Clin Exp Res 1991;15:45–66. 51. Marchesini G, Bugianesi E, Forlani G, et al. Nonalcoholic fatty liver, steatohepatitis, and the metabolic syndrome. Hepatology 2003;37:917–923.
232
52. Spengler U, Lichterfeld M, Rockstroh JK. Antiretroviral drug toxicity – a challenge for the hepatologist? J Hepatol 2002;36:283–294. 53. Badizadegan K, Perez-Atayde AR. Focal glycogenosis of the liver in disorders of ureagenesis: its occurrence and diagnostic significance. Hepatology 1997;26:365–373. 54. Benjaminov FS, Heathcote J. Liver disease in pregnancy. Am J Gastroenterol 2004;99:2479–2488. 55. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med 2002;346:1221–1231. 56. Baptista A, Bianchi L, De Groote J, et al. Alcoholic liver disease: morphological manifestations. Lancet 1981;1:707–711. 57. James OFW, Day CP. Non-alcoholic steatohepatitis (NASH): a disease of emerging identity and importance. J Hepatol 1998;29:495–501. 58. Diehl AM, Goodman Z, Ishak KG. Alcohol-like liver disease in nonalcoholics. A clinical and histologic comparison with alcoholinduced liver injury. Gastroenterology 1988;95:1056–1062. 59. Brunt EM, Janney CG, Di Bisceglie AM, et al. Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastroenterol 1999;94:2467–2474. 60. Kemeny MM, Battifora H, Blayney DW, et al. Sclerosing cholangitis after continuous hepatic artery infusion of FUDR. Ann Surg 1985;202:176–181. 61. Oh C-K, Pelletier SJ, Sawyer RG, et al. Uni- and multi-variate analysis of risk factors for early and late hepatic artery thrombosis after liver transplantation. Transplantation 2001;71:767–772. 62. Gocke DJ, Hsu K, Morgan C, et al. Vasculitis in association with Australia antigen. J Exp Med 1971;134 Suppl:330s–336s. 63. Lefkowitch JH, Mendez L. Morphologic features of hepatic injury in cardiac disease and shock. J Hepatol 1986;2:313–327. 64. Amitrano L, Guardascione MA, Brancaccio V, et al. Risk factors and clinical presentation of portal vein thrombosis in patients with liver cirrhosis. J Hepatol 2004;40:736–741. 65. Menon KVN, Shah V, Kamath PS. The Budd–Chiari syndrome. N Engl J Med 2004;350:578–585. 66. Cazals-Hatem D, Vilgrain V, Genin P, et al. Arterial and portal circulation and parenchymal changes in Budd–Chiari syndrome: a study in 17 explanted livers. Hepatology 2003;37:510–519. 67. Langlet P, Escolano S, Valla D, et al. Clinicopathological forms and prognostic index in Budd–Chiari syndrome. J Hepatol 2003;39:496–501. 68. Valla D-C. The diagnosis and management of the Budd–Chiari syndrome: consensus and controversies. Hepatology 2003;38:793–803. 69. Marubbio AT, Danielson B. Hepatic veno-occlusive disease in a renal transplant patient receiving azathioprine. Gastroenterology 1975;69:739–743. 70. Kishi M, Maeyama S, Iwaba A, et al. Hepatic veno-occlusive lesions and other histopathological changes of the liver in severe alcoholic hepatitis – a comparative clinicohistopathological study of autopsy cases. Alcohol Clin Exp Res 2000;24:74S–80S. 71. Khettry U, Backer A, Ayata G, et al. Centrilobular histopathologic changes in liver transplant biopsies. Hum Pathol 2002;33:270–276. 72. Neil DA, Hubscher SG. Histologic and biochemical changes during the evolution of chronic rejection of liver allografts. Hepatology 2002;35:639–651. 73. Demetris AJ. Central venulitis in liver allografts: considerations of differential diagnosis. Hepatology 2001;33:1329–1330. 74. Gerber MA, Thung SN, Bodenheimer HC Jr, et al. Characteristic histologic triad in liver adjacent to metastatic neoplasm. Liver 1986;6:85–88. 75. Winkler K, Poulsen H. Liver disease with periportal sinusoidal dilatation. A possible complication to contraceptive steroids. Scand J Gastroenterol 1975;10:699–704.
Chapter 13 LIVER BIOPSY HISTOPATHOLOGY
76. Czapar CA, Weldon-Linne CM, Moore DM, et al. Peliosis hepatis in the acquired immunodeficiency syndrome. Arch Pathol Lab Med 1986;110:611–613. 77. Fine KD, Solano M, Polter DE, et al. Malignant histiocytosis in a patient presenting with hepatic dysfunction and peliosis hepatis. Am J Gastroenterol 1995;90:485–488. 78. Scheuer PJ, Schachter LA, Mathur S, et al. Peliosis hepatis after liver transplantation. J Clin Pathol 1990;43:1036–1037. 79. Wanless IR, Mawdsley C, Adams R. On the pathogenesis of focal nodular hyperplasia of the liver. Hepatology 1985;5:1194–1200. 80. Larson AM. Liver disease in hereditary hemorrhagic telangiectasia. J Clin Gastroenterol 2003;36:149–158. 81. Ishak KG, Sharp HL. Metabolic errors and liver disease. In: MacSween RNM, Anthony PP, Scheuer PJ, et al., eds. Pathology of the liver, 3rd edn. Edinburgh: Churchill Livingstone, 1994: 123. 82. Deutsch J, Becker H, Auböck L. Histopathological features of liver disease in alpha-1-antitrypsin deficiency. Acta Paediatr 1994;393(Suppl):8–12. 83. Carrell RW, Lomas DA. Alpha-1-antitrypsin deficiency – a model for conformational diseases. N Engl J Med 2002;346:45–53. 84. Perlmutter DH. Alpha-1-antitrypsin deficiency. Semin Liver Dis 1998;18:217–225. 85. Pietrangelo A. Hereditary hemochromatosis – a new look at an old disease. N Engl J Med 2004;350:2383–2397. 86. Deugnier YM, Loréal O, Turlin B, et al. Liver pathology in genetic hemochromatosis: a review of 135 homozygous cases and their bioclinical correlations. Gastroenterology 1992;102: 2050–2059. 87. Tavill AS. Hemochromatosis. In: Schiff L, Schiff ER, eds. Diseases of the liver, 7th edn. Philadelphia: JB Lippincott, 1993: 669. 88. Fellows IW, Stewart M, Jeffcoate WJ, et al. Hepatocellular carcinoma in primary haemochromatosis in the absence of cirrhosis. Gut 1988;29:1603–1606. 89. Ludwig J, Hashimoto E, Porayko MK, et al. Hemosiderosis in cirrhosis: a study of 447 native livers. Gastroenterology 1997;112:882–888. 90. Kowdley KV, Trainer TD, Saltzman JR, et al. Utility of hepatic iron index in American patients with hereditary hemochromatosis: a multicenter study. Gastroenterology 1997;113:1270–1277. 91. Tao TY, Gitlin JD. Hepatic copper metabolism: insights from genetic disease. Hepatology 2003;37:1241–1247. 92. Sternlieb I. Fraternal concordance of types of abnormal hepatocellular mitochondria in Wilson’s disease. Hepatology 1992;16:728–732. 93. Scheinberg IH, Sternlieb I. Wilson’s disease. Major problems in internal medicine XXIII. Philadelphia: WB Saunders, 1984. 94. Roberts EA, Schilsky ML. A practice guideline on Wilson disease. Hepatology 2003;37:1475–1492. 95. Ludwig J, Moyer TP, Rakela J. The liver biopsy diagnosis of Wilson’s disease. Methods in pathology. Am J Clin Pathol 1994;102:443–446. 96. Anthony PP, Ishak KG, Nayak NC, et al. The morphology of cirrhosis: definition, nomenclature, and classification. Bull WHO 1977;55:521–540. 97. Schinoni MI, Andrade Z, De Freitas LAR, et al. Incomplete septal cirrhosis: an enigmatic disease. Liver Int 2004;24: 452–456. 98. Desmet VJ, Roskams T. Cirrhosis reversal: a duel between dogma and myth. J Hepatol 2004;40:860–867. 99. Anthony PP, Vogel CL, Barker LF. Liver cell dysplasia: a premalignant condition. J Clin Pathol 1973;26:217–223. 100. Mion F, Grozel L, Boillot O, et al. Adult cirrhotic liver explants: precancerous lesions and undetected small hepatocellular carcinomas. Gastroenterology 1996;111:1587–1592.
101. Sandor M, Weinstock JV, Wynn TA. Granulomas in schistosome and mycobacterial infections: a model of local immune responses. Trends Immunol 2003;24:44–52. 102. Klatskin G. Hepatic granulomata: problems in interpretation. Mt Sinai J Med 1977;44:798–812. 103. Lefkowitch JH. Hepatic granulomas. J Hepatol 1999;30:40–45. 104. Ishak KG: Granulomas of the liver. In: Ioachim HL. Pathology of granulomas. New York: Raven Press, 1983:307. 105. Sartin JS, Walker RC. Granulomatous hepatitis: a retrospective review of 88 cases at the Mayo Clinic. Mayo Clin Proc 1991;66:914–918. 106. Ferrell LD. Hepatic granulomas: a morphologic approach to diagnosis. Surg Pathol 1990;3:87–106. 107. Devaney K, Goodman ZD, Epstein MS, et al. Hepatic sarcoidosis. Clinicopathologic features in 100 patients. Am J Surg Pathol 1993;17:1272–1280. 108. Gaya DR, Thorburn KA, Oien KA, et al. Hepatic granulomas: a 10 year single centre experience. J Clin Pathol 2003;56: 850–853. 109. Simon HB, Wolff SM. Granulomatous hepatitis and prolonged fever of unknown origin: a study of 13 patients. Medicine 1973;52:1–21. 110. Knox TA, Kaplan MM, Gelfand JA, et al. Methotrexate treatment of idiopathic granulomatous hepatitis. Ann Intern Med 1995;122:592–595. 111. Dimashkieh HH, Mo JQ, Wyatt-Ashmead J, et al. Pediatric hepatic angiosarcoma: case report and review of the literature. Pediatr Dev Pathol 2004;7:527–532. 112. Okuda K, Peters RL (eds). Hepatocellular carcinoma. New York: John Wiley & Sons, 1976. 113. Okuda K, Ishak KG (eds). Neoplasms of the liver. New York: Springer-Verlag, 1987. 114. Stocker JT, Ishak KG. Focal nodular hyperplasia of the liver: a study of 21 pediatric cases. Cancer 1981;48:336–345. 115. Desmet VJ. Current problems in diagnosis of biliary disease and cholestasis. Semin Liver Dis 1986;6:233–245. 116. Wanless IR, Lentz JS, Roberts EA. Partial nodular transformation of liver in an adult with persistent ductus venosus. Review with hypothesis on pathogenesis. Arch Pathol Lab Med 1985;109: 427–432. 117. Stromeyer FW, Ishak KG. Nodular transformation (nodular ‘regenerative’ hyperplasia) of the liver. A clinicopathologic study of 30 cases. Hum Pathol 1981;12:60–71. 118. Steiner PE. Nodular regenerative hyperplasia of the liver. Am J Pathol 1959;35:943–953. 119. Wanless IR, Godwin TA, Allen F, et al. Nodular regenerative hyperplasia of the liver in hematologic disorders: a possible response to obliterative portal venopathy. A morphometric study of nine cases with an hypothesis on the pathogenesis. Medicine 1980;59:367–379. 120. Tung GA, Cronan JJ. Percutaneous needle biopsy of hepatic cavernous hemangioma. J Clin Gastroenterol 1993;16: 117–122. 121. Govindarajan S, Peters RL. The bile duct adenoma. A lesion distinct from Meyenburg complex. Arch Pathol Lab Med 1984;108:922–924. 122. Thommesen N. Biliary hamartomas (von Meyenburg complexes) in liver needle biopsies. Acta Pathol Microbiol Scand A 1978;86:93–99. 123. Jørgensen MJ. The ductal plate malformation. [Review]. Acta Pathol Microbiol Scand 1977;(Suppl 257):1–87. 124. Desmet VJ. Congenital diseases of intrahepatic bile ducts: variations on the theme ‘ductal plate malformation.’ Hepatology 1992;16:1069–1083. 125. Jörgensen MJ. The ductal plate malformation: a study of the intrahepatic bile-duct lesion in infantile polycystic disease and congenital hepatic fibrosis. Acta Pathol Microbiol Scand 1977;257(Suppl):1–88.
233
Section II. Approach to the Patient with Liver Disease
126. Allaire GS, Rabin L, Ishak KG, et al. Bile duct adenoma. A study of 152 cases. Am J Surg Pathol 1988;12:708–715. 127. Bhathal PS, Hughes NR, Goodman ZD. The so-called bile duct adenoma is a peribiliary gland hamartoma. Am J Surg Pathol 1996;20:858–864. 128. Nguyen B, Fléjou J-F, Terris B, et al. Focal nodular hyperplasia of the liver. A comprehensive pathology study of 305 lesions and recognition of new histologic forms. Am J Surg Pathol 1999;23:1441–1454. 129. Haber M, Reuben A, Burrell M, et al. Multiple focal nodular hyperplasia of the liver associated with hemihypertrophy and vascular malformations. Gastroenterology 1995;108:1256–1262. 130. Goodman ZD. Benign tumors of the liver. In: Okuda K, Ishak KG. Neoplasms of the liver. Tokyo: Springer-Verlag, 1990: 105. 131. Zucman-Rossi J. Genetic alterations in hepatocellular adenomas: recent findings and new challenges. J Hepatol 2004;40:1036–1039. 132. Bacq Y, Jacquemin E, Balabaud C, et al. Familial liver adenomatosis associated with hepatocyte nuclear factor 1a?inactivation. Gastroenterology 2003;125:1470–1475. 133. International Working Party. Terminology of nodular hepatocellular lesions. Hepatology 1995;22:983–993. 134. Tao L-C. Are oral contraceptive-associated liver cell adenomas premalignant? Acta Cytol 1992;36:338–344. 135. Berman MM, Libbey NP, Foster JH. Hepatocellular carcinoma. Polygonal cell type with fibrous stroma – an atypical variant with a favorable prognosis. Cancer 1980;46:1448–1455. 136. Craig JR, Peters RL, Edmondson HA, et al. Fibrolamellar carcinoma of the liver: a tumor of adolescents and young adults with distinctive clinico-pathologic features. Cancer 1980;46:372–379. 137. El-Serag HB, Davila JA. Is fibrolamellar carcinoma different from hepatocellular carcinoma? A US population-based study. Hepatology 2004;39:798–803. 138. Haga H, Egawa H, Shirase T, et al. Periportal edema and necrosis as diagnostic histological features of early humoral rejection in ABO-incompatible liver transplantation. Liver Transplant 2004;10:16–27. 139. Nagral A, Ben-Ari Z, Dhillon AP, et al. Eosinophils in acute cellular rejection in liver allografts. Liver Transplant Surg 1998;4:355–362.
234
140. Davern TJ, Lake JR. Recurrent disease after liver transplantation. Semin Gastrointest Dis 1998;9:86–109. 141. Ferrell LD, Wright TL, Roberts J, et al. Hepatitis C viral infection in liver transplant recipients. Hepatology 1992;16:865–876. 142. Hübscher SG, Elias E, Buckels JAC, et al. Primary biliary cirrhosis. Histological evidence of disease recurrence after liver transplantation. J Hepatol 1993;18:173–184. 143. Balan V, Batts KP, Porayko MK, et al. Histological evidence for recurrence of primary biliary cirrhosis after liver transplantation. Hepatology 1993;18:1392–1398. 144. Saxena R, Crawford JM, Navarro VJ, et al. Utilization of acidophil bodies in the diagnosis of recurrent hepatitis C infection after orthotopic liver transplantation. Mod Pathol 2002;15:897–903. 145. Hourigan LF, MacDonald GA, Purdie D, et al. Fibrosis in chronic hepatitis C correlates significantly with body mass index and steatosis. Hepatology 1999;29:1215–1219. 146. Davies SE, Portmann BC, O’Grady JG, et al. Hepatic histological findings after transplantation for chronic hepatitis B virus infection, including a unique pattern of fibrosing cholestatic hepatitis. Hepatology 1991;13:150–157. 147. Lim HL, Lau GKK, Davis GL, et al. Cholestatic hepatitis leading to hepatic failure in a patient with organ-transmitted hepatitis C virus infection. Gastroenterology 1994;106:248–251. 148. Delladetsima JK, Boletis JN, Makris F, et al. Fibrosing cholestatic hepatitis in renal transplant recipients with hepatitis C virus infection. Liver Transplant Surg 1999;5:294–300. 149. Case records of the Massachusetts General Hospital. Case 22000. N Engl J Med 2000;342:192–199. 150. Seehofer D, Rayes N, Tullius SG, et al. CMV hepatitis after liver transplantation: incidence, clinical course, and long-term follow-up. Liver Transplant 2002;8:1138–1146. 151. Theise ND, Conn M, Thung SN. Localization of cytomegalovirus antigens in liver allografts over time. Hum Pathol 1993;24: 103–108. 152. Mieli-Vergani G, Vergani D. De novo autoimmune hepatitis after liver transplantation. J Hepatol 2004;40:3–7.
Section II: Approach to the Patient with Liver Disease
14
LABORATORY TESTING FOR LIVER DISEASE Thierry Poynard and Françoise Imbert-Bismut Abbreviations AIDS acquired immunodeficiency syndrome bDNA AIH autoimmune hepatitis BSP ALP alkaline phosphatase ELISA ALT alanine aminotransferase GGT AMA antimitochondrial antibodies HBsAg ANA antinuclear antibodies HCV anti-LC1 liver specific cytosol antigen type 1 HDL anti-LKM1 antibodies to liver/kidney microsome type 1 ICG antisoluble liver antigen/liver pancreas IL SLA/LP INR LCAT AST aspartate aminotransferase AUROC area under the receiver characteristics curve LDH
branched DNA bromsulphalein enzyme-linked immunosorbent assay g-glutamyl transferase hepatitis B surface antigen hepatitis C virus high-density lipoprotein indocyanine green interleukin International Normalized Ratio lecithin-cholesterol acyltransterase lactate dehydrogenase
INTRODUCTION Because the liver performs multiple functions, no single laboratory test or battery of tests is sufficient to provide a complete estimate of the liver injury in every clinical situation. A broad array of biochemical tests is used to evaluate patients with suspected or established liver disease, but also for screening asymptomatic individuals. Abnormal elevations of serum liver chemistries may occur in 1–4% of the asymptomatic population.1,2 As with the evaluation of all medical tests, liver tests must be interpreted in the context of the subject’s risk factors for disease, symptoms, and historical and physical examination findings. There is a need to replace invasive tests such as liver biopsy by non-invasive biomarkers.3 In approaching patients with liver disease, it is helpful to classify laboratory tests into several broad categories. These include (1) the common serum liver chemistry tests that reflect hepatocellular damage, cholestasis or synthetic function; (2) tests that reflect fibrosis, such as hyaluronic acid and biochemical panels; (3) tests that contribute to accurate diagnosis in liver disease, including specific autoantibodies, genetic tests, and serologic tests for viral hepatitis; and (4) less common tests that measure the capacity of the liver to transport organic anions and clear endogenous or exogenous substances from the circulation, or that measure the capacity of the liver to metabolize drugs.
COMMON SERUM LIVER CHEMISTRY TESTS Five laboratory assays are commonly called liver function tests (LFT), although they are neither specific to the liver nor true meas-
LFT NAD pANCA PBC PCR PDH SGOT SGPT SMA TNF
liver function test nicotinamide adenine dinucleotide perinuclear antineutrophil cytoplasmic antibodies primary biliary cirrhosis polymerase chain reaction pyruvate dehydrogenase serum glutamic oxaloacetic transaminase serum glutamic pyruvic transaminase smooth muscle antibodies tumor necrosis factor
ures of liver function: alanine aminotransferase (ALT or SGPT), aspartate aminotransferase (AST or SGOT), g-glutamyl transferase (GGT), alkaline phosphatase (ALP) and bilirubin. LFTs are used to screen people for the presence of liver disease, suggest the underlying cause, estimate the severity, assess prognosis, and monitor the efficacy of therapy. Abnormal LFTs may be the first indication of subclinical liver disease and may thereby guide further diagnostic evaluation.2 After the existence of hepatic dysfunction is recognized, the specific pattern of liver test abnormalities may suggest the category of the underlying disease, such as hepatitis, biliary obstruction, or infiltrative liver disease (Table 141). Additional testing may permit an assessment of the severity of liver dysfunction, and certain tests, especially when performed repeatedly, may provide an estimate of prognosis. Prothrombin time and serum albumin are the most common tests used as prognostic factors, together with serum bilirubin. For example, a rising bilirubin level is an adverse prognostic feature of primary biliary cirrhosis. Prolongation of the prothrombin time correlates with an adverse prognosis in acute hepatitis. Finally, sequential LFTs may be useful in assessing response to therapy, as in the case of declining serum aminotransferase levels after therapy for chronic viral hepatitis. There are limitations to the use of LFTs in the diagnosis and management of liver disease. There are limitations in sensitivity, that is, the likelihood of an abnormal test result in patients known to have a disease. For example, patients with cirrhosis may have normal or minimally abnormal LFTs. Serum albumin, total bilirubin and GGT decrease and ALT slightly increases during a normal pregnancy.4 There are also limitations in specificity, that is, the likelihood of a normal test result in a patient without the disease. Elevated serum aminotransferase levels may indicate liver disease or cardiac or muscle disease. LFT abnormalities rarely provide a specific diagnosis, but rather suggest a general category of liver disorder. Often,
235
Section II. Approach to the Patient with Liver Disease
Table 14-1. Common Serum Liver Chemistry Tests
Table 14-2. Etiology of ALT or AST Elevations
Liver chemistry test
Clinical implication of abnormality
Mild ALT or AST elevations, <200 IU/l
Alanine aminotransferase Aspartate aminotransferase Bilirubin
Hepatocellular damage Hepatocellular damage Cholestasis, biliary obstruction, or impaired conjugation Cholestasis, biliary obstruction, infiltrative disease Cholestasis, biliary obstruction, infiltrative disease Synthetic function Synthetic function Cholestasis or biliary obstruction
g-Glutamyltransferase Alkaline phosphatase Prothrombin time Albumin Bile acids
further specific diagnostic testing is required. In general, the multiple functions of the liver combined with the large number of LFTs available preclude simple diagnostic algorithms that are applicable to the evaluation of all patients with suspected liver disease. Therefore it is important for the physician evaluating a patient with suspected or known liver disease to have a firm understanding of the diverse tests available, their indications and their limitations.
AMINOTRANSFERASES The aminotransferases, aspartate aminotransferase (AST, formerly serum glutamic oxaloacetic transaminase) and alanine aminotransferase (ALT, formerly serum glutamic pyruvic transaminase), are the most frequently used indicators of hepatic injury and represent markers of hepatocellular necrosis. These enzymes catalyze the transfer of the a-amino groups of aspartate and alanine to the a-keto group of ketoglutaric acid, resulting in the formation of oxaloacetic acid and pyruvic acid, respectively. The enzymes play a role in gluconeogenesis by facilitating the synthesis of glucose from non-carbohydrate sources. AST is present in both the mitochondria (80% of the total) and the cytosol (20%) of hepatocytes, but ALT is found only in the cytosol.5,6 Whereas ALT is localized primarily to the liver, AST is present in a variety of tissues, including liver, heart, skeletal muscle, kidney, brain, pancreas, lungs, leukocytes and erythrocytes. Serum AST levels are typically elevated in cardiac and muscle diseases. Both AST and ALT are present in the serum as holoenzymes and apoenzymes, depending on whether or not they are associated with their cofactor pyridoxal 5¢-phosphate. Thus potentially six different forms of AST and ALT can circulate in the serum: the apo- and holoenzymes of ALT, cytosolic AST, and mitochondrial AST. On rare occasions isolated elevations of AST in serum may be attributable to the formation of a macroenzyme with binding of AST to immunoglobulins.7,8 Numerous methods for assaying AST and ALT have been developed, and the normal range varies widely among laboratories and should be adjusted for sex and body mass index.9 In the absence of an international calibration, the expression of results in international units is recommended.10 The most sensitive and specific method for assaying the aminotransferases is enzymatic reduction of oxaloacetate and pyruvate to malate and lactate, respectively, coupled with oxidation of nicotinamide adenine dinucleotide, reduced form (NADH), to nicotinamide adenine dinucleotide (NAD). Because
236
Hepatic: ALT-predominant Chronic hepatitis C Chronic hepatitis B Acute viral hepatitis (A–E, EBV, CMV) Steatosis/steatohepatitis Hemochromatosis Medications/toxins/illicit drugs/herbs Autoimmune hepatitis Sclerosing cholangitis Primary biliary cirrhosis Acute bile duct obstruction a1-Antitrypsin deficiency Wilson’s disease Celiac disease Hepatic: AST-predominant Alcohol-related liver injury Steatosis/steatohepatitis Cirrhosis
Severe ALT or AST elevations, >600 IU/l Acute viral hepatitis (A–E, herpes) Autoimmune hepatitis Wilson’s disease Acute bile duct obstruction Acute Budd–Chiari syndrome Hepatic artery ligation
Medications/toxins/illicit drugs/herbs Ischemic hepatitis
Non-hepatic Hemolysis Myopathy Thyroid disease Strenuous exercise Sepsis Macro-AST
NADH absorbs light at 340 nm but NAD does not, the process can be followed spectrophotometrically. Serum levels of AST and ALT are elevated to some extent in almost all liver diseases. The highest elevations occur in severe viral hepatitis, drug- or toxin-induced hepatic necrosis, and circulatory shock (ischemic hepatitis) (Table 14-2).11,12 Although enzyme levels may reflect the extent of hepatocellular necrosis, they do not correlate with eventual outcome. In fact, declining AST and ALT levels may indicate either recovery or a poor prognosis, because of a paucity of remaining hepatocytes in the latter case. Moderately elevated levels of serum aminotransferase are typical of acute or chronic hepatitis, including viral hepatitis and autoimmune, druginduced and alcoholic hepatitis, whereas mild elevations are seen in fatty liver, non-alcoholic steatohepatitis, drug toxicity and chronic hepatitis C.13–21 In overweight people, mildly elevated levels of serum aminotransferase as a result of fatty liver may resolve with weight reduction of 10% or more.22,23 It appears in some instances that mild elevations in the serum ALT (and the AST) level may result from myositis24 or muscle injury after rigorous exercise.25 In cirrhosis, cholestatic liver diseases, and hepatic neoplasms, serum levels of AST and ALT may be elevated slightly. Acute biliary tract obstruction may result in elevations of AST and ALT of more than 300 U/l; these levels peak early and decline rapidly over 24–72 hours despite unresolved obstruction.26 Determinations of serum aminotransferase have proved useful as screening tests for subclinical liver disease in asymptomatic persons,27,28 and an abnormal result may lead to a diagnosis of
Chapter 14 LABORATORY TESTING FOR LIVER DISEASE
hemochromatosis,29 Wilson’s disease, or a1-antitrypsin deficiency and non-hepatic diseases such as celiac disease,30 Addison’s disease, and anorexia nervosa.33 Elevated serum aminotransferase levels have been found in as many as 54% of children31 and 40% of adult patients with celiac disease,32 but these prevalences were not observed in other studies.32–34 Liver biopsy in these patients may show variable degrees of steatosis, inflammation, and fibrosis.32 On the other hand, in one-third to one-half of otherwise healthy individuals with an isolated elevation of serum ALT, normal values are found when testing is repeated.18,35 The ratio of AST to ALT in serum may be helpful diagnostically. An AST:ALT ratio of more than 2.0 is characteristically observed in alcoholic liver disease.36 In this setting the ALT is often normal or minimally elevated. High ALT levels (>500 U/l) should suggest a diagnosis other than alcoholic liver disease, even when the AST:ALT ratio is greater than 2.0.37 In viral hepatitis the ratio of AST to ALT is typically less than 1.0. The ratio often rises (but not invariably38,39) to more than 1.0 as cirrhosis develops.40,41 The ratio of AST to ALT in patients with non-alcoholic fatty liver disease is typically less than 1.0 in the absence of fibrosis on liver biopsy; however, as in chronic viral hepatitis, the ratio exceeds 1.0 in some patients with cirrhosis.42 The lack of a substantial elevation in ALT levels in patients with alcoholic liver disease is thought to be related to pyridoxine deficiency. Pyridoxine deficiency decreases hepatic ALT more than hepatic AST, with corresponding changes in serum levels.43 The ratio of mitochondrial AST to total AST may be useful in the diagnostic of specific liver diseases, but the isoenzyme activity is not assayed routinely in clinical practice.44 Under certain circumstances the aminotransferase levels may be falsely diminished or inhibited. Markedly diminished serum levels of AST have been reported in patients on long-term hemodialysis and have been attributed either to dialysis of the enzyme or to pyridoxine deficiency, although neither possibility has been confirmed experimentally.45
ALKALINE PHOSPHATASE Alkaline phosphatase is a family of isoenzymes that catalyze the hydrolysis of a number of phosphate esters at an alkaline pH optimum. All enzymes in the family are glycoproteins that require zinc for activity. Alkaline phosphatases are coded for by four genes,46 including one that codes for alkaline phosphatase isoenzymes from liver, bone, first-trimester placenta, and kidney. Levels of alkaline phosphatase determined by different assay methods must be compared with caution because of the specific conditions of the different assays, the heterogeneity of the alkaline phosphatases, and their overlapping but distinctly different specific activities with different substrates. Although alkaline phosphatases are ubiquitous, their physiologic significance remains unclear. In the human body alkaline phosphatase has been identified in liver, bone, intestine, placenta, kidney, and leukocytes.47 In the liver two distinct forms of alkaline phosphatase are found. The enzyme in the liver is associated with sinusoidal and canalicular membranes, also present in the cytosol, and secreted in bile in large amounts.48 In healthy people most circulating alkaline phosphatase originates from liver or bone. In pregnant women, circulating placental alkaline phosphatase is often found. Average normal values of alkaline phosphatase vary with age: they are relatively high in childhood and
puberty (correlating with the rate of growth of bone), lower in middle age (higher in men than in women), and higher again in old age (particularly in women). Serum levels of alkaline phosphatase correlate with a person’s weight and the number of cigarettes smoked per day, and inversely with height.48–50 In patients with an elevated level of serum alkaline phosphatase, the source is the liver in a majority of cases; but in up to one-third of such individuals no evidence of liver disease can be found. Not uncommonly, isolated mildly elevated levels of alkaline phosphatase in otherwise normal persons return to normal on followup.50 Even in hospitalized patients non-specific elevations of serum alkaline phosphatase are not uncommon and are often transient.51 Bone disease characterized by increased osteoblastic activity also may be the source of an elevated serum alkaline phosphatase level, as may pregnancy. Only rarely is the intestine or kidney the source. The highest elevations of serum alkaline phosphatase in patients with liver disease occur in cholestatic disorders (Table 14-3). Elevations occur as a result of both intrahepatic and extrahepatic obstruction to bile flow, and the degree of elevation does not help to distinguish the two. The mechanism by which cholestasis increases serum alkaline phosphatase levels is thought to involve the induction of alkaline phosphatase synthesis secondary to enhanced translation of the mRNA of alkaline phosphatase, not from a failure to clear or excrete circulating alkaline phosphatase.52–54 That the passage of time is required for de novo synthesis of alkaline phosphatase after biliary obstruction is reflected in instances early in the course of acute suppurative cholangitis, in which serum alkaline phosphatase levels are normal but serum aminotransferase levels are markedly elevated.55 The serum alkaline phosphatase level may be normal in up to 3% of
Table 14-3. Causes of Elevated Serum Alkaline Phosphatase and g-Glutamyltransferase Causes of elevated serum alkaline phosphatase and g-glutamyltransferase
Elevated alkaline phosphatase only
Hepatobiliary Bile duct obstruction Primary biliary cirrhosis Primary sclerosing cholangitis Medications Hepatocellular carcinoma Hepatic metastasis Hepatitis Cirrhosis Vanishing bile duct syndromes Benign recurrent cholestasis Infiltrating diseases of the liver Sarcoidosis Tuberculosis Fungal infection Other granulomatous diseases Amyloidosis Lymphoma Non-hepatic Chronic renal failure Lymphoma and other malignancies Congestive heart failure Infection/inflammation
Bone disease Pregnancy Childhood growth
237
Section II. Approach to the Patient with Liver Disease
patients with primary sclerosing cholangitis, 56 probably less sensitive than GGT. Elevated levels of serum alkaline phosphatase in patients with cancer may result from hepatic or bony metastases. In hepatic metastasis elevation of the alkaline phosphatase in serum presumably results from localized biliary obstruction with induction of alkaline phosphatase and leakage into the serum. Elevated serum levels of alkaline phosphatase of hepatic origin also may result from other infiltrative liver diseases, such as abscesses, granulomatous liver disease, and amyloidosis. In hospitalized patients the highest elevations of serum alkaline phosphatase (>1000 IU/l) occur in those with malignant biliary obstruction, sepsis, and the acquired immunodeficiency syndrome (AIDS) with a systemic infection.57 Mildly elevated levels of serum alkaline phosphatase are non-specific and may be seen in cirrhosis, hepatitis, or congestive heart failure. Rarely, increased serum alkaline phosphatase occurs in a family on a genetic basis.58 Low levels of serum alkaline phosphatase may occur in hypothyroidism, pernicious anemia, zinc deficiency, and congenital hypophosphatasia, Wilson’s disease or in severe hepatic insufficiency.
g-GLUTAMYL TRANSFERASE (OR TRANSPEPTIDASE) g-Glutamyl transpeptidase catalyzes the transfer of g-glutamyl groups of peptides such as glutathione to other amino acids. gGlutamyl transpeptidase is widely distributed in membranes in a variety of tissues, including kidney, seminal vesicles, pancreas, liver, spleen, heart, and brain, and is thought to function in amino acid transport via a g-glutamyl cycle.59,60 a-Glutamyl transpeptidase has also been shown to catalyze the metabolism of S-substituted glutathione conjugates of various xenobiotics.61 g-Glutamyl transpeptidase has been localized to the entire hepatobiliary tree, from hepatocytes to common bile duct, and to pancreatic acini and ductules. The greatest concentration of enzyme is in epithelial cells lining fine biliary ductules. g-Glutamyl transpeptidase secreted into the bile is primarily of two types. One corresponds to the particulate fraction of biliary alkaline phosphatase. The other is a complex of biliary alkaline phosphatase and lipoproteins. g-Glutamyl transpeptidase occurs in the same membrane fragments that contain particulate alkaline phosphatase.62 Serum levels of g-glutamyl transpeptidase depend on age and sex; normal values are greater in men than in women, and in adults increase with age. Neonates may have values five to eight times higher than adults.63 Serum g-glutamyl transpeptidase levels are elevated in association with an array of pathologic states other than hepatobiliary disease; these include chronic alcoholism, pancreatic disease, myocardial infarction, renal failure, chronic obstructive pulmonary disease, and diabetes.63 In liver disease g-glutamyl transpeptidase activity in serum correlates well with serum alkaline phosphatase levels but is more sensitive. The diagnostic interest of serum g-glutamyl transpeptidase is its high sensitivity for hepatobiliary disease, particularly biliary tract disease, with an excellent negative predictive value (Table 14.3).63,64 Rarely is the serum g-glutamyl transpeptidase level normal in intrahepatic cholestasis.64–66 These exceptions tend to occur in infants or young children with some types of familial intrahepatic cholesta-
238
sis.67,68 Benign recurrent intrahepatic cholestasis can progress to progressive familial intrahepatic cholestasis with low serum g-glutamyl transpeptidase.69 The disadvantage is the low specificity. Moreover, the enzyme is inducible. Levels may be increased by a number of enzyme-inducing drugs, most notably alcohol and phenytoin, in the absence of other evidence of liver disease.70–72 One form of g-glutamyl transpeptidase has been proposed as a marker for hepatocellular carcinoma.73
BILIRUBIN Bilirubin is an endogenous organic anion derived primarily from the degradation of hemoglobin released from aging red blood cells. Measurement of bilirubin levels in serum is important in the assessment of hepatic function. Serum levels of bilirubin may be determined by the photometric detection of the azo derivatives obtained by the reaction of plasma with the diazonium ion of sulfanilic acid (the so-called diazo or van den Bergh reaction).74,75 The van den Bergh reaction separates bilirubin into two fractions: a water-soluble direct-reacting form representing conjugated bilirubin and a lipidsoluble indirect-reacting form representing unconjugated bilirubin.76 Direct-reacting bilirubin is a mixture of bilirubin monoglucuronides and diglucuronide that results from the conjugation of bilirubin in the liver. Bilirubin itself is insoluble in water and is bound to albumin; therefore, it does not appear in urine. In contrast, bilirubin glucuronides are water soluble and appear in urine when plasma levels are increased. When serum bilirubin glucuronides are elevated, some of the bilirubin may be bound covalently to serum albumin and is not filtered by the kidney. This mechanism accounts for the absence of bilirubinuria in some patients with direct hyperbilirubinemia, and for the slow resolution of jaundice in patients with otherwise normal liver function after recovery from acute liver disease.77 Measurement of bilirubin by the diazo method is not entirely accurate. At low serum levels the measurement of direct bilirubin by the diazo method overestimates conjugated bilirubin because some unconjugated bilirubin reacts directly with the reagent. This may lead to misinterpretation of mild unconjugated hyperbilirubinemia, as in Gilbert’s syndrome and hemolysis. More accurate measurement of bilirubin and its conjugates by high-performance liquid chromatography reveals that conjugated bilirubin constitutes only about 4–5% of total bilirubin in normal sera. This corresponds to a serum level of less than 1 mmol/l. By such methods, the upper limit of normal for direct bilirubin is 3 mmol/l. Such methods to measure bilirubin are not routinely employed, but provide the most sensitive means of detecting hepatobiliary disease and distinguishing the true unconjugated hyperbilirubinemia of Gilbert’s syndrome and hemolysis from the mild direct hyperbilirubinemia of liver disease. Normal levels of total bilirubin in serum or plasma are between 3 and 15 mmol/l and are significantly higher in men than in women. As previously indicated, hyperbilirubinemia is classified as either predominantly unconjugated or predominantly conjugated (Table 14-4). Levels between 17 and 70 mmol/l, representing unconjugated hyperbilirubinemia, may result from increased production of bilirubin, impaired transport of bilirubin into hepatocytes, or defective bilirubin conjugation in hepatocytes. Even in cases of severe hemolysis, the total serum bilirubin level is rarely above 70 mmol/l
Chapter 14 LABORATORY TESTING FOR LIVER DISEASE
Table 14-4. Causes of Hyperbilirubinemia Isolated unconjugated hyperbilirubinemia
Conjugated hyperbilirubinemia
Gilbert’s syndrome Neonatal jaundice Hemolysis Blood transfusion (hemolysis) Resorption of a large hematoma Shunt hyperbilirubinemia Crigler–Najjar syndrome Ineffective erythropoiesis Medications
Bile duct obstruction Hepatitis Cirrhosis Medications/toxins Primary biliary cirrhosis Primary sclerosing cholangitis Total parenteral nutrition Sepsis, benign postoperative jaundice Intrahepatic cholestasis of pregnancy Benign recurrent cholestasis Vanishing bile duct syndromes Dubin–Johnson syndrome Rotor syndrome
(5 mg/dl) in the presence of normal liver function. Serum bilirubin levels higher than 70 mmol/l or bilirubin levels between 17 and 70 mmol/l in association with other LFT abnormalities usually signify the presence of liver disease. In these situations at least 50% of the serum bilirubin is conjugated. Conjugated hyperbilirubinemia results from impaired intrahepatic excretion of bilirubin or extrahepatic obstruction. However, because of continued urinary excretion, maximum serum bilirubin levels plateau at around 500 mmol/l, even in complete bile duct obstruction. Extreme hyperbilirubinemia with levels above 500 mmol/l usually signifies severe parenchymal liver disease in association with hemolysis, as in sickle cell anemia or renal failure. The serum bilirubin level has prognostic value in chronic liver disease,78 particularly primary biliary cirrhosis and other cholestatic liver diseases,79 and in hepatic failure,80 in which deep jaundice is associated with increased mortality. In acute liver disease such as acute viral hepatitis, however, even profound jaundice is typically followed by complete recovery.
The highest concentrations of serum bile acids occur in viral hepatitis and extrahepatic obstruction.81–83 Serum bile acids are markedly elevated in cholestatic liver diseases, including primary biliary cirrhosis, primary sclerosing cholangitis, and intrahepatic cholestasis of pregnancy. Concentrations of endogenous bile acids in serum decrease during treatment with ursodeoxycholic acid, which becomes the predominant bile acid, accounting for 50% of the serum bile acid concentration.84
ALBUMIN Albumin is quantitatively the most important plasma protein synthesized by the liver and is a useful indicator of hepatic function. In adults the average size of the albumin pool is approximately 500 g; 12–15 g are synthesized daily. Because the half-life of albumin in serum is as long as 20 days, the serum albumin level is not a reliable indicator of hepatic protein synthesis in acute liver disease.85 Moreover, the serum albumin level at a single point in time may not reflect synthesis because the turnover of albumin can be affected by disturbances in distribution, catabolism, and synthesis. Albumin synthesis is affected not only by liver disease but also by nutritional status, hormonal balance, and osmotic pressure. In particular, albumin synthesis is exquisitely sensitive to the availability of amino acids, particularly tryptophan.86 Hypergammaglobulinemia may also lead to a decrease in the serum albumin level by increasing the contribution of serum immunoglobulins to the total plasma oncotic pressure.87 Serum albumin levels are typically depressed in patients with cirrhosis and ascites. In patients with or without ascites the serum albumin level correlates with prognosis.88 In the presence of ascites decreased levels of serum albumin may be a reflection of increased volume of distribution and impaired synthesis, and the actual size of the exchangeable albumin pool may be normal or increased.89 Albumin synthesis may also be depressed by heavy alcohol intake and poor dietary intake of protein. Other non-hepatic causes of hypoalbuminemia are nephrotic syndrome, protein-losing enteropathy, and burns.90
SERUM BILE ACIDS
PROTHROMBIN TIME AND SERUM COAGULATION FACTOR LEVELS
Bile acids are synthesized from cholesterol in the liver, conjugated with glycine or taurine, and excreted by the liver.81–83 Whereas serum bilirubin levels are influenced by factors such as the rate of bilirubin production and hepatic perfusion that do not directly reflect the function of the liver, bile acids more specifically reflect hepatic excretory function. Serum bile acids are more sensitive than bilirubin for detecting hepatobiliary disease. Fasting levels of serum bile acids are frequently elevated when the serum bilirubin level is normal.81 The greater sensitivity of serum bile acids is due in part to the much larger pool of bile acids compared to bilirubin. Bile acids (but not bilirubin) undergo enterohepatic recycling and storage in the gallbladder. In patients with hyperbilirubinemia fasting bile acid levels are usually elevated in serum, except in cases of hemolysis and congenital hyperbilirubinemia. The detection of normal serum bile acid levels in the presence of hyperbilirubinemia is thus a useful clue in the differential diagnosis of jaundice. However, this is also the same with the combination of bilirubin and normal g-glutamyltransferase.
The liver synthesizes coagulation factors I (fibrinogen), II (prothrombin), V, VII, IX, and X.91 Most of these are present in excess, and clotting abnormalities occur only when there is substantial impairment in the ability of the liver to synthesize these factors. The standard method of assessing impaired coagulation in liver disease is the one-stage prothrombin time of Quick,92 which evaluates the extrinsic coagulation pathway by measuring the rate of conversion of prothrombin to thrombin in the presence of a tissue extract (thromboplastin) and Ca2+ ions. The prothrombin time is prolonged when factors I, II, V, VII, and X are deficient, either singly or in combination. There is no advantage to using the International Normalized Ratio (INR) rather than the prothrombin time itself in patients with liver disease. In acute or chronic hepatocellular disease the prothrombin time may serve as a useful prognostic indicator. In acute hepatocellular disease a markedly prolonged and worsening prothrombin time suggests an increased likelihood of acute hepatic failure. The prognosis is particularly grave when the prothrombin time indicates decreases
239
Section II. Approach to the Patient with Liver Disease
of clotting factors to 10% of control or less. Prolongation of the prothrombin time also suggests a poor long-term prognosis in chronic liver disease and an increased risk of mortality from portosystemic shunt surgery.93,94 Prolongation of the prothrombin time is not specific for liver disease. It may occur as a result of congenital deficiencies of coagulation factors,95 acquired conditions such as consumptive coagulopathies, vitamin K deficiency, and the ingestion of drugs that antagonize the prothrombin complex (such as bishydroxycoumarin derivatives). Vitamin K deficiency primarily affects factors II, VII, IX, and X. Factor VII has the shortest half-life and decreases first, followed by factors X and IX. On the other hand, factor V is synthesized by the liver but not affected by vitamin K deficiency. Thus measurement of factor V levels can be used to distinguish hepatocellular injury from vitamin K deficiency, as occurs in obstructive jaundice or steatorrhea. In deficiency of vitamin K resulting from obstructive jaundice, steatorrhea, or therapeutic anticoagulation, coagulation factors are synthesized at the normal rate but lack procoagulant function because they have fewer of the g-carboxyglutamic acid residues necessary for binding Ca2+ to phospholipids.96 Vitamin K deficiency can be distinguished from impaired synthesis of coagulation factors by administering vitamin K parenterally.97 If the prothrombin time returns to normal or improves by at least 30% within 24 hours of administration of a single 10 mg injection of vitamin K, hepatic function is intact with regard to the synthesis of clotting factors, and vitamin K deficiency can be presumed to be responsible for the prolongation of the prothrombin time. Acetylcystein can artificially worsen prothrombin time in patients with uncomplicated paracetamol poisoning; a fall in this index might be misinterpreted as a sign of liver failure, leading to prolonged treatment time.98 Measurement of des-g-carboxy prothrombin (also known as prothrombin produced in the absence or antagonism of vitamin K II) also is a marker of liver dysfunction,99 and increased levels have been found in the plasma of patients with hepatocellular carcinoma.100 The abnormal prothrombin is presumably produced by the tumor. As a marker for hepatocellular carcinoma, the plasma level of desg-carboxy prothrombin can be complementary with levels of a-fetoprotein. Unfortunately, with currently available assays plasma levels of des-g-carboxy prothrombin are elevated in less than 50% of patients with tumors smaller than 3 cm in diameter.101–103 The test cannot be recommended for screening purposes.
OTHER ENZYMES 5¢-Nucleotidase catalyzes the hydrolysis of nucleotides by releasing inorganic phosphate from the 5¢-position of the pentose ring. 5¢Nucleotidase is present in the intestines, brain, heart, blood vessels, endocrine pancreas, and liver.104 In the liver 5¢-nucleotidase is associated primarily with canalicular and sinusoidal plasma membranes. Elevated serum levels of 5¢-nucleotidase are thought to be hepatobiliary in origin despite the widespread distribution of the enzyme in other body tissues. Serum levels of 5¢-nucleotidase correlate closely with serum alkaline phosphatase levels, and because serum 5¢-nucleotidase is so specific for liver diseases it is used to confirm the hepatic origin of elevated serum levels of alkaline phosphatase.
240
Measurements of 5¢-nucleotidase in serum can be useful in diagnosing liver disease in childhood and pregnancy. Alkaline phosphatase is increased physiologically in these settings, but 5¢-nucleotidase is not.104 Lactate dehydrogenase (LDH) is often included in liver biochemistry panels but has poor diagnostic specificity for liver disease. Markedly increased levels of LDH in serum may be seen in hepatocellular necrosis, shock liver, cancer, or hemolysis associated with liver disease. The ratio of ALT to LDH has been reported to distinguish acute viral hepatitis (>1.5) from shock liver and acetaminophen toxicity (<1.5) with a sensitivity of 94% and specificity of 84%.105 Glutamate dehydrogenase is a mitochondrial enzyme found particularly in centrilobular hepatocytes, whereas aminotransferases are distributed in a periportal location. It has been observed that serum glutamate dehydrogenase levels more closely reflect alcoholic hepatitis than do serum aspartate aminotransferase levels because the toxic effect of ethanol is directed primarily to the mitochondria in centrilobular hepatocytes.106 However, elevated levels of glutamate dehydrogenase in serum fall rapidly after cessation of alcohol intake, and may not discriminate between fatty liver and alcoholic hepatitis.107
TESTS THAT REFLECT FIBROSIS, SUCH AS HYALURONIC ACID AND BIOCHEMICAL PANELS One of the major clinical problems is how best to evaluate and manage the increasing numbers of patients infected with hepatitis C virus (HCV).108 Liver biopsy is still recommended in most patients.109,110 However, numerous studies strongly suggest that because of the limitations111–113 and risks of biopsy,114 as well as improvements in the diagnostic accuracy of biochemical markers,115,116 liver biopsy should no longer be considered mandatory. In a recent review of 66 studies of tests predicting biopsy findings, Gebo et al.117 concluded that panels of markers might have the greatest value in predicting the absence of or no more than minimal fibrosis on biopsy, and in predicting the presence of cirrhosis on biopsy. Serum ALT was the most commonly investigated marker, with sensitivity ranging from 61 to 71%. The diagnostic value was lower than that of a combination of markers. Among the extracellular matrix tests, hyaluronic acid correlated best with fibrosis stage overall, but has been demonstrated only for extensive fibrosis. The area under the receiver characteristics curve (AUROC) for extensive fibrosis ranges from 0.65 to 0.86. Markers of extracellular matrix degradation, such as tissue inhibitor of metalloproteinase-14, were less predictive than hyaluronic acid.117 Several cytokines and cytokine receptors were investigated, including tumor necrosis factors (TNF), TNF-R55, TNF-R75, and TNF-a, as well as serum interleukin (IL)-10, and IL-2 receptors. They were associated with fibrosis, but were less predictive than panel tests. TNF-a was associated with hepatic inflammation, but not with fibrosis. 117,118
Chapter 14 LABORATORY TESTING FOR LIVER DISEASE
Other tests were investigated, including glutathione, a-fetoprotein, prothrombin time, pseudocholinesterase, manganese superoxide dismutase, b-N-galactosidase, a2-macroglobin, haptoglobin, b-globulin, albumin, glutamyl transpeptidase, bilirubin, apolipoprotein A1, lactate dehydrogenase, AST, alkaline phosphatase, white blood cell count, platelet count, creatinine, total bile acids, and immunoglobulin G. Isolated, these markers were less useful than the panel of markers.109,117,118 Panels of markers have the greatest value in predicting the absence of or no more than minimal fibrosis on biopsy and in predicting the presence of cirrhosis on biopsy.117,118 A panel of matrix metalloprotein-2, 7S type IV collagen, and hyaluronic acid predicted no fibrosis/minimal fibrosis, with a sensitivity of 68% and specificity of 73%.109 An index combining age, platelets, GGT and cholesterol showed an AUROC at 0.81. One weakness of this index is the variability of serum cholesterol according to steatosis, particularly in patients infected by HCV genotype 3. A total of 16 publications115,116,118–134 have demonstrated the predictive value of two combinations of simple serum biochemical markers in patients infected with HCV (Table 14-5). One is called the fibrosis index (FibroTest FT) and combines the following five markers: a2-macroglobulin, haptoglobin, g-glutamyl transpeptidase (GGT), total bilirubin, and apolipoprotein A1.118 The second is called the activity index (ActiTest AT) and combines the same five markers plus alanine amino transferase (ALT), and has a high predictive value for the diagnosis of significant necroinflammatory histological activity.118,119 Since September 2002 these tests (FT-AT) are available in several countries, including the USA (FibroSure) as an alternative to liver biopsy. For the diagnosis of significant fibrosis by the METAVIR scoring system,129,130 the FT AUROC ranged from 0.73 to 0.87, significantly different from random diagnosis (AUROC = 0.50) in each study (Table 14-5).115,116,118–133 For the cut-off of 0.31, the FibroTest negative predictive value for excluding significant fibrosis (prevalence 0.31) was 91%.134 Similar results have not been obtained with other diagnostic tests. In four studies there was a direct comparison in the same patients of FT versus other biochemical markers, including hyaluronic acid,119 the Forns index, 123 the APRI index124 and the GlycoCirrhoTest. 133 All the comparisons were in favor of FT (Table 14-5), except for the GlycoCirrhoTest, which has a similar AUROC (0.87 vs 0.89 for FT).133 FT-AT has the same predictive values as were observed for patients co-infected with HIV, 121 and in patients with chronic hepatitis B.135 It is possible that biochemical markers such as those described here may provide a more accurate (quantitative and reproducible) picture of fibrogenic and necrotic events occurring within the liver than liver biopsy. The greater accuracies of FT-AT, when assessed with biopsy specimens greater than 15 mm versus smaller biopsies, suggest that some discordance between FT-AT and histology were due to biopsy specimen sampling error.115 Several case reports have observed false negatives of liver biopsy versus biochemical markers.115,116,118,121 The error was attributable to biopsy because there were overt clinical signs of cirrhosis, such as esophageal varices, low platelet counts, or a dysmorphic liver on ultrasound.115,116,118,121 In a recent prospective study it was estimated that
18% of discordances were attributable to biopsy failure (mostly due to small length) and 2% to FT-AT failure.116 Limitations of Fibrotest and Actitest are observed in less than 5%; they are false positive due to hemolysis (decrease of haptoglobin and increase of non-conjugated bilirubin), to Gilbert’s disease (increase in nonconjugated bilirubin), to acute inflammation (isolated increase in a2-macroglobulin), and to extrahepatic cholestasis (increase in total bilirubin and g-glutamyl transpeptidase); and false negative due to acute inflammation with isolated increase of haptoglobin.116 A weakness of these validations was that the same group that developed these tests performed most of the published studies. However, the independent published studies found the same significant diagnostic values as non-independent or multicenter studies. Among the nine studies estimating FT, five were performed by the same center (non-independent), two were performed in totally independent centers, and two were performed in multiple centers, mixing the non- and independent centers. The AUROCs for the diagnosis of F2F3F4 versus random AUROCs at 0.50 were all significant and similar between the three groups in a meta-analysis: mean difference in AUROC = 0.29, including 0.24 for independent, 0.25 for mixed and 0.36 for dependent studies.134 The advantages of the FT-AT are to give two quantitative and continuous estimates of both fibrosis and activity, with a conversion to the classic fibrosis stages and activity grades of the METAVIR scoring system.134 FibroTest permits the diagnosis of fibrosis in patients with normal transaminases.118 Recently, a study using profiles of serum protein N-glycans found that a profile had a similar AUROC to FT for the diagnosis of compensated cirrhosis. When combined with FT this marker had 100% specificity and 75% sensitivity for the diagnosis of compensated cirrhosis, which is not significantly different from the 92% specificity and 67% sensitivity of the FT.133
PLASMA LIPIDS AND LIPOPROTEINS The liver is important in the production and metabolism of plasma lipoproteins. The liver is the major source of plasma lipoproteins, with the exception of chylomicrons, which are synthesized by the intestine. Abnormalities in plasma concentrations of lipids and lipoproteins are common in liver disease.136 In acute hepatocellular injury increased levels of plasma triglycerides, decreased levels of cholesterol esters, and abnormal electrophoretic patterns of lipoproteins are typically observed. Many of these abnormalities can be related to deficiencies of enzymes of hepatic origin, including lecithin–cholesterol acyltransferase (LCAT) and hepatic lipase.137,138 Serum levels of apolipoprotein A1 decrease in acute viral hepatitis.139 In chronic liver disease serum levels of apolipoprotein A1 decrease with fibrosis progression.118 Low levels of cholesterol in serum may also reflect malnutrition. In genetic hepatic amyloidosis due to a mutation in apolipoprotein A1 the serum level is decreased and is associated with high g-glutamyl transferase serum activity and a small increase in creatinine level.140 In patients with intrahepatic or extrahepatic cholestasis, levels of cholesterol and phospholipids are typically elevated, often strikingly
241
Section II. Approach to the Patient with Liver Disease
Table 14-5. Summary of the Diagnostic Value of Fibrotest for the Staging of Hepatic Fibrosis and Comparisons with Hyaluronic Acid, the Forns Index and the Apri Index in Patients with Chronic Hepatitis C, from the Published Studies
242
First author
Number patients
Methodology
Marker
Stage prevalence
ImbertBismut, 2001
AUROC SE
189
Prospective Single center First year cohort
FibroTest
F2F3F4 0.38
0.84 0.03
ImbertBismut, 2001
134
Prospective Single center Validation cohort
FibroTest
F2F3F4 0.45
Poynard, 2001
165
Retrospective Randomized trial Multicenter
FibroTest
Poynard, 2001
165
Retrospective Randomized trial Multicenter
Poynard, 2003
352
Poynard, 2003
Cut-off
Sensitivity
Specificity
0.10 0.30 0.60 0.80
0.97 0.79 0.51 0.29
0.24 0.65 0.94 0.95
0.87 0.03
0.10 0.30 0.60 0.80
1.00 0.87 0.70 0.38
0.22 0.59 0.95 0.97
F3F4 Knodell 0.32
0.74 0.03
0.10 0.30 0.60 0.80
0.96 0.81 0.50 0.13
0.24 0.65 0.92 0.98
Hyaluronic
F3F4 Knodell 0.32
0.65 0.03
0.81 0.47 0.23
0.39 0.65 0.91
Retrospective Randomized trial Multicenter Before treatment
FibroTest
F2F3F4 0.39
0.73 0.03
0.10 0.30 0.60 0.80
0.97 0.86 0.50 0.20
0.08 0.45 0.79 0.95
352
Retrospective Randomized trial Multicenter After treatment
FibroTest
F2F3F4 0.32
0.77 0.03
0.10 0.30 0.60 0.80
0.98 0.85 0.46 0.16
0.15 0.39 0.81 0.97
Rossi, 2003
125
Prospective Multicenter Non-validated analyzers
FibroTest
F2F3F4 0.38
0.74 0.05
0.10 0.30 0.60 0.80
0.92 0.75 0.42 0.22
0.29 0.61 0.94 0.96
Myers, 2003
130
Retrospective Single center HCV–HIV Coinfection
FibroTest
F2F3F4 0.45
0.86 0.04
0.10 0.30 0.60 0.80
0.98 0.90 0.66 0.34
0.17 0.60 0.92 0.96
Thabut, 2003
249
Retrospective Single center From Imbert-Bismut publication
FibroTest
F2F3F4 0.38
0.84 0.02
0.10 0.30 0.60 0.80
0.98 0.84 0.58 0.29
0.22 0.65 0.93 0.95
Thabut, 2003
249
Retrospective Single center From Imbert-Bismut 2001
Forns Index
F2F3F4 0.38
0.78 0.03
1 3 6 8
1.00 1.00 0.55 0.19
0.04 0.26 0.86 0.97
Le Calvez 2004
323
Retrospective Single center From Imbert-Bismut 2001
FibroTest
F2F3F4 0.41
0.83 0.02
0.10 0.30 0.60 0.80
0.97 0.81 0.58 0.33
0.30 0.66 0.93 0.95
Le Calvez 2004
323
Retrospective Single center From Imbert-Bismut 2001
APRI Index
F2F3F4 0.41
0.74 0.03
0.50 1.00 1.50 2.00
0.81 0.54 0.36 0.24
0.56 0.84 0.91 0.95
Callewaert 2004
82
Prospective
FibroTest
F4 0.29
0.89 0.04
0.10 0.30 0.60 0.80
1.00 0.92 0.79 0.67
0.33 0.62 0.81 0.92
Callewaert 2004
82
Prospective
Glyco Cirrho Test
F4 compensated 0.29
0.87 0.04
1.00 0.79 0.21 0.17
0.12 0.88 0.95 1.00
20 40 100
-0.2 0.1 0.4 0.6
Chapter 14 LABORATORY TESTING FOR LIVER DISEASE
in patients with chronic cholestasis. Factors contributing to these changes include increased hepatic synthesis of cholesterol, regurgitation of biliary cholesterol into the plasma, decreased plasma LCAT activity, and regurgitation of biliary lecithin into the plasma.141,142 In addition, the serum of patients with obstructive jaundice contains an abnormal lipoprotein called lipoprotein X. Lipoprotein X is thought to represent non-esterified cholesterol and phospholipid regurgitated from bile. In contrast to hepatocellular disease, LCAT deficiency is a late manifestation of cholestasis and reflects a long duration of disease and evidence of hepatocellular dysfunction.143 A marked increase in high-density lipoprotein (HDL) has been observed early in the course of primary biliary cirrhosis, presumably because of the release of an inhibitor of hepatic lipase. In advanced primary biliary cirrhosis HDL levels are decreased and levels of lowdensity lipoproteins and lipoprotein X are increased.144
TESTS THAT CONTRIBUTE TO ACCURATE DIAGNOSIS IN LIVER DISEASE LABORATORY TESTS FOR HCV PCR Amplification The polymerase chain reaction (PCR) can detect 10–50 IU/ml of HCV RNA using the previous methods. Testing for HCV RNA is a reliable way of demonstrating HCV infection and is the most specific test. Testing for HCV RNA is particularly useful when transaminases are normal, several causes of liver disease are possible (i.e. alcohol consumption), in immunosuppressed patients (i.e. after transplantation, in HIV co-infected patients), and in acute hepatitis C before the development of antibodies (4–10 weeks).145
Enzyme Immunoassay Anti-HCV is detected by enzyme immunoassay. The third-generation test is usually very sensitive and very specific. In case of false positive or false negative doubts, the best test for confirmation of HCV infection is HCV RNA PCR. Immunosuppressed patients infected by HCV may test negative for anti-HCV. Antibody is usually present by 1 month (4–10 weeks) after the onset of acute illness. Anti-HCV is still detectable during and after treatment, whatever the response, and must not be tested for again.
Genotype and Serotype There are six major genotypes of hepatitis C and more than 50 subtypes. Knowing the genotype or serotype (genotype-specific antibodies) is helpful for the interferon plus ribavirin combination treatment of choice. Genotypes do not change during the course of infection and must not be tested for again. Knowing subtypes (i.e. 1a versus 1b) is currently not clinically helpful. There is no relationship between the severity of the disease (fibrosis stage) and genotype.
Quantification of HCV RNA in Serum Methods measuring the level of virus in serum use quantitative PCR and a branched DNA (bDNA) test. They are currently less sensitive than qualitative assays.
An effort was made to define clinically relevant HCV RNA loads in standardized international units (IU) for use in routine clinical and research applications based on standardized quantitative assays validated with appropriate calibrated panels. The semi-automated quantitative RT-PCR Superquant assay (National Genetics Institute, Los Angeles, CA), has a range from 50 to 1 470 000 IU/ml; the semiautomated Cobas Amplicor HCV Monitor assay version 2.0 (Cobas v2.0, Roche Molecular Systems, Pleasanton, CA) has a range from 600 to 2 630 000 IU/ml; the semi-automated HCV RNA quantitative assay LCx (Abbott Diagnostics, Chicago, IL) has a range of 25–2 630 000 IU/ml; the semi-automated branched DNA signal amplification Versant HCV RNA 3.0 assay (bDNA) (Bayer Corporation, Diagnostic Division, Tarrytown, NY), has a range 615–7 700 000 IU/ml; the manual branched DNA signal amplification (Quantiplex) Versant HCV RNA 2.0 assay (bDNA) (Bayer Corporation, Diagnostic Division, Tarrytown, NY), has a range 3200–19 000 000 IU/ml. In the more recent studies the median viral load ranged from 800 000 IU/ml (5.9 log10 IU/ml) to almost 1 300 000 IU/ml (6.1 log10 IU/ml). Knowing viral load is helpful for the pegylated or non-pegylated interferon plus ribavirin treatment choice. Patients with a high initial viral load have higher relapse rates and benefit more from a 48-week regimen than patients with a lower viral load. A 12-week stopping rule is also possible when the decrease in viral load is less than 2 logs compared with baseline value. Unlike HIV infection, viral load does not correlate with the severity of hepatitis (fibrosis progression).
HCV Core Antigen Assays HCV core antigen can be detected in the serum and quantified (Total HCV-core antigen assay, Ortho-Clinical Diagnostics, Raritan, USA). This quantification can be used as an indirect marker of the HCV viral load but is less sensitive than molecular HCV-RNA assays. However, it is less expensive than quantitative HCV-RNA testing.
LABORATORY TESTS FOR HBV Knowledge of the virus genome has permitted the development of several serological markers of HBV infection. Hepatitis B surface antigen (HBsAg and HBs), HbeAg, anti-Hbe and HBs are tested by enzyme-linked immunosorbent assay (ELISA). The presence of HBsAg in the serum for 6 months or more is indicative of chronic hepatitis B infection. The estimated annual incidence for clearance of HBsAg in chronically infected patients is low (0.1–0.8%) and usually due to a decrease in viremia rather than the emergence of HBsAg mutants. Commonly used commercial assays for HBV DNA levels are a branched DNA (bDNA) assay and a hybrid capture test. The lower limits of detection for these two assays are 700 000 and 140 000 copies/ml, respectively. One commercially available PCR assay allows for the detection of 200 copies/ml.146 The interpretation of HBV serological markers is described in Table 14-6.146,147 The best evaluable assay for HCV-RNA quantification is real-time PCR with log10 results in copies per ml. In clinical practice chronic HBV carriers may be divided into two easily identifiable serologic types: those who are positive for HBeAg
243
Section II. Approach to the Patient with Liver Disease
Table 14-6. Interpretation of Hepatitis B Serological Markers According to Symptoms, Transaminases and Histologic Features
Acute Chronic carrier wild type Chronic carrier pre-core mutant1 ‘Healthy carrier’2 ‘Immune–tolerance 2 Recovery/immunity Immunity from vaccination Occult infection3
HBsAg
HBsAb
HBeAg
HBeAb
Anti HBc IgG
Anti HBc-IgM
HBV-DNA
Symptoms
ALT
Histologic activity and fibrosis
+ + + + + – – –
– – – – – + + +/–
+ + – + + – – –
+/– – + – – + – –
+ + + + + + – +/–
+ –/+ –/+ – – – – –
+ + + – ++ – – +/–
+/– +/– +/– – – – – +/–
++ +/– +/– – – – – +/–
+ + + – – – – +/–
During flare-up antiHBc-IgM can be elevated. Carrier without symptoms, with normal transaminases and with normal biopsy are divided into ‘healthy carriers’ with undetectable HBV-DNA and subjects with detectable HBV-DNA (‘immune tolerance’). In these patients a risk of cirrhosis or hepatocellular carcinoma cannot be excluded. Pre-core mutant can be detected. 3 HBV DNA can be detected in the liver in absence of serological markers. 1 2
(‘wild type’) and those who are HBeAg negative and positive for anti-HBe. Recently genotype testing for HBV has been shown to be of value for treatment response and natural course. Despite the preliminary nature of HBV genotype testing, this may become important in the future as HBV genotype testing (major genotypes are A, B, C, D) will presumably become available.
LABORATORY TESTS FOR HDV All patients with positive HBsAg should be tested for co-infection with hepatitis D (delta virus). The best available test is anti-HDV antibody testing. The measurement of HDV-RNA is limited to specialist laboratories.
244
hepatitis C. Autoantibodies are a diagnostic hallmark, and the conventional serologic markers of AIH are antinuclear antibodies (ANA), smooth muscle antibodies (SMA), and antibodies to liver/kidney microsome type 1 (anti-LKM1). Diagnostic criteria have been codified and updated by an international panel (Table 147).151 Differences between a definite and a probable diagnosis of AIH relate mainly to the degree of serum g-globulin or immunoglobulin G elevation, levels of ANA, SMA, or anti-LKM1, and exposure to alcohol, medications, or infections that could cause liver injury. There is no time requirement to establish chronicity, and cholestatic clinical, laboratory, and histologic changes preclude the diagnosis. The presence of liver-specific cytosol antigen type 1 (anti-LC1), soluble liver antigen/liver pancreas (anti-SLA/LP), actin (anti-actin), and/or perinuclear antineutrophil cytoplasmic antibodies (pANCA), support a probable diagnosis if the other conventional markers are absent.152
LABORATORY TESTS FOR OTHER LIVER DISEASES Hemochromatosis
Primary Biliary Cirrhosis (PBC)
In patients with serum GGT or ALT elevations without obvious origin, serum ferritin levels, serum iron, and total iron-binding capacity should be measured so that the iron saturation (serum iron/ironbinding capacity) can be calculated. These tests, however, are not specific for hemochromatosis, and their specificity is particularly low in patients with other forms of liver disease. The serum tests that assay iron metabolism are age dependent and can be abnormal in the heterozygote state and in many other conditions.60 The genotype typically found in individuals of northern European descent can be diagnosed with a high degree of accuracy by detection of mutations in the HFE gene. The C282Y/C282Y homozygote is most likely to manifest the disease phenotype, whereas a minority of genotype C282Y/H63D (compound heterozygote) individuals may also develop the disease.148–150 A small population of patients have hemochromotosis due to genetic defects other than the HFE gene.
The specific diagnostic features are serologic: elevated immunoglobulins, especially IgM, and disease-specific autoantibodies – antimitochondrial antibodies (AMA) and other autoantibodies reacting with nuclear pore complex, such as gp210; and histologic: bile duct damage and portal inflammation with granulomas. PBC is a chronic non-suppurative destructive granulomatous cholangitis of uncertain etiology, affecting principally the medium-sized intrahepatic bile ducts. There is a female preponderance (over 90%) and a strong association with other autoimmune conditions, particularly Sjögren’s syndrome and thyroid disease. PBC is unique among autoimmune disease in that the condition has not been described in children.153 The diagnostic hallmark of PBC are so-called M2-AMA antibodies, the main target antigen of which is the E2 subunit of pyruvate dehydrogenase (PDH). Anti PDH-E2 antibodies have a sensitivity of 95% and a specifity of close to 100 % for PBC.
Autoimmune Hepatitis (AIH)
Primary Sclerosing Cholangitis (PSC)
Diagnosis requires the presence of characteristic features and the exclusion of other conditions that resemble AIH. The conditions most likely to be confused with AIH are Wilson’s disease, druginduced hepatitis, and chronic viral hepatitis, especially chronic
In contrast to AIH and PBC, there are no diagnostic immunologic features and biochemical, and serologic criteria define PSC only poorly. Bile ducts of all sizes may be affected by PSC, which also results in cholestasis and ultimately in cirrhosis. Unlike PBC and
Chapter 14 LABORATORY TESTING FOR LIVER DISEASE
Table 14-7. Diagnostic Criteria for Autoimmune Hepatitis
Positive laboratory testing Autoantibodies Laboratory features Globulin, g-globulin or immunoglobulin G level >=1.5 times normal Hypergammaglobulinemia of any degree Negative laboratory testing No active viral infection No genetic liver disease Normal serum ceruloplasmin, iron, and ferritin levels Non-specific serum copper, ceruloplasmin, iron, and/or ferritin abnormalities Other criteria No toxic or alcohol injury Histologic findings
Definite autoimmune hepatitis
Probable autoimmune hepatitis
SMA, ANA, or anti-LKM1, >=1:80 in adults and >=1:20 in children; no AMA Predominant serum aminotransferase abnormality Predominant serum aminotransferase abnormality
ANA, SMA, or anti-LKM1 >=1:40 in adults or other autoantibodies*
No markers of current infection with hepatitis A, B, and C viruses Normal a1-antitrypsin phenotype Partial a1-antitrypsin deficiency
No markers of current infection with hepatitis A, B, and C viruses
Daily alcohol <=25 g/day and no recent use of hepatotoxic drugs Interface hepatitis; no biliary lesions, granulomas, or prominent changes suggestive of another disease
Daily alcohol <=50 g/day and no recent use of hepatotoxic drugs Interface hepatitis; no biliary lesions, granulomas, or prominent changes suggestive of another disease
*Includes perinuclear antineutrophil cytoplasmic antibodies and the not generally available antibodies to soluble liver antigen/liver pancreas, actin, liver cytosol type 1, and asialoglycoprotein receptor. Based on recommendations of the International Autoimmune Hepatitis Group. SMA, smooth muscle antibodies; AMA, antimitochondrial antibodies; ANA, antinuclear antibodies; anti-LKM1 antibodies to liver/kidney microsome type 1.
AIH, PSC affects predominantly men and is probably associated with inflammatory bowel disease in around 70% of cases (ranging from 40 to 98% in different series). The diagnosis rests on distinctive cholangiographic appearances showing large intrahepatic and extrahepatic bile duct involvement.154 The most prominent autoantibodies are ANCA. These are not specific for PBC; however, some laboratories claim to have identified ANCA subtypes associated with PSC.
MISCELLANEOUS TESTS Serum g-Globulins Autoimmune hepatitis is characterized by greatly increased serum g-globulin levels, usually immunoglobulin (Ig) G.155 High g-globulin levels may be associated with false positive hepatitis C antibody results by enzyme immunoassay. The elevated g-globulin levels decline and the false positive hepatitis C antibody results are reversed on treatment with corticosteroids.156,157 Increased levels of g-globulins usually accompany alcoholic cirrhosis, specifically the so-called fast g-fraction, which produces a characteristic bridging pattern on electrophoresis. In addition, alcoholic cirrhosis is associated with elevated levels of serum IgA.158 Primary biliary cirrhosis is associated typically with hyperglobulinemia, predominantly of the IgM class. Serum globulins are not usually elevated in drug-induced cholestasis or extrahepatic obstruction. The detection of hyperglobulinemia may be a clue to the presence of chronic liver disease. However, quantitative measurements of
immunoglobulins, although helpful in supporting a clinical diagnosis, are not specific enough to be relied on to establish the precise cause of liver disease.
TESTS THAT MEASURE THE CAPACITY OF THE LIVER TO TRANSPORT Laboratory diagnosis of liver disease was introduced in 1913, when it was observed that a phthalein dye could be used to investigate hepatic function. Subsequently, it was demonstrated that intravenously administered sulfobromophthalein (bromsulphalein; BSP) or indocyanine green (ICG) or sorbitol were removed from the blood primarily by the liver, and that their rate of clearance was useful for evaluating liver function.159–162 The extraction ratio of ICG by the liver is higher than that for BSP (50–80%). Plasma levels of ICG can be determined directly by spectrophotometry, and repeated measurements without multiple blood sampling can be performed.163,164 The sensitivity of ICG for the detection of slight hepatic dysfunction is limited and less than for BSP. Clearance of sorbitol has also been used to measure hepatic blood flow, with the advantages of greater safety and less expense than ICG.165 However, these tests are usually not available in routine laboratories.
245
Section II. Approach to the Patient with Liver Disease
TESTS THAT MEASURE THE CAPACITY OF THE LIVER TO METABOLIZE DRUGS Hepatic function may be assessed with substances that are metabolized selectively by the liver. The most widely performed tests assess hepatic drug metabolism, such as the determination of plasma clearance of antipyrine and the 14C-aminopyrine breath test.166,167 Additional tests include the determination of caffeine clearance, galactose elimination, the maximum rate of synthesis of urea, and the formation of metabolites of lidocaine.168–170 Their precise role in practice remains uncertain. Compared to common liver biochemical tests, ‘quantitative’ tests are cumbersome, labor-intensive and expensive. They measure a specific function of a specific hepatic microsomal enzyme. However, in acute and chronic liver disease the quantitative measurement of the function of an individual microsomal enzyme does not reflect a test for global liver function. Hopefully, in the future a combination of several tests will allow us to obtain a global liver function test that is comparable to creatinine clearance in kidney disease.
ACKNOWLEDGMENT Supported by grants from ARMHV and ARECA.
REFERENCES 1. Pratt DS, Kaplan MM. Evaluation of abnormal liver-enzyme results in asymptomatic patients. N Engl J Med 2000;342: 1266. 2. Aranda-Michel J, Sherman KE. Tests of the liver: use and misuse. Gastroenterologist 1998;6:34. 3. Afdhal NH. Biopsy or biomarkers: is there a gold standard for diagnosis of liver fibrosis? Clin Chem 2004;50:1299–1300. 4. Bacq Y, Zarka O, Brechot JF, et al. Liver function tests in normal pregnancy: a prospective study of 103 pregnant women and 103 matched controls. Hepatology 1996;23:1030. 5. Rej R. Aspartate aminotransferase activity and isoenzyme proportions in human liver tissues. Clin Chem 1978; 24:1971. 6. Kew MC. Serum aminotransferase concentration as evidence of hepatocellular damage. Lancet 2000;355:591. 7. Foust RT, O’Brien JF, Schiff ER. Isolated aspartate aminotransferase elevation due to macroenzyme formation with liver biopsy correlation. Am J Gastroenterol 1990;85:88. 8. Ohtsuki T, Inui T, Andoh A, et al. Characteristics of aspartate aminotransferase binding immunoglobulin determined by the isotope method. J Lab Clin Med 1994;124:218–223. 9. Piton A, Poynard T, Imbert-Bismut F, et al. Factors associated with serum alanine transaminase activity in healthy subjects: consequences for the definition of normal values, for selection of blood donors, and for patients with chronic hepatitis C. Hepatology 1998;27:1213. 10. Halfon P, Imbert-Bismut F, Messous D, et al. A prospective assessment of the inter-laboratory variability of biochemical markers of fibrosis (FibroTest) and activity (ActiTest) in patients with chronic liver disease. Comp Hepatol 2002;2:3–7. 11. Johnson RD, O’Connor ML, Kerr RM. Extreme serum elevations of aspartate aminotransferase. Am J Gastroenterol 1995;90:1244.
246
12. Whitehead MW, Hawkes ND, Hainsworth I, et al. A prospective study of the causes of notably raised aspartate aminotransferase of liver origin. Gut 1999;45:129. 13. Flora KD, Keeffe EB. Evaluation of mildly abnormal liver tests in asymptomatic patients. J Insur Med 1990;22:206. 14. Hultcrantz R, Glaumann H, Lindberg G, et al. Liver investigation in 149 asymptomatic patients with moderately elevated activities of serum aminotransferases. Scand J Gastroenterol 1986;21:109. 15. Van Ness NM, Diehl AM. Is liver biopsy useful in the evaluation of patients with chronically elevated liver enzymes? Ann Intern Med 1989;111:473. 16. Friedman LS, Dienstag JL, Watkins E, et al. Evaluation of blood donors with elevated serum alanine aminotransferase levels. Ann Intern Med 1987;107:137. 17. Katkov WN, Friedman LS, Cody H, et al. Elevated serum alanine aminotransferase levels in blood donors: the contribution of hepatitis C virus. Ann Intern Med 1991;115:882. 18. Daniel S, Ben-Menachem T, Vasudevan G, et al. Prospective evaluation of unexplained chronic liver transaminase abnormalities in asymptomatic and symptomatic patients. Am J Gastroenterol 1999;94:3010. 19. Briere RO. Serum ALT levels: effect of sex, race, and obesity on unit rejection rate. Transfusion 1988;28:392. 20. Hetland G, Holme I, Stromme JH. Co-variation of alanine aminotransferase levels with relative weight in blood donors. Scand J Clin Lab Invest 1992;52:51. 21. Hay JE, Czaja AJ, Rakela J, et al. The nature of unexplained chronic aminotransferase elevations of a mild to moderate degree in asymptomatic patients. Hepatology 1989;9:193. 22. Palmer M, Shaffner F. Effect of weight reduction on hepatic abnormalities in overweight patients. Gastroenterology 1990;99:1408. 23. Hickman IJ, Jonsson JR, Prins JB, et al. Modest weight loss and physical activity in overweight patients with chronic liver disease results in sustained improvements in alanine aminotransferase, fasting insulin, and quality of life. Gut 2004;53:413–419. 24. Helfgott SM, Karlson E, Beckman E. Misinterpretation of serum transaminase elevation in ‘occult’ myositis. Am J Med 1993;95:447. 25. Fallon KE, Sivyer G, Sivyer K, et al. The biochemistry of runners in a 1600 km ultramarathon. Br J Sports Med 1999;33:264–269. 26. Patwardhan RV, Smith OJ, Farmelant MH. Serum transaminase levels and cholescintigraphic abnormalities in acute biliary tract obstruction. Arch Intern Med 1987;147:1249. 27. Ruttimann S, Dreifuss M, Clemencon D, et al. Multiple biochemical blood testing as a case-finding tool in ambulatory medical patients. Am J Med 1993;94:141. 28. Goddard CJ, Warnes TW. Raised liver enzymes in asymptomatic patients: investigation and outcome. Dig Dis 1992;10:218. 29. Bhavnani M, Lloyd D, Bhattacharyya A, et al. Screening for genetic haemochromatosis in blood samples with raised alanine aminotransferase. Gut 2000;46:707. 30. Bardella MT, Vecchi M, Conte D, et al. Chronic unexplained hypertransaminasemia may be caused by occult celiac disease. Hepatology 1999;29:654. 31. Rivera-Nieves J, Kozaiwa K, Parrish CR, et al. Marked transaminase elevation in anorexia nervosa. Dig Dis Sci 2000;45:1959. 32. Bonamico M, Pitzalis G, Culasso F. Hepatic damage during coeliac disease in childhood. Minerva Pediatr 1986;38:959. 33. Mugica F, Castiella A, Otazua P, et al. Prevalence of coeliac disease in unexplained chronic hypertransaminasemia. Rev Esp Enferm Dig 2001;93:707–714. 34. Volta U, Granito A, De Franceschi L, et al. Anti tissue transglutaminase antibodies as predictors of silent coeliac disease
Chapter 14 LABORATORY TESTING FOR LIVER DISEASE
35.
36. 37. 38.
39.
40.
41.
42.
43.
44.
45.
46. 47. 48.
49. 50. 51.
52.
53.
54.
55.
56.
57.
in patients with hypertransaminasaemia of unknown origin. Dig Liver Dis 2001;33:420–425. Kundrotas LW, Clement DJ. Serum alanine aminotransferase (ALT) elevation in asymptomatic US Air Force basic trainee blood donors. Dig Dis Sci 1993;38:2145. Cohen JA, Kaplan MM. The SGOT/SGPT ratio – an indicator of alcoholic liver disease. Dig Dis Sci 1979;24:835. Matloff DS, Selinger MJ, Kaplan MM. Hepatic transaminase activity in alcoholic liver disease. Gastroenterology 1980;78:1389. Reedy DW, Loo AT, Levine RA. AST/ALT ratio >1 is not diagnostic of cirrhosis in patients with chronic hepatitis C. Dig Dis Sci 1998;43:2156. Imperiale TF, Said AT, Cummings OW, et al. Need for validation of clinical decision aids: use of the AST/ALT ratio in predicting cirrhosis in chronic hepatitis C. Am J Gastroenterol 2000;95:2328. Giannini E, Botta F, Fasoli A, et al. Progressive liver functional impairment is associated with an increase in AST/ALT ratio. Dig Dis Sci 1999;44:1249. Sheth SG, Flamm SL, Gordon FD, et al. AST/ALT ratio predicts cirrhosis in patients with chronic hepatitis C virus infection. Am J Gastroenterol 1998;93:44. Sorbi D, Boynton J, Lindor KD. The ratio of aspartate aminotransferase to alanine aminotransferase: potential value in differentiating nonalcoholic steatohepatitis from alcoholic liver disease. Am J Gastroenterol 1999;94:1018. Ludwig S, Kaplowitz N. Effect of pyridoxine deficiency on serum and liver transaminases in experimental liver injury in the rat. Gastroenterology 1980;79:545. Nalpas B, Vassault A, Charpin S, et al. Serum mitochondrial aspartate aminotransferase as a marker of chronic alcoholism: diagnostic value and interpretation in a liver unit. Hepatology 1986;6:608. Yasuda K, Okuda K, Endo N, et al. Hypoaminotransferasemia in patients undergoing long-term hemodialysis: clinical and biochemical appraisal. Gastroenterology 1995;109:1295. Millan JL. On codevelopmental expression and structure of alkaline phosphatase genes. Anticancer Res 1988;8:995. Simko V. Alkaline phosphatases in biology and medicine. Dig Dis 1991;9:189. Hagerstrand I. Distribution of alkaline phosphatase activity in healthy and diseased human liver tissue. Acta Pathol Microbiol Scand A 1975;83:519. Gordon T. Factors associated with serum alkaline phosphatase level. Arch Pathol Lab Med 1993;117:187. Reichling JJ, Kaplan MM. Clinical use of serum enzymes in liver disease. Dig Dis Sci 1988;33:1601. Lieberman D, Phillips D. ‘Isolated’ elevation of alkaline phosphatase: significance in hospitalized patients. J Clin Gastroenterol 1990;12:415. Clubb JS, Neale FC, Posen S. The behavior of infused placental alkaline phosphatase in human subjects. J Lab Clin Med 1965;66:493. Kaplan MM, Righetti A. Induction of rat liver alkaline phosphatase: the mechanism of the serum elevation in bile duct obstruction. J Clin Invest 1970;49:508. Seetharam S, Sussman NL, Komoda T, et al. The mechanism of elevated alkaline phosphatase activity after bile duct ligation in the rat. Hepatology 1986;6:374. George GO, Spiegelman GA, Barkin JS. Normal serum alkaline phosphatase: an unusual finding in early suppurative biliary obstruction. Am J Gastroenterol 1993;88:771. Cooper JF, Brand EJ. Symptomatic sclerosing cholangitis in patients with normal alkaline phosphatase: two case reports and a review of the literature. Am J Gastroenterol 1988;83:308. Maldonado O, Demasi R, Maldonado Y, et al. Extremely high levels of alkaline phosphatase in hospitalized patients. J Clin Gastroenterol 1998;27:342.
58. Wilson JW. Inherited elevation of alkaline phosphatase activity in the absence of disease. N Engl J Med 1979;301:983. 59. Goldberg DM, Martin JV. Role of g-glutamyl transpeptidase activity in the diagnosis of hepatobiliary disease. Digestion 1975;12:232. 60. Meister A. Glutathione and the g-glutamyl cycle. In: Arias IM, Jakoby WB, eds. Glutathione: metabolism and function. New York: Raven Press, 1976:35. 61. Tate SS, Thompson GA, Meister A. Recent studies on gglutamyl transpeptidase. In: Arias IM, Jakoby WB, eds. Glutathione: metabolism and function. New York: Raven Press, 1976:45. 62. Wenham PR, Price CP, Sammons HG. g-Glutamyl-transferase isoenzymes in human bile. J Clin Pathol 1978;31:666. 63. Penn R, Worthington DJ. Is serum g-glutamyl-transferase a misleading test? BMJ 286:531, 1983. 64. Whitfield JB, Pounder RE, Neale G, et al. Serum g-glutamyl transpeptidase activity in liver disease. Gut 1972;13:702. 65. Kajiwara E, Akagi K, Tsuji H, et al. Low gammaglutamyltranspeptidase activity in patients with acute intrahepatic cholestasis. Gastroenterol Jpn 1981;16:76–81. 66. Yoshida H, Sakuma H, Morito T, et al. No elevation of serum gamma-glutamyl transpeptidase level in a polymyositis case with intrahepatic cholestasis. Nippon Shokakibyo Gakkai Zasshi. 1983;80:2224–2227. 67. Jansen PLM, Müller M. The molecular genetics of familial intrahepatic cholestasis. Gut 2000;47:1. 68. Lachaux A, Loras-Duclaux I, Bouvier R, et al. Benign recurrent cholestasis with normal gamma-glutamyl-transpeptidase activity. J Pediatr 1992;121:78. 69. Van Ooteghem NA, Klomp LW, van Berge-Henegouwen GP, et al. Benign recurrent intrahepatic cholestasis progressing to progressive familial intrahepatic cholestasis: low GGT cholestasis is a clinical continuum. J Hepatol 2002;36:439–443. 70. Keeffe EB, Sunderland MC, Gabourel JD. Serum gammaglutamyl transpeptidase activity in patients receiving chronic phenytoin therapy. Dig Dis Sci 1986;31:1056. 71. Zien M, Discombe G. Serum g-glutamyl transpeptidase as a diagnostic aid. Lancet 1970;2:748. 72. Moussavian SN, Becker RC, Pipmeyer JL, et al. Serum gamma-glutamyl transpeptidase and chronic alcoholism: influence of alcohol ingestion and liver disease. Dig Dis Sci 1985;30:211. 73. Jeng K-S, Sheen I-S, Tsai Y-C. Gamma-glutamyl transpeptidase messenger RNA may serve as a diagnostic aid in hepatocellular carcinoma. Br Med J 2001;88:986. 74. Tiribelli C, Ostrow JD. New concepts in bilirubin chemistry, transport and metabolism: report of the Second International Bilirubin Workshop, 9–11 April 1992, Trieste, Italy. Hepatology 1993;17:715. 75. van den Bergh HAA, Muller P. Uber ein direkte und indirekte Diazoreaktion auf Bilirubin. Biochem Zeitschr 1916;77:93. 76. Malloy HT, Evelyn KA. The determination of bilirubin with the photoelectric colorimeter. J Biol Chem 1937;119:481. 77. Weiss JS, Gautam A, Lauff JJ, et al. The clinical importance of a protein-bound fraction of serum bilirubin in patients with hyperbilirubinemia. N Engl J Med 1983;309:147. 78. Child CG III, Turcotte JG. Surgery and portal hypertension. In: Child CG III, ed. The liver and portal hypertension. Philadelphia, WB Saunders, 1964:50. 79. Dickson ER, Grambsch PM, Fleming TR, et al. Prognosis in primary biliary cirrhosis: model for decision making. Hepatology 1989;10:1. 80. O’Grady JG, Alexander GJM, Hayllar KM, et al. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989;97:439. 81. Hofmann AF. The continuing importance of bile acids in liver and intestinal disease. Arch Intern Med 1999;159:2647.
247
Section II. Approach to the Patient with Liver Disease
82. Trauner M, Meier PJ, Boyer JL. Molecular pathogenesis of cholestasis. N Engl J Med 1998;339:1217. 83. Berk PD, Javitt NB. Hyperbilirubinemia and cholestasis. Am J Med 1978;64:311. 84. Poupon RE, Chretien Y, Poupon R, et al. Serum bile acids in primary biliary cirrhosis: effect of ursodeoxycholic acid therapy. Hepatology 1993;17:599. 85. Skrede S, Blomhoff JP, Elgjo K, et al. Biochemical tests in evaluation of liver function. Scand J Gastroenterol 1973;19(Suppl):37. 86. Kirsch R, Frith L, Black E, et al. Regulation of albumin synthesis and catabolism by alteration of dietary protein. Nature 1968;217:578. 87. Keshgegian AA. Hypoalbuminemia associated with diffuse hypergamma-globulinemia in chronic diseases: lack of diagnostic specificity. Am J Clin Pathol 1984;81:477. 88. Ballmer PE, Walshe D, McNurlan MA, et al. Albumin synthesis rates in cirrhosis: correlation with Child–Turcotte classification. Hepatology 1993;18:292. 89. Rothschild MA, Oratz M, Zimmon D, et al. Albumin synthesis in cirrhotic subjects with ascites studied with carbonate 14C. J Clin Invest 1969;48:344. 90. Oratz M, Rothschild M A, Schreiber SS. Alcohol, amino acids, and albumin synthesis. III. Effects of ethanol, acetaldehyde, and 4-methylpyrazole. Gastroenterology 1978;74: 672–676. 91. Suttie JW, Jackson CM. Prothrombin structure, activation and biosynthesis. Physiol Rev 1977;57:1. 92. Quick AJ. Hemorrhagic diseases and thrombosis, 3rd edn. Philadelphia: Lea & Febiger, 1966: 391. 93. Maddrey WC, Boitnott JK, Bedine MS, et al. Corticosteroid therapy of alcoholic hepatitis. Gastroenterology 1978;75:193. 94. Pugh RN, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646. 95. Ratnoff OD. Hemostatic mechanisms in liver disease. Med Clin North Am 1963;47:721. 96. Stenflo J, Ferlund P, Egan W, et al. Vitamin K dependent modifications of glutamic acid residues in prothrombin. Proc Natl Acad Sci USA 1974;71:2730. 97. Lord JW, Andrus W deW. Differentiation of intrahepatic and extrahepatic jaundice. Response of the plasma prothrombin to intramuscular injection of menadione (2-methyl-1,4naphthoquinone) as a diagnostic aid. Arch Intern Med 1941;68:199. 98. Schmidt LE, Knudsen TT, Dalhoff K, et al. Effect of acetylcysteine on prothrombin index in paracetamol poisoning without hepatocellular injury. Lancet 2002;360:1151–1152. 99. Blanchard RA, Furie BC, Jorgensen M, et al. Acquired vitamin K-dependent carboxylation deficiency in liver disease. N Engl J Med 1981;305:242. 100. Weitz IC, Liebman HA. Des-g-carboxy (abnormal) prothrombin and hepatocellular carcinoma: a critical review. Hepatology 1993;18:990. 101. Ishii M, Gama H, Chida N, et al. Simultaneous measurements of serum a-fetoprotein and protein induced by vitamin K absence for detecting hepatocellular carcinoma. Am J Gastroenterol 2000;95:1036. 102. Nomura F, Ishijima M, Kuwa K, et al. Serum des-gammacarboxy prothrombin levels determined by a new generation of sensitive immunoassays in patients with small-sized hepatocellular carcinoma. Am J Gastroenterol 1999;94:650. 103. Okuda H, Nakanishi T, Takatsu K, et al. Measurement of serum levels of des-g-carboxy prothrombin in patients with hepatocellular carcinoma by a revised enzyme immunoassay kit with increased sensitivity. Cancer 1999;85:812. 104. Goldberg DM. 5¢-Nucleotidase: recent advances in cell biology, methodology and clinical significance. Digestion 1973;8:87.
248
105. Cassidy WM, Reynolds TB. Serum lactic dehydrogenase in the differential diagnosis of acute hepatocellular injury. J Clin Gastroenterol 1994;19:118. 106. Van Waes L, Lieber CS. Glutamate dehydrogenase: a reliable marker of liver cell necrosis in the alcoholic. Br Med J 1977;2:1508. 107. Jenkins WJ, Rosalki SB, Foo Y, et al. Serum glutamate dehydrogenase is not a reliable marker of liver cell necrosis in alcoholics. J Clin Pathol 1982;35:207. 107. Afdhal NH. Diagnosing fibrosis in hepatitis C: is the pendulum swinging from biopsy to blood tests? Hepatology 2003;37:972–974. 108. Dienstag J. The role of liver biopsy in chronic hepatitis C. Hepatology 2002;36:S152–S160. 109. Bravo AA, Sheth SG, Chopra S. Liver biopsy. N Engl J Med 2001;344:495–500. 110. Regev A, Berho M, Jeffers LJ, et al. Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002;97:2614–2618. 111. Colloredo G, Guido M, Sonzogni A, et al. Impact of liver biopsy size on histological evaluation of chronic viral hepatitis: the smaller the sample, the milder the disease. J Hepatol 2003;39:239–244. 112. Bedossa P, Dargère D, Paradis V. Sampling variability of liver fibrosis in chronic hepatitis C. Hepatology 2003;38:1449–1457. 113. Poynard T, Ratziu V, Bedossa P. Appropriateness of liver biopsy. Can J Gastroenterol 2000;14:543–548. 114. Poynard T, McHutchison J, Manns M, et al. Biochemical surrogate markers of liver fibrosis and activity in a randomized trial of peginterferon alfa-2b and ribavirin. Hepatology 2003;38:481–492. 115. Poynard T, Munteanu M, Imbert-Bismut F, et al. Prospective analysis of discordant results between biochemical markers and biopsy in patients with chronic hepatitis C. Clin Chem 2004;50:1344–1355. 116. Gebo KA, Herlong HF, Torbenson MS, et al. Role of liver biopsy in management of chronic hepatitis C: A systematic review. Hepatology 2002;36:S161–S172. 117. Imbert-Bismut F, Ratziu V, Laurence Pieroni L, et al., for the MULTIVIRC group. Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study. Lancet 2001;357:1069–1075. 118. Poynard T, Imbert-Bismut F, Ratziu V, et al. Biochemical markers of liver fibrosis in patients infected by Hepatitis C Virus: Longitudinal validation in a randomized trial. J Viral Hepatitis 2002;9:128–133. 119. Myers RP, Ratziu V, Imbert-Bismut F, et al. Biochemical markers of liver fibrosis: a comparison with historical features in patients with chronic hepatitis C. Am J Gastroenterol 2002;97:2419–2425. 120. Myers RP, Benhamou Y, Imbert-Bismut F, et al. Serum biochemical markers accurately predict liver fibrosis in HIV and hepatitis C virus-coinfected patients. AIDS 2003;17:1–5. 121. Myers RP, de Torres M, Imbert-Bismut F, et al. Biochemical markers of fibrosis in patients with chronic hepatitis C: a comparison with prothrombin time, platelet count and the age–platelet index. Dig Dis Sci 2003;48:146–153. 122. Thabut D, Simon M, Myers RP, et al. Non invasive prediction of fibrosis in patients with chronic hepatitis C. [Letter] Hepatology 2003;37:1220–1221. 123. Le Calvez S, Thabut D, Messous D, et al. Fibrotest has higher predictive values than APRI for fibrosis diagnosis in patients with chronic hepatitis C. [Letter] Hepatology 2004;39: 862–863. 124. Halfon P, Imbert-Bismut F, Messous D, et al. A prospective assessment of the inter-laboratory variability of biochemical markers of fibrosis (FibroTest) and activity (ActiTest) in patients with chronic liver disease. Comp Hepatol 2002;1:3–7.
Chapter 14 LABORATORY TESTING FOR LIVER DISEASE
125. Rossi E, Adams L, Prins A, et al. Validation of the FibroTest biochemical markers score in assessing liver fibrosis in hepatitis C patients. Clin Chem 2003;49:450–454. 126. Imbert-Bismut F, Messous D, Thibaut V, et al. Intra-laboratory analytical variability of biochemical markers of fibrosis (Fibrotest) and activity (Actitest) and reference ranges in healthy blood donors. Clin Chem Lab Med 2004; 42:323–333. 127. Munteanu M, Messous D, Thabut D, et al. Intra-individual fasting versus postprandial variation of biochemical markers of liver fibrosis (Fibrotest) and activity (Actitest). Comp Hepatol 2004;3:3. 128. The METAVIR cooperative group. Inter- and intra-observer variation in the assessment of liver biopsy of chronic hepatitis C. Hepatology 1994;20:15–20. 129. Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. The METAVIR Cooperative Study Group. Hepatology 1996;24:289–293. 130. Poynard T, Imbert-Bismut F, Ratziu V, et al. Fibrotest even better than liver biopsy? Clin Chem 2003. Electronic letter http://www.clinchem.org/cgi/eletters/49/3/450. Response: (21 March 2003) 131. Poynard T. Cost effectiveness of pegylated interferon alpha 2b and ribavirin combination in chronic hepatitis C. [Letter] Gut 2003;52:1532. 132. Callewaert N, Van Vlierberghe H, Van Hecke A, et al. Noninvasive diagnosis of liver cirrhosis using DNA sequencerbased total serum protein glycomics. Nature Med 2004; 10:1–6. 133. Poynard T, Imbert-Bismut F, Ratziu V. Serum markers of liver fibrosis. Hepatol Rev 2004;1:25–33. 134. Myers RP, Tainturier MH, Ratziu V, et al. Prediction of liver histological lesions with biochemical markers in patients with chronic hepatitis B. J Hepatol 2003;39:222–230. 135. McIntyre N. Plasma lipids and lipoproteins in liver disease. Gut 1978;19:526. 136. Day RC, Harry DS, Owen JS, et al. Plasma lecithin: cholesterol–acyltransferase activity and the lipoprotein abnormalities of liver disease. Scand J Clin Lab Invest 1978;150:223. 137. Freeman M, Kuiken L, Ragland JB, et al. Hepatic triglyceride lipase deficiency in liver disease. Lipids 1977;12:443. 138. Fukushima N, Yamamoto K, Ozaki I, et al. Apolipoprotein A-I, E, C-III and LDL-receptor mRNA expression in liver diseases. Nippon Rinsho 1993;51:407–413. 139. Obici L, Palladini G, Giorgetti S, et al. Liver biopsy discloses a new apolipoprotein A-I hereditary amyloidosis in several unrelated Italian families. Gastroenterology 2004;126:1416–4122. 140. Switzer S. Plasma lipoproteins in liver disease: I. Immunologically distinct low-density lipoproteins in patients with biliary obstruction. J Clin Invest 1967;46:1855. 141. Seidel D, Alauprovic P, Furman RH. A lipoprotein characterizing obstructive jaundice: I. Method for quantitative separation and identification of lipoproteins in jaundiced subjects. J Clin Invest 1969;48:1211. 142. Sörös P, Böttcher J, Maschek H, et al. Lipoprotein-X in patients with cirrhosis: its relationship to cholestasis and hypercholesterolemia. Hepatology 1998;28:1199. 143. Jahn CE, Schaefer EJ, Taam LA, et al. Lipoprotein abnormalities in primary biliary cirrhosis: association with hepatic lipase inhibition as well as altered cholesterol esterification. Gastroenterology 1985;89:1266. 144. Pawlotsky JM, Bouvier-Alias M, Hezode C, et al. Standardization of hepatitis C virus RNA quantification. Hepatology 2000;32:654–659. 145. Pawlotsky JM. Molecular diagnosis of viral hepatitis. Gastroenterologys 2002;122:1554–1568.
146. Brechot C, Thiers V, Kremsdorf D, et al. Persistent hepatitis B virus infection in subjects without hepatitis B surface antigen: clinically significant or purely ‘occult’? Hepatology 2001;34:194–203. 147. Bassett ML, Halliday JW, Ferris RA, et al. Diagnosis of hemochromatosis in young subjects: predictive accuracy of biochemical screening tests. Gastroenterology 1984;87: 628–633. 148. Bacon BR. Hemochromatosis: diagnosis and management. Gastroenterology 2001;120:718–725. 149. Tavill AS. Diagnosis and management of hemochromatosis. Hepatology 2001;33:1321–1313. 150. Alvarez F, Berg PA, Bianchi FB, et al. International Autoimmune Hepatitis Group report: Review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999;1:929–938. 151. Czaja AJ, Freese DK. Diagnosis and treatment of autoimmune hepatitis. Hepatology 2002;36:479–497. 152. Van de Water J, Cooper A, Surh SC, et al. Detection of autoantibodies ro recombinant mitochondrial proteins in patients with primary biliary cirrhosis. N Engl J Med 1989;320:1377–1380. 153. Wiesner RH, LaRusso NF, Ludwig J, et al. Comparison of the clinicopathological features of primary sclerosing cholangitis and primary biliary cirrhosis. Gastroenterology 1985;88:108–114. 154. MacLachlan MJ, Rodnan GP, Cooper WM, et al. Chronic active (‘lupoid’) hepatitis: a clinical, serological, and pathological study of 20 patients. Ann Intern Med 1965;62:425. 155. McFarlane IG, Smith HM, Johnson PJ, et al. Hepatitis C virus antibodies in chronic active hepatitis: pathogenetic factor or false-positive result? Lancet 1990;335:754. 156. Mitchel LS, Jeffers LJ, Reddy KR, et al. Detection of hepatitis C virus antibody by first and second generation assays and polymerase chain reaction in patients with autoimmune chronic active hepatitis types I, II, and III. Am J Gastroenterol 1993;88:1027. 157. Demeulenaere L, Wieme RJ. Special electrophoretic anomalies in the serum of liver patients: a report of 1145 cases. Am J Dig Dis 1961;6:661. 158. Rowntree LG, Hurwitz SH, Bloomfield AL. An experimental and clinical study of the value of phenoltetrachlorphthalein as a test of hepatic function. Bull Johns Hopkins Hosp 1913;24:327. 159. Rosenthal SM, White EC. Clinical application of bromsulphthalein test for hepatic function. JAMA 1925;84:1112. 160. Cherrick GR, Stein SW, Leevy CM, et al. Indocyanine green: observations on its physical properties, plasma decay and hepatic extraction. J Clin Invest 1960;39:592. 161. Leevy CM, Smith F, Longueville J, et al. Indocyanine green clearance as a test for hepatic function: evaluation by dichromatic ear densitometry. JAMA 1967;200:236. 162. Ishigami Y, Masuzawa M, Miyoshi E, et al. Clinical applications of ICG finger monitor in patients with liver disease. J Hepatol 1993;19:232. 163. Therapondos G, Plevris JN, Stanley AJ, et al. Cerebral near infrared spectroscopy for the measurement of indocyanine green elimination in cirrhosis. Aliment Pharmacol Ther 2000;14:923. 164. Jiao LR, El-Desoky AA, Seifalian AM, et al. Effect of liver blood flow and function on hepatic indocyanine green clearance measured directly in a cirrhotic animal model. Br J Surg 2000;87:568. 165. Poulsen HE, Loft S. Antipyrine as a model drug to study hepatic drug-metabolizing capacity. J Hepatol 1988;6:374. 166. Opekun AR Jr, Klein PD, Graham DY. [13C]Aminopyrine breath test detects altered liver metabolism caused by low-dose oral contraceptives. Dig Dis Sci 1995;40:2417.
249
Section II. Approach to the Patient with Liver Disease
167. Jover R, Carnicer F, Sánchez-Pajá J, et al. Salivary caffeine clearance predicts survival in patients with liver cirrhosis. Am J Gastroenterol 1997;92:1905. 168. Ballmer PE, Reichen J, McNurlan MA, et al. Albumin but not fibrinogen synthesis correlates with galactose elimination capacity in patients with cirrhosis of the liver. Hepatology 1996;24:53.
250
169. Rypins EB, Henderson JM, Fulenwider JT, et al. A tracer method for measuring rate of urea synthesis in normal and cirrhotic subjects. Gastroenterology 1980;78:1419. 170. Ercolani G, Grazi GL, Callivà R, et al. The lidocaine (MEGX) test as an index of hepatic function: its clinical usefulness in liver surgery. Surgery 2000;127:464.
Section II. Approach to the Patient with Liver Disease
15
CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING Rizwan Aslam, Yee-Li Sun and Judy Yee Abbreviations BCS Budd–Chiari syndrome CT computed tomography CTAP CT arterioportography CTHA CT hepatic arteriography DN dysplastic nodules EHE epithelioid hemangioendothelioma 18 F-FDG 18-fluorine 2-deoxy-2-D-glucose FLC fibrolamellar carcinoma
FNH focal nodular hyperplasia Gd-DTPA gadolinium diethylene triaminepentaacetic acid HA hepatic angiosarcoma HCC hepatocellular carcinoma HU hounsfield units IOUS intraoperative ultrasound Mn-DPDP mangafodipir trisodium
INTRODUCTION Diagnostic imaging today plays an important role in the accurate evaluation of the liver and its pathology, and should be part of the routine clinical assessment of patients, together with a thorough history, physical examination, and laboratory tests. Imaging is used to detect changes in the normal hepatic architecture, which may relate to underlying pathology. Some imaging modalities are more sensitive and specific than others for different pathologies. The main modalities used are ultrasound (US), computed tomography (CT), nuclear medicine (NM), positron emission tomography (PET), and magnetic resonance imaging (MRI). In this chapter we explain some basic concepts of radiology with regard to the efficacy of the various available imaging techniques. This approach should help the clinician select the appropriate diagnostic modality to investigate a patient’s symptoms and clinical signs. This evidence-based approach should help to improve the efficiency and cost-effectiveness of hepatic imaging.
FOCAL BENIGN LIVER LESIONS LESIONS OF HEPATOCELLULAR ORIGIN Hepatocellular Adenoma Hepatocellular adenoma is a relatively rare benign lesion often found in women of childbearing age. The tumor is associated with oral contraceptive and anabolic steroid use. Unlike other benign lesions of the liver, such as focal nodular hyperplasia (FNH), adenomas have a tendency to outgrow their vascular supply, leading to hemorrhage, necrosis, and possible rupture. These lesions are often resected because of the risk of complications as well as of malignant transformation. The sonographic features of adenomas are non-specific. These lesions most commonly appear as hyperechoic (bright) masses owing
MRI NM PET SPG SPIO 99m Tc US
magnetic resonance imaging nuclear medicine positron emission tomography spoiled gradient echo sequence superparamagnetic iron oxide particles Technetium-99m ultrasound
to their high lipid content. However, the echogenicity of the lesions can be variable, depending upon the presence and age of internal hemorrhage. Color Doppler is a useful technique for distinguishing these lesions from those of FNH. Hepatic adenomas will often demonstrate subcapsular vessels as well as intratumoral veins that are not seen with FNH.1 Hepatic adenomas have a variable appearance on non-contrast CT. Adenomas may be hypodense (dark) relative to the surrounding liver parenchyma, owing to the presence of intratumoral fat. They may also appear hyperdense (bright) when acute intratumoral hemorrhage is present. Most adenomas are isodense to surrounding liver tissue because they consist mainly of normal hepatocytes. Hepatic adenomas are hypervascular lesions and will exhibit intense enhancement during the arterial phase of contrast-enhanced CT. Small lesions often enhance homogeneously with rapid washout owing to arteriovenous shunting (Figure 15-1). Larger lesions tend to be more heterogeneous and may exhibit initial peripheral enhancement that moves centrally (because of the vascular supply that originates from the subcapsular feeding vessels), thus simulating the appearance of a hemangioma.2 The MR appearance of hepatic adenomas has also been reported to be highly variable. These lesions tend to have a heterogeneous appearance (93%)3 because they contain hemorrhage, fat, and areas of necrosis. On T1-weighted sequences lesions with a high fat content or acute hemorrhage appear hyperintense (bright), whereas those with large areas of necrosis or old hemorrhage will appear hypointense (dark). On T2-weighted sequences it has been reported that 47–74% of these lesions are hyperintense4 (Figure 15-2). In addition, studies have shown that approximately 30% of these lesions will have a peripheral hypointense rim on both T1- and T2weighted images, corresponding to the presence of a fibrous capsule.5 Dynamic post-gadolinium MR imaging typically reveals early arterial enhancement of these lesions, similar to the findings on CT.
251
Section II. Approach to the Patient with Liver Disease
A
C
Nuclear medicine studies, in particular scintigraphy using technetium (99mTc) sulfur colloid, may help differentiate hepatic adenomas from FNH. Adenomas appear as focal defects on (99mTc) sulfur colloid scans, whereas FNH appear as hot (bright) spots because they contain a high concentration of reticuloendothelial (Kupffer) cells that will take up the agent.
Focal Nodular Hyperplasia Focal nodular hyperplasia (FNH) is the second most common benign liver neoplasm after hemangioma. It occurs more frequently in
252
B
Figure 15-1. Multiphasic CT appearance of an adenoma. Contrast-enhanced axial CT images in different phases of enhancement reveal a pattern consistent with a hepatic adenoma. (A) A lobular lesion is seen in the posterior right lobe (arrow) that displays early and rapid homogeneous enhancement. (B, C) Images obtained during the subsequent portal–venous phase demonstrates rapid washout of contrast that is often seen in adenomas due to high levels of AV shunting. (C) On a further delayed image, the lesion is no longer visible, and has become isodense to the surrounding liver parenchyma.
women and is often discovered incidentally, although patients with FNH can present with symptoms such as pain. The typical sonographic appearance of FNH is a well-defined homogeneous mass. The lesion often has subtle differences in echogenicity compared to the surrounding liver parenchyma and may be difficult to detect. The presence of a large central scar (present in approximately 20% of cases) makes these lesions more detectable and appears as a linear hyperechoic (bright) band. Color Doppler imaging helps to demonstrate these hypervascular lesions, with blood flow radiating peripherally from a large central feeding vessel.6
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING Figure 15-2. MR appearance of a large heterogeneous adenoma. (A) In this T1-weighted image a large heterogeneous, but overall hypointense (dark) mass (M) is seen in the left hepatic lobe. (B) Enhancement is noted on the post-gadolinium images. (C) In the T2-weighted axial image the lesion is again seen to be heterogeneous with areas of hyperintensity (brightness). (D) The T2-weighted coronal image demonstrates the extent of the mass that was confirmed to be an adenoma.
A
B
C
D
FNH typically appears as an iso- or slightly hypodense (dark) solid mass with lobulated contours on unenhanced CT scan. Dynamic contrast-enhanced helical CT demonstrates a characteristic enhancement pattern. The lesions tend to enhance immediately and intensely, with the exception of the region of the central scar on arterial phase images (Figure 15-3). There is rapid washout of contrast, so that the lesion becomes isodense to the surrounding liver in the portal venous phase, with the exception of the central scar, which shows delayed enhancement6 (Figure 15-3). Persistent enhancement of the central scar on the delayed images helps to differentiate FNH from hepatocellular carcinomas. One series identified late enhancement of the central scar in approximately 80% of cases.7 MR imaging has a sensitivity of 70% and a specificity of 98% for the detection of FNH.6 T1-weighted sequences typically demonstrate an iso- or hypointense mass with a hypointense central scar. T2-weighted sequences demonstrate a mildly hyperintense lesion with a more hyperintense central scar (Figure 15-4). Following the administration of gadolinium, the enhancement characteristics of FNH parallel those seen on CT, with homogeneous enhancement of the majority of the lesion on the arterial phase and late enhancement of the central scar on the delayed phase images, while the remainder of the lesion becomes isointense (Figure 15-4). Liverspecific contrast agents such as superparamagnetic iron oxide
(SPIO) can also be used to distinguish FNH. SPIO is taken up by Kupffer cells, which are present in higher concentrations in FNH than in other hepatic lesions. On T2-weighted images with SPIO enhancement FNH lesions will take up the contrast agent (more so than the surrounding liver parenchyma), resulting in hypointensity of the lesion due to loss of signal.6 Mangafodipir trisodium (Mn-DPDP) can also be used to characterize FNH. Because FNH is also composed of normal hepatocytes, Mn-DPDP is taken up by the lesion, leading to hyper- or isointensity on T1-weighted sequences relative to normal parenchyma.3 Several atypical findings of FNH can lead to diagnostic dilemmas. A hypointense (dark) central scar on T2 sequences or on contrastenhanced T1 sequences can rarely be found in FNH. This may be confused with the collagenous scars that occur in hepatocellular carcinoma (HCC), fibrolamellar carcinoma and intrahepatic cholangiocarcinomas.3 FNH lesions may also occasionally be surrounded by a pseudocapsule. This may show enhancement on delayed images that can mimic the enhancement of the capsule of HCC. However, unlike the capsule of HCC, which consists mainly of fibrotic elements and has low intensity (dark) on both T1- and T2-weighted sequences, the pseudocapsule seen in FNH results from an inflammatory reaction due to compression of surrounding liver parenchyma and is thus usually hyperintense on T2-weighted sequences.6
253
Section II. Approach to the Patient with Liver Disease
A
B
Figure 15-3. CT appearance of FNH. (A) An arterial-phase contrast-enhanced CT scan reveals an enhancing lesion (arrow) with a central scar in the left hepatic lobe consistent with FNH. (B) The delayed-phase image demonstrates filling-in of the central scar.
Figure 15-4. MR appearance of FNH. (A) An axial T1-weighted MR image obtained during the arterial phase of gadolinium enhancement reveals two early enhancing hepatic lesions (arrows), the larger of which demonstrates a hypointense (dark) central scar. (B) A T1-weighted image obtained during the portal phase of enhancement reveals homogeneous enhancement of the lesions and the surrounding liver, making the lesion less distinct. (C) This delayed-phase T1 image demonstrates persistent enhancement of the central scar (arrow). (D) An axial T2-weighted image demonstrates an isointense lesion with a bright central scar (arrow) which is consistent with FNH.
254
A
B
C
D
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
Nodular Regenerative Hyperplasia Nodular regenerative hyperplasia is a benign entity that has been associated with a variety of systemic diseases (myeloproliferative, lymphoproliferative, and collagen vascular disorders) as well as drugs (steroids and anticancer agents).3 It is typically a diffuse process, although it can manifest with distinct areas of large, prominent nodules that can mimic focal masses. Histologically, nodular regenerative hyperplasia is made up of hyperplastic cells resembling normal hepatocytes that organize in nodules. The absence of significant fibrosis in these lesions is an important feature that helps to distinguish them from FNH and cirrhosis. Clinically, nodular regeneration is often associated with portal hypertension. Nodular regenerative hyperplasia appears as isoechoic nodules in the liver on ultrasound, as they are comprised of normal hepatic tissue. The lesions can also be hypo- to anechoic when there is associated internal hemorrhage. The CT appearance can be normal, particularly when there are small nodules. Regenerative nodules may also appear as focal hypodense nodules on CT, although with internal hemorrhage they may appear hyperdense. There have been few reports of the MR appearance of nodular regenerative hyperplasia. However, these lesions have been described as isointense on T2 and with foci of hyperintensity on T1, thought to be due to hemorrhage.3
LESIONS OF BILIARY ORIGIN Hepatic (Biliary) Cysts Hepatic cysts are common benign lesions that occur in 1–14% of the population.3 They may be solitary or multiple; in the latter case they may be associated with polycystic kidney disease (Figure 155). Although they are typically asymptomatic, large lesions can lead to clinical symptoms. Ultrasound is highly accurate in identifying cystic lesions of the liver. Simple hepatic cysts are typically seen as round anechoic lesions with smooth borders and posterior acoustic enhancement. If the cysts become complicated by infection or hemorrhage, they can develop septations and become more heterogeneous. Cysts in the liver typically appear on CT as homogeneous, hypodense, well delineated lesions with no perceptible wall. They do not exhibit enhancement following contrast administration (Figure 156). Small cysts, <1.5 cm in diameter, may be difficult to characterize on CT owing to partial voluming effects. Hepatic cysts exhibit the same characteristics as water on MRI and are low intensity on T1 (dark) and high intensity on T2 (bright) (Figure 15-7). There is no enhancement following the intravenous administration of gadolinium.
Figure 15-5. Hepatic cysts in adult polycystic kidney disease. (A,B) Axial contrast-enhanced CT images reveal multiple cystic lesions throughout the liver. (C,D) In these images obtained at a lower level, the patient is noted to have diffuse cystic change involving the left kidney as well. This patient was known to have adult polycystic kidney disease.
A
B
C
D
255
Section II. Approach to the Patient with Liver Disease
A
B
Figure 15-6. Contrast CT appearance of hepatic cysts. (A) In this non-contrast enhanced axial CT scan multiple low-density (dark) liver lesions are seen that are compatible with hepatic simple cysts. (B) On the post-contrast images the surrounding liver parenchyma enhances but the dark lesions do not. This is characteristic of simple cysts.
A
B
Figure 15-7. MR appearance of hepatic cysts. (A) This axial T1-weighted MR image reveals a dark lesion (arrow) in the right lobe of the liver that is consistent with a cyst. (B) The corresponding T2-weighted image at a slightly lower level demonstrates that the lesion is bright on this sequence, also consistent with the features of a simple cyst.
Biliary Cystadenoma Biliary cystadenomas are rare, slow-growing lesions that are benign, but can recur after resection. Transformation to malignant cystadenocarcinoma may occur. Biliary cystadenomas appear on ultrasound as anechoic lesions with internal septations giving them a multilocular cystic appearance. These lesions appear cystic on unenhanced CT as well. Unlike simple cysts, a surrounding wall and internal septations and soft
256
tissue nodules are often identified that typically enhance (Figure 158). The cystic components of biliary cystadenomas appear on MRI as low intensity (dark) on T1 images and high intensity (bright) on T2-weighted sequences. Septations appear hypointense on T2. However, there is variability in these findings, depending upon the protein content of the fluid and the presence or absence of soft tissue components.8 The cyst wall, septations, and nodular components tend to enhance on MRI also.
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
Figure 15-8. CT appearance of biliary cystadenoma. (A, B) Two axial contrast-enhanced CT images from different patients reveal hypodense (dark) cystic lesions with septations and focal areas of soft tissue density. The soft tissue areas enhance with intravenous contrast, consistent with a biliary cystadenoma.
Bile Duct Adenoma (Bile Duct Hamartoma, Von Meyenburg Complex) Bile duct adenomas are thought to arise from remnants of embryonic bile ducts. They are typically discovered incidentally and are usually stable, benign lesions. The imaging findings of bile duct adenomas are non-specific. When the lesions are multifocal, they can simulate metastases or abscesses and may therefore require pathologic diagnosis.9,10 The sonographic appearance is of small nodules of variable echogenicity. On CT, the lesions typically manifest as scattered small nodules (5–10 mm) that are hypo- to isodense and demonstrate minimal if any enhancement11 (Figure 15-9). On MR, the lesions tend to be hypointense on T1 and hyperintense on T2, with no specific pattern of enhancement after intravenous gadolinium administration3 (Figure 15-10).
LESIONS OF MESENCHYMAL ORIGIN Hemangioma Hemangiomas are the most common benign tumor of the liver and the second most common liver tumor overall after metastases. The lesions are often less than 4 cm in diameter and are commonly located peripherally.3 Hemangiomas appear on ultrasound as solitary, lobulated masses with a well demarcated border. The lesions are typically homogeneous and hyperechoic (bright), and characteristically exhibit increased through-transmission. Color Doppler studies often highlight peripheral feeding vessels with no evidence of flow centrally.3 The sensitivity of ultrasound for detecting hemangiomas has been reported to range from 60 to 75%, with a specificity of 60–80%.12
Hemangiomas are seen as well defined hypodense lesions on unenhanced CT scans. Dynamic multiphase contrast-enhanced CT is typically used for diagnosis. During the early contrast phase hemangiomas exhibit a characteristic pattern of peripheral globular, nodular enhancement which has been found to be 88% sensitive and 88–100% specific for differentiating these benign lesions from hypervascular metastases13,14 (Figure 15-11). On subsequent images hemangiomas exhibit peripheral to central filling with contrast. The lesions remain hyperdense to the surrounding tissue owing to their lack of intratumoral shunting on delayed images. Smaller lesions usually show complete enhancement, whereas larger lesions are more heterogeneous and may have central areas that remain hypodense on delayed images. Calcifications have been reported in 5–20% of lesions.3,11 Overall, the sensitivity of CT for the detection of hemangiomas ranges between 75 and 85%, with a specificity of 75–90%.12 MR imaging of hemangiomas has been reported to have an even higher sensitivity (85–95%) and specificity (85–95%) than CT.12 The lesions typically have a low T1 signal intensity and a high T2 intensity. The markedly high T2 intensity and the lobulations within the tumor are helpful in differentiating hemangiomas from malignant liver tumors (Figure 15-12). In addition, with the administration of gadolinium, the lesions show peripheral puddling with centripetal filling that parallels the CT enhancement pattern. Smaller lesions often enhance completely, whereas larger ones may have a persistent unenhanced central region corresponding to central fibrosis and/or necrosis. Specific liver-related MR contrast agents can also be used to evaluate hemangiomas. Nuclear medicine studies using 99mTc-labeled red blood cells have also been reported to be useful, particularly in the differentiation of
257
Section II. Approach to the Patient with Liver Disease
A
B
Figure 15-9. Biliary hamartomas on CT. (A,B) Contrast-enhanced axial CT images demonstrate multiple tiny hypodense (dark), non-enhancing lesions scattered throughout the liver. This appearance is consistent with that of a bile duct hamartoma.
A
B
Figure 15-10. Biliary hamartomas on MRI. (A) This axial T1-weighted post-contrast image of the liver reveals multiple small hypointense (dark) lesions throughout the liver parenchyma, consistent with biliary hamartomas. (B) The same lesions are bright on T2.
258
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
C
D
Figure 15-11. Hemangioma on CT. (A) A non-contrast axial CT through the abdomen reveals a hypodense (dark) lesion (arrow) in the right hepatic lobe. (B) During the arterial phase the lesion demonstrates the classic peripheral lobular enhancement pattern of a hepatic hemangioma. (C) In the portal phase image, the lesion is shown to be more homogeneously enhancing. (D) In the delayed phase images the lesion demonstrates complete homogeneous enhancement that persists after the washout of contrast from the surrounding parenchyma.
259
Section II. Approach to the Patient with Liver Disease
A
B
C
D
Figure 15-12. MR appearance of hemangioma. (A) This axial post-contrast MR image obtained during the arterial phase demonstrates a peripherally enhancing lesion in the right hepatic lobe (arrow). (B,C) Images obtained during the portal–venous phases demonstrate more homogeneous enhancement of the lesion with time. This pattern is again consistent with that of a hemangioma. (D) The T2-weighted image demonstrates increased hyperintensity (bright appearance), also in keeping with a hemangioma.
giant hemangiomas (>8 cm) from other solid hepatic masses. Giant hemangiomas may be difficult to diagnose based on ultrasound and CT alone because of the variable patterns that can be seen. However, 99mTc-labeled red blood cell SPECT scanning has been shown to be highly specific (Figure 15-13). In a series examining
260
ultrasound versus RBC scanning for giant liver lesions, a total of 23 lesions were detected by ultrasound. Eighteen were detected by RBC scanning and found to have two patterns – homogeneous increased pooling, or peripheral uptake with a lack of central pooling. The five giant liver lesions seen by ultrasound, but not by
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
C
red blood cell scanning, were pathologically proven to be lesions other than hemangiomas,15 suggesting poor specificity of ultrasound in comparison to RBC scanning.
(Mesenchymal) Hamartoma Hamartomas are comprised predominantly of cystic elements and mesenchymal septa. The relative proportions of these two elements play a key role in the imaging appearance. Hamartomas appear on ultrasound as either large anechoic cystic regions surrounded by internal septa (cystic predominant) or small cystic regions with thicker septa (mesenchymal predominant). Hamartomas are typically seen as large heterogeneous masses on non-contrast CT. The stromal elements appear mildly hypodense compared to the surrounding liver parenchyma, whereas the cystic elements show the typical low density of fluid. With contrast administration, the septa (stromal elements) enhance but the cystic portions do not.11 MR findings can vary considerably, depending on the protein content of the fluid and the relative proportion of stromal to cystic elements.
B
Figure 15-13. Ultrasound and nuclear medicine studies of a hemangioma. (A) Ultrasound reveals a homogeneous hyperechoic solitary mass with a welldemarcated border. (B) Contrast-enhanced CT demonstrates the mass (black arrow) in the left lobe of the liver. (C) Nuclear medicine red cell scan using 99mTclabeled red blood cells reveals progressively increased signal in the lesion (white arrow). This is helpful in confirming the diagnosis of hemangioma.
endothelioma can be quite aggressive and cause significant morbidity. Patients typically present with an abdominal mass that resolves spontaneously. However, the lesion can be complicated by congestive heart failure, platelet dysfunction, and hemorrhage.3 Malignant transformation has also been described. Infantile hemangioendothelioma presents on ultrasound as a complex mass with large draining hepatic veins (Figure 15-14). The lesions may be solitary, but are usually multiple and can be hypo- or hyperechoic.11 Multiple hypodense nodules are seen on CT scans. The low density is due to cystic degeneration and hemorrhage. Following the administration of intravenous contrast, these lesions typically show early peripheral enhancement with variable delayed central uptake that can appear similar to the enhancement pattern of hemangiomas in adults. The MR appearance of these lesions is highly variable, with heterogeneity on both T1 and T2 owing to the presence of fibrosis, necrosis, and hemorrhage. Dynamic gadoliniumenhanced imaging reveals enhancement characteristics similar to those of hemangiomas.3
Infantile Hemangioendothelioma
Lymphangioma
Infantile hemangioendothelioma is a benign lesion of the liver that occurs primarily in infancy, usually between the ages of 1 and 6 months. Although it is classified as benign, infantile hemangio-
Lymphangiomas are masses made up of dilated lymphatic channels that compress the normal hepatic parenchyma. They often occur as a part of a systemic syndrome involving multiple organs, including
261
Section II. Approach to the Patient with Liver Disease
A
C
bone, soft tissues, lung, and brain. They appear as well delineated cystic lesions in the liver that do not enhance with contrast administration (Figure 15-15).
Angiomyolipoma/Myelolipoma Angiomyolipomas (AMLs) are lesions composed of angiomyomatous and fatty elements. In the abdomen they most commonly occur in the kidneys and adrenals, but rarely can be found in the liver. Angiomyolipomas may occur as isolated cases, or they can be found in association with tuberous sclerosis (6–10% of cases).16
262
B
Figure 15-14. Infantile hemangioendothelioma. (A) This ultrasound image reveals multiple hypoechoic lesions throughout the liver in a 7-week-old child with infantile hemangioendothelioma. (B) Another ultrasound image illustrates large draining hepatic veins, also associated with this condition. (C) Contrastenhanced MRI reveals a markedly enlarged abnormal liver with multiple lobular lesions.
Angiomyolipomas are often heterogeneous on ultrasound appearance, with areas of increased echogenicity.3,16 AMLs are also heterogeneous on CT, with the angiomyomatous elements having higher density and the fatty ones having lower density. Unlike pure hepatic lipomas, angiomyolipomas can show partial enhancement owing to the angiomatous component3 (Figure 15-16). The fatty elements of AML can be detected on MRI as bright on T1 and can be confirmed with fat saturation images. However, the presence of fat is not a specific finding, as hepatic adenomas can contain fat and hepatocellular carcinomas have also been reported
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
Figure 15-15. Lymphangiomatosis on CT. Axial contrast-enhanced images from this patient with systemic lymphangiomatosis demonstrates multiple cystic lesions throughout the liver and spleen as well as in the vertebral body.
A
B
Figure 15-16. Angiomyolipoma. (A) A non-contrast axial CT image reveals a heterogeneous lesion in the liver with elements of both fat and soft tissue density. The density of the fatty elements measures -95 Hounsfield units (HU), similar to the density of subcutaneous fat (F). (B) A post-contrast image reveals enhancement in the soft tissue parts of the lesion. This enhancement is due to the angiomatous component of this hepatic angiomyolipoma.
263
Section II. Approach to the Patient with Liver Disease
to undergo fatty metamorphosis. In addition, a pathologic series of 30 cases of hepatic AMLs found that one-third of the cases had fat in less than 10% of the lesion, and one-sixth had no fat at all.17 The presence or absence of fat therefore appears to be neither specific nor sensitive for the diagnosis.
Lipoma Hepatic lipomas are uncommon and can usually be easily diagnosed on imaging. They appear as well delineated masses with attenuation similar to that of subcutaneous fat on CT scan. On MR, the fatty lesions exhibit high attenuation on T1- and T2-weighted sequences. There is no enhancement following the administration of intravenous contrast agents.3
Leiomyoma Leiomyomas are extremely rare liver tumors. The imaging characteristics are relatively non-specific. The lesions can appear on ultrasound as either solid or hypoechoic, with scattered internal echoes. Leiomyomas have lower density than the surrounding liver parenchyma on CT, and can have two distinct enhancement patterns: either peripheral rim enhancement or homogeneous enhancement. On MR, the lesions are typically hypointense on T1 and hyperintense on T2.3
Inflammatory Pseudotumor Inflammatory pseudotumor is a rare benign lesion that can occur anywhere in the body. Histologically, it consists primarily of chronic inflammatory cells and proliferating fibrovascular elements. It typically presents in the liver as a solitary mass, with a slight male predominance. Inflammatory pseudotumor appears as isolated hypodense areas with either no contrast enhancement or delayed persistent enhancement on contrast-enhanced CT. The MRI features of inflammatory pseudotumor are variable, but most commonly the lesions exhibit T2 hyperintensity. On gadolinium-enhanced MR images several patterns have been described in individual case reports. It has been hypothesized that the variability in the contrast enhancement characteristics is a function of the relative degree of fibrotic versus inflammatory elements, with the fibrotic elements demonstrating delayed but persistent enhancement following gadolinium administration.18
FOCAL MALIGNANT LESIONS Liver Metastases Metastases are a much more common cause of focal liver lesions than primary hepatic neoplasms. Ultrasound, CT, MRI, and/or PET imaging may be employed to identify and characterize hepatic metastatic disease. CT and ultrasound are readily available at most sites and are more cost-effective than MRI or PET, and are therefore the primary imaging tools employed to evaluate the liver for metastatic disease. MRI and PET are often reserved for problem solving. The most common tumors that metastasize to the liver are breast, lung, pancreas, and gastrointestinal tumors. A significant number of patients will have hepatic metastases at the time of diagnosis. The prognosis in these cases depends on the
264
number and size of metastases. The appearance of liver metastases varies considerably. They are most commonly seen as discrete focal liver lesions, but occasionally they may appear as a diffuse infiltrative process that can mimic cirrhosis on imaging studies and which has been termed the pseudocirrhosis sign.19 Melanoma, lymphoma, pancreatic islet cell, breast, and colon carcinomas are the more commonly encountered primary tumors that may have this presentation. In this setting metastatic disease may appear as diffuse parenchymal heterogeneity on both contrast-enhanced CT and MRI. Detection is often difficult and is based on indirect features such as vascular and architectural distortion or alterations of the liver contour. Diffuse involvement of the liver by metastases may appear as multiple discrete nodules on ultrasound. Occasionally the ultrasound findings may be very subtle, the only sign of metastatic disease being a slight increase in heterogeneity of the hepatic parenchyma, often accompanied by hepatomegaly.20 The coexistence of small benign lesions in the liver, such as hemangiomas and cysts, often makes the accurate diagnosis of metastases more difficult. When these lesions are less than 15 mm in size it is often particularly difficult to differentiate them from metastases. In a series of 209 patients with known primary malignancy, 51% had liver lesions less than 15 mm which were suspicious for metastases but were subsequently found to be benign.21 The accuracy of imaging for the detection of metastases has been difficult to document because only a small percentage of patients with liver metastases undergo liver transplantation. Autopsy correlation is frequently unavailable for these patients. Consequently, intraoperative ultrasound (IOUS) is often used as the reference standard in studies evaluating the sensitivity of other imaging tools for the detection of hepatic metastases. IOUS has been found to detect 10–15% more lesions than other imaging techniques, such as CT arterioportography (CTAP), which is reported to have a sensitivity of 81–94%.22,23 IOUS is able to detect very small cysts 1–3 mm in size and small solid lesions 3–5 mm in size. However, despite the superior sensitivity of IOUS for the detection of small lesions, it has a lower specificity than CT and MRI for lesion characterization. Hepatic metastases usually appear as low- or isoattenuating lesions on unenhanced CT. Calcification is uncommon, but may be seen in metastases from ovarian, breast, lung, renal, and thyroid cancers, as well as from mucinous adenocarcinoma. As metastatic lesions in the liver enlarge, central areas of necrosis may become evident. The use of intravenous contrast is essential for identifying the majority of metastases. Most metastases are hypovascular and enhance poorly in comparison to liver parenchyma. Hypervascular metastases are less common and are usually associated with the following primary tumors: choriocarcinoma, renal cell carcinoma, thyroid carcinoma, breast carcinoma, melanoma, carcinoid tumor, neuroendocrine tumors, and pheochromocytomas (Figure 15-17). Hypervascular metastases typically enhance during the arterial phase of the CT scan, and may become isodense to the surrounding liver parenchyma on the later portal venous phase. Dual-phase CT scanning of the liver should be performed with scans during the arterial and portal venous phase if hypervascular metastases are suspected. Liver metastases appear on MRI as focal, rounded lesions with decreased signal (dark) on T1 images and moderately increased signal (bright) on T224 (Figure 15-18). Metastases in the liver may be dif-
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
Figure 15-17. Metastases on CT. (A) Contrast-enhanced axial CT in a patient with known breast carcinoma demonstrates multiple scattered hypodense lesions consistent with metastatic disease. (B) Contrast-enhanced axial CT from a different patient with a known neuroendocrine tumor reveals multiple hyperenhancing lesions throughout the liver, some of which have areas of central hypodensity consistent with necrosis. The marked enhancement of these lesions is due to their hypervascularity.
A
B
Figure 15-18. Metastatic disease on MRI. (A) T1-weighted axial MR image demonstrates multiple hypointense (dark) lesions scattered throughout the liver parenchyma. (B) In the T2-weighted image the lesions are slightly hyperintense (bright), consistent with the diagnosis of metastatic disease.
265
Section II. Approach to the Patient with Liver Disease
ferentiated from cysts and hemangiomas on MRI because they often have ill-defined margins and are usually not as bright on T2-weighted images as hemangiomas and cysts. Variations of this characteristic appearance may occur. Hepatic metastases from malignant melanoma are often of high signal intensity (bright) on T1-weighted images owing to the T1-shortening effect of melanin. Metastases to the liver from colonic and ovarian adenocarcinoma, multiple myeloma, and pancreatic mucinous cystic neoplasms have all been reported to have high signal on T1-weighted images.25 Hypervascular metastases may hemorrhage, resulting in increased signal on both T1 and T2 images. This can mimic the appearance of hemangiomas or cysts, although the enhancement characteristics of metastases may aid in differentiation. Dynamic gadolinium-enhanced MRI has similar arterial and venous enhancement patterns as contrast-enhanced CT scanning for metastases to the liver, but is reported to be more sensitive.26 Various strategies have been developed to increase detection rates of hepatic metastases using MRI contrast agents. The use of tissuespecific contrast agents such as ferumoxide (Feridex; Berlex, Wayne, NJ), which is a superparamagnetic iron oxide (SPIO) and a reticuloendothelial system-negative agent, has further improved the sensitivity and specificity of MRI.27 Metastatic lesions will appear bright against a dark liver following intravenous ferumoxide use. Alternatively, liver metastases appear unenhanced (dark) against a bright background (positively enhanced) following intravenous mangafodipir trisodium, which is a manganese-based hepatobiliarypositive agent that is taken up by hepatocytes and mainly excreted in the bile.28 MRI is also employed in patients with diffuse fatty infiltration, where CT or ultrasound may not be able to differentiate metastases from areas of fatty sparing. In this situation, chemical shift imaging with MRI is helpful in showing signal loss in regions of fatty infiltration. Conventional CT and MRI-based techniques are reasonably successful in demonstrating metastases of 1 cm or larger, with a sensitivity of 95–100% using ferumoxide (SPIO)-enhanced MRI, and a sensitivity of 90–95% using contrast-enhanced helical CT.29, 30 The sensitivity of CT for the detection of lesions <1 cm remains disappointing at 56%.31 However, there is likely to be further improvement in the detection of small lesions owing to the narrower slice collimation possible with newer multidetector CT scanners. Positron emission tomography (PET) is a molecular imaging technique that reflects physiologic processes in the tissues being imaged and is a valuable technique for the detection of metastases.32 The most frequently used positron emitting radiopharmaceutical is the 18-fluorine 2-deoxy-2-D-glucose (18F-FDG), a radioactively labeled glucose analog. 18F-FDG accumulates predominantly in tumor tissue and decays by emitting positrons, which can be visualized as a bright spot clearly distinguishable from the surrounding parenchyma approximately 45 minutes after intravenous injection. 18F-FDG PET has high sensitivity for the detection of hepatic metastases, especially for those of colorectal cancer, with reported sensitivities and specificities as high as 100% for lesions >1 cm in diameter.33,34 A significant limitation of PET imaging is the poor anatomic detail seen on scans. Therefore, abnormalities identified on PET imaging need to be correlated with findings from other studies, such as CT or MRI.
266
PET-CT is a relatively new technology that combines both PET and CT, providing both metabolic and anatomical information regarding liver lesions. Several authors have reported improved lesion localization with PET-CT in comparison to CT and PET performed separately. This is especially true for lesions that cannot be identified by conventional CT.35
Hepatocellular Carcinoma Hepatocellular carcinoma accounts for 90% of all primary malignant hepatic tumors and often occurs in the setting of cirrhosis. The primary role of imaging is early detection of HCC, as solitary small lesions can be treated successfully with liver transplantation. Transplantation remains an option in patients with up to three discrete lesions if each is smaller than 3 cm, or if there is a solitary lesion less than 5 cm in diameter. Patients who have undergone liver transplantation have a reported 4-year survival rate of 85%.36 Transplantation is usually of little benefit when lesions are numerous or larger than 5 cm, as metastatic disease is often present at this stage. The detection of HCC is difficult in the cirrhotic liver because the generalized architectural distortion and fibrosis results in a heterogeneous appearance. When fibrosis becomes confluent, it can appear mass-like and may mimic HCC. HCC may present as a solitary mass, multiple focal masses, or as a diffuse infiltrating process. Lesions are often heterogeneous in appearance, with evidence of necrosis, hemorrhage, or calcification. Small regions of fatty metamorphosis within lesions have also been described.37 HCC also readily invades hepatic vessels, particularly the portal venous system (Figure 15-19). Tumors may be exophytic or pedunculated in nature, extending outside the liver. A pseudo-
Figure 15-19. Portal vein invasion by HCC. Contrast-enhanced axial CT image obtained during the portal venous phase demonstrates a large heterogeneous HCC with evidence of invasion (arrow) into the portal vein (PV).
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
capsule is visible in up to 60% of cases. Occasionally HCC develops in non-cirrhotic livers and typically presents as a solitary, welldefined mass (Figure 15-20). HCC may be difficult to detect on ultrasound, particularly in patients with severe cirrhosis. Lesions may be of decreased, increased, or mixed echogenicity relative to liver parenchyma. A fibrous pseudocapsule, if present, appears as a hypoechoic halo
around the lesion. The margins of the tumor are often irregular when there is no capsule. Tumors are often hypervascular and surrounded by a fine network of vessels.38 Extension of the tumor into hepatic vein branches, portal vein branches, or bile ducts is readily identified on ultrasound imaging. Solid lesions identified within the liver parenchyma on ultrasound are often non-specific and may represent HCC, regenerative
A
B
C
D
Figure 15-20. HCC in a patient with HBV. In this patient with hepatitis B infection, a well defined mass is seen that is consistent with hepatocellular carcinoma. (A) The arterial-phase axial CT image reveals a well circumscribed lesion with early enhancement. (B) In the subsequent image, the hypervascularity of the lesion is clearly evident. (C,D) In the delayed-phase images the lesion is relatively hypodense to the surrounding parenchyma, with a rim of continued enhancement which may represent the tumor capsule.
267
Section II. Approach to the Patient with Liver Disease
nodules, siderotic nodules, dysplastic nodules, focal fat, or hemangiomas. The sensitivity of ultrasound for nodule detection in livers affected by cirrhosis is 30–51% and only 33–45% for HCC detection.39,40 Because of this low sensitivity, ultrasound is often used in combination with serum AFP measurements. The use of sonography with serum AFP has improved the accuracy for HCC detection. However, a significant false positive rate remains, and in one study 25% of patients had benign lesions mistaken for malignancy and another 20% had lesions that could not be identified on subsequent imaging studies or repeat ultrasound.41 Dynamic contrastenhanced CT or MRI is often obtained to further characterize lesions identified on ultrasound. In advanced cases of cirrhosis all three imaging modalities will have poor detection rates, and biopsy may be required to confirm the presence of HCC. Various types of nodule may be present in the cirrhotic liver. Of the different types, regenerative nodules are present pathologically in all cirrhotic livers, although they are identified in only a minority of patients by CT imaging. Siderotic nodules contain focal iron deposits and are more likely to be detected. They appear as small hyperdense foci on unenhanced CT (Figure 15-21). Regenerative nodules may be difficult to identify on contrast-enhanced CT as they typically enhance homogeneously and to the same degree as the surrounding liver parenchyma. Dysplastic nodules are often not identified by CT imaging, as they have a similar appearance and enhancement characteristics as regenerative nodules. Dysplastic nodules may also contain iron deposits, giving them a similar appearance to siderotic nodules on unenhanced CT. HCC appears as a hypoattenuating mass on unenhanced CT scans and may contain central areas of high density due to intratumoral
Figure 15-21. Siderotic nodules on CT. Non-contrast axial CT image in this patient with cirrhosis demonstrates small hyperdense foci that are consistent with siderotic regenerative nodules. Such nodules may be difficult to differentiate from HCC, but the hyperdensity on the non-contrast images is helpful.
268
hemorrhage, or areas of low density due to necrosis. If the tumor is encapsulated, a thin rim of low attenuation is evident. Unencapsulated tumors often have ill-defined margins. HCC is difficult to detect on conventional contrast-enhanced CT scans performed during the portal venous phase of enhancement, with a reported sensitivity of 48% and a specificity of 70% for HCC detection.42 The tumor typically enhances early on arterial phase images (see Figure 15-20) and appears isoattenuating or hypoattenuating on portal venous phase images. Helical CT with the use of specific triple-phase scanning protocols consisting of non contrast, arterial, and portal venous phases has improved sensitivity for the detection of HCC up to 59–88%.43,44 The accuracy of triple-phase dynamic contrast-enhanced CT for detecting tumors <2 cm in cirrhotic livers is lower, at 60%, but improves to 82% for lesions measuring 2–5 cm.45 Direct CT hepatic arteriography (CTHA) and CT arterioportography (CTAP) are techniques that were used in the past for HCC detection. CTHA refers to CT performed during catheter injection of contrast into the hepatic artery. CTAP refers to CT performed during catheter injection of contrast into the superior mesenteric artery, with contrast returning to the liver via the portal circulation. Although CTAP and CTHA are considered to be the most sensitive CT techniques for HCC detection, false positives are common. Low specificity, high cost, and the increased invasiveness have caused these techniques to be largely replaced by multiphasic helical CT.46 Lipiodol-enhanced CT is another technique that was used in the past for HCC detection. This has also been replaced by multiphasic helical CT because studies have found no significant advantage of lipiodol-enhanced CT; a sensitivity of 53% and specificity of 88% have been reported.47 Advances in MRI technology have resulted in improved sensitivity and specificity of this modality for HCC detection, particularly in the cirrhotic liver. Studies have confirmed that multiphasic gadolinium-enhanced MRI is better for HCC detection than contrast-enhanced multiphasic CT.48,49 Both modalities still have difficulty in detecting small tumors (<2 cm) owing to the presence of other focal lesions in patients with late-stage cirrhosis.49 Regenerative nodules result from localized proliferation of hepatocytes and their supporting stroma. They are usually not detected by cross-sectional imaging because they are too small or too similar to adjacent liver parenchyma. MRI does, however, detect more regenerative nodules than CT, but is only reliably able to identify regenerative nodules that contain significant amounts of iron in the form of hemosiderin, which appear as hypointense (dark) foci on both T1- and T2-weighted images (Figure 15-22). Regenerative nodules without hemosiderin are similar in signal to normal liver parenchyma and will not be detected unless they are large and cause a bulge in the liver contour. Gradient echo sequences demonstrate the presence of hemosiderin better than do spin-echo sequences because of greater sensitivity to the magnetic susceptibility affects of iron in hemosiderin. This superparamagnetic affect is particularly conspicuous on T2-weighted spoiled gradient echo sequence (SPG) and T2*-weighted gradient echo imaging, which is the most sensitive sequence for the detection of iron and results in maximal loss of signal compared to other sequences. Dysplastic nodules (DN) are nodular regions of hepatocytes with dysplasia but without definite histologic criteria for malignancy.
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
Figure 15-22. Siderotic nodules on MR. (A) A T2-weighted MR image in this patient with cirrhosis demonstrates scattered low-intensity (dark) lesions (arrows) throughout the liver that represent siderotic nodules. (B) On T1-weighted imaging after the administration of contrast, the lesions (arrow) remain hypointense. Although HCC may have a similar pre-contrast T1 appearance, they tend to enhance with contrast administration.
They may be low or high grade. Dysplastic nodules appear as low signal intensity (dark) foci in the liver on T2-weighted and gradient echo sequences. They are mildly hyperintense (bright) on T1weighted sequences, in contrast to typical findings for HCC. These signal changes reflect the iron, copper, glycogen, and lipid content of the lesions.50 It is often not possible to differentiate dysplastic nodules from regenerative nodules on MRI: one study was only able to identify 15% of dysplastic nodules on pretransplant MRI studies.49 Identification is complicated because it is difficult to determine whether dysplastic nodules enhance on MRI, as they are already hyperintense (bright) on T1-weighted images. The MR imaging appearance of HCC is highly variable. This variability may be due in part to hemorrhage and necrosis, or to the copper, glycogen, or lipid content of these lesions.50 HCC is typically of increased signal intensity (slightly bright) on T2-weighted sequences and hypointense (dark) on T1-weighted images51 (Figure 15-23). However, these are often the signal characteristics of regenerating and dysplastic nodules as well, particularly on T2-weighted images.49,51 Therefore, the distinction between these malignant and benign lesions cannot be made reliably using signal characteristics alone. Other benign lesions occurring in the cirrhotic liver, such as hemangiomas, regions of hepatic infarction, and confluent hepatic fibrosis, may also appear slightly bright on T2-weighted sequences.51 The enhancement characteristics of hepatocellular carcinoma may be used to help improve the accuracy of diagnosis. Studies have shown that early arterial enhancement with gadolinium diethylene triaminepenta-acetic acid (Gd-DTPA) is an important imaging feature of HCC. Lesions are typically hyperintense during the arterial phase and are iso- to hypointense relative to surrounding liver parenchyma on delayed phase images. Although this is considered a
characteristic feature of HCC, it can also occasionally be seen with dysplastic and regenerating nodules, arteriovenous and arterioportal shunts, and metastases. Benign lesions such as cavernous hemangiomas, hepatocellular adenomas, and focal nodular hyperplasia may also have a similar appearance. HCC may have a rim of low signal intensity on T1-weighted images corresponding to the fibrous capsule. If present, it usually enhances on delayed images. Hepatic tissue-specific contrast agents such as ferumoxide, which consist of superparamagnetic iron oxide particles (SPIO), have been employed to help diagnose HCC. Although the use of SPIO improves detection on T2-weighted sequences, these agents do not significantly improve detection of HCC compared to Gd-DTPAenhanced MRI. Ferumoxide and gadolinium have also been used in combination to further aid HCC detection, and there may be some benefit to this approach.52
Fibrolamellar Carcinoma Fibrolamellar carcinoma (FLC) is a rare neoplasm of hepatocellular origin that occurs in a younger age group than HCC (mean age at presentation 23 years) and typically develops in the absence of cirrhosis. FLC carries a better prognosis than HCC, with a 5-year survival rate of 60%, compared to 30% for HCC.53 FLC appears as a well defined mass of mixed or increased echotexture with lobulated margins on ultrasound.54 Focal hypoechoic regions may be seen in the tumor and usually represent hemorrhage or necrosis. Ultrasound is only partially successful in demonstrating central scars (33–60% detection rate).55,56 When a central scar is present it may be calcified and is often hyperechoic in appearance, with posterior acoustic shadowing.
269
Section II. Approach to the Patient with Liver Disease
A
C
Fibrolamellar carcinoma is usually a well defined hypodense liver mass on unenhanced CT. Calcifications in the central scar can be seen in up to 55% of cases,55 and is a useful feature to help differentiate these lesions from HCC and FNH, in which calcification is much less common: 7.5% and 1.4%, respectively.56,57 The calcifications may be punctate, nodular, or stellate. The majority of FLC are hypervascular and enhance heterogeneously during the arterial phase of CT (80%).56 In the portal venous phase of imaging the tumors are often isoattenuating to the surrounding liver parenchyma, but some may remain hyperattenuating or become hypoattenuating. The
270
B
Figure 15-23. HCC on MRI. (A) Pre-contrast T1-weighted axial MR demonstrates a subtle low-signal lesion (arrow) in the periphery of the right lobe of the liver. (B) The early post-contrast T1-weighted image demonstrates marked enhancement. (C) On T2-weighted MR imaging the lesion is of slightly increased signal relative to the surrounding parenchyma. These findings are consistent with hepatocellular carcinoma.
central scar, when present, is usually low density relative to the tumor on both unenhanced and contrast-enhanced images, but has been reported to increase in density on delayed-phase images in 25–56% of cases.56, 58 In contrast to HCC, invasion of the portal venous system is uncommon with FLC. FLC may cause capsular retraction, which helps to differentiate it from benign lesions such as FNH.59 FLC appears on MRI as hypointense (dark) on T1-weighted images and isointense or slightly hyperintense (bright) on T2. The central scar and fibrous septa are usually well demonstrated on
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
MR imaging, and are hypointense on both T1- and T2-weighted sequences. There is often delayed enhancement of the central scar after intravenous Gd-DTPA.60 Areas of calcification are low signal or demonstrate a signal void. FNH can also present with a central scar and have an appearance similar to FLC; however, the central scar of FNH commonly has increased signal intensity on T2-weighted MRI, in contrast to the low signal intensity of the FLC scar.61 If a central scar is absent it is often not possible to differentiate FLC from other tumors, such as HCC, metastases, hepatocellular adenoma, or cholangiocarcinoma, based on imaging characteristics.
Cholangiocarcinoma Cholangiocarcinoma is an adenocarcinoma arising from bile duct epithelium. It is the second most common primary liver malignancy after HCC and is more common in males, typically occurring in the sixth or seventh decade of life. It is usually classified as either intrahepatic or extrahepatic, with intrahepatic cholangiocarcinoma further subdivided into peripheral or hilar in location. Cholangiocarcinomas are variable in appearance on ultrasound and may have a hypoechoic, hyperechoic, or mixed appearance. Calcification, if present, appear as echogenic foci with shadowing. Biliary ductal dilatation is often present peripheral to the tumor.62 Cholangiocarcinoma appears as a hypodense mass with irregular margins on unenhanced CT. There may be speckled areas of high attenuation seen within the mass, representing mucin in the tumor and bile ducts.63 Following intravenous contrast administration there is usually mild to moderate peripheral enhancement, with poorly enhancing central areas representing regions of dense fibrosis within the tumor (Figure 15-24). Delayed persistent enhancement is seen in 74% of lesions.64 The degree of contrast enhancement on delayed
A
images is variable and the lesions may enhance homogeneously or heterogeneously. Areas of necrosis and hemorrhage may also be present. Cholangiocarcinomas are usually unencapsulated and hypointense (dark) on T1-weighted MRI and hyperintense (bright) on T2, compared to normal liver parenchyma. Central areas of hypointensity on T2-weighted images represent areas of fibrosis within the tumor65. In addition to the main lesion, satellite nodules may also occur. With dynamic post-Gd-DTPA-enhanced MRI, progressive concentric enhancement is seen. This usually starts peripherally, with the lesion filling towards the center. The fibrotic regions of the tumor do not typically enhance to any significant degree, and this can help to differentiate these lesions from hemangiomas.
Biliary Cystadenocarcinoma Biliary cystadenocarcinomas are rare, slow growing cystic neoplasms, the majority of which are related to the intrahepatic bile ducts. They represent less than 5% of intrahepatic cystic masses of biliary origin.66 There are two subtypes: biliary cystadenocarcinoma with ovarian stroma, and biliary cystadenocarcinoma without ovarian stroma. Biliary cystadenocarcinoma with ovarian stroma is found predominantly in women and typically has a better prognosis. This form often develops from a pre-existing biliary cystadenoma. Biliary cystadenocarcinoma without ovarian stroma is found in both sexes and carries a worse prognosis, with 50% mortality.67 It is usually not possible to distinguish between these subtypes based on imaging, or to differentiate accurately between benign cystadenomas and cystadenocarcinomas. Biliary cystadenocarcinomas appear as cystic lesions on ultrasound and may contain fluid of varying echogenicity, depending upon
B
Figure 15-24. Cholangiocarcinoma on CT. (A) Pre-contrast axial CT scan reveals a hypodense mass arising from the hepatic hilum with peripheral intrahepatic bile duct dilatation (arrow). (B) The size and extent of the lesion are better appreciated after intravenous contrast enhancement, multiple hypodense areas being seen within the lesion representing fibrosis and regions of necrosis. Biopsy confirmed the diagnosis of cholangiocarcinoma.
271
Section II. Approach to the Patient with Liver Disease
whether the fluid is proteinaceous, mucinous, or hemorrhagic. The cysts may be septated and contain mural nodules. The septations and mural nodules are usually better visualized on ultrasound than with other imaging modalities.68 Cyst calcifications appear as hyperechoic foci with posterior acoustic shadowing. Biliary cystadenocarcinomas appear as multiloculated cysts on CT scans. The density of the cysts depends upon their content (Figure 15-25). Septations and nodules may enhance or be calcified. The presence of both nodules and septations is reported to favor the diagnosis of cystadenocarcinoma over cystadenoma.69 MRI can better characterize these lesions and determine their extent. The MR signal intensity of biliary cystadenocarcinoma is variable on both T1- and T2-weighted images, depending on the composition of the cyst fluid. Fluid–fluid levels, which probably represent the different stages of hemorrhage, occur more frequently in biliary cystadenocarcinoma than cystadenoma, and are best depicted by MRI.66,69
Lymphoma Primary hepatic lymphoma is rare in comparison to secondary involvement, which may be seen in up to 50% of patients with Hodgkin’s and non-Hodgkin’s lymphoma.70 Primary hepatic lymphoma may appear as solitary or multifocal masses. Diffuse infiltration of the liver is uncommon and is usually associated with secondary involvement. Focal lymphomatous lesions may be hypoechoic or anechoic on ultrasound relative to normal liver parenchyma, with posterior acoustic enhancement.71 The diffuse form of the disease is often difficult to detect sonographically because it appears as a generalized change in echogenicity (either hypoechoic or heterogeneous), which is often overlooked
Figure 15-25. Biliary cystadenocarcinoma. Contrast-enhanced axial CT reveals a large septated cystic lesion with internal calcifications. Although the imaging characteristics alone do not allow differentiation from a benign biliary cystadenoma, this lesion was pathologically proved to be a malignant cystadenocarcinoma.
272
and attributed to other commonly occurring liver pathologies, such as fatty infiltration or cirrhosis.70 Primary hepatic lymphoma appears on CT as a hypodense liver mass that does not enhance significantly following intravenous contrast administration. Areas of central necrosis may also be present.72 Secondary hepatic lymphoma is usually diffusely infiltrative, resulting in a generalized decreased attenuation of the liver indistinguishable from fatty infiltration. It is uncommon for primary hepatic lymphoma to have this presentation. The appearance of hepatic lymphoma on MR imaging is also nonspecific. Primary hepatic lymphoma has been described as iso- or hypointense on T1-weighted images and hyperintense on T2.70
Hepatic Angiosarcoma Hepatic angiosarcoma (HA) is the most common primary mesenchymal tumor of the liver and accounts for 2% of all primary liver tumors. It is aggressive and occurs predominantly in men. These lesions are highly vascular and percutaneous biopsy is associated with the risk of massive hemorrhage. Hepatic angiosarcoma has a well established association with prior exposure to carcinogens, such as the radiographic contrast agent Thorotrast (thorium dioxide). Radiation, vinyl chloride, and arsenic exposure have also been implicated as causes of hepatic angiosarcoma,73 although fortunately, exposure to these agents is now quite rare. Most tumors currently occur in the absence of known risk factors. Thorotrast is poorly excreted and accumulates in the liver, spleen, bone marrow, and lymph nodes, and may be visible on abdominal radiographs because of its density. After Thorotrast exposure there is usually a latency period of about 20–40 years before the tumor develops.74 If, over time, there is evidence on plain radiographs of displacement of Thorotrast particles overlying the liver, the diagnosis of hepatic angiosarcoma should be considered.75 The ultrasound features are non-specific. Single or multiple masses are often visible, with variable echotexture depending on the degree of hemorrhage and necrosis within the lesions. CT is very sensitive for identifying the metallic density of Thorotrast as well as demonstrating any relative displacement of Thorotrast particles by tumor. The tumors are usually low density on non-contrast enhanced CT, with hyperdense foci that probably represent hemorrhage. After contrast administration there may be areas of intense enhancement in the more vascular portions of the tumor during the arterial phase, which then become hypo- or isoattenuating during the portal venous phase of enhancement.75 Some angiosarcomas have an enhancement pattern similar to that of benign hepatic hemangiomas, with persistent centripetal filling on delayed images. MRI demonstrates the hemorrhagic and hypervascular nature of hepatic angiosarcomas. They are usually hypointense (dark) to normal liver on T1-weighted images, with high signal (bright) foci within the lesions suggesting hemorrhage. The lesions are hyperintense (bright) on T2-weighted sequences. After intravenous GdDTPA administration there is initial peripheral enhancement followed by persistent enhancement on delayed images.76
Epithelioid Hemangioendothelioma Epithelioid hemangioendothelioma (EHE) is a rare, low-grade malignant neoplasm of vascular origin that may arise in the liver, lung,
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
soft tissue, and bone. It occurs in adulthood, with a mean age of 46 years, and has a slight female predominance.77 Plain radiography demonstrates calcification in 15% of patients.78 The sonographic features of epithelioid hemangioendothelioma are variable. The tumor is typically seen as multiple hypoechoic lesions that correspond to the central core of myxoid stroma.79 However, lesions may also appear isoechoic or hyperechoic relative to the liver parenchyma. Calcifications, if present, are seen as echogenic foci with posterior acoustic shadowing. Different sonographic patterns may be seen in the same patient. Epithelioid hemangioendotheliomas are low density on contrastenhanced CT and demonstrate peripheral enhancement after intravenous contrast administration80 (Figure 15-26). There is often a fibrotic reaction in the surrounding liver parenchyma which results in capsular retraction.81 Evidence of calcification is seen in 20% of patients on CT.77 MRI depicts the internal architecture of EHE better than does CT.82 Lesions have a variable appearance on T1-weighted MR images but are often of low to normal intensity relative to normal liver, with a thin dark peripheral rim.83 On T2-weighted images the lesions are often hyperintense (bright) or heterogeneous in signal intensity. Again, a dark peripheral rim may be evident. Following the administration of intravenous gadolinium-DTPA these lesions demonstrate a central hypointense region, with enhancement of the peripheral rim and an outermost hypointense (dark) rim.82
DIFFUSE LIVER DISEASE Diffuse liver disease is usually the result of disease processes that result in cirrhosis, such as alcohol abuse, viral hepatitis, hemochromatosis, Wilson’s disease, Budd–Chiari syndrome, and steatosis. In
A
addition, systemic disorders such as sarcoidosis and amyloidosis may present with diffuse abnormality of the liver parenchyma. Other conditions that typically present as focal lesions may mimic the radiological appearance of diffuse liver disease, including hepatocellular carcinoma, metastases, and lymphoma. Imaging plays a key role in detecting and characterizing diffuse liver disease and may obviate the need for tissue biopsy.
Cirrhosis Cirrhosis is the most common cause of diffuse liver disease and is the result of repeated episodes of hepatocellular injury from a variety of different causes, such as impaired circulation, exposure to toxins, and disease processes causing hepatic inflammation. These repeated insults result in a process of nodular regeneration and fibrosis, causing permanent changes to the hepatic architecture. Changes in hepatic morphology related to cirrhosis may be detected by ultrasound, CT, or MRI. A common finding in viral and alcohol-related cirrhosis is volume loss of the right hepatic lobe and the medial segment of the left hepatic lobe with compensatory enlargement of the remainder of the left lobe and the caudate lobe. This change in lobe size has been quantified by assessing the caudate to right lobe ratio. A ratio >0.65 is reported to be 90% specific for the presence of cirrhosis.84 Cross-sectional imaging modalities such as CT and MRI are well suited to demonstrate this finding because both the right lobe and the caudate lobe are clearly seen. The pattern of lobar atrophy and regeneration is often different in patients with autoimmune causes of cirrhosis and Budd–Chiari syndrome, who typically present with atrophy of the lateral segment of the left lobe and enlargement of the caudate lobe. Widening of the hepatic interlobar fissure and expansion of the gallbladder fossa may also be associated with cirrhosis (Figure 15-27). Enlargement of the gallbladder
B
Figure 15-26. Epithelioid hemangioendothelioma. (A) Contrast-enhanced CT scan through the abdomen reveals a rim enhancing lesion (black arrow) that was found to be an epithelioid hemangioendothelioma. (B) A corresponding vascular lesion (white arrow) was found in the lung as well.
273
Section II. Approach to the Patient with Liver Disease
Figure 15-27. Widened gallbladder fossa. This axial CT image demonstrates a markedly widened gallbladder fossa (arrows) of the liver. This contains the gallbladder and also part of the stomach, duodenum, and colon in this patient. This finding is very suggestive of cirrhosis.
fossa has been shown to have a sensitivity, specificity, accuracy, and positive predictive value for the diagnosis of cirrhosis on MRI of 68%, 98%, 80%, and 98%, respectively.85 As cirrhosis progresses, the contour of the liver becomes irregular in outline in the presence of nodule formation and intrahepatic fibrosis causing capsular retraction (Figure 15-28). This distortion of the parenchymal tissues also affects the hepatic arterial and portal venous circulation, resulting in increased tortuosity of the intrahepatic vessels, described as a ‘corkscrew appearance’ (Figure 15-29). This finding was initially described on angiography, but can be seen on axial, multiplanar reformatted, and three-dimensional images using multidetector CT scanners as well as on MRI.86 Portal venous hypertension may develop, resulting in reduced portal vein blood flow, which may eventually reverse in direction. There is often increased flow within the hepatic arteries to compensate when this occurs. Indirect evidence of portal venous hypertension includes secondary signs such as splenomegaly, and the development of multiple portosystemic collateral vessels (Figures 15-30, 15-31). Arteriovenous and arterioportal shunts may also develop, and appear as early opacification of the intrahepatic veins during arterial phase contrast-enhanced CT. Occasionally, regions of fibrosis in the liver become more confluent and may be mistaken for focal mass lesions on CT and MRI. Confluent fibrosis is characteristically low in density on unenhanced CT.87 After contrast administration there is often delayed enhancement. Occasionally, confluent fibrosis may exhibit a more rapid or intense enhancement during the arterial and portal phases. In this situation, it may be difficult to differentiate it from hepatocellular carcinoma (HCC).88 There is loss of volume in the involved liver
274
Figure 15-28. Nodular shrunken liver. This axial CT image demonstrates a nodular, shrunken liver (arrows indicate nodular liver edge) consistent with cirrhosis. Ascites (A) is also present.
Figure 15-29. Corkscrew vessels. Contrast-enhanced axial CT in this cirrhotic patient demonstrates ascites, a nodular shrunken liver, and corkscrew vessels (arrow).
segments, with associated vascular crowding and capsular retraction. Similar changes may be seen on MRI. Regions of fibrosis are generally lower in signal on T1- and T2-weighted sequences than the surrounding liver parenchyma. There are numerous other focal lesions that can be seen in a cirrhotic liver that may also simulate HCC.
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
Figure 15-30. Recanalized paraumbilical vein. Two axial CT images reveal findings consistent with portal hypertension. (A) Splenomegaly (S) and prominent subcutaneous vessels. (B) Another image revealing a dilated subcutaneous vessel (arrows) in the anterior abdominal wall. This vessel is a recanalized paraumbilical vein, in this patient with cirrhosis and portal hypertension who had clinical evidence of a ‘caput medusae’.
These include regenerating and dysplastic nodules, as well as benign lesions such as cysts and hemangiomas. It is often a challenge to distinguish HCC accurately from these other lesions, especially in the setting of cirrhosis. Imaging findings should always be correlated with clinical disease markers, such as serum a-fetoprotein. In addition to the generalized parenchymal changes related to fibrosis within the cirrhotic liver, focal nodules are often seen. Imaging cannot reliably distinguish between micro- and macronodular forms of the disease. When nodularity is identified by imaging it is usually of the macronodular type, with nodules larger than 3 mm. Hepatic nodules have been categorized into two main types: regenerative, and dysplastic or neoplastic.89 Regenerative nodules result from local proliferation of hepatocytes surrounded by fibrous septa. They include monoacinar and multiacinar regenerative nodules, siderotic nodules, and regions of segmental or lobar hyperplasia. It is usually not possible to make the distinction between these different types of nodule by imaging. Dysplastic nodules are found in up to 25% of explanted livers,90 and are subdivided into low or high grade depending on their potential to develop into HCC. A stepwise pathway of carcinogenesis has been proposed, from regenerative nodule to low grade, then high-grade dysplastic nodules, and finally to HCC.90 Cirrhosis also results in extrahepatic changes that are often better visualized by CT and MRI than by ultrasound. Bowel wall thickening has been found in 64% of patients with cirrhosis, versus 7% of control subjects.91 Gallbladder wall thickening may occur not associated with cholelithiasis. Mesenteric edema and stranding are seen with increased frequency in up to 86% of patients with cirrhosis.92 There is often associated ascites, pleural effusions, and subcutaneous
edema. These changes have been attributed to the low mean serum albumin levels that exist in these patients.
Iron Storage Disorders Hepatic iron deposition may be due to hemochromatosis, cirrhosis, intravascular hemolysis, or following numerous blood transfusions. Increased hepatic iron deposition can be detected on unenhanced CT by examining the attenuation value of the liver (measured in Hounsfield units – HU), which approximates the relative liver density. The normal liver has a value of 50–65 HU. Iron deposition within the liver can increase this value to range from 70 to 140 HU. In these cases the liver vasculature often appears prominent, with increased contrast between the relatively lower-density vessels and the higher-density liver (Figure 15-32). CT has been shown to have 100% sensitivity in detecting hepatic iron overload when blood iron levels are raised to five times above normal. The sensitivity decreases to approximately 60% when levels are only 2.5 times above normal.93 However, the liver may be normal or low in density despite significant iron deposition, owing to concomitant fatty infiltration (which tends to lower the attenuation value (HU)). There are also other causes of a hyperdense liver on unenhanced CT, such as Wilson’s disease, type IV glycogen storage disease, Thorotrast administration, gold deposition, and amiodarone usage. MRI is the most sensitive and specific imaging modality for demonstrating increased hepatic iron stores, because the change in signal intensity is not based on density but the actual physical properties of iron atoms. A decrease in hepatic MR signal intensity on both T1- and T2-weighted sequences occurs in the presence of iron
275
Section II. Approach to the Patient with Liver Disease
A
B
C
D
Figure 15-31. Portal HTN and collaterals. (A) Contrast-enhanced axial CT reveals multiple findings consistent with cirrhosis, including splenomegaly (S), a shrunken and nodular liver (L), and a recanalized paraumbilical vein (arrow). (B) CT from a different patient reveals significant vascular collateralization (enhancing vessels in the center of the image). (C,D) Two more examples of collateralization in the form of esophageal and gastrohepatic ligament varices (arrows) are seen in these patients with portal hypertension.
overload because iron has superparamagnetic properties and leads to spin dephasing. This superparamagnetic effect is particularly conspicuous on T2-weighted spoiled gradient echo (SPG) and T*weighted gradient echo imaging, which maximize the signal loss compared to other MR sequences.94 The signal intensity in the liver can be compared to that of adjacent paraspinal muscles, because iron is not deposited within the skeletal muscle. Therefore, if the liver
276
signal intensity is much less than that of paraspinal muscle, iron deposition should be considered95 (Figure 15-33). MR imaging can also help to distinguish primary from secondary hemochromatosis. In primary hemochromatosis iron is also deposited in other organs, such as the pancreas, heart, and adrenal glands, changing their signal characteristics (Figure 15-33). In most cases the spleen is spared.96 The pattern of iron deposition is
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
Figure 15-32. Iron deposition on CT. Hepatic iron deposition was suspected in this patient who had received multiple transfusions. (A) A non-contrast axial CT demonstrates normal hepatic density in an earlier scan. (B) In this non-contrast study, obtained several years later, the liver (L) is noted to be markedly more dense (brighter, 80 HU) than the spleen (S) (50 HU). Also, note the increased contrast between the liver parenchyma and the darker hepatic vessels. This appearance is consistent with hepatic iron deposition.
A
B
Figure 15-33. Hemochromatosis on MRI. (A) A T2-weighted axial image shows the liver (L) to be diffusely low in signal (dark) in this patient with hemochromatosis. The pancreas (P) is also dark, owing to iron deposition within its parenchyma. These features are consistent with primary hemochromatosis. (B) The T2weighted image at a higher level reveals a similar loss in signal of the myocardium (C), further confirming that this is a case of primary rather than secondary hemochromatosis.
277
Section II. Approach to the Patient with Liver Disease
different in cases of secondary hemochromatosis, where abnormal signal intensity is seen initially only in organs with significant reticuloendothelial cell activity, such as the liver, spleen, and bone marrow. Eventually, when the reticuloendothelial system becomes saturated, iron deposition in other tissues occurs and may also be detected by MRI. Ultrasound has no role in detecting hepatic iron overload. If sonographic liver abnormalities are present they are usually secondary to cirrhosis.
Wilson’s Disease In Wilson’s disease copper accumulates in the liver because of impaired biliary excretion.97 The copper is deposited predominantly in a periportal distribution along the hepatic sinusoids, which incites an inflammatory reaction resulting in acute hepatitis with fatty change. This process results in fibrosis of the liver, and eventually in macronodular cirrhosis.97 The CT scan features of Wilson’s disease are predominantly those of cirrhosis. One exception is that the liver is diffusely increased in density, owing to the accumulation of copper, and will appear brighter on CT.98 The increased attenuation from copper deposition is in some cases offset by abnormally low density from coexisting fatty deposition in the liver, resulting in no overall change in density. Copper has paramagnetic properties, which may be detected by MRI. Focal areas of copper deposition appear as hyperintense (lighter) on T1-weighted images and hypointense (darker) on T2. These findings may only be seen in the early stages of the disease, before severe cirrhosis has developed.99 The sonographic findings in the liver related to Wilson’s disease are those of cirrhosis, and are indistinguishable from cirrhosis related to other causes. Overall, imaging does not play a major role in the diagnosis of this condition.
Steatosis Steatosis, or fatty change, within the liver can result from several common etiologies, including diabetes mellitus, alcohol, obesity, hepatitis, liver transplantation, and various drugs. Fatty deposition occurs as a result of either decreased hepatic clearance of fatty acids or increased production and mobilization of fatty acids. The distribution of steatosis or fatty change throughout the liver is often quite variable, ranging from focal or multifocal deposits to diffuse infiltration. Focal fatty change can mimic focal neoplasm, although it usually occurs in characteristic locations, such as adjacent to the intersegmental fissure or next to the gallbladder fossa. Focal areas of fatty infiltration appear as regions of low density (darker) on noncontrast and contrast-enhanced CT. If there are vessels passing through a region of focal fatty deposition they are usually not deflected or distorted, which may help to differentiate focal fatty change from other low-density lesions, such as tumor. It is easier to quantify fatty deposition with CT because density measurements may be obtained. The normal liver has an attenuation value of 50–65 HU and is approximately 8–12 HU higher than the spleen on unenhanced CT scans.100 Areas of fatty change show abnormally decreased attenuation – typically 10 HU lower than the spleen on unenhanced CT scan – and 25 HU less than the spleen on contrast-enhanced CT scan101 (Figure 15-34). The CT appearances of steatosis are usually easily identified. However, focal fatty change may occasionally have an atypical appearance and can mimic
278
a focal hepatic lesion. In this instance, either ultrasound or MRI may be employed to distinguish focal fat from tumor. MRI is the most sensitive technique for detecting fatty change within the liver. On T1-weighted spin echo sequences areas of fatty change are higher in signal intensity (brighter) than normal liver. The MR technique that best demonstrates fatty change is in-phase and opposed-phase imaging. This can be achieved with a T1-weighted gradient echo sequence. If there is evidence of steatosis within a region of the liver, that area will lose signal (become darker) on the opposed-phase sequence compared to the in-phase sequence102 (Figure 15-35). On ultrasound, steatosis may appear as focal or diffuse areas of increased echogenicity (brighter) of the liver parenchyma, compared to normal parenchyma. The sensitivity of ultrasound is limited, as changes often only become apparent once there is extensive fatty deposition.
Budd–Chiari Syndrome In Budd–Chiari syndrome (BCS) the liver has a mottled appearance on contrast-enhanced CT. There is delayed enhancement in the periphery of the liver and around the hepatic veins (Figure 15-36). The peripheral zones of the liver appear low in density on contrastenhanced CT because there is reversal of portal venous blood flow, related to increased postsinusoidal pressure caused by hepatic venous obstruction.103 Relative atrophy of the right and left hepatic lobes occurs with compensatory enlargement and increased enhancement of the caudate lobe; the caudate lobe is usually spared because it has a separate venous drainage direct to the IVC. Although the CT findings of BCS are often non-specific, when thrombosis of the hepatic veins and the IVC is identified in the appropriate clinical setting the diagnosis of BCS can be made. With MRI the liver appears atrophic and heterogeneous in 64% of patients and venous thrombosis is identified in 86%.104 The affected areas are low (darker) in signal intensity on T1-weighted images and high (brighter) in signal intensity on T2-weighted images, owing to hepatic congestion. The liver periphery enhances poorly on T1-weighted images following intravenous gadolinium contrast administration, particularly during the acute phase of the disease. In patients with chronic disease, nodular regenerative hyperplasia may develop. These nodules have increased signal intensity on T1weighted images and low to intermediate signal intensity on T2weighted images. MRI readily demonstrates other features of the disease as well, such as thrombosis, occlusion, or narrowing of the IVC and hepatic veins. Venous collaterals are readily identified, including comma-shaped intrahepatic varices, which are a characteristic finding, often not visualized on other imaging modalities. These collaterals form to bypass the level of obstruction.105 Ultrasound has a high sensitivity and specificity for the identification of BCS, and can depict the echogenic membrane or fibrous cord in the IVC, which is a common cause of chronic BCS. Initially there may be marked hepatomegaly, often associated with gallbladder wall thickening, ascites, and splenomegaly. Color and Doppler images help demonstrate the underlying vascular abnormality and are able to assess flow within vessels. Nuclear medicine imaging can also suggest the diagnosis of BCS as well as provide an estimate of the amount of normally functioning liver. Technetium-99m (99mTc) sulfur colloid uptake is increased
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
C
D
Figure 15-34. Steatosis on CT. Paired non-contrast and contrast-enhanced CT scans from two different patients reveal the characteristic changes of hepatic steatosis. The normal liver is approximately of similar or slightly increased density as the spleen on non-contrast CT scans. In these non-contrast images (A) and (C) of patients with steatosis, the liver (L) is noted to be less dense (darker) than the spleen (S). After the administration of intravenous contrast, images (B) and (D), the differences in density are further accentuated.
(seen as a region of increased activity or hot spot) in the caudate lobe compared to the rest of the liver, in which uptake may be reduced or absent. This technique has a low specificity, however, as hot spots depicted by 99mTc sulfur colloid scans may occur within residual normal liver segments in patients with liver disease due to other causes.
Sarcoidosis Liver involvement is found in 24–94% of patients with sarcoidosis.106 Changes in the liver occur predominantly in the periportal regions. This process may lead to periportal fibrosis and cirrhosis, resulting eventually in portal hypertension. The imaging findings of hepatic sarcoidosis are non-specific and include hepatomegaly, splenomegaly,
279
Section II. Approach to the Patient with Liver Disease
A
B
C
D
Figure 15-35. Focal fat. (A) Non-contrast axial CT image reveals a hypodense (dark) lesion in the right lobe. (B) The lesion does not enhance with contrast. The lack of local architectural distortion suggests the possibility of focal fat. (C) An in-phase gradient echo MR image through the lesion demonstrates a mildly hyperintense (bright) lesion. (D) On the subsequent opposed-phase image the lesion becomes hypointense (dark), which confirms the diagnosis of focal fatty change.
and eventually features of cirrhosis. Occasionally on contrastenhanced CT and MRI, multiple focal low-density lesions can be identified in the liver and spleen. On MRI, these lesions have low signal intensity (dark) on all sequences compared to normal liver parenchyma. The periportal increase in signal intensity on T2weighted images may reflect the tendency of sarcoid granulomas to appear along the portal tracts.107 On sonography, a pattern of either
280
diffuse increased homogeneous or heterogeneous echogenicity has been reported in patients with hepatic sarcoidosis.108
Amyloidosis The imaging appearance of hepatic amyloidosis is non-specific, and in most cases no focal abnormality can be identified. Diffuse
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
C
D
Figure 15-36. CT appearance of Budd–Chiari syndrome. Several images from a dynamic contrast-enhanced CT scan demonstrate features consistent with Budd–Chiari syndrome. (A) In the early arterial phase there is little hepatic enhancement. In subsequent phases (B,C) there is initial central enhancement within the liver parenchyma with sparing of the periphery. In the final delayed image (D) the periphery shows contrast enhancement.
hepatomegaly due to amyloid deposition is the most common finding, occurring in up to 81% of patients.109 Contrast-enhanced CT is able to depict hepatomegaly and show regions of diffuse or focal decreased parenchymal attenuation, which correspond to amyloid deposition. These areas may demonstrate delayed enhancement. These regions of amyloid deposition may eventually calcify, usually with a punctuate pattern110 (Figure 15-37).
The MRI features of amyloidosis are poorly defined. No significant signal intensity changes are reported on T2-weighted images. Diffuse increase in signal intensity on T1-weighted images has been described. The significance of this is unclear and may be related to hepatic steatosis.111 Ultrasound has a limited role, but a heterogeneous appearance to the hepatic parenchyma may be detected.
281
Section II. Approach to the Patient with Liver Disease
A
B
Figure 15-37. Amyloidosis. (A) Plain film of the abdomen demonstrates hepatosplenomegaly. (B) Axial CT image shows scattered areas of low density (dark areas) within the liver due to amyloid deposition. Calcifications are present in the spleen, which is also consistent with amyloidosis.
IMAGING OF HEPATIC INFECTIONS Pyogenic Liver Abscess Pyogenic liver abscesses may be solitary or multiple. The etiology of solitary abscesses is often elusive; multiple pyogenic abscesses are most often related to biliary disease, malignancy, trauma, or prior surgery. Multiple pyogenic abscesses may be present in clusters or in a diffuse, widely scattered pattern that mimics fungal disease. The clustered pattern is more common and is associated with a local spread of coliform organisms (such as Escherichia coli) via the portal circulation. The diffuse pattern is less common and is usually caused by hematogenous dissemination of Staphylococcus spp. Imaging plays an important role in the diagnosis of pyogenic abscesses as well as in their treatment, through image-guided drainage. CT and ultrasound have been reported to be very sensitive for the detection of pyogenic abscesses – 100% and 94%, respectively – but specificity for both is low.112 MRI can also detect pyogenic abscesses with high sensitivity but plays a limited role because of its inability to be used for image-guided drainage. Pyogenic abscesses are generally categorized as micro (<2 cm) or macro (>2 cm, typically due to a coalescence of smaller lesions). On ultrasound, microabscesses may appear as discrete hypoechoic (dark) lesions or as poorly defined areas with distorted echogenicity. Larger lesions are often ill defined, with a variable internal appearance, ranging from hypoechoic owing to liquefactive necrosis to hyperechoic due to fibrotic change.113 On unenhanced CT microabscesses are typically well defined small hypodense (dark) lesions. With the administration of contrast, these lesions may demonstrate faint wall enhancement as well as perilesional edema.113 These characteristics help to differentiate them from cysts. Larger abscesses are also hypodense but may have
282
Figure 15-38. Bacterial abscess on CT. In this contrast-enhanced CT scan a large irregular hypodense (dark) lesion is seen within the liver. There is subtle rim enhancement (brightness at the edges), which in the correct clinical context is consistent with an abscess.
a more variable appearance, often requiring aspiration for diagnosis (Figure 15-38). They may be either unilocular with smooth external borders, or multilocular with multiple internal septations and irregular margins. Rim enhancement with contrast is relatively uncommon (6%),114 but the administration of contrast and
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
A
B
Figure 15-39. Streptococcal abscess on CT. (A) An arterial-phase contrast-enhanced CT scan demonstrates multiple hypodense (dark) liver lesions. (B) In the portal phase of contrast enhancement the lesions remain dark, but a rim of enhancement becomes visible. These lesions were drained and found to be due to streptococcal abscesses.
enhancement of surrounding normal parenchyma can help to highlight the hypodense abscesses (Figure 15-39). Hepatic metastases may mimic the appearance of pyogenic abscesses on CT. However, unlike abscesses, metastases seldom coalesce.112 The MR appearance of abscesses is variable on both T1 and T2 sequences, depending on the protein content of the lesion. In cases where there is perilesional edema there may be increased T2 signal (bright) of the surrounding liver parenchyma.
Amebic Abscess Entamoeba histolytica is a parasite that is endemic worldwide, infecting approximately 10% of the world’s population. Both ultrasound and CT are sensitive but non-specific for the detection of amebic abscesses.114 Ultrasound is the initial method of choice for investigation. In one series, ultrasound detected 100% of amebic abscesses.112 Five sonographic features of amebic abscesses have been described. They include: no wall echoes; an oval or rounded shape; a homogeneously hypoechoic lesion with low-level internal echoes; location near or touching the liver capsule; and enhanced through transmission. Approximately 30% of amebic abscesses display all five characteristics.115 Amebic abscesses appear as well defined low-density round lesions (10–20 HU) on unenhanced CT.116 With the administration of intravenous contrast there is often enhancement of the wall, and a peripheral zone of edema is present. The central cavity itself may have septations as well as fluid levels (Figure 15-40). One advantage of CT over ultrasound is that CT can more easily detect extrahepatic extension of the abscess (to the pleura, chest wall, portal vein etc.), which is common.113
On MRI, amebic abscesses have homogeneous low intensity (dark) on T1 and high intensity on (bright) T2 sequences. The perilesional zone of edema, when present, appears bright on T2weighted images.
Fungal Abscess Fungal disease involving the liver typically occurs in immunosuppressed patients, such as those with HIV and those with hematologic malignancies. The most common causative agent of hepatic fungal disease is Candida albicans, although Cryptococcus, Histoplasma, and Aspergillus have also been implicated. Ultrasound or CT may be used for the identification of liver lesions greater than 1 cm. If the source of infection is unknown CT is preferred, as the whole body can be imaged, allowing for better localization of other sites of disease. Contrast-enhanced CT is also thought to be superior to ultrasound for the detection of small lesions. To date, no large series have been completed to determine the test characteristics of the various imaging modalities. However, several small studies in immunocompromised patients with hepatic candidiasis have suggested that dynamic contrast-enhanced CT, particularly with the use of arterial-phase images, is highly sensitive (90–100%),117 as is dynamic contrast-enhanced MRI (95–100%).118 Four patterns of appearance have been described for fungal abscesses on ultrasound imaging. The most common but least specific one is that of a discrete hypoechoic nodule. In patients with active disease and a normal white blood cell count, the lesion may also have a bull’s eye appearance with a central region of hyperechogenicity surrounded by a hypoechoic rim. A third pattern is that of a discrete echogenic focus with variable degrees of acoustic
283
Section II. Approach to the Patient with Liver Disease
A
B
C
D
Figure 15-40. Amebic abscess. (A) A contrast-enhanced axial CT image reveals a relatively well defined hypodense (dark) lesion in the right hepatic lobe, found to be an amebic abscess. (B) An image at a lower level reveals internal septations. (C) A subtle rim of contrast enhancement is also appreciated. (D) The thickened colonic wall (arrow) in this image indicates a coexisting amebic colitis.
shadowing. This is most commonly seen in patients with resolving disease. Finally, a ‘wheel within a wheel’ pattern has been described. The inner wheel is comprised of a central hypoechoic region of necrosis. This is surrounded by a hyperechoic rim of inflammatory cells, which is in turn encircled by a peripheral hypoechoic rim of fibrosis.119 On unenhanced CT, fungal abscesses appear as discrete hypodense lesions varying in size from 2 mm to 2 cm. They may also have
284
a target-like appearance consisting of a central hyperdense region surrounded by low density. Following intravenous contrast administration the lesions typically enhance centrally, although they can also exhibit rim enhancements.120 The appearance of these lesions on MRI varies depending upon the stage of disease. In untreated cases, fungal abscesses are dark on T1 and contrast-enhanced sequences but very bright on T2 sequences. In the subacute stage after treatment, the lesions are
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
mildly bright on both T1 and T2, and exhibit enhancement with contrast. A dark ring is usually seen surrounding the lesions on all sequences. Finally, in completely treated cases, the lesions are slightly dark on T1 and normal to slightly bright on T2.113
Echinococcal Disease There are two pathogenic organisms that cause echinococcal disease in humans, Echinococcus granulosum and E. multilocularis. E. granulosum is by far the more common and is the causative agent of hydatid cyst disease. The adult form of the echinococcal tapeworm lives in the intestines of dogs and its eggs are shed in the feces. Accidental ingestion of these eggs is followed by digestion in the small bowel and subsequent delivery via the portal circulation to the liver, which is the most frequently involved organ in this disease (approximately 75% of cases).120 Hydatid cysts are commonly located in the right hepatic lobe. The imaging appearance of these cysts varies depending on the stage of cyst growth. Ultrasound is the examination of choice for the diagnosis of echinococcal liver disease, as it is the most sensitive technique (88–98%) for detecting membranes, septa, and hydatid sand (contents of ruptured daughter cysts).120 Sonographically the cyst wall tends to have a layered appearance, with echogenic outlines separated by a hypoechoic middle layer. In the early stages of disease cysts are often unilocular and homogenously hypoechoic on sonography. However, even these ‘simple cysts’ may have scattered echogenic foci internally, representing hydatid sand. These foci settle dependently, but with patient repositioning they become agitated, creating the ‘snowstorm sign’. The formation of daughter cysts results in a septated, multilocular appearance on ultrasound. When the membranes detach from the pericyst linear serpentine
A
hyperechoic structures can be seen within the cyst, resulting in the ‘waterlily sign’, which is highly specific for hydatid disease.120 Calcifications often develop in the cyst wall and appear as a hyperechoic contour with a cone-shaped acoustic shadow on ultrasound. Although lesions are typically cystic, they can also appear as solid masses, especially as vesicles burst and the interior fills with scolices and hydatid sand. In order to help differentiate such lesions from other solid hepatic masses, it becomes important to look for daughter cysts and detached membranes. CT is often used for the detection of hydatid disease when ultrasound is equivocal. Typically, hydatid cysts appear as well defined hypodense lesions with a distinct wall. Ring-like calcifications of the pericyst are present in 20–30% of cases. As healing occurs, the entire cyst calcifies densely, an appearance that is usually representative of dead or inactive lesions.121 Cyst elements that are seen on ultrasound can be identified on CT as well. The detached laminated membranes are often seen as linear strands of increased density.120 Daughter cysts usually occur in a peripheral location and are typically slightly hypodense relative to the mother cyst on CT120 (Figure 15-41). In general, contrast is not required to diagnose hepatic hydatid cysts, although it is helpful for evaluating the complications of the disease. MRI may also be used to evaluate hydatid cysts. MR often best demonstrates the pericyst, the matrix, and the internal hydatid sand. The pericyst is usually dark on both T1 and T2 sequences. The matrix is dark on T1 and very bright on T2, and daughter cysts are typically dark relative to the matrix on both T1 and T2 sequences.122 E. multilocularis causes the rare multilocular or alveolar form of hydatid disease.113 A hailstorm pattern of indistinct echogenic nodules is often seen on ultrasound. When liquefactive necrosis is present the lesions may be hypoechoic centrally, with an irregular
B
Figure 15-41. Echinococcal disease. (A) The contrast-enhanced axial CT image shown here reveals a complicated cystic lesion found to be due to echinococcal liver disease. Note the linear bright strands (arrows) within the cyst. These represent detached laminated membranes. (B) An image obtained at a lower level further demonstrates the extent of this cystic lesion. Peripherally located daughter cysts (D) as well as detached membranes are also evident. All of these findings are consistent with the imaging appearance of echinococcal disease.
285
Section II. Approach to the Patient with Liver Disease
hyperechoic border. On CT, the lesions are usually irregular and hypodense and do not enhance following intravenous contrast administration. Multiple complications of hydatid cyst disease have been described, ranging from intrahepatic processes such as cyst rupture and infection, to exophytic growth, thoracic involvement, peritoneal seeding, portal vein involvement, and abdominal wall invasion. These complications are often not identified on ultrasound and are better visualized on cross-sectional imaging. Cyst rupture occurs in 50–90% of cases and can be categorized into direct, communicating, and contained ruptures. In contained ruptures, the pericyst remains intact. In communicating ruptures there is communication with the biliary system, and in direct rupture there is disruption of the pericyst and endocyst with perforation into the peritoneum, pleural space, or adjacent hollow viscus. The disruption of the cyst wall can usually be detected on both ultrasound and CT. In addition, in cases of communicating and direct rupture the cyst may shrink and assume a non-spherical shape.120 Another complication of hydatid cyst disease is infection, which usually occurs after rupture and often manifests as an abscess. CT is the method of choice for detecting this complication, as peripheral rim enhancement due to inflammation is often seen.
Schistosomiasis Schistosomiasis is a parasitic disease that may be caused by three species of schistosome: Schistosoma japonicum, S. mansoni, and S. hematobium. Chronic infection with S. japonicum or S. mansoni can cause presinusoidal portal hypertension, leading to cirrhosis and an increased risk of hepatocellular carcinoma. The organisms live in the gastrointestinal tract and lay eggs in the mesenteric veins. These eggs subsequently embolize to the portal vein, where they cause an inflammatory reaction leading to fibrosis. Imaging is often not helpful for the acute diagnosis of schistosomiasis because the characteristic findings are detected only years after the initial infection.123 In cases of S. japonicum infection, ultrasound typically reveals an irregular hepatic surface with an internal mosaic pattern of echogenic septa outlining polygonal areas of normal liver.113 The pathognomonic findings on CT imaging are calcified septa, usually occurring perpendicular to the liver capsule and resulting in a ‘tortoise shell’ or ‘turtle back’ appearance. The liver contour appears irregular, with capsular calcification. There is often extension of the periportal fat deep into the parenchyma, owing to the fibrotic reaction and subsequent retraction.113 Capsular calcifications are not well seen on MR imaging. Septa appear on MRI as dark on T1 and bright on T2 sequences.124 In cases of S. mansoni infection, ultrasound often shows portal vein wall thickening with increased echogenicity. This produces a bull’s eye appearance, with an anechoic portal vein surrounded by a thick fibrous echogenic wall. Other common ultrasound findings include a hypertrophic left hepatic lobe, splenomegaly, and the presence of collateral vessels. Low-density rings are seen on CT scans surrounding portal vein branches that enhance strongly following intravenous contrast administration. The periportal regions appear on MRI as isointense on T1-weighted images with enhancement post contrast, and bright on T2-weighted images.
286
Tuberculosis Tuberculosis involving the liver typically occurs as a result of hematogenous dissemination of the organism. In most cases, tubercular infection of the liver is not well detected on imaging and may simply cause non-specific hepatomegaly. When findings are present, a miliary pattern (multiple microabscesses) is the most common appearance. Focal lesions, known as tuberculomas, appear as round, hypoechoic masses on ultrasound. The CT appearance of tuberculomas is quite non-specific, with hypodense lesions that do not enhance with contrast. The lesions appear on MRI as dark on T1 and dark to isointense on T2. In the healing phase tuberculosis may cause diffuse hepatic calcifications, well seen on CT or ultrasound. The diagnosis of tuberculosis of the liver must be made by biopsy when imaging findings are non-specific.
Viral Hepatitis The role of imaging in the diagnosis of viral hepatitis is limited. In the presence of acute viral infection the findings are non-specific and imaging is obtained primarily to exclude other pathologic processes that may produce similar clinical and laboratory abnormalities, such as hepatic metastases, biliary obstruction, and cirrhosis. Acute viral hepatitis may cause hepatomegaly, often with decreased echogenicity of the liver parenchyma on ultrasound. CT can also demonstrate hepatomegaly and can reveal periportal edema. A diffusely heterogeneous enhancement pattern of the liver may also be observed on CT. Periportal edema appears on MRI as a high T2 signal (bright) around the portal vein. One small retrospective series (48 patients) examining the correlation of MR findings to elevated liver function tests found that irregular patchy enhancement of the liver on arterial-phase contrast-enhanced MR images was sensitive for detecting acute hepatitis (although this was not specific for viral etiologies).125 In the chronic phase of viral hepatitis the only imaging finding may be cirrhosis and possible periportal lymphadenopathy.
Bacillary Peliosis Bacillary angiomatosis is a rare systemic disease caused by the bacteria Bartonella henselae and is usually found in immunocompromised patients. Infection of the liver by this organism is known as bacillary peliosis and is seen almost exclusively in AIDS patients. Bacillary peliosis appears on ultrasound as multiple small round hypoechoic lesions in the liver and spleen. These lesions may be low or high density on contrast-enhanced CT and are typically scattered in location and smaller than 1 cm. Imaging findings tend to be nonspecific, and bacillary peliosis must be differentiated from abscesses due to other organisms, including fungi, viruses, and Pneumocystis carinii, as well as neoplastic lesions such as lymphoma and Kaposi’s sarcoma.113
REFERENCES 1. Golli M, Van Nhieu JT, Mathieu D, et al. Hepatocellular adenoma: color Doppler US and pathologic correlations. Radiology 1994;190:741–744. 2. Grazioli L, Federle MP, Brancatelli G, et al. Hepatic adenomas: imaging and pathologic findings. Radiographics 2001;21:877–894.
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
3. Mortele KJ, Ros PR. Benign liver neoplasms. Clin Liver Dis 2002;6:119–145. 4. Arrive L, Flejou JF, Vilgrain V, et al. Hepatic adenoma: MR findings in 51 pathologically proven lesions. Radiology 1994;193:507–512. 5. Chung KY, Mayo-Smith WW, Saini S, et al. Hepatocellular adenoma: MR imaging features with pathologic correlation. Am J Roentgenol 1995;165:303–308. 6. Hussain SM, Terkivatan T, Zondervan PE, et al. Focal nodular hyperplasia: findings at state-of-the-art MR imaging, US, CT and pathologic analysis. Radiographics 2004;24:3–19. 7. Brancatelli G, Federle MP, Grazioli L, et al. Focal nodular hyperplasia: CT findings with emphasis on multiphasic helical CT in 78 patients. Radiology 2001;219:61–68. 8. Gabata T, Kadoya M, Matsui O, et al. Biliary cystadenoma with mesenchymal stroma of the liver: Correlation between unusual MR appearance and pathologic findings. J Magn Reson Imag 1998;8:503–504. 9. Sada PN, Ramakrishna B. Computed tomography of von Meyenburg complex simulating microabscesses. Australas Radiol 1994;38:225–226. 10. Eisenberg D, Hurwitz L, Yu A. CT and sonography of multiple bile duct hamartomas simulating malignant disease. Am J Roentgenol 1986;147:279–280. 11. Horton KM, Bluemke DA, Hruban RH, et al. CT and MR imaging of benign hepatic and biliary tumors. Radiographics 1999;19:431–451. 12. Trotter JF, Everson GT. Benign focal lesions of the liver. Clin Liver Dis 2001;5:17–42. 13. Leslie DF, Johnson CD, MacCarty RL, et al. Single-pass CT of hepatic tumors: Value of globular enhancement in distinguishing hemangiomas from hypervascular metastases. Am J Roentgenol 1995;165:1403–1406. 14. Nino-Murcia M, Olcott EW, Jeffrey RB, et al. Focal liver lesions: Pattern-based classification scheme for enhancement at arterial phase CT. Radiology 2000;215:746–751. 15. Tsai CC, Yen TC, Tzen KY. The value of Tc-99m red blood cell SPECT in differentiating giant cavernous hemangioma of the liver from other liver solid masses. Clin Nucl Med 2002;27:578–581. 16. Hogemann D, Flemming P, Kreipe H, Galanski M. Correlation of MRI and CT findings with histopathology in hepatic angiomyolipoma. Eur Radiol 2001;11:1389–1395. 17. Tsui WM, Colombari R, Portmann BC, et al. Hepatic angiomyolipoma: a clinicopathologic study of 30 cases and delineation of unusual morphologic variants. Am J Surg Pathol 1999;23:34–48. 18. Mortele KJ, Wiesner W, deHemptinne B, et al. Multifocal inflammatory pseudotumor of the liver: dynamic gadoliniumenhanced, ferumoxides-enhanced, and mangafodipir trisodiumenhanced MR imaging findings. Radiology 2002;12:304–308. 19. Young ST, Paulson EK, Washington K, et al. CT of the liver in patients with metastatic breast carcinoma treated by chemotherapy: findings simulating cirrhosis. AJR Am J Roentgenol 1994;163:1385–1388. 20. Zweibel WJ. Sonographic diagnosis of diffuse liver disease. Semin Ultrasound CT MR 1995;16:8–15. 21. Jones EC, Chezmar JL, Nelson RC, Bernardino ME. The frequency and significance of small (less than or equal to 15 mm) hepatic lesions detected by CT. Am J Roentgenol 1992;158:535–539. 22. Soyer P, Levesque M, Elias D, et al. Detection of liver metastases from colorectal cancer: comparison of intraoperative US and CT during arterial portography. Radiology 1992;183:541. 23. Soyer P, Bluemke DA, Fishman EK. CT during arterial portography for the preoperative evaluation of hepatic tumors: How, when, and why? Am J Roentgenol 1994;163:1325–31.
24. Sica GT, Ji H, Ros PR. CT and MR imaging of hepatic metastases. Am J Roentgenol 2000;174:691–698. 25. Kelekis NL, Semelka RC, Woosley JT. Malignant lesions of the liver with high signal intensity on T1 weighted MR images. J Magn Reson Imag 1996;6:291–294. 26. Semelka RC, Worawattanakul S, Kelekis NL, et al. Liver lesion detection, characterization, and effect on patient management: Comparison of single-phase spiral CT and current MR techniques. J Magn Reson Imag 1997;7:1040–1047. 27. Vogl TJ, Hammerstingl R, Schwarz W, et al. Superparamagnetic iron oxide-enhanced versus gadolinium-enhanced MR imaging for differential diagnosis of focal liver lesions. Radiology 1996;198:881–887. 28. Gallez B, Bacic G, Swartz HM. Evidence for the dissociation of the hepatobiliary MRI contrast agent Mn-DPDP. Magn Reson Med 1996;35:14–19. 29. Valls C, Andía E, Sánchez A, et al. Hepatic metastases from colorectal cancer: preoperative detection and assessment of resectability with helical CT. Radiology 2001;218:55. 30. Ward J, Naik KS, Guthrie JA, et al. Hepatic lesion detection: comparison of MR imaging after the administration of superparamagnetic iron oxide with dual-phase CT by using alternative-free response receiver operating characteristic analysis. Radiology 1999;210:459. 31. Kuszyk BS, Bluemke DA, Urban BA, et al. Portal-phase contrastenhanced helical CT for the detection of malignant hepatic tumors: sensitivity based on comparison with intraoperative and pathologic findings. Am J Roentgenol 1996;166:91–95. 32. Kinkel K, Lu Y, Both M, et al. Detection of hepatic metastases from cancers of the gastrointestinal tract by using noninvasive imaging methods (US, CT, MR imaging PET): a meta-analysis. Radiology 2002;224:748–756. 33. Itola JV, Delbeke D, Sandler MP, et al. Positron emission tomography to stage suspected metastatic colorectal carcinoma to the liver. Am J Surg 1996;171:21–26. 34. Boykin KN, Zibari GB, Lilien DL, et al. The use of FDGpositron emission tomography for the evaluation of colorectal metastases to the liver. Am Surg 1999;65:1183–1185. 35. Bar-Shalom R, Yefremov N, Guralnik L, et al. Clinical performance of PET/CT in evaluation of cancer: additional value for diagnostic imaging and patient management. J Nucl Med 2003;44:1200–1209. 36. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas patients with cirrhosis. N Engl J Med 1996;334:693–699. 37. Yoshikakawa J, Matsui O, Takashima T, et al. Fatty metamorphosis in hepatocellular carcinoma: Radiologic features in 10 cases. Am J Roentgenol 1988;151:717–720. 38. Tanaka S, Kitamura T, Fujita M, et al. Color Doppler flow imaging of liver tumors. Am J Roentgenol 1990:154;509–514. 39. Kim CK, Lim JH, Lee WJ. Detection of hepatocellular carcinoma and dysplastic nodules in cirrhotic liver: accuracy of ultrasonography in transplant patients. J Ultrasound Med 2001;20:47–55. 40. Rode A, Bancel B, Douek P, et al. Small nodule detection in cirrhotic livers: evaluation with US, spiral CT and MR and correlation with pathologic examination of explanted liver. J Comput Assist Tomogr 2001;25:327–336. 41. Kim CK, Lim JH, Lee WJ. Detection of hepatocellular carcinoma and dysplastic nodules in cirrhotic liver: accuracy of ultrasonography in transplant patients. J Ultrasound Med 2001;20:99–104. 42. Farinati F, Gianni S. Surveillance for hepatocellular carcinoma in cirrhosis: Is it cost effective? Eur J Cancer Prev 2001;10:111–115. 43. Miller WJ, Baron RL, Dodd GD 3rd, et al. Malignancies in patients with cirrhosis: CT sensitivity in and specificity in 200 consecutive transplant patients. Radiology 1994;193:645–650.
287
Section II. Approach to the Patient with Liver Disease
44. Petersin MS, Baron RL, Marsh JW, et al. Pretransplantation surveillance for possible hepatocellular carcinoma in patients with cirrhosis: epidemiology and CT based tumor detection rate in 430 cases with surgical and pathologic correlation. Radiology 2000;217:743–749. 45. Lim JH, Kim CK, Lee WJ, et al. Detection of hepatocellular carcinoma and dysplastic nodules in cirrhotic livers: accuracy of helical CT in transplant patients. Am J Roentgenol 2000;175:693–698. 46. Jang HJ, Lim JH, Lee SJ, et al. Hepatocellular carcinoma: are combined CT during arterial portography and CT hepatic arteriography in addition to triple-phase helical CT all necessary for preoperative evaluation? Radiology 2000; 215:373. 47. Taourel PG, Pageaux GP, Coste V, et al. Small hepatocellular carcinoma in patients undergoing liver transplantation: detection with CT after injection of iodized oil. Radiology 1995;197:377–380. 48. Yamashita Y, Mitsuzaki K, Yi T, et al. Small hepatocellular carcinoma in patients with chronic liver damage: Prospective comparison of detection with dynamic MR imaging and helical CT of the whole liver. Radiology 1996;200:79–84. 49. Krinsky GA, Lee VS, Theise ND, et al. Hepatocellular carcinoma and dysplastic nodules in patients with cirrhosis: prospective diagnosis with MR imaging and explantation correlation. Radiology 2001;219:445–454. 50. Ebara M, Fukada H, Kojima Y, et al. Small hepatocellular carcinoma: relationship of signal intensity to histopathologic findings and metal content of the tumor and surrounding hepatic parenchyma. Radiology 1999;210:81–88. 51. Ward J, Robinson PJ. How to detect hepatocellular carcinoma in cirrhosis. Eur Radiol 2002;12:2258–2272. 52. Halvaara J, Tervahartiala P, Isoniemi H, Hockerstadt K. Efficacy of sequential use of superparamagnetic iron oxide and gadolinium in liver MR imaging. Acta Radiol 2002;43: 180–185. 53. Wong LK, Link DP, Frey CF. Fibrolamellar hepatocarcinoma: Radiology, management and pathology. AJR Am J Roentgenol 1982;139:172–175. 54. Adam A, Gibson RN, Soreide O, et al. The radiology of fibrolamellar hepatoma. Clin Radiol 1986;37:355–358. 55. Brandt DJ, Johnson CD, Stephens DH, Weiland LH. Imaging of fibrolamellar hepatocellular carcinoma. AJR Am J Roentgenol 1988;151:295–299. 56. Ichikawa T, Federle MP, Grazioli L, et al. Fibrolamellar hepatocellular carcinoma: Imaging and pathologic findings in 31 recent cases. Radiology 1999;213:352–361. 57. Freeny PC, Baron RL, Teefey SA. Hepatocellular carcinoma: Reduced frequency of typical findings with dynamic contrastenhanced CT in a non-Asian population. Radiology 1992;182:143–148. 58. McLarney JK, Rucker PT, Bender GN, et al. Fibrolamellar carcinoma of the liver: Radiologic–pathologic correlation. Radiographics 1999;19:453–71. 59. Soyer P, Roch A, Levesque M, Legmann P. CT of fibrolamellar hepatocellular carcinoma. J Comput Assist Tomogr 1991;14:533–538. 60. Corrigan K, Semelka RC. Dynamic contrast enhanced MR imaging of the fibrolamellar hepatocellular carcinoma. Abdom Imag 1995;20:122–125. 61. Mattison GR, Glazer GM, Quint LE, et al. MR imaging of hepatic focal nodular hyperplasia: Characterization and distinction from primary malignant hepatic tumors. AJR Am J Roentgenol 1987;148:711–715. 62. Wibulpolprasert B, Dhiensiri T. Peripheral cholangiocarcinoma: Sonographic evaluation. J Clin Ultrasound 1992;20:303–14. 63. Choi BI, Park JH, Kim YI, et al. Peripheral cholangiocarcinoma and clonorchiasis: CT findings. Radiology 1988;169:149–153.
288
64. Lacomis JM, Baron RL, Oliver JH III, et al. Cholangiocarcinoma: Delayed CT contrast enhancement patterns. Radiology 1997;2003:98–104. 65. Fan ZM, Yamashita Y, Harada M, et al. Intrahepatic cholangiocarcinoma: Spin-echo and contrast-enhanced dynamic MR imaging. AJR Am J Roentgenol 1993;161:313–317. 66. Mortele KJ, Ros PR. Cystic focal liver lesions in the adult: Differential CT and MR imaging features. Radiographics 2001;21:895–910. 67. Devaney K, Goodman ZD, Ishak KG. Hepatobiliary cystadenoma and cystadenocarcinoma. A light microscopic and immunohistochemical study of 70 patients. Am J Surg Pathol 1994;18:1078–1091. 68. Korobkin M, Stephens DH, Lee JK, et al. Biliary cystadenoma and cystadenocarcinoma: CT and sonographic findings. AJR Am J Roentgenol 1989;153:507–511. 69. Buetow PC, Buck JL, Pantongrag-Brown L, et al. Biliary cystadenoma and cystadenocarcinoma: Clinicalimaging–pathologic correlations with emphasis on the importance of ovarian stroma. Radiology 1995;196:805–810. 70. Shirkhoda A, Ros PR, Farah J, et al. Lymphoma of the solid abdominal viscera. Radiol Clin North Am 1990 28:785–799. 71. Ginaldi S, Bernardino M, Jing B, et al. Ultrasonographic patterns of hepatic lymphoma. Radiology 1980;136:427–431. 72. Sanders LM, Botet JF, Straus DJ, et al. CT of primary lymphoma of the liver. AJR 1989;152:973–976. 73. Levy DW, Rindsberg S, Friedman AC, et al. Thorotrast-induced hepatosplenic neoplasia: CT identification. AJR Am J Roentgenol 1986;146:997–1004. 74. Naka N, Ohsawa M, Tomita Y, et al. Angiosarcoma in Japan. A review of 99 cases. Cancer 1995;75:989–996. 75. Buetow PC, Buck JL, Ros PR, et al. Malignant vascular tumors of the liver: Radiologic–pathologic correlation. Radiographics 1994;14:153–166. 76. Powers C, Ros PR, Stoupis C, et al. Primary liver neoplasms: MR imaging with pathologic correlation. Radiographics 1994;14:459–482. 77. Makhlouf, HR, Ishak KG, Goodman ZD. Epithelioid hemangioendothelioma of the liver: a clinicopathologic study of 137 cases. Cancer 1999;85:562–582. 78. Ishak KG, Sesterham IA, Goodman MZD, et al. Epithelioid hemangioendothelioma of the liver: A clinicopathologic and follow-up study of 32 cases. Hum Pathol 1994;15:839–852. 79. Radin DR, Craig JR, Colletti PM, et al. Hepatic epithelioid hemangioendothelioma. Radiology 1988;169:145–148. 80. Furui S, Itai Y, Ohtomo K, et al. Hepatic epithelioid hemangioendothelioma: report of five cases. Radiology 1989;171:63–68. 81. Miller WJ, Dodd GD III, Federle MP, Baron RL. Epithelioid hemangioendothelioma of the liver: imaging findings with pathologic correlation. AJR Am J Roentgenol 1992;159: 53–57. 82. Van Beers B, Roche A, Mathieu D, et al. Epithelioid hemangioendothelioma of the liver: MR and CT findings. J Comput Assist Tomogr 1992;16: 420–424. 83. Bartolozzi C, Cioni D, Donati F, et al. Focal liver lesions: MR imaging–pathologic correlation. Eur Radiol 2001;11:1374– 1388. 84. Harbin WP, Robert NJ, Ferrucci JT. Diagnosis of cirrhosis based on regional changes in hepatic morphology: Radiological and pathologic analysis. Radiology 1980;135:273. 85. Ito K, Mitchell DG, Gabata T. Expanded gallbladder fossa: simple MR imaging sign of cirrhosis. Radiology 1999;211:723–726. 86. Takayasu K, Yoshie K, Muramatsu Y, et al. Haemodynamic changes in non-alcoholic (viral) liver cirrhosis studied by computed tomography (CT) arterial portography and CT arteriography. J Gastroenterol Hepatol 1999;14:908–914.
Chapter 15 CRITICAL EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF LIVER IMAGING
87. Ohtomo K, Baron RL, Dodd GD 3rd, et al. Confluent hepatic fibrosis in advanced cirrhosis: appearance at CT. Radiology 1993;188:31–35. 88. Ohtomo K, Baron RL, Dodd GD 3rd, et al. Confluent hepatic fibrosis in advanced cirrhosis: evaluation with MR imaging. Radiology 1993;189:871–874. 89. International Working Party: Terminology of nodular hepatocellular lesions. Hepatology 1995;22:983–993. 90. Sakamoto M, Hirohashi S, Shimosato Y. Early stages of multistep hepatocarcinogenesis: adenomatous hyperplasia and early hepatocellular carcinoma. Hum Pathol 1991;22:172–178. 91. Karahan OI, Dodd GD 3rd, Chintapalli KN, et al. Gastrointestinal wall thickening in patients with cirrhosis: frequency and patterns at contrast-enhanced CT. Radiology 2000;215:103–107. 92. Chopra S, Dodd GD 3rd, Chintapalli KN, et al. Mesenteric, omental, and retroperitoneal edema in cirrhosis: frequency and spectrum of CT findings. Radiology 1999;211:737–742. 93. Bonkovsky HL, Slaker DP, Bills EB, et al. Usefulness and limitations of laboratory and hepatic imaging studies in ironstorage disease. Gastroenterology 1990;99:1079–1091. 94. Siegelman ES, Mitchell DG, Semelka RC. Abdominal iron deposition: Metabolism, MR findings and clinical importance. Radiology 1996;199:13. 95. Jensen PD, Jensen FT, Christensen T, Ellegaard J. Non-invasive assessment of tissue iron overload in the liver by magnetic resonance imaging. Br J Haematol 1994;87:171–184. 96. Bonkovsky HL, Rubin RB, Cable EE et al. Hepatic iron concentration: noninvasive estimation by means of MR imaging techniques. Radiology 1999;212:227–234. 97. Eugene RS, Michael FS, Willis CM. Schiff ’s diseases of the liver, 9th edn. Baltimore: Lippincott Williams & Wilkins, 2002. 98. Dixon A, Walsche J. Computed tomography of the liver in Wilson’s disease. J Comput Assist Tomogr 1984;8:46–48. 99. Ko S, Lee T, Ng S, et al. Unusual liver MR findings of Wilson’s disease in an asymptomatic 2-year-old girl. Abdom Imag 1998;23:56–59. 100. Piekarski J, Goldberg HI, Royal SA, et al. Difference between liver and spleen CT numbers in the normal adult: its usefulness in predicting the presence of diffuse liver disease. Radiology 1980;137:727–729. 101. Alpern MB, Lawson TL, Foley WD, et al. Focal hepatic mass and fatty infiltration detected by enhanced dynamic CT. Radiology 1986;158:45. 102. Siegelman ES, Outwater EK, Vinitski S et al. Fat suppression by saturation/opposed phase hybrid technique: Spin-echo versus gradient-echo imaging. Magn Reson Imaging 1995;13:545–548. 103. Mitchell DG, Nazarian LN. Hepatic vascular diseases: CT and MRI. Semin Ultrasound, CT, MR 1995;16:49–68. 104. Soyer P, Rabenandrasana A, Barge J, et al. MRI of Budd–Chiari syndrome. Abdom Imag 1994;19:325–329. 105. Stark DD, Hahn PF, Trey C, et al. MRI of the Budd–Chiari syndrome. AJR Am J Roentgenol 1986;146:1141–1148.
106. Lehmuskallio E, Hannuksela M, Halme H. The liver in sarcoidosis. Acta Med Scand 1977;202:289–293. 107. Warshauer DM, Lee JK. Imaging manifestations of abdominal sarcoidosis. AJR Am J Roentgenol 2004;182:15–28. 108. Kessler A, Mitchell DG, Israel HL, Goldberg BB. Hepatic and splenic sarcoidosis: ultrasound and MR imaging. Abdom Imag 1993;18:159–163. 109. Brunt EM, Tiniakos DG. Metabolic storage diseases: amyloidosis. Clin Liver Dis 2004;8:915–930. 110. Jacobs JE, Birnbaum BA, Furth EE. Abdominal visceral calcification in primary amyloidosis: CT findings. Abdom Imag 1997;22:519–521. 111. Benson, Hemmingsson A, Ericsson A, et al. Magnetic resonance imaging in primary amyloidosis. Acta Radiol 1987;28:13–15. 112. Barnes PF, Decock KM, Reynolds TN, et al. A comparison of amebic and pyogenic abscess of the liver. Medicine 1987;66:472–483. 113. Mortele KJ, Segatto E, Ros PR. The infected liver: radiologic–pathologic correlation. Radiographics 2004;24:937–955. 114. Ralls PW. Inflammatory disease of the liver. Clin Liver Dis 2002;6:203–225. 115. Ralls PW, Barnes PF, Radin DR, et al. Sonographic features of amebic and pyogenic abscesses: A blinded comparison. Am J Roentgenol 1987;149:499–501. 116. Radin DR, Ralls PW, Colletti PM, Halls JM. CT of amebic liver abscess. Am J Roentgenol 1988;150:1297–1301. 117. Metser U, Haider MA, Dill-Macky M, et al. Fungal liver infection in immunocompromised patients: depiction with multiphasic contrast-enhanced helical CT. Radiology 2005;235:97–105. 118. Semelka RC, Kelekis NL, Sallah S, et al. Hepatosplenic fungal disease: diagnostic accuracy and spectrum of appearances on MR imaging. Am J Roentgenol 1997;169:1311–1316. 119. Pastakia B, Shawker TH, Thaler M, et al. Hepatosplenic candidiasis: wheels within wheels. Radiology 1988;166: 417–444. 120. Pedrosa I, Saiz A, Arrazola J, et al. Hydatid disease: radiologic and pathologic features and complications. Radiographics 2000;20:795–817. 121. Beggs I. The radiology of hydatid disease. Am J Roentgenol 1985;145:639–648. 122. Kalovidouris A, Gouliamos A, Vlachos L, et al. MRI of abdominal hydatid disease. Abdom Imag 1994;19:489–494. 123. Cesmeli E, Vogelaers D, Voet D, et al. Ultrasound and CT of liver parenchyma in acute schistosomiasis. Br J Radiol 1997;70:758–760. 124. Patel SA, Castillo DF, Hibbeln JF, Watkins JL. Magnetic resonance imaging appearance of hepatic schistosomiasis, with ultrasound and computed tomography correlation. Am J Gastroenterol 1993;88:113–116. 125. Martin DR. Magnetic resonance imaging of diffuse liver diseases. Top Magn Reson Imag 2002;13:151–164
289
Section II. Approach to the Patient with Liver Disease
16
EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF BILIARY IMAGING Steven Goldschmid Abbreviations CT computed tomography ERCP endoscopic retrograde cholangiopancreatography EUS endoscopic ultrasound
HIDA MRCP
nuclear scintigraphy magnetic resonance cholangiopancreatography
INTRODUCTION Over the past two to three decades a variety of hepatobiliary imaging techniques have emerged. Non-invasive imaging techniques include magnetic resonance cholangiopancreatography (MRCP), oral cholecystography (OCG), nuclear scintigraphy (HIDA), transabdominal ultrasonography, and computed tomography (CT). Invasive techniques include angiography, direct percutaneous transhepatic cholangiography (PTC), endoscopic ultrasound (EUS), and endoscopic retrograde cholangiopancreatography (ERCP). The technique chosen often depends on the problem being evaluated and the equipment and expertise available. Each imaging technique has its own strengths, limitations, costs, and risks to the patient. The utility of each in a variety of clinical situations will be examined. Studies in which techniques were compared will be the focus of this chapter.
RIGHT UPPER QUADRANT PAIN When patients present with right upper quadrant pain they generally undergo an initial ultrasound or CT examination of the abdomen. Investigators at the University of Pennsylvania reviewed 128 patients who underwent both examinations for acute right upper quadrant pain.1 In 66 patients, ultrasound was performed first. Eighteen patients had either cholecystitis or choledocholithiasis that was eventually confirmed. The sensitivity and specificity of ultrasound in correctly identifying the problem were 83% and 95%, respectively. The sensitivity and specificity of CT in correctly identifying the problem were 39% and 93%, respectively. In those patients who underwent CT examination first, ultrasound examination changed the management in six. In those patients who underwent ultrasound examination first, CT did not change the management in any case. Ultrasound has long been the preferred method of evaluation of patients with acute right upper quadrant pain.2 The advantages of ultrasound are that it is inexpensive, readily available, and noninvasive. Disadvantages mostly relate to the experience of the oper-
OCG PTC
oral cholecystography percutaneous transhepatic cholangiography
ator; however, obesity and the presence of bowel gas make for a less satisfying examination. The common bile duct is considered dilated on ultrasound in most patients if its diameter is greater than 6 mm. Ultrasound has a sensitivity and specificity of 87% and 99%, respectively, in visualizing dilated bile ducts.3 Localization of the level of obstruction is possible in 90% of cases.4 Ultrasound has been compared to MRI in the evaluation of acute right upper quadrant pain as well.5 In this recent study, there was no difference in the diagnosis of gallbladder wall thickening, the presence of gallstones or pericholecystic fluid, or the diagnosis of acute cholecystitis. The sensitivity for acute cholecystitis for ultrasound and MRI was 50%. The specificity was 89% for ultrasound and 86% for MRI. Although the authors concluded that MRI was equivalent to ultrasound in making a diagnosis in patients with acute right upper quadrant pain, it is doubtful that MRI will replace ultrasound in this setting. Ultrasound is a powerful and effective diagnostic method for evaluating acute right upper quadrant pain.6 For this reason, it is doubtful that other studies comparing the sensitivity and specificity of different biliary imaging modalities in acute right upper quadrant pain will be performed.
GALLSTONES AND CHOLECYSTITIS Plain films of the abdomen can diagnose gallstones but lack sensitivity and specificity. Eighty percent of gallstones in western populations are cholesterol stones. Only 20% of cholesterol stones have enough calcium to be seen on plain films. Only 50% of pigment stones are visible. Overall, only 25% of gallstones are visible on an abdominal film. The principal imaging modality for gallstones is transabdominal ultrasonography. Ultrasonography is simple, accurate, and portable; requires no preparation; and requires no exposure to ionizing radiation. For stones measuring more than 2 mm ultrasound has a sensitivity and specificity greater than 95% when the stones exhibit shadowing.7 Ultrasonography can also diagnose acute cholecystitis. The presence of pericholecystic fluid (in the absence of ascites) and gall-
291
Section II. Approach to the Patient with Liver Disease
Figure 16-2. Multiple common bile duct stones identified by MRCP (white arrows). (Courtesy of Dr John Cunningham.) Figure 16-1. Gallbladder sonogram of patient with acute cholecystitis with thickened gallbladder wall (yellow arrow) and multiple gallstones (red arrow).
bladder wall thickening greater than 4 mm (in the absence of hypoalbuminemia) is a good indicator of acute cholecystitis8 (Figure 161). The presence of Murphy’s sign by ultrasonography is highly suggestive of acute cholecystitis. It has a positive predictive value of greater than 90% when cholelithiasis is present.9 Cholescintigraphy (hepatobiliary scan, HIDA, PIPIDA) can also be used to diagnose acute cholecystitis. In fact, this radionuclide imaging test has its greatest utility in acute cholecystitis, as it gives an assessment of cystic duct patency. A positive test will diagnose acute cholecystitis with a sensitivity of 95% and a specificity of 90%. False positives are seen when patients fail to fast before the examination, and in patients that are critically ill. A normal hepatobiliary scan virtually excludes acute cholecystitis.10 There are few direct comparison studies evaluating the sensitivity and specificity of all the different biliary imaging techniques. In this chapter, evidence from the current literature was therefore gathered systematically to assess the sensitivity and specificity of some of the different biliary imaging techniques.11
COMMON BILE DUCT AND GALLBLADDER STONES Many studies have evaluated the value of MRCP in the identification of common bile duct stones12–15 (Figure 16-2). Many authors have proclaimed that MRCP should replace ERCP as a diagnostic procedure, as it is fast, non-invasive, requires no sedation, involves no radiation, and has a sensitivity and specificity close to those of the gold standard, ERCP.13,16,17 Taylor et al.18 proposed that MRCP should precede ERCP to select those patients that require a therapeutic procedure. In their study, the sensitivity and specificity of MRCP for diagnosing choledocholithiasis were 97.9% and 89%, respectively. There are concerns about this study including a significant false positive rate of MRCP. Nine of 46 patients were diagnosed with stones on MRCP that were not found on ERCP. These
292
presumed stones were all small – <5 mm – and could have been missed during ERCP. Of the total 149 patients, ERCP was unsuccessful in 20 and MRCP could not be performed in 8. Once again, these are issues that cloud our true knowledge of the sensitivity and specificity of these tests. Furthermore, the technology associated with MRCP is rapidly changing.19 Comparison of different MR cholangiography techniques produce similar sensitivities and specificities, yet clearly some images are better than others.20 A variety of diagnostic pitfalls are associated with MRCP that can lead to nonvisualization of gallstones, pseudo-obstruction, false duplication, and dilatation.21 Obviously, an experienced team is necessary to avoid these pitfalls. The same rules also apply to ERCP.22 The most comprehensive evaluation of MRCP versus ERCP was reported by Kaltenthaler et al.23 This study reviewed all the literature relating to the clinical and cost effectiveness of MRCP. Studies were excluded if they were published before 1995, or if they did not directly compare MRCP to ERCP. Fifteen studies were included in their evaluation of the sensitivity and specificity of ERCP and MRCP in diagnosing choledocholithiasis. Two studies were clearly outliers, as their sensitivities were extremely poor compared to the other 13. Overall, the mean sensitivity of MRCP was 0.87 (95% CI 0.85–0.91). The median sensitivity was 0.91, with a range of 0.5–1.00. If the two outlier studies are excluded the overall sensitivity for diagnosing choledocholithiasis becomes 0.92 (95% CI 0.89–0.95). The median sensitivity is then 0.93, with a range of 0.81–1.00. Similarly, the overall specificity of MRCP in diagnosing choledocholithiasis is 0.95 (95% CI 0.94–0.97). The median specificity is 0.96, with a range of 0.83–0.99. If the two outlier studies are excluded, the mean specificity is 0.95 (95% CI 0.93–0.97) and the median specificity becomes 0.94, with a range of 0.83–0.99. Very few studies directly compare the sensitivity and specificity of multiple biliary imaging techniques. A study from Ireland revealed MRCP to be comparable to ERCP in the diagnosis of choledocholithiasis, as would be expected, but it was far superior to transabdominal ultrasound24 (Table 16-1). In this study, choledocholithiasis was found in 34 patients. MRCP resulted in three false positive and three false negative studies. There was difficulty with interpretation around the ampullary region, a problem
Chapter 16 EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF BILIARY IMAGING
common with MRCP. A more recent study compared ultrasound, CT, MRCP, and MRI plus MRCP to direct cholangiography (Table 16-2).25 Again, results comparable to those of ERCP are obtained with MRCP. MRCP was superior to CT, which was superior to ultrasound in the diagnosis of choledocholithiasis. Some authors have advocated utilizing EUS in the initial evaluation of patients with suspected choledocholithiasis (Figure 16-3). In a study by Meroni et al.26 8% (47) of patients scheduled for laparoscopic cholecystectomy were found to have one or more liver enzyme abnormalities, with a normal-appearing common bile duct on transabdominal ultrasound. EUS was performed in these patients and common bile duct stones were found in nine; the common bile duct was declared free of stones in the other 38. Only five of the patients with stones by EUS were confirmed to have common bile duct stones, for a positive predictive value of 0.55. In the patients without stones, two of the 38 were eventually found to have common duct stones. At least in the patients without stones, the negative predictive value is reasonable, at 0.95. Although EUS does have fewer complications than ERCP, it seems to be a better test if it is negative and should perhaps be reserved for patients with a low likelihood of disease in their common bile duct. This seemed to hold true in a study performed at the University of Michigan comparing EUS to ERCP in patients with suspected biliary tract disease.27 In this study, 30 patients referred for ERCP were studied with MRCP and EUS 24 hours before their ERCP. EUS was more sensitive than MRCP in the evaluation and detection of choledocholithiasis and biliary strictures. Because the negative predictive value of both tests is excellent, the authors compared the proportions of patients correctly identified as having no biliary tract disease and found EUS to be significantly better (p < 0.02). In patients with a low likelihood of disease, EUS may be the best test to confirm there is a normal biliary tree. In both studies, all patients were unable to successfully complete their tests and significant complications occurred in those undergoing ERCP.
patients with intrahepatic stones. In a Korean study, 49 patients were proved to have intrahepatic stones.28 The sensitivity and specificity of MRCP for detecting intrahepatic stones were 97% and 93%, respectively. The sensitivity and specificity of ERCP for detecting intrahepatic stones were 59% and 93%, respectively. Therefore, in this study MRCP had a statistically significantly higher sensitivity than ERCP in the diagnosis of intrahepatic biliary stones. MRCP and ERCP had equivalent sensitivities and specificities for detecting stones in the common bile duct and gallbladder. It is important to remember that a radiologist was present throughout the MRCP to optimize the images, and in the absence of that level of observation it is difficult to know whether MRCP will prove to be that much better than ERCP. The authors also point out that the low sensitivity of ERCP for detecting intrahepatic stones was due to severe ductal stenosis, which led to incomplete filling of the ducts that contained stones.28 This is a reflection of operator dependence in ERCP, and a more aggressive endoscopist might have been able to visualize the biliary tree more completely. This study highlights some of the difficulties encountered when comparing studies that examine biliary imaging. There will be technical differences related to both the equipment and the technique used to obtain the images. There will be differences attributed to how carefully images are scruti-
INTRAHEPATIC STONES Intrahepatic stones can be difficult to diagnose. There is some controversy as to which imaging technique is best suited to evaluate
Table 16-1. Diagnostic Accuracy of MRCP and Ultrasound Compared to ERCP in the Evaluation of Choledocholithiasis24
Sensitivity Specificity Diagnostic accuracy
Abdominal Ultrasound (%)
MRCP (%)
38 100 89
91 98 97
Figure 16-3. Common bile duct stone (white arrow) identified on initial evaluation for abdominal pain by EUS. (Courtesy of Dr John Cunningham.)
Table 16-2. Diagnostic Accuracy of Biliary Imaging Methods (%)25
Cholelithiasis Choledocholithiasis Intrahepatic duct stone
US
CT
ERCP/PTC
MRCP
MRCP+MRI
71.4 52.9 85.7
75 63.6 100
80 87.5 100
78.5 88.2 71.4
92.9 94.1 85.7
293
Section II. Approach to the Patient with Liver Disease
nized, for example, whether or not a radiologist is present when images are actually being produced. There will be differences in the abilities of the operators, as in the above study, in the ability of the physician to obtain the desired images. And lastly, not all patients will be able to have complete studies performed, for a variety of reasons, including claustrophobia, inability to be adequately sedated, anatomic variance, and inexperience of the staff performing the procedure. The poor sensitivity of ERCP in this study may be related to some of these factors. In other studies, ERCP has been very successful in identifying intrahepatic bile duct stones. Patients with recurrent pyogenic cholangitis often present with intrahepatic stones and biliary strictures. ERCP is frequently the first test of choice and is often successful in identifying intrahepatic strictures and stones.29 Better training and technological advances will continue to improve the ability to diagnose, as well as treat, intrahepatic bile duct stones with ERCP.30
BILIARY STRICTURES Once again, there are no direct comparisons of all the various imaging modalities examining biliary strictures. A meta-analysis reported in the Annals of Internal Medicine included 67 studies of 498 identified that used MRCP to diagnose patients with suspected biliary tract disease.15 Most of the studies included were blinded, had consecutive enrollment, and had some type of gold standard. Overall, MRCP had an adjusted sensitivity of 95% (spread (±2 SD): 75–95%) and an adjusted specificity of 94% (CI 86–99%) in identifying biliary tract abnormalities. However, looking at different imaging endpoints, and taking into account confounders that effect sensitivity and specificity, the authors concluded that MRCP was highly accurate for diagnosing the presence of obstruction, but less accurate at differentiating malignant from benign obstruction. Adjusted sensitivity was 88% (spread (±2 SD): 70–96%), and adjusted specificity was 95% (CI 82–99%) for differentiating malignant from benign biliary obstruction, respectively. Rösch and colleagues31 studied 50 patients with jaundice who were suspected of having biliary strictures. Forty underwent direct cholangiography (ERCP or PTC), MRCP, CT, and EUS. Twenty-six had malignant strictures. The sensitivity and specificity of each test compared to direct cholangiography for the diagnosis of malignancy did not show any statistically significant difference. The authors pointed out, however, that invasive tests may provide the same imaging information and make it possible to obtain tissue or perform therapeutic maneuvers, obviating the need for MRCP in many patients. They concluded that MRCP has a limited clinical role in the differential diagnosis of malignant strictures. In yet another study evaluating biliary strictures, ERCP was compared to MRCP looking at quality of imaging, complete visualization of the biliary tract, and differentiation of malignant from benign disease. Although no statistically significant differences were found, ERCP correctly identified malignant lesions in 76% of patients and MRCP correctly identified malignant lesions in 58% (p = 0.057). After supplementing ERCP with intraductal ultrasound, the accuracy of differentiating malignant from benign lesions increased to 88%, which was significantly better than MRCP (p = 0.0047).32
294
On the other hand, Courbière et al. studied 49 patients with benign and malignant biliary stenosis and compared MRCP to ERCP.33 They reported that 42 (86%) MRCP studies were of excellent quality, compared to only 26 (55%) excellent direct cholangiograms, and in two patients cholangiograms were completely unsuccessful. The sensitivity and specificity of MRCP to detect common bile duct stenosis were 88% and 100%, respectively. The ability of MRCP to identify bile duct obstruction and the level of obstruction was also excellent compared to ERCP, at 97% and 96%, respectively. These authors concluded that MRCP should replace ERCP in biliary tract disease. In a recent review of malignant hilar biliary obstruction, Freeman and Sielaff34 suggest that MRCP, CT, and EUS can supplant the use of invasive cholangiography. They propose that ERCP may be detrimental to patients with malignant biliary obstruction because ERCP may increase the risk of postoperative infection and stenting may interfere with the surgeon’s ability to feel the tumor intraoperatively. Also, there is a significant morbidity associated with ERCP. In good-risk patients they feel a tissue diagnosis is not always necessary, and ERCP frequently fails to provide a tissue diagnosis. There are many that oppose this view: in fact, there is some evidence that a high bilirubin preoperatively worsens outcome, therefore some kind of preoperative stenting is advantageous. Also, few patients are candidates for cure and the morbidity and mortality associated with resection are very poor.35 In a Japanese study with no hospital mortality and no positive ductal margins, morbidity was 40% and the median survival 27 months in patients with malignant bile duct tumors.36
SUMMARY Good tools exist to evaluate abnormalities of the biliary tree. Generally, ERCP and MRCP have distinct advantages over the other imaging modalities mentioned in this chapter. It seems safe to say that ultrasound is a uniformly reasonable first test in patients with acute right upper quadrant pain, or if one is simply looking for evidence of biliary obstruction in the jaundiced patient.37 The low cost and ease of performing ultrasound will probably keep it in this position for some time to come. Overall, MRCP and ERCP have comparable sensitivity and specificity, regardless of the diagnostic abnormality in question. Clearly, differences in studies exist and conflicting information can be found in the literature, some more supportive of ERCP over MRCP, and some more supportive of MRCP over ERCP. We can account for some of these differences by recognizing some of the pitfalls of MRCP imaging described above, realizing that some patients cannot complete an MRCP because of claustrophobia, and knowing that technique, operator, and technology all play a role in obtaining and interpreting images. The frequency with which we are introduced to newer and better techniques is increasing and the time taken to develop better machines keeps shortening, which will lead to improvements in the sensitivity and specificity of MRCP. ERCP is quite operator dependent and the number of true experts in this technique is remarkably smaller than the pool of ERCP endoscopists available. Studies where the failure or complication rate of ERCP exceeds 4–5% should be interpreted cautiously. New tech-
Chapter 16 EVALUATION OF THE SPECIFICITY AND SENSITIVITY OF BILIARY IMAGING
niques and equipment, and better-trained endoscopists, will also lead to improved sensitivity and specificity of ERCP. Which biliary imaging study to use in a given clinical situation depends on many factors, including the quality of the equipment, the experience and expertise of the operators, and the preferences of the surgeon. 38 Also, the need to obtain a tissue diagnosis or perform a therapeutic procedure will play a major role in decision making, as this can only be done with an invasive procedure. In our own institution, the success rate for ERCP is 98%, correct tissue diagnosis is made in 84%, and major complications are seen in less than 4%. This clearly makes ERCP attractive to us in appropriate patients. Therefore, a team of the involved subspecialties interested in biliary disorders, recognizing the nuances inherent to their institution’s expertise and equipment, will bring about the best results in biliary imaging and in the management of these patients. Combining modalities such as hybrid PET/CT, adding new contrast agents, using specific labeled markers, and making direct intraductal ultrasound and imaging more readily available, will add to the armamentarium and improve our ability to diagnose biliary disorders.39–41 We can look forward to ever better imaging modalities in the future, with sensitivities and specificities approaching 100%.
REFERENCES 1. Harvey RT, Miller WT. Acute biliary disease: Initial CT and follow-up US versus intial US and follow-up CT. Radiology 1999;213:831–836. 2. Weltman DI, Zeman RK. Acute diseases of the gallbladder and bile ducts. Radiol Clin North Am 1994;32:933–950. 3. Cooperberg PL, Li D, Wong P, et al. Accuracy of hepatic duct size in the evaluation of extrahepatic biliary obstruction. Radiology 1980;135:141–144. 4. Laing FC, Jeffrey RB, Wing VW, Myberg DA. Biliary dilatation: defining the level and cause by real-time US. Radiology 1986;160:39–42. 5. Oh KY, Gifeather M, Kennedy A, et al. Limited abdominal MRI in the evaluation of acute right upper quadrant pain. Abdom Imag 2003;28:643–651. 6. Nino-Murcia M, Jeffrey RB. Imaging the patient with right upper quadrant pain. Semin Roentgenol 2001;36:81–91. 7. Shea JA, Berlin JA, Escarce JJ, et al. Revised estimates of diagnostic test sensitivity and specificity in suspected biliary tract disease. Arch Intern Med 1994;154:2585–2596. 8. Boland GWL, Slater G, Lu DSK, et al. Prevalence and significance of gallbladder abnormalities seen on sonography in intensive care unit patients. AJR Am J Roentgenol 2000;174:973–977. 9. Ralls PW, Colletti PM, Lapin SA, et al. Real-time sonography in suspected acute cholecystitis. Prospective evaluation of primary and secondary signs. Radiology 1985;155:767–771. 10. Prevot N, Mariat G, Mahul P et al. Contribution of cholescintigraphy to the early diagnosis of acute acalculous cholecystitis in intensive-care-unit patients. Eur J Nucl Med 1999;26:1317–1325. 11. Matowe L, Gilbert FJ. How to synthesize evidence for imaging guidelines. Clin Radiol 2004;59:63–68. 12. Soto JA, Barish MA, Yucel EK, et al. Magnetic resonance cholangiography: Comparison with endoscopic retrograde cholangiopancreatography. Gastroenterology 1996;110:589–597. 13. Barish MA, Yucel EK, Ferrucci JT. Current concepts: Magnetic resonance cholangiopancreatography. N Engl J Med 1999;341:258–264.
14. Zidi SH, Prat F, Le Guen O, et al. Use of magnetic resonance cholangiography in the diagnosis of choledocholithiasis: Prospective comparison with a reference imaging method. Gut 1999;44:118–122. 15. Romagnuolo J, Dardou M, Rahme E, et al. Magnetic resonance cholangiopancreatography: A meta-analysis of test performance in suspected biliary disease. Ann Intern Med 2003;139:547–563. 16. Chan Y-L, Chan ACW, Lam WW, et al. Choledocholithiasis: comparison of MR cholangiography and endoscopic retrograde cholangiography. Radiology 1996;200:85–89. 17. Fulcher AS, Turner MA, Capps GW, et al. Half-Fourier RARE MR cholangiopancreatography: experience in 300 subjects. Radiology 1998;207:21–32. 18. Taylor ACF, Little AF, Hennessy OF, et al. Prospective assessment of magnetic resonance cholangiopancreatography for non-invasive imaging of the biliary tree. Gastrointest Endosc 2002;55:17–22. 19. Fulcher AS, Turner MA, Capps GW. MR cholangiography: Technical advances and clinical applications. Radiographics 1999;19:25–44. 20. Soto JA, Barish MA, Alvarez O, Medina S. Detection of choledochlithiasis with MR cholangiography: Comparison of three-dimensional fast spin-echo and single and multisection halfFourier rapid acquisition with relaxation enhancement sequences. Radiology 2000;215:737–745. 21. Watanabe Y, Dohke M, Takayoshi I, et al. Diagnostic pitfalls of MR cholangiopancreatography in the evaluation of the biliary tract and gallbladder. Radiographics 1999;19:415–429. 22. Baron TH, Fleischer DE. Past, present, and future of endoscopic retrograde cholangiopancreatography: perspectives on the national institutes of health consensus conference. Mayo Clin Proc 2002;77:407–412. 23. Kaltenthaler E, Vergel YB, Chilcott J, et al. A systematic review and economic evaluation of magnetic resonance cholangiopancreatography compared with diagnostic endoscopic retrograde cholangiopancreatography. Health Tech Assess 2004;8:1–89. 24. Varghese JC, Liddell RP, Farrell MA, et al. Diagnostic accuracy of magnetic resonance cholangiopancreatography and ultrasound compared with direct cholangiography in the detection of choledocholithiaisis. Clin Radiol 2000;55:25–35. 25. Zhong L, Yao Q, Li L, Xu J. Imaging diagnosis of pancreatobiliary diseases: A control study. World J Gastroenterol 2003;9:2824–2827. 26. Meroni E, Bisagni P, Fumagalli U, et al. Pre-operative endoscopic ultrasonography can optimize the management of patients undergoing laparoscopic cholecystectomy with abnormal liver function tests as the sole risk factor for choledocholithiasis: a prospective study. Dig Liver Dis 2003;36:73–77. 27. Scheiman, JM, Carlos RC, Barnett JL, et al. Can endoscopic ultrasound or magnetic resonance cholangiopancreatography replace ERCP in patients with suspected biliary disease? A prospective trial and cost analysis. Am J Gastroenterol 2001;96:2900–2904. 28. Kim TK, Kim BS, Kim JH, et al. Diagnosis of intrahepatic stones: Superiority of MR cholangiography over endoscopic retrograde cholangiopancreatography. Am J Roentgenol 2002;179:429–434. 29. Liu CL, Fan ST, Wong J. Primary biliary stones: Diagnosis and management. World J Surg 1998;22:1162–1166. 30. Fairbanks KD, Kalloo AN. Therapeutic endoscopic retrograde cholangiopancreatography: what the future holds. Gastrointest Endosc Clin North Am 2003;13:799–809. 31. Rösch T, Meining A, Frümorgen S, et al. A prospective comparison of the diagnostic accuracy of ERCP, MRCP, CT, and EUS in biliary strictures. Gastrointest Endosc 2002;55:870–876. 32. Domagk D, Wessling J, Reimer P, et al. Endoscopic retrograde chloangiopancreatography, intraductal ultrasonography, and magnetic resonance cholangiopancreatography in bile duct strictures: A prospective comparison of imaging diagnostics with
295
Section II. Approach to the Patient with Liver Disease
33.
34. 35.
36.
296
histopathological correlation. Am J Gastroenterol 2004;99:1684–1689. Courbiere M, Pilleul F, Henry L, et al. Value of magnetic resonance cholangiography in benign and malignant biliary stenosis: comparative study with direct cholangiography. J Comput Assist Tomogr 2003;27:315–320. Freeman ML, Sielaff TD. A modern approach to malignant hilar biliary obstruction. Rev Gastroenterol Disord 2003;3:187–201. Zervos EE, Pearson H, Durkin AJ, et al. In-continuity hepatic resection for advanced hilar cholangiocarcinoma. Am J Surg 2004;188:584–588. Kondo S, Hirano S, Yoshiyasu A, et al. Forty consecutive resections of hilar cholangiocarcinoma with no postoperative
37. 38. 39. 40. 41.
mortality and no positive ductal margins. Ann Surg 2004;240:95–101. Rubens DJ. Hepatobiliary imaging and its pitfalls. Radiol Clin North Am 2004;42: 257–278. Baillie J, Paulson EK, Vitellas KM. Biliary imaging: A review. Gastroenterology 2003;124:1686–1699. Clarke JC. PET/CT ‘Cometh the hour, cometh the machine?’ Clin Radiol 2004;59:775–776. Stahl A, Wieder H, Wester HJ, et al. PET/CT molecular imaging in abdominal oncology. Abdom Imag 2004;29:388–397. Farrell RJ, Agarwal B, Brandwein SL, et al. Intraductal US is a useful adjunct to ERCP for distinguishing malignant from benign biliary strictures. Gastrointest Endosc 2002;56:681–687.
Section II: Approach to the Patient with Liver Disease
TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT (TIPS)
17
Thomas D. Boyer Abbreviations ALT alanine aminotransferase BCS Budd–Chiari syndrome DSRS distal splenorenal shunt FHVP free hepatic vein pressure GAVE gastric antral vascular ectasia
HRS HVPG INR LVP MELD
hepatorenal syndrome hepatic venous pressure gradient international normalized ratio large-volume paracentesis Model for End-Stage Liver Disease
INTRODUCTION Portal hypertension develops commonly in patients with cirrhosis and is manifested by the development of varices, ascites, or hepatic encephalopathy. Once it was realized that portal hypertension was the cause of many of the above problems, therapies were developed in an attempt to relieve the portal hypertension. The most effective way to lower portal pressure is by use of a surgical shunt that decompresses the portal circulation into the systemic circulation. Unfortunately, patients with cirrhosis are a poor surgical risk, especially if they are actively bleeding or have poor hepatic function. Thus, alternatives to surgical shunts have been sought. In 1971 Rosch and colleagues1 suggested that the creation of a portosystemic shunt within the liver itself may be a way to lower portal pressure. However, it was not until the development of stents to keep the intrahepatic tract open that TIPS became a useful method for the treatment of portal hypertension.2–4 Despite the ability to create an intrahepatic shunt, the role of TIPS in the management of most of the complications of cirrhosis was unclear as recently as 1995.5 Variceal bleeding that was refractory to medical management was clearly an uncontroversial use of TIPS. However, the role of TIPS in the prevention of rebleeding and in the control of cirrhotic ascites was not well defined. Subsequently, a number of controlled trials have better defined the role of TIPS in the management of a variety of the complications of cirrhosis. Uncertainties, however, remain as to how TIPS should be used for the treatment of some complications of portal hypertension.
PROCEDURE A TIPS is created in most cases by an interventional radiologist, or on occasion by a specially trained physician. The procedure is performed most commonly using conscious sedation, but if the procedure is going to be prolonged or the patient is unstable (actively bleeding varices) then general anesthesia may be required. Before the procedure, patency of the portal vein and the absence of a hepatoma should be ascertained by ultrasound. The success of achieving portal decompression is >90% in most series.6–12 Recently
PHG PTFE TIPS VOD WHVP
portal hypertensive gastropathy polytetrafluoroethylene transjugular intrahepatic portosystemic shunt veno-occlusive disease wedged hepatic vein pressure
the Society for Interventional Radiology has suggested that successful portal decompression should be achieved in >95% of cases, and if the success rate is lower than this number then the program should review how they are performing the procedure.13,14 TIPS can be placed successfully in the sickest of patients, and yet many will die in the immediate postoperative period (30 days). The cause of death in most of these patients is due to disease progression and perhaps to diversion of portal vein flow (TIPS is a side-toside shunt; Figure 17-1) but is not usually due to complications of the procedure itself, such as intraperitoneal bleeding.12,15–17 In one large series the incidence of fatal complications directly related to the procedure was 1.7% (range 0.6–4.3%). Institutions performing fewer than 150 TIPS had a higher rate of fatal complications than those with more experience.12 The exact number of TIPS that need to be performed by an individual before they are proficient has not been determined, but this is one of the more difficult procedures an interventional radiologist performs, and so extensive training and good support by physicians experienced in the management of patients with cirrhosis is required to achieve success. The gold standard for successful reduction of the hepatic venous pressure gradient (HVPG) in the treatment of varices is to achieve a value of less than 12 mmHg.18–20 This goal has been used to define success with both pharmacologic therapy and TIPS. However, with pharmacologic therapy even a 20% reduction of HVPG has a positive effect on the risk of bleeding. Similarly, in one series rebleeding following TIPS revision was 18% in those whose pressure failed to fall, 7% in those with a 25–50% decline in pressure, and 1% in those whose pressure fell by >50%.21 In another series a reduction of the pressure by at least 50% was associated with a rebleeding rate of 11%, whereas lesser falls in pressure were associated with rebleeding rates of 31%.22 A gradient of <12 mmHg is achievable in most patients, but this appears to increase the risk of hepatic encephalopathy.22 If the shunts could be tailored to the patients, i.e. those with a high risk of encephalopathy might receive a smaller shunt, perhaps the outcome in these higher-risk patients could be improved.22 Pending further studies, the goal of a TIPS when used for bleeding varices is to lower the portal pressure to <12 mmHg. The HVPG that needs to be achieved when the indication is refractory ascites is unclear. This uncertainty is not surprising, as
297
Section II. Approach to the Patient with Liver Disease
A
B
Figure 17-1. (A) Portogram of a patient prior to TIPS creation is shown with a liver parenchymal blush present showing portal perfusion. (B) Following TIPS creation all of the portovenous blood goes through the shunt and the liver is not perfused, i.e. a total shunt.
ascites forms in the cirrhotic patient not only because of the portal hypertension but also because of changes in renal and hepatic function. Thus, simple decompression may not cure the ascites if hepatic and renal function fails to improve. One suggestion was that the gradient needs to be <8 mmHg to achieve control of the ascites in many of these patients.23 However, there is no widespread acceptance of this value and further study is required. The lack of correlation between pressure changes and clinical outcomes is because there is a lack of uniformity on how the pressure is measured when creating a TIPS. When determining portal pressure using the wedged hepatic vein pressure (WHVP), the pressure is measured in the wedged position and then the free hepatic vein pressure (FHVP) is determined. This allows for the measurement of the HVPG (WHVP-FHVP), which corrects for intraabdominal pressure.18 After the creation of a TIPS the portal and hepatic veins are connected and most radiologists use the right atrial pressure to calculate the HVPG. The right atrium is in the chest, and thus the pressure gradient is influence by the difference in pressure between abdomen and chest. A true HVPG is not determined when using the right atrial pressure. More accurate pressures would be obtained if the inferior vena cava pressure were used as the intra-abdominal reference point, but this has not been accepted by the radiologic community.
CONTRAINDICATIONS AND PREDICTORS OF SURVIVAL The contraindications to the creation of a TIPS are shown in Table 17-1. Most of these contraindications must be considered relative rather than absolute. For example, patients with congestive heart failure or severe pulmonary hypertension (mean pulmonary artery pressure >45 mmHg) are not candidates for a TIPS as they are not candidates for liver transplantation, and may have worsening of their heart failure or pulmonary hypertension following a TIPS (see Chapters 23 and 24). However, patients with mild pulmonary hypertension or a history of heart failure that is currently well controlled may
298
Table 17-1. Contraindications to Placement of a TIPS Absolute Primary prevention of variceal bleeding Congestive heart failure Multiple hepatic cysts Uncontrolled systemic infection or sepsis Unrelieved biliary obstruction Severe pulmonary hypertension Relative Hepatoma, especially if central Obstruction of all hepatic veins Portal vein thrombosis Severe coagulopathy (INR > 5) Bilirubin > 5 mg/dl Thrombocytopenia < 20 000/cm3 Moderate pulmonary hypertension
well benefit from a TIPS. Similarly, the presence of portal vein thrombosis or a large hepatic tumor is a relative contraindication, but in skilled hands a TIPS can be safely created in these patients. When there are contraindications it is important that experienced physicians are involved in the decision as to whether or not to create a TIPS in this difficult group of patients. The level of hepatic and renal function is an important predictor of post-TIPS survival, and laboratory tests performed before the decision to create a TIPS should include serum electrolytes, BUN and creatinine, a complete blood count, serum bilirubin, albumin, AST, ALT, and prothrombin time. A Doppler ultrasound should be obtained to document hepatic and portal vein patency and the absence of liver masses. Although pulmonary hypertension and cardiac dysfunction are common in this group of patients, routine performance of a cardiac echo is not required unless the patient has signs or symptoms suggestive of cardiac disease. One-year survival rates for patients undergoing a TIPS for bleeding varices vary from 48 to 90%. Survival rates following TIPS
Chapter 17 TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT (TIPS)
Table 17-2. Concordance Statistics for Prediction of Survival Following Tips Using Three Models
Table 17-3. Complications of TIPS
Mortality
MELD (95% CI)
Child–Pugh (95% CI)
Emory25 (95% CI)
Complications
3 month 12 month 36 month
0.77 (0.61–0.94) 0.78*# (0.67–0.89) 0.79 # (0.68–0.9)
0.77 (0.63–0.91) 0.67 (0.55–0.80) 0.70 (0.57–0.82)
0.81 (0.68–0.94) 0.65 (0.55–0.75) 0.64 (0.54–0.73)
TIPS dysfunction Thrombosis Occlusion/stenosis Transcapsular puncture Intraperitoneal bleed Hepatic infarction Fistulas Hemobilia Sepsis Infection of TIPS Hemolysis Encephalopathy New/worse Chronic Stent migration or placement into portal vein or IVC
*MELD significantly better than Child–Pugh. #MELD significantly better than Emory. Adapted from Schepke M, Roth F, Fimmers R, et al. Comparison of MELD, Child–Pugh, and Emory Model for the prediction of survival in patients undergoing transjugular intrahepatic portosystemic shunting. Am J Gastroenterol 2003;98:1167–1174.
creation when the indication has been ascites are similar, at 48–76%.24–29 A number of models have been created to predict the survival of patients before TIPS creation.24–28 The modified Model for End-Stage Liver Disease (MELD) was the original model.26 MELD utilizes serum bilirubin, International Normalized Ratio (INR) for prothrombin time, and serum creatine. The three variables are used in the following equation: [3.8 loge (bilirubin [g/dl) + 11.2 loge (INR) + 9.6 loge (creatinine [mg/dl])]. Another model was developed based on the findings that four variables independently predicted survival following TIPS. These were: bilirubin >3.0 mg/dl (1 point), ALT >100 IU/l (1 point), pre-TIPS encephalopathy (1 point), and urgency of TIPS (2 points). Patients were divided into three groups (low risk – 0 points, medium risk – 1–3 points, high risk – 4–5 points) and survival was significantly worse in those who were considered high risk.25 These two models and the Child–Pugh score were used to predict survival following TIPS in a subsequent study.29 The concordance statistics for each of the models are shown in Table 17-2. All of the models are good at predicting 30-day mortality following a TIPS, whereas MELD is better at predicting long-term survival. Short-term survival has also been predicted using only bilirubin, the APACHE-II score, and the need for emergency TIPS.30,31 Irrespective of which tests are used to determine mortality, the information should be used to decide whether or not the risk of death is too great to perform the procedure in any environment other than a transplant center. Also, if the patient has a poor 1-year survival prognosis they should be referred to a transplant center following completion of the TIPS.
COMPLICATIONS (Table 17-3) Dysfunction is the most common complication of TIPS, and it is this that creates the need for frequent monitoring and reintervention to maintain shunt patency. TIPS dysfunction is said to occur when there is a loss of the portal system decompression that had been originally achieved by the shunt. Most physicians feel that when the HVPG rises to above 12 mmHg or the complication for which the TIPS was originally created recurs, TIPS dysfunction is present and intervention is required.32,33 TIPS dysfunction may be due to either thrombosis or endothelial hyperplasia. Thrombosis of the TIPS may occur within the first 24 hours after creation of the TIPS, and usually within the first few weeks. Thrombosis is observed in 10–15% of cases and is thought to be due to leakage of bile into the stent, hypercoagulable syndromes, or inadequate
Frequency (%) 10–15 18–78 33 1–2 ~1 Rare <5 2–10 Rare 10–15 10–44 5–20 10–20*
*Data taken from experience at time of transplant and incidence probably overestimated Data from references 70 and 71.
coverage of the TIPS tract with the stent.34–38 The presence of the thrombosis is determined by Doppler ultrasound, and repeat catheterization is required to restore patency. There is little compelling evidence that the use of anticoagulation reduces the incidence of shunt thrombosis.38,39 Pseudointimal hyperplasia is common following TIPS and leads to stenosis or occlusion of the shunt. The reported incidence varies from 18 to 78%, and the latter is more reflective of the experience of most large centers.7,9,10,40–44 The occluded stent is covered by a fibrinous exudate composed of connective tissue, mesenchymal cells, and a single layer of endothelial cells on the luminal surface.36,37,45,46 Doppler ultrasound is commonly used to identify TIPS dysfunction due to pseudointimal hyperplasia. Unfortunately, the sensitivity of this test is not very good, varying between 10 and 26%. The specificity is better at 88–100%, and the positive and negative predictive values are poor to acceptable.47,48 When the ultrasound findings are correlated with the presence or absence of TIPS insufficiency based on the pressure gradient, the sensitivity and specificity are even worse, being 86% and 48%, respectively.49 The most predictive factor for TIPS dysfunction is the recurrence of the portal hypertension for which the TIPS was originally created, including the endoscopic reappearance of varices.47 A normal ultrasound in the presence of recurrent ascites or varices, or rebleeding from the varices, does not rule out TIPS dysfunction and repeat catheterization of the shunt is required. TIPS dysfunction has been the complication that has most limited the usefulness of TIPS in the treatment of complications of portal hypertension. In a recently completed controlled trial in which TIPS was compared to distal splenorenal shunt for the prevention of rebleeding from varices, both TIPS and the surgical shunt were excellent methods, with similar rates of survival. However, only 11% of the surgical patients required reintervention to maintain patency, whereas 82% of the TIPS patients required a reintervention.72 The development of coated stents may reduce the need for reintervention. In one series of 71 patients who received coated stents, the primary rates of patency at 6 and 12 months were 87%
299
Percentage patients without shunt dysfunction
Section II. Approach to the Patient with Liver Disease
120
Table 17-4. Indications for TIPS
100
Indication
Efficacy
Primary prevention variceal bleeding Secondary prevention variceal bleeding Refractory cirrhotic ascites Refractory acutely bleeding varices Portal hypertensive gastropathy Bleeding gastric varices Gastric antral vascular ectasia Refractory hepatic hydrothorax Hepatorenal syndrome Type 1 Type 2 Budd–Chiari syndrome Veno-occlusive disease Hepatopulmonary syndrome
None Excellent (controlled trials) Good (controlled trials) Excellent Good Excellent None Good
80 60 40 20
PTFE Bare
0 0
6
12
18
Month following TIPS Figure 17-2. Frequency of TIPS insufficiency using the PTFE-coated stents vs the bare stents. (Data taken from Bureau C, Garcia-Pagan JC, Otal P, et al. Improved clinical outcome using polytetrafluoroethylene-coated stents for TIPS: Results of a randomized study. Gastroenterology 2004;126:469–475).
and 81%, respectively.50,51 Primary patency rates with bare stents would be 30–50% at best. A recently completed randomized controlled trial has further clarified the role of coated stents in the prevention of TIPS insufficiency. Eighty patients were randomized to receive either bare or polytetrafluoroethylene (PTFE)-covered stent grafts and were followed at frequent intervals with Doppler ultrasound and venography. The probability of remaining free of TIPS dysfunction in the two groups is shown in Figure 17-2. Shunt dysfunction was seen in 12.8% of the coated shunts compared to 43.9% of the bare stents (p <0.001). Recurrences of the complications of portal hypertension were also seen more commonly in those receiving bare rather than coated stents, but survival was the same.32 The PTFE-coated stents are available in Europe, and with the completion of a trial in the USA they also should become available in this country in the near future.52 The use of the PTFE-coated stents does not mean that the shunts need not be monitored. Clearly, TIPS dysfunction will occur irrespective of the type of stent used, and if the patients are not monitored then failure will occur. At this time the sensitivity and specificity of Doppler ultrasound for identifying TIPS dysfunction in patients with the coated TIPS are unknown. Hepatic encephalopathy is another complication of TIPS that has been of concern. De novo or worsening of pre-existing encephalopathy is seen in 20–41% of cases.24,32,53,54 Factors predictive of postTIPS encephalopathy include female gender, hypoalbuminemia, increasing age, previous encephalopathy, and encephalopathy at the time of TIPS creation,53,54 none of which is an absolute contraindication to TIPS placement. In addition, encephalopathy that occurred because of variceal bleeding is unlikely to recur if the TIPS successfully reduces the portal pressure, and therefore even severe encephalopathy following a variceal bleed should not be considered a contraindication to a TIPS. If encephalopathy develops following TIPS placement most patients respond to standard treatment with lactulose or neomycin. If the patient develops refractory encephalopathy, then either the TIPS can be occluded or its diameter can be reduced to help with the encephalopathy.55–57
300
Uncertain Good Good None Uncertain
During the initial stages of TIPS creation it is common for the liver capsule to be punctured. Fortunately, serious intraperitoneal bleeding is uncommon, at 1–2%. Creation of a biliary–venous or hepatic artery–portal vein fistula is rare, as is hepatic infarction due to injury of the hepatic artery.58–60 Before the TIPS stent is covered with endothelial cells, red blood cells may be damaged during passage through the stent, leading to hemolysis.61–63
INDICATIONS FOR TIPS (Table 17-4) The indications for TIPS creation are shown in Table 17-4. The efficacy of TIPS is judged to be excellent if the complication of portal hypertension is controlled significantly better than with other therapies; good if the benefit is balanced by complications such as encephalopathy compared to other therapies; none if outcomes are no better or worse than with alternative therapies; and unknown if there are insufficient data to make a judgment.
PRIMARY PREVENTION OF VARICEAL BLEEDING Treatments used to prevent bleeding in patients with varices that have never bled are termed primary prevention/prophylaxis, and bblockers are currently considered the best therapy for this group of patients (see Chapter 20). Previously, when surgical shunts were developed for the treatment of portal hypertension it was thought that they would be an effective way to prevent bleeding in patients with varices that had never bled. Unfortunately, although less bleeding was seen in the patients who underwent surgical shunting, higher rates of mortality due to hepatic failure were observed compared to the control groups of patients.64 The increased incidence of hepatic failure was thought to be due to the diversion of portal venous blood away from the liver and into the shunt. As a TIPS is a side-to-side shunt (Figure 17-1) a similar diversion of portal blood will occur, thereby increasing the risk of hepatic failure. The use of TIPS to decompress the portal venous system before abdominal surgery such as a liver transplant is also without merit, as there is no evidence that this reduces operative time or the need for blood transfu-
Chapter 17 TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT (TIPS)
sion.65,66 Thus the use of TIPS to prevent bleeding in patients who have never bled is not warranted.
TIPS IN THE ACUTELY BLEEDING PATIENT WHO HAS FAILED MEDICAL THERAPY Acute variceal bleeding is one of the dreaded complications of portal hypertension and is associated with significant morbidity and mortality. Although progress has been made in salvaging this group of patients, 20% will die within 6 months of the variceal bleed.67 One of the factors most predictive of death is continued bleeding despite medical therapy, or early rebleeding in the hospital.68 It has been in the patients who are medical failures that TIPS has been used as salvage therapy. In one report analyzing 15 studies in which TIPS was used to control acutely bleeding varices, bleeding was controlled in 94% of cases.69 Unfortunately, mortality at 6 weeks was 36%. This high rate of mortality is not unexpected, as urgency is an independent predictor of postprocedure mortality from TIPS.25 High mortality rates were also observed when surgical shunts were used to control acutely bleeding varices.68 Preoperative tests of liver function can be used to predict postprocedure mortality (see above), and these determinations should be made before deciding whether or not a TIPS should be created in the actively bleeding patient. A high risk of death following a TIPS is not an absolute contraindication to the procedure in the patient bleeding to death from varices, but the risks should be conveyed to the patient and/or their family before undertaking the procedure.
SECONDARY PREVENTION OF ESOPHAGEAL VARICEAL BLEEDING IN PATIENTS WHO HAVE FAILED MEDICAL THERAPY The prevention of recurrent bleeding in the patient who has bled at least once from varices is termed secondary prevention or prophyl-
axis. It was this indication for which TIPS was originally developed. The hope for TIPS compared to surgical shunts was that both approaches would be effective in the prevention of rebleeding, but with lower rates of encephalopathy as the size of the TIPS could be varied to allow for portal decompression but continued perfusion of the liver. Unfortunately, in most patients, to achieve a pressure of <12 mmHg 10 mm stents are required, and portal flow is diverted through the shunt. In addition, no controlled trials comparing outcomes according to the size of stent used have been published. TIPS has been compared to surgical shunts in only one controlled trial, published in preliminary form, and both TIPS and surgical shunts have been compared to endoscopic therapy in a large number of patients. It is useful to compare the two decompressive procedures to endoscopic therapy (Table 17-5). Irrespective of whether a surgical shunt or a TIPS is used to decompress the portal venous system, the rebleeding rates with decompression were significantly better than those seen with endoscopic therapy, but at the cost of an increase in the incidence of encephalopathy and with no impact on survival (Table 17-5).68,74 The cost of TIPS was greater than endoscopic therapy because of the need for reintervention to maintain TIPS patency.73 Similar results were observed when TIPS was compared to pharmacologic therapy. The risk of rebleeding was three times greater in those who received pharmacologic therapy, but the risk of encephalopathy was twofold less than in those who received a TIPS. Survival was the same in both groups, but the cost was significantly higher for TIPS.75 Most experts agree that TIPS should be used as salvage therapy for patients with variceal rebleeding who have failed medical therapy, and not as the initial approach to management. Therefore, the most important issue with TIPS is how it compares to surgical therapies. There is one published trial in which TIPS was compared to H-graft shunts. Unfortunately, the patients were not randomized but done as pairs, which created bias in the study. The investigators found rebleeding rates of 16% in the TIPS group and 3% in the shunted group, with no effect on survival.76 A randomized controlled trial has been published in preliminary form in which TIPS was compared to distal splenorenal shunt (DSRS).72 The results of the study are summarized in Table 17-5 and rebleeding rates, the incidence of encephalopathy and survival were the same in both groups.
Table 17-5. Controlled Trials of Endoscopic Therapy vs Surgical Shunts or TIPS and TIPS vs Surgical Shunt Number of patients 376 811 140
Rebleeding rate (%) Endo 49.8 Endo 46.6 DSRS 5.5
PCS 12.4* TIPS 18.9* TIPS 9
Encephalopathy Endo 8.6 Endo 18.7 DSRS 37
Mortality PCS 17.2** TIPS 34.0** TIPS 24
Endo 28.8 Endo 26.5 DSRS 38
PCS 28.8 TIPS 27.3 TIPS 36
PCS, portocaval shunt; Endo, endoscopic therapy; DSRS, distal splenorenal shunt. *By meta-analysis rebleeding was significantly less with PCS or TIPS than with Endo. **By meta-analysis the incidence of encephalopathy was greater with PCS or TIPS than with Endo. Data taken from DAmico G, Pagliaro L, Bosch J. The treatment of portal hypertension: A meta-analytic review. Hepatology 1995;22:332–353. Henderson JM, Boyer TD, Kutner M, et al. DSRS vs TIPS for refractory variceal bleeding: a prospective randomized controlled trial. Hepatology 2004;40:presented as late breaking abstract AASLD. Papatheodoridis GV, Goulis J, Leandro G, et al. Transjugular intrahepatic portosystemic shunt compared with endoscopic treatment for prevention of variceal rebleeding: A meta-analysis. Hepatology 999;30:612–622.
301
Section II. Approach to the Patient with Liver Disease
In both series72,76 frequent reinterventions were required in the TIPS patients to assure patency and thus prevent rebleeding. With the need for reintervention the cost of TIPS may be greater than the cost of surgical shunting, and this issue is examined in the study by Henderson.72 It can be expected that with the use of PTFE-coated stents the incidence of shunt dysfunction will be significantly decreased, and thus the need for reintervention should decline as well.32 Unfortunately, despite the use of the coated stents, shunt dysfunction will occur and monitoring will be required. Given the poor sensitivity and specificity of Doppler ultrasound in identifying shunt dysfunction, reinterventions will still be required with the new stents to document patency and prevent rebleeding. Therefore, how much cost saving will be seen with the coated stents is unclear. Based on the published data and those of Henderson et al.,72 both TIPS and surgical shunts are equally effective for the prevention of variceal rebleeding. The choice as to which approach to use should be based on the availability of physicians trained in the procedure and the cost.
SECONDARY PREVENTION OF GASTRIC VARICEAL BLEEDING Gastric varices are difficult to treat because standard therapies such as variceal band ligation and sclerotherapy are ineffective, and in many countries agents such as glue are not available. TIPS has therefore been considered an excellent method for the management of gastric varices that have bled at least once. Given the absence of good alternatives, no controlled trials comparing TIPS to other therapies have been published. In small published series TIPS was equally effective in controlling acute bleeding and preventing rebleeding when the indication was either esophageal or gastric variceal bleeding.77–80 Care must be exercised when using TIPS in the treatment of gastric varices. If after the creation of the TIPS there is persistent filling of the gastric varices there is a risk of early rebleeding and the varices should be embolized, as shown in Figure 17-3.
A
PREVENTION OF REBLEEDING FROM PORTAL HYPERTENSIVE GASTROPATHY (PHG) AND GASTRIC ANTRAL VASCULAR ECTASIA (GAVE) PHG occurs commonly in patients with portal hypertension, is most prominent in the fundus and body of the stomach, and may be associated with acute and chronic blood loss. GAVE can be seen in patients with and without portal hypertension, is localized to the antrum, and can be associated with severe blood loss in the patient with portal hypertension.81 TIPS has been used in patients with both PHG and GAVE, and has been associated with endoscopic improvement and a decrease in bleeding in those with PHG but not in those with GAVE.82,83 Thus, if patients have recurrent bleeding from PHG they are candidates for creation of a TIPS, whereas with GAVE the only option is liver transplantation.
CIRRHOTIC ASCITES The development of ascites in the cirrhotic patient is due to both portal hypertension and disorders of renal function (see Chapters 19 and 22). Ascites can be controlled in most patients with a combination of diuretics and sodium restriction. However, about 10% will develop refractory ascites, defined as a failure to respond to a high dose of diuretics (400 mg/day spironolactone plus 160 mg/day furosemide), or they develop complications of therapy, such as hyponatremia or renal insufficiency, that prevent a successful diuresis.84 Once patients develop refractory ascites their prognosis is poor, with 50% dying within a year.85 A number of approaches have been taken to manage these patients, including peritoneovenous shunts, repeated large-volume paracentesis (LVP) and TIPS. Peritoneovenous shunts have been abandoned because of a lack of efficacy and a high rate of complications.86 The efficacy of TIPS versus LVP in the control of refractory cirrhotic ascites has been compared in four
B
Figure 17-3. (A) Portogram following TIPS creation with continued filling of gastric varices (arrow). (B) Following embolization of the collateral, no further flow into the gastric varices is seen (arrow).
302
Chapter 17 TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT (TIPS)
60 Percentage of patients
TIPS
LVP
50 40 30 20 10 0 Ascites improved
2 year survival
Encephalopathy
Figure 17-4. Comparison of TIPS vs LVP in the treatment of refractory cirrhotic ascites. (Data taken from references 23, 86–88 and combined, and % of patients who showed improvement in ascites, survival and encephalopathy calculated.) TIPS, transjugular portosystemic shunt; LVP, large-volume paracentesis.
controlled trials23,86–88 (Figure 17-4). Ascites was shown to improve in 38–84% of patients following TIPS, and in 0–43% following LVP. Survival was unaffected irrespective of which treatment was selected, and encephalopathy tended to be slightly more common in those receiving a TIPS (Figure 17-4). None of the studies looked at cost as a variable. In conclusion, TIPS appears to be good at reducing the need for repeated LVP; however, there is no improvement in survival and an increased risk of hepatic encephalopathy. Therefore, the use of TIPS in the patient with refractory cirrhotic ascites should be limited to those who can no longer tolerate LVP or who are developing renal insufficiency. TIPS should be considered a temporizing measure until the patient can undergo liver transplantation.
HEPATIC HYDROTHORAX REFRACTORY TO MEDICAL TREATMENT Ascitic fluid that enters the pleural space via pores in the diaphragm is termed hepatic hydrothorax. In the majority of patients hepatic hydrothorax develops in the right lung and is responsive to diuretic therapy. Some patients, however, undergo repeated thoracenteses to control their symptoms despite the use of diuretics, and are said to have hepatic hydrothorax that is refractory to medical therapy.89 Peritoneovenous shunts or thoracotomy, in an attempt to identify and repair the defect in the diaphragm, have had little success in this group of patients. TIPS has been used for the treatment of hepatic hydrothorax in a series of small studies.90–92 All found that TIPS reduced or eliminated the need for thoracentesis as long as the shunt remained patent. Survival, however, was poor in this group of patients, and they should also be considered for liver transplantation irrespective of the ability of TIPS to control the hydrothorax.
HEPATORENAL SYNDROME (HRS) (Chapter 22) The development of renal insufficiency in the patient with cirrhosis is termed HRS if there is no definable cause such as sepsis or expo-
sure to nephrotoxic drugs. Patients with HRS have a very poor prognosis, and thus serum creatinine is one of the critical components of the MELD scoring system.26 If the renal failure develops rapidly, over a period of 2 weeks or less, then the patient is said to have type 1 HRS and this group has a very poor prognosis. The more gradual development of HRS is termed type 2, and these patients have a better prognosis.84,85 Creation of a TIPS is associated with improvements in glomerular filtration rates and renal plasma flow and falls in plasma renin activity and plasma aldosterone levels in patients with cirrhotic ascites, suggesting that TIPS may be an effective therapy for HRS.93–95 However, despite the creation of a TIPS in patients with type 1 and 2 HRS, survival was still poor, being 20% and 45%, respectively, at 1 year.93 Of greater concern is the fact that many patients with type 1 HRS have high MELD scores or are jaundiced, and thus have a high 30-day mortality following TIPS. To better define how TIPS should be used in the management of HRS, controlled trials comparing TIPS to other therapies are needed.
BUDD–CHIARI SYNDROME (BCS) (see Chapter 45) BCS develops because of obstruction to the venous outflow from the liver, due to obstruction of the hepatic veins, the inferior vena cava or both.96–98 Previously it was believed that the best management of these patients was portal venous decompression using surgically placed side-to-side portacaval shunts. However, more recently it has become apparent that many patients can be managed with anticoagulation alone, whereas for those with more advanced disease liver transplantation is the only option.98 Thus, the number of patients with BCS who require decompression is limited. Experience suggests that those with chronic BCS and refractory ascites are reasonable candidates for a TIPS. In contrast, in those who received a TIPS for acute hepatic failure due to BCS the rate of early mortality was 50%.99–100 Given the rarity of BCS it is unlikely that controlled trials will be performed in which TIPS is compared to alternative forms of therapy. Given this lack of data, TIPS should only be used in patients with an intermediate prognosis and refractory ascites, whereas those with less advanced disease can receive anticoagulation and those with more severe disease a liver transplant. For patients with acute hepatic failure due to BCS liver transplantation is the best option, and TIPS should only be considered a bridge to transplantation.
VENO-OCCLUSIVE DISEASE (VOD) (see Chapter 46) VOD is seen most commonly following hematopoietic stem cell transplantation or after the ingestion of certain plants, such as bush tea. The syndrome varies in severity, with most patients having mild disorders of liver function and weight gain. For others, however, the disease is severe, with the development of ascites and hepatic failure. TIPS has been used in a few patients with severe disease, and in most cases ascites and liver tests improved but the patients still died of multiorgan system failure.101–104 Thus, TIPS is not considered an effective form of therapy for VOD.
303
Section II. Approach to the Patient with Liver Disease
HEPATOPULMONARY SYNDROME (see Chapter 24) Patients with cirrhosis may develop hypoxemia because of the shunting of blood through arteriovenous fistulas in the lung. TIPS has been used in a few patients with hepatopulmonary syndrome and oxygenation improved.105 Why creating a portosystemic shunt should reduce shunting in other organs is unclear, and the use of TIPS for hepatopulmonary syndrome is not warranted.
CONCLUSIONS TIPS has become a useful therapy to control the complications of portal hypertension in patients with cirrhosis. The successful creation of a TIPS leading to decompression of the portal venous system requires significant skill on the part of the interventional radiologist. In addition, the decision to create a TIPS in a patient with cirrhosis should be made by a physician skilled in the management of patients with liver disease, in concert with the interventional radiologist. If the patient is at high risk of dying following a TIPS, a liver transplantation center should be contacted to determine their eligibility for a transplant prior to a TIPS procedure, except in an emergency. In controlled trials TIPS has been shown to be as effective as surgical shunts in the prevention of rebleeding from esophageal varices. TIPS is also an excellent way to control acute bleeding in the patient who has failed medical therapy, to prevent rebleeding from gastric varices, and to control refractory bleeding from portal hypertensive gastropathy. TIPS is better than large-volume paracentesis in the control of refractory cirrhotic ascites, but this improvement has no impact on survival. The role of TIPS in the management of patients with Budd–Chiari syndrome and hepatorenal syndrome is unclear. TIPS is not indicated in patients with varices that have never bled, for the control of bleeding from GAVE, and in patients with severe VOD or those with hepatopulmonary syndrome.
REFERENCES 1. Rösch J, Hanafe W, Snow H, et al. Transjugular intrahepatic portocaval shunt. Am J Surg 1971;121:588–592. 2. Rösch J, Keller FS. Transjugular intrahepatic portosystemic shunt: present status, comparison with endoscopic therapy and shunt surgery, and future prospectives. World J Surg 2001;25:337–346. 3. Rössle M, Richter GM, Noldge G, et al. New non-operative treatment for variceal haemorrhage. Lancet 1989;2:153. 4. Rösch J. The creation of TIPS: A brief history. In: Conn HO, Palmaz JC, Rosch J, Rossle M, eds. TIPS: transjugular intrahepatic portosystemic shunts. New York: Igaku-Shoin, 1996: 55–64. 5. Shiffman ML, Jeffers L, Hoofnagle JH, et al. Role of transjugular intrahepatic portosystemic shunt treatment of portal hypertension and its complication: Conference sponsored by the National Digestive Diseases Advisory Board. Hepatology 1995;22:1591–1597. 6. Boyer TD. Transjugular intrahepatic portosystemic shunt: Curr status. Gastroenterol 2003;124:1700–1710.
304
7. Rössle M, Haag K, Ochs A, et al. The transjugular intrahepatic portosystemic stent–shunt procedure for variceal bleeding. N Engl J Med 1994;330:165–171. 8. Luketic VA, Sanyal AJ. Esophageal varices. II. TIPS (transjugular intrahepatic portosystemic shunt) and surgical therapy. Gastroenterol Clin North Am 2000;29:387–421. 9. Cello JP, Ring EJ, Olcott EW, et al. Endoscopic sclerotherapy compared with percutaneous transjugular intrahepatic portosystemic shunt after initial sclerotherapy in patients with acute variceal hemorrhage. A randomized, controlled trial. Ann Intern Med 1997;126:858–865. 10. Sanyal AJ, Freedman AM, Luketic VA, et al. Transjugular intrahepatic portosystemic shunts compared with endoscopic sclerotherapy for the prevention of recurrent variceal hemorrhage. A randomized, controlled trial. Ann Intern Med 1997;126:849–857. 11. Cabrera J, Maynar M, Granados R, et al. Transjugular intrahepatic portosystemic shunt versus sclerotherapy in the elective treatment of variceal hemorrhage. Gastroenterology 1996;110:832–839. 12. Barton RE, Rosch J, Saxon RR, et al. TIPS: Short- and long-term results: A survey of 1750 patients. Semin Intervent Radiol 1995;12:364–367. 13. Haskal ZJ, Martin L, Cardella JF, et al. Quality improvement guidelines for transjugular intrahepatic portosystemic shunts. J Vasc Intervent Radiol 2001;12:131–136. 14. Haskal ZJ, Martin L, Cardella JF, et al. Quality improvement guidelines for transjugular intrahepatic portosystemic shunts. J Vasc Intervent Radiol 2003;14:S265–S270. 15. Martin M, Zajko AB, Orons PD, et al. Transjugular intrahepatic portosystemic shunt in the management of variceal bleeding: indications and clinical results. Surgery 1993;114:719–726. 16. Ring EJ, Lake JR, Roberts JP, et al. Using transjugular intrahepatic portosystemic shunts to control variceal bleeding before liver transplantation. Ann Intern Med 1992;116: 304–309. 17. Burroughs AK, Patch D. Transjugular intrahepatic portosystemic shunt. Semin Liver Dis 1999;19:457–473. 18. Groszmann R, Wongcharatrawee S. The hepatic venous pressure gradient: Anything worth doing should be done right. Hepatology 2004;39:280–282. 19. Boyer TD Changing clinical practice with measurements of portal pressure. Hepatology 2004;39;283–285. 20. Thalheimer U, Mela M, Patch D, et al. Targeting portal pressure measurements: A critical reappraisal. Hepatology 2004;39: 286–290. 21. Rossle M, Siegersterrer V, Olschewski M, et al. How much reduction in portal pressure is necessary to prevent variceal rebleeding? A longitudinal study in 225 patients with transjugular intrahepatic portosystemic shunts. Am J Gastroenterol 2001;96:3379–3383. 22. Casado M, Bosch J, Garcia-Pagan JC, et al. Clinical events after transjugular intrahepatic portosystemic shunt: correlation with hemodynamic findings. Gastroenterology 1998;114:1296–1303. 23. Sanyal A, Genning C, Reddy KR, et al. and the NASTRA group. The North American study for the treatment of refractory ascites. Gastroenterology 2003;124:634–631. 24. Jalan R, Elton RA, Redhead DN, et al. Analysis of prognostic variables in the prediction of mortality, shunt failure, variceal rebleeding and encephalopathy following the transjugular intrahepatic portosystemic stent–shunt for variceal haemorrhage. J Hepatol 1995;23:123–128. 25. Chalasani N, Clark WS, Martin LG, et al. Determinants of mortality in patients with advanced cirrhosis after transjugular intrahepatic portosystemic shunting. Gastroenterology 2000;118:138–144. 26. Malinchoc M, Kamath PS, Gordon FD, et al. A model to predict poor survival in patients undergoing transjugular
Chapter 17 TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT (TIPS)
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
intrahepatic portosystemic shunts. Hepatology 2000;31:864–871. Patch D, Nikolopoulou V, McCormick A, et al. Factors related to early mortality after transjugular intrahepatic portosystemic shunt for failed endoscopic therapy in acute variceal bleeding. J Hepatol 1998;28:454–460. Ferral H, Vasan R, Speeg KV, et al. Evaluation of a model to predict poor survival in patients undergoing elective TIPS procedures. J Vasc Intervent Radiol 2002;13:1103–1108. Schepke M, Roth F, Fimmers R, et al. Comparison of MELD, Child–Pugh, and Emory Model for the prediction of survival in patients undergoing transjugular intrahepatic portosystemic shunting. Am J Gastroenterol 2003;98:1167–1174. Rajan DK, Haskal Z, Clark TW. Serum bilirubin as a predictor of early mortality after transjugular intrahepatic portosystemic shunt placement: A multivariate analysis. J Vasc Intervent Radiol 2002;13:155–161. Brensing KA, Raab P, Textor J, et al. Prospective evaluation of a clinical score for 60-day mortality after transjugular intrahepatic portosystemic stent–shunt. Eur J Gastroenterol Hepatol 2002;14:723–731. Bureau C, Garcia-Pagan JC, Otal P, et al. Improved clinical outcome using polytetrafluoroethylene-coated stents for TIPS: Results of a randomized study. Gastroenterology 2004;126:469–475. Haskal ZJ, Carroll JW, Jacobs JE, et al. Sonography of transjugular intrahepatic portosystemic shunts: detection of elevated portosystemic gradients and loss of shunt function. J Vasc Intervent Radiol 1997;84:549–56. Rössle M, Siegerstetter V, Huber M, et al. The first decade of the transjugular intrahepatic portosystemic shunt (TIPS): state of the art. Liver 1998;18:73–89. LaBerge JM, Somberg KA, Lake JR, et al. Two-year outcome following transjugular intrahepatic portosystemic shunt for variceal bleeding: results in 90 patients. Gastroenterology 1995;108:1143–1151. LaBerge JM, Ferrell LD, Ring EJ, et al. Histopathologic study of stenotic and occluded transjugular intrahepatic portosystemic shunts. J Vasc Intervent Radiol 1993;4:779–786. Saxon RR, Mendel-Hartvig J, Corless CL, et al. Bile duct injury as a major cause of stenosis and occlusion in transjugular intrahepatic portosystemic shunts: comparative histopathologic analysis in humans and swine. J Vasc Intervent Radiol 1996;7:487–497. Sauer P, Theilmann L, Herrmann S, et al. Phenprocoumon for prevention of shunt occlusion after transjugular intrahepatic portosystemic stent shunt: a randomized trial. Hepatology 1996;24:1433–1436. Siegerstetter V, Huber M, Ochs A, et al. Platelet aggregation and platelet-derived growth factor inhibition for prevention of insufficiency of the transjugular intrahepatic portosystemic shunt: a randomized study comparing trapidil plus ticlopidine with heparin treatment. Hepatology 1999;29:33–38. Jalan R, Forrest EH, Stanley AJ, et al. A randomized trial comparing transjugular intrahepatic portosystemic stent–shunt with variceal band ligation in the prevention of rebleeding from esophageal varices. Hepatology 1997;26:1115–1122. Sauer P, Theilmann L, Stremmel W, et al. Transjugular intrahepatic portosystemic stent shunt versus sclerotherapy plus propranolol for variceal rebleeding. Gastroenterology 1997;113:1623–1631. Cabrera J, Maynar M, Granados R, et al. Transjugular intrahepatic portosystemic shunt versus sclerotherapy in the elective treatment of variceal hemorrhage. Gastroenterology 1996;110:832–839. Merli M, Salerno F, Riggio O, et al. Transjugular intrahepatic portosystemic shunt versus endoscopic sclerotherapy for the prevention of variceal bleeding in cirrhosis: a randomized
44.
45.
46.
47.
48.
49.
50. 51.
52. 53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
multicenter trial. Gruppo Italiano Studio TIPS (GIST). Hepatology 1998;27:48–53. Garcia-Villarreal L, Martinez-Lagares F, Sierra A, et al. Transjugular intrahepatic portosystemic shunt versus endoscopic sclerotherapy for the prevention of variceal rebleeding after recent variceal hemorrhage. Hepatology 1999;29:27–32. Ducoin H, El-Khoury J, Rousseau H, et al. Histopathologic analysis of transjugular intrahepatic portosystemic shunts. Hepatology 1997;25:1064–1069. Sanyal AJ, Contos MJ, Yager D, et al. Development of pseudointima and stenosis after transjugular intrahepatic portasystemic shunts: characterization of cell phenotype and function. Hepatology 1998;28:22–32. Sanyal AJ, Freedman AM, Luketic VA, et al. The natural history of portal hypertension after transjugular intrahepatic portosystemic shunts. Gastroenterology 1997;112:889–898. Owens CA, Bartolone C, Warner DL, et al. The inaccuracy of duplex ultrasonography in predicting patency of transjugular intrahepatic portosystemic shunts. Gastroenterology 1998;114:975–980. Haskal ZJ, Carroll JW, Jacobs JE, et al. Sonography of transjugular intrahepatic portosystemic shunts: detection of elevated portal systemic gradients and loss of shunt function. J Vasc Intervent Radiol 1997;84:549–556. Saxon R. A new era for transjugular intrahepatic portosystemic shunts? J Vasc Intervent Radiol 2004;15:217–219. Hausegger KA, Karnel F, Georgieva B, et al. Transjugular intrahepatic portosystemic shunt creation with the Viatorr expanded polytetrafluoroethylene-covered stent–graft. J Vasc Intervent Radiol 2004;15:239–248. Haskal ZJ. Interim US multicenter results of Viatorr ePTFE TIPS endoprosthesis trial. Hepatology 2003;38:293A. Sanyal AJ, Freedman AM, Shiffman ML, et al. Portosystemic encephalopathy after transjugular intrahepatic portosystemic shunt: results of a prospective controlled study. Hepatology 1994;20:46–55. Somberg KA, Riegler JL, LaBerge JM, et al. Hepatic encephalopathy after transjugular intrahepatic portosystemic shunts: incidence and risk factors. Am J Gastroenterol 1995;90:549–555. Haskal ZJ, Cope C, Soulen MC, et al. Intentional reversible thrombosis of transjugular intrahepatic portosystemic shunts. Radiology 1995;195:485–488. Kerlan RK Jr, LaBerge JM, Baker EL, et al. Ring EJ. Successful reversal of hepatic encephalopathy with intentional occlusion of transjugular intrahepatic portosystemic shunts. J Vasc Intervent Radiol 1995;6:917–921. Haskal ZJ, Middlebrook MR. Creation of a stenotic stent to reduce flow through a transjugular intrahepatic portosystemic shunt. J Vasc Intervent Radiol. 1994;5:827–829. Pattynama PM, van Hoek B, Kool LJ. Inadvertent arteriovenous stenting during transjugular intrahepatic portosystemic shunt procedure and the importance of hepatic artery perfusion. Cardiovasc Intervent Radiol 1995;18:192–195. Mallery S, Freeman ML, Peine CJ, et al. Biliary-shunt fistula following transjugular intrahepatic portosystemic shunt placement. Gastroenterology 1996;111:1353–1357. Haskal ZJ, Pentecost MJ, Rubin RA. Hepatic arterial injuries after transjugular intrahepatic portosystemic shunt placement: report of two cases. Radiology 1993;188:85–88. Sanyal AJ, Freedman AM, Purdum PP, et al. The hematologic consequences of transjugular intrahepatic portosystemic shunts. Hepatology 1996;23:32–39. Jalan R, Redhead DN, Allan PL, et al. Prospective evaluation of haematological alterations following the transjugular intrahepatic portosystemic stent–shunt (TIPSS). Eur J Gastroenterol Hepatol 1996;8:381–385.
305
Section II. Approach to the Patient with Liver Disease
63. Conn HO. Hemolysis after transjugular intrahepatic portosystemic shunting: the naked stent syndrome. Hepatology 1996;23:177–181. 64. Conn HO, Lindenmuth WW, May CJ, et al. Prophylactic portacaval anastomosis. A tale of two studies. Medicine 1971;51:27–40. 65. Moreno A, Meneu JC, Moreno E, et al. Liver transplantation and transjugular intrahepatic portosystemic shunt. Transplant Proc 2003;35:1869–1870. 66. Tripathi D, Therapondos G, Redhead DN, et al. Transjugular intrahepatic portosystemic stent–shunt and its effects on orthotopic liver transplantation. Eur J Gastroenterol Hepatol 2002;14:827–832. 67. Chalasani N, Kahi C, Francois F, et al. Improved patient survival after acute variceal bleeding: a multicenter, cohort study. Am J Gastroenterol 2003;98:653–659. 68. DAmico G, Pagliaro L, Bosch J. The treatment of portal hypertension: A meta-analytic review. Hepatology 1995;22:332–353. 69. Vangeli M, Patch D, Burroughs AK. Salvage tips for uncontrolled variceal bleeding. J Hepatol 2003;37:703–704. 70. Rössle M, Siegerstetter V, Huber M, et al. The first decade of the transjugular intrahepatic portosystemic shunt (TIPS): state of the art. Liver 1998;18:73–89. 71. Boyer TD, Haskal ZJ. Transjugular intrahepatic portosystemic shunt (TIPS). Practice Guidelines. Hepatology 2004 (submitted). 72. Henderson JM, Boyer TD, Kutner M, et al. DSRS vs TIPS for refractory variceal bleeding: a prospective randomized controlled trial. Hepatology 2004;40:presented as late breaking abstract AASLD. 73. Meddi P, Merli M, Lionetti R, et al. Cost analysis for the prevention of variceal rebleeding: a comparison between transjugular intrahepatic portosystemic shunt and endoscopic sclerotherapy in a selected group of Italian cirrhotic patients. Hepatology 1999;29:1074–1077. 74. Papatheodoridis GV, Goulis J, Leandro G, et al. Transjugular intrahepatic portosystemic shunt compared with endoscopic treatment for prevention of variceal rebleeding: A meta-analysis. Hepatology 1999;30:612–622. 75. Escorsell AM, Banares R, Garcia-Pagan JC, et al. TIPS versus drug therapy in preventing variceal rebleeding in advanced cirrhosis: A randomized controlled trial. Hepatology 2002;35:385–392. 76. Rosemurgy AS, Serafini FM, Zweibel BR, et al. Transjugular intrahepatic portosystemic shunt vs. small-diameter prosthetic H-graft portacaval shunt: extended follow-up of an expanded randomized prospective trial. J Gastrointest Surg 2000;4:589–597. 77. Chau TN, Patch D, Chan YW, et al. ‘Salvage’ transjugular intrahepatic portosystemic shunts: gastric fundal compared with esophageal variceal bleeding. Gastroenterology 1998;114:981–987. 78. Rees CJ, Nylander DL, Thompson NP, et al. Do gastric and esophageal varices bleed at different portal pressures and is TIPS an effective treatment? Liver 2000;20:253–256. 79. Tripathi D, Therapondos G, Jackson D, et al. The role of the transjugular intrahepatic portosystemic stent shunt (TIPSS) in the management of bleeding gastric varices: clinical and haemodynamic correlations. Gut 2002;51:270–274. 80. Barange K, Peron JM, Imani K, et al. Transjugular intrahepatic portosystemic shunt in the treatment of refractory bleeding from ruptured gastric varices. Hepatology 1999;30:1139–1143. 81. Burak KW, Lee SS, Beck PL. Portal hypertensive gastropathy and gastric vascular ectasia (GAVE) syndrome. Gut 2001;49:866–872. 82. Kamath PS, Lacerda M, Ahlquist DA, et al. Gastric mucosal responses to intrahepatic portosystemic shunting in patients with cirrhosis. Gastroenterology 2000;118:905–911.
306
83. Urata J, Yamashita Y, Tsuchigame T, et al. The effects of transjugular intrahepatic portosystemic shunt on portal hypertensive gastropathy. J Gastroenterol Hepatol 1998;13:977–979. 84. Runyon BA. Management of adult patients with ascites due to cirrhosis. Hepatology 2004;39:841–856. 85. Gines P, Cardenas A, Arroyo V, et al. Management of cirrhosis and ascites. N Engl J Med 2004;350:1646–1654. 86. Lebrec D, Giuily N, Hadengue A, et al. Transjugular intrahepatic portosystemic shunts: comparison with paracentesis in patients with cirrhosis and refractory ascites: a randomized trial. J Hepatol 1996;25:135–144. 87. Rossle M, Ochs A, Gulberg V, et al. A comparison of paracentesis and transjugular intrahepatic portosystemic shunting in patients with ascites. N Engl J Med 2000;342:1701–1707. 88. Gines P, Uriz J, Calaborra B, et al. Transjugular intrahepatic portosystemic shunting versus paracentesis plus albumin for refractory ascites in cirrhosis. Gastroenterology 2002;123:1839–1847. 89. Strauss RM, Boyer TD. Hepatic hydrothorax. Semin Liver Dis 1997;17:227–232. 90. Strauss RM, Martin LG, Kaufman SL, et al. Transjugular intrahepatic portal systemic shunt (TIPS) for the management of symptomatic cirrhotic hydrothorax. Am J Gastroenterol 1994;89:1520–1522. 91. Gordon FD, Anastopoulos HT, Crenshaw W, et al. The successful treatment of symptomatic, refractory hepatic hydrothorax with transjugular intrahepatic portosystemic shunt. Hepatology 1997;25:1366–1369. 92. Siegerstetter V, Deibert P, Ochs A, et al. Treatment of refractory hepatic hydrothorax with transjugular intrahepatic portosystemic shunt: long-term results in 40 patients. Eur J Gastroenterol Hepatol 2001;13:529–534. 93. Brensing KA, Textor J, Perz J, et al. Long term outcome after transjugular intrahepatic portosystemic stent–shunt in nontransplant cirrhotics with hepatorenal syndrome: a phase II study. Gut 2000;47:288–295. 94. Wong F, Sniderman K, Liu P, et al. Transjugular intrahepatic portosystemic stent shunt: effects on hemodynamics and sodium homeostasis in cirrhosis and refractory ascites. Ann Intern Med 1995;122:816–822. 95. Guevara M, Gines P, Bandi JC, et al. Transjugular intrahepatic portosystemic shunt in hepatorenal syndrome: effects on renal function and vasoactive systems. Hepatology 1998;28:416–422. 96. Valla DC. The diagnosis and management of the Budd–Chiari syndrome: consensus and controversies. Hepatology 2003;38:793–803. 97. Okuda K, Kage M, Shrestha SM. Proposal of a new nomenclature for Budd–Chiari syndrome: hepatic vein thrombosis versus thrombosis of the inferior vena cava at its hepatic portion. Hepatology 1998;28:1191–1198. 98. Murad SD, Valla DC, de Groen PC, et al. Determinants of survival and the effect of portosystemic shunting in patients with Budd–Chiari syndrome. Hepatology 2004;39:500–508. 99. Perello A, Garcia-Pagan JC, Gilabert R, et al. TIPS is a useful long-term derivative therapy for patients with Budd–Chiari syndrome uncontrolled by medical therapy. Hepatology 2002;35:132–139. 100. Mancuso A, Fung K, Mela M, et al. TIPS for acute and chronic Budd–Chiari syndrome: a single center experience. J Hepatol 2003;38:751–754. 101. Fried MW, Connaghan DB, Sharma S, et al. Transjugular intrahepatic portosystemic shunt for the management of severe venoocclusive disease following bone marrow transplantation. Hepatology 1996;24:588– 591. 102. Azoulay D, Castaing D, Lemoine A, et al. Transjugular intrahepatic portosystemic shunts (TIPS) for severe
Chapter 17 TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT (TIPS)
venoocclusive disease of the liver following bone marrow transplantation. Bone Marrow Transplant 2000;25:987–992. 103. Zenz T, Rossle M, Bertz H, et al. Severe veno-occlusive disease after allogenic bone marrow or peripheral stem cell transplantation – role of transjugular intrahepatic portosystemic shunt (TIPS). Liver 2001;21:31–36. 104. Smith FO, Johnson MS, Scherer LR, et al. Transjugular intrahepatic portosystemic shunting (TIPS) for treatment of
severe hepatic veno-occlusive disease. Bone Marrow Transplant 1996;18:643–646. 105. Paramesh AS, Husain SZ, Shneider B, et al. Improvement in hepatopulmonary syndrome after transjugular intrahepatic portasystemic shunting: Case report and review of literature. Pediatr Transplant 2003;7:157–162.
307
Section III. Clinical Consequences of Liver Disease
18
HEPATIC ENCEPHALOPATHY Kevin D. Mullen Abbreviations BCAA branched-chain amino acid BZ benzodiazepine EEG electroencephalography
GABA HE IRBs
g-amino butyric acid hepatic encephalopathy human investigation review boards
INTRODUCTION Hepatic encephalopathy (HE) is a term used to describe a spectrum of neuropsychiatric abnormalities occurring in patients with significant liver disease and/or major portosystemic shunting of blood. Strictly speaking the term should not be employed in liver disease of any other potential cause if encephalopathy is identified. However, in clinical practice the diagnosis is often made in the presence of other causes of alterations of brain function, such as the use of sedatives. Patients with advanced liver disease are more likely to exhibit HE, but they are also more prone to present with multiple simultaneous causes of encephalopathy. Herein lies one of the many difficulties in designing clinical treatment trials in patients with HE. We will return to this issue of difficulties in study design throughout the chapter, for two main reasons. All the previously published literature on HE has to be viewed from our current understanding of HE. In addition, lessons learned from prior difficulties with study design in HE can teach us a great deal about this enigmatic and poorly understood disorder.
DEFINITION AND NOMENCLATURE Since the classic papers of Adams and Foley1,2 and Summerskill and colleagues,3 there have been few attempts to standardize terminology in HE. Growing appreciation of the problems created by inexact or loosely applied terminology led to the development of a consensus panel at a meeting in Newcastle, England in 1996.4 This ultimately led to a publication of a paper in 1998 which for the first time standardized nomenclature with a multiaxial-type classification.5 Virtually every individual with a major interest in HE was involved in the generation of this consensus statement, and the final product represented a compromise from many. As such, it was felt that all subsequent publications in this field should use this terminology until it is officially modified at some future time. Table 181 represents a minor modification of the consensus statement. Some comments are needed on this terminology. First, the confusing terms ‘acute’ and ‘chronic’ HE were replaced by ‘episodic’ and ‘persistent’, respectively. Much of the older literature describing treatment trials in chronic HE in fact concerned treatment of episodes of HE in patients with advanced chronic liver disease. Worse still, many did not even specify whether precipitating factors for HE were sought or treated. We now recognize that most episodes of HE in patients with chronic liver disease are precipitated by
LOLA MARS PSE
L-ornithine–L-aspartate
molecular absorbent recirculating system portal–systemic encephalopathy
events such as bleeding, infection, or the use of sedatives/hypnotics.5 Accordingly, it was specified that a list of precipitating factors should be carefully sought and treated as part of the normal medical care for patients with HE. Only if these were not present could the term ‘spontaneous HE’ be used. The rest of the terminology is, hopefully, self-evident. However, it should be noted that this new terminology does not define precisely when HE should be called persistent (?4 weeks), and it does not specify a category for patients with persistent neurological signs, such as parkinsonian-type abnormalities without alterations in mental status. Although it is highly desirable to use more precise terminology, it is important that we do not go too far. If, for instance, we specify a large list of exclusionary criteria for the diagnosis of HE we may make it impossible to perform clinical treatment trials. This may have already happened with the hepatorenal syndrome.6
GENERAL PATHOPHYSIOLOGY Before addressing the fundamental molecular mechanisms thought to be responsible for the pathogenesis of HE it is important to outline some major pathophysiological aspects of the syndrome. It is not commonly known that some of the more famous cases published in the literature did not have significant structural liver disease.7,8 Instead, they represented what we now term type B HE (B stands for bypass). Indeed, the term portal–systemic encephalopathy (PSE) became popular partly because of one of these cases.7 It should be mentioned that PSE as a term is not endorsed in the new terminology. As a basic premise nearly all investigators agree that the ‘toxin’ causing HE arises in the gut. Failure of the liver to detoxify portal blood toxins leads to overt HE when the functional capacity of the liver and hepatic perfusion by portal blood are diminished. This is seen in everyday practice, where the vast majority of HE is observed in patients with decompensated cirrhosis and well established portosystemic shunts. However, not emphasized nearly enough has been a very important observation that there is a virtual absence of overt HE in patients with total diversion of portal blood around the liver who have normal liver histology. Portal vein thrombosis is quite prevalent in certain parts of the world, and HE is rare even after massive variceal bleeding. Defining why these patients rarely develop HE unless decompressive shunts are created is extremely important to our understanding of HE and has been a subject of great debate.9 Of equal importance are the type B HE patients who
311
Section III. Clinical Consequences of Liver Disease
Table 18-1. Classification of Hepatic Encephalopathy (HE)
Table 18-3. Hypotheses of Pathogenesis of HE
Type A: Associated with Acute liver failure Type B: Associated with major portosystemic shunting and no intrinsic liver disease Type C: Associated with Chronic liver disease/Cirrhosis Patterns of presentation in type B and C HE: Overt episode: spontaneous or precipitated Recurrent episodes: more than two in 1 year Persistent Minimal Mild to moderate stage I–II Severe stage III–IV Treatment suppressed Hepatocerebral degeneration
Direct ammonia neurotoxicity Multiple synergistic neurotoxins, including ammonia False neurotransmitter/plasma amino acid imbalance Monoamine disturbances True neurotransmitter disturbances GABA/benzodiazepine Severe zinc deficiency Endogenous opiates Astrocyte dysfunction Manganese neurotoxicity Histamine Cerebral edema/disturbed osmolyte pattern Cytokines – as a cofactor Neurosteroids
Table 18-2. Mechanisms of Post-Shunt HE Worsening liver function Loss of hepatotrophic portal perfusion Ischemia Poor hepatic arterial buffer response Enhanced absorption of ammonia or other gut toxins Increased portosystemic bypass
have recurrent bouts of HE despite no intrinsic liver disease. Table 18-2 outlines some possible mechanisms, but a precise understanding of this phenomenon is still lacking. One of the difficult aspects of HE is the concept of the sensitized brain. It is widely accepted that gut toxins, especially ammonia, not only play a major role in inducing bouts of overt HE but also may sensitize the brain to a variety of insults. This demonstrates the problem with the ammonia hypothesis, which was often dismissed because overt bouts of apparent HE seemed to occur without significant elevations in blood ammonia. The sensitized brain in longstanding cirrhotic patients may be akin to the brain in those with compensated low-grade dementia. Everything seems to be working fairly well until sepsis or dehydration occurs, at which point there is a marked deterioration in mental status. Similarly, in patients with cirrhosis these non-hyperammonemia bouts of HE are thought to be classic HE, but in fact may represent a secondary insult to the sensitized brain, leading to HE. As hyperammonemia is not the cause of the altered mental state it is very easy to conceive that correcting these types of precipitating factors without using lactulose may improve bouts of encephalopathy with no major impact on ammonia levels. As noted earlier, the specificity of HE, or the confidence one can have in assigning a bout of mental status change to HE, is difficult in the face of multiple precipitating factors or independent causes of encephalopathy (i.e. sepsis). We will return to this issue intermittently throughout this chapter. Before going into more specific discussion of the molecular mechanism thought to be responsible for the pathogenesis of HE, it is important to note that HE may not be just a reversible metabolic encephalopathy. Brain atrophy and parkinsonian signs are a definite element of this syndrome.10–13 They may have been more prevalent in the days of frequent creation of decompressive portosystemic shunts,14 but are still typically observed in patients with large long-
312
standing spontaneous portosystemic shunts. Successful liver transplantation can improve or possibly reverse some of these changes in brain structure,15 but as yet this is not often factored into the decision to list patients for liver transplantation.
PATHOGENESIS OF HEPATIC ENCEPHALOPATHY One of the major problems in writing a chapter on the syndrome of hepatic encephalopathy is the extraordinary plethora of publications on possible mechanisms for this neuropsychiatric condition. As listed in Table 18-3, there are numerous hypotheses as to the cause of HE. Just describing the basic scientific concepts behind some of these hypotheses could fill an entire chapter. We will primarily discuss theories of pathogenesis that are still felt to be viable, or at least which are still being investigated actively. There is no question that once again ammonia is the toxin with the best support for a major role in the pathogenesis of HE. Based on recent investigations it appears that it is the interactions of ammonia with other systems that may lead to HE. This has resulted in a unifying concept for ammonia’s role in HE.
ORIGINAL AMMONIA HYPOTHESIS A series of publications in the 1950s for the first time seriously raised the possibility that ammonia played a major role in HE.16–18 One particularly compelling paper reported induction of the classic signs of HE in cirrhotic patients taking ammonia exchange resins.16 This therapy was being investigated as a possible treatment for ascites because of its ability to exchange sodium for ammonia in the resins.16 However, the appearance of what was then called impending hepatic coma after taking these resins indicated that perhaps ammonia was an important toxin in HE. Throughout the 1950s and 1960s, using probably inaccurate blood ammonia assays, it proved difficult to establish a correlation between the levels of blood ammonia and the severity of HE. Hidden within this literature were two particularly large studies that showed a reasonable correlation.19,20 However, only in the last few years have three studies using modern ammonia assays found a consistent correlation (r = 0.6 or
Chapter 18 HEPATIC ENCEPHALOPATHY
higher) between blood ammonia levels and the severity of HE as judged by clinical criteria21–23 (see later section on Diagnosis: Ammonia testing). The basic premise of the original ammonia hypothesis was that ammonia was directly neurotoxic.24 There is no question that hyperammonemia does cause major CNS disturbances, as seen typically in patients with hereditary urea cycle enzyme defects.25 However, one major objection to this theory of direct neurotoxicity was the dissimilarity between HE in chronic liver disease patients and that in patients with ‘pure hyperammonemia’, where seizures were noted to be common.26 Also, two controlled trials of oral ammonia administration in cirrhotic patients failed to induce HE.27,28 Because of these issues and the failure for years to find a correlation of blood ammonia with the severity of HE, many investigators began seriously to question the ammonia hypothesis. This led to an amazing amount of disparate research on other possible mechanisms for the pathogenesis of HE. In many publications all of this work was carefully detailed and remains of interest.29–32 However, one inescapable point kept arising, in that most treatments thought to be effective in HE were best explained by an ability to reduce the generation of ammonia and its absorption from the gut.
THE UNIFYING AMMONIA HYPOTHESIS Rather than describe the complex evolution of this hypothesis we will attempt to summarize its current status. The majority of HE investigators now accept this concept, which was originally suggested by Haussinger and colleagues in 2000.33 The basic idea is the entire spectrum of HE, from that seen in acute liver failure to chronic liver disease, is associated with varying degrees of cerebral edema. The earliest initial event causing cerebral edema is most likely astrocyte swelling due to influx of ammonia into the brain. Astrocytes contain glutamine synthetase, which catalysis glutamates in combination with ammonia and represents the sole mechanism for handling excessive levels of ammonia in the brain. The accumulation of osmotically active glutamine causes astrocyte swelling, which leads to a host of events (Table 18-4), many of which can be construed to play a role in the pathogenesis of HE. In acute liver failure the rapid ingress of ammonia does not allow adequate compensation to occur in response to this generation of intracellular osmolytes. This compensation phenomenon features the pumping-out of existing intracellular osmolytes such as myoinositol and taurine.30 These acute effects can escalate to the point where sufficient cerebral edema occurs to cause intracranial hypertension. In contrast, in chronic liver disease cerebral edema tends
Table 18-4. Some Effects of Astrocyte Swelling Activation of extracellular regulated protein kinases (erks) Up-regulation of peripheral-type benzodiazepine receptors Protein phosphorylation Amino acid transport Intra- and subcellular pH Increased production of neurosteroids Down-regulation of osmolyte transporters
to be low grade because of a compensating efflux of myoinositol.34 In the 1990s nuclear magnetic spectroscopy studies identified decreased brain myoinositol levels,35 but it took the perception of Haussinger to understand that this was an indicator of low-grade cerebral edema.33 Various techniques have now confirmed low-grade cerebral edema, particularly in brain white matter, which disappears with successful liver transplantation.36 Astrocyte swelling has also been noted to occur with exposure to manganese,37 and ligands that bind to peripheral-type benzodiazepine receptors and other substance that have been thought to be a factor in the genesis of HE.38–40 Accordingly, we now have a central theme to most research on the pathogenesis of HE, i.e. that ammonia is at its core. Given this belief, ammonia reduction remains the target for most treatments being tested for efficacy in the reversal of HE.
SYNERGISTIC NEUROTOXIN HYPOTHESIS Championed by Zieve,41 this hypothesis enjoyed considerable popularity for a time. At its core was the idea that mercaptans, a byproduct of methionine metabolism, in conjunction with other gut-derived toxins, caused HE. Reports of induction of HE by oral methionine in animal models of liver disease and human cirrhotics have periodically appeared in the literature.42,43 Fetor hepaticus, commonly present in patients with or likely to develop HE, has been conclusively proved to be derived from mercaptans.43 However, little has been heard in recent times about these interesting compounds, despite refinements in their measurement.43 Nonetheless, the concept that many compounds apart from ammonia may contribute to HE is widely accepted.
FALSE NEUROTRANSMITTER/ PLASMA AMINO ACID IMBALANCE HYPOTHESES As can be seen from the heading, these two hypotheses were originally distinct. False neurotransmitters were believed to be generated in liver failure and to replace normal neurotransmitters (e.g. dopamine) in the brain.44 This could lead to a failure of neurotransmission, because false or weak neurotransmitters had activity levels far lower than those of normal neurotransmitters. Some data appeared suggesting that this concept might also underlie the hyperdynamic circulation seen in advanced liver disease.44 However, the rather direct approach of Zieve and colleagues of injecting large quantities of octopamine into the cerebral ventricles of rats reduced enthusiasm for this hypothesis when no discernible effects were observed.45 The observation that an abnormal pattern of plasma amino acids was found in cirrhotic patients both with and without HE led to the plasma amino acid imbalance hypothesis, which later incorporated elements of the false neurotransmitter hypothesis.46 Elevated brain levels of tyrosine (a precursor for some proposed false neurotransmitter) and phenylalanine were thought to occur because of a fall
313
Section III. Clinical Consequences of Liver Disease
in the blood concentrations of branched-chain amino acids (BCAA). The influence of ammonia on glutamine transport across the blood–brain barrier using the amino acid carrying system was also thought to be important.46 The ‘proof of the pudding’, as it were, would have been reversal of HE with normalization of this abnormal ratio of amino acids in blood. Nutritional products for enteral and parenteral use were developed that were low in tyrosine and phenylalanine and enriched in branched-chain amino acids. The normal value for the ratio of branched-chain amino acids to phenylalanine and tyrosine in blood is 2–3.5. Cirrhotic patients usually had a ratio of 1 or less. However, despite succeeding in normalizing this ratio, clearcut improvement of HE was difficult to prove. Indeed, this whole area became very controversial, and little remains of these special and expensive nutritional products except for their continuing enthusiastic use in Japan.47 Ingenious as this hypothesis was, it attracts little interest today.
GABA/BENZODIAZEPINE HYPOTHESIS The endogenous benzodiazepine (BZ) hypothesis is not distinct from the g-amino butyric acid (GABA) hypothesis. Benzodiazepines generally exert their CNS effects by augmenting neuroinhibitory GABA-ergic neurotransmission. The main difference, if there is one, between these two hypotheses is that the GABA hypothesis proposed that increased GABA neurotransmission was due to excess brain GABA,48 whereas the BZ hypothesis proposed that increased GABA-ergic neurotransmission was due to endogenous benzodiazepines.49 Indeed, it was during a series of neuropharmacologic experiments on the galactosamine rabbit model of HE to test whether GABA-ergic neurotransmission was increased that flumazenil (a receptor antagonist) was accidentally observed to improve HE.50 It has been forgotten that these experiments also provided evidence that GABA-ergic tone was increased in the rabbit model. However, the effects of flumazenil, which was expected to be inactive, were so dramatic that this theme was embraced.49 CGS-8216, another benzodiazepine antagonist like flumazenil, was also reported by Baraldi and colleagues51 to be able to reverse HE in a rat model. However, their interpretation of this was that the supersensitivity of the BZ receptor was responsible for this HE reversal effect. We, on the other hand, began a search for endogenous benzodiazepines, which we thought could be the only possible explanation for our findings.49 Barbaro and colleagues,52 many years later, published a very large clinical trial of the efficacy of flumazenil as an ultrafast therapy for HE. They reported that about 30% of patients displayed some degree of reversal of HE with flumazenil, but not with placebo. Over the years a number of other studies have indicated similar results to those of Barbaro et al.,52–59 but only Ferenci and colleagues8 used long-term flumazenil in a patient when all other therapies had failed to control her HE. The search for the identification of endogenous benzodiazepines has not been very fruitful. Despite reports to the contrary, endogenous benzodiazepines are a constant finding in HE.60 The real question is what role do they play in HE, and where do they come from?
314
The issue of occult ingestion of prescription benzodiazepines as an explanation for the finding of benzodiazepines in cirrhotic patients cannot be easily resolved, as endogenous and prescription benzodiazepines are largely identical. These compounds are present in the food supply and there is evidence that they arise in the gut.61,62 As with ammonia, treatments that are effective in HE reduce blood levels of benzodiazepines.63 Recently Zeneroli and colleagues63 have reported the presence of these compounds once again, so this hypothesis of HE is not yet dead.
OTHER HYPOTHESES There are a number of other ideas as to what might contribute to the pathogenesis of HE. Neurotoxicity from manganese is of major interest in light of the finding of excessive levels of this metal in certain brain regions.64 Also, the ubiquitous finding of hyperintensity on T1-weighted MRI scans of the brain is largely attributable to manganese.65 As mentioned earlier, manganese has been shown to induce astrocyte swelling in vitro.37 One wonders whether chelation therapy will ever be tried for the treatment of HE. The endogenous opiate system may play a role in the pathogenesis of HE, based primarily on studies of Yurdaydin and colleagues.66 Case reports of reversal of human HE with narcotic antagonists have appeared.67 Fatigue to some extent, and pruritus for certain, can be related to the accumulation of endogenous opiates.68 Indeed, Bergasa69 has championed narcotic antagonist therapy for severe pruritus in liver disease, and has data from randomized double-blind placebo-controlled trials in support of this concept. It would be inappropriate not to recognize that Michael Norenberg for years suggested that the key to understanding HE was to study astrocytes.39,40 The unified ammonia hypothesis would appear to have proved him right. A major amount of his long-standing work has explored the mechanisms whereby astrocyte function can be disrupted in HE. Butterworth and his many colleagues have devoted decades to extensive research on the possible pathogenic mechanisms of HE, and much of our current understanding comes from work done in his laboratory. Strangely, he could never find endogenous benzodiazepines, but his recent work on neurosteroids is arousing great interest.70 These compounds are made by astrocytes and enhance GABA-ergic neurotransmission.71 His laboratory has also explored autopsy material from brains with HE and has examined nearly every system in the brain. Perhaps like all of us he feels that many simultaneous mechanisms may be responsible for HE. His excellent reviews on pathogenesis are required reading for all in the field.31,38,72 Comprehensive reviews of the possible mechanism for the pathogenesis of HE are already published.29–32,73,74 However, from a historical perspective the book by Conn and Liebenthal is still the most interesting reading one can find in this area.74
DIAGNOSIS OF HEPATIC ENCEPHALOPATHY GENERAL POINTS Entertaining the idea that a patient may have HE is not a challenge when a change in mental status is noted and obvious liver disease is
Chapter 18 HEPATIC ENCEPHALOPATHY
Table 18-5. Other Causes of Encephalopathy in Advanced Liver Disease Condition
Method of detection
Treatment
Metabolic Hypoxia Hypercoprie Hypoglycemia Hypernatremia Hyponatremia Hypothyroidism Hyperthyroidism Hypercalemia
Blood gas analysis Blood gas analysis Blood test Blood test Blood test Blood test Blood test Blood test
Treat course Treat course IV glucose Insulin IV free water Water restriction Thyroid Hormone Propylthiouracil etc
CNS Cerebrovascular accident Intracranial hematoma Seizure (postictal) Metastatic cancer
Brain imaging Brain imaging Electroencephalogram/clinical pictures Brain Imaging
Hydration Tx course Tx course Evaluation of Hematoma Antiseizure medications
Toxins Alcohol CNS depressants
Blood alcohol Drug screen
Control cerebral edema Supportive Tx
Infection Sepsis (general) Meningitis Encephalitis Intracranial abscess Delirium tremens
Blood, urine, ascites culture Spinal fluid examination Clinical picture, brain biopsy Brain imaging Clinical picture, history
Antagonists, time Antibiotics Antibiotics Antiviral agents etc. Antibiotics/drainage
*This list is based on Seen in last 25 years in one hospital. Other conditions can and do occur but are less common than the above list. Full biochemical profile, brain imaging, extensive search for sepsis and drug screen advisable in any advanced liver patient with coma.
present. However, one has no assurance that one is dealing with HE until other potential causes of ‘encephalopathy’ are ruled out. Table 18-5 lists the important entities to rule out, especially in patients with advanced cirrhosis. This list arises from the authors’ own experience with encephalopathy in liver patients over the years. Of course, any condition that alters CNS function could conceivably occur in liver patients. One major problem for clinicians and HE investigators is that technically a patient cannot be labeled as having HE if one or more other conditions are present. Generally, the clinician treats HE and attempts to correct the other potential cases of encephalopathy. The HE investigator often has to exclude patients with concomitant causes of brain dysfunction from clinical trials, which makes finding adequate number of patients with ‘pure’ HE difficult. Equally important is the issue of a thorough search for precipitating factors in this particular episode of HE. The consensus statement on HE has specified the precipitating factors that should be identified (Table 18-6). Most are easily discerned, but occult sepsis and dehydration can be difficult. A good rule of thumb is that every advanced cirrhotic displaying a change in mental status is septic until proved otherwise. The more severe the episode of HE, the more true this statement is. Free water dehydration is grossly underdiagnosed in HE, for a number of reasons. The first is the widespread acceptance that hyponatremia is an aggravating factor for HE.75 Indeed, the astrocyte swelling mentioned earlier is significantly aggravated by the presence of hyponatremia (both in vitro and in vivo studies support this). This leads to a sense of relief when serum sodium levels are normal, high normal, or even elevated. However, this may be one
Table 18-6. Precipitating Factors of Episodes of HE Definite Gastrointestinal bleeding* Sepsis* Hypokalemia* Azotemia* Hyponatremia Dehydration CNS-active drugs* Constipation* High protein load* Portosystemic shunt Superimposed liver injury Less certain Zinc deficiency Hypernatremia b-Blockers Anemia Oral methionine *Mentioned in HE consensus.
of the few indications of free water depletion in cirrhotics. Free water depletion is most commonly caused by excessive lactulose use, or poorly controlled diabetes and the use of diuretics. It is not associated with hypotension or prerenal azotemia in its pure form, which is another reason it is not usually considered to be a precipitating factor in severe HE. However, there is a phenomenon that all clinicians recognize yet ignore in patients with bouts of HE. This is a very rapid (i.e. within hours) arousal from HE with simple IV
315
Section III. Clinical Consequences of Liver Disease
water (D5W). It is impossible to know how many cases of free water depletion precipitate HE, but in the patient with overenthusiastic use of lactulose it may be common. Our colleagues in the International Ascites Club have recognized this problem by requiring patients on lactulose to be excluded from the diagnosis of hepatorenal syndrome,6 as they realize how common free water and true dehydration are in patients on lactulose. There is a major need to investigate the frequency of free water depletion in patients with HE. This has major implications for patients entering treatment trials. Frank hypernatremia in a cirrhotic patient must be corrected before arousal from HE can be expected. The reason for such prominent hyperammonemia after gastrointestinal bleeding has been ascertained. Hemoglobin is deficient in the essential amino acid isoleucine,76,77 and when hemoglobin is digested and absorbed it causes an unbalanced amino acid profile that ultimately leads to disproportionate production of ammonia. Conceivably, infusions of isoleucine to cirrhotic patients after they bleed may reduce the ammonia load. Most of the precipitating factors listed are easily identified and correctable. True spontaneous HE does exist, but is inversely proportional to the degree of completeness of the search for precipitating factors. Correction of precipitating factors has a major impact on the rapidity of reversal of HE. Controlling for this in clinical treatment trials remains a huge challenge, and is really only overcome by enrolling large numbers of patients to equalize the profile of precipitating factors in each treatment arm. Having discussed the fairly straightforward case of earlydiagnosed HE, we will now turn to a relatively uncommon, but real problem. This is a bout of changed mental status in a patient with no obvious signs of liver disease. Table 18-7 lists the situations when this may arise. This phenomenon is seen most often in patients with large portosystemic shunts and well preserved liver function. They may have underlying compensated cirrhosis, or may have no intrinsic liver disease. The latter group would be classified as type B (for bypass) HE patients. Despite the old name for HE, which was portosystemic encephalopathy, there is no good evidence that portosystemic shunting is the major factor in the appearance of overt HE. In fact, patients with long-standing portosystemic shunts because of splanchnic venous thrombosis (i.e. portal vein) but normal liver biopsies do not often experience HE. This remains the case even after a major GI bleed. Theories as to why these patients are not very susceptible to HE are numerous,9 but perhaps an even better question to ask is why the small minority experience bouts of HE. If we could understand why this occurs, as illustrated by some famous cases in the literature,7,8 we might be much closer to understanding the pathogenesis of HE. In any event, although liver function will be normal in the rare shunt-only HE case, the patient
Table 18-7. Conditions when Underlying Cause for HE may be Obscure Long-standing well compensated cirrhosis Congenital portosystemic shunts Portosystemic shunts due to splanchnic venous thrombosis Non-cirrhotic portal hypertension Coincidental partial urea cycle enzyme deficiency Mild liver disease with other entities such as zinc deficiency, Addison’s disease, or hypothyroidism
316
still will display features of portosystemic shunting, such as spider angiomata. Historical, clinical, and biochemical features that may point to underlying cryptic liver disease or major spontaneous portosystemic shunting are listed in Table 18-8. A far more common situation is the patient with extant but compensated cirrhosis and significant spontaneous portosystemic shunts. Large shunts of any type (i.e. surgical, TIPS, or spontaneous) are associated with a higher incidence of HE in patients over 65 years of age. Unlike alcoholic cirrhosis, where shunting and loss of liver function progress at roughly the same rate, we see a different pattern in patients with hepatitis C or non-alcoholic steatohepatitis (NASH)induced cirrhosis. Large shunts develop over decades, and we may start to observe HE before major loss of liver function is obvious. This will pose a significant challenge to clinicians. Patients may be diagnosed as having dementia when underlying HE is the real issue.
DIAGNOSIS Specific Issues – Mental Status Changes HE was stated by Adams and Foley2 to be a syndrome with two major components: altered mental status, and generalized motor
Table 18-8. Clues to the Presence of Underlying Liver Dysfunction (A) History Risk factors for viral hepatitis Lifetime alcohol consumption Family history of liver disease Originally from endemic area for hepatitis B Exposure to hepatotoxins – industrial, drugs etc. Personal history of long-standing diabetes Past history of hepatitis/jaundice Jejunoileal bypass procedure (B) Clinical signs on examination Spider angiomata (especially in men) Palmar erythema Dupuytren’s contractures Gynecomastia/loss of body hair Testicular atrophy Hepatomegaly or small liver Splenomegaly Ascites Peripheral edema Fetor hepatis Caput medusa (C) Imaging studies (US, CT, or MRI) Irregular liver contour Collateral vessels – in esophagus, liver hilum, and elsewhere Ascites Hepatomegaly or shriveled liver Splenomegaly Hyperintensity of basal ganglia or T1-weighted images of the brain (D) Laboratory testing Hypoalbuminemia Prolonged prothrombin time Hypergammaglobulinemia AST > ALT (when elevated) Hyperbilirubinemia Leukopenia Thrombocytopenia Properly performed serum ammonia
Chapter 18 HEPATIC ENCEPHALOPATHY
disturbance. Mental status changes can vary from deep coma to subtle changes in orientation. Most physicians, without many supporting data, feel that patients with well compensated cirrhosis (i.e. normal prothrombin time and albumin levels) should not develop overt or clinically detectable HE unless they have a major precipitating factor, such as a large variceal bleed. This may or may not be true, but a recent study by Gentile et al.78 identified stage II HE in a large group of Child’s class A cirrhotic patients. The form of HE detected was the persistent type, and this has major implications for future studies of the treatment of HE. This type of stable persistent HE in stable cirrhotics is almost ideal for treatment trials and should be sought in our clinics. Krieger et al.,60 in a very provocative paper, suggested that many neurological abnormalities could be found in ambulatory outpatient cirrhotics. It seems entirely possible that detailed neurological evaluation by physicians adept at such examinations will confirm these findings, especially in patients with prior (even remote) attacks of overt HE. Krieger et al. did not just find signs of extrapyramidal dysfunction, which have been reported on a number of occasions, but actually found evidence of cortical and subcortical deficits even in patients on continuing active HE treatment (e.g. lactulose, low-protein diet, etc.).79,80 The conclusion of the editorial commentary on this paper was that no-one else had previously searched for these neurological deficits in this population.59 Clearly, we need to search for these patients in the ambulatory setting so that they can be treated effectively and enrolled in clinical trials. Table 18-9 details the system for staging of HE, which essentially was a modification of the system originally discussed by Parsons Smith et al.81 This is the so-called West Haven Scale,74 and it has been used for three decades. There is general agreement that stages III and IV are consistently assessed. However, stage II to some degree, and stage I in particular, are assessed variably. To our knowledge, no interobserver variability study has ever been carried out to establish how consistently clinicians assign patients to stages I or II HE. Nonetheless, this staging system is used universally in clinical practice. Recently an amalgamation of the West Haven and Glasgow Coma Scales appeared, and this might be a major improvement but should be tested prospectively.82
Table 18-9. West Haven Criteria for Staging of HE Stage 0: No abnormality detected Stage 1: Trivial lack of awareness Shortened attention span Euphoria or anxiety Impairment of addition and subtraction Stage 2: Lethargy Disorientation for time Obvious personality changes Inappropriate behavior Stage 3: Somnolence to semistupor Response to stimuli Confusion Gross disorientation Bizarre behavior Stage 4: Coma – tests of mental state impossible
Generalized Motor Abnormalities Hyperreflexia, asterixis, and extensor plantar reflexes are often seen in HE. Asterixis is not specific to HE, but its presence in a patient with known liver disease of some consequence is a fairly reliable sign. Occasionally decerebrate and even decorticate posturing is seen. The latter may sometimes be due to overt intracranial hypertension, even in patients with chronic liver disease.83 It was originally thought that cerebral edema severe enough to cause intracranial hypertension was only seen in acute liver failure (i.e. any type A HE). However, it is seen – albeit uncommonly – in chronic liver disease patients with ominous prognostic complications. Recently, a report from France has suggested that persistent neurological deficits (e.g. lasting days) can be seen in HE in chronic liver disease without brain imaging abnormalities indicative of cerebrovascular disease.84 This type of phenomenon has appeared periodically in the literature over the years. Nonetheless, clear lateralizing signs in a patient thought to have HE may be due to entities other than HE. The ubiquitous use of CT or MRI scans of the brain has been very helpful in identifying non-HE pathology, such as subdural hematoma, in patients originally thought to only have HE.
Clinical Evaluation Generally, if the patient is obtunded only family members can provide historical clues as to what may have precipitated a bout of HE. Sometimes the injudicious use of tranquilizers or sleeping tablets is discovered by interviewing the family. Excessive diarrhea from lactulose is an important suggestion of dehydration, and paradoxically tends to arise in settings when the family is making sure the patient is compliant with therapy. As noted earlier, an extensive evaluation is needed to sift through the differential diagnosis.85 Clinical examination of patients thought to have HE should start with a systematic neurological evaluation and detailed documentation of overall mental status. Equally importantly, stigma of chronic liver disease should be sought (see Table 18-8). There is a very extensive differential diagnosis in patients suspected to have HE, which is detailed in Table 18-5. Laboratory testing and imaging studies are very important in helping discern the cause of mental status changes in liver patients.
Electrophysiological Studies Despite the relatively extensive use of electroencephalography (EEG) and evoked potentials in clinical trials,81,86–92 these electrophysiological techniques are rarely applied in clinical practice unless other problems are suspected. Research techniques such as transcranial magnetic stimulation are beginning to be used more frequently.90,93 However, of all the electrophysiologic techniques that have been applied to HE, the so-called flicker fusion detection system seems to have the best possibility of entering the clinical arena.94 These devices have been used by pharmacologists for years to detect the effects of sedatives etc. on brain function.95 One major study has been published showing their usefulness in detecting/measuring HE.94 Generally this device can only be used to measure HE in the minimal–stage II spectrum of severity. This, however, may well be the population of patients targeted for HE treatment trials in the future.
317
Section III. Clinical Consequences of Liver Disease
Brain Imaging The increase in availability of high-quality imaging of the brain is important to HE in a number of ways (Table 18-10). First, findings of intracranial pathology in patients thought to have HE have been a major advance in our attempts to exclude other concomitant causes of ‘encephalopathy’ in cirrhotic patients. Patients with advanced HE or severe ascites are subject to falls, and can easily develop intracranial hematomas because of their coexisting coagulopathies. In addition, we now have new information on alterations in the brain that may relate directly to the pathogenesis of HE. T1weighted MRI scans have frequently identified hyperintense basal ganglia in patients with established portosystemic shunts.96–104 There has been considerable debate as to what was causing this,85 and it is now widely accepted to be due to deposits of manganese. In vitro this metal causes swelling of astrocytes and can be implicated as a factor in the pathogenesis of HE.37 Loss of cortical volume has been reported in cirrhotic patients ever since the era of CT scanning of the brain began.11,105 Indeed, in those earlier studies some findings were noted that suggested cerebral edema was present,105 which, years later with more sophisticated MR techniques, was shown to occur frequently in patients with chronic liver disease. MR spectroscopy identified for the first time in HE the universal finding of myoinositol depletion in brain tissues.33,35,106,107 This was another indication, only realized later, of the presence of some degree of cerebral edema in most patients with HE.33 PET scanning has repeatedly identified regional specific alterations in brain blood flow in HE, which also may have major implications in its pathogenesis.108–114
Diagnosis: Blood Ammonia Testing No quantitative estimate has ever been made of the number of ammonia blood tests done each year in the USA. Blood ammonia levels are frequently used as a diagnostic test for HE without any proof of their diagnostic accuracy.115 Before examining the diagnostic value of blood ammonia testing it is important to look at the use of this test in patients known to have underlying cirrhosis. If a patient has an altered mental status compatible with HE and the ammonia assay reveals a level in the normal range, patients are still treated for HE. The same applies if the ammonia levels are elevated. Accordingly, the influence of a blood ammonia test on treatment of an encephalopathic patient with known cirrhosis is negligible. One must still rule out other causes of encephalopathy no matter what the ammonia level, as it does not rule out other causes of mental status changes.
In three recent studies21–23 rigorously performed ammonia assays using modern techniques have confirmed a correlation of blood levels with the severity of HE (r > 0.6). This is not sufficient proof to use ammonia testing as a diagnostic strategy, however.115 In clinical practice, where ammonia assays are not as rigorously performed, it is not uncommon to find improbably high ammonia levels. These arise from failure to put the blood specimen on ice and run the assays quickly. In vitro generation of ammonia is a well described phenomenon.116 One scenario often encountered is blood ammonia elevations in patients after alcohol withdrawal seizures. These elevated levels arise from massive muscular activity during seizures, but are often responsible for labeling postictal confusion as HE. Hence ammonia testing can be misleading from a number of perspectives. One potential use of random ammonia assays is for the identification of unsuspected cirrhosis, congenital portosystemic shunts, or urea cycle disorders. Cirrhosis is frequently associated with hyperammonemia without clinical signs to suggest HE. So, even with a high blood ammonia (assuming the assay was run correctly) one cannot diagnose HE unless mental status changes are present. If they are, then of course the differential diagnosis of what causes elevated blood ammonia levels should be explored. Too often physicians are contacted urgently about ‘panic’ values for blood ammonia levels when the blood ammonia is 2–3 times normal level or high, but the patient has no mental status changes. In this situation one wonders why the test was done in the first place (Table 18-11). To summarize, blood ammonia testing has little role in the diagnosis of HE. Clinical suspicion is the main guide to diagnosis, and normal blood ammonia does not rule out HE. The only use of blood ammonia assays is in patients without any clear indication of underlying liver disease who have marked mental status changes. Even in this setting one must be very cautious to avoid overusing the test results to direct future care. Recently the excellent equivalency between venous and arterial ammonia levels in a large series of patients with cirrhosis and/or HE suggests that venous blood ammonia levels also correlate with the severity of HE.22
TREATMENT OF HEPATIC ENCEPHALOPATHY Before discussing the many therapies available to treat HE some basic points need to be made. The first is that in clinical practice
Table 18-11. Current Status of Ammonia Testing as a Diagnostic Aid
Table 18-10. Findings on Brain Imaging in HE Patients CT Cortical atrophy (worse in alcoholic liver disease) Subtle suggestion of cerebral edema Rarely frontal cerebral edema seen MRI Hyperintensity of basal ganglia on T1-weighted images Newer techniques identify frank cerebral edema Cerebral and cerebellar atrophy
318
Good correlation between blood ammonia and the severity of HE has been established Because of frequency of false positives and false negatives an elevated ammonia level does not meet the criteria for a reliable diagnostic test Ammonia assays are often poorly performed If a patient is felt to have HE and ammonia level is normal the patient is still usually treated as having HE, especially if has known severe liver disease If a patient is felt to have HE and ammonia level is elevated then patient is still treated for HE Potentially the only use for ammonia testing is in cases where there is no clear cause for a change in mental status
Chapter 18 HEPATIC ENCEPHALOPATHY
one of the most important aspects to reversing a clinical bout of overt HE is the identification and correction of the precipitating factor(s). Indeed, a four-pronged approach is the basis of treatment of all bouts of HE. This includes (1) supportive care for patients with an altered sensation; (2) a search for and correction of precipitating factors; (3) careful exclusion and/or treatment of concomitant medical illnesses and causes of encephalopathy; (4) the commencement of empirical therapy. This type of care requires an ICU for patients with severe (stage III/IV) HE. Nowadays these strategies overlap somewhat. For instance, the empirical use of lactulose also corrects the precipitating factor of constipation, or eliminates large gut loads of blood after successful attempts to quell variceal bleeding. Maintenance of hydration in the comatose patient may in fact lead to rapid recovery from HE. Having said this, one needs to look at the specific agents used to treat HE over the years. Treatment will be discussed first from a more general historical perspective, and then the literature will be examined in more detail.
HISTORICAL REVIEW As ammonia began to be suspected as the toxin that caused brain dysfunction in cirrhotic patients in the 1930s,117,118 and more definitively in the 1950s,16,17,18 various treatments were employed. Initially antibiotics such as tetracyclines were used to reduce the production of ammonia by gut flora.119 Later, poorly absorbed antibiotics such as neomycin became the standard treatment.120–122 Neomycin continued to be used extensively until the arrival of lactulose in the 1960s.123 Interestingly, neomycin was subjected to its first comparison with placebo in two trials in the early 1990s,124,125 and was found to be equivalent in efficacy! Nonetheless, it continued to be used as a comparison drug in many more trials in the years that followed.126–128 Perhaps one of the reasons for this was the lack of diarrhea with neomycin, which helped blinding in these studies, often against other antibiotic therapies.129–131 After the initial report of probable efficacy in 1966, lactulose was featured in a number of RCTs in the 1970s. Two early trials versus placebo showed little efficacy.132,133 Following these studies, major RCTs were performed comparing lactulose to neomycin.126–128 None of these showed a superiority of lactulose over neomycin, but complicating matters considerably was the use of sorbitol with neomycin, thereby causing diarrhea equivalent to that with the use of lactulose. Because diarrhea itself134 and sorbitol135 may be somewhat effective as agents in the treatment of HE, this may explain why neomycin was in many studies found to be slightly superior to lactulose. In any event, despite the results of these trials lactulose was adopted as the drug of choice to treat HE, presumably because it was thought to be equivalent to neomycin (even though insufficient data supported this) and had fewer side effects. Modern studies of the treatment of HE have been dominated by two issues. The first is the belief that lactulose is a well proven, highly effective treatment, and that therefore all new therapies should be compared to it. Secondly, Human Investigation Review Boards (IRBs) began to believe that placebo treatment was unethical. The first issue was a real concern to IRB members, who felt that lactulose was daily shown to be effective in the wards. The problem with this perspective is no patient on the ward with overt HE was treated just with lactulose: all had the full totality of HE treatment
applied in every circumstance. One could never tell whether lactulose was the reason patients recovered from HE episodes, or whether the recovery was due to the treatment of other factors. The other concern was what to do with the patient receiving placebo in whom the HE worsens. The publication of a series of systemic reviews has called into question virtually every therapy that has been employed in the treatment of HE.136–141 A plea was made to do what is normally done to establish the efficacy of drugs, i.e. to compare them to placebo in a controlled situation, and such studies are being planned. One could dismiss the entire body of literature on HE treatment as we restart to establish efficacy for each agent; however, there is much to be learned from the literature, in terms of both lessons for future trials and of genuine clinical findings that are relevant for the practicing physician.
GENERAL ISSUES The first three items in Table 18-12 are very important in the management of overt HE. No data are available to specify how frequently these measures, either singly or in combination, correct a bout of overt HE. The spontaneous improvement rate with or without these measures is also unknown. However, figures in the range of 40–70% have been suggested.125,142 One problem with assessing the impact of correcting precipitating factors is the near universal use of lactulose as the laxative of choice to cleanse the gut. Hence the correction of precipitating factors contains an element of use of an active treatment agent. Many decades ago laxative treatment was shown to have some benefit in HE,134 although other studies have refuted this claim. Recent studies have once again suggested a benefit of gut lavage using bowel cleansing agents such as polyethylene glycol, along with a motility agent such as neostigmine.143 As there have been at least two reports of HE being associated with slowed intestinal motility,144,145 this new version of bowel cleansing makes some sense. Meantime, copious amounts of lactu-
Table 18-12. Treatment of Hepatic Encephalopathy Supportive care of unconscious or confused patient Identification and treatment of concomitant disease Careful search for and correction of precipitating factors Reduction of ammonia production and absorption Lactulose, lactitol, lactose orally or in enema form Dietary protein restriction Poorly absorbed antibiotics Disaccharide inhibitors Probiotics Promotion of waste nitrogen excretion L-Ornithine-aspartate (LOLA) IV and PO Sodium benzoate IV and PO Correction of neurotransmitter abnormalities in the brain Flumazenil Branched-chain amino acid-enriched formulations Dopaminergic agents Narcotic antagonists – naloxone Zinc repletion (many possible modes of action) Portosystemic shunt suppression Liver support systems Liver transplantation
319
Section III. Clinical Consequences of Liver Disease
lose are introduced into the gut of HE patients either orally, via a nasogastric tube, or by enema. As mentioned earlier, this approach is only part of a multimodality therapy applied to all patients with overt HE. This uncontrolled situation seems to work well in the majority of cases, but does not prove a specific efficacy for lactulose. However, diarrhea can lead to severe hypernatremia (e.g. serum sodium > 150 mmol/L), which is associated with a high mortality rate, and thus lactulose, like all drugs, has side effects that may be associated with morbidity and mortality.146
TREATMENTS THAT REDUCE AMMONIA PRODUCTION AND ABSORPTION IN THE GUT Non-absorbed disaccharides, such as lactulose or lactitol, have been tested as treatments for episodes of HE.70,72,126–128,136,137,147–156 Their mechanism of action is complex. First, they are quite effective osmotic laxatives, but the production of an acidic environment after metabolism in the colon is felt to be a crucial aspect of their efficacy in HE.157–163 This luminal acidification has been shown to trap ammonia in the colon and promote its incorporation into bacterial proteins.163 However, the acidification theory has been questioned in one study.159 The possibility that lactulose may inhibit intestinal glutaminase activity has been raised as another possible mechanism of action.164 Increased glutaminase activity in the small intestine has been noted in patients with advanced cirrhosis, possibly providing another explanation for the blood ammonia-lowering action of lactulose.165 Although these possible mechanisms of action of lactulose are interesting, the fact remains that its credentials as the standard of care in the treatment of HE are very weak (Tables 18-13 and 1814): so weak, in fact, that in this age of evidence-based medicine lactulose would be considered an unproven therapy.137 The recent systematic review of all poorly absorbed disaccharides seriously questions the overall validity of lactulose as a standard of care for HE. In the only placebo-controlled trials of lactulose in the treatment of HE there was no statistical superiority of lactulose over placebo. Somewhat larger RCTs comparing lactulose to neomycin Table 18-13. RCTs of Lactulose versus Placebo for Overt HE Study
n
Comparator
Number improved/total lactulose vs placebo Crossover no difference in efficacy 10/14 vs 7/12 Crossover no difference in efficacy 5/9 vs 6/9
Elkington147
7
Sorbitol
Simmons132 Rodgers148
26 6
Glucose Sorbitol
Germain133
18
Saccharose
Table 18-14. RCTs of Lactulose versus Treatment for Minimal HE Study
n
Comparator
Number improved/total lactulose vs no tx
Watanabe150 Li132 Phiman153
36 86 26
No Tx No Tx No Tx
10/22 vs 3/14* 26/48 vs 11/38* 8/14 vs 0/12*
*Lactulose significently better than no treatment
320
(usually with added sorbitol)126–128 have been stated to show an equivalence in efficacy of these two agents. However, virtually none of these studies had sufficient patients to demonstrate equivalence by today’s standards, as equivalence studies require far more patients than studies where one therapy is superior to another. Moreover, poorly absorbed antibiotics in some of the RCTs were more effective than lactulose.126 Exactly how lactulose came to be the standard of care is difficult to comprehend, but would appear to have been more because of its lower toxicity relative to neomycin rather than any demonstrated superior efficacy. The call for placebo-controlled trials to clarify the efficacy of lactulose (or lactitol) in the treatment of HE137 awaits the resolution of the ethical constraints in performing such trials. The solution may be to study patients with stage II HE, as has been done by a number of investigators already. This alleviates the major problems encountered with more severe HE in patients with advanced cirrhosis, where there is concern that failure to treat the HE may be associated with unacceptable risks to the patient.166
Poorly Absorbed Antibiotics Despite the lack of definite proof that neomycin was more effective than placebo, it is still used for the treatment of HE and needs to be discussed.124,125 The mechanism of action of neomycin is at least twofold. First, the antibiotic suppresses the growth of aerobic intestinal flora, which in turn reduces ammonia production in the gut.167,168 More interestingly, neomycin was noted in a number of reports to induce malabsorption and villous atrophy in the small intestine.169,170 If one proposes that neomycin actively helps HE, then a reduction of intestinal glutaminase activity may be an important mode of action of the drug. Weber et al.171 demonstrated in dogs that the small bowel contributes quantitatively more ammonia to the portal vein than does the colon. This ammonia arises primarily from mucosal glutaminase activity, and if neomycin reduces glutaminase activity this would reduce ammonia production. Also, the possibility that small bowel bacterial overgrowth in cirrhotic patients was treated by oral neomycin therapy has also been considered.172 A number of studies have been performed comparing rifaximin to either other antibiotics or lactulose/lactitol (Table 18-15). Like neomycin, rifaximin is considered non-absorbable from the intestine. However, only a tiny amount is absorbed, unlike the 3% or so of neomycin.173,174 Rifaximin has a much broader range of antibacterial action, yet has not been reported to cause fungal overgrowth.173,174 Enteric flora develop a resistance to this antibiotic by a non-plasmid mediated mechanism, but the resistance is insufficient to prevent antibacterial action by the high levels of rifaximin in the gut. A cycling system of 2 weeks on and 2 weeks off rifaximin for 6 months has been tested for long-term control of HE.175 Apparently this works well, which may be useful for long-term treatment. Most studies, however, have been shorter, and all the data have found equivalence between rifaximin and neomycin or lactulose or lactitol.129–131,176–183 One study compared rifaximin to lactitol in over 102 patients and confirmed the efficacy of rifaximin reported in earlier studies.184 Some of the design issues mentioned with other treatments of HE also apply to rifaximin. Most comparison trials have had insufficient numbers to compare equivalency with other treatments. This is a moot point, as most other therapies are of
Chapter 18 HEPATIC ENCEPHALOPATHY
Table 18-15. RCTs of Non-absorbable Disaccharide vs Antibiotics for Overt HE Study
n
Disaccharide/Other
Antibiotic/Other
Number improved/total
Conn128 Atterbury127 Orlandi126 Russo154 Blanc125 Bucci176 Fera177 Festi131 Massa178 Song179 Longuerico180 MAS184
33 47 190 15 60 58 40 21 40 64 27 103
Lact/placebo Lact/placebo Lactulose Lactulose Lactitol Lact + placebo Lact + placebo Lactulose Lact + placebo Lactitol Lact + placebo Lact + placebo
Neomycin/sorbitol Neomycin/sorbitol Neomycin/MgSo4 Ribostamycin Vancomycin Rifaximin/sorbitol Rifaximin/placebo Rifaximin Rifaximin/sorbitol Rifaximin Rifaximin/placebo Rifaximin/placebo
15/18 vs 13/15 19/23 vs 20/24 28/91 vs 34/82 7/8 vs 5/7 20/29 vs 21/31 Equivalent results 16/20 vs 20/20 Equivalent results 18/20 vs 20/20 18/25 vs 31/39 2/13 vs 8/14 41/53 vs 40/50
unproven value. At this point it is important to state that although most treatments of HE have not been proved effective in placebocontrolled RCTs, this is not to say that they are without effect. The probability is that most have some efficacy, but deficiencies in study design have led to equivocal results in many trials. A well designed placebo-controlled trial is still awaited. Other antibiotics have been used to treat HE. Metronidazole became very popular in the early 1980s as a result of a study published by Morgan et al.185 They reported a similar efficacy for metronidazole compared to neomycin despite 90% absorption of the drug in the upper intestine. In many centers metronidazole is the preferred second-line drug if lactulose in ineffective or poorly tolerated. However, like most treatments for HE it has not been compared in efficacy to placebo. Also, the number of patients in Morgan et al.’s trial was insufficient to declare equivalency of treatment effect. Nonetheless, as with lactulose, many physicians are passionate believers in metronidazole as an effective (if unproven) therapy for HE. Accumulation of metronidazole in advanced liver disease is a significant risk. CNS toxicity and peripheral neuropathy are well recognized side effects, especially if long-term (>2 weeks) therapy is used. Vancomycin was used extensively after a report by Tarao et al.186 that it had efficacy in lactulose-resistant HE. Its popularity as perhaps the third-line drug after lactulose and metronidazole faded with the advent of vancomycin-resistant enterococcus. Most physicians voluntarily ceased its use for this indication.
DIETARY PROTEIN RESTRICTION Returning to the theme of lessons learned from earlier treatment trials in HE, dietary protein restriction has an interesting history. The precipitation of overt episodes of HE by dietary protein loading was first reported in the 1950s.17,18 Many patients in subsequent RCTs had prior portosystemic shunts and were felt to be exquisitely sensitive to even normal dietary protein intake.75 Accordingly, in many trials a standard 40 g/day protein intake was employed,155,158 in addition to whatever therapies were being investigated for efficiency in the treatment of HE. Although it was not the aim of these studies, the basic concept was to establish whether agents such as neomycin or lactulose could improve oral protein tolerance in the long term. What was actually tested was the ability of these treatments to improve an episode of HE. As few placebo-controlled trials
were performed, it is difficult to gauge whether protein restriction or the active agent under test was responsible for any improvement in HE. In comparison studies, for instance, the efficacy of neomycin versus lactulose was not actually investigated.155,158 Instead, the addition of these agents to dietary protein restriction was tested in patients experiencing a bout of overt HE. Because dietary protein overload was frequently the dominant precipitating factor in these episodes,75 one can see the problems created by the use of standardized oral protein restriction for all patients. Indeed, one wonders whether the failure of agents such as neomycin or lactulose to be proved superior to placebo124,125,132,133 in RCTs was in part due to an improvement in HE in the placebo arm using oral protein restriction alone. Dietary protein restriction still is used in patients with recurrent bouts of overt HE,187 despite strong statements from experts that this mode of therapy should not be employed.188 Dietary protein restriction below the maintenance level (e.g. 0.8 g/kg/day) may in fact lead to lean body mass catabolism and may ultimately increase the nitrogen load to the systemic circulation. Patients truly found to have thresholds for the induction of HE below maintenance daily protein requirements should be treated with alternative protein sources, such as amino acid formulations.189 Vegetable protein85 or enteral branched-chain amino acid-supplemented regimens190 may permit an adequate daily intake of protein without aggravating HE. Recently a study was published showing that emergence from a bout of overt HE was not delayed if a normal protein diet was delivered in addition to standard HE therapy, and so the efficacy of such therapy in patients receiving a standard protein diet should be examined.191 Before summarizing the role of protein restriction in the treatment of HE it is worth remembering that we frequently provide minimal nutritional support to patients in the first few days of a bout of severe HE. One of the potential virtues of the branched-chain amino acid-enriched formulation was the perceived need to rapidly initiate and then maintain adequate nutritional support in such patients. The data on the efficacy of these formulations are controversial.139,192–197 Suffice it to say at this point that the data in support of these special parenteral formulations as a specific treatment for HE are not compelling. Currently, oral protein restriction is not advocated as a treatment for HE. Of necessity it occurs in the first few days of treatment of severe disease, but should virtually be
321
Section III. Clinical Consequences of Liver Disease
never used in long-term care. If oral dietary protein intolerance is documented then a vegetable protein diet is recommended,198,199,200 with additional pectin or fiber.201–203 Branched-chain amino acidenriched supplements47,110,204 or formulations are still useful in highly selected protein-intolerant patients, but again at a high financial cost.205,206
DISACCHARIDASE INHIBITORS Inhibitions of disaccharidase enzyme activity in the small intestine have been reported in at least two trials to be efficacious in the treatment of HE.78,207 This approach essentially causes carbohydrate malabsorption, which in many ways works in an equivalent fashion to non-absorbable disaccharides such as lactulose. Both trials featured patients with relatively mild HE, but improvements were fairly well documented. The most recent study exclusively enrolled diabetic patients with HE and demonstrated an improvement in glycemic control as well.78 As both trials were double-blind RCTs they should be given attention, especially as they both had a placebo-controlled arm. However, as is often the case in HE trials, the characterization of the patients enrolled in the study is rather sparse (e.g., how many on therapy before enrollment, etc.). In addition, as in all studies with treatment agents that cause diarrhea, one has concerns about the blinding process. Nonetheless, we may see more advances in this form of treatment in the years to come.
PROBIOTICS As in many disciplines in medicine, there has been renewed interest in modifying the gut flora to treat HE.208–212 Early studies with Lactobacillus acidophilus appear to have been discouraging, but may have been confounded by the difficulty of maintaining this bacterium in the gut flora.213 Another study included neomycin, which might have contributed to the reported modest efficacy.214 Repeated oral administration of encoated Enterococcus SF was deemed as effective as lactulose in both short-term215 and one long-term controlled trial.216 This type of bacterium is fermentative, lactic-acid producing, urease negative, and resistant to certain antibiotics. More studies are beginning to appear in this interesting approach to HE therapy.
PROMOTION OF WASTE NITROGEN EXCRETION Hyperammonemia in patients with advanced liver disease arises by multiple mechanisms. Prevention of the generation and absorption of ammonia in the gut has been the primary focus of the commonly used treatments for HE (e.g. lactulose, neomycin etc.); however, another approach to this problem is to promote the excretion of waste nitrogen. This can be approached by enhancing what remains of the liver’s capacity for ureagenic and glutamine synthesis, or by providing substances that fix ammonia and are excreted in the urine. More drugs working along these lines are being developed. Some of the existing data will be reviewed here. L-Ornithine–L-Aspartate (LOLA) Apart from its catchy name, there are many virtues to this therapy for HE. First, its potential mechanism of action is quite well understood and it promotes ureagenesis and glutamine synthetase activity in the liver.217 There is also evidence that it promotes glutamine
322
synthesis and possibly protein anabolism in skeletal muscle.218 Most importantly from the clinical perspective, this drug has been tested in both a parenteral and oral form for efficacy in HE in placebocontrolled randomized trials.142,219 The study of infused LOLA is perhaps one of the better studies to be published for decades.142 It clearly demonstrated significant improvement in stage II HE using multiple measures. Interestingly, it also demonstrated a 40% improvement rate in the placebo arm of the study. Several other trials have been performed with this agent versus placebo with, by now, a total enrollment of 370 or more patients.220 The drug has not officially been released for treatment of HE in any country, which is unfortunate given the findings in the above studies.
Sodium Benzoate Metabolic fixation of ammonia using sodium benzoate to form hippurate, originally used to promote alternative paths of waste nitrogen excretion in patients with inborn errors of urea synthesis, has been explored as a treatment for HE.221 The only published RCT compared sodium benzoate to lactulose in the treatment of overt episodes of HE.222 The two agents were judged to be equivalent in efficacy, but the study may have lacked the numbers to prove equivalency by today’s standards. Its low cost and availability in oral and parenteral form suggests that sodium benzoate should be studied further as a treatment for HE. Unfortunately, it has an unpleasant taste and its sodium content is a concern in patients with advanced liver disease. A number of physicians use this agent at a dose of 5 g orally twice daily as a second-line drug to treat HE. They obtain it from a commercial pharmacy.
CORRECTION OF NEUROTRANSMITTER ABNORMALITIES IN THE BRAIN As numerous abnormalities in brain chemistry were identified in HE primarily in animal models, therapy was designed to try and connect these derangements. Some agents showed promise, but currently none is used frequently. Consequently, even though there is a large body of information on these potential approaches to the treatment of HE only selective comments will be made.
Flumazenil The reversal of HE in an animal model without access to prescription benzodiazepine drugs was interpreted to indicate that the accumulation of endogenous benzodiazepines might be a factor in HE.50 This possibility was later examined in a series of clinical studies.223–225 The subsequent discovery of benzodiazepines detected by assays in cirrhotic patients verified to be drug free supported this hypothesis.60,73,226 Because flumazenil is a very selective benzodiazepine antagonist it was logical to perform studies to evaluate its efficacy as a treatment for HE. Table 18-16 briefly outlines the reported efficacy in patients with overt and minimal HE, respectively.52–59 Unlike non-absorbable disaccharides, the efficacy of flumazenil appeared to be far more demonstrable in overt versus minimal HE. This tends to suggest that minimal HE is not mediated to any significant degree by endogenous benzodiazepines. Although the consistent reporting of improvements in overt HE with flumazenil is encouraging, it has not translated into widespread use of this agent. There are valid reasons for this, not the least of
Chapter 18 HEPATIC ENCEPHALOPATHY
Table 18-16. RCTs of Flumazenil versus Placebo for Overt and Minimal HE
Study
n
Pomier-Layrargues53 Cadranel55 Gyr56 Barbaro52 Zhu57 Lacetti58
21 18 49 527 25 21
(A) Overt Study design Crossover Crossover Parallel Crossover Parallel Parallel
Improvement in HE Drug vs placebo 5/11 vs 0/11* 6/10 vs 0/8* 7/28 vs 0/21* 66/265 vs 9/262* 3/13 vs 0/12* 5/11 vs 0/10
*Signifies superior to placebo.
Table 18-17. Current Status of BCAA Supplementation in HE Parenteral use of branched-chain amino acid-enriched formulation is rare Enteral supplements of BCAAs in a standard diet are still used selectively In general enteral studies show benefit in terms of preventing or ameliorating HE while improving total protein intake The cost and skepticism regarding efficacy has limited use of this therapy in the USA
Table 18-18. More Common Reasons for HE Resistant to Treatment (B) Minimal
Kapczinski54 Gooday138 Amodio138 Dursun59
20 10 13 40
Crossover Crossover Crossover Parallel
No significant effect Flumazenil superior No effect 8/20 vs 0/20* Flumazenil superior ? All minimal HE
which is a lack of a long-acting oral preparation to treat patients who respond to an initial dose.8 Also, many physicians believe that the flumazenil response rate is not only modest (~30%), but may be due to occult ingestion of prescription benzodiazepines. There is good evidence against the latter, but the problem with detection assays for endogenous and pharmaceutical benzodiazepines is that both are variably detected and indeed may be identical.60 One assumes that this form of treatment will be used in selected cases in the ICU, but will not enter the regular HE treatment arena until the identity and source of endogenous benzodiazepines are clarified and a long-acting antagonist becomes available. One interesting aspect to flumazenil treatment trials in HE is that it illustrates that placebo-controlled trials can be performed in very severe HE if an ultrafast therapy (working in minutes/hours) is being tested.
Branched-Chain Amino Acid (BCAA)-Enriched Formulations This form of therapy has been subjected to detailed analysis.139,192–197 The results of these trials can be summarized by stating that the parenteral form of therapy has never been validated as a treatment for HE. In contrast, the enteral form of BCAA supplements has some reasonable evidence in its favor.205,206 Whether the gains in preventing HE occurrences are worth the cost is debatable. As mentioned earlier, vegetable protein-based diets are now employed for patients with dietary intolerance to a typical American diet. The evidence in favor of this approach is not strong either. The extensive systematic review by Ahls-Nielsen et al.139 is excellent reading for those interested in this field. Rather than trying to repeat their excellent review, the salient points are shown in Table 18-17.
Dopaminergic Agents Also the subject of a recent systematic review,140 dopaminergic agents have a very limited role to play in the management of HE. However, the data available do not preclude some benefit from this
End-stage liver disease only – still should be able to rouse Excessive purgation leading to dehydration/free water loss Failure to identify and treat sepsis Ileus, especially in association with azotemia (may need dialysis) Long-acting sedative drug intake Undiagnosed concomitant CNS problem, e.g. hypothyroidism Too effective portosystemic shunt procedure Profound zinc deficiency
form of treatment. L-Dopa and bromocriptine, the dopaminergic agents used in these trials, may have a role to play in the management of the apparently newly discovered cirrhotic patient with extrapyramidal symptoms and no overt signs of HE.79,80 It is distinctly possible that more rigorously designed studies might in future change the view that these therapies are ineffective.
Opiate Antagonists No RCTs of narcotic antagonist treatment of HE have been published. Data from animal models of HE show a reversal of disease with naloxone.66,227 Moreover, there is considerable evidence of accumulation of endogenous opiates, particularly enkephalins, in patients with advanced liver disease.228 Opiate antagonists may play a significant role in pruritus in patients with cholestasis,68,69 but the jury is still out with regard to HE. The rapidity of action of narcotic antagonists such as naloxone would make it appear to be relatively easy to perform an RCT of this type of therapy.
Zinc Repletion Two major studies have been published regarding the effect of zinc supplementation on HE,229–231 but after one encouraging report229 enthusiasm for the treatment has waned.230 In resistant cases zinc has merit as a treatment, especially if leukocyte zinc levels are low.232 A report appeared of HE that was resistant to all therapy until zinc repletion was tried.233
HE RESISTANT TO THERAPY Having discussed most of the agents used over the years to treat HE, a few comments about resistant HE are in order (Table 18-18). It is actually very rare to be unable to rouse a patient from severe HE, no matter how bad their liver function has become. It may be difficult to keep patients out of overt bouts of HE, but at least temporary arousal should be achievable with one or more of the
323
Section III. Clinical Consequences of Liver Disease
treatments outlined in the previous sections. If no arousal is seen a number of explanations should be considered. The first and perhaps most common contributory factors to resistant HE are inadvertent dehydration and untreated sepsis. The former has already been mentioned and is a particular problem in patients with advanced liver disease. If serum sodium levels are in the high normal to above normal range the patient may be dehydrated, leading to resistant HE. Intravenous D5W will generally rouse these patients. Untreated sepsis prevents emergence from severe HE and should be vigorously sought in all patients with severe HE who do not arouse after 3–4 days of treatment. The combination of renal failure and ileus, together with severe spontaneous bacterial peritonitis, often results in HE resistant to therapy. Indeed, little other than enemas can be given in this situation. Hemofiltration may be needed to control the situation until bowel function returns. Occasionally long-lasting metabolites of sedative drugs can lead to protracted HE. Drug screens can help in identifying this population, even though we still cannot distinguish between endogenous and pharmaceutical benzodiazepines. Coincidental hypoadrenalism and hypothyroidism can give the appearance of resistant HE and need to be uncovered before liver transplantation is attempted. Sustained HE after a portosystemic shunt can be improved by measures that reduce the shunt size or flow. Finally, severe zinc deficiency has been reported to cause resistant HE.233 Small bowel loss of zinc can occur with excessive purgation.
LIVER SUPPORT SYSTEMS The quest for liver support systems that either stabilize or correct some of the manifestations of liver failure continues.234 Most of the literature on this topic features uncontrolled trials, which initially show promise. Subsequently, the support systems often disappear from view or fail to perform satisfactorily in an RCT.235–242 The molecular absorbent recirculating system (MARS) seems to have broken this trend, with the report of Heeman et al.,235 who reported apparent benefit using this device in patients with acute or chronic liver failure. The study was terminated prematurely because the MARS-treated group were doing ‘too well’ and thus the study was underpowered. Nonetheless, it represents a positive outcome in an area where few studies show clear benefit in favor of liver support devices. One other device, the so-called Biologic DT system, also seemed to be effective in one study.243 The ‘mistake’ with testing of these systems was their original use in end-stage liver disease patients in either acute or chronic liver failure.234 As mentioned earlier, it is difficult to perform controlled trials in this population. The acute-on-chronic liver failure syndrome, especially if the liver support system is applied earlier in the course of liver failure, may represent good testing situations with liver support systems. Interestingly, in virtually all reports the improvement in HE with these systems is striking.244,245 Survival, however, is the usual endpoint in such studies, and a survival benefit has been difficult to demonstrate. Systematic reviews of the work to date have appeared in the literature,246,247 and the current status of these systems is briefly described in Table 18-19.
324
Table 18-19. Current Status of Liver Support Systems in HE Reasonably good evidence exists that these devices improve HE Most studies have been in very advanced liver failure. This results in multiple uncontrollable events affecting outcome The molecular albumin recirculating system (MARS) has been the most tested system to date. It shows promise as a safe and fairly effective liver support system Early use in acute on chronic liver failure may be associated with a better outcome The precise mechanism whereby these systems improve HE is not known
Table 18-20. Issues of Liver Transplantation and HE HE is not included in the MELD score No priority is given to patients with severe recurrent or resistant HE Failure to show HE independently predicts survival may be related to its close association with liver function status Older data on the failure of HE to predict survival in advanced liver disease patients may have occurred because of overstaging of HE to improve priority listing New association of the independent predictive power of HE on survival may be more valid in the MELD era. However, HE status needs to be reported more exactly
CLOSURE OF PORTOSYSTEMIC SHUNTS Despite the paucity of signs of HE in most patients with portosystemic shunts who do not have liver disease, there are those with liver disease who experience severe and recurrent bouts of HE attributable to portosystemic shunting.85 The classic patient has excellent liver synthetic function and yet has bouts of severe HE. When this type of patient is encountered, imaging studies of the abdomen should be performed to identify large collateral vessels. Closure of these vessels by either radiological or surgical interventions is sometimes feasible.248–255 However, before proceeding to attempt these maneuvers it is best to consult a transplant surgeon to ensure that this will not preclude or interfere with later liver transplantation. Reduction of pre-existing transhepatic stent diameters, or closure of large collateral vessels, can have a dramatic effect on the course of HE.249
LIVER TRANSPLANTATION Although it is not commonly the primary indication for liver transplantation, HE is generally improved by a successful graft. Table 1820 lists some of the current issues regarding HE and liver transplantation. Rather than reiterate these points in detail, the issue that will be discussed is the timing of referral for transplantation assessment in patients with drug and alcohol abuse problems. Successful completion of a rehabilitation program to qualify for liver transplantation requires well preserved cognitive skills, which may be limited because of HE. Tragically, we see too many patients with drug problems referred when they already have difficult-to-control HE. At present, recurrent or persistent HE does not give patients
Chapter 18 HEPATIC ENCEPHALOPATHY
priority for liver transplantation. Moreover, the impact of HE episodes can be such that patients often fail to complete drug rehabilitation programs satisfactorily.
SUMMARY The terminology of HE has officially been changed, and this system should be used in future publications in this field. After years of discord there is general agreement that cerebral edema is present to some degree in all types of HE. Ammonia, either via its own action or via astrocyte function, is the major theory under review in studies on the pathogenesis of HE. Other theories thought to be distinct are now being found to fit into this unifying hypothesis of disturbances in cellular function centering on the astrocyte. The state of affairs as regards evidence in support of most of the therapies for HE is poor. Newer perspectives are being brought to bear on improving study design for future trials. New treatments are being developed that will benefit from better designed RCTs in the future.
REFERENCES 1. Adams RD, Foley JM. The neurological changes in the more common types of severe liver disease. Trans Am Neurol Assoc 1949;74:217–219. 2. Adams RD, Foley JM. The neurological disorder associated with liver disease. Proc Assoc Res Nerv Ment Dis 1953;32: 1998–237. 3. Summerskill WHJ, Davidson EA, Sherlock S, Steiner RE. The neuropsychiatric syndrome associated with hepatic cirrhosis and/or extensive portal collateral circulation. Q J Med 1956;25:245–266. 4. Mullen KD. Terminology of hepatic encephalopathy: Work in progress. In: Yurdaydin C, Bozkaya H, eds. Advances in hepatic encephalopathy and metabolism in liver disease. Ankara: Ankara University Press, 2000: 375–380. 5. Ferenci P, Lockwood A, Mullen K, et al. Hepatic encephalopathy – definition, nomenclature, diagnosis, and quantification: final report of the working party at the 11th World Congress of Gastroenterology, Vienna, 1998. Hepatology 2002;35:716– 721. 6. Arroyo V, Gines P, Gerbes AL, et al. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. International Ascites Club. Hepatology 1996;23:164–176. 7. McDermott WV Jr, Adams RD. Episodic stupor associated with Eck fistula in the human with particular reference to the metabolism of ammonia. J Clin Invest 1954;33:1–9. 8. Ferenci P, Grimm G, Meryn S, Gangl A. Successful long term therapy of portal–systemic encephalopathy by the benzodiazepine antagonist flumazenil. Gastroenterology 1989;96:240–243. 9. Mullen KD. Interplay of portal pressure, portal perfusion and hepatic arterial inflow in patients with spontaneous or artificially created portosystemic shunts. Indian J Gastroenterol 2003;22(Suppl 2):S25–S27. 10. Zeneroli ML, Cioni G, Vezzelli C, et al. Prevalence of brain atrophy in liver cirrhosis patients with chronic persistent encephalopathy. Evaluation by computed tomography. J Hepatol 1987;4:283–292. 11. Zeneroli ML, Cioni G, Crisi G, et al. Globus pallidus alterations and brain atrophy in liver cirrhosis patients with encephalopathy: an MR imaging study. Magn Reson Imag 1991;9:295–302.
12. Mullen KD, Cole M, Foley JM. Neurological deficits in ‘awake’ cirrhotic patients on hepatic encephalopathy treatment: missed metabolic or metal disorder? Gastroenterology 1996;111:256–257. 13. Krieger S, Jaub M, Jansen O, et al. Neuropsychiatric profile and hyperintense globus pallidus on T1 weighted magnetic resonance images in liver cirrhosis. Gastroenterology 1996;111:147–155. 14. Victor M, Adams RD, Cole M. The acquired (non-Wilsonian) type of chronic hepatocerebral dysfunction. Medicine 1965;44:345–396. 15. Powell EE, Pender MP, Chalk JB, et al. Improvement in chronic hepatocerebral degeneration following liver transplantation. Gastroenterology 1990;98:1079–1082. 16. Gabuzda GJ, Phillips GB, Davidson CS. Reversible toxic manifestations in patients with cirrhosis of the liver given cation exchange resions. N Engl J Med 1952;246:124–130. 17. Phillips GB, Schartze R, Gabuzda GJ, Davidson CS. Syndrome of impending coma in patients with cirrhosis of the liver given certain nitrogenous substances. N Engl J Med 1952;247:239–246. 18. Schwartz R, Phillips G, Seegmiller JE, et al. Dietary protein in the genesis of hepatic coma. N Engl J Med 1954;251:685– 689. 19. Stahl J. Studies of blood ammonia in liver disease. Ann Intern Med 1963;98:1–24. 20. Muting D, Kalk JF, Koussouris P, et al. The role of protein metabolism in 204 liver cirrhotics with and without hepatic encephalopathy. I. Clinical and general biochemical findings. Hepatogastroenterology 1986;33:61–65. 21. Kramer L, Tribl B, Gendo A, et al. Partial pressure of ammonia versus ammonia in hepatic encephalopathy. Hepatology 2000;31:30–34. 22. Ong JP, Aggarawal A. Krieger D, et al. Correlation between ammonia levels and the severity of hepatic encephalopathy. Am J Med 2003;114:188–193. 23. Nicolao F, Ferati C, Masini A, et al. Role of determination of portal pressure of ammonia in cirrhotic patients with and without hepatic encephalopathy. J Hepatol 2003;38:441–446. 24. Cooper AJL, Plum F. Biochemistry and physiology of brain ammonia. Physiol Rev 1987;67:440–519. 25. Msall M, Batshaw ML, Suss R, et al. Neurologic outcome in children with inborn errors of urea synthesis. Outcome of ureacycle enzymopathies. N Engl J Med 1984;310:1500–1505. 26. Iles JF, Jack JJB. Ammonia: assessment of its action on postsynaptic inhibitions as a cause of convulsions. Exp Brain Res 1970;11:431–437. 27. Eichler M, Bessman SP. A double-blind study of the effects of ammonium infusion or psychological functioning in cirrhotic patients. J Nerv Ment Dis 1962;134:539–542. 28. Cohn R, Castell DO. The effect of acute hyperammonemia on the electroencephalogram. J Lab Clin Med 1982;2:495–506. 29. Gitlin N. Hepatic encephalopathy. In: Zakim D, Boyer TD, eds. Hepatology. A textbook of liver disease. Philadelphia: WB Saunders, 1996: 605–617. 30. Blei A. Hepatic encephalopathy. In: Bircher J, Benhamou JP, McIntyre N, et al., eds. Clinical hepatology. Oxford: Oxford University Press, 1999: 765–783. 31. Butterworth RF. The neurobiology of hepatic encephalopathy. Semin Liver Dis 1996;16:235–244. 32. Ferenci P, Puspok A, Steindl P. Current concepts in the pathophysiology of hepatic encephalopathy. Eur J Clin Invest 1992;22:573–581. 33. Haussinger D, Kircheis G, Fischer R, et al. Hepatic encephalopathy in chronic liver disease: a clinical manifestation of astrocyte swelling and low-grade cerebral edema? J Hepatol 2000;32:1035–1038. 34. Blei AT. Pathophysiology of brain edema in fulminant hepatic failure, revisited. Metab Brain Dis 2001;16:85–94.
325
Section III. Clinical Consequences of Liver Disease
35. Ross BD, Jacobson S, Villamil F, et al. Subclinical hepatic encephalopathy: proton MR spectroscopic abnormalities. Radiology 1994;193:457–463. 36. Cordoba J, Alonso J, Rovira A, et al. The development of lowgrade cerebral edema in cirrhosis is supported by the evolution of (1)H-magnetic resonance abnormalities after liver transplantation. J Hepatol 2001;35:598–604. 37. Jayakumar AR, Rama Rao KV, Kalaiselvi P, Norenberg MD. Combined effects of ammonia and manganese on astrocytes in culture. Neurochem Res 2004;29:2051–2056. 38. Butterworth RF. The astrocytic (peripheral type) benzodiazepine receptor: Role in pathogenesis of portal–systemic encephalopathy. Neurochem Int 2000:36;411–416. 39. Norenberg MD. The role of astrocytes in hepatic encephalopathy. Neurochem Pathol 1987;6:13–33. 40. Norenberg MD. Astrocyte–ammonia interactions in hepatic encephalopathy. Semin Liver Dis 1996;216:245–243. 41. Zieve L, Doizaki WM, Zieve FJ. Synergism between mercaptans and ammonia and fatty acids in the production of coma, a possible role for mercaptans in the pathogenesis of hepatic coma. J Lab Clin Med 1974;83:16–28. 42. Blom HJ, Chamuleau RA, Rothwizer J, et al. Methanethiol metabolism and its role in the pathogenesis of hepatic encephalopathy in rats and dogs. Hepatology 1990;11:682–689. 43. Blom HJ, Ferenci P, Grimm G, et al. The role of methanethiol on the pathogenesis of HE. Hepatology 1991;13:445–454. 44. Fischer JE, Baldessari RJ. False neurotransmitters and hepatic failure. Lancet 1971;2:75–79. 45. Zieve L, Olsen RL. Can hepatic coma be caused by a reduction of brain noradrenaline in dopamine? Gut 1977;18:688–691. 46. James JH, Zipparo V, Jeppson B, Fischer JE. Hyperammonemia, plasma amino acid imbalance and blood–brain amino acids transport: a unified theory of portal–systemic encephalopathy. Lancet 1979;2:772–775. 47. Sato S, Watanabe A, Muto Y, et al. Effect of oral branched chain amino acid granules on event free survival in patients with liver cirrhosis. Clin Gastroenterol Hepatol 2005;3:705–713. 48. Schafer DF, Jones EA. Hepatic encephalopathy and the g-aminobutyric acid neurotransmitter system. Lancet 1982;1:18–19. 49. Mullen KD, Martin JV, Mendelson WB, et al. Could an endogenous benzodiazepine ligand contribute to hepatic encephalopathy? Lancet 1988;27;1:457–459. 50. Bassett ML, Mullen KD, Skolnick P, Jones EA. Amelioration of hepatic encephalopathy by pharmacological antagonists of the GABAA–benzodiazepine reception complex in a rabbit model of fulminant hepatic failure. Gastroenterology 1987;93: 1069–1077. 51. Baraldi M, Zeneroli ML, Ventura E, et al. Supersensitivity of benzodiazepine receptors in hepatic encephalopathy due to fulminant hepatic failure in the rat: reversal by a benzodiazepine antagonist. Clin Sci (Lond) 1984;67:167–175. 52. Barbaro G, Di Lorenzo G, Soldini M, et al. Flumazenil for hepatic coma in patients with liver cirrhosis: an Italian multicentre double-blind, placebo-controlled, crossover study. Eur J Emerg Med 1998;5:213–218. 53. Pomier-Layrargues G, Giguere JF, Lavoie J, et al. Flumazenil in cirrhotic patients in hepatic coma: a randomized double-blind placebo-controlled crossover trial. Hepatology 1995;21:604–605. 54. Kapczinski F, Sherman D, Williams R, et al. Differential effects of flumazenil in alcoholic and nonalcoholic cirrhotic patients. Psychopharmacology Berlin 1995;120:220–226. 55. Cadranel JF, el Younsi M, Pidoux B, et al. Flumazenil therapy for hepatic encephalopathy in cirrhotic patients: a double-blind pragmatic randomized, placebo study. Eur J Gastroenterol Hepatol 1995;7:325–329. 56. Gyr K, Meier R, Haussler J, et al. Evaluation of the efficacy and safety of flumazenil in the treatment of portal systemic
326
57.
58.
59.
60. 61.
62.
63.
64.
65. 66. 67. 68.
69.
70.
71.
72.
73.
74. 75.
76.
77.
encephalopathy: a double blind, randomized, placebo controlled multicentre study. Gut 1996;39:319–324. Zhu C, Wang J, Liu T. Flumazenil in the treatment of cirrhotic patients with hepatic encephalopathy: A randomized doubleblind clinical trial. Chin J Digestion 1998;18:355–358. Laccetti M, Manes G, Uomo G, et al. Flumazenil in the treatment of acute hepatic encephalopathy in cirrhotic patients: a double blind randomized placebo controlled study. Dig Liver Dis 2000;32:335–338. Dursun M, Caliskan M, Canoruc F, et al. The efficacy of flumazenil in subclinical to mild hepatic encephalopathic ambulatory patients. A prospective, randomised, double-blind, placebo-controlled study. Swiss Med Wkly 2003;133:118–123. Mullen KD, Jones EA. Natural benzodiazepines and hepatic encephalopathy. Semin Liver Dis 1996;16:255–264. Yurdaydin C, Walsh TJ, Engler HD, et al. Gut bacteria provide precursors of benzodiazepine receptor ligands in a rat model of hepatic encephalopathy. Brain Res 1995;679:42–48. Aronson LR, Gacad RC, Kaminsky-Russ K, et al. Endogenous benzodiazepine activity in the peripheral and portal blood of dogs with congenital portosystemic shunts. Vet Surg 1997;26:189–194. Zeneroli ML, Venturini I, Stefanelli S, et al. Antibacterial activity of rifaximin reduces the levels of benzodiazepine-like compounds in patients with liver cirrhosis. Pharmacol Res 1997;35:557–560 Pomier-Layrargues G, Shepcott D, Spahn L, Butterworth RF. Accumulation of manganese and copper in palladium of cirrhotic patients: role in the pathogenesis of hepatic encephalopathy? Metab Brain Dis 1995;10:351–354. Krieger D, Krieger S, Jansen O, et al. Manganese and chronic hepatic encephalopathy. Lancet 1995;346:270–274. Yurdaydin C. The central opioid system in liver disease and its complications. Metab Brain Dis 2001;16:79–83. Ozsoylu S, Kocak N. Naloxone in hepatic encephalopathy. Am J Dis Child 1985;139:749–750. Bergasa NV, Thomas DA, Vergalla J, et al. Plasma from patients with the pruritus of cholestasis induces opioid receptormediated scratching in monkeys. Life Sci 1993:53:1253–1257. Bergasa NV, Alling DW, Talbot TL, et al. Effects of naloxone infusions in patients with the pruritus of cholestasis. A doubleblind, randomized, controlled trial. Ann Intern Med 1995;123:161–167. Ahbouche S, Butterworth RF. Pathophysiology of hepatic encephalopathy: a new look at GABA from the molecular standpoint. Metab Brain Dis 2004;19:331–343. Ahbouche S, Pomier-Layrargues C, Butterworth RF. Increased brain concentration of endogenous (non-benzodiazepine) GABAA receptor ligands in human hepatic encephalopathy. Metab Brain Dis 2004;19:241–251. Butterworth RK. Neurotransmitter dysfunction in hepatic encephalopathy. New approaches and new findings. Metab Brain Dis 2001;16-55–65. Basile AS, Jones EA, Skolnick P. The pathogenesis and treatment of hepatic encephalopathy: Evidence for the involvement of benzodiazepine receptor ligands. Pharmacol Rev 1991;43:27–71. Conn H, Lieberthal M. The hepatic coma syndromes and lactulose. Baltimore: Williams & Wilkins, 1979. Cordoba J, Gottstein J, Blei AT. Chronic hyponatremia exacerbates ammonia-induced brain edema in rats after portacaval anastomosis. J Hepatol 1998;29:589–594. van Berlo CL, van de Bogaard AE, van der Heijden MA, et al. Is increased ammonia liberation after bleeding in the digestive tract the consequence of complete absence of isoleucine in hemoglobin? A study in pigs. Hepatology 1989;10:315–323. Olde Damink SWM, Jalan R, Dejong CHC, et al. Metabolic consequences of an upper gastrointestinal bleed in patients with cirrhosis. In: Jones EA, Meijer AJ, Chamuleau RAFM, eds.
Chapter 18 HEPATIC ENCEPHALOPATHY
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88. 89. 90.
91.
92.
93.
94.
95.
96. 97.
98.
Encephalopathy and nitrogen metabolism in liver failure. Dordrecht: Kluwer Academic, 2003: 21–29. Gentile S, Guarino G, Romano M, et al. A randomized controlled trial of acarbose in hepatic encephalopathy. Clin Gastroenterol Hepatol 2005;3:184–191. Spahr L, Vingerhoets F, Lazeyras F, et al. Magnetic resonance imaging and proton spectroscopic alterations correlate with parkinsonian signs in patients with cirrhosis. Gastroenterology 2000;119:774–781. Jover R, Company L, Gutierrez A, et al. Minimal hepatic encephalopathy and extrapyramidal signs in patients with cirrhosis. Am J Gastroenterol 2003;98:1599–1604. Parsons-Smith BG, Summerskill WHJ, Dawson AM, Sherlock S. The electroencephalogram in liver disease. Lancet 1957;2867–2871. Vaquero J, Oliver D, Hassanein T, et al. Evaluation of mental state in a clinical trial of MARS for patients with acute hepatic encephalopathy: Comparison of two scales. In: Jones EA, Meijer AJ, Chamuleau RAFM, eds. Encephalopathy and nitrogen metabolism in liver failure. Dordrecht: Kluwer Academic, 2003: 55–69. Donovan JP, Schafer DF, Shaw BW Jr, Sorrell MF. Cerebral oedema and increased intracranial pressure in chronic liver disease. Lancet 19987;351:719–721. Candranel JF, Lebietz E, DiMartino V, et al. Focal neurological signs in hepatic encephalopathy in cirrhotic patients: underestimated entity? Am J Gastroenterol 2001;96:515–518. Mullen KD, Dasarathy S. Hepatic encephalopathy. In: Schiff ER, Sorrell MF, Maddrey WC, eds. Schiff ’s Diseases of the liver, 8th edn. Philadelphia: Lippincott Raven, 1999: 545–581. Weissenborn K, Scholz M, Hinrichs H, et al. Neurophysiological assessment of early hepatic encephalopathy. Electroencephalogr Clin Neurophysiol 1990;75:289–295. Van der Rijt CC, Schalm SW, deGroot AH, deKlieger M. Objective measurement of hepatic encephalopathy by means of automated EEG analysis. Electroencephalogr Clin Neurophysiol 1989;57:423–426. Brenner RP. The electroencephalogram in altered states of consciousness. Neurol Clin 1985;3:615–631. Bickford RG, Butt HR. Hepatic coma: the electroencephalographic pattern, J Clin Invest 1955;34:790–799. Nolano M, Guardascione MA, Amitrano L, et al. Cortico-spinal pathways and inhibitory mechanisms in hepatic encephalopathy. Electroencephalogr Clin Neurophysiol 1997;105:72–78. Sanford SL, Tarter RE, Sclabassi R, Van Theil DH. Sensory information processing in patients with non-alcoholic cirrhosis: short latency, visual, auditory, and somatosensory event related potentials. J Neurol Sci 1987:80;269–276. Kullman F, Hollerbach S, Holstege A, Scholmerich J. Subclinical hepatic encephalopathy: the diagnostic value of evoked potentials. J Hepatol 1995;22:101–110. Cordoba J, Raguer N, Flavia M, et al. T2 hyperintensity along the cortico-spinal tract in cirrhosis relates to functional abnormalities. Hepatology 2003;38:1026–1033. Kircheis G, Wettstein M, Timmermann L, et al. Critical flicker frequency for quantification of low-grade hepatic encephalopathy. Hepatology 2002;35:357–366. Bakti G, Fisch HU, Karlaganis G, et al. Mechanism of the excessive sedative response of cirrhotics to benzodiazepines: model experiments with triazolam. Hepatology 1987;7:629–638. Morgan MY. Non-invasive neuro investigation in liver disease. Sem Liv Dis 1996;16:293–314. Brunberg JA, Kanal E, Hirsch W, Van Thiel DH. Chronic acquired hepatic failure: MR imaging of the brain at 1.5T. Am J Neuroradiol 1991;12;909–914. Inoue E, Hori S, Narumi Y, et al. Portal–systemic encephalopathy: presence of basal ganglia lesions with high signal intensity on MR images. Radiology 1991;179:551–555.
99. Pujol A, Pujol J, Graus F, et al. Hyperintense globus pallidus on T1-weighted MRI in cirrhotic patients is associated with severity of liver failure. Neurology 1993;43:65–69. 100. Syh HW, Chu WK, Mar N, McConnell JR. An image analysis on MR imaging of the brain for hepatic encephalopathy. Biomed Sci Instrum 1991;27:29–33. 101. Kulisevsky J, Pujol J, Balanzo J, et al. Pallidal hyperintensity on magnetic resonance imaging in cirrhotic patients: clinical correlations. Hepatology 1992;16:1382–1388. 102. Norton NS, McConnell JR, Zetterman RK, Rodriguez-Sierra JF. A quantitative evaluation of magnetic resonance image signal changes of the brain in chronic hepatic encephalopathy. J Hepatol 1994;21:764–770. 103. Taylor-Robinson SD, Oatridge A, Hajnal JV, et al. MR imaging of the basal ganglia in chronic liver disease: correlation of T1weighted and magnetization transfer contrast measurements with liver dysfunction and neuropsychiatric status. Metab Brain Dis 1995;10:175–188. 104. Thuluvath PJ, Edwin D, Yue NC, et al. Increased signals seen in globus pallidus in T1-weighted magnetic resonance imaging in cirrhotics are not suggestive of chronic hepatic encephalopathy. Hepatology 1996;24:282–283. 105. Bernthal P, Hays A, Tarter RE, et al. Cerebral CT scan abnormalities in cholestatic and hepatocellular disease and their relationship to neuropsychologic test performance. Hepatology 1987;7:107–114. 106. Kanamori K, Blumi S, Ross B. Magnetic resonance spectroscopy in the study of hyperammonemia and hepatic encephalopathy. In: Felipo V, Grisolia S, eds. Advances in cirrhosis, hyperammonemia and hepatic encephalopathy. New York: Plenum Press, 1997: 185–194. 107. Ross BD, Danielson ER, Blumi S. Proton magnetic resonance spectroscopy: the new gold standard for diagnosis of clinical and subclinical hepatic encephalopathy. Dig Dis Sci 1996;4(Suppl 1):30–39. 108. Lockwood AH, Yap EWH, Rhoades HM, Wong WH. Altered cerebral blood flow and glucose metabolism in patients with liver disease and minimal encephalopathy. J Cerebr Blood Flow Metab 1991;11:331–336. 109. O’Carroll RE, Hayes PC, Ebmeier KP, et al. Regional cerebral blood flow and cognitive function in patients with chronic liver disease. Lancet 1991;337:1250–1253. 110. Trzepacz PT, Tarter RE, Shah A, et al. SPECT scan and cognitive findings in subclinical hepatic encephalopathy. J Neuropsychiatry Clin Neurosci 1994;6:170–175. 111. Kawatoko T, Murai K, Ibayashi S, et al. Marked reduction of cerebral oxygen metabolism in patients with advanced cirrhosis: a positron emission tomography study. Fukuoka Igaku Zasshi 1992;83:33–42. 112. Burra P, Pizzolato G, Orlando F, et al. Single-photon emission computed tomography with 99mTChexamethylpropyleneamineoxide in cirrhotic patients before and after liver transplantation. Transplant Proc 1994;26:3677–3678. 113. Lockwood AH, Murphy BW, Donnelly KZ, et al. Positronemission tomographic localization of abnormalities of brain metabolism in patients with minimal hepatic encephalopathy. Hepatology 1993;18:1061–1068. 114. Lockwood AH, Yap EWH, Wong W-H. Cerebral ammonia metabolism in patients with severe liver disease and minimal hepatic encephalopathy. J Cerebr Blood Flow Metab 1991;11:337–341. 115. Ong JP, Dasarathy S, Aggarwal A, et al. Arterial ammonia levels in hepatic encephalopathy: Usefulness in diagnosis. In: Yurdaydin C, Bozkayo H, eds. Advanced hepatic encephalopathy and metabolism in liver disease. Ankara: Ankara University Press, 2000: 397–402. 116. Huizerga JR, Tangerman A, Gips CH. Determination in biological fluids. Ann Clin Biochem 1996;33:23–30.
327
Section III. Clinical Consequences of Liver Disease
117. van Caulaert C, Deviller C, Hofstein J. Epreuve de l’ammoniémie provoqure. Compt Rend Soc Biol 1932;111:737–788. 118. Kirk E. Amino acid and ammonia metabolism in liver disease. Acta Med Scand 1936;77(Suppl):1–147. 119. Farquhar JD, Stokes J Jr, Whitlock CM Jr, et al. Studies on the use of aureomycin in hepatic disease. III. A note on aureomycin therapy in hepatic coma. Am J Med Sci 1950;220:166–172. 120. Sherlock S, Summerskill WHJ, Dawson AM. The treatment and prognosis of hepatic coma. Lancet 1956;2:689–694. 121. Fisher CJ, Faloon WW. Blood ammonia levels in hepatic cirrhosis: their control by the oral administration of neomycin. N Engl J Med 1957;257:1030–1055. 122. Dawson AM, McLaren J, Sherlock S. Neomycin in the treatment of hepatic coma? Lancet 1957;273:1263–1268. 123. Bircher J, Muller J, Guggenheim P, Hammerli UP. Treatment of chronic portal–systemic encephalopathy with lactulose. Lancet 1966;1:890–892. 124. Strauss E, Tramote R, Silva EP, et al. Double-blind randomized clinical trial comparing neomycin and placebo in the treatment of exogenous hepatic encephalopathy. Hepatogastroenterology 1992;39:542–545. 125. Blanc P, Daures JP, Liautard J, et al. Lactulose–neomycin combination versus placebo in the treatment of acute hepatic encephalopathy. Results of a randomized controlled trial. Gastroenterol Clin Biol 1994;18:1063–1068. 126. Orlandi F, Freddara U, Candelaresi MT, et al. Comparison between neomycin and lactulose in 173 patients with hepatic encephalopathy: a randomized clinical study. Dig Dis Sci 1981;26:498–506. 127. Atterbury CE, Maddrey WC, Conn Ho. Neomycin–sorbitol and lactulose in the treatment of acute portal–systemic encephalopathy. A controlled, double-blind clinical trial. Am J Dig Dis 1978;23:398–406. 128. Conn HO, Leevy CM, Vlacevic ZR, et al. Comparison of lactulose and neomycin in the treatment of chronic portal–systemic encephalopathy. A double-blind controlled trial. Gastroenterology 1977;72573–583. 129. Di Piazza S, Gabriella Filippazzo M, Valenza LM, et al. Rifaximin versus neomycin in the treatment of portosystemic encephalopathy. Ital J Gastroenterol 1991;23:403–407. 130. Parini P, Cippola A, Ponchi M, et al. Effect of rifaximin and paromomycin in the treatment of portal systemic encephalopathy. Curr Ther Res 1992;52:34–39. 131. Festi D, Mazzella G, Orsini M. Rifaximin in the treatment of chronic hepatic encephalopathy: results of multi-center study of efficacy and safety. Curr Ther Res 1993;54:598–609. 132. Simmons F, Goldstein H, Boyle JD. A controlled clinical trial of lactulose in hepatic encephalopathy. Gastroenterology 1970;59:827–832. 133. Germain L, Frexinos J, Louis A, Ribet A. A double blind study of lactulose in 18 patients with chronic hepatic encephalopathy after portocaval shunt. Arch Fr Mal App Dig 1973;62:293–302. 134. Gertman PM, Gagnon O, Iber FL. Controlled diarrhea in the treatment of cirrhosis. JAMA 1966;197;257–260. 135. McClain CJ, Kromhout JP, Zieve L, Duane WC. Effect of sorbitol on psychomotor function: its use in alcoholic cirrhosis. Arch Intern Med 1981;141:901–903. 136. Als-Nielsen B, Gluud LL, Gluud C. Nonabsorbable disaccharides for hepatic encephalopathy (Cochrane Review). In: The Cochrane Library, Issue 2, 204. Chichester: John Wiley & Sons. 137. Als-Nielsen B, Gluud LL, Gluud C. Non-absorbable disaccharides for hepatic encephalopathy: systemic review of randomized trials. Br Med J 2004;328:1046. 138. Als-Nielsen B. Gluud LL, Gluud C. Benzodiazepine receptor antagonists for hepatic encephalopathy (Cochrane Review). In: The Cochrane Library, Issue 2, 2004. Chichester: John Wiley & Sons.
328
139. Als-Nielsen B, Koretz RL, Kjaergard LL, Gluud C. Branchedchain amino acids for hepatic encephalopathy (Cochrane Review). In: The Cochrane Library, Issue 2, 2003. Chichester: John Wiley & Sons. 140. Als-Nielsen B, Gluud LL, Gluud C. Dopaminergic agents for hepatic encephalopathy (Cochrane Review). In: The Cochrane Library, Issue 4, 2004. Chichester: John Wiley & Sons. 141. Liu JP, Gluud LL, Als-Nielsen B, Gluud C. Artificial and bioartificial support systems for liver failure. In: The Cochrane Library, Issue 1, 2004. Chichester: John Wiley & Sons. 142. Kircheis G, Nilius R, Held C, et al. Therapeutic efficacy of Lornithine-L-aspartate infusions in patients with cirrhosis and hepatic encephalopathy: results of a placebo-controlled, doubleblind study. Hepatology 1997;25:1351–1360. 143. Kiba T, Numata K, Saito S. Neostigmine and polyethylene glycol solution for the therapy of acute hepatic encephalopathy with liver cirrhosis and ascites. Hepatogastroenterology 2002;50:823–826. 144. Van Thiel DH, Fagiuoli S, Wright HI, et al. Gastrointestinal transit in cirrhotic patients: Effect of hepatic encephalopathy and its treatment. Hepatology 1994;17:67–71. 145. Bouin M, Vincent C, Bouhier K, et al. Increased oro-cecal transit time in grade I or II hepatic encephalopathy. Gastroenterol Clin Biol 20;28:1240–1244. 146. Nelson DC, McGrew WERG Jr, Holimpa AM Jr. Hypernatremia and lactulose therapy. JAMA 1983;249:1295–1298. 147. Elkington SG, Floch MH, Conn HO. Lactulose in the treatment of chronic portal–systemic encephalopathy. A double-blind clinical trial. N Engl J Med 1969;281:408–412. 148. Rodgers JB Jr., Kiley JE, Balint JA. Comparison of results of long-term treatment of chronic hepatic encephalopathy with lactulose and sorbitol. Am J Gastroenterol 1973;60:459–465. 149. Uribe M, Campollo O, Vargas F, et al. Acidifying enemas (lactitol and lactose) vs. nonacidifying enemas (tap water) to treat acute portal–systemic encephalopathy: a double-blind, randomized clinical trial. Hepatology 1987;7:639–643. 150. Watanabe A, Sakai T, Sato S, et al. Clinical efficacy of lactulose in cirrhotic patients with and without subclinical hepatic encephalopathy. Hepatology 1997;26:1410–1414. 151. Shi H, Liu HY, Fu Z, et al. Lactitol in treatment of subclinical hepatic encephalopathy: a double blind placebo-controlled randomized trial (Chinese). Chin J Dig 1997;17:221–223. 152. Li Z, Zhang H, Hong Y, Clinical effect of lactulose in the treatment of subclinical hepatic encephalopathy. Chin J Integr Trad West Med Liver Dis 1999;9:13–15. 153. Dhiman RK, Sawhney MS, Chawla YK, et al. Efficacy of lactulose in cirrhotic patients with subclinical hepatic encephalopathy. Dig Dis Sci 2000;45:1549–1552. 154. Russo M, Galanti B, Nardiello S, et al. Ribostamycin for the treatment of hepatic encephalopathy: A crossover study with lactulose. Curr Ther Res Clin Exp 1989;45:133–141. 155. Orlandi F, Brunelli E, Benedetti A, Macani A. Clinical trials of non-absorbable disaccharide therapy, in hepatic encephalopathy. In: Conn HO, Bircher J, eds. Hepatic encephalopathy: syndromes and treatment. Bloomington IL: Medi Ed Press, 1994: 209–212. 156. Uribe M. Management of portal systemic encephalopathy with lactose in patients with lactose deficiency. In: Conn HO, Bircher J, eds. Hepatic encephalopathy: syndromes and therapies. Bloomington, IL: Medi Ed Press, 1994: 311–327. 157. Bircher J, Ulrich D. Clinical pharmacology of lactulose, lactitol and related disaccharides. In: Conn HO, Bircher J, eds. Hepatic encephalopathy: syndrome and therapies. Bloomington IL: Medi Ed Press, 1994: 195–208. 158. Conn HO, Floch MH. Effects of lactulose and Lactobacillus acidophilus on the fecal flora. Am J Clin Nutr 1970;23:1588–1594.
Chapter 18 HEPATIC ENCEPHALOPATHY
159. Price JB, Sawoda M, Voorhees AB. Clinical significance of intraluminal pH in intestinal ammonia transport. Am J Surg 1970;119:595–598. 160. Fessel JM, Conn HO. Lactulose in the treatment of acute hepatic encephalopathy. Am J Med Sci 1973;266:103–110. 161. Weber FL Jr. The effect of lactulose on urea metabolism and nitrogen excretion in cirrhotic patients. Gastroenterology 1979;77:118–123. 162. Vince A, Killingsley M, Wrong OM. Effect of lactulose on ammonia production in a fecal incubation system. Gastroenterology 1978;76:544–549. 163. Mortensen PB. The effect of orally administered lactulose on colonic nitrogen metabolism and excretion. Hepatology 1992;16:1350–1356. 164. Van Leeuwen PAM, Van Berlo CL, Soeters PB. New mode of action for lactulose. Lancet 1988;1:55–56. 165. Romero-Gomez M, Ramos-Guerrero R, Grande L, et al. Intestinal glutaminase activity is increased in liver cirrhosis and correlates with minimal hepatic encephalopathy. J Hepatol 2004;41:49–54. 166. Sanaka MR, Ong JF, Mullen KD. Challenges of designing hepatic encephalopathy treatment trials. Hepatology 2003;38:527–529. 167. Blei AT, Cordoba J. Hepatic encephalopathy. Am J Gastroenterol 2001;96:1968–1976. 168. Silen W, Harper H, Mawdsley Dl, Weirich Wl. Effect of antibacterial agents on ammonia production within the intestine. Proc Soc Exp Biol Med 1955;88:138–140. 169. Dobbins WO, Herrero BA, Mansback CM. Morphologic alterations associated with neomycin induced malabsorption. Am J Med Sci 1968;255:63–77. 170. Jacobson ED, Prior JT, Faloon WW. Malabsorptive syndrome induced by neomycin. Morphologic alterations with jejunal mucosa. J Lab Clin Med 1960;50:245–250. 171. Weber FLJ, Veach GL. The importance of the small intestine in gut ammonia production in the fasted dog. Gastroenterology 1979;77:235–240. 172. Gunnarsdottir SA, Sadik R, Shev S, et al. Small intestinal motility disturbances and bacterial overgrowth in patients with liver cirrhosis and portal hypertension. Am J Gastroenterol 2003;98:1362–1370. 173. Williams R, Bass N. Rifaximin, a nonabsorbed oral antibiotic, in the treatment of hepatic encephalopathy: antimicrobial activity, efficacy and safety. Rev Gastroenterol Disord 2005:5(Suppl):10–18. 174. Zeneroli ML, Avallone R, Corsi L, et al. Management of hepatic encephalopathy: role of rifaximin. Chemotherapy 2005;51(Suppl):90–95. 175. Miglio F, Valpiani D, Rosselini SR, Ferrini A. Rifaximin, a nonabsorbable rifaximycin for the treatment of hepatic encephalopathy. A double blind randomized trial. Curr Med Res Opin 1997;13:593–601. 176. Bucci L, Palmieri GC. Double-blind, double-dummy comparison between treatment with rigaximin and lactulose in patients with medium to severe degree hepatic encephalopathy. Curr Med Res Opin 1993;12:109–118. 177. Fera G, Agostinacchio F, Nigro M, et al. Rifaximin in the treatment of hepatic encephalopathy. Eur J Clin Res 1993;4:57–66. 178. Massa P, Vallerino E, Dodero M. Treatment of hepatic encephalopathy with rifaximin: double-blind, double dummy study versus lactulose. Eur J Clin Res 1993;4:7–18. 179. Song H, Lee KS, Kim MH, et al. The clinical efficacy of rifaximin in the treatment of hepatic encephalopathy (comparison with lactulose) [abstract]. Hepatology 2000;32:407. 180. Loguercio C, Federico A, De Girolamo V, et al. Cyclic treatment of chronic hepatic encephalopathy with rifaximin. Results of a double-blind clinical study. Minerva Gastroenterol Dietol 2003;49:53–62.
181. Pedretti G, Calzetti C, Missale C, Fiackadon F. Rifaximin versus neomycin on hyperammonemia in chronic portal systemic encephalopathy of cirrhosis: a double-blind randomized trial. Ital J Gastroenterol 1991;23:175–178. 182. Williams R, James OE, Warners TW, Morgan MY. Evaluation of the efficacy and safety of rifaximin in the treatment of hepatic encephalopathy: a double-blind, randomized, dose finding multicenter study. Eur J Gastroenterol Hepatol 2000;12:203–208. 183. Puxeddu A, Quartini M, Massimetti A, Ferrieri A. Rifaximin in the treatment of chronic hepatic encephalopathy. Curr Med Res Opin 1995;13:274–281. 184. Mas A, Rodes J, Sunyer L, et al. Comparison of rifaximin and lactitol in the treatment of acute hepatic encephalopathy: results of a randomized, double-blind, double-dummy, controlled clinical trial. J Hepatol 2003;38:51–58. 185. Morgan M, Read AE, Speller PCE. Treatment of hepatic encephalopathy with metronidazole. Gut 1982;23:1–7. 186. Tarao K, Ikeda T, Hayashi K, et al. Successful use of vancomycin hydrochloride in the treatment of lactulose resistant chronic hepatic encephalopathy. Gut 1990;31:702–706. 187. Soulsby CT, Morgan MY. Dietary management of hepatic encephalopathy in cirrhotic patients: survey of current practice in United Kingdom. Br Med J 1999;318:1391. 188. Plauth M, Merli M, Kondrup J, et al. ESPEN guidelines for nutrition in liver disease and transplantation (Consensus statement). Clin Nutr 1997;16:43–55. 189. Plauth M, Merli M, Kondrup J. Management of hepatic encephalopathy. N Engl J Med 1997;337:1921–1922. 190. Seymour CA, Whelan K, Dietary management of hepatic encephalopathy. Br Med J 1999;318:1364–1365. 191. Cordoba J, Lopez-Hellin J, Planas M, et al. Normal protein diet for episodic hepatic encephalopathy: results of a randomized study. Hepatology 2004;41:38–43. 192. Erikkson LS, Conn HO. Branched-chain amino acids in hepatic encephalopathy. Gastroenterology 1990;99:604–607. 193. Gluud C. Branched-chain amino acids for hepatic encephalopathy? Hepatology 1991;13:812–813. 194. Naylor CD. Branched-chain amino acids in hepatic encephalopathy. Continuing controversy (letter; comment). Int J Technol Assess Health Care 1991;7:648–650. 195. Ferenci P. Branched-chain amino acids in hepatic encephalopathy. Gastroenterology 1990;98:1395–1396. 196. Erikkson LS, Conn HO. Branched-chain amino acids in the management of hepatic encephalopathy an analysis of variants. Hepatology 1989;10:228–246. 197. Naylor CD, O’Rourke K, Detsky AS, Baker JP. Parenteral nutrition with branched-chain amino acids in hepatic encephalopathy. A meta-analysis. Gastroenterology 1989;97:1033–1042. 198. Shaw S, Worner TM, Lieber CS. Comparison of animal and vegetable protein sources in the dietary management of hepatic encephalopathy. Am J Clin Nutr 1983;38:59–63. 199. Keshavarzian A, Meek J, Sutton C, et al. Dietary protein supplementation from vegetable sources in the management of chronic portal systemic encephalopathy. Am J Gastroenterol 1984;79:945–949. 200. Uribe M, Dibildox M, Malpica S, et al. Beneficial effect of vegetable protein diet supplemented with psyllium plantago in patients with hepatic encephalopathy and diabetes mellitus. Gastroenterology 1985;88:901–907. 201. Herman R, Shakoor T, Weber FL. Beneficial effect of pectin in chronic hepatic encephalopathy. Gastroenterology 1987;92:1795. 202. Garcia-Compean D, Uribe M. Fiber rather than protein determines tolerance to nitrogen load in chronic portal systemic encephalopathy: a randomized trial. Hepatology 1987;7:1034. [abstract]. 203. Uribe M, Conn HO. Dietary management of portal–systemic encephalopathy. In: Conn HO, Bircher J, eds. Hepatic
329
Section III. Clinical Consequences of Liver Disease
204.
205.
206.
207.
208. 209.
210.
211.
212. 213.
214.
215.
216.
217.
218.
219.
220.
221.
222.
223.
330
encephalopathy: Syndromes and therapy. Bloomington, IL: Medi-Ed Press, 1994: 331–349. Marchesini G, Dioguardi FS, Bianchi GP, et al. Long-term oral branched-chain amino acid treatment in chronic hepatic encephalopathy. A randomized double-blind casein-controlled trial. The Italian Multicenter Study Group. J Hepatol 1990;11:92–1001. Egberts EH, Schomerus H, Hamster W, Jurgens P. Branchedchain amino acids in the treatment of latent portosystemic encephalopathy. A double blind placebo controlled crossover study. Gastroenterology 1985;88:887–895. Horst D, Grace ND, Conn HO, et al. Comparison of dietary protein with an oral, branched chain enriched amino acid supplement in chronic portal–systemic encephalopathy. Hepatology 1984;4:279–287. Uribe M, Moran S, Poo JL, et al. Beneficial effect of carbohydrate maldigestion induced by a disaccharidase in inhibitor (AO-128) in the treatment of chronic portal–systemic encephalopathy. A double-blind randomized controlled trial. Scand J Gastroenterol 1998;33:1099–1106. Solga SF. Probiotics can treat hepatic encephalopathy. Med Hypoth 2003;61:307–313. Bongaerts G, Severijinam R, Timmerman H. Effect of antibiotics, prebiotics and probiotics in treatment for hepatic encephalopathy. Med Hypoth 2005;64:64–68. Jic L, Zhang MH. Comparison of probiotics and lactulose in the treatment of hepatic encephalopathy in rats. World J Gastroenterol 2005;14:908–914. Liu Q, Duan ZP, Ha da K, et al. Symbiotic modulation of gut flora on minimal hepatic encephalopathy in patients with cirrhosis. Hepatology 2004;39:1441–1449. Solga SF, Diehl AM. Gut flora based therapy in liver disease? The liver cares about the gut. Hepatology 2004;39:199–200. Macbeth WA, Kass EH, McDermott WV Jr. Treatment of hepatic encephalopathy by alteration of intestinal flora with lactobacillus acidophilus. Lancet 1965;Vol 1, 399–403. Reed AE, McCarthy CF, Heaton KW, Laidlaw J. Lactobacillus acidophilus in treatment of hepatic encephalopathy. Br Med J 1966;1:1267–1269. Loguerico C, Del Vecchio Blanco C, Coltarti M. Enterococcus lactic acid bacterial strain SF 68 and lactulose in hepatic encephalopathy: A controlled trial. J Int Med Res 1987;15:335–343. Loguerico C, Abbiati R, Ranaldi M, et al. Long-term efficacy of enterococcus fecium SF 68 versus lactulose in the treatment of patients with cirrhosis. J Hepatol 1995;23:39–46. Kircheis G, Quack G, Erbler H. L-ornithine-L-aspartate in the treatment of hyperammonemia and hepatic encephalopathy. In: Conn HO, Bircher J, eds. Hepatic encephalopathy: Syndromes and therapies. Bloomington, IL: Medi-Ed Press, 1994: 373–383. Kircheis G, Wettstein M, Dahl S, Haussinger D. Clinical efficacy of L-ornithine-L-aspartate in the management of hepatic encephalopathy. Metab Brain Dis 2002;17:453–462. Stauch S, Kircheis G, Adler F, et al. Oral L–ornithine-L-aspartate therapy of chronic hepatic encephalopathy: results of a placebocontrolled double-blind study. J Hepatol 1998;28:856–864. Delcker AM, Jalan R, Schumacher M, Comes G. L-Ornithine-Laspartate vs placebo in the treatment of hepatic encephalopathy: a meta-analysis of randomized placebo-controlled trials using individual data. Hepatology 2000;32:310A [abstract]. Mendenhall CL, Rouster S, Marshall L, Weisner R. A new therapy for portal systemic encephalopathy. Am J Gastroenterol 1986;82:540–543. Sushma S, Dasarathy S, Tandon RK, et al. Sodium benzoate in the treatment of acute encephalopathy: a double blind randomized trial. Hepatology 1982;16:138–144. Mullen KD, Martin JV, Mendelson WB, et al. Evidence for the presence of benzodiazepines receptor binding substances in
224.
225.
226.
227.
228.
229.
230.
231.
232. 233.
234. 235.
236.
237.
238.
239. 240.
241.
242.
243.
244.
cerebrospinal fluid of a rabbit model of hepatic encephalopathy. Metab Brain Dis 1989;4:253–260. Basile AS, Parnell L, Jasuri T, et al. Brain concentrations of benzodiazepines are elevated in an animal model of hepatic encephalopathy. Proc Natl Acad Sci USA 1990;87:5263–5267. Mullen KD, Szauter KM, Kaminsky Russ K. Endogenous benzodiazepine activity in physiological fluids of patients with hepatic encephalopathy. Lancet 1990;336:81–83. Basile AS, Hughes RD, Harrison PM, et al. Elevated brain concentrations of 1:4-benzodiazepines in fulminant hepatic failure. N Engl J Med 1991;325;473–478. Yurdaydin C, Li Y, Ha JH, et al. Brain and plasma levels of opioid peptides are altered in rats with thioacetamide-induced liver failure: implications for the treatment of hepatic encephalopathy. J Pharmacol Exp Ther 1995;273:185–192. Yurdaydin C, Karavelioglu D, Onaran O, et al. Opioid receptor ligands in human hepatic encephalopathy. J Hepatol 199;29:796–801. Reding P, Duchateau J, Bataile C. Oral zinc supplements improve hepatic encephalopathy. Results of a randomized controlled trial. Lancet 1984;2:493–495. Riggio O, Oriosto F, Merli M, et al. Short-term oral zinc supplementation does not improve chronic hepatic encephalopathy: results of a double-blind crossover trial. Dig Dis Sci 1991;36:1204–1208. Riggio O, Merli M, Capocaccia L, et al. Zinc supplementation reduces blood ammonia and increased liver transcarbamylase activity in experimental cirrhosis. Hepatology 1992;16:785–788. Keeling PW, Jones RB, Hilton PJ, Thompson RH. Reduced leucocyte zinc in liver disease. Gut 1980;21:561–564. Van der Rijt CCD, Scholm SW, Wchat H, et al. Overt hepatic encephalopathy precipitated by zinc deficiency. Gastroenterology 1991;100:1114–1118. Mullen KD, Dasarathy S. MARS – Does it stand the test of time? Metab Brain Dis 2004;19:223–228. Mitzner SR, Klammt S, Peszynski P, et al. Improvement of multiple organ functions in hepatorenal syndrome during albumin dialysis with the molecular adsorbent recirculating system. Ther Apher 2001;5:417–422. O’Grady JG, Gimson AE, O’Brien CJ, et al. Controlled trials of charcoal hemoperfusion and prognostic factors in fulminant hepatic failure. Gastroenterology 1988;94:1186–1192. Hughes RD, Pucknell A, Routley D, et al. Evaluation of the Biologic-DT sorbent–suspension dialyser in patients with fulminant hepatic failure. Int J Artif Org 2001;24:434–442. Ellis AJ, Hughes RD, Nicholl D, et al. Temporary extracorporeal liver support for severe acute alcoholic hepatitis using the BioLogic-DT. Int J Artif Org 1999;22:27–34. Wilkinson AH, Ash SR, Nissenson AR. Hemodiabsorption in treatment of hepatic failure. J Transpl Coord 1998;8:43–50. He JQ, Chen CY Deng JTT, et al. Clinical study on the treatment of fatal hepatitis with artificial liver support system. Chin Crit Care Med 2000;12:105–108. Ellis AJ, Hughes RD, Wendon JA, et al. Pilot-controlled trial of the extracorporeal liver assist device in acute liver failure. Hepatology 1996;24:1446–1451. Stevens C, Busuttil R, Han S, et al. An interim analysis of a phase II/III prospective randomized multicenter controlled trial of the Hepat Assist Bioartificial Liver Support System for the treatment of fulminant hepatic failure. Hepatology 2001;4 [abstract]. Mitzner SR, Stange J, Klammt S, et al. Improvement of hepatorenal syndrome with extracorporeal albumin dialysis MARS: results of a prospective, randomized, controlled clinical trial. Liver Transpl 2000;6:287–289. Kramer L, Gendo A, Madl C, et al. A controlled study of sorbent suspension dialysis in chronic liver disease and hepatic encephalopathy. Int J Artif Org 2001;24:434–442.
Chapter 18 HEPATIC ENCEPHALOPATHY
245. Sen S, Jalan R, Williams R. Extracorporeal albumin dialysis in acute-on-chronic liver failure: will it stand the test of time? Hepatology 2002;36:1014–1016. 246. Liu JP, Gluud LL, Als-Nielsen B, Gluud C. Artificial and bioartificial support systems for liver failure. In: The Cochrane Library, Issue 1, 2004. Chichester: John Wiley & Sons. 247. Kjaergard LL, Liu J, Als-Nielsen B, Gluud C. Artificial and bioartificial support systems for acute and acute-on-chronic liver failure: a systematic review. JAMA 2003;289:217– 222. 248. Clarke B, Ellis MJ, Leung V, et al. Reversal of hepatic encephalopathy and alteration in amino acid profiles after blocking a surgical splenorenal shunt by interventional radiological techniques. J Hepatol 1989;8:325–329. 249. Hiroka A, Kubose K, Hamada M. Hepatic encephalopathy due to intrahepatic portosystemic venous shunt successfully treated by interventional radiology. Intern Med 2004;44:212–216.
250. Sachdev A, Duseja A. Decompressive shunts and hepatic encephalopathy. Indian J Gastroenterol 2003;22(Suppl 2): S21–S24. 251. Ong JP, Mullen KD. Hepatic encephalopathy. Eur J Gastroenterol Hepatol 2001;13:325–334. 252. Gerbes AL, Waggershauser T, Holl J, et al. Experiences with novel techniques for reduction of start flow in transjugular intrahepatic portosystemic shunts. Z Gastroenterol 1998;36:373–377. 253. Rossle M, Siegerstetter V, Huber M, Ochs A. The first decade of the transjugular intrahepatic portosystemic shunt (TIPS): State of the art. Liver 1998;1:73–89. 254. Nishie A, Yoshimitsu K, Honda H. Treatment of hepatic encephalopathy by retrograde and transcaval coil embolization of an ilea vein-to-right gonadal vein portosystemic shunt Cardiovasc Intervent Radiol 1997;20:222–224. 255. Mullen KD. Hepatic encephalopathy after portosystemic shunts: Any clues from TIPS. Am J Gasterol 1995;70:531–533.
331
Section III: Clinical Consequences of Liver Disease
19
ASCITES Guadalupe Garcia-Tsao Abbreviations AFB acid fast bacilli Catn catheterization COX-2 cyclooxygenase-2 CT computed tomography FHVP free hepatic vein pressure HVPG hepatic venons pressure gradient IV intravenous
LVP NO PCD PRA PTFE PVS
large-volume paracentesis nitric oxide post-paracentesis circulatory dysfunction plasma renin activity polytetrafluoroethylene peritoneovenous shunt
INTRODUCTION Ascites is the accumulation of fluid in the peritoneal cavity and its most common cause is cirrhosis. That fluid accumulates in the abdominal cavity has been known since ancient times, and it was Hippocrates who recognized that ascites (from the Greek askos, meaning a leather bag used to carry wine, water or oil) derived from a diseased liver and that it carried a grim prognosis.1 The development of ascites is one of the complications that marks the transition from compensated to decompensated cirrhosis. Other complications that mark this transition are variceal hemorrhage, hepatic encephalopathy or jaundice; however, ascites is the most common.2,3
EPIDEMIOLOGY At the time of diagnosis, ascites is present in 20–60% of patients with cirrhosis, depending on the referral pattern.2,4 In prospective studies of patients with compensated cirrhosis of all etiologies, the cumulative probability of developing ascites ranges from 35 to 50% in 5 years (Figure 19-1),2,5 not unlike the annual incidence rate of around 6% in patients with viral cirrhosis.6 In a large cohort of patients with decompensated HCV-related cirrhosis, ascites was the most frequent first decompensating event, occurring in 48% of cases.3
PATHOGENESIS In cirrhosis, leakage of ascites into the peritoneal space occurs as a result of sinusoidal hypertension, which in turn results from hepatic venous outflow block secondary to regenerative nodules and fibrosis. The other essential factor in the pathogenesis of cirrhotic ascites is plasma volume expansion, through sodium and water retention, which allows for the replenishment of the intravascular volume, thereby maintaining the formation of ascites (Figure 19-2).
RAAS SAAG SBP TIPS WHVP
renin-angiotensin-aldosterone system serum ascites albumin gradient spontaneous bacterial peritonitis transjugular intrahepatic portosystemic shunt wedged hepatic vein pressure
SINUSOIDAL HYPERTENSION Similar to gastroesophageal varices, in which a minimal portal pressure gradient of 10–12 mmHg is needed for their development, the development of ascites also seems to require a minimal portal pressure gradient of 12 mmHg.7,8 In a recent consensus conference on portal hypertension, a portal pressure gradient of 10 mmHg or more was defined as ‘clinically significant portal hypertension’ because complications of cirrhosis do not occur below this threshold pressure.9
SODIUM RETENTION Sinusoidal hypertension alone is not sufficient to maintain ascites formation. Without replenishment of the intravascular space, leakage of fluid into the peritoneal cavity would be a self-limited process. Replenishment of the intravascular space, that is, plasma volume expansion, is accomplished through sodium retention, which has been shown to precede the development of ascites.10 Splanchnic and systemic arteriolar vasodilatation is the most likely mechanism that leads to sodium retention.11 This vasodilatation results in a reduction of the ‘effective’ arterial blood volume, which in turn leads to stimulation of neurohumoral systems, specifically the renin–angiotensin–aldosterone system (RAAS), the sympathetic nervous system, and the non-osmotic release of antidiuretic hormone. Activation of RAAS and the sympathetic nervous system results in sodium retention and, in extreme cases, renal vasoconstriction. Activation of antidiuretic hormone leads to free water retention and hyponatremia.12 An increased production of the vasodilator nitric oxide (NO) is currently considered the main cause of splanchnic and systemic vasodilatation in cirrhosis. In fact, NO blockade has been shown to raise systemic blood pressure, increase sodium excretion and decrease the volume of ascites in experimental cirrhosis and portal hypertension while down-regulating RAAS activation.13,14 The administration of vasodilators such as prazosin, angiotensin receptor blockers and phosphodiesterase inhibitors to cirrhotic patients leads to further RAAS activation,15–17 with associated sodium retention,15,17 ascites,15 and decreases in creatinine clearance.16
333
Section III. Clinical Consequences of Liver Disease
The presence of normal or low levels of plasma renin activity in some patients with cirrhosis and ascites suggests that in some cases sodium retention occurs unrelated to vasodilatation.
CLINICAL FEATURES The most frequent symptoms associated with the presence of ascites are increased abdominal girth (described as tightness of the belt or garments around the waist) and recent weight gain.18 Ascites induces abdominal distention, but this sign in itself has a poor specificity19 as other conditions, including obesity, gas, tumors, and pregnancy, will also induce abdominal distention. When present in small to moderate amounts, ascites can be identified by bulging flanks, flank dullness and shifting dullness. The last two are the most sensitive
100 0 Ascites
90
Jaundice
80
Encephalopathy GI hemorrhage
xxxxxx
70
50 40 30 20 10 0 20
40
60
DIAGNOSIS Abdominal ultrasonography is the most cost-effective and least invasive method to confirm the presence of ascites and is therefore considered the gold standard in its diagnosis. It can detect amounts as small as 100 ml,22 and even as small as 1–2 ml when Morison’s pouch and the pelvic cul-de-sac are scanned.23 Abdominal ultrasound is also useful in determining the best site to perform a diagnostic or therapeutic paracentesis, particularly in patients with a small amount of ascites or in those with loculated ascites. Additionally, although not very sensitive in the diagnosis of cirrhosis, ultrasound is the most useful initial test to investigate the presence of hepatic vein obstruction, an important and frequently overlooked cause of ascites.23 Therefore, in patients with new onset of ascites abdominal ultrasound should always include Doppler examination of the hepatic veins.
DIFFERENTIAL DIAGNOSIS
60
0
physical maneuvers in the diagnosis of ascites.19,20 Ascites can be classified into three grades:21 1, mild ascites only detectable by ultrasound examination; 2, moderate ascites manifest by moderate symmetrical distention of the abdomen; and 3, large or gross ascites with marked abdominal distention.
80
100
120
140
160
xxxxxx Figure 19-1. Probability of developing decompensated cirrhosis in 257 patients with compensated cirrhosis of different etiologies. Of the complications of portal hypertension, ascites is the most frequent decompensating event.5
Although cirrhosis is the cause of ascites in more than 75% of patients, other causes, such as peritoneal malignancy (12% of cases), cardiac failure (5%) and peritoneal tuberculosis (2%)24 should be considered in the differential diagnosis of ascites. A diagnostic paracentesis should be the first test performed in the diagnostic workup of a patient with ascites. It is a safe procedure with a very low incidence of serious complications, mostly transfusion-requiring hematomas that occur at a rate of 0.2–0.9%.25,26 Renal dysfunction appears to be more predictive of bleeding complications than clotting abnormalities, and therefore coagulopathy is not a contraindication to perform a diagnostic paracentesis.25,26 Care should always be taken to avoid abdominal wall collaterals and to avoid the area of the inferior hypogastric artery, which lies midway between the anterior superior iliac spine and the
Figure 19-2. Pathophysiology of cirrhotic ascites. Two main pathogenic mechanisms in the formation of ascites are sinusoidal hypertension and vasodilatation leading to activation of neurohumoral systems and sodium and water retention.
Cirrhosis
Intrahepatic resistance
Sinusoidal pressure
Ascites
334
Arteriolar resistance (vasodilation)
(HVPG ≥ 10 mmHg)
Sodium and water retention
Effective arterial blood volume
Activation of neurohumoral systems (aldosterone, renin angiotensin, epinephrine)
Chapter 19 ASCITES
three main causes of ascites – cirrhosis, peritoneal pathology (malignancy or tuberculosis) and heart failure – can be distinguished by combining the results of both the SAAG and the ascites total protein content, and the workup of the patient with ascites can thus be further refined (Figure 19-4). In cirrhosis, the SAAG is high and
30
HVPG (mmHg)
pubic tubercle. Technically, the midline below the umbilicus is often recommended as a site for paracentesis because of its presumed avascularity. However, in patients with portal hypertension this area is commonly vascular,27 and therefore the flanks are preferable. Uncomplicated ascitic fluid is transparent, straw colored to slightly yellow. The presence of blood in a non-traumatic tap (in which blood does not clot) may indicate the presence of malignant ascites. Milky fluid is indicative of chylous ascites, and although cirrhosis is the most common cause of non-surgical chylous ascites it represents only 0.5–1% of cases of cirrhotic ascites.28 Ascites total protein and serum ascites albumin gradient (SAAG) are two inexpensive tests that, taken together, are most useful in determining the etiology of ascites and hence in guiding the workup of patients with ascites. A high (>2.5 g/dl) ascites total protein occurs in peritoneal processes (malignancy, tuberculosis) owing to leakage of high-protein mesenteric lymph from obliterated lymphatics and/or from an inflamed peritoneal surface. A high ascites total protein also occurs in cases of postsinusoidal or posthepatic sinusoidal hypertension when sinusoids are normal and protein-rich lymph ‘leaks’ into the peritoneal cavity.29,30 In hepatic cirrhosis an abnormally low protein content of liver lymph has been demonstrated as a result of deposition of fibrous tissue in the sinusoids (‘capillarization of the sinusoid’) that renders the sinusoid less leaky to macromolecules.31,32 On the other hand, the SAAG, which involves measuring the albumin concentration of serum and ascitic fluid specimens and subtracting the ascitic fluid value from the serum value, has been shown to correlate with hepatic sinusoidal pressure (Figure 19-3).33,34 A SAAG cut-off value of 1.1 g/dl has been shown to be the best to distinguish patients in whom ascites is secondary to liver disease and those with malignant ascites.35 Interestingly, this cut-off corresponds to a portal pressure gradient of 11–12 mmHg,33 the threshold pressure necessary for the development of ascites in cirrhotic patients (Figure 19-3). Therefore, the
11
20
10
y = 7.08X + 3.62 0 0.0
Figure 19-3. A direct, very significant correlation between the serum–ascites albumin gradient (SAAG) and the hepatic venous pressure gradient (HVPG), a measure of sinusoidal pressure, is present.33 A cut-off SAAG of >1.1 g/dl has been identified as one that distinguishes ascites secondary to sinusoidal hypertension from that secondary to peritoneal causes. This cut-off corresponds to an HVPG of 11 mmHg, identified by other studies as the threshold pressure for the formation of ascites.
Peritoneum SAAG < 1.1
Normal ‘leaky’ sinusoid Ascites protein > 2.5
Peritoneal lymph Ascites protein > 2.5
Sinusoidal hypertension –Cirrhosis –Late Budd Chiari
Post-sinusoidal hypertension –Congestive heart failure –Constrictive pericarditis –Early Budd Chiari syndrome –Veno-occlusive disease
Peritoneal pathology –Malignancy –Tuberculosis
–Echocardiogram –Right heart catheter and HVPG
3.0
SAAG (g/dl)
‘Capillarized’ sinusoid Ascites protein < 2.5
–Ultrasound or CT scan –Endoscopy
2.0
1.1
Source of ascites
Hepatic sinusoids SAAG > 1.1
1.0
Figure 19-4. Differential diagnosis of ascites depending on the source of ascites. The serum– ascites albumin gradient (SAAG) is high (>1.1 g/dl) when the source is hepatic sinusoids and low when the source is other than the sinusoids. Ascites total protein is high (>2.5 g/dl) when ascites is coming either from normal ‘leaky’ sinusoids or from the peritoneum. Workup is directed accordingly.
–Cytology/AFB –Laparoscopy with peritoneal biopsy
335
Section III. Clinical Consequences of Liver Disease
ascites total protein is low; in posthepatic or postsinusoidal causes of portal hypertension (e.g. heart failure, constrictive pericarditis), SAAG is high and ascites total protein is high; and in ascites secondary to peritoneal causes, SAAG is low and ascites total protein is high (Figure 19-5).36–38 In patients with mixed ascites (e.g. cirrhosis with superimposed peritoneal malignancy) the SAAG is high and the ascites protein is low, that is, the findings of ascites due to cirrhosis predominate.38 Of note, massive hepatic metastasis can lead to the development of ascites, but as the mechanism of ascites formation is sinusoidal hypertension, these cases of ‘malignant ascites’ will have a high SAAG.37 The definitive test to determine whether ascites is the result of sinusoidal hypertension is to actually measure hepatic sinusoidal pressure. The hepatic venous pressure gradient (HVPG), obtained by subtracting the free hepatic vein pressure (FHVP) from the wedged hepatic vein pressure (WHVP), is a measure of sinusoidal pressure. In cases of cirrhotic ascites the HVPG will be ≥10–12 mmHg. In cases of cardiac ascites, both the WHVP and the FHVP will be elevated (reflecting elevated systemic pressures) and hence the HVPG will be normal. In cases of peritoneal ascites (i.e. malignancy or tuberculosis), all hepatic venous pressure measurements (WHVP, FHVP and HVPG) will be normal, unless the patient has coexisting cirrhosis or heart failure. When performed properly, HVPG measurements are reproducible and safe.39 Additionally, hepatic vein catheterization for measurement of hepatic vein pressures allows for the performance, in the same procedure, of a transjugular liver biopsy that will further define the etiology of ascites.
ASSOCIATED CONDITIONS Hyponatremia develops in approximately 30% of cirrhotic patients with ascites and is defined as a serum sodium concentration <130 mEq/l.12 Hyponatremia in cirrhosis is dilutional and results mostly from the non-osmotic secretion of antidiuretic hormone, which in turn is secondary to vasodilatation. Although hyponatremia is usually asymptomatic, some patients may complain of anorexia, nausea and vomiting, lethargy, and occasionally seizures. Hyponatremia has also been associated with a further and important reduction in the already low levels of brain organic osmolytes, particularly myoinositol, suggesting that it may play a role in the pathogenesis of hepatic encephalopathy.40 Umbilical hernias develop in about 20% of cirrhotic patients with ascites (a rate significantly greater than 3% in patients without ascites) and may increase to up to 70% in patients with longstanding recurrent tense ascites.41 The main risks of these hernias are rupture42 and incarceration, a complication that has been observed mostly in patients in whom ascites is resolved after paracentesis, peritoneovenous shunt or transjugular intrahepatic portosystemic shunt.43,44 Hepatic hydrothorax develops in approximately 5–10% of patients with cirrhosis, most probably as a result of the transdiaphragmatic movement of fluid from the peritoneum to the pleural space through diaphragmatic defects.45,46 It usually develops in patients with ascites; however, hepatic hydrothorax may develop
336
in patients without detectable ascites.47 Although large amounts of ascites can accumulate in the peritoneal cavity before resulting in significant patient discomfort, the accumulation of smaller amounts of fluid (1–2 l) in the pleural space results in severe shortness of breath and hypoxemia. Pleural effusion is right-sided in 85%, left-sided in 13%, and bilateral in 2% of cases.48 The diagnosis of hepatic hydrothorax is established by radionuclide scanning of the chest after the intraperitoneal injection of 99mTc-labeled sulfur colloid or macroaggregated serum albumin. In hepatic hydrothorax the presence of radiotracer in the pleural space is demonstrated generally within 2 hours of its intraperitoneal injection.49,50
DISEASE COMPLICATION – SPONTANEOUS BACTERIAL PERITONITIS Spontaneous bacterial peritonitis (SBP) is a potentially lethal complication of ascites. It is an infection of ascites that occurs in the absence of a contiguous source of infection (e.g. intestinal perforation, intraabdominal abscess) and in the absence of an intra-abdominal inflammatory focus (e.g. abscess, acute pancreatitis, cholecystitis). Since its first description in the English-language literature in 1963/1964 as a ‘rarely recognized syndrome’, great strides have been made in its recognition and therapy, leading to a decrease in its mortality from >80% in initial series to a current rate of 10–20%.51
EPIDEMIOLOGY SBP is the most common type of infection in hospitalized cirrhotic patients, occurring in about 9% of cases and accounting for about 25% of all infections.52 The prevalence of SBP appears to be lower in the outpatient setting, where a 3.5% rate has been reported in patients undergoing serial therapeutic paracenteses.53 In prospective studies, the 12-month probability of developing a first episode of SBP in cirrhotic patients with ascites has ranged between 11%54 and 29%,55 incidence that is highly dependent on ascites total protein content (0% in patients with an ascites protein >1 g/dl vs 20% in patients with an ascites protein <1 g/dl).54 Spontaneous bacterial empyema is an entity akin to SBP in which hepatic hydrothorax becomes infected; it can occur in the absence of SBP or ascites, and its diagnosis and management are the same as for SBP.56
CLINICAL PICTURE The typical features of SBP consist of symptoms and signs of a generalized peritonitis, that is, diffuse abdominal pain, fever, abdominal tenderness with rebound tenderness, and decreased bowel sounds. However, patients rarely present with the complete picture and single elements of the typical presentation are more frequent, with isolated fever or abdominal pain being the most common presenting manifestations. The presence of unexplained encephalopathy and/or deterioration in renal function in a patient with ascites should always raise the suspicion of SBP.
DIAGNOSIS A diagnostic paracentesis to investigate SBP should be performed not only in cirrhotic patients with ascites who have clinical features or laboratory abnormalities (e.g. leukocytosis) suggestive
Chapter 19 ASCITES Figure 19-5. Results of ascites fluid analysis of patients in whom both the serum–ascites albumin gradient (SAAG) (A) and the ascites total protein (B) were determined.38 As is clearly shown, patients with cirrhosis have a high (>1.1 g/dl) SAAG and a low ascites protein (<2.5 g/dl), patients with cardiac ascites have high SAAG and high ascites protein, and patients with peritoneal carcinomatosis have a low SAAG and a high ascites protein. (Reproduced from Figure 1 in Runyon BA et al. The Serum Ascites Albumin Gradient is Superior to the ExudateTransudate Concept in the Different Diagnosis of Ascites. Ann Intern Med 1992;117:215–220.)
40
30
20
10
0 Cardiac ascites
Cirrhosis
A
Peritoneal carcinomatosis
(76) 70
60
50
40
30
20
10
0 B
Cirrhosis
Cardiac ascites
Peritoneal carcinomatosis
of SBP, but also in those admitted to the hospital for any other reason and in those with unexplained encephalopathy and renal dysfunction.57 Despite the use of more sensitive bacteriological culture methods, such as the inoculation of ascites into blood culture bottles, ascites culture is negative in approximately 40% of patients with clinical manifestations suggestive of SBP and increased ascites PMN.57
Therefore, the diagnosis of SBP is established when objective evidence of a local inflammatory reaction is present, based on the ascites PMN count. A cut-off of >250/mm3 has been identified as having the greatest diagnostic accuracy.58 In hemorrhagic ascites (i.e. ascites red blood cell count >10 000/mm3), subtracting one PMN for every 250 red blood cells will correct for the excess blood in ascites.
337
Section III. Clinical Consequences of Liver Disease
To maximize the possibilities of isolating an infecting organism, both ascites and blood bacteriological cultures should be performed whenever SBP is suspected.57
TREATMENT Once an ascites PMN count >250/mm3 is detected, antibiotic therapy needs to be started before obtaining the results of ascites or blood cultures. Antibiotic therapy would also be justified in patients in whom ascites cultures are twice positive despite a PMN count <250/mm3 (Figure 19-6). The most effective (~90% resolution rate) and safe drugs are intravenous (IV) cefotaxime or other third-generation cephalosporins.59–62 The combination of amoxicillin and clavulanic acid administered intravenously has been shown to be as effective and safe as cefotaxime in the treatment of SBP.63 Patients who develop SBP on prophylactic quinolones have been shown to respond as well to cefotaxime as patients not on prophylaxis.64 Doses of cefotaxime used in clinical trials have ranged between 2 g IV every 4 hours to 2 g IV every 12 hours, with equal efficacy.62 Ceftriaxone has been used successfully at a dose of 1–2 g IV every 24 hours, and ceftazidime at a dose of 1 g IV every 12–24 hours. The only study assessing the combination of amoxicillin and clavulanic acid used a dose of 1g/0.2 g IV every 8 hours.63 After 5 days of antibiotic therapy, ascites PMN have decreased below 250/mm3 in the majority of patients.65 In fact, 5-day antibiotic therapy has been shown to be as effective as 10-day therapy,60 and is therefore recommended as the minimum duration of therapy.57 However, prospective controlled studies have shown that the median time to SBP resolution is 8 days, and therefore this length of treatment is probably more recommendable.66
The antibiotics recommended above have been associated with very few side effects and no renal toxicity. Cirrhotic patients have an increased propensity to develop aminoglycoside-induced nephrotoxicity,67 and these antibiotics have been shown to be independent predictors of the development of renal dysfunction in such patients.68 Therefore, aminoglycosides should be considered a last resort in the treatment of infections in cirrhotic patients. In patients with community-acquired uncomplicated SBP (i.e. no renal dysfunction, no encephalopathy), oral ofloxacin (or other fully absorbed quinolone) is a good alternative.69 However, the use of quinolones for the treatment of SBP will depend on the local prevalence of quinolone-resistant organisms. Renal impairment, a main cause of death in SBP, occurs as a result of a further decrease in effective arterial blood volume that in turn probably results from a cytokine-mediated aggravation of vasodilatation as occurs in sepsis, in addition to a possible impairment in cardiac function.70 The addition of IV albumin to antibiotic therapy as a measure to increase effective blood volume has been shown to be associated to lower rates of renal dysfunction and in-hospital mortality.71 The dose of albumin used is arbitrary: 1.5 g/kg of body weight during the first 6 hours, followed by 1 g/kg on day 3, and may not be applicable to overweight patients. Patients who appear more likely to benefit from the addition of albumin are those with a serum bilirubin >4 mg/dl and evidence of renal impairment at baseline (BUN >30 mg/dl and/or creatinine >1.0 mg/dl).71 It is therefore this subgroup of patients in whom albumin can be recommended.72 On the other hand, there is a subgroup of patients with SBP – those with community-acquired infection, no encephalopathy and normal renal function – that have a 100% cure
Figure 19-6. Algorithm in a patient with suspected spontaneous bacterial peritonitis (SBP). Once a polymorphonuclear cell count (PMN) >250/mm3 is detected, antibiotic treatment should be initiated. If ascites culture is positive (in a patient with an initial PMN count <250/mm3), diagnostic paracentesis should be repeated and the patient treated if PMN >250/mm3. If the count is still <250/mm3 patient should not be treated unless culture remains positive for the same organism.
SBP suspected
Diagnostic paracentesis
338
PMN < 250
PMN > 250
No action
Treat
Culture negative
Culture positive
No action
Repeat tap
Culture negative Continue therapy
PMN < 250
PMN > 250
No action
Treat
Culture negative
Culture positive
No action
Treat
Culture positive
Chapter 19 ASCITES
rate and 100% survival with antibiotic therapy alone.69 Albumin would not be indicated in these patients.72 A control paracentesis performed 48 hours after starting therapy is recommended to assess the response to therapy and the need to modify antibiotic therapy (depending on the isolation of a causative organism) and/or to initiate investigations to rule out secondary peritonitis.57 This failure of initial therapy occurs in up to 23% of cases. In the presence of an obvious clinical improvement, control paracentesis may not be necessary. Intravenous antibiotics can be safely switched to oral after 2 days of therapy and once a response to therapy is demonstrated by a decrease in ascites PMN.73
bleeding and/or who have not had a previous episode of SBP, particularly as the rate of infection in this patient population is low.53 Limiting the use of prophylactic antibiotics to those at the highest risk of developing SBP is important, given the increased rate of infections with quinolone-resistant and trimethoprim–sulfamethoxazole-resistant organisms observed in patients on long-term norfloxacin prophylaxis.52,79 The results of placebo-controlled trials of antibiotic prophylaxis performed in populations at a higher risk of developing SBP, identified by low ascites protein, high serum bilirubin and a low platelet count,80 are awaited. Investigations of nonantibiotic measures to prevent SBP and other bacterial infections in cirrhosis are ongoing.72
PROPHYLAXIS In patients who survive an episode of SBP the 1-year cumulative recurrence rate is high, at about 70%. Recurrence, particularly from Gram-negative organisms, is significantly and markedly lower with the use of oral norfloxacin at a dose of 400 mg/day.74,75 It is therefore essential that patients surviving an episode of SBP be started on antibiotic prophylaxis to prevent recurrence. The use of weekly quinolones is not recommended, as they have been shown to be less effective in preventing SBP recurrence and are associated with a higher rate of development of quinolone-resistant organisms.75 Prophylaxis should be continuous until the disappearance of ascites (i.e. patients with alcoholic hepatitis), death or transplant. Another group of cirrhotic patients in whom antibiotic prophylaxis should be used routinely is those admitted with gastrointestinal (GI) hemorrhage, in whom the rate of bacterial infection is as high as 45%. In these patients, short-term antibiotic prophylaxis has been shown to be effective not only in reducing the rate of bacterial infections, but also in reducing in-hospital mortality76 and variceal rebleeding.77 The preferred antibiotic is norfloxacin at a dose of 400 mg orally twice a day for 7 days (or less if the patient is to be discharged from the hospital). Data from the only placebo-controlled trial of primary prophylaxis of SBP78 are inconclusive, and therefore there is insufficient information to support the use of long-term antibiotic prophylaxis in outpatient cirrhotic patients with ascites who do not have GI
TREATMENT OF ASCITES The treatment of cirrhotic ascites is important, not only because it improves the patient’s quality of life, but because SBP, one of the most lethal complications of cirrhosis, does not occur in the absence of ascites. Therapies for ascites include sodium restriction, diuretics, large-volume paracentesis (LVP), the transjugular intrahepatic portosystemic shunt (TIPS) and the peritoneovenous shunt (PVS) (Figure 19-7). The development of ascites in a cirrhotic patient denotes a poor prognosis and is an indication to initiate liver transplant evaluation. Therefore, transplantation constitutes the ultimate treatment for ascites and its complications. Ascites responds appropriately in 80–90% of patients upon the attainment of a negative sodium balance through the use of sodium restriction and/or diuretics. Even though this treatment takes longer and may have a higher complication rate than LVP, it still constitutes the mainstay of therapy, given its general applicability, low cost and ease of administration. In fact, the categorization of cirrhotic patients with ascites is based on their response to diuretic therapy. In this scheme, uncomplicated ascites assumes an uninfected ascites with a good response to diuretics, and refractory ascites assumes either diuretic-resistant ascites (ascites that is not eliminated even with maximal diuretic therapy) or diuretic-intractable ascites
Transplant
Figure 19-7. Different treatments of ascites placed in the context of its pathophysiology.
Cirrhosis
Intrahepatic resistance
Sinusoidal pressure
Arteriolar resistance (vasodilation)
Albumin TIPS PVS
TIPS
Diuretics Ascites PVS
LVP
Sodium and water retention Sodium restriction
Effective arterial blood volume
Activation of neurohumoral systems (aldosterone, renin, angiotensin, epinephrine)
339
Section III. Clinical Consequences of Liver Disease
(ascites that is not eliminated because maximal doses of diuretics cannot be reached given the development of renal and/or electrolyte abnormalities).81
SODIUM RESTRICTION Sodium restriction is recommended in all cirrhotic patients with ascites. Although dietary sodium should be restricted to levels lower than urinary sodium excretion, sodium restriction to ~90 mEq/day (i.e. 2 g sodium/day or 5.2 g dietary salt/day, considering that 1 mEq sodium = 23 mg sodium = 58.5 mg dietary salt) is a realistic goal, particularly in the outpatient setting.21 Further restriction of sodium is unrealistic and difficult to achieve. Patients with a baseline urinary sodium excretion >50 mEq/24 hours, an infrequent setting, may respond to salt restriction alone. There are virtually no complications associated with sodium restriction. However, clinicians should be cautious about the nutritional status of patients on sodium restriction, as the non-palatability of a salt-restricted diet may lead to an inadequate food intake. In these cases, liberalizing sodium restriction and adding diuretics is preferable to compromising the already compromised nutrition of the cirrhotic patient with ascites.
DIURETICS Spironolactone is the diuretic of choice. Even though loop diuretics such as furosemide are more potent natriuretics, randomized trials have shown a significantly lower efficacy of furosemide used alone than of spironolactone alone82,83 or the combination spironolactone/furosemide.82 When furosemide is used alone, sodium that is not reabsorbed in the loop of Henle is taken up at the distal and collecting tubules as a result of the hyperaldosteronism present in most cirrhotic patients with ascites. Therefore, furosemide should not be used as the sole agent in the treatment of cirrhotic ascites. Diuretic therapy can be initiated with spironolactone alone or with spironolactone plus furosemide. Both schemes are equally effective; however, dose adjustments are needed more frequently in patients in whom treatment is initiated with combination therapy because of more rapid increases in blood urea nitrogen and/or decreases in serum sodium.82,84 Therefore, and particularly in the outpatient setting, it is preferable to initiate therapy with spironolactone alone at a daily dose of 100 mg orally, and to increase it in a stepwise fashion to a maximum of 400 mg/day. Because spironolactone takes several days to take effect, it can be administered in a single daily dose and adjustments should only be made after the patient has been ≥4 days on a stable dose. If weight loss is not optimal or if hyperkalemia develops, furosemide should be added at an initial single daily dose of 40 mg, increased in a stepwise fashion to a maximum of 160 mg/day. An insufficient diuretic response necessitating increases in the dose of diuretics is defined as a weight loss <1 kg in the first week and <2 kg/week in subsequent weeks.21 Before considering that ascites is refractory to diuretics, it is necessary to ascertain whether the patient has adhered to the prescribed sodium-restricted diet and has restrained from using non-steroidal anti-inflammatory drugs, known to blunt the response to diuretics. Non-adherence to dietary sodium restriction and/or diuretics should be suspected if patients fail to lose weight despite an adequate 24hour urine sodium excretion (>50 mEq/l or greater than daily sodium intake).
340
Complications The more commonly described complications of diuretic therapy are renal impairment due to intravascular volume depletion (25%), hyponatremia (28%) and hepatic encephalopathy (26%).85–87 As renal dysfunction due to diuretics is caused by a reduction in intravascular volume, diuretics should not be initiated in patients with a rising creatinine or in those with concomitant complications of cirrhosis known to be associated with a decreased effective arterial blood volume, such as variceal hemorrhage and SBP. In patients who develop renal dysfunction (elevation in creatinine >50% to >1.5 g/dl) diuretics should be temporarily discontinued and restarted at a lower dose once creatinine returns to baseline. Patients who develop hyponatremia (serum sodium <130 mEq/l) while on diuretics should be managed with fluid restriction and a decrease in the dose of diuretics. In order to minimize the rate of complications, weight loss in patients without edema should be maintained at a maximum of 1 lb/day (0.5 kg/day), whereas in patients with edema a weight loss of 2 lb/day (1 kg/day) is allowable. Spironolactone is often associated with adverse events related to its antiandrogenic activity, mainly painful gynecomastia. Potassium canrenoate, one of the major metabolites of spironolactone, has a comparable diuretic effect and a lower antiandrogenic activity and could be used in cases in which gynecomastia and mastalgia are side effects of spironolactone therapy. However, this drug is not available in the USA. Amiloride, another potassium-sparing diuretic, does not produce gynecomastia and is recommended in patients with intolerable painful gynecomastia, but it has significantly less natriuretic effect than spironolactone.88 Amiloride is used at an initial dose of 20 mg/day and can be increased to 60 mg/day. In patients in whom the natriuretic response on amiloride is suboptimal it may be worthwhile to attempt retreatment with spironolactone.
Contraindications Non-steroidal anti-inflammatory drugs or aspirin blunt the natriuretic effect of diuretics and should therefore not be used in cirrhotic patients with ascites.89,90 Although selective cyclooxygenase-2 (COX-2) inhibitors have not been shown to impair natriuresis or to induce renal dysfunction in cirrhotic rats,91 preliminary data in humans indicate that celecoxib may be related to a decrease in renal function,92 and therefore the use of COX-2 inhibitors should be avoided until more clinical data become available.
LARGE-VOLUME PARACENTESIS (LVP) LVP associated with IV albumin has been shown to be as effective as standard therapy with diuretics, but with a significantly faster resolution and the same or a lower rate of complications.85,86,93 As this therapy is significantly more expensive and requires more resources than the administration of diuretics, it is reserved for patients not responding to diuretics. However, in hospitalized patients with moderate/tense ascites in whom other complications have been resolved, it is reasonable to initiate therapy with total paracentesis as this will accelerate discharge from the hospital. Currently, LVP plus albumin is the standard therapy for refractory ascites. Although initially the recommendation was to perform daily 5 l paracenteses until ascites disappearance, it was subse-
Chapter 19 ASCITES
quently determined that total paracentesis – that is, the removal of all ascites in a single procedure accompanied by the concomitant infusion of 6–8 g albumin per liter of ascites removed – was as safe as repeated partial paracenteses.93 Because LVP is a local therapy that does not act on any of the mechanisms that lead to the formation of ascites, recurrence of ascites is the rule rather than the exception. The frequency of LVP is determined by the rate of ascites reaccumulation and, ultimately, on the need to relieve the patient’s discomfort. The rate of ascites reaccumulation depends largely on patient’s compliance with salt restriction and diuretics as well as the degree of sodium retention. In all LVP studies, diuretics are discontinued before and restarted after the procedure. The need to discontinue diuretics prior to LVP has not been well analyzed, and in practice it is not performed routinely. The administration of diuretics after LVP lengthens the time to ascites recurrence without any differences in complications.94 Therefore, sodium restriction and diuretics at the maximal tolerated dose should be used in conjunction with serial LVP. However, diuretics should be discontinued if associated with complications, or if urinary sodium is <30 mEq/l.
should be administered at a dose of 6–8 g IV per liter of ascites removed. For LVP <5 l a synthetic plasma expander (Haemaccel, Dextran-70) or even saline solution can be used instead of albumin, and it has been suggested that no plasma expansion may be necessary in this setting.21,95,99 Procedure-related complications consist mainly of bleeding and ascites leakage. Major bleeding occurs rarely but may be lethal,100 and is mostly related to rupture of mesenteric vessels rather than as a result of coagulopathy. In fact, in a series of over 1000 LVPs there was no significant bleeding, even in patients with marked thrombocytopenia or prolongation of prothrombin time.101 Clotting abnormalities should therefore not be considered a contraindication to LVP. Leakage of ascitic fluid is rare and occurs when extraction of ascites is incomplete. Therefore, this complication can be solved by completing the LVP, preferably in a site remote from the leaking puncture site. Similarly, another complication of paracentesis that is rare but which should be recognized is the development of sudden scrotal edema that results from subcutaneous tracking of peritoneal fluid into the scrotum; this can be solved by elevation of the scrotum.102
Complications
Contraindications
One of the main complications of LVP is the development of postparacentesis circulatory dysfunction (PCD), defined as a significant increase in plasma renin activity (PRA) 6 days after LVP. The development of PCD is associated with a faster recurrence of ascites, the development of renal dysfunction and a higher mortality.95,96 Two factors are independent predictors of the development of PCD: the amount of ascites extracted and the type of volume expander utilized in association with LVP,95 with the lowest rates observed when less than 5 l are removed and/or when albumin is used as a plasma volume expander (rate around 16%) (Figure 19-8).95,97–99 Albumin
The pathogenesis of PCD appears to be a worsening of the vasodilatory state of the patient, with a consequent further decrease in effective arterial blood volume and marked activation of neurohumoral systems that leads to further sodium retention, renal vasoconstriction, renal dysfunction and death.96 Therefore LVP should not be performed in the setting of conditions that have been associated with a worsening in the vasodilatory state of cirrhosis, such as SBP.
No expander
Saline
Synthetic
Albumin
Development of PCD (%)
70 60 50 40 30 20 10 0 Overall
<5–6l
>5–6l
Figure 19-8. Summary of studies exploring plasma volume expansion and development of post-paracentesis circulatory dysfunction (PCD). Two factors are independent predictors of the development of PCD: the amount of ascites removed and the type of volume expander utilized in association with largevolume paracentesis (LVP).95 Overall, when no volume expander is used, PCD occurs in the majority of patients;97 when synthetic plasma volume expanders or saline are used, PCD occurs in about 30% of patients.95,98,99 and when albumin is used, the lowest rates of PCD (approximately 16%) occur after LVP.95,99 Differences are more obvious with removal of >5–6l. However, with removal of <5l there are no significant differences in the development of PCD among different plasme volume expanders.
TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT (TIPS) Since the publication in 1993 of the first uncontrolled study of TIPS for the treatment of refractory ascites,103 advances have been made in our understanding of the effect of TIPS in cirrhotic patients with refractory ascites, including the publication of five prospective randomized trials comparing TIPS with LVP.104–108 TIPS acts as a sideto-side portocaval shunt thereby decompressing the sinusoids, the source of ascites (Figure 19-9). TIPS placement is associated with a normalization of sinusoidal pressure and a significant improvement in urinary sodium excretion that correlates with suppression of plasma renin activity (indicative of an improvement in effective arterial blood volume).109 Although it prevents the recurrence of ascites, the efficacy of TIPS is offset by an increase in the incidence of severe hepatic encephalopathy, a high incidence of shunt dysfunction and a higher cost. Except for the most recent trial,108 all other trials have shown that TIPS is not associated with a survival benefit or with a quality of life benefit compared to LVP.110 It should be noted that all these studies used uncovered TIPS stents. The use of polytetrafluoroethylene (PTFE)-covered stents has been shown to improve TIPS patency and to decrease the number of clinical relapses and reinterventions, without increasing the risk of encephalopathy.111 It has even been suggested that patients undergoing TIPS with covered stents have higher 2-year survival rates than patients with conventional TIPS.112 This benefit needs to be prospectively evaluated in patients with refractory ascites.
341
Section III. Clinical Consequences of Liver Disease
Hepatic vein
Hepatic vein
IVC
IVC Hepatic venous outflow block
Hepatic venous outflow block
Pressure
Pressure Hepatic sinusoid
Hepatic sinusoid
Hepatic artery
Hepatic artery
A
Portal vein
B
Portal vein
Figure 19-9. Although an end-to-side portocaval shunt (A) would decrease the development of ascites by decompressing splanchnic capillaries and decreasing blood flow into the sinusoids, it can also lead to greater ascites formation, particularly in patients with advanced cirrhosis, in whom the ouflow block is such that the portal vein becomes the outflow tract. On the other hand, the side-to-side portocaval shunt (B) (and the mesocaval shunt), by connecting the side of the portal vein (or the mesenteric vein) to the low-pressure-inferior vena cava, effectively decompress not only the collaterals but also the sinusoids. TIPS is physiologically a side-to-side portocaval shunt.
Until then, the consensus recommendation is to consider TIPS placement when the frequency of LVP is greater than two or three times per month.21
Complications Long-term complications of TIPS are new onset or worsening of prior hepatic encephalopathy and shunt dysfunction. Both occur in about a third of patients over a mean follow-up of about 1 year.109 The procedure-related complication rate is around 9%, with the most common being intraperitoneal hemorrhage.113 Other important complications are heart failure and hemolysis, which may develop in 10–15% of patients.
TREATMENT OF CONDITIONS ASSOCIATED WITH ASCITES TREATMENT OF HYPONATREMIA
A good predictor of post-TIPS survival is the Child–Pugh score, and a score higher than 11 should be considered a contraindication for TIPS placement. TIPS should also be avoided in elderly patients and in those with heart dysfunction.21 Patients with alcoholic cirrhosis who are drinking alcohol may improve with abstinence, and therefore TIPS should be delayed in these patients.
Water restriction to approximately 1 l/day prevents the progressive decrease in serum sodium concentration but does not correct hyponatremia. The administration of hypertonic saline solutions is not recommended as it will lead to further accumulation of ascites and edema. As hyponatremia results primarily from a decrease in effective arterial blood volume, plasma volume expansion with albumin is a reasonable therapy. Preliminary studies show that antidiuretic hormone V2-receptor antagonists increase free water excretion and improve serum sodium concentration in hyponatremic patients with cirrhosis and ascites.116,117 Controlled clinical trials using these drugs are awaited.
PERITONEOVENOUS SHUNT (PVS)
TREATMENT OF HEPATIC HYDROTHORAX
PVS is an alternative to LVP plus albumin. Both therapies have been shown to be equally effective, to have a similar rate of complications and a comparable survival.114,115 However, owing to its high obstruction rate, PVS requires frequent admissions for shunt revision or for the management of other more serious complications. The use of PVS has practically been abandoned because LVP plus albumin is a simpler procedure that can be performed in the outpatient setting. Additionally, placement of a PVS may hinder future placement of TIPS and may complicate liver transplant surgery, given its ability to produce peritoneal adhesions. Therefore, PVS is
Hepatic hydrothorax should be treated in the same manner as cirrhotic ascites, that is, the mainstay of therapy is sodium restriction and diuretics. Before determining that hydrothorax is refractory, a trial of in-hospital diuretic therapy should be attempted. In patients with refractory hepatic hydrothorax other therapeutic options, such as repeated thoracenteses, TIPS or pleurodesis, should be considered. Regarding thoracentesis, and given that no more than 2 l may be removed at a time because of the risk of re-expansion pulmonary edema, the procedure may need to be repeated very frequently. When thoracentesis is required every 2–3 weeks, alternative strate-
Contraindications
342
mostly indicated in patients who require LVP frequently and who are not candidates for TIPS or for transplant.
Chapter 19 ASCITES
gies such as TIPS should be considered. Uncontrolled studies of TIPS have shown resolution of the pleural effusion or a decrease in the need for thoracentesis in 67% of patients; however, mortality is high, particularly in non-responders to TIPS.109 In patients who have not responded to TIPS, an option is video-assisted thoracoscopy to repair diaphragmatic defects and to perform pleurodesis. However, information is available on only a small number of patients and the procedure is associated with significant morbidity and mortality.46 Placement of a chest tube should be avoided in patients with hepatic hydrothorax as it has been associated with multiple complications, mainly volume and electrolyte disturbances.
ated. The mainstay of therapy is sodium restriction and diuretics. Patients in whom ascites is refractory to diuretic therapy should be treated with LVP accompanied by IV infusion of albumin. Until further studies are available evaluating the efficacy of PTFE-covered stents, TIPS should be reserved for patients who require frequent LVP. SBP is a complication of ascites with a high mortality. Prompt diagnosis and early therapy with antibiotics is essential. In patients with SBP and renal dysfunction or jaundice, IV albumin will prevent the development of renal dysfunction and death.
REFERENCES PROGNOSIS AND NATURAL HISTORY Patients usually go through a sequence of uncomplicated ascites Æ refractory ascites Æ hepatorenal syndrome (HRS), each representing a more advanced stage with a worse prognosis. Whereas median survival in patients with compensated cirrhosis is around 9 years,5 once decompensation occurs median survival decreases to 1.6–1.8 years.2,5 Specifically for the presence of ascites, mortality is about 20% per year.118,119 In cirrhotic patients with moderate to tense ascites, four parameters have been found to be independent predictors of survival: impaired water excretion, mean arterial pressure, Child–Pugh score, and serum creatinine.120 Except for the Child–Pugh score, which is indicative of poor liver function, all the other parameters indicate a worsened hemodynamic status (that is, a more vasodilated state) and are consistent with other studies showing that hyponatremia and renal dysfunction are predictors of a poor survival in cirrhosis.121,122 Prognosis is worse in patients who develop refractory ascites, in whom the 1-year probability of death is 43%118 with a median survival of 9–12 months.106,114,118 Child–Pugh score has been consistently shown to be a poor prognostic factor in patients with refractory ascites.106,118,123 The development of PCD (see above) associated with LVP has also been shown to affect survival adversely.95,96 Of all the complications of cirrhosis, HRS has the worst prognosis. This complication is described in detail in Chapter 22. The main determinant of survival is the type of HRS. In type 1, hospital survival is less than 10% and the expected median survival time is only 2 weeks.124 In contrast, patients with HRS type 2 have a longer median survival of around 6 months,125 although still lower than that for refractory ascites. Patients with ascites have 1- and 5-year probabilities of developing HRS of around 20% and 40%, respectively.124 The likelihood of developing HRS is highest in patients with more marked sodium and water retention and more marked activation of vasoconstrictive systems (renin–angiotensin and sympathetic nervous system),124 indicative of marked vasodilatation. In a recent study performed in patients referred for liver transplantation, persistent ascites and low serum sodium identified patients with cirrhosis with high mortality risk despite a low MELD score.126
CONCLUSIONS Ascites is the most common decompensating event in cirrhosis and is associated with a poor prognosis. In cirrhotic patients with ascites evaluation to determine liver transplant candidacy should be initi-
1. Reuben A. My cup runneth over. Hepatology 2004;40:503–507. 2. D’Amico G, Morabito A, Pagliaro L, Marubini E. Survival and prognostic indicators in compensated and decompensated cirrhosis. Dig Dis Sci 1986;31:468–475. 3. Planas R, Balleste B, Alvarez MA, et al. Natural history of decompensated hepatitis C virus-related cirrhosis. A study of 200 patients. J Hepatol 2004;40:823–830. 4. Pagliaro L, D’Amico G, Pasta L, et al. Portal hypertension in cirrhosis: Natural history. In: Bosch J, Groszmann RJ, eds. Portal hypertension. Pathophysiology and treatment. Oxford: Blackwell Scientific, 1994: 72–92. 5. Gines P, Quintero E, Arroyo V. Compensated cirrhosis: natural history and prognosis. Hepatology 1987;7:122–128. 6. Gentilini P, Laffi G, La Villa G, et al. Long course and prognostic factors of virus-induced cirrhosis of the liver. Am J Gastroenterol 1997;92:66–72. 7. Morali GA, Sniderman KW, Deitel KM, et al. Is sinusoidal portal hypertension a necessary factor for the development of hepatic ascites? J Hepatol 1992;16:249–250. 8. Casado M, Bosch J, Garcia-Pagan JC, et al. Clinical events after transjugular intrahepatic portosystemic shunt: correlation with hemodynamic findings. Gastroenterology 1998;114: 1296–1303. 9. D’Amico G, Garcia-Tsao G, Cales P, et al. Diagnosis of portal hypertension: how and when. In: De Franchis R, ed. Portal Hypertension III. Proceedings of the Third Baveno International Consensus Workshop on Definitions, Methodology and Therapeutic Strategies. Oxford: Blackwell Science, 2001:36–64. 10. Wong F, Liu P, Blendis L. Sodium homeostasis with chronic sodium loading in preascitic cirrhosis. Gut 2001;49:847–851. 11. Schrier RW, Arroyo V, Bernardi M, et al. Peripheral arterial vasodilation hypothesis – a proposal for the initiation of renal sodium and water retention in cirrhosis. Hepatology 1988;8:1151–1157. 12. Gines P, Berl T, Bernardi M, et al. Hyponatremia in cirrhosis: from pathogenesis to treatment. Hepatology 1998;28:851–863. 13. Martin PY, Ohara M, Gines P, et al. Nitric oxide synthase (NOS) inhibition for one week improves renal sodium and water excretion in cirrhotic rats with ascites. J Clin Invest 1998;101:235–242. 14. Lee FY, Colombato LA, Albillos A, Groszmann R J. Nw-nitro-Larginine administration corrects peripheral vasodilation and systemic capillary hypotension, and ameliorates plasma volume expansion and sodium retention in portal hypertensive rats. Hepatology 1993;17:84–90. 15. Albillos A, Lledo J L, Rossi I, et al. Continuous prazosin administration in cirrhotic patients: effects on portal hemodynamics and on liver and renal function. Gastroenterology 1995;109:1257–1265. 16. Schepke M, Werner E, Bieker E, et al. The angiotensin II receptor antagonist irbesartan for the treatment of portal hypertension: a placebo-controlled, double-blind study. Hepatology 2000;32:406A.
343
Section III. Clinical Consequences of Liver Disease
17. Thiesson HC, Jensen BL, Jespersen B, et al. Inhibition of cGMP-specific phosphodiesterase type 5 reduces sodium excretion and arterial blood pressure in patients with NaCl retention and ascites. Am J Physiol Renal Physiol 2005;288:F1044–F1052. 18. Simel DL, Halvorsen RA Jr, Feussner JR. Quantitating bedside diagnosis: clinical evaluation of ascites. J Gen Intern Med 1988;3:423–428. 19. Cummings S, Papadakis M, Melnick J, et al. The predictive value of physical examinations for ascites. West J Med 1985;142:633–636. 20. Cattau EL Jr, Benjamin SB, Knuff TE, Castell DO. The accuracy of the physical examination in the diagnosis of suspected ascites. JAMA 1982;247:1164–1166. 21. Moore KP, Wong F, Gines P, et al. The management of ascites in cirrhosis: report on the consensus conference of the International Ascites Club. Hepatology 2003;38:258–266. 22. Goldberg BB, Goodman GA, Clearfield HR. Evaluation of ascites by ultrasound. Radiology 1970;96:15–22. 23. Black M, Friedman AC. Ultrasound examination in the patient with ascites. Ann Intern Med 1989;110:253–255. 24. Runyon BA. Ascites. In: Schiff L, Schiff ER, eds. Diseases of the liver, 7th edn. Philadelphia: Lippincott Williams & Wilkins,1993: 990–1015. 25. Runyon BA. Management of adult patients with ascites caused by cirrhosis. Hepatology 1998;27:264–272. 26. McVay PA, Toy PT. Lack of increased bleeding after paracentesis and thoracentesis in patients with mild coagulation abnormalities. Transfusion 1991;31:164–171. 27. Oelsner DH, Caldwell SH, Coles M, Driscoll CJ. Subumbilical midline vascularity of the abdominal wall in portal hypertension observed at laparoscopy. Gastrointest Endosc 1998;47:388–390. 28. Rector WG Jr. Spontaneous chylous ascites of cirrhosis. J Clin Gastroenterol 1984;6:369–372. 29. Witte CL, Witte MH, Dumont AE, et al. Lymph protein in hepatic cirrhosis and experimental hepatic and portal venous hypertension. Ann Surg 1968;168:567–577. 30. Witte CL, Witte MH. The congested liver. In: Lautt WW, ed. Hepatic circulation in health and disease. New York: Raven Press, 1981: 307–323. 31. Dumont AE, Witte CL, Witte MH. Protein content of liver lymph in patients with portal hypertension secondary to hepatic cirrhosis. Lymphology 1975;8:111–113. 32. Henriksen JH, Horn T, Christoffersen P. The blood–lymph barrier in the liver. A review based on morphological and functional concepts of normal and cirrhotic liver. Liver 1984;4:221–232. 33. Hoefs JC. Serum protein concentration and portal pressure determine the ascitic fluid protein concentration in patients with chronic liver disease. J Lab Clin Med 1983;102:260–273. 34. Henriksen JH. Colloid osmotic pressure in decompensated cirrhosis. A ‘mirror image’ of portal venous hypertension. Scand J Gastroenterol 1985;20:170–174. 35. Pare P, Talbot J, Hoefs JC. Serum–ascites albumin concentration gradient: a physiologic approach to the differential diagnosis of ascites. Gastroenterology 1983;85:240–244. 36. Rector WG Jr, Reynolds TB. Superiority of the serum–ascites albumin difference over the ascites total protein concentration in separation of ‘transudative’ and ‘exudative’ ascites. Am J Med 1984;77:83–85. 37. Albillos A, Cuervas-Mons V, Millan I, et al. Ascitic fluid polymorphonuclear cell count and serum to ascites albumin gradient in the diagnosis of bacterial peritonitis. Gastroenterology 1990;98:134–140. 38. Runyon BA, Montano AA, Akriviadis EA, et al. The serum–ascites albumin gradient is superior to the exudate–transudate concept in the differential diagnosis of ascites. Ann Intern Med 1992;117:215–220.
344
39. Groszmann RJ, Wongcharatrawee S. The hepatic venous pressure gradient: Anything worth doing should be done right. Hepatology 2004;39:280–283. 40. Restuccia T, Gomez-Anson B, Guevara M, et al. Effects of dilutional hyponatremia on brain organic osmolytes and water content in patients with cirrhosis. Hepatology 2004;39: 1613–1622. 41. Belghiti J, Durand F. Abdominal wall hernias in the setting of cirrhosis. Semin Liver Dis 1997;17:219–226. 42. Kirkpatrick S, Schubert T. Umbilical hernia rupture in cirrhotics with ascites. Dig Dis Sci 1988;33:762–765. 43. Lemmer JH, Strodel WE, Eckhauser FE. Umbilical hernia incarceration: a complication of medical therapy of ascites. Am J Gastroenterol 1983;78:295–296. 44. Trotter JF, Suhocki PV. Incarceration of umbilical hernia following transjugular intrahepatic portosystemic shunt for the treatment of ascites. Liver Transplant Surg 1999;5: 209–210. 45. Lieberman FL, Hidemura R, Peters RL, Reynolds TB. Pathogenesis and treatment of hydrothorax complicating cirrhosis with ascites. Ann Intern Med 1966;64:341–351. 46. Cardenas A, Kelleher T, Chopra S. Review article: hepatic hydrothorax. Aliment Pharmacol Ther 2004;20:271–279. 47. Rubinstein D, McInnes IE, Dudley FJ. Hepatic hydrothorax in the absence of clinical ascites: diagnosis and management. Gastroenterology 1985;88:188–191. 48. Strauss RM, Boyer TD. Hepatic hydrothorax. Semin Liv Dis 1997;17:227–232. 49. Schuster DM, Mukundan S Jr, Small W, Fajman WA. The use of the diagnostic radionuclide ascites scan to facilitate treatment decisions for hepatic hydrothorax. Clin Nucl Med 1998;23: 16–18. 50. Bhattacharya A, Mittal BR, Biswas T, et al. Radioisotope scintigraphy in the diagnosis of hepatic hydrothorax. J Gastroenterol Hepatol 2001;16:317–321. 51. Garcia-Tsao G. Spontaneous bacterial peritonitis: a historical perspective. J Hepatol 2004;41:522–527. 52. Fernandez J, Navasa M, Gomez J, et al. Bacterial infections in cirrhosis: epidemiological changes with invasive procedures and norfloxacin prophylaxis. Hepatology 2002;35:140–148. 53. Evans LT, Kim WR, Poterucha JJ, Kamath PS. Spontaneous bacterial peritonitis in asymptomatic outpatients with cirrhotic ascites. Hepatology 2003;37:897–901. 54. Llach J, Rimola A, Navasa M, et al. Incidence and predictive factors of first episode of spontaneous bacterial peritonitis in cirrhosis with ascites: relevance of ascitic fluid protein concentration. Hepatology 1992;16:724–727. 55. Andreu M, Sola R, Sitges-Serra A, et al. Risk factors for spontaneous bacterial peritonitis in cirrhotic patients with ascites. Gastroenterology 1993;104:1133–1138. 56. Xiol X, Castellvi JM, Guardiola J, et al. Spontaneous bacterial empyema in cirrhotic patients: a prospective study. Hepatology 1996;23:719–723. 57. Rimola A, Garcia-Tsao G, Navasa M, et al. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. J Hepatol 2000;32:142–153. 58. Garcia-Tsao G. Spontaneous bacterial peritonitis. Gastroenterol Clin North Am 1992;21:257–275. 59. Felisart J, Rimola A, Arroyo V, et al. Cefotaxime is more effective than is ampicillin–tobramycin in cirrhotics with severe infections. Hepatology 1985;5:457–462. 60. Runyon BA, McHutchison JG, Antillon MR, et al. Short-course versus long-course antibiotic treatment of spontaneous bacterial peritonitis. Gastroenterology 1991;100:1737–1742. 61. Gomez-Jimenez J, Ribera E, Gasser I, et al. Randomized trial comparing ceftriaxone with cefonicid for treatment of spontaneous bacterial peritonitis in cirrhotic patients. Antimicrob Agents Chemother 1993;37:1587–1592.
Chapter 19 ASCITES
62. Rimola A, Salmeron JM, Clemente G, et al. Two different dosages of cefotaxime in the treatment of spontaneous bacterial peritonitis in cirrhosis: results of a prospective, randomized, multicenter study. Hepatology 1995;21:674–679. 63. Ricart E, Soriano G, Novella M, et al. Amoxicillin–clavulanic acid versus cefotaxime in the therapy of bacterial infections in cirrhotic patients. J Hepatol 2000;32:596–602. 64. Llovet JM, Rodriguez-Iglesias P, Moitinho E, et al. Spontaneous bacterial peritonitis in patients with cirrhosis undergoing selective intestinal decontamination. A retrospective study of 229 spontaneous bacterial peritonitis episodes. J Hepatol 1997;26:88–95. 65. Fong TL, Akriviadis EA, Runyon BA, Reynolds TB. Polymorphonuclear cell count response and duration of antibiotic therapy in spontaneous bacterial peritonitis. Hepatology 1989;9:423–426. 66. Garcia-Tsao G. Spontaneous bacterial peritonitis. In: Weinstein WM, Hawkey CJ, Bosch J, eds. Clinical Gastroenterology and Hepatology. Philadelphia: Elsevier, 2005:723–728. 67. Garcia-Tsao G. Further evidence against the use of aminoglycosides in cirrhotic patients. Gastroenterology 1998;114:612–613. 68. Hampel H, Bynum GD, Zamora E, El-Serag HB. Risk factors for the development of renal dysfunction in hospitalized patients with cirrhosis. Am J Gastroenterol 2001;96:2206–2210. 69. Navasa M, Follo A, Llovet JM, et al. Randomized, comparative study of oral ofloxacin versus intravenous cefotaxime in spontaneous bacterial peritonitis. Gastroenterology 1996;111:1011–1017. 70. Ruiz-del-Arbol L, Urman J, Fernandez J, et al. Systemic, renal, and hepatic hemodynamic derangement in cirrhotic patients with spontaneous bacterial peritonitis. Hepatology 2003;38:1210–1218. 71. Sort P, Navasa M, Arroyo V, et al. Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med 1999;341:403– 409. 72. Garcia-Tsao G. Bacterial infections in cirrhosis: treatment and prophylaxis. J Hepatol 2005;42(Suppl):S85–S92. 73. Terg R, Cobas S, Fassio E, et al. Oral ciprofloxacin after a short course of intravenous ciprofloxacin in the treatment of spontaneous bacterial peritonitis: results of a multicenter, randomized study. J Hepatol 2000;33:564–569. 74. Gines P, Rimola A, Planas R, et al. Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis: results of a double-blind, placebo-controlled trial. Hepatology 1990;12:716–724. 75. Bauer TM, Follo A, Navasa M, et al. Daily norfloxacin is more effective than weekly rufloxacin in prevention of spontaneous bacterial peritonitis recurrence. Dig Dis Sci 2002;47:1356–1361. 76. Bernard B, Grange JD, Khac EN, et al. Antibiotic prophylaxis for the prevention of bacterial infections in cirrhotic patients with gastrointestinal bleeding: a meta-analysis. Hepatology 1999;29:1655–1661. 77. Hou MC, Lin HC, Liu TT, et al. Antibiotic prophylaxis after endoscopic therapy prevents rebleeding in acute variceal hemorrhage: a randomized trial. Hepatology 2004;39:746–753. 78. Grange JD, Roulot D, Pelletier G, et al. Norfloxacin primary prophylaxis of bacterial infections in cirrhotic patients with ascites – a double-blind randomized trial. J Hepatol 1998;29:430–436. 79. Novella M, Sola R, Soriano G, et al. Continuous versus inpatient prophylaxis of the first episode of spontaneous bacterial peritonitis with norfloxacin. Hepatology 1997;25:532–536. 80. Guarner C, Sola R, Soriano G, et al. Risk of a first communityacquired spontaneous bacterial peritonitis in cirrhotics with low ascitic fluid protein levels. Gastroenterology 1999;117: 414–419.
81. Arroyo V, Gines P, Gerbes AL, et al. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. Hepatology 1996;23:164–176. 82. Fogel MR, Sawhney VK, Neal A, et al. Diuresis in the ascitic patient: a randomized controlled trial of three regimens. J Clin Gastroenterol 1981;3(Suppl 1):73–80. 83. Perez-Ayuso RM, Arroyo V, Planas R, et al. Randomized comparative study of efficacy of furosemide versus spironolactone in nonazotemic cirrhosis with ascites. Relationship between the diuretic response and the activity of the renin–aldosterone system. Gastroenterology 1983;84:961–968. 84. Santos J, Planas R, Pardo A, et al. Spironolactone alone or in combination with furosemide in the treatment of moderate ascites in nonazotemic cirrhosis. A randomized comparative study of efficacy and safety. J Hepatol 2003;39:187–192. 85. Gines P, Arroyo V, Quintero E, et al. Comparison of paracentesis and diuretics in the treatment of cirrhotics with tense ascites: Results of a randomized study. Gastroenterology 1987;93:234– 241. 86. Salerno F, Badalamenti S, Incerti P, et al. Repeated paracentesis and i.v. albumin infusion to treat ‘tense’ ascites in cirrhotic patients: A safe alternative therapy. J Hepatol 1987;5:102–108. 87. Sola R, Vila MC, Andreu M, et al. Total paracentesis with dextran 40 vs. diuretics in the treatment of ascites in cirrhosis: a randomized controlled study. J Hepatol 1994;20:282–288. 88. Angeli P, Dalla Pria M, DeBei E, et al. Randomized clinical study of the efficacy of amiloride and potassium canrenoate in nonazotemic cirrhotic patients with ascites. Hepatology 1994;19:72–79. 89. Mirouze D, Zipser RD, Reynolds TB. Effects of inhibitors of prostaglandin synthesis on induced diuresis in cirrhosis. Hepatology 1983;3:50–55. 90. Planas R, Arroyo V, Rimola A, et al. Acetylsalicylic acid suppresses the renal hemodynamic effect and reduces the diuretic action of furosemide in cirrhosis with ascites. Gastroenterology 1983;84:247–252. 91. Bosch-Marce M, Claria J, Titos E, et al. Selective inhibition of cyclooxygenase 2 spares renal function and prostaglandin synthesis in cirrhotic rats with ascites. Gastroenterology 1999;116:1167–1176. 92. Guevara M, Abecasis R, Terg R. Effect of celecoxib on renal function in cirrhotic patients with ascites. A pilot study. Scand J Gastroenterol 2004;39:385–386. 93. Tito L, Gines P, Arroyo V, et al. Total paracentesis associated with intravenous albumin management of patients with cirrhosis and ascites. Gastroenterology 1990;98:146–151. 94. Fernandez-Esparrach G, Guevara M, Sort P, et al. Diuretic requirements after therapeutic paracentesis in non-azotemic patients with cirrhosis. A randomized double-blind trial of spironolactone versus placebo. J Hepatol 1997;26:614–620. 95. Gines A, Fernandez-Esparrach G, Monescillo A, et al. Randomized trial comparing albumin, dextran-70 and polygeline in cirrhotic patients with ascites treated by paracentesis. Gastroenterology 1996;111:1002–1010. 96. Ruiz del Arbol L, Monescillo A, Jimenez W, et al. Paracentesisinduced circulatory dysfunction: mechanism and effect on hepatic hemodynamics in cirrhosis. Gastroenterology 1997;113:579–586. 97. Gines P, Tito L, Arroyo V, et al. Randomized comparative study of therapeutic paracentesis with and without intravenous albumin in cirrhosis. Gastroenterology 1988;94:1493–1501. 98. Planas R, Gines P, Arroyo V, Llach J, Panes J, Vargas V, Salmeron JM, Gines A, Toledo C, Rimola A, Jimenez W, Asbert M, Gassull MA, Rodes J. Dextran-70 versus albumin as plasma expanders in cirrhotic patients with tense ascites treated with total paracentesis. Results of a randomized study. Gastroenterology 1990;99:1738–1744.
345
Section III. Clinical Consequences of Liver Disease
99. Sola-Vera J, Minana J, Ricart E, et al. Randomized trial comparing albumin and saline in the prevention of paracentesisinduced circulatory dysfunction in cirrhotic patients with ascites. Hepatology 2003;37:1147–1153. 100. Arnold C, Haag K, Blum HE, Rossle M. Acute hemoperitoneum after large-volume paracentesis. Gastroenterology 1997;113: 978–982. 101. Grabau CM, Crago S F, Hoff LK, et al. Performance standards for therapeutic abdominal paracentesis. Hepatology 2004;40:484–488. 102. Conn HO. Sudden scrotal edema in cirrhosis: a postparacentesis syndrome. Ann Intern Med 1971;74:943–945. 103. Ferral H, Bjarnason H, Wegryn SA, et al. Refractory ascites: early experience in treatment with transjugular intrahepatic portosystemic shunt. Radiology 1993;189:795–801. 104. Lebrec D, Giuily N, Hadengue A, et al. and a French Group of Clinicans and a Group of Biologists. Transjugular intrahepatic portosystemic shunts: comparison with paracentesis in patients with cirrhosis and refractory ascites: a randomized trial. J Hepatol 1996;25:135–144. 105. Rossle M, Deibert P, Haag K, et al. Randomised trial of transjugular–intrahepatic–portosystemic shunt versus endoscopy plus propranolol for prevention of variceal rebleeding. Lancet 1997;349:1043–1049. 106. Gines P, Uriz J, Calahorra B, et al. for the International Study Group on Refractory Ascites in Cirrhosis. Transjugular intrahepatic portosystemic shunting versus repeated paracentesis plus intravenous albumin for refractory ascites in cirrhosis: A multicenter randomized comparative study. Gastroenterology 2002;123:1839–1847. 107. Sanyal AJ, Genning C, Reddy KR, et al. and the North American Study Group for the Treatment of Refractory Ascites. The North American Study for the Treatment of Refractory Ascites. Gastroenterology 2003;124:634–641. 108. Salerno F, Merli M, Riggio O, et al. Randomized controlled study of TIPS versus paracentesis plus albumin in cirrhosis with severe ascites. Hepatology 2004;40:629–635. 109. Garcia-Tsao G. Transjugular intrahepatic portosystemic shunt for the management of refractory ascites in cirrhosis. In: Gines P, Arroyo V, Rodes J, Schrier RW, eds. Ascites and Renal Dysfunction in Liver Disease. 2nd edn. Oxford: Blackwell Publishing, 2005:251–259. 110. Saab S, Nieto J, Ly D, Runyon B. TIPS versus paracentesis for cirrhotic patients with refractory ascites. Cochrane Database Syst Rev 2004;3:CD004889:CD004889. 111. Bureau C, Garcia-Pagan JC, Otal P, et al. Improved clinical outcome using polytetrafluoroethylene-coated stents for TIPS: results of a randomized study. Gastroenterology 2004;126:469–475. 112. Angermayr B, Cejna M, Koenig F, et al. for the Vienna TIPS Study Group. Survival in patients undergoing transjugular
346
113. 114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125. 126.
intrahepatic portosystemic shunt: ePTFE-covered stentgrafts versus bare stents. Hepatology 2003;38:1043–1050. Boyer TD. Transjugular intrahepatic portosystemic shunt: current status. Gastroenterology 2003;124:1700–1710. Gines P, Arroyo V, Vargas V, et al. Paracentesis with intravenous infusion of albumin as compared with peritoneovenous shunting in cirrhosis with refractory ascites. N Engl J Med 1991;325:829–835. Gines A, Planas R, Angeli P, et al. Treatment of patients with cirrhosis and refractory ascites by LeVeen shunt with titanium tip. Comparison with therapeutic paracentesis. Hepatology 1995;22:124–131. Wong F, Blei A T, Blendis LM, Thuluvath PJ. A vasopressin receptor antagonist (VPA-985) improves serum sodium concentration in patients with hyponatremia: a multicenter, randomized, placebo-controlled trial. Hepatology 2003;37:182–191. Gerbes AL, Gulberg V, Gines P, et al. Therapy of hyponatremia in cirrhosis with a vasopressin receptor antagonist: a randomized double-blind multicenter trial. Gastroenterology 2003;124:933–939. Salerno F, Borroni G, Moser P, et al. Survival and prognostic factors of cirrhotic patients with ascites: a study of 134 outpatients. Am J Gastroenterol 1993;88:514–519. D’Amico G. Natural history of compensated cirrhosis and varices. In: Boyer TD, Groszmann RJ (course directors). Complications of cirrhosis: pathogenesis consequences and therapy. American Association for the Study of Liver Diseases 2001;118–123. Fernandez-Esparrach G, Sanchez-Fueyo A, Gines P, et al. A prognostic model for predicting survival in cirrhosis with ascites. J Hepatol 2001;34:46–52. Arroyo V, Rodes J, Gutierrez Lizarraga MA, Revert L. Prognostic value of spontaneous hyponatremia in cirrhosis with ascites. Am J Dig Dis 1976;21:249–256. Llach J, Gines P, Arroyo V, et al. Prognostic value of arterial pressure, endogenous vasoactive systems, and renal function in cirrhotic patients admitted to the hospital for the treatment of ascites. Gastroenterology 1988;94:482–487. Guardiola J, Baliellas C, Xiol X, et al. External validation of a prognostic model for predicting survival of cirrhotic patients with refractory ascites. Am J Gastroenterol 2002;97: 2374–2378. Gines A, Escorsell A, Gines P, et al. Incidence, predictive factors, and prognosis of the hepatorenal syndrome in cirrhosis with ascites. Gastroenterology 1993;105:229–236. Gines P, Guevara M, Arroyo V, Rodes J. Hepatorenal syndrome. Lancet 2003;362:1819–1827. Heuman DM, Abou-Assi SG, Habib A, et al. Persistent ascites and low serum sodium identify patients with cirrhosis and low MELD scores who are at high risk for early death. Hepatology 2004;40:802–810.
Section III. Clinical Consequences of Liver Disease
20
PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES Thomas D. Boyer Abbreviations DSRS distal splenorenal shunt GAVE gastric antral vascular ectasia GI gastrointestinal GOV gastroesophageal
GOV1 GOV2 IGV HVPG
Type 1 gastroesophageal varices Type 2 gastroesophageal varices isolated gastric varices hepatic venous pressure gradient
INTRODUCTION A portal venous system is defined as one that begins and ends in capillaries. The term portal vein is applied to the venous system that begins in the capillaries of the intestine and terminates as the hepatic sinusoids. This vascular system provides for an intimate relationship between the intestines, the pancreas, and the liver. Thus, nutrients absorbed from the gastrointestinal tract, as well as hormones such as insulin and glucagon released by the pancreas secondary to ingestion of food, are delivered to the liver directly and in high concentration. The delivery of these hormones and nutrients to the liver has an impact on the whole organism as well as on the viability of the hepatocytes per se. Unfortunately, the physiologic significance of the portal vein is not of the greatest interest to physicians. Cirrhosis and other conditions cause an increased resistance to portal venous blood flow, which leads to portal hypertension. The major consequences of portal hypertension are venous collaterals, bleeding varices, and ascites. Many patients who bleed from varices will die during hospitalization for that condition. Thus, portal hypertension is one of the most serious sequelae of chronic liver disease. This chapter discusses the pathophysiology of portal hypertension. Included are descriptions of the diseases that cause portal hypertension and the diagnosis and management of bleeding esophageal and gastric varices.
NORMAL ANATOMY OF THE PORTAL VENOUS SYSTEM The liver arises from the foregut and extends into the septum transversum, at which point the liver lobules develop. The blood supply to the liver arises from a separate anlage. The portal vein is a derivative of the omphalomesenteric veins. With regression of the yolk sac, the principal tributaries of the portal vein come from the intestines. Soon after birth the umbilical vein is obliterated and normal adult circulation is established. The portal vein is formed by the confluence of the splenic, superior mesenteric, and inferior mesenteric
IGV IMN TIPS
isolated gastric varices isosorbide-5-mononitrate transjugular intrahepatic portal–systemic shunt
veins (Figure 20-1). The superior mesenteric vein drains the entire small bowel and right colon. The inferior mesenteric vein drains the remainder of the colon and the rectum. Additional contributions to portal venous blood flow are provided by the left gastric (coronary), gastroepiploic, and pancreatic veins. The portal vein begins at the second lumbar vertebra, is posterior to the pancreas, and on average extends for 6.4 cm (4.8–8.8 cm) before entering the liver. Within the liver, it divides into right and left branches that serve the right and left lobes, respectively. The portal venous blood mixes with blood from the hepatic arteries, either in portal venules or in the sinusoids. The blood is collected from the sinusoids by the hepatic veins. The right and left hepatic veins enter the inferior vena cava separately, just before it penetrates the diaphragm. The caudate lobe may drain into the inferior vena cava via a separate hepatic vein.
PHYSIOLOGIC SIGNIFICANCE OF THE PORTAL VEIN Portal venous blood has a high concentration of nutrients and hormones, but a relatively low concentration of oxygen. This allows the liver to play a central role in carbohydrate, fat, and protein metabolism. In addition, the liver acts as a filter of portal venous blood by removing toxic substances that are absorbed by the intestines. Chemical toxins are removed by hepatocytes, and particulate elements (bacteria) are removed by Kupffer cells. Normal hepatic blood flow is approximately 1500 ml/min.1 Thirty percent of the flow and 30–60% of the oxygen consumed by the liver come from the hepatic artery.1–3 The dual hepatic blood supply makes the normal liver resistant to anoxia. Ligation of the portal vein, for example, will not cause hepatocellular necrosis. Similarly, accidental ligation of the hepatic artery or its major branches does not necessarily lead to hepatic failure. The intrahepatic circulation has two unique features. The first relates to outflow resistance. The ratio of pre- to postcapillary resistance in skeletal muscle is 4 : 1, compared to 49 : 1 in the liver. The high ratio in the liver reflects the lack of outflow resistance from the hepatic sinusoids.4 In skeletal muscle, arteriolar vasodilation results
347
Section III. Clinical Consequences of Liver Disease
Stomach Liver
HV
LGV
SV
PV
Spleen LRV RRV IMV SMV IVC Figure 20-1. The portal-venous system. HV, hepatic vein; IVC, inferior vena cava; IMV, inferior mesenteric vein; LGV, left gastric vein; LRV, left renal vein; PV, portal vein; RRV, right renal vein; SV, splenic vein; SMV, superior mesenteric vein.
in a marked rise in capillary pressure because of the high outflow resistance. In the liver, vasodilation causes a minimal rise in sinusoidal pressure, as outflow resistance is low. Thus, the sinusoidal pressure remains low despite changes in blood flow. Maintenance of a low sinusoidal pressure is important because of the character of the endothelial lining of the sinusoids. For example, the endothelial lining of most capillaries is continuous and therefore acts as a semipermeable membrane, allowing water but few proteins to pass. As a consequence, increases of pressure in the capillary, with the resultant transudation of fluids, are counteracted by the rises of intravascular oncotic pressure, decreases in tissue oncotic pressure, and increases in mechanical pressures in the extravascular space (Starling forces). The endothelial lining of the liver, in contrast, contains pores that allow the free passage of proteins into the tissue fluid (Figure 20-2). There is a permeability barrier to the efflux of proteins from hepatic sinusoids to lymphatics only at low venous pressure, for example, 0–2 mmHg. The nature of this effect is uncertain, but it could be secondary to restrictive properties of the interstitium. Nevertheless, the permeability barrier is reduced when hepatic venous pressure exceeds 2 mmHg, and does not exist at pressures greater than 10 mmHg.5 Therefore, any rise in sinusoidal pressure results in the rapid movement of large amounts of fluid into the extracellular space (space of Disse). This fluid is accompanied by proteins. Thus, Starling forces that operate to control movement of fluid between the intra- and the extravascular space in other tissues do not apply to the liver. One can appreciate the importance of maintaining a low sinusoidal pressure by recalling that a rapid rise in sinusoidal pressure, as observed in patients with occlusion of the hepatic veins, leads to massive ascites. The second unique feature of the hepatic circulation is the relationship between hepatic artery and portal vein blood flow.6,7 In both animals and humans, a decrease in portal venous flow or sinusoidal pressure causes a reflex increase in hepatic arterial flow. Conversely,
348
Figure 20-2. Scanning electron micrograph of a hepatic sinusoid. Notice the nucleus of the endothelial cell (EN). The endothelium is fenestrated, and numerous microvilli (MV) from the underlying hepatocyte can be seen protruding into the vascular space through the fenestrae (¥14 000). (Reprinted from Jones AL, Schmucker D. Current concepts of liver structure as related to function. Gastroenterology 1977;73:833–851. ©1977, with permission from Elsevier.)
an increase in sinusoidal flow or pressure causes a reflex decrease in hepatic arterial flow. This reflex change appears to be intrinsic to the liver, inasmuch as transplanted livers that are denervated still demonstrate this response.7 This buffer response may be mediated by adenosine, and the response maintains a constant hepatic blood flow despite changes in portal venous flow that occur during digestion. Changes in hepatic arterial flow, however, do not cause alterations in portal venous blood flow. The latter is controlled by splanchnic hemodynamics, which appear to be independent of metabolic events in the liver. The one exception to this may be that a fall in portal flow with hemorrhage will cause splanchnic vasodilation.7
PORTAL HYPERTENSION DEFINITION The normal portal venous pressure is 5–10 mmHg (7–14 cmH2O).8 A wedged hepatic venous pressure or direct portal venous pressure that is more than 5 mmHg greater than the inferior vena caval pressure, a splenic pressure of more than 15 mmHg, or portal venous pressure, measured at surgery, of greater than 30 cmH2O, is abnormal and indicates the presence of portal hypertension.8 In the vast majority of patients the cause of portal hypertension is increased resistance to flow within the hepatic sinusoids. However, portal hypertension can result from diseases as varied as portal vein thrombosis and constrictive pericarditis. As portal hypertension is a complication of many different disease states, the diagnosis should not be considered an end to the work-up but rather a signpost that provides direction.
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
Figure 20-4. Endoscopic appearance of large distal esophageal varices.
Figure 20-3. Barium esophagogram, demonstrating large varices involving the lower two-thirds of the esophagus. (Courtesy of T Munyer.)
Table 20-1. Portal–Systemic Collaterals Located at transition zones between squamous and glandular epithelium, for example gastroesophageal junction, anus, ileostomies Obliterated fetal circulation in the falciform ligament, for example umbilical and periumbilical veins Retroperitoneal channels originating in the splenic vein and anastomosing with the adrenal gland or renal veins, especially the left renal vein Areas in which the gastrointestinal tract becomes retroperitoneal, that is, when a somatic surface is in contact with an area not covered by peritoneum. These areas include the duodenum; descending, ascending and sigmoid colon; spleen; liver Modified from McIndoe AH. Vascular lesions of portal cirrhosis. Arch Pathol 1928;5:23–42.
CLINICAL FEATURES Portal–Systemic Collaterals The portal system has numerous collaterals that interconnect with the systemic circulation (Table 20-1). The collaterals that lead to the greatest clinical problems lie within the mucosae of the stomach and esophagus. When dilated, they form gastric and esophageal varices, which may appear as abnormalities on upper gastrointestinal series (Figure 20-3), or they may be seen at endoscopy (Figure 20-4). The umbilical vein may dilate, as will the veins of the abdominal wall, with the development of visible collaterals. If the inferior cava is also occluded (as may be seen in the Budd–Chiari syndrome), the direction of blood flow in the abdominal wall collaterals will be cephalad. When the inferior vena cava is patent, the direction of
flow of collaterals will be cephalad above the umbilicus and caudad below it. Without markedly dilated vessels, the direction of flow is usually difficult to determine. In the author’s experience, the existence of posterior collaterals has been the most reliable diagnostic finding in patients with obstruction of the inferior vena cava. If flow in the umbilical and periumbilical veins becomes great enough a caput medusae will form, and there may be an audible venous hum (Cruveilhier–Baumgarten murmur) over the course of the umbilical vein. The origin of the umbilical vein is the left portal vein; therefore, if the umbilical vein is dilated, the cause of the portal hypertension must be intrahepatic. Large venous shunts may also form between the portal vein and the left renal vein. On occasion, these shunts may appear to be as large as surgical portacaval shunts. They appear, however, to be ineffective in reducing portal venous pressure and preventing bleeding from esophageal varices.9 In some cirrhotic patients with chronic encephalopathy very large spontaneous portal–systemic collaterals are present, and they shunt a large portion of the splanchnic venous blood into the systemic circulation. Esophageal varices may never develop in these patients.10 The remaining collaterals listed in Table 20-1 usually do not cause symptoms. However, varices may develop in adhesions from previous intra-abdominal surgery or in the small bowel or colon, and the patient may seek treatment because of gastrointestinal hemorrhage.11,12 A particularly difficult problem is bleeding enterostomal varices in patients who have received ileostomies or colostomies for inflammatory bowel disease, and in whom portal hypertension develops because of liver disease, for example sclerosing cholangitis.11,13 Bleeding from rectal varices is infrequent. Hemorrhoids are different from rectal varices, and the frequency of hemorrhoids is not increased in patients with portal hypertension.14
Splenomegaly Although splenomegaly is commonly associated with portal hypertension, there is a poor correlation between portal venous pressure and the size of the spleen.15 Splenomegaly may be associated with hypersplenism. Leukopenia, thrombocytopenia, and anemia are common clinical features of diseases such as schistosomiasis. The degree of depression of the formed elements is infrequently of sufficient severity to cause clinical problems, that is, bleeding or
349
Section III. Clinical Consequences of Liver Disease
infection, and splenectomy is rarely indicated for hypersplenism resulting from portal hypertension.16
ASSESSMENT OF THE PORTAL VENOUS SYSTEM Identification of Varices Esophageal varices can be seen either endoscopically or radiographically. The most accurate method is endoscopy, as upper gastrointestinal series frequently will fail to identify varices unless they are large. Agreement between different endoscopic observers as to the size of varices is good.17 Investigators have attempted to correlate the size of esophageal varices with portal venous pressure. Studies showing both a correlation and a lack of a correlation have been published.18,19 Different collateral circulations develop in response to portal hypertension,9 and it would not be unexpected that, for a given level of portal hypertension, some patients will have large varices whereas others will have small ones. Valves are present in the perforating veins of the esophagus. They are thought to prevent the flow of blood from periesophageal veins into vessels within the mucosa, but the valves may become incompetent in some patients with portal hypertension. The failure of the valves may increase flow into the varices that are in continuity with the periesophageal veins via perforating veins,20 leading to an increase in variceal size. If the presence of these incompetent valves is an important determinant of the size of varices, it becomes easier to understand the lack of correlation between portal pressure and the size of varices.
Visualization of the Portal Vein and Hepatic Veins Visualization of the portal vein is necessary for several reasons, the most important being to exclude thrombosis of the portal vein or its tributaries. In addition, some surgeons feel that the presence of hepatopetal (toward the liver) or hepatofugal (away from the liver) blood flow should determine the type of portal–systemic shunt performed.21 Also, knowledge of portal vein anatomy is necessary to decide what type of shunt is feasible, or whether the vein is patent for liver transplantation.
Umbilical Vein Catheterization, Splenoportograms, and Visceral Angiography Previously, umbilical vein catheterization or splenoportography were used to visualize the portal vein and its tributaries. Splenoportography was used more frequently because of its ease of performance (Figure 20-5). However, it is associated with a significant incidence of splenic bleeding and a high incidence of false positive results for portal vein thrombosis, because all of the splenic blood flow enters the collaterals, leaving the splenic vein unopacified. Umbilical portography is difficult to perform and its use is limited to specialized centers. The technique most commonly used is venous phase angiography following the selective injection of contrast material into the celiac trunk or the splenic artery. This provides excellent visualization of the portal vein and its tributaries (Figure 20-6). It is also excellent for assessing the patency of portacaval shunts and the development of hepatofugal flow. However, a false diagnosis of splenic vein thrombosis can be made by angiography in patients with cirrhosis if there is extensive shunting of splenic venous blood into large collaterals, which leaves the patent splenic and portal veins unopacified. Magnetic resonance angiography is also excellent for visualizing the portal and splenic veins and may replace angiography as the preferred method for visualizing the portal and hepatic veins. (Figure 20-7 A, B).
Ultrasonography Gray-scale ultrasonography is a simple method of visualizing the portal vein and collaterals (Figure 20-8). It has been suggested that the diameter of the portal vein increases in patients with portal hypertension, and many such patients have a portal vein diameter of >13 mm.22 Either the identification of collaterals or perhaps the lack of response to respiration appears to be an acceptable way to identify patients with portal hypertension. Gray-scale ultrasonography, however, is an unsatisfactory method for estimating pressure, as there is a poor relationship between the diameter of the splanchnic vessels and the level of portal pressure in patients with liver disease.23 The best use of ultrasound is for the identification of clots within the portal vein (Figure 20-9) or its tributaries.
Figure 20-5. Splenoportogram of a patient with splenic vein thrombosis. The needle is within the spleen (closed arrow). Contrast material can be seen to flow from the spleen to the portal vein (PV) via large collaterals (C). The pool of contrast material (open arrow) is within numerous gastric varices.
350
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES Figure 20-6. Following injection of contrast material into the superior mesenteric artery, the portal vein (PV), left gastric vein (LGV), and splenic vein (SV) can be seen. (Courtesy of G Gooding.)
Measurements of Portal Venous Pressure Hepatic Vein Catheterization This is the easiest technique to perform and provides reliable and reproducible measurements. A curved, end-hole catheter or a balloon catheter is passed via an antecubital or a femoral vein into the hepatic veins24 and moved under fluoroscopic control. When it is passed from below it is necessary to use deflectors to enter the hepatic veins, as the veins are directed caudad. When the catheter is passed from above, the hepatic vein usually can be entered with ease. When a patient has moderate to tense ascites, however, the liver is tipped anteriorly and catheterization may be difficult. For this reason, failure to catheterize the hepatic veins in patients with ascites should be considered only suggestive, rather than conclusive, of hepatic vein thrombosis. The use of the femoral approach is preferred by some because the right side of the heart is not entered, which may reduce the incidence of arrhythmias. In the author’s experience, catheterization of the right atrium is essentially without complication and on occasion has provided useful clinical information. The success rate of hepatic vein catheterization exceeds 95%, and it is essentially without morbidity.24 The only difficulty is that most interventional radiologists are not trained in measuring the pressure correctly and it is essential that a tracing be obtained in all patients.24 In addition, there may some variability of the pressure within the liver itself, leading to an underestimation of portal pressure.25 Following entrance into the hepatic veins, either the balloon is inflated or, if using an end-hole catheter, its tip is advanced as far as possible (wedged position). Wedging of the catheter creates a wide area of stasis, and the pressure is then recorded. Following this, a small amount of contrast material is injected (Figure 20-10) to confirm that the catheter is in the wedged position. The catheter is withdrawn and the procedure is repeated in several hepatic veins. The highest pressure recorded is used as the estimate of portal
venous pressure. The inferior vena caval pressure or hepatic vein pressure is recorded and subtracted from the wedged hepatic venous pressure, so that the effects of increased abdominal pressure resulting from ascites can be removed and the hepatic venous pressure gradient (HVPG) obtained.24 The wedged hepatic venous pressure provides an accurate measurement of portal venous pressure in alcoholic liver disease.26 In non-alcoholic liver disease, however, the wedged hepatic venous pressure may underestimate portal venous pressure, although it appears to measure portal pressure accurately in cirrhosis due to hepatitis C.26,27 In conditions causing presinusoidal portal hypertension, wedged hepatic venous pressure is normal or less than portal venous pressure. Measurement of wedged hepatic venous pressure has several clinical uses. An elevated wedged hepatic venous pressure is indicative of intrahepatic portal hypertension and establishes the presence of liver disease in a patient with ascites or splenomegaly. The wedge pressure also may be of some use in the management of the patient with the all too common condition of upper gastrointestinal hemorrhage and endoscopic demonstration of only non-bleeding varices. In such a case, the question is whether the varices are the source of the hemorrhage. Studies28 have reported that variceal hemorrhage does not occur in patients whose HVPG is less than 12 mmHg. In fact a HVPG of at least 12 mmHg appears to be required for the formation of varices.18 However, the HVPG is not significantly different between those with large and small varices. Measurement of the HVPG may potentially be useful in monitoring therapy and in deciding which patients with acute variceal hemorrhage are likely to fail medical therapy. Recent studies have shown that early rebleeding is seen in patients with high portal pressures obtained at the index bleed.29 This information has been used to determine therapy: 116 patients with acute variceal bleeding had their HVPG determined; in 64 the gradient was less than 20 mmHg and they were considered at low risk for rebleeding. Of these low-
351
Section III. Clinical Consequences of Liver Disease
Figure 20-8. Longitudinal ultrasonogram in which the normal main portal vein (MPV) is visible adjacent and anterior to the inferior vena cava (IVC). H, head. (Courtesy of PW Callen.) A
B
Figure 20-7. (A) Portal vein as it enters the liver. (B) Hepatic vein entering the IVC.
risk patients 12% were treatment failures; 52 had pressures above 20 mmHg and were considered at high risk for bleeding. Half of this group had TIPS insertion, whereas the others were managed medically. Treatment failure was seen in 12% of those who had a TIPS and in 50% of those managed medically.29 Hospital mortality was also greater in the high-risk patients managed medically than in the low-risk patients and those undergoing TIPS placement. This study is important because it confirms the value of measuring HVPG in the acute setting and demonstrates how that information can be used to make therapeutic decisions.
352
Figure 20-9. Coronal ultrasonogram from a patient with a hepatoma with a tumor thrombus in the portal venous system. The main portal vein (MPV) is filled with low-amplitude echoes secondary to thrombosis. Notice the clear inferior vena cava (IVC). H, head; A, aorta; L, liver. (Courtesy of PW Callen.)
Perhaps the greatest value of obtaining HVPG will be in determining the response to pharmacologic therapy. A number of studies have shown that if HVPG can be reduced to below 12 mmHg or by more than 20%, bleeding is less likely to occur than in patients who fail to achieve this hemodynamic response.30,31 A limited number of studies have been performed examining HVPG in primary prophylaxis. In one study a significant difference in bleed-
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
Good responders
Poor responders
Cumulative probability
100 90 80 70 60 50 40 30
Responder = 20% fall or HVPG < 12 mmHg
P = 0.008
20 0
12
24
36
48
60
Months
Figure 20-11. Probability of bleeding in responders and non-responders to pharmacologic therapy. Responders had to have a > 20% fall in the HVPG or to a value of £ 12 mmHg. (From Merkel C, Bolognesi M, Sacerdoti D, et al. The hemodynamic response to medical treatment of portal hypertension as a predictor of clinical effectiveness in the primary prophylaxis of variceal bleeding in cirrhosis. Hepatology 2000;32:930–934, ©2000, with permission of WileyLiss.)
Table 20-2. Rebleeding in Patients who are Responders and Nonresponders to b-Blockade (Data from reference 31.)
Figure 20-10. The catheter (C) is in the wedged position. Contrast material has been injected, and beyond the tip of the catheter (arrow), the sinusoids are filled. There is no reflux of contrast material back along the catheter, confirming that it is properly wedged. There is reflux into the portal vein (PV).
ing risk was seen in those with compared to those without a hemodynamic response (Figure 20-11).32 Similar findings have been reported when the hemodynamic response has been examined in secondary prophylaxis studies (Table 20-2). In those with a hemodynamic response the risk of rebleeding varied from 7% to 43%, whereas in those without a response the risk of rebleeding was 22–52%. Even fewer of the patients re-bled when their HVPG was £12 mmHg.31 All of these reports are retrospective analyses of studies in which therapy was not altered based on the hemodynamic data. In addition, in the secondary prophylaxis studies there were significant numbers of patients who failed to get the second pressure measurement, leading to concern about bias in the data set.31 Despite these concerns the data are quite consistent, in that a hemodynamic response is predictive of a reduced risk of bleeding, and in future prospective studies this response should be used to determine the therapeutic approach. Recent analyses suggest this approach may well be cost effective.33,34
Direct Transhepatic Portal Venous Pressure Direct transhepatic puncture of the portal veins and the recording of pressure were first described in the 1950s. This method has been modified by using the same thin needle that is used for transhepatic cholangiography,26 and is performed in a similar manner. The hepatic veins also can be entered, allowing for determination of the portal vein–hepatic vein gradient. It is most useful in identifying patients with presinusoidal portal hypertension or hepatic vein thrombosis.26
Groups Total (n = 279)
Rebleeding average % (range of % in five studies)
Hemodynamic responders n = 101 Hemodynamic non-responders n = 110
35 (26–37) 20 (7–43) 40 (22–52) 16 (0–50)
Repeat HVPG £ 12 mmHg n = 49
Hemodynamic responders are those whose pressure fell to £12 mmHg or who had a 20% fall in HVPG. HVPG, hepatic venous pressure gradient.
Currently, most obtain a direct portal pressure using the hepatic vein approach during the performance of a TIPS (see Chapter 17).35
Variceal Pressure and Flow Measurement Pressure within esophageal varices can be measured directly during esophagoscopy by inserting a needle into the varix. The puncture of varices is associated with minimal bleeding, and the pressure obtained is proportional to the portal pressure. However, when measured this way, variceal pressure was 1.3–14.5 mmHg below the corrected wedged hepatic venous pressure in 8 of 11 patients with cirrhosis.36 Hence variceal pressure provides only an approximation of portal pressure. Varices are thin walled, lack external supporting tissues, and may therefore behave like elastic structures. Mosimann37 suggested that these properties mean that the pressure required to compress a varix will equal the venous pressure within the varix. Based on this principle, he developed a device that passes through a standard endoscope and allows for the measurement of variceal pressure. In a subsequent study, variceal pressure obtained with the endoscopic gauge was very similar to splenic pulp pressure or uncorrected wedged hepatic venous pressure.38 Variceal pressure obtained with the endoscopic gauge was found to be equal to directly measured variceal pressure; however, both were significantly lower than the corrected wedged hepatic venous pressure.36 Variceal pressure measured by a gauge was higher in patients with varices that had
353
Section III. Clinical Consequences of Liver Disease
bled than in those with non-bleeding varices. These differences were found despite the fact that the portal–hepatic vein gradient was the same in both bleeders and non-bleeders.39 The finding of a high variceal pressure on endoscopy also has been found to be predictive of the first bleed from varices.40 Thus, the endoscopic measurement of variceal pressure may have some clinical utility. However, the use of the endoscopic gauge to measure variceal pressure is limited somewhat by the need for an experienced operator and a cooperative patient. In addition, pressure measurements are difficult to obtain in the presence of frequent esophageal contractions.41 The most likely application of this technique will be in the assessment of newer therapies used to treat bleeding varices.42 Blood flow in esophageal varices has also been measured by use of a Doppler flowmeter. A Doppler probe is passed into the esophagus via an endoscope and is placed over a varix. The direction and velocity of flow can then be determined. This technique is most useful in distinguishing a fundal mass from fundal varices.
Measurements of Hepatic Artery Flow and Portal Venous Blood Flow During hepatic vein catheterization, clearance studies can be performed for the measurement of hepatic blood flow. It is not possible to obtain accurate clearance studies in patients with liver disease without catheterization of the hepatic veins because of variable extraction of compounds by the liver. Sulfobromophthalein and indocyanine green are the agents most commonly used to estimate hepatic blood flow. Indocyanine green is preferred because it is cleared only by the liver, and is non-toxic. Following the injection or infusion of indocyanine green or sulfobromophthalein, the disappearance of the compound is followed in both a peripheral vein and a hepatic vein, and the hepatic blood flow is calculated, based on the Fick principle.2 Using these methods, hepatic blood flow has been found to be normal, reduced, and increased in patients with liver disease.43 These methods thus measure total hepatic blood flow but provide no information as to the relative contributions of flow in the portal vein and hepatic artery (see later). Knowledge of the relative contributions of the hepatic artery and portal vein blood flow to total hepatic blood flow may be important in selecting patients for portacaval shunts. For example, it has been suggested that patients with normal portal venous blood flow are poor candidates for shunts that divert portal venous blood. Also, the response of the hepatic arterial blood flow to portal diversion may be important, as a small increase in hepatic arterial flow may be indicative of a poor outcome following a portal–systemic shunt. Portal venous blood flow can be measured during surgery with an electromagnetic flowmeter: it will be normal to stagnant.44 Estimates of portal venous blood flow also can be obtained without surgery. One method requires catheterization of the superior mesenteric artery, hepatic veins, and umbilical vein, with injection of Crlabeled red cells and 125I-labeled albumin microaggregates. The portal venous fraction of hepatic blood flow can also be estimated by computed radionuclide angiography. Following the injection of 99m Tc-pertechnetate, 100 images at 1-second intervals are collected from heart, kidney, lung, spleen, and liver. The portal venous fraction can then be calculated by computer. In patients with cirrhosis, the portal venous fraction varies from normal (66%) to essentially
354
nothing and correlates well with findings on venous-phase angiography.45 However, it has not been established that this technique is measuring portal venous inflow; hence its use in the evaluation of patients with portal hypertension is limited. Hepatic arterial blood flow may also be determined during surgery, using a flowmeter. In one series of 47 patients with cirrhosis, hepatic arterial blood flow was 366 ± 21.3 ml/min.46 Doppler ultrasound is also used to measure the velocity of flow in blood vessels. The average velocity of blood flow in the vessels, determined by use of a pulsed Doppler flowmeter, is multiplied by the cross-sectional area of a vessel as determined by ultrasound to obtain the volume of flow in the vessel. This technique has been used in several studies to measure portal blood flow both in normal individuals and in patients with portal hypertension. Results by Doppler ultrasound compare favorably with measurements obtained by flowmeter and cineangiography. There are, however, a number of technical problems associated with the use of Doppler technology that can lead to errors or variability in estimates of portal venous blood flow.47,48 Irrespective of these technical problems, and given the difficulty in obtaining accurate measurements of portal venous blood flow non-invasively, together with the reasonable accuracy and reproducibility of Doppler ultrasound, measurement of portal blood flow by this technique should provide new insights into the pathophysiology of portal hypertension. Doppler ultrasound may also be used to measure hepatic arterial flow and resistance.
PATHOGENESIS OF PORTAL HYPERTENSION The pressure within a vessel is determined by the flow and resistance within that vessel. This relationship is expressed by Ohm’s law: DP = Q ¥ R, where P is pressure, and Q and R are flow and resistance, respectively. Therefore, an increase in either flow or resistance can lead to an increase in pressure. Resistance depends on a number of factors, as defined by Poiseuille’s law: R = 8nL/pr4, where n is the coefficient of viscosity, L the length of the vessel, and r the radius. The length of the vessel and viscosity are usually constant, although the latter may change during resuscitation of a patient who has recently bled. Radius appears to be the most important factor, as small changes in radius are associated with large changes in resistance.
Increased Blood Flow in the Portal Vein Increased portal venous blood flow is an uncommon cause of portal hypertension. There is little outflow resistance from the liver, and the increase in flow must be large before the sinusoidal pressure rises significantly. It is difficult, for example, to induce portal hypertension in dogs by increasing portal venous blood flow. Canine portal venous pressure rises only after the resistance in the liver increases in response to the high-flow state.49–51 Increased portal venous blood flow in some patients appears to cause portal hypertension only when the vascular resistance in the liver has increased.52 As mentioned above, an increase in portal venous blood flow is an uncommon cause of portal hypertension when the liver is normal. However, when the vascular resistance is increased, as is observed in cirrhosis (see below), then small increases in portal venous inflow may be associated with significant increases in portal vein pressure.
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
Portal–systemic collaterals that develop in response to the rise in portal pressure will tend to minimize the rate of rise in pressure.53 However, large portal–systemic collaterals are seen in patients with portal hypertension, and the higher the pressure the more extensive the collaterals. These observations suggest that either the resistance in the collaterals is very high or there is increased portal venous flow. Although resistance in the portal–systemic collaterals may be high,54 it seems unlikely that it is unusually high, as the collaterals persist after pressure in the portal vein is normalized, either by removing the portal vein obstruction by performing a portal–systemic shunt,55 or following liver transplantation.56 Portal venous flow is normal or reduced in patients with cirrhosis.57,58 However, in cirrhosis portal venous flow does not reflect portal venous inflow, as much of the latter is diverted into portal–systemic collaterals. Only by measuring total inflow can the role of portal venous flow in the pathogenesis of portal hypertension be determined. Studies by Groszmann and colleagues55,59–61 in partial portal veinligated and cirrhotic rats have shown that total portal inflow (portal system plus collaterals) increases as the collateral system develops. They measured portal venous inflow using radiolabeled microspheres in portal vein-ligated rats. Immediately following constriction of the portal vein, portal pressure increased and portal venous inflow decreased. Within a few days, however, portal venous inflow increased to greater than normal values because of a fall in splanchnic arteriolar resistance.62 The increase in portal venous inflow is thought to account for about 40% of the increase in portal pressure that develops in this model.61 These studies were performed within a short time of constriction of the portal vein. Portal hypertension develops over a prolonged period in most patients with liver disease, and it is unclear how closely this acute model mimics such cases. Rats were studied both 20 and 180 days after constriction of the portal vein.62 The high portal venous inflow observed in rats 20 days after constriction was gone by 180 days. This fall was due both to an increase in splanchnic arteriolar resistance to normal levels and an increase in resistance in portal–systemic collaterals to greater than normal values. The cause of the latter increase was unclear, but the authors suggested that the initial high flow in collaterals might have caused structural changes in the vessels, leading to an increase in resistance. Increased thickness of portal vein radicals has been described in patients with arteriovenous (portal vein) fistulae, and this structural change is thought to be due to increased flow.52 Whether similar changes occur in the portal–systemic collaterals is unknown. In addition, in humans only 2% of the increase in portal pressure is felt to be due to the increase in portal flow.54 Attempts have been made to lower portal pressure in patients with cirrhosis by decreasing splenic venous inflow, and hence portal venous inflow, to the liver. Witte and co-workers63 studied seven patients with cirrhosis, portal hypertension, large spleens with increased blood flow and bleeding varices, or hypersplenism. They measured portal pressures in these patients before and after ligation of the splenic artery. Although splenic artery ligation may have benefited the patients, portal venous pressures were only slightly reduced, from 26.6 ± 2.1 mmHg to 22.9 ± 2.1 mmHg. The author believes that this small change in portal venous pressure indicates that high-flow states in the portal (splenic) vein contribute little to the severity of portal hypertension seen in patients with cirrhosis. In addition, reducing portal blood flow to
normal with propranolol in rats with stenosis of the portal vein failed to normalize portal pressure. The drug failed because of an increase in resistance of the portal–collateral vascular system.64 These studies do not support a role for high portal venous flow as a significant factor in the severity of portal hypertension found in patients with cirrhosis. They do suggest that the increase in flow contributes to the formation of portal–systemic collaterals, helps maintain high portal pressure in the face of an increasing collateral circulation, and represents a compensatory mechanism for the decrease in portal flow to the liver that is a sequela of the rise in intrahepatic resistance. With the development of portal hypertension there is also development of the hyperdynamic circulation, the pathogenesis of which is discussed in detail in Chapter 23. The following sequence of events appears to occur in patients with portal hypertension. Pressure in the portal vein is increased because of either hepatic fibrosis or partial or complete occlusion of the portal vein. Total portal venous inflow declines and hepatic blood flow is partially maintained by an increase in hepatic arterial flow (see below). In response to the rise in portal pressure, portal–systemic collaterals develop and vascular resistance falls in the splanchnic bed, leading to the development of a hyperdynamic circulation. Splanchnic and portal venous inflow increase and portal pressure continues to be elevated, despite the opening of the collateral circulation. As resistance in the liver continues to increase, there is a further increase in portal pressure, a fall in liver perfusion by the portal vein, and an increasing percentage of blood that is shunted through the collateral circulation. The liver is therefore deprived of portal blood, which over time will tend to accelerate the progression of the liver disease even if the cause (i.e. alcohol) has been removed. The hyperdynamic circulation contributes not only to the development of portal hypertension but also to the development of the hepatopulmonary syndrome (see Chapter 24), cirrhotic cardiomyopathy (Chapter 23) and ascites and hepatorenal syndrome (Chapter 22).
Increased Hepatic Arterial Flow Hepatic arterial flow also contributes to sinusoidal pressure. There is an intimate relationship between hepatic artery and portal vein blood flow and pressure. Portal venous flow is known to be low in cirrhosis, but sinusoidal pressure is high. The primacy of flow versus pressure can be determined by measuring hepatic artery resistance in humans and animals with cirrhosis. Hemodynamics was measured in patients with cirrhosis both pre and post liver transplant.65 Pretransplant hepatic artery flow was low and resistance high despite low portal vein flow. Following transplantation, hepatic artery resistance was low despite increased portal vein flow. These findings suggest that sinusoidal pressure, not flow, is the principal determinant of arterial resistance in cirrhosis. In an animal model of schistosomiasis a fall in portal venous flow was accompanied by increased hepatic arterial flow despite a rise in portal pressure.66 However, the increase in hepatic arterial flow only partially compensated for the fall in portal venous inflow. In humans, diversion of portal venous flow in some but not all patients with cirrhosis is associated with an increase in hepatic arterial flow.67 Thus, the normal ‘buffer’ response of the hepatic artery is preserved in cirrhosis, albeit probably at a reduced level compared with the normal liver.
355
Section III. Clinical Consequences of Liver Disease
Increased Resistance An increase in vascular resistance is the most common cause of portal hypertension. The portal venous system lacks valves, and any increase in pressure is transmitted quantitatively to structures on the intestinal side of the obstruction. Thus, diseases as different as splenic vein thrombosis, cirrhosis, and constrictive pericarditis cause portal hypertension, but their clinical manifestations differ. The site of the increased resistance in cirrhosis was thought by early investigators to be postsinusoidal. They based this conclusion on the fact that the wedged hepatic venous pressure was increased in patients with cirrhosis. The obstructing catheter creates an area of stasis that extends toward the portal venules (see Figure 20-10). The stasis reaches to the first collateral that leads to an unobstructed sinusoid. Early workers assumed that these collaterals were preserved in cirrhosis, and that the pressure recorded by the wedged catheter therefore reflected pressures near the terminal hepatic venules. Thus, the site of obstruction was considered to be postsinusoidal. However, if this were case the portal venous pressure would have to be greater than the wedged hepatic venous pressure to maintain prograde flow in the portal vein, which is known to be
Normal
Normal HVC
mmHg
Hepatic vein
mmHg
1
3
Sinusoid
3
3
Portal vein
6
6
A
B
Alcoholic Cirrhosis HVC Hepatic vein
Chronic Active Hepatitis HVC
mmHg
mmHg
20
8
Sinusoid
20
8
Portal vein
20
C
356
present in cirrhosis. This idea was supported when one group of investigators found that the portal venous pressure exceeded the wedged hepatic venous pressure by about 20% in patients with alcoholic cirrhosis. However, later studies showed that the portal venous pressure is identical to the wedged hepatic venous pressure in these patients.26,27 and this led to a re-evaluation of the location of the resistance to flow in cirrhosis. Reynolds and colleagues suggested two reasons for the equality of the wedged hepatic venous pressure and portal venous pressure in alcoholic liver disease: (1) the resistance to flow was present throughout the sinusoid, and (2) there was a loss of or reduction in collaterals between the hepatic sinusoids (Figure 20-12). In several investigations, all forms of alcoholic liver disease were found to have equal wedged hepatic venous and portal venous pressures.26,68 Although the pressures were equal in many patients with non-alcoholic liver disease, in some the portal venous pressure exceeded the wedged hepatic venous pressure. In the latter subgroup, either there is an element of presinusoidal obstruction or the collaterals near the terminal hepatic venules are patent, preventing the area of stasis created by the catheter from extending to the portal venules (Figure 20-12).
20 D
Figure 20-12. Schematic representation of the hepatic pressures recorded during wedged hepatic vein catheterization. The pressure recorded by the hepatic vein catheter (HVC) represents an area of stasis created by the catheter (cross-hatched areas) that extends to intersinusoidal collaterals leading to unobstructed sinusoids (normal; A and B). In alcoholic liver disease (C) there is an increase in resistance throughout the entire sinusoid, so that there is a uniform pressure drop. Also, there is a loss of intersinusoidal collaterals, so that the area of stasis extends to the portal vein and the wedge pressure equals the portal vein pressure. With diseases such as chronic active hepatitis (D) the resistance within the sinusoid may be greater near the portal tracts than near the terminal hepatic veins. Also, intersinusoidal collaterals may be preserved, and the wedge pressure would then underestimate the portal vein pressure. (Modified from Reynolds TD, Ito S, Iwatsuki S. Measurement of portal pressure and its clinical application. Am J Med 1970;49:649–657, ©1970, with permission from Elsevier.)
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
In early studies of cirrhotic livers, distortion and reduction of the hepatic microcirculation were observed. These vascular changes, together with compression of the portal and – less probably – the hepatic veins by regenerative nodules, were felt to be a major cause for the increase in vascular resistance observed in cirrhotic livers.69–71 Portal hypertension may, however, be found without cirrhosis,72,73 and this observation has led investigators to question whether deposition of collagen alone is an important factor in increasing intrahepatic vascular resistance. Alcoholic liver injury is marked by deposition of collagen in the space of Disse and around the terminal hepatic veins (Figure 20-13). Liver biopsy specimens were obtained from 70 patients with alcoholic liver disease, and the amount of collagen in the space of Disse was scored on the basis of electron microscopic examination of tissue.74 A significant positive correlation was found between the amount of collagen present and the level of intrahepatic pressure. Studies in alcohol-fed baboons also found that animals with the highest portal pressures had the most extensive perivenular (terminal hepatic vein) fibrosis.75 There also appears to be loss of vessels, leading to what is termed extinction.76 Thrombosis of the portal and hepatic veins is thought to lead to ischemia, loss of parenchyma, and worsening of the fibrosis. The broad band of collagen seen in the cirrhotic liver is thought to be due to parenchymal extinction. Portal pressure in alcoholic liver disease is known to fall with abstinence.77 However, because the fall occurs over 3–4 weeks, a loss of collagen or alteration in the hepatic vasculature is an unlikely explanation for an improvement in portal pressure in this setting. A change in the size of hepatocytes may account for the fall in pressure. In support of this concept is an observation that treatments known to increase the size of hepatocytes also increase portal pressure.78 Studies in alcoholic patients also support the idea that the size of hepatocytes has an effect on sinusoidal volume and hence on portal pressure. In a series of studies, investigators estimated the surface area of hepatocytes and sinusoids in liver biopsy specimens obtained from patients with both alcoholic and non-alcoholic liver
Figure 20-13. Transmission electron micrograph of a hepatic sinusoid. Tissue obtained from a patient with alcoholic liver disease. Notice the red cell (RBC) in the vascular space and the fenestrae (FN) in the endothelium. There are numerous collagen bundles (CB) within the space of Disse. (Courtesy of M Barker.)
disease.79,80 They found a correlation between pressure and size of the hepatocytes with the highest pressures, and the largest cells were found in patients with cirrhosis. Studies by other investigators of a small number of patients with alcoholic liver disease and alcohol-fed baboons have failed to find a correlation between hepatocyte size and portal pressure, perhaps because of variations in technique.81,82 All of the changes in resistance that have been considered are largely passive: that is, they reflect changes in cell size or are secondary to scarring. Recent work suggests that intrahepatic resistance may also change in an active manner. Present in perivenous and perisinusoidal areas of cirrhotic but not normal livers are contractile cells called myofibroblasts83 (see Chapter 6). These cells appear to develop from activated hepatic stellate cells.84 They respond to compounds such as endothelins and nitric oxide, and may relax or contract, thereby altering intrahepatic resistance. Endothelins are a group of compounds that are potent vasoconstrictors. They bind to two different types of receptor, termed ETA and ETB. Binding of endothelins to ETA receptors on vascular smooth muscle cells leads to vasoconstriction, whereas binding to ETB receptors on endothelial cells leads to the release of nitric oxide and vasorelaxation.84 Levels of endothelin are increased in cirrhotics, especially those with ascites.85 Infusion of endothelin into isolated perfused rat livers leads to a rise in portal pressure.86 Upon activation, hepatic stellate cells become myofibroblasts, express ETA and ETB receptors, and contract upon exposure to endothelin 1.84,86 These and other data suggested that endothelins may increase portal pressure in liver disease by binding to hepatic stellate cells, leading to their contraction and a rise in resistance within the liver microcirculation. In support of this idea are the findings that the acute administration of an ETA/B antagonist leads to a fall in portal pressure in cirrhotic rats.87 Chronic administration of an ETA/B antagonist in an animal model, however, failed to lower portal pressure.88 Studies in humans using these receptor antagonists are awaited. There also is intense interest in the role nitric oxide plays in portal hypertension as well as the hyperdynamic circulation. The reader is referred to excellent reviews on this subject.89,90 Nitric oxide overproduction appears to contribute to the development of the hyperdynamic circulation. This conclusion is based on the findings in humans that in exhaled air levels of nitric oxide are increased in cirrhotics before but not after liver transplantation. The exhaled levels of nitric oxide correlate with cardiac index.91,92 Patients with cirrhosis have increased plasma concentrations of nitric oxide as well, and the splanchnic bed appears to be its principal source.90 In animals blockade of nitric oxide synthesis ameliorates the hyperdynamic state.89,90 However, when portal hypertension has been induced in knockout mice for both eNOS and iNOS, the hyperdynamic circulation still developed, indicating that factors other than nitric oxide play an important role in the development of the hyperdynamic circulation (see Chapter 23).93 In contrast to the belief that the nitric oxide is overproduced in the splanchnic bed and systemically contributes to the hyperdynamic circulation, in the portal system it is believed there is an underproduction of nitric oxide, which makes the portal hypertension worse.89 The release of nitric oxide from the endothelial cells of cirrhotic animals is reduced, especially in the face of increased
357
Section III. Clinical Consequences of Liver Disease
levels of endothelin 1.89,94 If this relative nitric oxide deficiency could be corrected then there should be a fall in portal pressure. Nitric oxide synthase was overexpressed in normal and cirrhotic rat livers. This was associated with an increase in nitric oxide levels and a fall in intrahepatic resistance and portal pressure in cirrhotic animals.95 This latter work suggests that targeted increases in nitric oxide levels in the portal circulation could be used to lower portal pressure. If drugs, such as nitrates, can be developed that increase nitric oxide levels in the portal system without spillover into the systemic circulation, then new therapies will be possible.
CLASSIFICATION OF DISEASES CAUSING PORTAL HYPERTENSION Classifications of diseases that cause portal hypertension are usually based on whether the wedged hepatic venous pressure is normal or less than the portal pressure (presinusoidal cause), or whether it is increased and equals the portal pressure (intrahepatic or sinusoidal cause). If the site of obstruction to flow is distal to the sinusoids (i.e. hepatic veins, heart, and so on), the disease is considered to be a postsinusoidal form of portal hypertension.96 Unfortunately, there is a great deal of variation among patients with the same disease as to whether they appear to have sinusoidal, presinusoidal, or a mixed form of portal hypertension, based on the wedged hepatic vein pressure. For example, 43% of patients with non-alcoholic cirrhosis have a portal venous pressure that exceeds the wedged hepatic venous pressure by at least 4 mmHg (there is a presinusoidal component to their portal hypertension), whereas in the remainder wedged hepatic venous pressure equals portal venous pressure.26 Because localization of the site of obstruction has important therapeutic implications, the classification used in this chapter groups diseases that cause portal hypertension by the organ or vessel that contains the obstructing lesion. It is assumed that an increase in resistance, rather than an increase in portal venous inflow, is the principal pathophysiologic event. Diseases that are thought to cause portal hypertension principally by increased blood flow in the portal vein are grouped separately. Those conditions in which the wedged hepatic venous pressure may be less than the actual portal vein pressure are so indicated (Table 20-3).
Table 20-3. Causes of Portal Hypertension Increased portal venous blood flow Arteriovenous fistula Splenomegaly (not caused by liver disease) Thrombosis or occlusion of portal or splenic veins Liver diseases Acute Alcoholic hepatitis Alcoholic fatty liver Fulminant hepatitis Chronic Alcoholic liver disease Chronic hepatitis* Primary biliary cirrhosis* Wilson’s disease* Other forms of cirrhosis, including biliary, hemochromatosis, cryptogenic, cystic fibrosis, a-antitrypsin deficiency Idiopathic portal hypertension* Liver disease caused by arsenic, vinyl chloride, or copper salts* Congenital hepatic fibrosis* Schistosomiasis* Sarcoidosis* Metastatic carcinoma Diseases of hepatic venules or veins and the inferior vena cava Veno-occlusive disease Hepatic vein thrombosis Thrombosis of the inferior vena cava Web lesion, inferior vena cava Cardiac diseases Cardiomyopathy Valvular heart disease Constrictive pericarditis *Diseases in which the wedged hepatic venous pressure may be normal or less than portal venous pressure.
CAUSES OF PORTAL HYPERTENSION Increased Portal Venous Blood Flow Arteriovenous Fistula A fistula between an artery and the portal vein may occur either intra- or extrahepatically (Figure 20-14). These fistulae may be congenital in origin, may be associated with Rendu–Osler–Weber syndrome, or may follow trauma, liver biopsy, or rupture of an aneurysm of the hepatic artery. They may also coexist with hepatocellular carcinoma.97–100 Patients may seek treatment for portal hypertension; not infrequently, however, the fistula is first suspected when an abdominal bruit is heard. In 35% of affected patients abdominal pain is a major complaint. Unlike patients with large peripheral arteriovenous fistulae, those who have arterial–portal vein fistulae rarely have cardiac failure (16% of patients) because of the protective effect of the liver.101 The pathogenesis of the portal hypertension that develops in patients with arteriovenous fistulae should be obvious. The rate of delivery of
358
Figure 20-14. Superior mesenteric artery (SMA)–portal vein (PV) fistula. The SMA (open arrow) and PV (closed arrow) are seen, as is the fistula (curved arrow).
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
blood to the liver exceeds the rate of outflow; accordingly, there is a rise in pressure. Although increased portal blood flow is a consequence of the fistula, not all patients have portal hypertension,26 and factors other than increased flow may be required before portal hypertension develops. In dogs, anastomosis of the hepatic artery to the portal vein leads to muscular hyperplasia and thickening of the portal venules, with the late development of mild portal hypertension.49,51 This research suggests that the liver reacts to increased portal venous blood flow or pressure by increasing its vascular resistance. Increased resistance to blood flow within the liver also has been found in a patient with portal hypertension caused by an arteriovenous fistula. The portal hypertension persisted in this patient despite ligation of the artery feeding the fistula.52 In the case shown in Figure 20-14, embolization of the fistula was followed by portal vein thrombosis because with an increase in intrahepatic resistance the loss of arterial inflow into the portal vein led to static portal vein flow, followed by thrombosis of the portal vein. Thus, in both experimental animals and humans the liver plays an active role in the genesis of portal hypertension that develops secondary to arterial–portal vein fistulae. Treatment of portal hypertension resulting from a hepatic artery–portal vein fistula would appear to be straightforward, that is, ligation or embolization of the fistula.98,100 Unfortunately, significant increases in intrahepatic vascular resistance may become fixed. The portal hypertension may be permanent, so that ligation of the fistula will not relieve the increased pressure.52 A portal–systemic shunt may then be performed. The decision as to which treatment (ligation with or without a portal–systemic shunt) is necessary should be based on the change in portal pressure following clamping of the feeding artery. Performing a portal–systemic shunt without ligation of the fistula is hazardous because this establishes a systemic arteriovenous shunt and could cause congestive heart failure.
Splenomegaly not Caused by Liver Disease Portal hypertension with varices and ascites occasionally develops in patients with diseases that cause splenomegaly, such as polycythemia rubra vera, myelofibrosis, Gaucher’s disease, leukemia, and lymphoma.102–104 This is especially true for patients who have myeloid metaplasia. These patients have either bleeding varices or ascites during the course of their underlying hematologic disease. The uncertainty is not over the existence of the association but over the pathogenesis of the disease. The controversy stems from the findings of increased portal venous blood flow, elevated portal venous pressure, and normal or elevated wedged hepatic venous pressure.105 It is thought that there is an element of liver disease that contributes to the portal hypertension in those patients who have elevated wedged hepatic venous pressure. If the major pathologic event is an increased portal venous blood flow resulting from splenomegaly, splenectomy should be curative. If there is significant irreversible intrahepatic resistance, splenectomy should not be successful. In some cases splenectomy cures the portal hypertension, with resolution of the esophageal varices.103,104 The principal event in these patients, therefore, would appear to be increased splenic blood flow. The reason a few patients with splenomegaly develop portal hypertension is unknown, but it may relate to both the volume of blood flow and the response of the hepatic vasculature to the increased portal venous flow.
Thrombosis of Splenic and Portal Veins (see Chapter 47) Liver Disease Idiopathic or Non-cirrhotic Portal Hypertension The syndrome of idiopathic portal hypertension consists of portal hypertension and splenomegaly without portal vein obstruction or significant liver disease. This syndrome was originally described in 1882 by Guido Banti, and it carries his name. Other names used to describe similar conditions are non-cirrhotic portal fibrosis and hepatopedal sclerosis.106,107 The pathogenesis of the portal hypertension in these patients is unknown, although an episode of portal vein bacteremia or a toxin may be the initiating factor. Banti originally believed that the splenomegaly was the primary event, resulting in increased portal blood flow and portal hypertension. However, subsequent studies have suggested that there is sclerosis of the intrahepatic portal veins. Collagen deposition in periportal areas and in the space of Disse is also a prominent pathologic finding, and the number of portal vein branches is also reduced.106,108 Portal and splenic venous blood flow was measured by Doppler ultrasound in 17 patients with idiopathic portal hypertension.109 Splenic venous blood flow was greater than normal in all cases. However, portal venous blood flow was high in only eight of 17 patients. Compared with patients with normal portal venous flow, those with high flow had significantly lower portal pressures and presinusoidal resistance, and less severe vascular changes in liver biopsy specimens. The patients with high portal flow were felt to be in an early stage of disease in which increased blood flow resulting from splenomegaly was the principal cause of the increase in portal pressure. With time, however, the intrahepatic resistance increases and becomes the predominant cause of high portal pressure.109 To establish that this formulation is correct, serial measurements of pressure and flow need to be made in a large number of such patients. The majority of patients with idiopathic portal hypertension have either splenomegaly or gastrointestinal hemorrhage when they seek medical attention. Ascites also develops in these patients, despite the lack of significant liver disease. Patients who have idiopathic portal hypertension have few of the stigmata of chronic liver disease. Hepatic test results, when abnormal, are only slightly so.106–108 Hepatic failure may, however, develop late in the illness. The prognosis, however, is excellent, with a mean (50%) survival rate of 25 years from the onset of clinical illness.110 The pathologic characteristics of idiopathic portal hypertension are variable. In a few cases there is no histologic abnormality. In most, moderate to marked periportal fibrosis is found. Collagen bundles may be visible in the space of Disse. Phlebosclerotic changes in the portal veins may also be seen. The liver may be shrunken with extensive fibrosis in a patient with longstanding disease.106–108 The diagnosis of idiopathic portal hypertension is suggested when the biopsy specimen of the liver of a patient with features of portal hypertension shows either no abnormality or periportal fibrosis without cirrhosis. The biopsy specimen must contain portal areas so that the possibility that it represents only a regenerative nodule can be excluded. Celiac angiography can exclude portal or splenic vein thrombosis, and measurement of wedged hepatic venous pressure and portal venous pressure can then establish the diagnosis with cer-
359
Section III. Clinical Consequences of Liver Disease
tainty. Several investigators have observed that the wedged hepatic venous pressure is normal to moderately elevated; however, the direct portal venous pressure is always greater than the wedged hepatic venous pressure in this condition.106
Portal Hypertension Resulting from Arsenic, Vinyl Chloride, or Copper Many cases of idiopathic portal hypertension have now been found to result from environmental toxins. The first to be defined was chronic arsenical intoxication. Since 1950, prolonged ingestion of arsenicals has been known to cause hepatic fibrosis and cirrhosis. However, it was 20 years before the association between idiopathic portal hypertension and chronic exposure to arsenic was reported. High hepatic arsenic levels were found in four of nine patients from India with idiopathic portal hypertension. The hemodynamic findings (normal to increased wedged hepatic venous pressure, increased portal venous pressure, normal hepatic blood flow) were identical to those reported for idiopathic portal hypertension,111,112 and the pathologic characteristics were also similar.106 In 1974, Creech and Johnson reported that angiosarcomas developed in workers exposed to vinyl chloride.113 Exposure to vinyl chloride also causes periportal fibrosis and collagen deposition in the space of Disse, similar to that described for arsenic.114 Not unexpectedly, exposure to vinyl chloride has now been reported to cause a syndrome identical to idiopathic portal hypertension.115 The use of copper sulfate has also been associated with unusual hepatic fibrosis and presinusoidal portal hypertension.116 This increasing evidence suggests that many cases of idiopathic portal hypertension are the result of exposure to environmental toxins.
Schistosomiasis Hepatic schistosomiasis is a worldwide public health problem (see also Chapter 36). It is a major cause of portal hypertension and bleeding esophageal varices, and may account for 88% of the liver disease seen in some countries. The development of portal hypertension follows infestation of the liver by the parasite’s eggs. The adult worms shed their eggs into the portal vein, and the eggs stop in the portal venules. The presence of the eggs results in granuloma formation, portal fibrosis, and obstruction of the intrahepatic branches of the portal vein by fibrotic tissue.117–119 Portal hypertension is the natural consequence of these events. Because the abnormality is limited to the portal areas, the majority of the hepatocytes are preserved and hepatic function is well maintained. As the principal area involved in hepatic schistosomiasis is the portal area, a presinusoidal form of portal hypertension would be expected. Early hemodynamic investigations confirmed that wedged hepatic venous pressure was normal in the presence of significant portal hypertension.120 More recent studies, however, have shown that some patients with hepatic schistosomiasis have elevated wedged hepatic venous pressures.121 Patients with such elevated pressures may have more advanced disease, that is, extension of the fibrotic process into the sinusoids and the development of sinusoidal portal hypertension. This would explain the elevated pressures. However, an alternative explanation has been proposed: patients who have hepatic schistosomiasis have enlarged hepatic arteries and increased hepatic arterial blood flow.121 The latter finding may be a secondary response to decreases of portal venous blood flow. The question is, then,
360
whether the increased hepatic arterial blood flow elevates the wedged hepatic venous pressure. A later investigation suggested that this may be so. Clamping the hepatic artery in such patients leads to a significant decline in the wedged pressure but does not change portal venous pressure.121 The increased hepatic arterial blood flow might be an important factor in the maintenance of normal liver function in hepatic schistosomiasis. Alternatively, others suggest that because of massive splenomegaly there is high portal venous flow, and the latter causes a reflex decrease in hepatic arterial flow.120 It seems to this author that high portal venous flow is an unlikely explanation for the hemodynamic abnormalities found in patients with hepatic schistosomiasis. The use of Doppler ultrasound to measure splanchnic blood flow should resolve this question.
Alcoholic Liver Disease Cirrhosis caused principally by alcohol ingestion is the most common cause of portal hypertension in the western world. Much of the portal hypertension that develops in cirrhosis is caused by the loss of normal liver architecture. The normal hepatic vasculature is lost and is replaced by new, distorted channels that provide increased resistance. Also, the injury in alcoholic liver disease is most severe in zone 3 of Rappaport, near the terminal hepatic venule. The injury consists not only of liver cell death and inflammation, but also of sclerosis and obliteration of the terminal hepatic venules (acute alcoholic hepatitis).122 The hemodynamic sequela of this injury is similar to that in patients with occlusion of the hepatic veins. If the obliteration of the hepatic veins is severe enough, portal hypertension with ascites and esophageal varices develops without the presence of cirrhosis.123 Resolution of the acute alcoholic hepatitis may result in complete or partial resolution of the portal hypertension.26,77 If significant hepatic fibrosis follows the acute injury, the portal hypertension will probably be fixed, although possibly at a lower level. For example, the author studied one patient with acute alcoholic hepatitis in whom the wedged hepatic venous pressure was 10 mmHg greater than the portal venous pressure, and blood flow in the portal vein was away from the liver. With resolution of the acute injury, the two pressures became similar (wedged hepatic venous pressure decreased), and portal venous blood flow was prograde.26 During the acute illness this patient had a marked outflow blockage, and the portal vein became an exit route. When the acute injury abated, the outflow blockage was reduced. However, the patient was left with a fixed level of portal hypertension. Alcoholic fatty liver has also been reported to cause portal hypertension.124 The increase in pressure is apparently secondary to hepatic enlargement and compression of the sinusoids (see previous discussion). The pressure elevation is usually slight, and complete recovery is the rule in such cases. The wedged hepatic venous and portal venous pressures are equal in most patients with alcoholic liver disease, irrespective of the stage or activity of the disease (Figure 20-15 A).24, 26,125 The two pressures are equal in alcoholic liver disease because (1) the obstruction to flow is along the entire sinusoid, and (2) there may be a reduction in collaterals between sinusoids. Therefore, the area of stasis created by the catheter extends to the portal venules, and the pressure recorded is equal to the pressure in the portal vein (see Figure 2012).126
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
Non-alcoholic Liver Disease (Hepatitis C, Primary Biliary Cirrhosis, Wilson’s Disease) Portal hypertension also develops in patients with non-alcoholic liver disease. The location of the abnormal resistance to flow in these conditions is variable. Patients who have primary biliary cirrhosis are believed to have presinusoidal blockage early in the course of the illness, with sinusoidal blockage developing later.127 The author measured both portal venous and wedged hepatic venous pressures in eight patients with primary biliary cirrhosis and found the portal venous pressure to be significantly higher in only two.26 Patients with chronic hepatitis also may have an element of presinusoidal obstruction (Figure 20-15B). The fact that the wedged hepatic venous pressure may be normal when the portal venous pressure is increased limits the usefulness of measuring the former when attempting to determine the presence or absence of fibrosis. Measurements in patients with hepatitis C have found the two pressures to be similar.27 Wilson’s disease has been reported to cause presinusoidal portal hypertension. However, in other studies the pressures were similar.26 There are two reasons why portal venous pressure may be higher than wedged hepatic venous pressure in diseases such as primary biliary cirrhosis and chronic hepatitis. First, in both of these conditions the disease is most severe in the portal area. The resistance to flow would therefore be at the level of the portal venules and not within the sinusoids. Periportal fibrosis in animals is associated with alterations in portal venules that could lead to a selective increase in resistance in peripheral branches of the portal vein.128 Second, even with involvement of the sinusoids, if collaterals between sinusoids near the hepatic venules are preserved the wedge pressure would tend to underestimate the portal pressure (see Figure 20-12).
Metastatic Carcinoma to the Liver The development of ascites in a patient with widespread metastatic disease is presumed to be the result of peritoneal seeding by
tumor. Peritoneal seeding is the most common cause of ascites in this group of patients, but they also may have ascites because of portal hypertension. The ascites that forms is a transudate. Varices also may form, and on occasion bleed.129 Hepatic metastases can cause portal hypertension secondary to tumor emboli in the portal venules, or secondary to extensive replacement of liver by tumor. When ascites is due to extensive hepatic metastases, and not to peritoneal seeding, the intraperitoneal instillation of chemotherapeutic agents is not appropriate.
Congenital Hepatic Fibrosis This condition is a common cause of portal hypertension in children. An occasional patient with this disease may first seek treatment as an adult.130,131 The principal abnormality is in the portal triads. There is a proliferation of fibrous tissue and abnormal bile ducts; the lobular architecture is preserved, and there is no cirrhosis. The patient frequently has an associated renal lesion that is similar, if not identical, to medullary sponge kidney.131 Patients have signs and symptoms of portal hypertension – bleeding varices or splenomegaly. Patients with portal hypertension may have normal or elevated wedged hepatic venous pressures (see also Chapter 70).26 Some patients with congenital hepatic fibrosis may have a defect in protein glycosylation.132
Sarcoidosis This condition commonly involves the liver (see Chapter 38). Serious liver disease in patients with sarcoidosis is uncommon, but has been well described.133,134 In one patient with hepatic sarcoid and portal hypertension the sinusoidal pressure was normal, whereas the splenic pulp pressure was increased, leading the authors to conclude that sarcoidosis caused a presinusoidal form of portal hypertension. In a more extensive study, patients with mild hepatic sarcoid had normal sinusoidal pressures and patients with severe involvement had elevated sinusoidal pressures. In two patients both 30 Wedged hepatic vein pressure (mmHg)
Wedged hepatic vein pressure (mmHg)
30
20
10
N=29 r=96
(2) 0
20
10
N=25 r=78 0
20 10 Portal vein pressure (mmHg)
30
10 20 Portal vein pressure (mmHg)
30
Figure 20-15. Portal vein pressures and wedged hepatic vein pressures were measured in patients with alcoholic liver disease (A) and in patients with chronic active hepatitis (B). In alcoholic liver disease the portal vein pressure and wedged hepatic vein pressure are equal, whereas in chronic active hepatitis the portal vein pressure may exceed the wedged hepatic vein pressure. (Reproduced from Boyer TD, et al. Gastroenterology 1977;72:584–589, ©1977, with permission from Elsevier.)
361
Section III. Clinical Consequences of Liver Disease
wedged hepatic and portal venous pressure were measured, and both were similarly elevated.134 The appearance of portal hypertension is probably a sequela of hepatic fibrosis and cirrhosis that develops in severe sarcoidosis.134
Cystic Fibrosis This is a childhood disease that may involve the liver, with the development of serious hepatic disease (see also Chapter 69). In two large series 2.2% of patients with cystic fibrosis developed portal hypertension, which was manifested by bleeding varices, hypersplenism, or ascites.135,136 In another series, 10% of patients developed hepatic fibrosis or cirrhosis. Complications were seen most commonly following puberty.137 Investigation of the livers of these children revealed focal biliary cirrhosis, which had developed secondary to obstruction of bile ducts by inspissated mucus. Hepatic function may be well preserved because some areas of the liver are not involved. Children with cystic fibrosis and bleeding varices have been treated effectively with portal–systemic shunts.135,136 Sclerotherapy may also be effective.
Nodular Regenerative Hyperplasia and Focal Nodular Hyperplasia Nodular regenerative hyperplasia of the liver is a rare condition characterized by small regenerative nodules composed of hepatocytes that compress the surrounding parenchyma and are not surrounded by fibrous tissue.138,139 It is the lack of fibrosis that differentiates nodular regenerative hyperplasia from cirrhosis. The cause of the hyperplasia is unclear, but patients frequently have another serious chronic disease, most commonly rheumatoid arthritis or a hematologic disorder.138 This condition has been seen in association with ingestion of an adulterated rapeseed oil.140 Familial occurrence also has been described.141 Also, ischemia leading to apoptosis followed by nodular hyperplasia may play an important role.138,143 Routine liver test results are usually normal, or the alkaline phosphatase level is mildly increased. Patients are identified because of the development of splenomegaly, hepatomegaly, ascites, or bleeding varices.138,139 The wedged hepatic venous pressure may be normal or increased, and there appears to be a presinusoidal component to the portal hypertension in some cases.138,142 The cause of the portal hypertension is unclear, but compression of portal vessels by nodules, obliteration of portal veins, and increased splenic blood flow have all been thought to be important factors. Partial nodular transformation of the liver is also an uncommon clinical entity. The cause of this disease also is unknown. Hypoplasia of the major portal veins, resulting in atrophy in underperfused portions of liver and hyperplasia in areas with adequate blood supply, has been offered as an explanation for the unusual pathologic findings.144 The pathologic lesion consists of large (0.2–8.0 cm) nodules that are perihilar in location and occupy up to two-thirds of the liver. There is minimal fibrosis, except for central scars in some nodules. The remaining normal liver is compressed by the nodules. Hepatic test results are minimally abnormal, and patients usually present with bleeding varices and splenomegaly.145,146 The portal hypertension is felt to be secondary to venous compression by the nodules, although frequently there is associated portal vein thrombosis. The wedged hepatic venous pressure was increased in the single patient in whom it was measured.145
362
Heart Disease (see Chapter 56) Patients who have severe right-sided heart failure may present a clinical picture similar to that seen with hepatic vein thrombosis. The pathophysiology of the hepatic injury and the development of portal hypertension are really no different from hepatic vein thrombosis. The venous pressure rises in the right side of the heart and is transmitted quantitatively to the hepatic vein, the liver sinusoids, and the portal vein and its tributaries. Liver biopsy specimens show centrilobular congestion identical to that seen early in the Budd–Chiari syndrome. The clinical setting is a patient with severe underlying cardiac disease (cardiomyopathy, mitral stenosis complicated by tricuspid insufficiency, or constrictive pericarditis) who has longstanding heart failure and gradually develops ascites. Hepatic biochemical tests vary in the extent of their abnormalities. DIAGNOSIS AND MANAGEMENT OF BLEEDING ESOPHAGEAL AND GASTRIC VARICES Cause of Bleeding The two most popular theories for why varices bleed are erosion of varices secondary to acid reflux, and spontaneous variceal rupture. Ascites and plasma volume also have been thought to be important factors in the genesis of bleeding. Plasma volume is increased in cirrhosis, and changes in plasma volume may be followed by changes in portal venous pressure. Increases in plasma volume raise portal venous pressure and increase the size of esophageal varices.147,148 Sudden changes in plasma volume may occur infrequently. For example, loss of blood with bleeding or with a diuresis will reduce the plasma volume, with an associated fall in portal pressure, and may be an important factor in the cessation of hemorrhage.149–151 Conversely, overexpansion of the plasma volume during hospitalization will increase portal pressure and may be associated with hemorrhage. Studies in animals with portal hypertension showed that repletion of the blood volume following phlebotomy is associated with a rise in portal pressure to greater than control values.149 This worsening of portal hypertension following transfusion appears to be due to a rise in portal vascular resistance.150 If this animal model is representative of patients with cirrhosis, it may help to explain the common clinical event of patients who initially stop bleeding from their varices only to rebleed following hospitalization and the initiation of treatment. Ascites may be a factor in variceal hemorrhage because it is thought to increase blood flow and pressure in esophageal varices. Iwatsuki and Reynolds152 studied 15 patients with cirrhosis and determined that although paracentesis lowered intra-abdominal pressure, portal venous pressure and hepatic blood flow were unchanged. Subsequent studies have shown a fall in the HVPG (10%) following total paracentesis and a fall in hepatic blood flow as well.153 Intravariceal pressure, variceal size, and wall tension also fall after paracentesis, and an increase in intra-abdominal pressure is associated with an increase in size of the varices and an increase in wall tension.154,155 Thus, pressure in the varices does appear to be influenced by intra-abdominal pressure, but whether this increases the risk of bleeding and, more importantly, whether a total paracentesis would decrease the risk of rebleeding, is unknown. Erosion of varices secondary to reflux esophagitis has been suggested to be an important factor in precipitating bleeding. Post-
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
mortem and endoscopic studies in the past supported a role for esophageal erosion. However, studies show that patients who bled from varices had normal lower esophageal pressures and responded normally to alkalinization. There was no evidence of acid reflux when it was tested for with an esophageal pH electrode. Also, treatment with cimetidine did not affect rates of rebleeding from varices.156–158 The most convincing evidence against the erosion mechanism of variceal hemorrhage comes from histologic studies of esophageal specimens obtained from patients undergoing esophageal transection for acutely bleeding varices. When patients who have received therapies known to cause esophagitis (i.e. use of a Sengstaken–Blakemore tube or sclerotherapy) were excluded from analysis, few of the remaining patients had histologic evidence of esophagitis.159,160 An esophageal varix can be considered to be an elastic structure, and the tension on the vessel wall is thought to be an important factor in determining when rupture occurs. Wall tension (T) varies as a function of transmural pressure (TP), vessel radius (r), and wall thickness (W), as shown in the following equation: T = TP ¥ r/W. The wall tension resists the expanding forces (TP ¥ r/W), and when the latter exceeds the former, rupture occurs (Figure 20-16). Polio and Groszmann161 studied this relationship in artificial varices. Large artificial varices ruptured at a lower intraluminal pressure than did smaller ones that had the same initial wall thickness. Similarly, an initially thin-walled varix ruptured at a lower pressure than did a thicker-walled varix, despite having the same radius. This concept predicts that both variceal size and intravariceal pressure should be important in the genesis of variceal hemorrhage. The relationship between these two variables and the risk of hemorrhage is more likely to be qualitative, however, as it is not possible to control for other variables, such as wall thickness, the adequacy of supporting structures, and the presence of incompetent valves in the perforating veins that connect mucosal and periesophageal vessels. For example, the above relationship predicts that the tissues surrounding the varix that resist the expanding force (balloon in a bottle) may play an important role in preventing hemorrhage. Esophageal varices rupture most commonly at or near the cardioesophageal junction. This is the area of the esophagus where the veins are most
superficial and hence least surrounded by other tissues.162 It is not surprising, therefore, that the distal esophagus is the most frequent site of bleeding. In addition, large retroperitoneal varices never bleed, probably because they are supported by surrounding tissues that help resist the expanding forces and thus maintain the integrity of the vessel wall. There appears to be a relationship between the risk of bleeding from varices and portal pressure. A minimal level of pressure (≥11–12 mmHg) is required for the development of varices.18 As the presence of varices is required for bleeding to occur, it is therefore not surprising that bleeding is largely seen in patients with pressures greater than 11–12 mmHg (Figure 20-17).18,28 Also, patients who have bled tend to have higher pressures than those who have not.31 Many patients with pressures greater than 11–12 mmHg have not bled from varices, however, and there are a few patients with varices or a history of bleeding who have relatively low portal pressures. The level of portal pressure appears to be most important, in that varices will form only in those patients with relatively high pressures. Once the varices form, however, the level of portal pressure is a poor predictor of future hemorrhage. Large varices are found more frequently in patients with a history of prior variceal hemorrhage than in those who have never bled.18,163,164 When followed prospectively, patients with large varices are also more likely to bleed than are those with small varices. Rector and Reynolds164 reviewed four studies that graded the size of varices in a group of cirrhotic patients who had not bled previously and then followed them for a period of 6–86 months. About one in three of those patients with large varices and 1 in 10 of those with small varices bled during the period of observation. Conversely, two of three patients with large varices did not bleed. More recently the risk of bleeding from small varices has been examined in a large prospective trial. The 2-year risk of bleeding in those with small varices was 12%, compared to a 2% risk of bleeding in those who lacked varices.165 Hence, large varices increase the risk of
30
Bled from varices ALD Other
Never bled from varices ALD Other
Esophageal lumen
Varix
Portal vein pressure mmHg above IVC
25 20 15 10
Pressure 5 Tension 0 Tension = TP x r/W Figure 20-16. Tension (T) on varix wall is determined by transmural pressure (TP), radius (r) and thickness of the wall (W) according to the relationship T = TP ¥ r/W.
N=11
N=2
N=33 N=37
Figure 20-17. Portal vein pressures of patients with alcoholic liver disease (ALD) and patients with other forms of liver disease were recorded. Charts were retrospectively reviewed for a history of bleeding varices. Patients bleeding from varices had a minimal pressure of 11.5 mmHg above inferior vena caval pressure (IVC).
363
Section III. Clinical Consequences of Liver Disease
bleeding, but it is not zero even in those with small varices, and other factors, such as severity of liver disease, contribute to the risk of bleeding.
Risk Factors and Bleeding from Esophageal Varices166 The above relationship predicts that large thin-walled varices are more likely to bleed than are small thick-walled ones. Based on these concepts, a number of groups have developed grading systems for the endoscopic appearance of varices in order to judge the risk of bleeding. The systems most commonly used were developed by the Japanese Research Society for Portal Hypertension,167 the North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices,168 and more recently one developed by the Italian Liver Cirrhosis Project. In this most recent report the size of varices, severity of gastropathy, and the presence of gastric varices were independent predictors of bleeding and a better index than that developed by the North Italian Endoscopic Club.169 None of these scoring systems is perfect, and each has varying degrees of sensitivity and specificity.170 However, they are quite useful when comparing different study populations and the effects of therapy. The results of one of these studies are shown in Table 20-4. It is apparent from Table 20-4 that the endoscopic appearance of the varices and the clinical state of the patient provide a relative risk of bleeding during a given period of observation. An index was calculated using three variables (Child’s class, size of varices, and presence of red wale markings). Patients were placed into six risk groups, and in a retrospective study observed bleeding rates were highest in the group with the highest index (bleeding rate in this high-risk group was 58.5% at 6 months, 68.9% at 1 year, and 68.9% at 2 years, compared with a 0%, 1.6%, and 6.8% risk of bleeding, respectively, in the lowest-risk group).168 The results shown in Table 20-4 are similar to those of a number of other studies in that it is the appearance of the varices and the clinical condition of the patient that determine the risk of bleeding. Of note is that once the varices have
Table 20-4. Bleeding Rates from Esophageal Varices during 24 Months of Follow-up (Data from reference 168.) Variable Child class A B C Size of varices Small Medium Large Red wale markings Absent Mild Moderate Severe Cherry-red spots Absent Mild Moderate Severe
364
% Who bled 17.0 31.1 38.9 18.1 28.6 48.9 19.1 32.9 39.3 80.0 23.0 32.0 40.0 55.0
been identified, the greatest risk of bleeding is during the following 6 months. Beyond 6 months the risk tends to diminish in those who have not bled.171 Why this should be is not clear, but if the patient is a liver transplant candidate and high-risk varices are identified, prevention of bleeding while the patient is awaiting transplantation may be a critical determinant of survival.
Risk Factors and Bleeding from Gastric Varices In one series gastric varices were identified in 20% of patients with esophageal varices and were most frequent in patients undergoing endoscopy for acute bleeding.172 Gastric varices also develop in patients who have undergone sclerotherapy for bleeding esophageal varices. They can be classified as gastroesophageal (GOV) or isolated gastric varices (IGV).172 Type 1 gastroesophageal varices (GOV1) are in continuity with the esophagus and extend only 2–5 cm below the gastroesophageal junction. Type 2 (GOV2) extend into the fundus of the stomach. Isolated gastric varices are thought to be suggestive of splenic vein thrombosis; however, they are usually a reflection of portal hypertension.173 Type 1 isolated gastric varices (IGV1) are located in the gastric fundus, whereas type 2 are present in the antrum or body of the stomach. The type of varix is an important risk factor for bleeding (Table 20-5). In the above study 75% of the gastric varices were GOV1, and therefore the number of patients with other types of gastric varices was small and the risk may be overestimated. Despite the small number of patients in some groups, the results suggest that certain gastric varices are at higher risk for bleeding. The finding of GOV2 or IGV1 type varices in a patient who has bled from esophageal varices, or has developed these gastric varices following successful sclerotherapy, should lead to consideration of treatment with a transjugular intrahepatic portal–systemic shunt or a surgical shunt, depending on the hepatic function of the patient, or injection therapy (see below).
Portal Hypertensive Gastropathy and Risk of Bleeding Portal hypertensive gastropathy is a complication of portal hypertension. Increase in gastric mucosal blood flow develops with an increase in portal pressure, and there are flow- and pressure-induced changes in the mucosal vessels of portal hypertensive animals and patients.174,175 The primacy of portal hypertension in the cause of the gastropathy is demonstrated by the cessation of bleeding from gastropathy following relief of portal hypertension.175 A history of previous sclerotherapy appears to be a risk factor for the development of gastropathy, but patients frequently have gastropathy in the absence of a history of sclerotherapy.175,176 Portal hypertensive
Table 20-5. Risk of Bleeding from Gastric Varices (Data from reference 172. See text for abbreviations.) Type of varix Total GOV1 GOV2 IGV1 IGV2
Risk of bleeding (%) 24 11.8 55 78 9
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
gastropathy is seen most commonly in those with advanced disease (87%) and less so in those with compensated cirrhosis (13%). Most cirrhotics have mild disease.175 The endoscopic appearance of portal hypertensive gastropathy is predictive of bleeding. Portal hypertensive gastropathy can be divided into mild (mosaic pattern) and severe (cherry-red spots and granular mucosa) forms.177 Mild or chronic bleeding was observed in 35% of patients with mild gastropathy versus 90% of patients with severe gastropathy. Overt bleeding occurred in 30% of those with mild gastropathy and 60% with severe gastropathy.177 Another system describes four basic lesions: mosaic-like pattern, red point lesions, cherry-red spots, and black-brown spots.178 Using this classification, gastropathy was present in 80% of patients and correlated with severity of disease, size of varices, and history of sclerotherapy.179 The severity of the gastropathy varied over time. Acute bleeding from the gastropathy occurred in 2.5% of patients and chronic bleeding in 10.8%.179 The administration of propranolol decreases gastric mucosal blood flow and is effective in reducing bleeding in patients with portal hypertensive gastropathy. Also, decompression of the portal system with a TIPS or liver transplantation are effective in stopping bleeding from portal hypertensive gastropathy.175 Another form of gastropathy, termed gastric antral vascular ectasia (GAVE), has been characterized in patients with portal hypertension. GAVE is localized to the antrum, is associated with poor hepatic function, and has a high risk of bleeding.175 Although observed in patients with severe portal hypertension, it is unclear how much the portal hypertension contributes to GAVE. For example, shunting procedures such as TIPS are ineffective in preventing rebleeding from GAVE, and the use of propranolol also lacks efficacy.175,180 In one patient bleeding resolved only following a liver transplant.181 GAVE is best managed with endoscopic therapy until the patient can undergo liver transplantation.
Predicting Bleeding from Other Sites in the Gastrointestinal Tract Although most patients with portal hypertension will bleed from the distal esophagus or stomach, a small percentage bleed from other lesions. Ectopic varices have been described in a number of locations, especially the colon. Bleeding from these varices is uncommon and there are no features of the varices that predict bleeding. The development of peristomal varices in patients with portal hypertension is an exception, in that bleeding from these varices is quite common. Of 10 patients with sclerosing cholangitis and portal hypertension who developed peristomal varices following proctocolectomy, all bled. The only effective management of this group of patients was surgical decompression.182 Portal colopathy also may be a source of occult or overt gastrointestinal (GI) blood loss. The colonic mucosa demonstrates multiple vascular ectasias on endoscopy, and edema and vascular congestion on biopsy.183 Portal colopathy is an uncommon cause of bleeding in patients with portal hypertension, but on occasion bleeding can be significant enough to require treatment.
Diagnosis of Bleeding Varices Patients bleeding from varices may have hematemesis, hematochezia, or melena. Although the onset is generally acute, patients
may bleed slowly and chronic iron-deficiency anemia may develop. The patient who needs treatment for a gastrointestinal hemorrhage resulting from varices may have features of liver disease, that is, ascites, spider nevi, jaundice, or hepatomegaly. However, many will have no external evidence that varices are the source of the hemorrhage. This will be especially true for those in whom liver disease is not the cause of their portal hypertension. Also, not all patients with cirrhosis who are bleeding will be bleeding from varices. These patients not infrequently bleed from gastritis, Mallory–Weiss tears, and ulcer disease. The initial step in diagnosing a variceal hemorrhage is to establish its location, that is, the upper or lower part of the gastrointestinal tract. This may be accomplished rapidly by inserting a nasogastric tube, followed by lavage of the stomach. A nasogastric tube should be passed in any patient with liver disease and gastrointestinal bleeding, even when hematochezia is present, because bleeding may be so brisk and transit so rapid that the blood is not altered while in the intestines. The failure to obtain blood from the nasogastric tube on entry into the stomach does not exclude varices as the source of hemorrhage, as frequently the bleeding is intermittent. The most accurate method for diagnosing bleeding varices and excluding other bleeding lesions is endoscopy (Figure 20-18). Studies have found that upper gastrointestinal tract endoscopy has an accuracy exceeding 90%.184,185 Endoscopy is an especially important diagnostic modality because gastropathy is a major cause of hemorrhage in patients who have varices. This lesion is missed by the standard gastrointestinal series.
Natural History of Varices and Screening The incidence of varices in patients with cirrhosis is about 50%, but in only 20% are the varices large and therefore most at risk for bleeding.186,187 Large varices are found most commonly in patients with advanced diseases. Varices develop at a rate of 4–12% per annum in patients with cirrhosis.165,188 Recent studies have also clarified the rate at which small varices increase in size. In one series of patients with small varices at the index endoscopy, at repeat endoscopy 16 months later 16% had resolved, 42% remained unchanged, and 42% had become larger.189 In another series 93 patients with small varices were followed and at 1 year 12% had developed large varices, and by 3 years varices had increased in size in 31%.165 The risk of bleeding from these small varices appears to be quite low: ~12% after 2 years.165 Once the varices become large then the risk of bleeding increases four- to fivefold (see above). With the development of therapies that prevent bleeding from varices that have never bled (see below), current recommendations are that patients with cirrhosis be screened for large varices and then placed on prophylactic treatment if large varices are found. Repeat endoscopy is recommended every 1–2 years to look for the development of large varices.190 As large varices are at greatest risk for bleeding, and as many patients with cirrhosis lack varices, are there any clinical findings that are predictive of the presence of varices? A number of different studies have examined this question.187 In one report a number of variables were examined to determine whether they were predictive of the presence of varices.191 The authors observed that a prolonged prothrombin time, a portal vein diameter greater than 13 mm, and a platelet count <100 000/ml were independent predictors of the presence of varices. If patients had
365
Section III. Clinical Consequences of Liver Disease
Figure 20-18. Endoscopic findings of active or recent bleeding from gastroesophageal varices. (A) A jet of blood from an esophageal varix is seen crossing the lumen of the esophagus and hitting the opposite wall. (B) A white nipple is evident on a large gastric varix that recently bled. Note the presence of portal hypertensive gastropathy in the surrounding mucosa.
A
B
all three variables the likelihood of varices was 90%, whereas in the absence of all three less than 10% of patients had varices. Thus, the presence of any of these features increases the likelihood of finding varices by endoscopy. Of greatest interest is identifying the patient with large varices, as they are most likely to bleed and are targets for preventive therapy. The presence of platelet counts of <88 000/ml and splenomegaly has been found to be predictive of the presence of large varices, and screening only patients with these features may be cost effective.186,187 However, other studies using model systems have challenged this suggestion and have recommended the use of b-blockers in all patients with cirrhosis and no screening for varices as the most cost-effective approach.192,193 Before we can embrace the latter approach further controlled trials comparing screening with no screening are required. Currently, cirrhotic patients with splenomegaly or thrombocytopenia are most likely to have large varices and should be screened. Also, if small or no varices are found, then rescreening at intervals of 2–3 years would appear to be most appropriate.
ACUTELY BLEEDING VARICES Non-endoscopic and Non-surgical Management General Measures When a patient needs treatment for an acute gastrointestinal hemorrhage, the initial step in management is to determine the status of the blood volume. The lack of hypotension or tachycardia in the supine patient is not adequate, as 25% of the blood volume may be lost without any change in these values. The blood pressure and pulse determinations therefore must be repeated with the patient in the upright position. Failure to realize that the patient has suffered significant blood loss early in the course of the illness may slow the tempo of management, so that clinical shock develops. The second step is to restore the patient’s blood volume. This should precede any diagnostic work-up. If endoscopy is performed on a hypotensive patient, the varices may be collapsed and undetectable. Restoration of the blood volume should be determined by improvement in the patient’s cardiovascular status, with the disappearance of hypotension and any orthostatic abnormalities. Transfusion to a predetermined hematocrit value irrespective of the status of the patient should be avoided, because overexpansion of the plasma volume will increase portal pressure and may precipitate
366
Table 20-6. Antibiotic Prophylaxis in Cirrhotics with Gastrointestinal Bleeding: Meta Analysis (Data from reference 197.) Groups
Free of infection Free SBP or bacteremia Survival
Mean % of patients Antibiotics
Controls
86 92 85
55 73 76
Odds ratio (CI intervals)
4.64 (3.19–6.75)* 3.86 (2.45–6.06)* 1.88 (1.22–2.89)*
*Difference is significant.
variceal hemorrhage (see above).147,150 Also, care must be exercised in the volume of fluid and amount of saline given to the patient with cirrhosis. The injudicious use of saline or glucose in water in these patients causes the rapid development of edema, ascites, or hyponatremia, because of the associated abnormalities of salt and water metabolism. The use of central venous pressure or pulmonary arterial wedged pressure, or both, to assess cardiovascular status is probably not necessary and may be misleading. Patients with cirrhosis have reduced systemic vascular resistance and increased pooling of blood in the splanchnic bed. Larger than usual volumes of fluid are therefore needed to cause changes in central venous pressure or pulmonary arterial wedged pressure, even when the patient has a normal plasma volume. The lack of hypotension and orthostatic changes in blood pressure, and an adequate urinary output, are sufficient indicators of a normal blood volume. The prognosis for patients with bleeding varices has improved over the last 20 years. Previously mortality rates at 6 weeks following the index bleed were 30–50%, but in a recent report had fallen to 17%.194 In a second series in-hospital mortality had fallen from 42.6% in 1980 to 14.5% in 2000.195 It is not entirely clear what accounts for the improved survival, but one important change in therapy is the use of prophylactic antibiotics in these patients. Bacterial infections are common in patients with cirrhosis, and especially in those who suffer GI tract bleeding.196,197 Because of the high risk of systemic infection antibiotics have been given prophylactically to patients with cirrhosis and gastrointestinal bleeding, and the results of a meta-analysis of this trial are given in Table 20-6. The antibiotics were given either orally or intravenously and used for 2–7 days. Use of antibiotics was associated with improvement in survival
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
and a significant reduction in the incidence of infection (Table 20-6). A subsequent meta-analysis of studies giving in-patients with cirrhosis prophylactic antibiotics for a number of reasons has also shown an improvement in survival and a lower incidence of bacterial infections.198 The use of antibiotics appears to reduce the risk of rebleeding as well.199 Given these findings, all patients with cirrhosis who have gastrointestinal bleeding should receive antibiotics prophylactically.
Treatment of Bleeding Esophageal Varices The treatment options for the management of bleeding varices are listed in Table 20-7. Table 20-8 gives the pooled odds ratios for pharmacologic therapy and endoscopic therapy.200–204 The use of TIPS and surgical shunts will be discussed in a later section.
Vasopressin and Nitroglycerin The use of vasopressin plus nitroglycerin will not be discussed in any detail as these drugs are no longer used in the management of acute variceal bleeding, having been replaced by drugs with fewer side effects and with similar efficacy (Table 20-8).
Table 20-7. Treatment Options for Acute Variceal Bleeding Medical Octreotide/somatostatin Vasopressin/nitrates Terlipressin Endoscopic therapy-sclerotherapy/banding Balloon tamponade* TIPS Surgical Emergency shunt Devascularization*
Somatostatin and Analogs Like vasopressin, somatostatin is a peptide hormone that acts as a vasoconstrictor when given in supraphysiologic doses. In contrast to vasopressin, the hemodynamic effects of somatostatin appear to be limited to the splanchnic circulation, and its use should be associated with fewer side effects. The effects of somatostatin on portal hemodynamics are unclear. Doses of 250 mg/h are associated with only transient falls in portal pressure, and only with doses of 500 mg/h are sustained reductions in portal pressure seen. The dose used therapeutically is a 250 mg bolus, followed by 250 mg/h given
Table 20-9. Meta-analysis of Terlipressin in Control of Acute Bleeding (Taken from reference 199.)
*Use limited to uncontrolled bleeding, failed TIPS, or with portal and splenic vein thrombosis.
Table 20-8. Pharmacologic and Endoscopic Treatment of Acutely Bleeding Esophageal Varices (if Pooled Odds Ratio (POR) <1.0 then First Therapy Favored over Second) (Data from references 200, 201 and 204.) A vs B Failure to control bleeding POR for A vs B Pharmacologic treatment Vasopressin vs no active Rx. Vasopressin + NTG vs vasopressin Terlipressin vs no active Rx. Terlipressin vs vasopressin Somatostatin vs vasopressin Octreotide vs vasopressin/terlipressin Endoscopic treatment Sclerotherapy vs pharmacologic# Sclerotherapy vs octreotide Banding vs sclerotherapy
Terlipressin. There are analogs of vasopressin that have different pharmacologic effects. The use of one of these analogs, triglycyl lysine–vasopressin (terlipressin), in the control of variceal hemorrhage has been investigated. This hormonogen is metabolized slowly to the active hormone (vasopressin). It has a long duration of action, no action on plasminogen activator, and less cardiac toxicity than does vasopressin. The dose is 2 mg intravenously every 4 hours, and with control of bleeding the dose can be reduced to 1 mg every 4 hours. Terlipressin administration leads to a fall in portal pressure and azygous blood flow.205 In a recent meta-analysis terlipressin was more effective than placebo but less effective than octreotide in controlling bleeding. The use of terlipressin was associated with an improved survival compared to placebo (Table 20-9).206 Terlipressin has also been given in combination with glyceryl trinitrate to patients who are being transported to hospital. Patients who received active treatment had better bleeding control and lower mortality than a placebo group.207
Death rate
0.22* 0.39* 0.31* 0.64* 0.68 0.58*
1.11 0.94 0.32* 1.48 0.92 —
0.61* 0.94 1.14
0.71 —
NTG, nitroglycerin. * Significantly better. #Pharmacologic includes vasopressin, somatostatin, and octreotide, and the data are combined.
Terlipressin vs Placebo Death Failure hemostasis Rebleeding Balloon tamponade Death Failure hemostasis Rebleeding Sclerotherapy Death Failure hemostasis Rebleeding Octreotide Death Failure hemostasis Somatostatin Death Failure hemostasis Rebleeding Vasopressin Death Failure hemostasis Rebleeding
Relative risk (95% CI)
Number needed to treat
0.66 (0.49–0.88)* 0.66 (0.53–0.82)* 0.99 (0.6–1.61)
8 (5–25) 6 (4–11) NA
0.94 (0.5–1.77) 1.61 (0.69–3.74) 0.84 (0.46–1.55)
NA NA NA
1.49 (0.88–2.52) 1.09 (0.62–1.90) 0.97 (0.62–1.54)
NA NA NA
1.2 (0.66–2.15) 1.62 (1.05–2.5)*
NA 7 (4–50) favors octreotide
0.95 (0.62–1.48) 1.05 (0.67–1.63) 1.2 (0.8–1.81)
NA NA NA
1.39 (0.96–2.0) 0.85 (0.65–1.13) 1.44 (0.93–2.25)
NA NA NA
*Significant difference NA, not available.
367
Section III. Clinical Consequences of Liver Disease
intravenously.205 The efficacy of somatostatin in the control of variceal bleeding is unclear. In a recent meta-analysis of somatostatin compared to control or no treatment, the use of somatostatin was associated with improved initial hemostasis but no effect on survival or rebleeding.208 The effect corresponded to a saving of 0.5 unit of blood per patient, and so somatostatin at currently recommended doses appears to have limited efficacy. Octreotide. This is a synthetic analog of somatostatin that has a much longer half-life and can therefore be given as a bolus or by infusion. The effect of octreotide on portal pressure is quite brief and its most consistent effect appears to be to damp the diurnal variation in portal pressure.205 In a recently published meta-analysis octreotide administration was more effective than no treatment or treatment with vasopressin/terlipressin. In addition, octreotide was as effective as sclerotherapy for the control of bleeding203 (Table 208). No mortality benefit was seen with octreotide compared to other forms of therapy. Another somatostatin analog (vapreotide) was administered for 5 days to patients acutely bleeding from varices. Patients received either the drug or a placebo preceding the index endoscopy. All patients underwent banding of their varices. Less bleeding at the index endoscopy and during the subsequent 5 days was observed in those who received vapreotide.210 In conclusion, somatostatin and its analogs or terlipressin appear to be effective in the control of variceal bleeding, and have few side effects but only a limited impact on patient survival. They are the preferred form of pharmacologic treatment for the patient bleeding acutely from varices.
Endoscopic Therapy This section will discuss the effectiveness of endoscopic therapy in controlling active bleeding from varices. In the section on primary prevention of variceal bleeding a more detailed discussion of endoscopic therapy is presented. The results of controlled trials that addressed the efficacy of sclerotherapy and banding in stopping acute hemorrhage are given in Table 20-8.200,201,204,211 Sclerotherapy is as good as balloon tamponade and better than treatment with vasopressin or terlipressin. Control of bleeding with somatostatin or octreotide is similar to that achieved with sclerotherapy. It is important to note that only 30–50% of patients are bleeding actively at the time of treatment, with the remainder having stopped bleeding.212 Thus, it is difficult to conclude that any of the therapies are 90% effective in controlling hemorrhage. Increasingly, esophageal band ligation has replaced sclerotherapy as the preferred form of management of esophageal varices. In a recently published meta-analysis, band ligation was as effective as sclerotherapy in the control of acute bleeding (Table 20-10).204 As with sclerotherapy, band ligation plus pharmacologic therapy appears to be more effective than monotherapy.210 Because of the less frequent side effects and the more rapid disappearance of varices associated with band ligation compared to sclerotherapy, the former is recommended. However, if acutely bleeding varices are discovered at endoscopy, hemostasis can be achieved with sclerotherapy, following by band ligation. If the patient is started on pharmacologic therapy preceding endoscopy, the likelihood of finding actively bleeding varices is reduced,211 which will reduce the need for sclerotherapy.
368
Table 20-10. Comparison of Endoscopic Sclerotherapy and Band Ligation of Esophageal Varices (Data from reference 204.) Outcome
POR
Conclusion
Control of active bleeding Variceal obliteration Rebleeding Rebleeding secondary varices Mortality secondary bleeding Esophageal strictures Bleeding esophageal ulcers Complications leading to death
1.14 1.24 0.52 0.47 0.49 0.1 0.54 0.47
Two therapies equal Two therapies equal Banding better Banding better Banding better Banding better Banding better Banding better
Balloon Tamponade of Varices Perhaps 10% of patients who present with bleeding varices will be refractory to pharmacologic and endoscopic management and will require a decompressive procedure. For this group of patients balloon tamponade of the varices is the only option available to control the bleeding before the decompressive procedure, most commonly a TIPS (see Chapter 17). Varices lie in the mucosa of the esophagus and stomach and are therefore susceptible to tamponade. Because blood to the esophageal varices flows through submucosal vessels in the fundus of the stomach, tamponade of either gastric or esophageal vessels should be effective in controlling hemorrhage. This is accomplished by placing a balloon in either the stomach or the esophagus. There are three types of tube available for this purpose. The Sengstaken–Blakemore is a triple-lumen tube that has a small gastric balloon (200–40 ml), an esophageal balloon, and a tube for aspirating gastric contents. An accessory nasogastric tube is placed above the esophageal balloon to remove saliva. Similar to the Sengstaken–Blakemore is the Minnesota tube. Its gastric balloon is larger (500 ml) than that of the Sengstaken–Blakemore tube. In addition, it has a lumen for esophageal aspiration. The Linton–Nicholas tube has a single large (600 ml) gastric balloon and lumina for aspiration of the stomach and esophagus. Balloon tamponade of esophageal varices is an effective form of therapy for the control of bleeding. In one series hemostasis for at least 24 hours was obtained in 88–91.5% of patients. Aspiration occurred in 10% of cases and was prevented by endotracheal intubation before tamponade.216 In a recent survey balloon tamponade was used by most physicians as a bridge to TIPS.217 We believe that complications can be minimized if balloon tamponade is performed by experienced personnel and all patients undergo endotracheal intubation prior to placement of the balloon. Its use is warranted in the management of a patient who is actively bleeding from varices and in whom both pharmacologic and endoscopic therapies have failed. Following control of the bleeding with the balloon, the patient should be referred for urgent TIPS placement.
Treatment of Bleeding Gastric Varices Bleeding gastric varices may be treated differently than esophageal varices. There are few data on the efficacy of pharmacologic therapy in the treatment of bleeding gastric varices. However, despite this lack most feel that pharmacologic therapy should be used in such patients. The use of standard sclerosing agents has not been shown to be effective in gastric variceal bleeding because of a high risk of rebleeding. The efficacy of butyl cyanoacrylate injection has been
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
compared to band ligation in the treatment of bleeding gastric varices. In a controlled trial initial hemostasis (no bleeding for 72 hours) was achieved in 87% of those receiving injection therapy, but in only 45% of those who were banded; the difference was significant. Obliteration of varices was achieved with both approaches, but there was significantly more rebleeding in the banding group – 54% versus 31% in the injection group.213 Butyl cyanoacrylate is not available in the US, but 2-octyl cyanoacrylate is available for skin closure. This agent has been used in 25 patients with gastric varices, some of whom had bled. Hemostasis was achieved in 100% of actively bleeding patients, and recurrent bleeding from gastric varices was seen in 4% of patients.214 The alternative to endoscopic therapy is performance of a TIPS or surgical shunt. TIPS is effective in the control of bleeding from gastric varices and in the prevention of rebleeding. In one series TIPS controlled bleeding from esophageal and gastric varices in 98% and 96% of patients, respectively. Rebleeding was observed in 24% of those who received a TIPS for bleeding esophageal varices and 29% of those who had bled from gastric varices.215 Butyl cyanoacrylate appears to have an efficacy similar to that of TIPS in the control of gastric variceal bleeding, but further studies using 2-octyl cyanoacrylate are needed before the latter’s role in the management of gastric variceal bleeding is determined.
Approach to Acutely Bleeding Varices The patient needs to be placed in an ICU and have their blood pressure, pulse and urine output normalized by infusing fluids and blood. However, they should not be transfused to a hematocrit of more than 25–28%, as bleeding may resume and the prognosis may worsen.218 Patients should receive antibiotics following the collection of cultures from blood, urine, and ascitic fluid, if present. They should be started on terlipressin or octreotide, and when hemodynamically stable should undergo an endoscopy. If actively bleeding esophageal varices are found, the patient should undergo sclerotherapy or band ligation. When the bleeding is from gastric varices and active hemorrhage is found, a Minnesota tube (preceded by tracheal intubation) should be placed and the patient sent for a TIPS. If it is available, the gastric varices may be injected with butyl cyanoacrylate as an alternative to TIPS placement. If the varices are not bleeding and no other source of bleeding is found, band ligation of esophageal varices is performed and the patient left on somatostatin or octreotide for 5 days.219 During this period a non-selective bblocker is started as well (see below). Should the patient rebleed in hospital, a TIPS should be considered. If the source of bleeding is thought to be gastric varices, then either a TIPS should be placed or the varices should be injected with butyl cyanoacrylate.
PREVENTION OF VARICEAL BLEEDING PRIMARY PROPHYLAXIS b-Blockers, Nitrates and Other Drugs, and Endoscopic Therapy Effects of b-Blockers and Nitrates on Portal Pressure
In 1980 Lebrec and colleagues220 reported that propranolol lowered portal pressure in patients with cirrhosis. Since that report numer-
ous controlled trials have been published showing that non-selective b-blockers reduce the risk of bleeding from esophageal varices.200,201 To be effective the HVPG must fall by at least 20% or to below 12 mmHg (see above section on hepatic vein pressure). As many as two-thirds of patients do not meet the above goals during treatment with b-blockers, which helps to explain why this treatment is not more effective.24,30,31 Despite these variable results, propranolol and other non-selective b-blockers do decrease portal pressure in many cirrhotic patients with portal hypertension. The effects of propranolol on the splanchnic circulation appear to be mediated via b1- and b2-adrenergic receptors in the heart and splanchnic circulation, respectively. Kroeger and Groszmann221 administered either propranolol or a selective b1- or b2-blocker to rats with portal hypertension. All three drugs caused a fall in portal venous flow and pressure, the greatest reductions being seen with propranolol. This fall was observed without changes in cardiac output. The enhanced effect of propranolol relative to more selective agents therefore reflects that propranolol blocks both b1-receptors (to decrease cardiac output) and b2-receptors. The latter effect causes splanchnic vasoconstriction by preventing vasodilation. Selective b1-blockers given to patients also are less effective in lowering portal pressure than is propranolol for a given level of b-blockade.222,223 When nitrates are given alone there is a variable but significant fall in the portal–hepatic vein gradient and a decline in azygous blood flow. If nitrates are given in combination with b-blockers, the falls in pressure are even greater, which provides the basis for the use of these drugs in preventing variceal bleeding.223
Therapy With the identification of agents that lower portal pressure chronically it was natural that they should be used in the management of patients with varices that have never bled. There have been thousands of patients enrolled in randomized controlled trials comparing b-blockers (propranolol and other non-selective b-blockers) with placebo and other agents in the prevention of variceal bleeding in patients who have never bled.200–202,224,226 Table 20-11 shows the results from an analysis of many of these studies. With the use of b-blockers there was a significant reduction in the risk of bleeding of about 40% compared to untreated controls or patients receiving a placebo, and a much smaller and not significant reduction in mortality. The calculated odds ratio is less than 1.0 (0.54), and therefore favors treatment with b-blockers.200 Overall mortality was unaffected, pooled odds ratio = 0.75. The uniformity of the results suggests that b-blockers are an effective form of therapy in the prevention of bleeding from varices. It is estimated that as many as 30 and as few as five patients need to be treated to prevent one episode of bleeding.225 This variability in number required to treat reflects the influence of the size of the varices and Child’s class on the risk of bleeding (Table 20-12). This efficacy is similar to results obtained in other prophylactic drug trials, but could be improved by documentation showing that the portal–hepatic vein gradient has fallen to less than 12 mmHg or by 20%.30,31 There are several contraindications to the use of b-blockers, including bradycardia, congestive heart failure, hypotension, and severe asthma. In one recent report 27% of patients with cirrhosis had a contraindication to their use.226 Frequent side effects from bblockers have also been observed in 10–21% of patients, with
369
Section III. Clinical Consequences of Liver Disease
Table 20-11. Primary Prophylaxis of Variceal Bleeding (Data taken from references 200, 201, 224, 226.) Bleeding rate Mortality (%) A vs B A vs B
A
vs
B
NAT NAT
vs vs
NAT IMN b-Blockers NAT VBL + b-blockers
vs vs vs vs vs
b-blockers (all patients) b-blockers (medium/ large varices) IMN b-blockers IMN + b-blocker sclerotherapy VBL 7/11
24/15* 30/14*
27/23 30/28
7/16 31/14 10.6/12.5 36/23#
15/22 37/42 89/88 37/30
NAT, no active therapy or placebo; IMN, isosorbide mononitrate; VBL, variceal band ligation. b-Blocker either propranolol or nadolol. *Difference between A and B significant. #Studies too heterogeneous to allow for conclusion of benefit. Bleeding rates and mortality are for 1 or 2 years of follow–up.
Table 20-12. Effect of Child’s Class and Risk of Bleeding on Cost of Treatment (Data from reference 225.) Child’s class/bleeding risk NNT ($ cost over 5 years)
Propranolol
Sclerotherapy
A–S A–L B–S B–L C–S C–L
30 ($77,000) 8 ($20,000) 16 ($41,000) 6 ($15,000) 10 ($26,000) 5 ($13,000)
50 ($411,000) 14 ($115,000) 23 ($189,000) 10 ($82,000) 16 ($131,000) 10 ($82,000)
Bleeding risk is based on the endoscopic appearance of the varices and is either small (S) or large (L) (see Table 20–4). NNT is the number needed to treat to prevent one episode of bleeding.
therapy being withdrawn in 0–8% because of side effects.227 The complications of b-blocker therapy have led to the search for other drugs.
Other Agents that Affect Pressure or Flow in Varices (Table 20-13) Isosorbide-5-mononitrate (IMN) is not an effective drug for the prevention of bleeding from varices when used as a single agent. In a placebo-controlled trial IMN was no more effective than placebo in the prevention of variceal bleeding, and the use of IMN alone cannot be recommended (Table 20-11).224 It is unclear whether bblockers plus nitrates are more effective than b-blockers alone in the prevention of the first bleed from varices. In the most recently completed large trial combination therapy was no more effective than monotherapy (Table 20-11),226 but not all investigators believe that combination therapy is no more efficacious than monotherapy.228 Given the current uncertainty and the increase in side effects with combination therapy, the use of b-blockers plus nitrates does not appear to be warranted outside controlled trials. There is increased sympathetic nervous system activity in cirrhosis, and it was thought that inhibition of this activity by clonidine, a centrally acting a2 agonist, might lower portal pressure. Clonidine was administered to 12 patients with cirrhosis and ascites, and
370
Table 20-13. Agents That May Reduce the Risk of Variceal Bleeding Clonidine Diuretics Verapamil Pentoxifyline Metoclopramide Endothelin receptor antagonists NO–URSO (NOX–1000)
hemodynamic measurements were made both before and after treatment.229 Clonidine reduced the pressure gradient between the portal vein and the hepatic vein by 20% and reduced azygos vein blood flow by 27%. However, the drug also caused significant falls in cardiac output and mean arterial pressure. Verapamil has been administered to patients with postnecrotic cirrhosis. Its use was associated with a fall of 14% in the portal vein–hepatic vein pressure gradient, with few systemic hemodynamic effects.230 Esophageal varices pass through the lower esophageal sphincter, and the pressure in the sphincter can be increased by metoclopramide and domperidone. Mastai and colleagues231 measured azygos vein blood flow in patients with cirrhosis before and after administration of these agents. Both caused a significant fall in azygos vein blood flow, with no associated change in the portal vein–hepatic vein pressure gradient. These drugs, therefore, may selectively reduce flow in esophageal varices by increasing the pressure in the lower esophageal sphincter. Serotonin can increase portal vascular resistance. Attempts have been made to lower portal pressure by administering a serotonin antagonist, ketanserin. Ketanserin given to 11 patients with cirrhosis reduced the portal vein–hepatic vein pressure gradient by 23% and the azygos vein blood flow by 26%, with few other effects on the systemic circulation.232 Most recently there has been intense interest in endothelins and nitric oxide as possible therapeutic targets.84 Endothelin levels are increased in cirrhosis, and when infused into animals led to a rise in portal pressure (see above discussion in Pathogenesis). Unfortunately, the administration of endothelin antagonists has not led to a fall in portal pressure in experimental animals.88 Nitric oxide is thought to be overproduced in the systemic circulation, thereby contributing to the hyperdynamic circulation. However, nitric oxide levels are thought to be reduced in the portal circulation, leading to a relative deficiency.89,90 The recent demonstrations that expressing nitric oxide in the livers of cirrhotic animals ameliorates their portal hypertension suggest that if one could target drugs such as nitrates to the portal circulation, this might be an effective form of therapy.94,95,233 The development of a drug in which nitric oxide has been linked to ursodeoxycholic acid (NCX-1000) has allowed for the targeted delivery of nitric oxide to the liver. In an initial study in animals the administration of NCX-1000 prevented the formation of ascites and reduced intrahepatic resistance in carbon tetrachloride-treated rats.234 In subsequent studies in cirrhotic rats NCX-1000 administration was associated with limited effects on portal pressure, and further studies in humans are awaited.235,236 If we are to extrapolate from the data obtained with b-blockers these
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
new drugs will have to lower the HVPG by at least 20% or to below 12 mmHg to reduce the risk of bleeding from varices. All will require controlled trials to establish their efficacy.
Injection Sclerotherapy of Varices Procedure The injection of varices with sclerosing agents was first reported in 1939. The method involves visualization of the varix during endoscopy, followed by injection of a sclerosing agent such as ethanolamine oleate or sodium morrhuate, into either the varix or the adjacent tissue. Each method of injection has its proponents. Given the difficulty in determining exactly at which point the sclerosing agent is injected, most patients probably receive a combination of both para- and intravariceal injections. A number of varices are injected during a session, and the patient returns at frequent intervals (1–4 weeks) for repeated injections. The goal of therapy is obliteration of the varices. Complications associated with sclerotherapy, although frequent, are not usually life-threatening. To a certain extent, the incidence of complications is a function of how vigorously certain complications are sought. For example, if patients undergo endoscopy within 24 hours of sclerotherapy 93% will be found to have esophageal ulcers. After 2 weeks this has decreased to 29%, and after 3 weeks only 12.5% will have esophageal ulcers.237 The risk of developing an esophageal stricture is low. Strictures are thought to develop because of the chemical esophagitis caused by the sclerosant and the dysmotility-related acid reflux. If the latter is an important factor, the use of antacids or H2-blockers after sclerotherapy would appear to be warranted. If strictures do develop, they can be treated easily by bougienage.238 Unfortunately, the use of agents that cause esophagitis appears to be necessary, as the agents that are most effective in eradicating varices are also the most toxic. Varices can be eradicated more quickly if a weekly rather than a 3-weekly schedule of injections is used. However, more frequent injections are associated with more esophagitis, and the benefits of this more aggressive approach are unclear.239 The reported frequency of bacteremia following sclerotherapy is variable, but in most cases it is transient and not associated with clinical evidence of septicemia.240,241
Treatment Prophylactic sclerotherapy has been evaluated in at least 21 trials, the results of which are given in Table 20-11. Although the data
suggested that there was less bleeding with sclerotherapy versus no active treatment, the heterogeneity of the trials made use of metaanalysis difficult.200–202 Perhaps more importantly, when the costeffectiveness of treatment with b-blockers was compared to that of sclerotherapy, b-blockers were shown to be cost-effective whereas sclerotherapy was not (Table 20-12).225 In conclusion, sclerotherapy is not considered appropriate treatment for the primary prevention of bleeding from esophageal varices.
Band Ligation of Esophageal Varices Procedure Band ligation of varices was first reported in 1989. The technique involves the placement of rubber bands around a portion of the varix containing esophageal mucosa. Initially, only a single band could be applied during each esophageal intubation, leading to the need for an overtube, the use of which was associated with esophageal perforation. Subsequently, devices have been developed that allow for the application of as many as 10 bands during a single intubation. Also, the device that holds the rubber bands is now made of clear plastic, which greatly increases one’s ability to see in the blood-filled esophagus. In general the complications are less severe than those seen following sclerotherapy, with less rebleeding, and band ligation should be performed in preference to sclerotherapy in the prevention of variceal bleeding (Table 20-10). At issue is whether variceal band ligation is better than b-blocker therapy in the primary prevention of variceal bleeding.242 There have been seven controlled trials published as complete papers, and two meta-analyses of these data and published abstracts (Table 2014).243–251 Both meta-analyses concluded that bleeding with variceal band ligation was significantly less frequent than with b-blocker therapy, and that more complications were seen with drug therapy.243,244 When only the published complete papers are analyzed the odds ratio favors band ligation (0.64, CI 0.4, 1.01) but not significantly. Unfortunately, all of the studies are not adequately powered to answer the question, and in those with the largest numbers of patients and the longest follow-up there was no difference in bleeding rates between the two groups.242 In addition, there is no cost analysis that favors one approach over the other. Lastly, the combination of b-blockers plus band ligation does not appear to be any more effective than band ligation alone, and cannot be recommended.252 Pending further studies, this authors believes that esophageal variceal band ligation as primary prophylaxis for
Table 20-14. Complete Published Studies Comparing Variceal Band Ligation to b-Blockers in Primary Prophylaxis of Variceal Bleeding Reference 250 248 246 249 245 247 251
Number of patients
% CP–C
% ETOH
30 100 152 89 62 110 31
18 16 13 33 23–26 33 19
17 20 49–53 22 10–13 65–70 19
Size varices
VBL repeat EGD
M–L M–L L M–L M–L M–L M–L
Q6W Q3–4W QW QW Q4–5W QW Q2–3W
b-Blockers % decrease pulse
F/U months
≥25% ≥25% ≥25% ≥25% ≥25% 160 mg# ≥25%
17.6 21.8 34 13–14 11–18 19.7 27
VB B vs L % 7 18 29 27 13 14 7
13 10 25 9* 0* 7 12
Mortality B vs L % NS 22 24 43 45 11 11 13 0* 25 23 20 37
B, b-blocker vs L, variceal band ligation; CP–C, Child–Pugh class C; NS, not stated; ETOH, cirrhosis due to alcohol; VB–cumulative rate of variceal bleeding; w, weeks.
371
Section III. Clinical Consequences of Liver Disease
bleeding varices should be limited to patients who are intolerant of b-blockers.
Approach to Primary Prophylaxis Patients with cirrhosis should be screened for esophageal varices as discussed above (see Natural history of varices and screening). If medium to large varices are found the patient should be placed on a non-selective b-blocker (propranolol or nadolol). This author prefers the latter because of its ease of administration. The dose of the drug should be titrated until the heart rate has fallen by 25% or is 55 beats/min. If the patient bleeds while on b-blocker therapy then band ligation should be performed and b-blockers continued, so as to help in the treatment of portal hypertensive gastropathy; their use may also slow the return of the varices (see below). In patients who are intolerant of b-blockers, band ligation should be performed.
PRE-PRIMARY PROPHYLAXIS Treatment of patients who have portal hypertension but who have no or only small varices in the hope of preventing their development or growth is termed pre-primary prophylaxis. Observations indicate that after 2–3 years of follow-up 16–44% of patients with cirrhosis who lack varices at the initial endoscopy will develop them, but few will bleed.253 One study published in preliminary form254 has examined whether a non-selective b-blocker can prevent the formation of varices in patients with portal hypertension. There was no reduction in the rate of formation of varices in the treated compared to the control group. The other alternative is to try to slow the progression of varices. In one series about 25% of patients with small varices developed large ones during a 2-year follow-up period, but only 12% bled.165 More recently the impact of nadolol compared to a placebo on the growth of varices has been examined.255 In this series 161 patients with small varices were followed for a mean of 36 months. Nine patients in the nadolol group and 29 in the placebo group had growth of their varices. The risk of bleeding was also reduced in the nadolol group but there was no improvement in survival.255 Thus, in one report the use of b-blockers to prevent the development of varices was without benefit, whereas in the other b-blockers prevented the growth of small varices. It is difficult to explain these different results, but only 50% of cirrhotics develop varices and it is possible that the two populations were intrinsically
different and thus different results were obtained. Based on these two small studies b-blockers cannot be recommended in the cirrhotic patient to prevent or slow the growth of varices, especially as a significant number of patients will experience side effects and the risk of bleeding in this group is quite small. Further studies are required to resolve this issue.
SECONDARY PROPHYLAXIS b-Blockers have also been used to prevent recurrent bleeding in patients who have bled at least once from varices. Table 20-15 gives the results of a meta-analysis of published controlled trials comparing b-blockers with placebo. Treatment with b-blockers reduced the risk of recurrent bleeding by about 35% and reduced the death rate by 29% (both significant). The studies were heterogeneous in their design as to when the b-blocker therapy was initiated and the severity of the liver disease. Although these are important variables and should have been considered in the design of all of the studies,256 bblockers do appear to be an effective form of therapy for reducing the risk of bleeding from varices that have bled at least once. Endoscopic sclerotherapy has been used in many controlled trials to prevent rebleeding from varices. As shown in Table 20-15, compared to no active treatment, sclerotherapy reduced the risk of rebleeding and death. Compared to b-blockers, sclerotherapy was somewhat more effective in preventing bleeding, with no difference in survival.200,201 There was significant heterogeneity in the studies and most feel that they are comparable forms of treatment.200 When sclerotherapy has been compared to sclerotherapy plus b-blockers, there is a significant reduction in the risk of bleeding with combination therapy and no decrease in mortality (Table 22-15). Endoscopic variceal ligation has been compared to sclerotherapy in a number of trials. As shown in Table 20-10, control of acutely bleeding varices and variceal obliteration are the same with these two forms of therapy.204 Band ligation is clearly superior to sclerotherapy as regards the risk of rebleeding, mortality secondary to bleeding, and complications. Endoscopic variceal band ligation should be used in place of sclerotherapy in the prevention of rebleeding from esophageal varices. Whether band ligation is superior to b-blockers in the prevention of rebleeding is unclear. However, combination therapy with b-blockers plus band ligation appears to be superior to band ligation alone. In two controlled trials band ligation was compared to such combination therapy.257,258 After
Table 20-15. b-Blockers and Sclerotherapy in the Secondary Prophylaxis of Varices (Data from references 200, 201 and 257.) Treatment
Rebleeding rate vs
B
A/B (%)
POR
A/B (%)
POR
NAT NAT Beta Blockers EST EVL
vs vs vs vs vs
Beta Blockers EST EST EST + beta blockers EVL + beta blockers
63/41 66/53 57/49 40/30 47/23*
0.4* 0.62* 0.66# 0.66
24/17 55/48 35/32 15/13 32/17
0.7* 0.77* 0.96 0.83
NAT, no active treatment; EST, endoscopic sclerotherapy; POR, pooled odds ratio. If POR is <1 then treatment B is better than A. #Significant heterogeneity. *Treatment B significantly better than treatment A.
372
Death rate
A
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
1–2 years of follow-up, rebleeding and recurrence were significantly less in the patients who received combination therapy, with no impact on survival.
TRANSJUGULAR INTRAHEPATIC PORTAL–SYSTEMIC SHUNT (TIPS) (see Chapter 17) The greatest interest in TIPS has been as a form of treatment for secondary prophylaxis of varices. Between 1988 and 1999, at least 11 controlled trials of 811 patients were published that compared TIPS to endoscopic treatment in the prevention of rebleeding from esophageal varices.259 Most of the studies used sclerotherapy, but in two variceal ligation was used. The results from a recent metaanalysis are shown in Table 20-16 and are strikingly similar to what has been observed when surgical shunts are compared to endoscopic therapy (see below). The incidence of bleeding is less in the group who received a TIPS, but death rates are unaffected and the incidence of hepatic encephalopathy is increased in those who received TIPS. Death from variceal bleeding was more frequent in the endoscopic group, and an average of 18% of patients were crossed over to TIPS because of refractory variceal bleeding. Death from other causes was significantly more common in the TIPS patients.259 A cost analysis of TIPS versus endoscopic sclerotherapy has also been performed.260 In this study the cumulative cost of patients who had received a TIPS was significantly greater at 12 and 18 months compared to those who had undergone sclerotherapy. Although rebleeding was more frequent in the sclerotherapy group, the number of days in hospital for the two groups was similar because of the increase in encephalopathy and shunt insufficiency in the TIPS group. TIPS has also been compared to drug therapy in the prevention of rebleeding. In this trial rebleeding occurred in 13% of TIPS-treated patients and 39% of those receiving propranolol plus IMN. More encephalopathy was seen in those receiving a TIPS, and survival was the same in both groups. Cost was significantly greater for the TIPS group.261 Based on these studies, TIPS should not be considered as a primary therapy to prevent rebleeding but instead should be used in those who have failed medical therapy.
SURGICAL MANAGEMENT OF PORTAL HYPERTENSION The surgical procedures for managing portal hypertension fall into three broad groups: decompressive shunts, total, partial, or selec-
tive; devascularization procedures; and liver transplantation. Devascularization procedures are rarely used and will not be discussed. The role of surgical shunts in the prevention of rebleeding will be reviewed, and liver transplantation is discussed in Chapters 49–53.
DECOMPRESSIVE SHUNTS These fall into three different types, which have the common feature of decompression of gastroesophageal varices. The difference between the three types of shunt depends on whether the portal and superior mesenteric venous systems are decompressed. Total portal–systemic shunts decompress all parts of the portal venous system and the cirrhotic liver is deprived of all portal flow. Partial shunts provide partial reduction of portal hypertension in all segments of the portal venous system, with the goal of decompressing varices sufficiently to prevent rebleeding but not totally depriving the liver of all its portal flow. Selective shunts decompress the spleen and gastroesophageal junction, but the portal and superior mesenteric venous systems are not decompressed: portal hypertension is maintained as a means of maintaining prograde portal flow to the cirrhotic liver. The following sections will look at each of these approaches in more detail. Total and selective portal–systemic shunts can be made in a variety of ways, as illustrated in Figures 20-19 and 20-20. Portacaval shunt may be either end-to-side or side-to-side. Pathophysiologically these are different, in that the former does not decompress the high pressure of the obstructed liver sinusoids whereas the latter does. In an end-to-side shunt the hepatic end of the portal vein is ligated so that the only outflow from the sinusoids remains the hepatic veins. In contrast, with a side-to-side portacaval shunt greater than 10 mm in diameter, the hepatic end of the portal vein serves as an outflow tract from the obstructed sinusoids, thus decompressing both the hepatic and the splanchnic beds. Thus, a side-to-side shunt alleviates ascites, whereas an end-to-side portacaval shunt may not.262 The most commonly used selective shunt is the distal splenorenal shunt (DSRS) that selectively decompresses the spleen and gastroesophageal junction to the left renal vein, while maintaining portal hypertension in the superior mesenteric and portal veins to maintain portal flow to the liver.263 The second type of selective shunt is the coronary caval shunt, as popularized by Inokuchi,264 which anastomoses the left gastric vein to the inferior vena cava. This is usually accompanied by a splenectomy. Both selective and total shunts have been effective in the prevention of rebleeding (Table 20-17). Rebleeding rates are 5–7% and hospital mortality is less than 10%. Unfortunately, there is a high incidence of hepatic encephalopathy, and mortality is unaffected by the approach used to manage these patients. With the development
Table 20-16. Efficacy of TIPS vs Endoscopic Therapy in the Prevention of Rebleeding from Esophageal Varices (Data taken from reference 259.) Variceal rebleeding TIPS 19%
EST 47%
OR 3.8*
Death NNT 4
TIPS 27%
EST 27%
Encephalopathy OR 0.97
TIPS 34%
EST 19%
OR 0.43*
EST, endoscopic sclerotherapy; NNT, number needed to treat to prevent one episode of rebleeding; OR, odds ratio. OR greater than 1 favors TIPS and less than one favors EST. *Significant differences between two forms of therapy.
373
Section III. Clinical Consequences of Liver Disease
Normal A
B
Coronary vein Portal vein Inferior vena cava
Inferior Splenic vein mesenteric vein
Superior mesenteric vein
C
D
E
Splenectomy
Figure 20-19. Total portal–systemic shunts. (A) Normal. (B), End-to-side portacaval shunt ligates the portal vein at the liver, so it does not decompress the liver sinusoids. (C) Side-to-side portacaval shunt (>10 mm diameter) leaves the portal vein intact, so this shunt decompresses the liver as well as the varices. (D), Mesocaval shunt requires an interposition graft – either PTFE (Gore-Tex) or autogenous jugular vein – to bridge the distance from the SMV to the IVC. Physiologically, this is the same as a side-to-side PCS. (E), Central splenorenal shunt entails splenectomy and anastomosis of the central end of the splenic vein to the left renal vein. This is a side-to-side total shunt.
374
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
Splenectomy A
B
Table 20-17. Consecutive Series of Major Surgical Methods in Prevention of Rebleeding Published in the 1990s
No. of patients Hospital mortality Rebleeding Encephalopathy Survival (5 yrs)
Total shunt265
DSRS266
Devasc267
378 6.4% 7.7% 45% 57%
147 4.8% 5.6% 17% 77% (3 yr)
549 4.9% 6%* — 66%
DSRS, distal splenorenal shunt; Devasc., devascularization procedures. *Rebleeding as cause of death.
of TIPS the role of surgical shunts in the prevention of rebleeding has become unclear. Two studies have compared surgical shunts to TIPS in the prevention of rebleeding in patients who have failed medical therapy. One trial has compared TIPS to a surgical 8 mm portocaval H graft interposition shunt.268 This study entered ‘all comers’ and included 46% Child–Pugh class C patients, and showed a benefit to surgical shunt compared to TIPS. However, the patients were not truly randomized but were entered into the study sequentially. The most recent study has been published in preliminary form and patients were randomized to DSRS or TIPS. All patients had failed medical therapy and had Child–Pugh class A or B cirrhosis. Rebleeding rates were 5.5% in the DSRS group and 10.5% in the TIPS group (difference not significant). Survival rates and incidence of hepatic encephalopathy was the same in both groups. The patients undergoing TIPS required significantly more interventions than those who had a DSRS, but bare not coated stents were used in the study. The authors believe these are equivalent forms of therapy but that patients receiving a TIPS will require more surveillance to ensure its patency.269
Approach to Secondary Prophylaxis Once the patient has been diagnosed as having bled from varices a decision needs to be reached as to whether to use b-blockers alone, or to use both endoscopic and pharmacologic therapies. The advantage of b-blockers is the proven safety and efficacy of a single agent that is inexpensive. The need for repeated endoscopies is eliminated. However, most patients are started on endoscopic therapy at the time of their index bleed. In addition, banding is clearly superior to
Figure 20-20. Selective variceal decompression. (A) Distal splenorenal shunt decompresses the gastroesophageal junction and the spleen to control bleeding. The splanchnic and portal venous systems maintain hypertension and flow to the liver. (B) Coronary caval shunt decompresses the gastroesophageal junction via the left gastric vein. Splenectomy is also performed. Portal hypertension, and prograde portal flow, is maintained.
sclerotherapy, with a lower incidence of rebleeding. In controlled trials sclerotherapy is at least as good if not better than b-blockers in preventing rebleeding (Table 20-15). Thus, banding should be better than b-blockers. In addition, combination therapy – endoscopic plus pharmacologic – is better in preventing rebleeding than either therapy alone (Table 20-15).200,201,257,258 Lastly, combined therapy appears to slow the recurrence rate for varices.257 This author believes that the preferable approach to prevent rebleeding is a combination of endoscopic variceal ligation and a non-selective b-blocker. Once the varices have been obliterated, the b-blocker is continued to reduce the risk of bleeding from portal-hypertensive gastropathy as well as to slow the rate at which the varices reform. It the patient fails therapy and rebleeds, then the choice is a surgical shunt or a TIPS. Based on the recently completed controlled trial, both TIPS and DSRS appear to be effective in patients with good liver function.269 In patients with more advanced disease TIPS is a better choice, followed by referral to a liver transplant center. A similar choice needs to be made if the patient has bled from gastric varices.
REFERENCES 1. Greenway CV, Stark RD. Hepatic vascular bed. Physiol Rev 1971;51:23–65. 2. Chiandussi L, Greco E, Sardi G, et al. Estimation of hepatic arterial and portal venous blood flow by direct catheterization of the vena porta through the umbilical cord in man. Acta Heptato-splenol 1968;15:166–171. 3. Schenk WG, McDonald J, McDonald K, et al. Direct measurement of hepatic blood flow in surgical patients with related observations on hepatic flow dynamics in experimental animals. Ann Surg 1962;156:463–469. 4. Lautt WW, Greenway C. Conceptual review of the hepatic vascular bed. Hepatology 1987;7:952–963. 5. Granger D, Miller T, Allen R, et al. Permselectivity of cat liver blood–lymph barrier to endogenous macromolecules. Gastroenterology 1979;77:103–109. 6. Henderson JM, Gilmore GT, Mackay GJ, et al. Hemodynamics during liver transplantation: the interactions between cardiac output and portal venous and hepatic arterial flows. Hepatology 1992;16:715–718. 7. Henderson JM, Mackay GJ, Lumsden AB, et al. The effect of liver denervation on hepatic hemodynamics during hypovolemic shock in swine. Hepatology 1992;15:130–133.
375
Section III. Clinical Consequences of Liver Disease
8. Reynolds TB, Redeker AG, Geller HM. Wedged hepatic vein pressure. Am J Med 1957;22:341–350. 9. Rousselot L, Moreno A, Panke W. Studies on portal hypertension. IV. The clinical and physiopathologic significance of self-established (non-surgical) portal venous shunts. Ann Surg 1959;150:384–410. 10. Takashi M, Igarashi M, Hino S, et al. Portal hemodynamics in chronic portal systemic encephalopathy. Angiographic study in seven patients. J Hepatol 1985;1:467–476. 11. Ricci R, Lee K, Greenberger N. Chronic gastrointestinal bleeding from ileal varices after total proctocolectomy for ulcerative colitis: correction by mesocaval shunt. Gastroenterology 1980;78:1053–1058. 12. Wilson S, Stone R, Christie J, et al. Massive lower gastrointestinal bleeding from intestinal varices. Arch Surg 1979;114:1158–1161. 13. Weisner RH, LaRusso NF, Dozois RR, et al. Peristomal varices after proctocolectomy in patients with primary sclerosing cholangitis. Gastroenterology 1986;90:316–322. 14. Jacobs DM, Bubrick MP, Onstad GP, et al. The relationship of hemorrhoids to portal hypertension. Dis Colon Rectum 1980;23:567–569. 15. Westaby S, Wilkinson S, Williams R. Spleen size and portal hypertension in cirrhosis. Digestion 1978;17:63–68. 16. El-Khishen MA, Henderson JM, Millikan WJ, et al. Splenectomy is contraindicated for thrombocytopenia secondary to portal hypertension. Surg Gynecol Obstet 1985;160:233–238. 17. Cales P, Zabotto B, Meskens C, et al. Gastroesophageal endoscopic features in cirrhosis. Gastroenterology 1990;98:156–162. 18. Garcia-Tsao G, Groszmann RJ, Fisher RL, et al. Portal pressure, presence of gastroesophageal varices and variceal bleeding. Hepatology 1985;5:419–524. 19. Bandoh T, Mitarai Y, Kitano S, et al. Clinical significance of esophageal variceal pressure in patients with esophageal varices. J Hepatol 1994;21:326–331. 20. McCormack TT, Rose JD, Smith PM, et al. Perforating veins and blood flow in esophageal varices. Lancet 1983;2:1441– 1444. 21. Warren W, Restrepo J, Respress J, et al. The importance of hemodynamic studies in management of portal hypertension. Ann Surg 1963;158:387–400. 22. Escorsell A, Garcia-Pagan JC, Bosch J. Assessment of portal hypertension in humans. Clin Liver Dis 2001;5:575–589. 23. Lafortune M, Marleau D, Breton G, et al. Portal venous system measurements in portal hypertension. Radiology 1984;151:27–30. 24. Groszmann R, Wongcharatrawee S. The hepatic venous pressure gradient: anything worth doing should be done right. Hepatology 2004;39:280–282. 25. Keiding S, Vilstrup H. Intrahepatic heterogeneity of hepatic venous pressure gradient in cirrhosis. Scand J Gastroenterol 2002;37:960–964. 26. Boyer TD, Triger D, Horisawa M, et al. Direct transhepatic measurement of portal vein pressure using a thin needle: comparison with wedged hepatic vein pressure. Gastroenterology 1977;72:584–589. 27. Perello A, Escorsell A, Bru C, et al. Wedged hepatic venous pressure adequately reflects portal pressure in hepatitis C virusrelated cirrhosis. Hepatology 1999;30:1393–1397. 28. Viallet A, Marleau D, Huet M, et al. Hemodynamic evaluation of patients with intrahepatic portal hypertension: relationship between bleeding varices and the portohepatic gradient. Gastroenterology 1975;69:1297–1300. 29. Monescillo A, Martinez-Lagares F, Ruiz-del-Arbol L, et al. Influence of portal hypertensión and its early decompression by TIPS placement on the outcome of variceal bleeding. Hepatology 2004;40:793–801.
376
30. Boyer TD. Changing clinical practice with measurements of portal pressure. Hepatology 2004;39:284–285. 31. Thalheimer U, Mela M, Patch D, et al. Targeting portal pressure measurements: A critical reappraisal. Hepatology 2004;39:286–290. 32. Merkel C, Bolognesi M, Sacerdoti D, et al. The hemodynamic response to medical treatment of portal hypertension as a predictor of clinical effectiveness in the primary prophylaxis of variceal bleeding in cirrhosis. Hepatology 2000;32:930–934. 33. Imperiale TF, Chalasani N, Klein RW. Measuring the hemodynamic response to primary pharmacophylaxis of variceal bleeding: A cost-effectiveness analysis. Am J Gastroenterol 2003;98:2742–2750. 34. Targownik L, Spiegel B, Dulai G, Karsan H, Gralnek I. The costeffectiveness of hepatic venous pressure gradient monitoring in the prevention of recurrent variceal hemorrhage. Am J Gastroenterol 2004;99:1306–1315. 35. Boyer TD Transjugular intrahepatic portosystemic shunt: current status. Gastroenterology 2003;124:1700–1710. 36. Bosch J, Bordas JM, Rigau J, et al. Noninvasive measurement of the pressure of esophageal varices using an endoscopic gauge: comparison with measurements by variceal puncture in patients undergoing endoscopic sclerotherapy. Hepatology 1986;6:667–672. 37. Mosimann R. Non-aggressive assessment of portal hypertension using endoscopic measurement of variceal pressure. Preliminary report. Am J Surg 1982;143:212–214. 38. Dawson J, Gertsch P, Mosimann F, et al. Endoscopic variceal pressure measurements: response to isosorbide dinitrate. Gut 1985;26:843–847. 39. Rigau J, Bosch J, Bordas J, et al. Endoscopic measurement of variceal pressure in cirrhosis. Correlation with protal pressure and variceal hemorrhage. Gastroenterology 1989;96: 873–880. 40. Nevens F, Bustami R, Scheys I, et al. Variceal pressure is a factor predicting the risk of a first variceal bleeding: A prospective cohort study in cirrhotic patients. Hepatology 1998;27:15–19. 41. Bosch J, Mastai R, Kravetz D, et al. Hemodynamic evaluation of the patient with portal hypertension. Semin Liver Dis 1986;6:309–317. 42. Nevens F, Van Steenbergen W, Yap SH, et al. Assessment of variceal pressure by non-invasive endoscopic registration: A placebo controlled evaluation of the effect of terlipressin and octreotide. Gut 1996;38:129–134. 43. Redeker A, Geller H, Reynolds T. Hepatic wedge pressure, blood flow, vascular resistance, and oxygen consumption in cirrhosis before and after end-to-side portacaval shunt. J Clin Invest 1958;37:606–618. 44. Moreno AH, Burchell A, Rousselot L, et al. Portal blood flow in cirrhosis of the liver. J Clin Invest 1967;46:436–445. 45. Sarper R, Fajman WA, Rypins EB, et al. A noninvasive method for measuring portal venous/total hepatic blood flow by hepatosplenic radionuclide angiography. Radiology 1981;141:179–184. 46. Burchell A, Moreno A, Panke W, et al. Hepatic artery flow improvement after portacaval shunt: a single hemodynamic clinical correlate. Ann Surg 1976;184:289–302. 47. Burns P, Taylor K, Blei AT. Doppler flowmetry and portal hypertension. Gastroenterology 1987;92:824–826. 48. Bosch J. Diagnostic tools in portal hypertension. In: Schmid R, Bianchi L, Gerok W, Maier K-P, eds. Extrahepatic manifestations of liver disease. Boston: Kluwer Academic, 1993: 41– 49. Zuidema G, Gaisford W, Abel M, et al. Segmental portal arterialization of canine liver. Surgery 1975;53:689–698. 50. Siderys H, Judd D, Herendeen T. The experimental production of elevated portal pressure by increasing portal vein flow. Surg Gynecol Obstet 1965;120:514–516.
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
51. Schilling J, McKee F. Late follow-up on experimental hepatic–portal arteriovenous fistula. Surg Forum 1953;4:392–397. 52. Donovan AJ, Reynolds TB, Mikkelsen W, et al. Systemic–portal arteriovenous fistulas: pathologic and hemodynamic observations in two patients. Surgery 1969;66:474–482. 53. Benoit J, Womack WA, Hernandez L, et al. ‘Forward’ and ‘backward’ flow mechanisms of portal hypertension. Gastroenterology 1985;89:1092–1096. 54. Benoit JN, Granger N. Splanchnic hemodynamics in chronic portal hypertension. Semin Liver Dis 1986;6:287–298. 55. Groszmann RJ, Vorobioff J, Riley E. Splanchnic hemodynamics in portal-hypertensive rats: measurement with a-labeled microspheres. Am J Physiol 1982;242:G156–G160. 56. Chezmar JL, Redvanly RD, Nelson RC, et al. Persistence of portal systemic collaterals and splenomegaly on CT after orthotopic liver transplantation. AJR Am J Roentgenol 1992;159:317–320. 57. Moreno A, Burchell A, Reddy R, et al. The hemodynamics of portal hypertension revisited: determinants and significance of occluded portal pressures. Surgery 1975;77:167–179. 58. Moriyasu F, Nishida O, Nobuyuki B, et al. Measurement of portal vascular resistance in patients with portal hypertension. Gastroenterology 1986;90:710–717. 59. Sikuler E, Kravetz D, Groszmann RJ. Evolution of portal hypertension and mechanisms involved in its maintenance in a rat model. Am J Physiol 1985;248:G618–G625. 60. Vorobioff J, Bredfeldt JE, Groszmann RJ. Hyperdynamic circulation in portal-hypertensive rat model: a primary factor for maintenance of chronic portal hypertension. Am J Physiol 1983;244:G52–G57. 61. Vorobioff J, Bredfeldt JE, Groszmann RJ. Increased blood flow through the portal system in cirrhotic rats. Gastroenterology 1984;87:1120–1126. 62. Sikuler E, Groszmann RJ. Hemodynamic studies in long- and short-term portal hypertensive rats: the relation to systemic glucagon levels. Hepatology 1986;6:414–418. 63. Witte C, Witte M, Renert W, et al. Splenic artery ligation in selected patients with hepatic cirrhosis and in Sprague–Dawley rats. Surg Gynecol Obstet 1976;142:1–12. 64. Kroeger RJ, Groszmann RJ. Increased portal venous resistance hinders portal pressure reduction during the administration of Badrenergic blocking agents in a portal hypertensive model. Hepatology 1985;5:97–101. 65. Paulsen AW, Klintmalm BG. Direct measurement of hepatic blood flow in native and transplanted organs, with accompanying systemic hemodynamics. Hepatology 1992;16:100–111. 66. Morgan JS, Groszmann RJ, Rojkind M, et al. Hemodynamic mechanisms of emerging portal hypertension caused by schistosomiasis in the hamster. Hepatology 1990;11:98–104. 67. Zimmon D, Kessler R. Effect of portal venous blood flow diversion on portal pressure. J Clin Invest 1980;65:1388–1397. 68. Pomier-Layrargues G, Kusielewicz D, Willems B, et al. Presinusoidal portal hypertension in non-alcoholic cirrhosis. Hepatology 1985;5:415–418. 69. Kelty RH, Baggenstoss AH, Butt HR. The relation of the regenerated liver nodule to the vascular bed in cirrhosis. Gastroenterology 1950;15:285–295. 70. Baldus WP, Hoffbauer FW. Vascular changes in the cirrhotic liver as studied by injection technic. Am J Dig Dis 1963;8:689– 700. 71. Shibayama Y, Nakata K. Localization of increased hepatic vascular resistance in liver cirrhosis. Hepatology 1985;5:643–648. 72. Orrego H, Medline A, Blendis LM, et al. Collagenisation of the Disse space in alcoholic liver disease. Gut 1979;20:673–679. 73. Edmondson H, Peters R, Reynolds T, et al. Sclerosing hyaline necrosis of the liver in the chronic alcoholic. Ann Intern Med 1963;59:646–673.
74. Orrego H, Blendis LM, Crossley IR, et al. Correlation of intrahepatic pressure with collagen in the Disse space and hepatomegaly in humans and in the rat. Gastroenterology 1981;80:546–556. 75. Mikakawa H, Iida S, Leo MA, et al. Pathogenesis of precirrhotic portal hypertension in alcohol-fed baboons. Gastroenterology 1985;88:143–150. 76. Wanless I, Wong F, Blendis L, et al. Hepatic and portal vein thrombosis in cirrhosis: possible role in development of parenchymal extinction and portal hypertension. Hepatology 1995;21:1238–1247. 77. Reynolds TB, Geller H, Kuzma O, et al. Spontaneous decrease in portal pressure with clinical improvement in cirrhosis. N Engl J Med 1960;263:734–739. 78. Colman JC, Britton RS, Orrego H, et al. Relation between osmotically induced hepatocyte enlargement and portal hypertension. Am J Physiol 1983;245:G382–G387. 79. Blendis LM, Orrego H, Crossley IR, et al. The role of hepatocyte enlargement and hepatic pressure in cirrhotic and noncirrhotic alcoholic liver disease. Hepatology 1982;2: 539–546. 80. Vidins EI, Britton RS, Medline A, et al. Sinusoidal caliber in alcoholic and nonalcoholic liver disease: diagnostic and pathogenic implications. Hepatology 1985;5:408–414. 81. Krogsgaard K, Gluud C, Henriksen JH, et al. Correlation between liver morphology and portal pressure in alcoholic liver disease. Hepatology 1984;4:699–703. 82. Blendis LM. Hepatocyte swelling and portal hypertension. J Hepatol 1992;15:4–5. 83. Rockey DC, Boyles JK, Gabbiani G, et al. Rat lipocytes express smooth muscle actin upon activation in vivo and in culture. J Submicrosc Cytol Pathol 1992;24:193–203. 84. Rockey D. The cellular pathogenesis of portal hypertension: stellate cell contractility, endothelin, and nitric oxide. Hepatology 1997;25:2–5. 85. Asbert M, Gines A, Gines P, et al. Circulating levels of endothelin in cirrhosis. Gastroenterology 1993;104:1485– 1491. 86. Pinzani M, Gentilini P. Biology of hepatic stellate cells and their possible relevance in the pathogenesis of portal hypertension in cirrhosis. Semin Liver Dis 1999;19:397–410. 87. Rockey DC, Weisinger RA. Endothelin-induced contractility of stellate cells from normal and cirrhotic rat liver: Implications for regulation of portal pressure and resistance. Hepatology 1996;24:233–240. 88. Poo J-L, Jimenez W, Munoz RM, et al. Chronic blockade of endothelin receptors in cirrhotic rats: hepatic and hemodynamic effects. Gastroenterology 1999;116:161–167. 89. Wiest R, Groszmann R. Nitric oxide and portal hypertension: Its role in the regulation of intrahepatic and splanchnic vascular resistance. Sem Liver Dis 19:411–426, 1999. 90. Martin P-Y, Gines P, Schrier R. Nitric oxide as a mediator of hemodynamic abnormalities and sodium and water retention in cirrhosis. N Engl J Med 1998;339:533–541. 91. Matsumoto A, Ogura K, Hirata Y, et al. Increased nitric oxide in the exhaled air of patients with decompensated liver cirrhosis. Ann Intern Med 1995;123:110–113. 92. Rolla G, Brussino L, Colagrande P, et al. Exhaled nitric oxide and impaired oxygenation in cirrhotic patients before and after liver transplantation. Ann Intern Med 1998;129:375–378. 93. Iwakiri Y, Cadelina G, Sessa WC, Groszmann RJ. Mice with targeted deletion of eNOS develop hyperdynamic circulation associated with portal hypertension. Am J Physiol 2002;283:G1074–G1081. 94. Rockey, DC, Chung JJ. Reduced nitric oxide production by endothelial cells in cirrhotic rat liver: endothelial cell dysfunction in portal hypertension. Gastroenterology 1998;114:344–351.
377
Section III. Clinical Consequences of Liver Disease
95. Yu Q, Shao R, Qian HS, et al. Gene transfer of the neuronal NO synthase isoform to cirrhotic rat liver ameliorates portal hypertension. J Clin Invest 2000;105:741–748. 96. Groszmann RJ, Atterbury CE. The pathophysiology of portal hypertension: a basis for classification. Semin Liver Dis 1982;2:177–186. 97. Matsumoto S, Mori H, Yamada Y, et al. Intrahepatic portohepatic venous shunts in Rendu–Osler–Weber disease: imaging demonstration. Eur J Radiol 2004;14:592–596. 98. Yamagami T, Nakamura T, Nishimura T. Portal hypertension secondary to spontaneous arterio-portal venous fistulas: transcatheter arterial embolization with n-butyl cyanoacrylate and microcoils. Cardiovasc Intervent Radiology 2000;23:400–402. 99. Lane M, Jeffrey RB Jr, Katz DS. Spontaneous intrahepatic vascular shunts. AJR am J Roentgenol 2000;174:125–131. 100. Kirchgatterer A, Baldinger C, Lenglinger F, et al. Portal hypertension due to traumatic hepatic arterioportal fistula: report of successful embolization. Eur J Gastroenterol Hepatol 1998;10:1057–1059. 101. Van Way C, Crane J, Riddell D, et al. Arteriovenous fistula in the portal circulation. Surgery 1971;70:876–890. 102. Lukie BD, Card RT. Portal hypertension complicating myelofibrosis: reversal following splenectomy. Can Med Assoc J 1977;117:771–771. 103. Tefferi A, Mesa R, Nagorney DM, et al. Splenectomy in myelofibrosis with myeloid metaplasia: a single-institution experience with 223 patients. Blood 2001;95:2226–2233. 104. Sullivan A, Rheinlander H, Weintraub L. Esophageal varices in angiogenic myeloid metaplasia: disappearance after splenectomy. Gastroenterology 1974;66:429–432. 105. Rosenbaum D, Murphy G, Swisher S. Hemodynamic studies on the portal circulation in myeloid metaplasia. Am J Med 1966;41:360–368. 106. Sarin SK, Kapoor D. Non-cirrhotic portal fibrosis: Current concepts and management. J Gastroenterol Hepatol 2002;17:526–534. 107. Ibarrola C, Colina F. Clinicopathological features of nine cases on non-cirrhotic portal hypertension: current definitions and criteria are inadequate. Histopathology 2003;42:251–264. 108. Ohba KT, Zen Y, Sato Y, et al. A comparative histological and morphometric study of vascular changes in idiopathic portal hypertension and alcoholic fibrosis/cirrhosis. Histopathology 2003;43:55–61. 109. Ohnishi K, Saito M, Sato S, et al. Portal hemodynamics in idiopathic portal hypertension (Banti’s syndrome): comparison with chronic persistent hepatitis and normal subjects. Gastroenterology 1987;92:751–758. 110. Okuda K, Kono K, Ohnishi K, et al. Clinical study of eighty-six cases of idiopathic portal hypertension and comparison with cirrhosis with splenomegaly. Gastroenterology 1984;86:600–610. 111. Morris J, Schmid M, Newman S, et al. Arsenic and non-cirrhotic portal hypertension. Gastroenterology 1974;64:86–94. 112. Huet P-M, Guillaume E, Cote J, et al. Non-cirrhotic presinusoidal portal hypertension associated with chronic arsenical intoxication. Gastroenterology 1975;68:1270–1277. 113. Creech J, Johnson M. Angiosarcoma of liver in the manufacture of polyvinyl chloride. J Occup Med 1974;16:150–151. 114. Berk P, Martin J, Young R, et al. Vinyl chloride-associated liver disease. Ann Intern Med 1976;84:717–731. 115. Blendis L, Smith P, Lawrie B, et al. Portal hypertension in vinyl chloride monomer workers. Gastroenterology 1978;75:206–211. 116. Primentel J, Menezes A. Liver disease in vineyard sprayers. Gastroenterology 1977;72:275. 117. Raia S, Mies S, Macedo AL. Portal hypertension in schistosomiasis. Clin Gastroenterol 1985;14:57–82. 118. Coutinho A. Hemodynamic studies of portal hypertension in schistosomiasis. Am J Med 1968;44:547–556.
378
119. Siqueira ES, Rohr MR, Libera ED, et al. Band ligation or sclerotherapy as endoscopic threatment for oesophageal varices in schistosomotic patients: results of a randomized study. HBP Surg 1998;11:27–32. 120. Coutinho A. Hemodynamic studies of portal hypertension in schistosomiasis. Am J Med 1968;44:547–556. 121. Paes-Alves C, Alves A, Abreu W, et al. Hepatic artery hypertrophy and sinusoidal hypertension in advanced schistosomiasis. Gastroenterology 1977;72:126–128. 122. Edmondson H, Peters R, Reynolds T, et al. Sclerosing hyaline necrosis of the liver in the chronic alcoholic. Ann Intern Med 1963;59:646–673. 123. Reynolds T, Hidemura R, Michel H, et al. Portal hypertension without cirrhosis in alcoholic liver disease. Ann Intern Med 1969;70:497–506. 124. Leevy C. Fatty liver: a study of 270 patients with biopsy-proven fatty liver and a review of the literature. Medicine 1962;41:249–276. 125. Moriyasu F, Nishida O, Ban N, et al. Measurement of portal vascular resistance in patients with portal hypertension. Gastroenterology 1986;90:710–717. 126. Reynolds TD, Ito S, Iwatsuki S. Measurement of portal pressure and its clinical application. Am J Med 1970;49:649–657. 127. Kew M, Varma R, Dos Santos H, et al. Portal hypertension in primary biliary cirrhosis. Gut 1971;12:830–834. 128. Shibayama Y, Nakata K. Localization of increased hepatic vascular resistance in liver cirrhosis. Hepatology 1985;5:643–648. 129. Hyun BH, Singer EP, Sharrett RH. Esophageal varices and metastatic carcinoma of liver: a report of three cases and review of the literature. Arch Pathol 1976;77:292–298. 130. Zeitoun D, Brancatelli G, Colombat M, et al. Congenital hepatic fibrosis: CT findings in 18 adults. Radiology 2004;231: 109–116. 131. Poddar U, Thapa BR, Vashishta RK, et al. Congenital hepatic fibrosis in Indian children. J Gastroenterol Hepatol 1999;14:1192–1196. 132. Freeze H. New diagnosis and treatment of congenital hepatic fibrosis. J Parenteral Gen Nutr 1999;29:104–106. 133. Rudzki C, Ishak Z, Zimmerman H. Chronic intrahepatic cholestasis in sarcoidosis. Am J Med 1975;59:373–387. 134. Maddrey W, Johns C, Boitnott J, et al. Sarcoidosis and chronic hepatic disease: a clinical and pathologic study of 20 patients. Medicine 1970;49:375–395. 135. Stern R, Stevens D, Boat T, et al. Symptomatic hepatic diseases in cystic fibrosis: incidence, course, and outcome of portal systemic shunting. Gastroenterology 1976;70:645–649. 136. Schuster S, Shwachmann H, Toyawa W, et al. The management of portal hypertension in cystic fibrosis. J Pediatr Surg 1977;12:201–206. 137. Lindbald A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis. Hepatology 1999;30:1151–1158. 138. Wanless IR, Godwin TA, Allen F, et al. Nodular regenerative hyperplasia of the liver in hematologic disorders: a possible response to obliterative portal venopathy. Medicine 1980;59:367–379. 139. Trenschel G, Schubert A, Dries V, Benz-Bohm G. Nodular regenerative hyperplasia of the liver: case report of a 13-year-old girl and review of the literature. Pediatr Radiol 2000;30:64–68. 140. Solis-Herruzo JA, Vidal JV, Colina F, et al. Nodular regenerative hyperplasia of the liver associated with the toxic oil syndrome: report of five cases. Hepatology 1986;6:687–693. 141. Dumortier J, Boillot O, Chevallier M. et al. Familial occurrence of nodular regenerative hyperplasia of the liver: a report on three families. Gut 1999;45:289–294. 142. Tanabe U, Sueyoshi K, Yoshinaka H, et al. Hepatic hemodynamics in a patient with nodular regenerative hyperplasia. Am J Gastroenterology 1996;91:1012–1015.
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
143. Shimamatsu K, Wanless R. Role of ischemia in causing apoptosis, atrophy and nodular hyperplasia in human liver. Hepatology 1997;26:343–350. 144. Wanless IR, Lentz JS, Roberts EA. Partial nodular transformation of liver in an adult with persistent ductus venosus. Review with hypothesis on pathogenesis. Arch Pathol Lab Med 1985;109:427–432. 145. Sherlock S, Feldman CA, Moran B, et al. Partial nodular transformation of the liver with portal hypertension. Am J Med 1966;40:195–203. 146. Classen M, Elster K, Pesch HJ, et al. Portal hypertension caused by partial nodular transformation of the liver. Gut 1970;11:245–249. 147. Boyer JL, Chatterjee C, Iber F, et al. Effects of plasma-volume expansion on portal hypertension. N Engl J Med 1966;275:750–755. 148. Preger L, Maddison F, Won G, et al. The erection and detumescence of esophageal varices. Am J Roentgenol Radium Ther Nucl Med 1969;107:77–80. 149. Kravetz D, Sikuler E, Groszmann RJ. Splanchnic and systemic hemodynamics in portal hypertensive rats during hemorrhage and blood volume restitution. Gastroenterology 1986;90:1232–1240. 150. Kravetz D, Bosch J, Arderiu M, et al. Hemodynamic effects of blood volume restitution following hemorrhage in rats with portal hypertension due to cirrhosis of the liver: influence of the extent of portal systemic shunting. Hepatology 1989;9: 808–814. 151. Okumura H, Aramaki T, Katsuta Y, et al. Reduction in hepatic venous pressure gradient as a consequence of volume contraction due to chronic administration of spironolactone in patients with cirrhosis and no ascites. Am J Gastroenterol 1991;86: 46–52. 152. Iwatsuki S, Reynolds T. Effects of increased intra-abdominal pressure on hepatic hemodynamics in patients with chronic liver disease and portal hypertension. Gastroenterology 1973;65:294299. 153. Luca A, Feu F, Garcia-Pagan JC, et al. Favorable effects of total paracentesis on splanchnic hemodynamics in cirrhotic patients with tense ascites. Hepatology 1994;20:30–33. 154. Kravetz D, Romero G, Argonz J, et al. Total volume paracentesis decreases variceal pressure, size and wall tension. Hepatology 1997;25:59–62. 155. Escorsell A, Gines A, Llach J, et al. Increasing intra-abdominal pressure increases pressure, volume, and wall tension in esophageal varices. Hepatology 2002;36:936–940. 156. Eckardt V, Grace N, Kantrowitz P. Does lower esophageal sphincter incompetency contribute to esophageal variceal hemorrhage? Gastroenterology 1976;71:185–189. 157. Eckardt V, Grace N. Gastroesophageal reflux and bleeding esophageal varices. Gastroenterology 1979;76:39–42. 158. MacDougall BRD, Williams R. A controlled clinical trial of cimetidine in the recurrence of variceal hemorrhage: implications about the pathogenesis of hemorrhage. Hepatology 1983;3:69–73. 159. Spence RAJ, Sloan JM, Johnston GW. Oesophagitis in patients undergoing oesophageal transection for varices – a histological study. Br J Surg 1983;70:332–334. 160. Ponce J, Froufe A, dela Morena A, et al. Morphometric study of the esophageal mucosa in patients with variceal bleeding. Hepatology 1981;1:641–646. 161. Polio J, Groszmann RJ. Hemodynamic factors involved in the development and rupture of esophageal varices: a pathophysiologic approach to treatment. Semin Liver Dis 1986;6:318–331. 162. Noda T. Angioarchitectural study of esophageal varices with special reference to variceal rupture. Virchows Arch [A] 1984;404:381–392.
163. Lebrec D, DeFleury P, Bernard R, et al. Portal hypertension, size of esophageal varices and risk of gastrointestinal bleeding in alcoholic cirrhosis. Gastroenterology 1980;79:1139–1144. 164. Rector WG, Reynolds TB. Risk factors for hemorrhage from esophageal varices and acute gastric erosions. Clin Gastroenterol 1985;14:139–153. 165. Merli M, Nicolini G, Angeloni S, et al. Incidence and natural history of small esophageal varices in cirrhotic patients. J Hepatology 2003;38:266–272. 166. Boyer TD. Portal hypertensive hemorrhage: Pathogenesis and risk factors. Semin Gastrointest Dis 1995;6:125–133. 167. Japanese Research Society for Portal Hypertension. The general rules for recording endoscopic findings of esophageal varices. Jpn J Surg 1980;10:84–87. 168. North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices. Prediction of the first variceal hemorrhage in patients with cirrhosis of the liver and esophageal varices. N Engl J Med 1988;319:983–989. 169. Zoli M, Merkel C, Magalotti D, et al. Evaluation of a new endoscopic index to predict first bleeding from the upper gastrointestinal tract in patients with cirrhosis. Hepatology 1996;24:1047–1052. 170. Rigo GP, Merighi A, Chahin NJ, et al. A prospective study of the ability of three endoscopic classifications to predict hemorrhage from esophageal varices. Gastrointest Endosc 1992;38:425–429. 171. Siringo S, Bolondi L, Gaiani S, et al. Timing of the first variceal hemorrhage in cirrhotic patients: prospective evaluation of Doppler flowmetry, endoscopy and clinical parameters. Hepatology 1994;20:66–73. 172. Sarin SK, Lahoti D, Saxena S, et al. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients. Hepatology 1992;16:1343–1349. 173. Levine MS, Kieu K, Rubesin SC, et al. Isolated gastric varices: splenic vein obstruction or portal hypertension? Gastrointest Radiol 1990;15:188–192. 174. Albillos A, Colombato LA, Enriquez R, et al. Sequence of morphological and hemodynamic changes of gastric microvessels in portal hypertension. Gastroenterology 1992;102:2066–2070. 175. Burak KW, Lee SS, Beck PL. Portal hypertensive gastropathy and gastric antral vascular ectasia (GAVE) syndrome. Gut 2001;49:866–872. 176. Tanoue K, Hashizume M, Wada H, et al. Effects of endoscopic injection sclerotherapy on portal hypertensive gastropathy: a prospective study. Gastrointest Endosc 1992;38:582–585. 177. D’Amico G, Montalbano L, Traina M, et al. Natural history of congestive gastropathy in cirrhosis. Gastroenterology 1990;99:1558–1564. 178. Carpinelli L, Primignami M, Preatoni P, et al. Portal hypertensive gastropathy: reproducibility of a classification, prevalence of elementary lesions, sensitivity and specificity in the diagnosis of cirrhosis of the liver. A NIEC multicenter study. Ital J Gastroenterol Hepatology 1997;29:533–540. 179. Primignani M, Carpinelli M, Preatoni P, et al. Natural history of portal hypertensive gastropathy in patients with liver cirrhosis. Gastroenterology 2000;119:181–187. 180. Payen J-L, Cales P, Voigt J-J, et al. Severe portal hypertensive gastropathy and antral vascular ectasia are distinct entities in patients with cirrhosis. Gastroenterology 1995;108:138–144. 181. Spahr L, Villeneuve JP, Dufresne M, et al. Gastric antral vascular ectasia in cirrhotic patients with portal hypertension. Gut 1999;44:739–742. 182. Wiesner RH, LaRusso NF, Dozois RR, et al. Peristomal varices after proctocolectomy in patients with primary sclerosing cholangitis. Gastroenterology 1986;90:316–322.
379
Section III. Clinical Consequences of Liver Disease
183. Kozarek RA, Botoman VA, Bradfeldt JE, et al. Portal colopathy: prospective study of colonoscopy in patients with portal hypertension. Gastroenterology 1991;101:1192–1197. 184. Dagradi A, Arguello J, Weingarten Z. Failure of endoscopy to establish a source of upper gastrointestinal bleeding. Am J Gastroenterol 1979;72:395–402. 185. Dagradi A, Ruiz R, Weingarten Z. Influence of emergency endoscopy on the management and outcome of patients with upper gastrointestinal hemorrhage. Am J Gastroenterol 1979;72:403–415. 186. Chalasani N, Imperiale TF, Ismail A, et al. Predictors of large esophageal varices in patients with cirrhosis. Am J Gastroenterol 1999;94:3285–3291. 187. Pankaj R, Kowdley K. Prediction of varices in patients with cirrhosis: A high-stakes numbers game? J Clin Gastroenterol 2002;34:4–5. 188. Boyer TD. Natural history of portal hypertension. Clin Liver Dis 1997;1:31–44. 189. Cales P, Desmorat H, Vinel JP, et al. Incidence of large esophageal varices in patients with cirrhosis. Application to prophylaxis of first bleeding. Gut 1990;31:1298–1302. 190. Grace ND Diagnosis and treatment of gastrointestinal bleeding secondary to portal hypertension. American College of Gastroenterology Practice Parameters Committee. Am J Gastroenterol 1997;92:1081–109 191. Schepis F, Camma C, Niceforo D, et al. Which patients with cirrhosis should undergo endoscopic screening for esophageal varices detection? Hepatology 2001;33:333–338. 192. Arguedas MR, Heudebert GR, Eloubeidi MA, Abrams GA, Fallon MB. Cost-effectiveness of screening, surveillance, and primary prophylaxis strategies for esophageal varices. Am J Gastroenterol 2002;97:2441–2452. 193. Spiegel BMR, Targownik L, Dulai GS, Karsan HA, Gralnek IM. Endoscopic screening for esophageal varices in cirrhosis: Is it ever cost effective? Hepatology 2003;37:366–377. 194. Chalasani N, Kahi C, Francois F, et al. Improved patient survival after acute variceal bleeding: a multicenter, cohort study. Am J Gastroenterology 2003; 98:653–659. 195. Carbonell N, Pauwels A, Serfaty L, et al. Improved survival after variceal bleeding in patients with cirrhosis over the past two decades. Hepatology 2004;40:652–659. 196. Fernandez J, Navasa M, Gomez J, et al. Bacterial infections in cirrhosis: Epidemiological changes with invasive procedures and norfloxacin prophylaxis. Hepatology 2002;35:140–148. 197. Bernard B, Grange J-D, Khac EN, et al. Antibiotic prophylaxis for the prevention of bacterial infections in cirrhotic patients with gastrointestinal bleeding: A meta-analysis. Hepatology 1999;29:1655–1661. 198. Soares-Weiser K, Brezus N, Tur-Kaspa RT, et al. Antibiotic prophylaxis of bacterial infections in cirrhotic inpatients: a metaanalysis of randomized controlled trials. Scand J Gastroenterol 2003;38:193–200. 199. Hou M-C, Lin H-C, Liu T-T, et al. Antibiotic prophylaxis after endoscopic therapy prevents rebleeding in acute variceal hemorrhage: A randomized trial. Hepatology 200439;: 746–753. 200. D’Amico G, Pagliaro L, Bosch J. The treatment of portal hypertension: a meta-analytic review. Hepatology 22:332–354,1995 201. D’Amico G, Pagliaro L, Bosch J. Pharmacological treatment of portal hypertension: an evidence-based approach. Semin Liver Dis 1999;19:475–505. 202. Luketic V, Sanyal A. Esophageal varices. I. Clinical presentation, medical therapy, and endoscopic therapy. Gastroenterol Clin North Am 2000;29:337–385. 203. Corley DA, Cello JP, Adkisson W, et al. Octreotide for acute esophageal variceal bleeding. Gastroenterology 2001;120:946–954.
380
204. Laine L, Cook DJ. Endoscopic ligation compared with sclerotherapy for treatment of esophageal variceal bleeding. Ann Intern Med 1995;123:280–287. 205. Nevens F. Review article: a critical comparison of drug therapies in currently used therapeutic strategies for variceal bleeding. Aliment Pharmacol Ther 2004;20(Suppl. 3):18–22. 206. Ioannou GN, Doust J, Rockey DC. Systematic review: terlipressin in acute oesophageal variceal haemorrhage. Aliment Pharmacol Ther 2003;17:53–64. 207. Levacher S, Letoumelin P, Pateron D. et al. Early administration of terlipressin plus glyceryl trinitrate to control active upper gastrointestinal bleeding in cirrhotic patients. Lancet 1995;346:865–868. 208. Gotzsche PC, Hrobjartsson A. Somatostatin analogues for acute bleeding oesophageal varices. Cochrane Database Syst Rev 2:2005. 209. Corley DA, Cello JP, Adkisson W, et al. Octreotide for acute esophageal variceal bleeding: A meta-analysis. Gastroenterology 2001;120:946–954. 210. Cales P, Masliah C, Bernard B, et al. Early administration of vapreotide for variceal bleeding in patients with cirrhosis. N Engl J Med 2001;344:23–28. 211. DeFranchis R, Primignani M. Endoscopic treatments of portal hypertension. Semin Liver Dis 1999;19:439–455. 212. Sung JJ, Chung SC, Lai C-W, et al. Octreotide infusion or emergency sclerotherapy for variceal haemorrhage. Lancet 1993;342:637–641. 213. Lo G-H, Lai K-H, Cheng J-S. et al. A prospective, randomized trial of butyl cyanoacrylate injection versus band ligation in the management of bleeding gastric varices. Hepatology 2001;33:1060–1064. 214. Rengstorff DS, Binmoeller KF. A pilot study of 2-octyl cyanoacrylate injection for treatment of gastric fundal varices in humans. Gastrointest Endosc 2004;9:553–558. 215. Chau TN, Patch D, Chan YW, et al. ‘Salvage’ transjugular intrahepatic portosystemic shunts: gastric fundal compared with esophageal variceal bleeding. Gastroenterology 1998;114:981–987. 216. Panes J, Teres J, Posch J, Rodes J. Efficacy of balloon tamponade in treatment of bleeding gastric and esophageal varices. Results in 151 consecutive episodes. Dig Dis Sci 1988;33:454–459. 217. Bajaj JS, Ananthakrishnan A, Saeian K. Survey of attitudes of AASLD members toward balloon tamponade. Hepatology 2005;41:1435–1436. 218. Castaneda B, Morales J, Lionetti R, et al. Effects of blood volume restitution following a portal hypertensive-related bleeding in anesthetized cirrhotic rats. Hepatology 2001;33:821–825. 219. Banares R, Albillow A, Rincon D, et al. Endoscopic treatment versus endoscopic plus pharmacologic treatment for acute variceal bleeding: A meta-analysis. Hepatology 2002;35:609–615s. 220. Lebrec D, Nouel O, Corbic M, et al. Propranolol. A medical treatment for portal hypertension? Lancet 2:180–182, 1980. 221. Kroeger RJ, Groszmann RJ. Effect of selective blockade of b2adrenergic receptors on portal and systemic hemodynamics in a portal hypertensive rat model. Gastroenterology 1985;88:896–900. 222. Hillon P, Lebrec D, Munoz C, et al. Comparison of the effects of a cardioselective and nonselective b-blocker on portal hypertension in patients with cirrhosis. Hepatology 1982;2:528–531. 223. Garcia-Pagan JC, Escorsell A, Mortinho E, et al. Influence of pharmacological agents on portal hemodynamics: basis for its use in the treatment of portal hypertension. Semin Liver Dis 1999;19:427–438.
Chapter 20 PORTAL HYPERTENSION AND BLEEDING ESOPHAGEAL VARICES
224. Garcia-Pagan JC, Villanueva C, Vila MC, et al. Isosorbide mononitrate in the prevention of first variceal bleed in patients who cannot receive b-blockers. Gastroenterology 2001;121:908–914. 225. Teran JC, Imperiale TF, Mullen KD, et al. Primary prophylaxis of variceal bleeding in cirrhosis. A cost-effectiveness analysis. Gastroenterology 1997;112:473–482. 226. Garcia-Pagan JC, Morrialls R, Banares R, et al. Propranolol plus placebo versus propranolol plus isosorbide-5-mononitrate in the prevention of a first variceal bleed: a double-blind RCT. Hepatology 2003;37:1260–1266. 227. Boyer TD. The patient who cannot receive beta-blockers. In: Groszmann RJ, Bosch J, eds. Portal hypertension in the 21st century. Boston: Kluwer Academic, 2004: 301–307. 228. Merkel C. Nonselective b-blockers plus nitrates in portal hypertension: An open question. Hepatology 2003;37:1254–1255. 229. Moreau R, Lee S, Hadengue A. Hemodynamic effects of clonidine-induced decrease in sympathetic tone in patients with cirrhosis. Hepatology 1987;7:149–154. 230. Kong CW, Lay CS, Tsai YT, et al. The hemodynamic effect of verapamil on portal hypertension in patients with postnecrotic cirrhosis. Hepatology 1986;6:423–426. 231. Mastai R, Grande L, Bosch J, et al. Effects of metoclopramide and domperidone on azygous venous blood flow in patients with cirrhosis and portal hypertension. Hepatology 1986;6:1244–1247. 232. Hadengue A, Lee SS, Moreau R, et al. Beneficial hemodynamic effects of ketanserin in patients with cirrhosis: possible role of serotonergic mechanisms in portal hypertension. Hepatology 1987;7:644–647. 233. Groszmann RJ, Abraldes JC. Portal hypertension: From bench to bedside. J Clin Gastroenterol 2005;39:S125–S130. 234. Fiorucci S, Antonelli E, Morelli O, et al. NCX-1000, a NOreleasing derivative of ursodeoxycholic acid, selectively delivers NO to the liver and protects against development of portal hypertension. PNAS 98:8897–8902, 2001. 235. Loureiro-Silva MR, Cadelina GW, Iwakiri Y, et al. A liverspecific nitric oxide donor improves the intra-hepatic vascular response to both portal blood flow increase and methoxamine in cirrhotic rats. J Hepatol 2003;39:940–946. 236. Fiorucci S, Antonelli E, Brancaleone V, et al. NCX-1000, a nitric oxide-releasing derivative of ursodeoxycholic acid, ameliorates portal hypertension and lowers norepinephrine-induced intrahepatic resistance in the isolated and perfused rat liver. J Hepatol 2003;39:932–939. 237. Sarin SK, Nanda R, Vij JC, et al. Oesophageal ulceration after sclerotherapy – a complication or an accompaniment? Endoscopy 1986;18:44–45s. 238. Reilly JJ Jr, Schade RR, Van Thiel DS. Esophageal function after injection sclerotherapy: pathogenesis of esophageal stricture. Am J Surg 1984;147:85–88. 239. Westaby D, Melai WM, MacDougall BRD, et al. Injection sclerotherapy for oesophageal varices: a prospective randomised trial of different treatment schedules. Gut 1984;25:129–132. 240. Low DE, Shoenut JP, Kennedy JK, et al. Infectious complications of endoscopic injection sclerotherapy. Arch Intern Med 1986;146:569–571. 241. Sauerbruch T, Holl J, Ruckdeschel G, et al. Bacteraemia associated with endoscopic sclerotherapy of oesophageal varices. Endoscopy 1985;17:170–172. 242. Boyer TD Primary prophylaxis for variceal bleeding. Are we there yet? Gastroenterology 2005;128:1120–1122. 243. Imperiale T, Chalasani N. A meta-analysis of endoscopic variceal ligation for primary prophylaxis of esophageal variceal bleeding. Hepatology 2001;3:802–807. 244. Khuroo MS, Khuroo NS, Farahat KLC, et al. Meta-analysis: endoscopic variceal ligation for primary prophylaxis of
245.
246.
247.
248.
249.
250.
251.
252.
253.
254.
255.
256.
257.
258.
259.
260.
261.
oesophageal variceal bleeding. Aliment Pharmacol Ther 2005;21:347–361. Jutabha R, Jensen DM, Martin P, et al. Randomized study comparing banding and propranolol to prevent initial variceal hemorrhage in cirrhotics with high risk esophageal varices. Gastroenterology 2005;128:870–881. Schepke M, Kleber G, Nurnberg D, et al. Ligation versus propranolol for the primary prophylaxis of variceal bleeding in cirrhosis. Hepatology 2004;40:65–72. Lui HF, Stanley AJ, Forrest EH, et al. Primary prophylaxis of variceal hemorrhage: a randomized controlled trial comparing band ligation, propranolol, and isosorbide mononitrate [see comment]. Gastroenterology 2002;123:735–744. Lo GH, Chen WC, Chen MH, et al. Endoscopic ligation vs. nadolol in the prevention of first variceal bleeding in patients with cirrhosis. Gastrointest Endosc 2004;59:333–338. Sarin SK, Lamba GS, Kumar M, Misra A, Murthy NS. Comparison of endoscopic ligation and propranolol for the primary prevention of variceal bleeding. N Engl J Med 1999;340:988–993. De BK, Ghoshal UC, Das T, Santra A, Biswas PK. Endoscopic variceal ligation for primary prophylaxis of oesophageal variceal bleed: preliminary report of a randomized controlled trial. J Gastroenterol Hepatol 1999;14:220–224. Thuluvath PJ, Maheshwari A, Jagannath S, Arepally A. A randomized controlled trial of b–blockers versus endoscopic band ligation for primary prophylaxis: a large sample size is required to show a difference in bleeding rates. Dig Dis Sci 2005;50:407–410. Sarin SK, Wadhawan M, Agarwal SR. et al. Endoscopic variceal ligation plus propranolol versus endoscopic variceal ligation alone in primary prophylaxis of variceal bleedin. Am J Gastroenterol s2005;100:797–804. Merkel C, Bolognesi M, Gatta A. The cirrhotic patient with no varices and with small varices. In: Groszmann RJ, Bosch J, eds. Portal hypertension in the 21st century. Boston: Kluwer Academic, 2004: 271–276. Groszmann RJ, Garcia-Tsao G, Makuch R, et al. Multicenter randomized placebo-controlled trial of non-selective betablockers in the prevention of the complications of portal hypertension: final results and identification of a predictive factor. Hepatology 2004;38(Suppl. 1):206 [abstract]. Merkel C, Marin R, Angeli P, et al. A placebo–controlled clinical trial of nadolol in the prophylaxis of growth of small esophageal varices in cirrhosis. Gastroenterology 2004;127:476–484. Burroughs AK, D’Heygere F, McIntrye N. Pitfalls in studies of prophylactic therapy for variceal bleeding in cirrhosis. Hepatology 1986;6:1407–1413. Lo G-H, Lai KH, Cheng JS, et al. Endoscopic variceal ligation plus nadolol and sucralfate compared with ligation alone for the prevention of variceal rebleeding: A prospective, randomized trial. Hepatology 2000;32:461–465. de la Pena J, Brullet E, Sanchez-Hernandez E, et al. Variceal ligation plus nadolol compared with ligation for prophylaxis of variceal rebleeding: A multicenter trial. Hepatology 2005;41:572–578. Papatheodoridis GV, Goulis J, Leandro G, et al. Transjugular intrahepatic portosystemic shunt compared with endoscopic treatment for prevention of variceal rebleeding: A meta-analysis. Hepatology 1999;30:612–622. Meddi P, Merli M, Lionetti R, et al. Cost analysis for the prevention of variceal rebleeding: A comparison between transjugular intrahepatic portosystemic shunt and endoscopic sclerotherapy in a selected group of Italian cirrhotic patients. Hepatology 1999;29:1074–1077. Escorsell A, Banares R, Garcia-Pagan JC, et al. TIPS versus drug therapy in preventing variceal rebleeding in advanced
381
Section III. Clinical Consequences of Liver Disease
262.
263.
264. 265.
266.
382
cirrhosis: A randomized controlled trail. Hepatology 2002;35:385–392. Boyer TD, Henderson JM. Portal hypertension and bleeding esophageal varices. In: Zakim D, Boyer TD, eds. Hepatology: A textbook of liver disease, 4th edn. Philadelphia: WB Saunders, 2003: 581–629. Warren WD, Zeppa R, Foman JS. Selective transplenic decompression of gastroesophageal varices by distal splenorenal shunt. Ann Surg 1967;166:437–455. Inokuchi K. A selective portocaval shunt. Lancet 1968;2:51–52. Stipa S, Balducci G, Ziparo V, et al. Total shunting and elective management of variceal bleeding. World J Surg 1994;18:200–204. Henderson JM, Gilmore GT, Hooks MA, et al. Selective shunt in the management of variceal bleeding in the era of liver transplantation. Ann Surg 1992;216:248–254.
267. Idezuki Y, Kokudo N, Sanjo K, Bandai Y. Sugiura procedure for management of variceal bleeding in Japan. World J Surg 1994;18:216–221. 268. Rosemurgy AS, Serafini FM, Zweibel BR, et al. Transjugular intrahepatic portosystemic shunt vs. small-diameter prosthetic H-graft portacaval shunt: extended follow-up of an expanded randomized prospective trial. J Gastroinest Surg 2000;4:589–597. 269. Henderson JM, Boyer TD, Kutner M, et al. Multi-center randomized trial of distal spleno-renal shunt vs. transjugular intrahepatic portal systemic shunt for control of refractory variceal bleeding in patients with Child’s class A or B cirrhosis. (submitted)
Section III: Clinical Consequences of Liver Disease
21
ACUTE LIVER FAILURE Arun J. Sanyal and R. Todd Stravitz Abbreviations AFP a-fetoprotein AIH autoimmune hepatitis ALF acute liver failure APACHE acute physiology and chronic health evaluation APAP acetaminophen ARDS adult respiratory distress syndrome BCAA branched chain amino acids BCS Budd Chiari Syndrome CMV cytomegalovirus CPP cerebral perfusion pressure DIC disseminated intravascular coagulation EBV Epstein-Barr virus
EEG FFP G6DP GABA GLNase GS GSH HAV HBeAg HBsAg HBV HCV HDV
electroencephalogram fresh frozen plasma glucose 6-phosphate dehydrogenase g-aminobutyric acid glutaminase glutamine synthetase glutathione hepatitis A virus hepatitis B e antigen hepatitis B surface antigen hepatitis B virus hepatitis C virus hepatitis D virus
INTRODUCTION AND DEFINITIONS The syndromes of acute liver failure (ALF), characterized by jaundice, coagulopathy, and altered sensorium (hepatic encephalopathy), are among the most catastrophic afflictions of humans. Without rapid hepatocyte regeneration, complications involving almost every organ system ensue, including cerebral edema, renal failure, respiratory failure, sepsis, gastrointestinal bleeding, and cardiovascular collapse. Fortunately, ALF is uncommon, with approximately 2000 cases per year in the United States; however, ALF remains a disease with an intimidating mortality of 40–95%. Because of its rarity, heterogeneity, and rapidity of evolution, ALF is an extremely difficult syndrome to study. Consequently, its management remains largely center specific, and many questions regarding the pathogenesis of the complications of ALF and their management remain unanswered. Lucke and Mallory first described ALF in 1946 as three phases of liver disease: a prodromal/preicteric phase, an intermediate phase with the onset of jaundice, and a final phase of encephalopathy. These and subsequent authors recognized that the interval between the onset of icterus and that of encephalopathy conveyed important etiologic and prognostic information. Accordingly, several groups have subcategorized patients with ALF by the time interval between the latter two phases. As most commonly defined, the term fulminant hepatic failure denotes the development of hepatic encephalopathy within 8 weeks of the onset of symptoms. A more gradually evolving ‘subfulminant’ course (also referred to as lateonset hepatic failure) was subsequently recognized by the King’s College group, and defined as encephalopathy developing between 8 and 24 weeks following the onset of jaundice.1 This latter subtype of ALF was characterized by a greater likelihood of developing signs of chronic liver disease, such as ascites and renal failure, a lower inci-
HEV ICP INR MAP NAC NAPQI OLT PCR PCWP PgI SIRS TTV
hepatitis E virus intracranial pressure international normalized ratio mean arterial pressure N-acteylcysteine N-acteyl-p-benzoquinone imine orthotopic liver transplantation polymerase chain reaction pulmonary capillary wedge pressure prostacyclin systemic inflammatory response stage transfusion-transmitted virus
dence of cerebral edema, but, paradoxically, a higher mortality than the fulminant subtype (Table 21-1). In order to better predict the clinical course, complications, and prognosis of patients with ALF, O’Grady et al.2 recently proposed categorizing ALF into hyperacute (jaundice to encephalopathy interval of less than 7 days), acute (interval of 8–28 days), and subacute (interval more than 28 days). This terminology was proposed after reviewing the cases of 538 patients seen in the Liver Failure Unit of King’s College, London.1,3 Patients with acetaminophen hepatotoxicity universally progressed to hepatic encephalopathy within 7 days of the onset of jaundice, a hyperacute time course (Table 21-2), and regardless of etiology, patients with hyperacute liver failure had a relatively favorable spontaneous survival rate (Figure 21-1). In contrast, cryptogenic liver injury dominated as the cause of subacute liver failure, and was associated with a dismal survival rate (~14%).4 ALF due to viral hepatitis A and B usually followed a hyperacute time course, and idiosyncratic drug toxicity a more delayed course, with survival rates that were inversely proportional to the jaundiceto-encephalopathy interval (Figure 21-1).5 According to its strict definition, the diagnosis of ALF requires liver failure in a patient with a previously healthy organ.6 However, some patients present with ALF superimposed on unrecognized chronic liver disease. Although purists do not consider this to represent true ALF, patients with acute decompensation of previously subclinical liver disease should be considered under the umbrella of ALF for the purposes of management.5
ETIOLOGY AND EPIDEMIOLOGY It is critical to identify the etiology of ALF in order to administer appropriate antidotes and to better anticipate prognosis and hence the need for liver transplantation. Many diverse insults to the liver
383
Section III. Clinical Consequences of Liver Disease
Figure 21-1. Clinical outcome of 228 patients with non-acetaminophen-induced ALF according to the time between onset of jaundice and onset of hepatic encephalopathy in weeks. (From O’Grady JG, Schalm SW, Williams R. Acute liver failure: redefining the syndromes. Lancet 1993;342:273, with permission).
90 Number of patients
80
Patients with cerebral edema
Survivors
Number of patients
70 60 50 40 30 20 10 0 1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
22 24
Weeks
Table 21-2. Prognosis of ALF According to the Interval Between Onset of Jaundice and Development of Encephalopathy: Proposed New Classification
50 N=690
Percent
40
Etiology Interval n
30 20
Acetaminophen (%) Hepatitis A (%) Hepatitis B (%) Hepatitis non-A, non-B (%) Drug (idiosyncratic; %) Survival (%)
10
In
ch
Is
ug H BV
e at
Dr
in
AI H em ia H AV W ils on ’s Pr BC eg S na nc y O th er
de
te
rm
AP AP
0
Etiology
Figure 21-2. Etiologies of ALF in the United States 1997–June 2004 reported to the Acute Liver Failure Study Group Registry. (Table 2 from Seminars in Liver Disease 2003; 23;217–226, ©2003, with permission of Thieme.)
Table 21-1. Complications of ALF According to Traditional Classification Feature Onset of encephalopathy after jaundice Patient age (y) Ascites (%) Renal failure (%) Cerebral edema (%) Chronic liver disease
Fulminant < 8 weeks 25 7 35 67 Rare
Subfulminant 8–24 weeks 45 62 62 9 Infrequent
(From Gimson AE, O’Grady JG, Ede RJ, et al. Late onset hepatic failure: clinical, serological and histological features. Hepatology 1986;6:288, with permission)
384
Hyperacute Liver Failure 0–7 days 391
Acute Liver Failure 8–28 days 89
Subacute Liver Failure >28 days 59
100 55 63 14 35 35
0 31 29 39 53 7
0 14 8 48 12 14
(From O’Grady JG, Schalm SW, Williams R. Acute liver failure: redefining the syndromes. Lancet 1993;342:273, and Williams R. Classification and clinical syndromes of acute liver failure. In: Lee WM, Williams R, eds. Acute liver failure. Cambridge: Cambridge University Press, 1997:2, with permission)
may result in ALF, including viral infections, idiosyncratic drug reactions, toxins, metabolic abnormalities, and vascular catastrophes. In the multicenter US Acute Liver Failure Study Group Registry, acetaminophen overdose accounts for almost 45% of ALF cases reported since 1997, with indeterminate/cryptogenic and idiosyncratic drug reactions in second and third place, being 13–15% of cases each (Figure 21-2). Globally, hepatitis B remains the most common cause of ALF. However, the relative contribution of different etiologies varies greatly by geography (Table 21-3). In developing nations, viral hepatitis (hepatitis A, B, and E) is the principal causes of ALF, whereas in the UK and USA acetaminophen toxicity is overwhelmingly the most common cause.7,8 Isoniazid hepatotoxicity is one of the most common drug-induced causes of ALF in regions where tuberculosis is endemic.12 In the west, the etiology of ALF remains unknown in a large proportion of cases.9–11 In some series specifically studying patients with subacute liver failure up to 70%
Chapter 21 ACUTE LIVER FAILURE
had an unknown cause of liver injury despite extensive serologic evaluation and polymerase chain reaction (PCR) testing of their serum.4
HEPATOTROPHIC VIRUSES All of the commonly recognized hepatotrophic viruses (hepatitis A, B, C, D, and E) have been reported to cause ALF, although the relative risk of ALF in acute infection, the clinical course of ALF, and the prognosis vary significantly depending on the specific viral infection (Table 21-4).
Hepatitis A Acute infection with hepatitis A virus (HAV) is an important cause of ALF because of its worldwide distribution. The risk of ALF after acute hepatitis A infection ranges between 0.01 and 0.1%, and is higher in older patients and those living in, or traveling to, endemic areas.13 ALF due to hepatitis A is diagnosed by the presence of IgM anti-HAV antibodies, which are present in 95% of cases; repeat testing may be required in suspected cases of acute hepatitis A to detect the remaining 5%.14 The prevalence of IgM anti-HAV antibodies in ALF ranges between 2 and 6% in most series,15 although it has been reported as high as 20–31% in northwestern Europe.16–18 The temporal pattern of ALF in acute hepatitis A is usually hyperacute, with only very rare subfulminant cases being reported.4,5 Accordingly, spontaneous survival in patients with ALF due to hepatitis A is rela-
Table 21-3. Prevalence of ALF Etiologies According to Geography Country UK France Denmark Argentina Japan India
APAP
HBV
HAV
Drug
Indeterminate
Other
73 2 19 0 0 0
2 32 31 22 18 31
2 4 2 8 3 2
2 17 17 14 0 5
8 18 15 25 71 0
9 27 13 31 8 62
APAP, acetaminophen; HBV, hepatitis B virus; HAV, hepatitis A virus. (From Lee WM. Acute liver failure in the United States. Semin Liver Dis 2003;23:217, with permission)
tively high (40–60%) compared to that of ALF due to hepatitis B or cryptogenic ALF.17,19 Because the age of exposure to HAV has been delayed with improved sanitation and the mortality from HAV-associated ALF increases with age, the relative importance of HAV in ALF has increased in some areas of Europe.20 Acute hepatitis A is more likely to progress to ALF in patients over 40 years of age21 with a history of homosexuality or IV drug abuse,22 and with underlying chronic viral hepatitis B23 or C24 or alcoholic liver disease.25 These observations support universal immunization against HAV in HBVand HCV-infected patients, or in patients with alcoholic liver disease. Liver transplantation for HAV-induced ALF may uncommonly result in the recurrence of hepatitis in the graft,21,26 suggesting that HAV immunoglobulin should be administered in such patients.
Hepatitis B The absolute risk of ALF after acute hepatitis B is approximately 1%.27 Women and older patients may be at greater risk for ALF from hepatitis B virus (HBV).28,29 Recent series from the USA have shown a marked reduction in HBV-induced ALF to only 10% of cases, perhaps due to successful vaccination programs.8 In areas of the world endemic for HBV, viral superinfection of chronic HBV carriers with HCV, HDV, or a cryptic viral agent may be the most common etiology of ALF.30,31 In Taiwan, for example, the incidence of ALF after acute viral hepatitis was 2%, but was 10.6% in chronic HBsAg carriers.32 The overall spontaneous survival rate after HBVinduced ALF ranges between 15 and 36%.29,33,34 The diagnosis of HBV in ALF is frequently hampered by a vigorous immunologic response, which rapidly clears the virus.28,35 Consequently, serum hepatitis B surface antigen (HBsAg), hepatitis B e antigen (HBeAg), and HBV DNA may be absent in ALF. Because HBsAg may clear within a few days of the onset of illness in as many as 30–50% of ALF patients,36 the diagnosis of acute HBV often relies on detecting indirect serologic evidence of infection (IgM anti-HBc and/or anti-HBs).37 The role of occult HBV infection in cryptogenic ALF remains uncertain. ALF occasionally follows withdrawal of immunosuppression or chemotherapy in patients with chronic HBV infection.38,39 Presumably, immunosuppression in such patients increases HBV replication and hepatocyte infection, which may then provoke massive immune-mediated cytotoxicity after the restitution of immune competence.
Table 21-4. ALF due to Hepatotrophic Viruses: Comparison of Clinical and Diagnostic Features Feature
Risk of ALF in acute infection Risk factors for ALF Clinical course Spontaneous survival after ALF Diagnostic tests
Hepatitis Virus A
B
C
D
E
Non-A–E
0.01–0.1%
1.0%
Very rare
Pregnant: 20%
?
HBV, HCV, age>40 yrs, IVDA, EtOH Hyperacute 40–60%
Females, HCV
HBV
Co: 1–10% Super: 5–20% Chronic HBV
Pregnancy
?
Hyperacute/acute 15–39%
Subacute ?
Acute 31–45%
Acute ?
Subacute 13–33%
IgM-anti HBc, HBsAg, HBV DNA
Anti-HCV, HCV RNA
IgM-anti HDV; Co: IgM-anti HBc
IgM-anti HEV
None
IgM-anti HAV
IVDA, intravenous drug abuse; EtOH, ethanol; Co, HBV/HDV co-infection; Super, HDV superinfection of chronic HBV; OLT, orthotopic liver transplantation.
385
Section III. Clinical Consequences of Liver Disease
Hepatitis D Hepatitis D virus (HDV; the delta agent) is an adventitious RNA virus that requires concomitant HBV infection to complete its life cycle.40 HDV and HBV are acquired by similar parenteral means, and HDV may infect a patient simultaneously with HBV (co-infection) or may superinfect a patient with pre-existing HBV infection; both forms of infection frequently lead to ALF.36 Although both coand superinfection with HDV increase the risk of ALF in hepatitis B two- to fivefold,27,41 the risk appears to be highest in superinfected subjects (acute mortality 1–10% and 5–20%, respectively), in whom the replicative machinery of HBV is well established.42,43 Paradoxically, the mortality from HDV/HBV co-infection may be less than for HBV alone,36,43 suggesting mutual interference in replication between the viruses. The prevalence of HDV antibodies in patients with HBV-associated ALF is highest in areas of the Mediterranean and Middle East.40 In the USA, HDV is most common in intravenous drug abusers (up to 34%),36,43 where epidemics of severe HDV hepatitis and ALF have been described.27
Hepatitis C Whether hepatitis C virus (HCV) independently causes ALF remains controversial. Most series from centers in western nations report few cases of ALF attributable to HCV as the sole etiology,4,37,44–46 with the exception of one study from Los Angeles that reported a 60% prevalence of HCV RNA in serum from ALF of indeterminate cause.47 Whereas HCV may be an uncommon etiology of ALF in western nations, studies from Japan have detected HCV markers (antibody and/or RNA) in as many as 50% of non-A, non-B cases of ALF.48,49 Although HCV does not appear to be an important cause of ALF in most western series, chronic HCV infection probably plays an important role as a cofactor in ALF triggered by viral superinfection.24,30,45,50
Hepatitis E Hepatitis E virus (HEV), a single-stranded RNA virus, has been identified as the agent responsible for enterically transmitted epidemic hepatitis.51 Infection with HEV occurs almost exclusively in developing nations and only rarely in the west, usually in émigrés or travelers from endemic areas.45,46,52,53 Areas of endemic HEV infection include parts of northern Africa and southern Asia, where HEV is one of the two most common etiologies of ALF, being second only to HBV.12,54 Young adults and pregnant women appear to be particularly vulnerable to ALF after acute HEV infection, the latter usually presenting in their third trimester.55 In such cases the overall mortality approaches 20%.56
Non-A–E viruses Recent studies have attempted to identify other viral etiologies of ALF. Putative agents have included togavirus,57 paramyxovirus,58 human papilloma virus 6,59 GB virus-C,60 and transfusion-transmitted virus (TTV).61 A paramyxovirus may be one of several causes of ALF related to syncytial giant-cell hepatitis.58,62 Patients present with severe chronic hepatitis or subacute hepatic failure, and nucleocapsids resembling measles virus have been detected in infected hepatocytes by electron microscopy.58 Although evidence of hepati-
386
tis G virus infection or TTV is common in patients with ALF, most studies refute a possible role of these viruses in non-A–E ALF because markers are equally common in the general population.61,63–65 Although recurrence of hepatitis after liver transplantation for ALF of presumed viral cryptic origin has been reported in a few cases,66 none of these putative agents have rigorously fulfilled Koch’s postulates.
Systemic Viral Infections All members of the herpes virus family have been anecdotally incriminated in ALF, usually in neonates and immunocompromised adults, including post-transplant patients.67–69 The clinical presentation of ALF due to varicella zoster virus infection usually involves the typical vesicular rash, which may be delayed in comparison to abdominal symptoms. Similarly, herpes simplex viral hepatitis usually presents as part of a disseminated infection including mucocutaneous vesicles, and frequently progresses to ALF, disseminated intravascular coagulation, and death.68 Rare reports of viral ALF have been ascribed to human herpes virus-6,70 Epstein–Barr virus (EBV),71 and cytomegalovirus (CMV),72 but have also been challenged.73 The diagnosis of ALF due to herpes virus is usually made by serologies and the detection of DNA in blood; liver biopsy specimens may be helpful in showing eosinophilic intranuclear inclusions. Prompt administration of IV aciclovir may be life saving.69,74 Other systemic viral infections, such as adenovirus75 and Coxsackie B virus76 have been incriminated in rare cases of ALF. Hemorrhagic fever viruses (Ebola, Yellow fever, Lassa fever, Dengue) cause ALF in endemic areas. Acute parvovirus B19 may cause ALF in children, and is associated with high spontaneous recovery rate.77
DRUG-INDUCED ALF Drug hepatotoxicity accounts for 10–20% of ALF in developed nations, and a much higher proportion if acetaminophen is included. Almost 1000 drugs have been incriminated in liver injury, many of which can cause ALF.78 Drugs that cause ALF may be intrinsic or idiosyncratic toxins. Intrinsic hepatoxins such as acetaminophen cause ALF in a dose-dependent and predictable manner, whereas idiosyncratic hepatotoxins cause ALF rarely and in a doseindependent manner78 (Table 21-5). Drug-induced ALF may be categorized by the primary pathological lesion observed on liver biopsy79 (Table 21-6). Most cause hepatocellular necrosis, others are mitochondrial toxins and lead to microvesicular steatosis, and yet others damage endothelial cells of
Table 21-5. Features of ALF Induced by Intrinsic vs Idiosyncratic Drug Hepatoxicity Feature of ALF
Intrinsic hepatotoxin
Idiosyncratic hepatotoxin
Dose dependence Incidence Latent period Clinical course Survival without transplantation
Yes High Consistent, usually days Hyperacute Relatively good
No Low variable, often weeks Subacute Poor
(Adapted from Zimmerman HJ. Hepatotoxicity: the adverse effects of drugs and other chemicals on the liver, 2nd edn. Philadelphia Lippincott Williams & Wilkins, 1999:446, with permission)
Chapter 21 ACUTE LIVER FAILURE
Table 21-6. A Partial List of Drugs Causing ALF According to Primary Pathological Findings Pathologic lesion/Drug
Frequency of ALF
Potentiating agents
Hepatocellular necrosis Acetaminophen
Dose-dependent
Ethanol, isoniazid, barbiturates
Anesthetics: halothane Infrequent Antiepileptics: Phenytoin Infrequent Carbamazepine Rare Antibiotics: Amoxicillin Isoniazid Infrequent Nitrofurantoin Ketoconazole Rare Ofloxacin Rare Sulfonamides Infrequent Antihypertensives: a-Methyldopa Hydralazine Labetalol Rare Nicotinic acid (slow release) Rare Non-steroidal anti-inflammatory drugs: Diclofenac Infrequent Bromfenac Infrequent Microvesicular steatosis Amiodarone Fialuridine Tetracycline Valproate
and anticonvulsants, particularly phenytoin.78 ALF due to recreational drugs, such as Ecstasy (3,4-methyl-dioxymethamphetamine), appears to be increasing in Europe.83–85 In Spain, 8% of patients admitted to a specialized liver unit with ALF (31% with drug-induced disease) had a recent history of Ecstasy ingestion, the second most common etiology of ALF in patients under the age of 25.85
Acetaminophen Clavulanic acid Rifampin
Trimethoprim
Infrequent
Infrequent
Veno-occlusive disease Azathioprine Busulfan Cyclophosphamide Dacarbazine 6-Thioguanine (Adapted from Sze G, Kaplowitz N. Drug hepatotoxicity as a cause of acute liver failure. In: Lee WM, Williams R eds. Acute liver failure. Cambridge: Cambridge University Press, 1997:27, and Lee WM. Acute liver failure. N Engl J Med 1993;330:584, with permission)
terminal hepatic venules, leading to veno-occlusive disease. The clinical course of ALF due to an idiosyncratic drug reaction usually follows a subacute tempo, with a high mortality without liver transplantation. In addition to ALF, drugs that cause microvesicular steatosis result in progressive lactic acidosis (e.g. fialuridine),80 and those that cause hepatic veno-occlusive disease result in acute right upper quadrant pain, tender hepatomegaly, and ascites. Age, gender, nutritional state, concomitant diseases, other drugs, ethanol consumption, and genetic polymorphisms of drug-metabolizing hepatic enzymes – most importantly the cytochrome P450s – all contribute to the risk of idiosyncratic drug-induced ALF.19,79 The risk of ALF after ingestion of some drugs increases in the presence of potentiating agents, often another drug (e.g. rifampin and trimethoprim, which enhance the hepatotoxic effects of isoniazid and sulfamethoxazole, respectively).81,82 Idiosyncratic drug reactions lead to ALF infrequently (~1/10 000 prescriptions) or rarely (~1/50 000 prescriptions (Table 21-6).79 The most common agents include non-steroidal anti-inflammatory drugs, antibiotics (especially sulfonamides, isoniazid and rifampin),
The analgesic acetaminophen is an intrinsic hepatotoxin with a narrow therapeutic window. When used in recommended doses (<4 g/day), acetaminophen rarely causes hepatotoxicity. However, acetaminophen overdose is the single most common cause of ALF in the UK, accounting for 60–70% of cases,3,86 and the USA (nearly 45% in the ALF Study Group Registry).8 In an earlier review of acetaminophen-induced hepatotoxicity at Parkland Hospital in Dallas, over two-thirds of patients consumed toxic doses of acetaminophen with suicidal intent, and the remaining one-third were accidental overdoses.87 More recent experience from the ALF Study Group Registry, however, suggests that unintentional overdose has become much more common, accounting for 55% of acetaminopheninduced ALF, and that a significant proportion of this increase has come from overdoses with acetaminophen-containing narcotics (38% of cases).88 The clinical presentation of ALF due to acetaminophen includes a hyperacute progression from jaundice to hepatic encephalopathy.5 Hepatic transaminases rise within 12–24 hours of ingestion, often peaking at strikingly high levels with AST>ALT, significantly higher than for other etiologies of ALF (Table 21-7).8 Peak transaminases and prothrombin times are usually observed 3 days after ingestion, and then rapidly resolve if the patient spontaneously recovers; peak bilirubin levels are lower than for other etiologies of ALF (Table 217). The spontaneous survival rate is higher and the need for liver transplantation lower with acetaminophen than for ALF of other etiologies.8 Liver biopsy characteristically reveals centrilobular (zone 3) necrosis without inflammation.89 Renal failure, either oliguric or non-oliguric, develops in 70% of patients with ALF within 24–72 hours of ingestion of the acetaminophen overdose.90 The mechanism of hepatic injury in acetaminophen overdose and the rationale behind the use of its antidote, N-acetylcysteine (NAC), was first elucidated in experimental animals by Mitchell and co-workers in 1973.91–94 Under non-toxic conditions, 80% of an acetaminophen dose is glucuronidated and sulfated, and eliminated in the urine (Figure 21-3). A small fraction (<5%) is oxidized to the reactive intermediate N-acetyl-p-benzoquinone imine (NAPQI) by cytochrome P450 enzymes, most importantly cytochrome P450 2E1. NAPQI may be rendered harmless by binding to reduced glutathione (GSH), a reaction catalyzed by glutathione-S-transferases. However, under conditions where the supply of NAPQI exceeds available GSH, the former covalently binds hepatocellular proteins, initiating hepatocyte necrosis. Disruption of mitochondrial membrane integrity and function may be the primary target of NAPQI.78 The toxicity of acetaminophen may be enhanced by agents that either increase the production of NAPQI or reduce the supply of GSH. Higher doses of acetaminophen saturate the enzymes involved in conjugation, thereby increasing the substrate for the
387
Section III. Clinical Consequences of Liver Disease
Table 21-7. Clinical and Laboratory Features of the 690 Patients Enrolled in the Acute Liver Failure Study Group Registry Variable Sex (% female) Age (yrs) Encephalopathy Stage III/IV (%) Peak ALT (U/L) PT (s) Bilirubin (mg/dl) Transplanted (%) Spontaneous Survival (%) Overall Survival (%)
APAP (n = 299)
Drug (n = 88)
Indeterminate (n = 99)
HAV/HBV (n = 27/52)
All Others (= 125)
73 37
70 40
56 37
48/48 47/40
77 41
51 4100 28 4.6 8
40 580 25 22 45
47 904 27 23 41
48/60 2661/1415 25/30 12/19 26/50
42 677 26 15 32
64
19
25
59/27
30
70
61
64
81/69
56
NS, not statistically significant. p Values represent comparisons between patients with acetaminophen-induced ALF and all other cases. (Data courtesy of WM Lee, PI, ALF Study Group. Data shown are for patients enrolled as of June 2004)
O
O
O
HN – C – CH3
HN – C – CH3
HN – C – CH3
Glucuronyl Transferase
Sulfotransferase
B O-Gluc
OH
Glucuronide
APAP
SO4
P-450 2E1 (P-450 1A2) (P-450 3A4)
A O HN – C – CH3
O NAPQI GSH Transferase
Covalent Binding C
O
O
HN – C – CH3
HN – C – CH3
Necrosis S-G OH Mercapturic Acid
388
S-Prot
Figure 21-3. Pathways of acetaminophen (APAP) metabolism. Pathway A, the pathway toward increased formation of the toxic APAP metabolite, NAPQI, is enhanced by larger doses of acetaminophen and inducers of cytochrome P450 2E1 (ethanol, isoniazid, phenobarbital, phenytoin). Pathway A may also be enhanced by conditions that decrease glucuronidation (fasting, malnutrition). Pathway B is inhibited by fasting and glucuronyl transferase deficiency (Gilbert’s syndrome), leading to an increase in pathway A. Pathway C, a major pathway toward detoxification of NAPQI, is inhibited by conditions that deplete the hepatocyte of glutathione (GSH), such as fasting, malnutrition, and ethanol consumption. Inhibition of pathway C leads to increased covalent binding of NAPQI to hepatocyte proteins, and hence necrosis. (From Zimmerman H. Hepatotoxicity: the adverse effects of drugs and other chemicals on the liver. Philadelphia: Lippincott Williams & Wilkins, 1999, with permission).
Chapter 21 ACUTE LIVER FAILURE
oxidative pathway. Ethanol and certain drugs (e.g. isoniazid and barbiturates) induce the activity of cytochrome P450s, thereby increasing NAPQI production; the increased toxicity of acetaminophen due to ingestion of ethanol appears to be predominantly due to induction of cytochrome P450 2E1.95 In contrast, fasting and malnutrition such as are seen with chronic alcohol abuse decrease GSH synthesis, reducing the hepatocyte’s ability to detoxify NAPQI. The dose of acetaminophen required to induce hepatic necrosis varies greatly between patients and may depend on nutritional status, ethanol consumption, co-administered drugs, and probably environmentally and genetically determined expression of cytochrome P450 2E1.96 In average-risk patients ingesting acetaminophen with suicidal intent, a minimal dose of 7–8 g is necessary to induce hepatocyte necrosis. Doses of 15 g often cause hepatotoxicity, whereas 21 g consistently cause severe liver injury;97 the mean ingested dose of acetaminophen in the ALF Study Group Registry was 13.2 g.88 The potentiation of acetaminophen hepatotoxicity by chronic alcohol consumption occurs at doses considered ‘therapeutic’.87,89,98 In alcoholics with acute hepatic injury after ingestion of acetaminophen with therapeutic intent, 40% ingested <4 g/day, the recommended dose, and 60% ingested <6 g/day, considered ‘non-toxic’.89
Biological Toxins
Reye’s syndrome, a disorder of hepatocyte mitochondrial metabolism, has become an extremely rare cause of ALF in the USA, with no more than two cases per year reported to the Centers for Disease Control between 1994 and 1997.108,109 Reye’s syndrome usually presents in children with an influenza-like viral prodrome and a history of salicylate ingestion, and is followed by encephalopathy, cerebral edema, and frequently death. Liver biopsy shows characteristic microvesicular steatosis with little necrosis, reflecting mitochondrial injury, which impairs both urea cycle disposal of ammonium and b-oxidation of fatty acids.110 Other metabolic causes of ALF in neonates or children include galactosemia, fructosemia, tyrosinemia, a1-antitrypsin deficiency, and Niemann–Pick disease type II.111 ALF presenting in pregnant women may be caused by disease entities specific to pregnancy or non-specific agents such as viruses or drugs.76,112 Acute fatty liver of pregnancy is a disorder of hepatocyte mitochondrial metabolism occurring in the late third trimester. The high mortality due to ALF for both mother and fetus can be completely avoided by prompt delivery.113 Recent studies have attributed acute fatty liver of pregnancy and the HELLP syndrome (hemolysis, elevated liver chemistries, low platelets) to concomitant defects in maternal and fetal mitochondrial long-chain 3-hydroxyacyl-coenzyme A dehydrogenase.114
Ischemic Causes of ALF
ALF caused by ingestion of the mushrooms of the genus Amanita (A. phalloides, verna, and virosa) occurs occasionally in Europe (50–100 fatal cases per year) but rarely in the USA, usually in California and the Pacific Northwest (fewer than 100 fatal cases between 1900 and 1994).99–101 Three medium-sized mushrooms (50 g) contain sufficient toxin, a-amanitin and phalloidin, to cause ALF; the toxins are heat stable and not degraded by cooking. Symptoms of gastroenteritis (abdominal pain, nausea, vomiting, and diarrhea) precede liver dysfunction, and renal failure and pancreatitis are common. A mitochondrial toxin isolated from the foodborne pathogen Bacillus cereus was recently incriminated in a case of ALF in which autopsy of the liver showed microvesicular steatosis.102 ALF due to herbal remedies has been reported with increasing frequency,103,104 and all patients with ALF should be specifically queried about ingestion of alternative medicines.
Acute hepatic vein thrombosis – the Budd–Chiari syndrome – may present as ALF.115 The classic findings of chronic hepatic vein thrombosis, tender hepatomegaly and ascites may be absent in patients presenting with ALF. Pathologically, the liver shows extensive hemorrhagic infarction. Circulatory collapse – so-called ‘shock liver’ – may evolve into ALF if hepatic ischemia is prolonged, often after sepsis or surgery,116 or in patients with severe heart failure.117 Severe intrahepatic sickling in patients with sickle cell anemia can also precipitate ischemic ALF,118,119 as can veno-occlusive disease after systemic chemotherapy, usually in the setting of bone marrow transplantation. In the early post liver-transplant recipient ALF may be a manifestation of hepatic artery or portal vein thrombosis. Finally, status epilepticus rarely may result in ALF due to ischemic liver injury, possibly in addition to chronic anticonvulsant therapy.120
Metabolic Causes of ALF
Diffuse Malignant Infiltration of the Liver
Acute Wilson’s disease, a rare presentation of the autosomal recessive defect in canalicular copper transport, accounts for about 3% of ALF cases in the US ALF Study Group Registry. Wilson’s disease presenting as ALF typically occurs in a young patient (first two decades of life), and is often accompanied by Coombs’negative hemolytic anemia, the result of massive copper release from necrotic hepatocytes and subsequent injury to erythrocyte membranes. Hypouricemia and low serum alkaline phosphatase may accompany strikingly high serum bilirubin (largely indirect) and low ceruloplasmin levels. Serum ceruloplasmin concentrations may be normal or high in patients presenting with ALF, however, and serum or urine copper may be helpful.105,106 A suspicion of acute Wilson’s disease as the cause of ALF should immediately lead to liver transplant evaluation, as spontaneous recovery without transplantation is rare.107
Rarely, massive hepatocellular necrosis may follow infiltration of the liver by several metastatic malignancies, most commonly breast carcinoma and lymphomas. Other malignancies include melanoma, gastric carcinoma, small-cell lung carcinoma, pancreatic carcinoma, and leukemia.121,122 Hepatic CT scanning may not reveal nodular infiltration if there is diffuse intrasinusoidal spread. Pathologic examination most often reveals diffuse infiltration of the liver with tumor cells rather than nodular aggregates.122
Rare Causes of ALF These include Q fever, falciparum malaria, and infections with hemorragic fever viruses. Primary non-function after liver transplantation causes the syndrome of ALF; multiorgan failure may be abated by graft hepatectomy while a patient awaits retransplantation.123
389
Section III. Clinical Consequences of Liver Disease
PATHOGENESIS AND CLINICAL FEATURES OF ALF Early Presentation of the Acute Injury Depending on the cause of ALF, the illness may either present without warning or be preceded by a prodromal illness.124 Patients often complain of non-specific epigastric and upper abdominal distress accompanied by anorexia and nausea in the early stages of the illness. Marked abdominal pain may occur in those with Budd–Chiari syndrome.125,126 On examination, the liver is usually normal in size or small. These symptoms are accompanied by marked elevations of serum AST and ALT and only modest elevations in alkaline phosphatase.
FAILURE OF LIVER FUNCTION Failure of Hepatobiliary Excretion Marked impairment in hepatobiliary excretory function results in hyperbilirubinemia and jaundice. The canalicular excretion of bilirubin is the rate-limiting step in bilirubin excretion; consequently, ALF causes conjugated hyperbilirubinemia.127 The degree of hyperbilirubinemia is accentuated if hemolysis coexists,128,129 which may result from the oxidant stress associated with the cause of ALF, or to ALF itself. Patients with an underlying predisposition to hemolysis, for example glucose 6-phosphate dehydrogenase (G6DP) deficiency, experience the highest degrees of hyperbilirubinemia.130 ALF associated with Wilson’s disease is often accompanied by hemolytic anemia and extraordinarily high serum bilirubin levels.131
Failure to Metabolize Toxic Substances The liver metabolizes many potentially toxic endogeneous substrates which accumulate in ALF, the most clinically important of which is ammonia. The increase in serum ammonia levels in ALF is primarily due to the failure of the liver to convert ammonia to urea via the urea cycle,132–134 and has been implicated in the pathogenesis of hepatic encephalopathy and intracranial hypertension.135–138 Sources of retained ammonia in ALF include the intestine, muscle, and less importantly the kidneys (Figure 21-4). The metabolism of drugs constitutes the second important detoxification function of the liver. Most drugs undergo some degree of hepatic modification. Because ALF impairs drug metabolism, the biological half-life of many drugs increases in these patients.139 The volume of distribution of many drugs is also altered in ALF, which further affects the elimination of both drugs and their metabolites.139 Renal failure further impairs the excretion of water-soluble drugs or drug metabolites.140 These changes in pharmacokinetics enhance the probability of drug toxicity or worsened liver injury.141,142 Therefore, the use of all medications in the setting of ALF must be carefully considered in terms of necessity, dosage, and toxicity.
Failure of Intermediary Metabolism The metabolic consequences of ALF include alterations in carbohydrate, lipid, and protein metabolism (Table 21-8). Spontaneous hypoglycemia frequently complicates ALF because of decreased glycogen stores, decreased ability to mobilize glycogen, and
390
Glutamine NH3
Glutamine Urea
Glutamine
? Ur NH3 Figure 21-4. Sources and metabolic fate of ammonia and glutamine in ALF. The primary source of ammonia is the gut, which under normal conditions is cleared by urea (major pathway) and glutamine (minor pathway) synthesis in the liver. Hepatocellular insufficiency in ALF results in the accumulation of ammonia in peripheral tissues, particularly brain and muscle, which detoxify ammonia by synthesizing glutamine from glutamate. In turn, glutamine released into blood is taken up by the intestines, liberating ammonia, or cleared by the kidneys. The capacity of renal excretion of glutamine in ALF is adversely affected by renal dysfunction, which often accompanies ALF. (Figure 1 from Seminars in Liver Disease 2003; 23:259–269, ©2003, with permission of Thieme.)
Table 21-8. Metabolic Consequences of Acute Liver Failure Carbohydrate metabolism Hypoglycemia: decreased glycogen stores decreased gluconeogenesis Hyperglycemia (usually mild): insulin resistance Lipid metabolism Increased plasma free fatty acids: increased peripheral lipolysis decreased lipogenesis Altered arterial ketone body ratio: altered mitochondrial redox potential Protein metabolism Increased protein breakdown Increased plasma amino acid levels Relative decrease in branched chain amino acid levels
decreased gluconeogenesis within the liver.143–146 Serum concentrations of free fatty acids increase in ALF,147 resulting in a decrease in acetoacetate/3-b-hydroxybutryate (arterial ketone body ratio), which may contribute to altered sensorium.148,149 ALF is also associated with negative nitrogen balance,134,150 which results from the enhanced catabolism of proteins, including muscle proteins.151,152
Failure of Biosynthetic Function of the Liver The two most clinically relevant synthetic products of the liver include albumin and coagulation factors. Albumin has a half-life of
Chapter 21 ACUTE LIVER FAILURE
15–20 days,153 and serum concentrations therefore do not usually drop until late in the course of ALF. Coagulopathy exists universally in patients with ALF.51 Plasma activities of factors II, V, VII, IX and X, which are synthesized in the liver, are invariably reduced. Factor V and VII activities, which have the shortest half-lives, offer important prognostic information,154 and deficiency of these factors results in a prolongation of the prothrombin time.154,155 Although the prothrombin time is expressed in many hospitals as the international normalized ratio (INR), several studies have concluded that the INR may be misleading in the setting of ALF; therefore, prothrombin time should be reported.156,157
IMPACT OF ALF ON OTHER PHYSIOLOGIC SYSTEMS ALF affects the function of virtually all organ systems, not only from the direct consequences of hepatic necrosis and failure of liver function, but also from microcirculatory dysfunction, which causes inadequate oxygenation of peripheral tissues. As the function of one organ system declines, a domino effect occurs in others, often compounded by sepsis. Eventually, multiorgan failure results in death.
Table 21-9. Pathogenesis of Impaired Tissue Respiration in Acute Liver Failure Impaired pulmonary oxygenation of blood Decreased delivery (dysfunctional airway, e.g. mucus plugs) Impaired ventilation V/Q mismatch Altered oxygen carriage and delivery capacity of blood Decreased hemoglobin concentration Altered oxygen association–dissociation properties of hemoglobin: Increased association hypocarbia alkalosis decreased red cell DPG in stored blood Increased dissociation: acidosis hypercarbia increased red cell DPG due to anemia and hypoxemia Altered ability to deliver oxygenated blood to tissues Decreased cardiac output (advanced ALF) Changes in regional vascular resistance Opening of peripheral arteriovenous shunts Impaired oxygen utilization by tissues Changes in transit time in peripheral microcirculation Pathologic oxygen dependency
Microcirculatory Dysfunction (Table 21-9) The pathogenesis of impaired peripheral tissue oxygenation in ALF includes defects in oxygenation of blood in the lungs, oxygen carriage in arterial blood, and oxygen extraction from the microcirculation. Decreased oxygenation of blood in the lungs may result from atelectasis, volume overload, pneumonia, or the development of the systemic inflammatory response state (SIRS). Oxygen delivery to peripheral tissues via the blood is adversely affected by decreased cardiac output and shifting of the oxygen dissociation curve of hemoglobin (Figure 21-5) as a consequence of acid–base and ventilatory disturbances and altered 2,3-diphosphoglycerate levels.158,159 Finally, ALF compromises peripheral tissue perfusion as a consequence of vasodilation of the microcirculation.160–164 Vasoactive factors such as
% Saturation of hemoglobin
100
75
50
Increased affinity Decreased affinity
–alkalosis –hypothermia –hypocarbia –decreased 2,3-DPG
–acidosis –fever –hypercarbia –increased 2,3-DPG
25
0 0
25
50
75
100
125
Blood pO2 (mmHg) Figure 21-5. Oxyhemoglobin dissociation curve. The affinity of hemoglobin for oxygen is increased (curve shifted to the left) with alkalosis, hypothermia, hypocarbia, and decreased red cell 2,3-DPG; the affinity of hemoglobin for oxygen is decreased (curve shifted to the right) with acidosis, hypercarbia, and increased red cell 2,3-DPG.
nitric oxide and tumor necrosis factor165–167 appear to be responsible for the microcirculatory dysfunction in ALF, which creates a functional diffusion barrier against oxygen delivery because of decreased contact time with tissue capillaries.168,169 In addition, peripheral vascular shunts open, diverting blood away from tissue.170–172
Cardiovascular Consequences The initial cardiovascular hallmark of ALF is a hyperdynamic state, characterized by an increased cardiac output and decreased systemic vascular resistance.170,173 This decrease in systemic vascular resistance results from peripheral arteriolar dilatation, capillary recruitment in peripheral tissues, and arteriovenous shunting, predominantly the result of increased nitric oxide activity.173–175 As a result, cardiac output often increases to values of 7–10 l/min,176 owing to both tachycardia and increased stroke volume.177 In early ALF, central venous pressure usually remains low, reflecting decreased central blood volume,178 but hypervolemia usually follows due to infused intravenous fluids and the development of renal failure.177 Similarly, mean arterial pressure is maintained by the increase in cardiac output in early ALF, but drops once cardiac output succumbs to intravascular volume depletion, arrythmias, or myocardial depression.177,178 The final stages of ALF are characterized by profound peripheral vasodilation, failing cardiac output, and eventually hemodynamic collapse.
Pulmonary Consequences (Table 21-10) Central hyperventilation with a respiratory alkalosis characterizes the initial pulmonary presentation of ALF.173,179 As intracranial pressure (ICP) rises, hyperventilation often progresses and the development of severe hyperventilation may precede sudden respiratory arrest. Initially, oxygenation is relatively preserved in patients with ALF,173 but progressive hypoxemia supervenes due to ventilation– 391
Section III. Clinical Consequences of Liver Disease
Table 21-10. Pulmonary Dysfunction in ALF
Table 21-12. Mechanisms Underlying Bleeding Diathesis in ALF
Airway problems Mucus plugs Aspiration Endotracheal tube cuff leak Pulmonary alveolar and parenchymal processes: Pneumonia Pulmonary hemorrhage Atelectasis Pneumothorax Pulmonary vascular processes: Volume overload Heart failure ARDS
Altered production of coagulation factors Decreased activity of factors II, V, VII, IX and X Increased activity of factor VIII (endothelial activation) Increased consumption of coagulation factors Disseminated intravascular coagulation (DIC) Increased fibrinolysis Thrombocytopenia ?Decreased production vs. sequestration DIC Qualitative platelet dysfunction De novo Renal failure
Table 21-11. Mechanisms of Renal Failure in ALF Prerenal azotemia Increased GI losses (GI bleeding, nasogastric drainage, diarrhea from lactulose) Inadequate volume replacement Acute tubular necrosis Volume depletion Iatrogenic (aminoglycosides, NSAIDs) Sepsis-related Decreased renal perfusion Cortical necrosis Urinary tract infection Hepatorenal syndrome
perfusion mismatch within the lungs, volume overload, left ventricular failure, intrapulmonary arteriovenous shunting, increased pulmonary capillary permeability, and pneumonia.174,178 Pulmonary edema in the presence of a normal pulmonary capillary wedge pressure suggests the development of adult respiratory distress syndrome (ARDS), which can also emerge as the pulmonary component of the systemic inflammatory response syndrome (SIRS).180 In the final stages of ALF, respiratory failure can develop in the form of both hypoventilation and failure of oxygenation.181
Renal and Electrolyte Disturbances (Table 21-11) Acute renal failure occurs in 40–80% of patients with ALF173,182–184 and presents as oliguria and rising serum creatinine. The four major causes include effective blood volume depletion, acute tubular necrosis, sepsis, and hepatorenal syndrome.185,186 A decrease in effective circulating blood volume frequently results from marked systemic vasodilation, gastrointestinal bleeding, fluid loss due to inappropriately aggressive lactulose therapy, and inadequate volume replacement. Severe, prolonged hypovolemia, or the use of nonsteroidal anti-inflammatory drugs or aminoglycosides can lead to acute tubular necrosis, a complication documented in 22.5% of patients with ALF.90,182 Sepsis exacerbates peripheral vasodilation and may precipitate renal failure from circulatory collapse,187 or cause diffuse cortical necrosis by initiating disseminated intravascular coagulation (DIC).188,189 Hepatorenal syndrome (type I)146,186,190,191 in ALF appears to be initiated by pathophysiologic mechanisms similar to those present in cirrhotic individuals.186 The 392
prognosis of patients with ALF and renal failure is very poor unless they receive a liver transplant,192,193 but spontaneous renal recovery can occur with recovery of liver function.186 Severe fluid and electrolyte abnormalities always accompany ALF. Free water retention occurs early, resulting in hyponatremia,194 and contributes to cerebral edema if severe. In general, the degree of hyponatremia is proportional to the severity of liver failure. Hypokalemia accompanies hyponatremia, and is due to gastrointestinal losses, aggressive diuresis, and alkalosis. Hypophosphatemia occurs commonly and results from a shift of phosphate from the extracellular to the intracellular compartment in response to glucose infusions,195 and perhaps due to use in regenerating hepatocytes.196 In the presence of oliguric renal failure, however, hyperkalemia and hyperphosphatemia usually develop. Finally, hypocalcemia can follow the transfusion of large amounts of citrated blood products.
Hematologic Disturbances (Table 21-12) Patients with ALF have a multifactorial bleeding diathesis owing to decreased synthesis of coagulation factors, increased factor consumption, and quantitative as well as qualitative platelet dysfunction. The presence of DIC consumes coagulation factors and should be suspected in a patient with microangiopathic hemolytic anemia, increased fibrinogen degradation products and fibrin D-dimer levels, and decreased fibrinogen levels.155,197,198 The degree of hypofibrinogenemia reflects the severity of the DIC, and is most severe in ALF complicated by sepsis. Platelet defects, both qualitative and quantitative, further compound the bleeding diathesis of the coagulopathy present in ALF.199,200 Platelets from patients with ALF exhibit poor adhesion and aggregation, especially in the setting of renal failure.201,202 Thrombocytopenia commonly accompanies ALF and results from DIC-induced consumption rather than low serum thrombopoietin concentrations, as might be expected in the setting of massive hepatic necrosis.201,203 Coagulopathy and platelet dysfunction in ALF therefore increase the risk of spontaneous hemorrhage from mucosal surfaces, the female reproductive organs, and the gastrointestinal tract. However, the risk of clinically significant spontaneous bleeding is less than 10%.204,205 Breakdown of Host Immune Defenses (Table 21-13) Several abnormalities in normal immune defense mechanisms occur in patients with ALF and markedly increase the susceptibility to
Chapter 21 ACUTE LIVER FAILURE
Table 21-13. Factors Contributing to Infection in ALF
Table 21-14. Clinical Stages of Hepatic Encephalopathy in ALF
Breakdown of natural barriers to infection Skin: intravenous lines Airway: Aspiration of pharyngeal and gastric contents Intubation Urinary tract: indwelling catheters Cranial: ICP monitors GI tract: increased translocation from edematous and hemorrhagic mucosa Colonization Skin and pharynx: hospital flora Stomach: use of antacid therapy promotes colonization Colon: use of antibiotics Impaired host defenses Decreased complement levels Decreased neutrophil chemotaxis Impaired ability to opsonize bacteria Decreased neutrophil capacity to produce superoxide ions Decreased reticuloendothelial capacity to clear bacteria and bacterial products
Grade
Symptoms
Signs
EEG
0 1
Normal Lack of awareness, short attention span, altered sleep pattern Agitation, lethargy seizures Asleep, arousable by pain, confused when aroused Unarousable
Normal Tremor, constructional apraxia, asterixis
Normal Symmetric slowing
Asterixis, hyperreflexia Hyperreflexia
Symmetric slowing, triphasic waves Triphasic waves
Babinski, ankle clonus, decerebrate posture
Delta (very slow) activity
2 3
4
ALF (Based on Conn HO. Quantifying the severity of hepatic encephalopathy. In: Conn HO, Bircher J, eds. Hepatic encephalopathy: syndromes and therapies. Philadelphia: Medi-Ed Press, 1994)
Table 21-15. Proposed Mechanisms Underlying Hepatic Encephalopathy in ALF
infection. Over 80% of patients with ALF have bacteriologic evidence of infection at some point during their illness.206–208 Natural host barriers are breached by the process of caring for critically ill patients. The use of acid-suppressive therapy increases the ability of pharyngeal bacteria to colonize the stomach,209 and in the setting of altered mental status they may be aspirated. Abnormal antibacterial defenses further contribute to the susceptibility to infection, such as depressed complement concentrations, impaired opsonization of bacteria, and decreased neutrophil chemotaxis and superoxide production.210–213 Clinically, pneumonia, septicemia and urinary tract infections are the most common types of infection encountered in patients with ALF.207
Gastrointestinal Consequences Nausea and vomiting occur frequently early in the course of ALF, and in the later stages an ileus may develop. The cause of the ileus is often multifactorial, and includes electrolyte and acid–base disturbances, sepsis, and the inappropriate use of narcotics to control agitation. Although pancreatic enzyme levels are elevated in a third of patients,214 clinically significant pancreatitis occurs only infrequently. Gastrointestinal bleeding from mucosal petechial lesions can also occur, especially in the setting of thrombocytopenia, DIC, and sepsis. The hepatic venous pressure gradient can become elevated, and both varices and ascites have been reported to develop late in ALF;215,216 however, variceal hemorrhage is distinctly rare.215
Neurologic Consequences By definition, neurologic dysfunction follows liver injury in patients with ALF. Although patients with ALF and cirrhosis experience a similar altered mental status, the clinical features of the neurologic dysfunction in ALF differ importantly from those associated with cirrhosis.217,218 Seizures and agitation frequently complicate the hepatic encephalopathy of ALF, but patients with cirrhosis do not manifest these symptoms. Furthermore, patients with ALF develop
Circulating neurotoxins Ammonia Gut-derived false neurotransmittors, e.g. Octopamine g-Aminobenzoic acid (GABA) Endogeneous GABA receptor agonists Miscellaneous: mercaptans, fatty acids, others Altered neurotransmission in brain GABA receptor complex-mediated neurotransmission Glutamatergic system Altered cerebral energy homeostasis
cerebral edema and intracranial hypertension, which occur very rarely in patients with cirrhosis.
Hepatic Encephalopathy (Tables 22-14, 22-15) The biochemical basis of hepatic encephalopathy in ALF remains incompletely understood, but certainly involves the accumulation of endogenous or gut-derived toxins in the central nervous system that alter energy balance and neurotransmission.133,218–220 Circulating neurotoxins that have been incriminated in the genesis of hepatic encephalopathy include ammonia, endogenously derived benzodiazepine receptor agonists, and others.221–225 Ammonia has been most strongly implicated in the pathogenesis of hepatic encephalopathy in ALF (Figure 21-6), as high serum and brain concentrations of ammonia136,137,226 and brain concentrations of glutamine, the principal metabolic product of ammonia detoxification, are usual in patients with ALF.133,227–229 Although circulating concentrations of ammonia correlate only weakly with the degree of neurologic dysfunction,230 arterial rather than venous concentrations may be more accurate.134 Ammonia in the patient with ALF originates predominantly from the gut, from bacterial metabolism of urea and amino acids, as well as from the utilization of glutamine as an energy source by the intestinal mucosa (see Figure 21-4). The accumulation of putative neurotoxins within the brain of patients with ALF causes encephalopathy by two interrelated mech393
Section III. Clinical Consequences of Liver Disease
anisms: altered brain metabolism and altered neurotransmission. Although early studies supported a role for altered brain energy homeostasis in the encephalopathy of ALF, more recent studies provide conflicting data;231–234 thus, altered energy homeostasis may be a terminal rather than a primary mechanism.235 The two neurotransmitter systems that appear to be most adversely affected in ALF are the g-aminobutyric acid (GABA) and glutamatergic systems. The GABA receptor complex serves as the site of action of benzodiazepines and inhibits neurotransmission.224 In patients with ALF, increased circulating endogenous ligands for GABA receptors have been detected236 and ammonia increases the affinity of such ligands for this receptor.237 Decreased turnover of GABA in the brain may also contribute to the increase in GABA-ergic tone in the brain.238 In contrast to the increased tone of the inhibitory GABA system, intracellular concentrations of glutamate, the major excitatory neurotransmitter of the mammalian brain, are decreased in
Presynaptic neuron
to CSF
Astrocyte
Glutamine GLNase Glutamate
cap.
Glutamine GS
NH3
Glutamate Glutamate AM PA
AMPA
KA
KA
NM
GLT-1 GLAST
DA
Postsynaptic neuron
M
Figure 21-6. Relationship of ammonia to hepatic encephalopathy in ALF. A cartoon demonstrating the relationship between arterial ammonia levels and glutamatergic neurotransmission in cerebral neurons and astrocytes in the genesis of hepatic encephalopathy. (From Butterworth RF. Hepatic encephalopathy: disorder of multiple neurotransitter systems. In: Record C, Mardini AH, eds. Advances in hepatic encephalopathy and metabolism in liver disease. Newcastle upon Tyne: Ipswich Book Company, 1997, with permission.)
ALF.239–242 Diminished glutamate concentrations probably result from increased consumption rather than decreased production, as glutamate is used to detoxify ammonia within astrocytes (Figure 21-6).
Intracranial Hypertension and Cerebral Edema The adult cranium is a rigid compartment with very poor compliance; consequently, a small increase in the volume of intracranial contents can cause marked increases in intracranial pressure. Mechanisms of autoregulation of cerebral blood flow often collapse in patients with ALF, which may result in cerebral hyperperfusion and injury to the frontal cortex,243,244 or hypoperfusion and cerebral ischemia.170,245–247 In ALF, the rise in intracranial pressure occurs mainly because of cerebral edema.248,249 Cortical astrocytes, which appear swollen on histologic examination of brains from patients with ALF,217,221,250,251 account for most of the increase in cerebral volume, not only because they are quantitatively one of the major cell types in the brain, but because they also regulate brain volume as the principal component of the blood–brain barrier. Astrocyte swelling due to an increase in intracellular osmolarity (cytotoxic edema) probably constitutes the single most important pathogenetic mechanism of cerebral edema in ALF. The most likely candidate for the increased intracellular osmole is glutamine (see Figure 21-7), derived by the addition of ammonia to glutamate via glutamine synthetase within astrocytes.252 Indeed, marked increases in brain glutamine concentrations occur in patients with ALF.242 Neurons normally adapt to increased intracellular osmolarity and cell volume by increasing export of inorganic ions and endogenous organic osmolytes such as myoinositol (Figure 21-7);253–255 the export of organic osmolytes predominates. In contrast to patients with cirrhosis, patients with ALF appear not to have time to compensate for the accumulation of glutamine by exporting organic osmolytes, which probably accounts for the fact that cerebral edema occurs frequently in ALF but seldom in cirrhosis.256,257
MANAGEMENT OF ACUTE LIVER FAILURE GENERAL MANAGEMENT Initial Evaluation and Triage (Table 21-16) Every patient with acute hepatocellular necrosis can potentially progress to ALF. Once a patient’s mental status begins to deteriorate, one may lose the opportunity to obtain vital information that
NH3 Glutamine
Organic osmolyte export Osmolarity + Cell volume Ion export
394
Figure 21-7. Astrocyte swelling in the pathogenesis of cerebral edema in ALF: mechanisms of cell swelling and compensation. Intracerebral concentrations of ammonia are detoxified by the amidation of glutamate to glutamine, a reaction catalyzed by glutamine synthetase. Glutamine, an osmotically active solute, increases astrocyte cell volume, which may be attenuated by two mechanisms, exporting ions (minor pathway) or organic osmolytes (major pathway).
Chapter 21 ACUTE LIVER FAILURE
Table 21-16. Initial Evaluation of Acute Liver Failure
Table 21-17. General Management of Acute Liver Failure
History Medications Recreational drugs Prodrome Travel Alcohol Past medical history, including psychiatric Physical examination Vital signs Size of liver Mental status Rash Laboratory studies To identify cause of ALF Hepatitis A IgM antibody Hepatitis B surface antigen and IgM core antibody HCV antibody HDV antibody (if HbsAg positive) Drug screen, including APAP level Doppler sonogram Antinuclear and antismooth muscle antibodies Ceruloplasmin, serum copper To assess severity Prothrombin time, bilirubin ABG for arterial pH Factors V and VII Phosphate, arterial lactate To assess complications Serum creatinine X-ray chest Cultures (blood, urine, sputum, wound) Electrolytes, blood sugar, plasma osmolarity Arterial ammonia Laboratory tests/studies required for OLT listing Anti-HIV Anti-CMV, EBV Echocardiogram Blood type and cross-match
Administration of etiology-specific therapy Acetaminophen: N-acetyl cysteine (NAC) Amanita: penicillin and silibinin Carbon tetrachloride: NAC Herpes simplex: aciclovir Hepatitis B: lamivudine? Lassa fever, yellow fever: ribavirin Malaria: quinine Giant cell hepatitis: steroids Autoimmune hepatitis: steroids Wilson’s disease: high-dose penicillamine? Management of concomitant diseases Diabetes: sliding scale insulin Hypothyroidism: thyroxine Asthma: continue b-agonists Fluid and electrolyte disturbances requiring management Hyponatremia Hypo- and hyperkalemia Hypocalcemia and -magnesemia Hyperosmolarity Acid–base disturbances requiring management Respiratory acidosis and severe respiratory alkalosis Metabolic acidosis with increased anion gap Nutrition Caloric goals: 35–40 kcal/kg/day Proteins: 40 g/day, titrate as necessary
could guide management, including the administration of life-saving antidotes. Therefore, upon initial contact with a medical team a careful drug ingestion history should be obtained, including prescription medications, herbal remedies, over-the-counter medications, and recreational drugs. Confounding conditions, such as alcohol use, malnutrition, and drug–drug interactions must be considered.258 Although by definition ALF occurs in a patient with a previously normal liver, chronic liver disease may manifest initially as acute liver necrosis, including autoimmune hepatitis,259,260 Wilson’s disease,105,261,262 and viral superinfections. Therefore, past medical history should also include a search for signs, symptoms, and risk factors for chronic liver diseases. A travel history should be obtained, as exposure to certain exotic etiologies of ALF may otherwise be overlooked. Finally, a detailed psychiatric history may provide clues about a possible surreptitious ingestion of hepatotoxins, particularly acetaminophen. Initial laboratory tests should include tests to assess the degree of physiologic dysfunction, the risk of mortality and the potential need for liver transplantation. Patients in whom the initial prothrombin time is more than 4–5 seconds prolonged should be admitted to
hospital for observation; the clinician should not be lulled into a false sense of security by a normal mental status in the presence of significant coagulopathy. Furthermore, patients with altered mental status, hemodynamic compromise, renal insufficiency, decreased oxygenation, acidosis or hypoglycemia should be admitted to an intensive care facility, and should be discussed with the nearest liver transplant center. Whether to proceed with liver transplantation evaluation remains perhaps the most important decision in the initial evaluation of a patient with ALF.3,20,263 Patients with grade II or higher encephalopathy should be evaluated unless contraindications exist. In subjects with lesser degrees of encephalopathy, profound coagulopathy (prothrombin time >50 s) or acidosis (pH <7.3) should also lead to early consideration for listing for transplantation. Older subjects, those with idiosyncratic drug-induced ALF, and patients with a subacute presentation should be evaluated relatively early in the disease course because of their poor prognosis for spontaneous recovery.1
Etiology-Specific Therapies (Table 21-17) The use of an antidote may decrease hepatic injury and reverse ALF in specific circumstances. N-acetyl cysteine (NAC) remains the treatment of choice for acetaminophen overdose, and in theory may protect the liver from other toxins, such as carbon tetrachloride or trichloroethylene, which cause hepatotoxicity by generating free radicals.264,265 The administration of NAC for acetaminophen overdose replenishes GSH, thereby detoxifying NAPQI.266 In addition, NAC improves oxygen delivery and utilization in ALF.267 The risk of acetaminophen toxicity in an individual can be estimated by a nomogram plot of the initial plasma acetaminophen concentration versus the time after ingestion (Figure 21-8).268 Patients with plots 395
300
25
-3
0%
ch
an
ce
of
he
pa
to
Hepatotoxicity unlikely
to
xic
ity
Standard treatment line
30 4
6
8
10
12
14
<12 hrs
12-24 hrs
>24 hours**
No NAC** at Referring Hospital
100
90% chance of hepatotoxicity
16
Cumulative Survival (%)
Plasma acetaminophen concentration
Section III. Clinical Consequences of Liver Disease
75
50
Time post overdose (hours) Figure 21-8. Nomogram depicting the risk of hepatotoxicity from acetaminophen according to plasma acetaminophen concentration (g/ml) and time after ingestion. A standard treatment line for administering N-acetylcysteine (NAC) was derived empirically. Plots above the standard treatment line have significant risk of hepatotoxicity and should receive NAC immediately; plots below the line have a low risk of hepatotoxicity and do not require NAC. Such nomograms should be interpreted with caution, however (see text). (From Makin AJ, Williams R. Acetaminophen-induced acute liver failure. In: Lee WM, Williams R, eds. Acute liver failure. Cambridge: Cambridge University Press, 1997, with permission.)
lying above an empirically derived treatment line have a high risk of hepatic necrosis and should definitely receive NAC. Below a zone of intermediate risk, a patient is unlikely to experience serious hepatotoxicity and therefore does not require NAC. However, one must be aware of the potential pitfalls in the use of nomograms,96 especially inaccurate assessment of the time since ingestion and the possibility of toxicity in individuals who accidentally ingest acetaminophen over several days. Early administration of NAC (within 8 hours of overdose) minimizes hepatotoxicity regardless of the initial plasma concentration of acetaminophen.265 The time after which NAC administration is no longer effective, however, remains controversial, with some studies documenting benefit of administration up to 24–36 hours after ingestion.265,269–271 Conversely, the worst survival rates after ALF due to acetaminophen overdose are observed in patients not given NAC at the referring hospital prior to transfer (Figure 21-9).270 Based on these observations, the King’s College group recommends that NAC ‘be used whenever there is any doubt concerning the timing, dose ingested, or plasma concentration, since the use of the antidote is much less hazardous than the consequences of withholding it’.97 In the USA, NAC is usually administered orally at a loading dose of 140 mg/kg, followed by 70 mg/kg every 4 hours for a total of 17 doses. For patients who cannot tolerate oral administration, intravenous NAC has recently been approved in the same doses for a total of 48 hours. NAC should be diluted 1:5 in 5% dextrose in a glass flask,269 and is as effective as orally administered NAC.272,273 In contrast to oral administration, intravenous NAC carries a small risk of allergic reaction and therefore the oral route is preferred. NAC should not be discontinued prematurely and a full course should be given, even after acetaminophen levels have become undetectable.
396
0 1
12
24
35
Time (Hours) Figure 21-9. Survival of patients with acetaminophen overdose according to the time at which they received N-acetylcysteine (NAC). Note that the survival of patients receiving NAC even >24 hours after ingestion was significantly higher than that of patients who were not treated with NAC before transfer to the Liver Failure Unit at King’s College Hospital (p <0.0001). (From Makin AJ, Wendon J, Williams R. A 7-year experience of severe acetaminopheninduced hepatotoxicity (1987–1993). Gastroenterology 1995;109:1907, with permission.)
Suspected Amanita mushroom poisoning should be treated initially with ipecac and charcoal to decrease the toxin load if the ingestion has occurred recently (within 30 minutes to a few hours).101 A combination of penicillin (300 000–1 000 000 units/kg/day, or 250 mg/kg/day IV) and silibinin (20–50 mg/kg/day IV) has been used as a specific antidote in those with evidence of liver injury due to Amanita poisoning.274,275 These agents are hypothesized to interrupt the enterohepatic circulation of toxins and also to compete at the hepatocyte membrane for transmembrane transport.101 Owing to the rarity of this etiology of ALF, the benefits of this regimen remain unproven. Other disease-specific treatments warrant consideration because of their potential to improve a patient’s clinical course and possibly avoid the need for liver transplantation. These include the use of aciclovir for herpes simplex hepatitis,74,276 ribavirin for Lassa fever,277,278 and antimalarials for falciparum malaria.128,279
Management of Fluids, Electrolytes, and Acid–Base Abnormalities Fluid and electrolyte abnormalities occur frequently in patients with ALF. Hyponatremia develops from decreased excretion of free water, and is exacerbated by movement of sodium to the intracellular compartment194 and the administration of diuretics and mannitol. In the absence of renal failure, the management of hyponatremia depends upon the volume status of the patient;280 hyponatremia associated with euvolemia or hypervolemia should be managed by restriction of free water, whereas hypovolemic patients should receive normal saline. Severe hyponatremia ([Na+]
Chapter 21 ACUTE LIVER FAILURE
<115 mEq/l) may require infusion of small amounts of 3% sodium chloride intravenously, but only with careful monitoring of the hemodynamic and neurologic status of the patient in an intensive care setting. Once renal failure sets in, severe hyponatremia is best treated in conjunction with dialysis.281 Hypokalemic alkalosis occurs early in the course of ALF, whereas hyperkalemic acidosis dominates the late stages. The former condition requires intravenous infusion of potassium, whereas the latter mandates hemodialysis. Hypophosphatemia also occurs frequently in patients with ALF, and may be precipitated by dextrose infusions and respiratory alkalosis, which drives phosphate into the intracellular compartment.282 Hypophosphatemia may contribute to worsening mental status and respiratory failure and should therefore be corrected by intravenous repletion. In the setting of renal failure and acidosis, the development of significant hyperphosphatemia requires dialysis.283 Hypocalcemia from repeated transfusions of citrated blood products may cause tetany or cardiac arrythmias.284,285 Hypocalcemia and hypomagnesemia may present concurrently and interfere with the correction of hypokalemia; these abnormalities should be corrected by intravenous replacement therapy.
PREVENTION AND MANAGEMENT OF COMPLICATIONS Bleeding Patients with mild to moderate coagulopathy and absence of bleeding do not require specific intervention,300 but the administration of vitamin K will ensure that deficiency does not contribute to the bleeding diathesis. The use of fresh frozen plasma (FFP) in patients with severe but asymptomatic coagulopathy remains controversial, as few data document its efficacy in preventing bleeding. Indeed, overzealous infusion of FFP may result in volume overload,263 transmission of cytomegalovirus infection, and will obscure the very important prognostic information of the prothrombin time. It therefore seems reasonable to withhold FFP in patients without overt bleeding. The limitations of FFP infusion can be overcome by exchange plasmapheresis,301–303 which avoids volume overload and rapidly corrects even severe coagulopathy304–309 (Figure 21-10). Plasmapheresis should be considered before central line or intracranial pressure monitor placement or liver transplantation. Recently, recombinant factor VII has been advocated for similar indications, and appears safe and very effective. DIC does not usually require specific intervention unless severe and accompanied by bleeding.263 Although limited studies have shown that heparin
Management of Nutrition
48
42
36
Seconds
ALF is a catabolic state characterized by negative nitrogen balance, muscle wasting, and aminoaciduria.286 Although the clinical value of nutritional support in ALF has not been carefully studied, several physiological considerations support the correction of negative nitrogen balance. First, protein–calorie malnutrition impairs the immune system, thereby increasing susceptibility to infection.286 Second, severe malnutrition impairs wound healing,287 which may negatively affect the outcome of liver transplantation. Approximately 40 g/day of protein should be administered initially and the dose modified based on an assessment of the metabolic state.288 However, increasing protein supplementation can contribute to hyperammonemia and hepatic encephalopathy,133 and arterial ammonia levels should be monitored during aggressive nutritional support and the protein load modified appropriately. Many studies have suggested that decreased levels of branched chain amino acids (BCAA) contribute to hepatic encephalopathy,289–292 forming the basis for the administration of enteral or parenteral nutritional products supplemented with BCAA. Although numerous studies have evaluated the utility of BCAA as a treatment for hepatic encephalopathy,290,293–297 results remain inconclusive owing to the heterogeneous patient populations studied, and to differences in study design and methodologies. Given their cost, the available data do not support the routine use of BCAA-supplemented feeds or infusions in the management of ALF. The initial caloric goal for the patient with ALF is approximately 35–40 Kcal/day295 by either the enteral or the parenteral route. As a general rule, patients with normal gastrointestinal function should be fed enterally. Patients with volume overload may meet caloric needs by the ingestion of lipid emulsions, which may be used safely in those with ALF.298 Hypoglycemia is a common and potentially fatal complication of ALF,299 and blood glucose should be monitored at 2–3-hour intervals by finger stick, and 10% dextrose infused to maintain levels over 80 mg/dl.
30
24 * 18
0 Before
After
Figure 21-10. Prothrombin time in 19 patients with ALF before and after a single exchange plasmapheresis. (From Munoz SJ, Ballas SK, Moritz MJ, et al. Perioperative management of fulminant and subfulminant hepatic failure with therapeutic plasmapheresis. Transplant Proc 1989;21:3535, with permission.)
397
Section III. Clinical Consequences of Liver Disease
400 Oxygen Consumption (ml · min-1 · m-2)
may be used to treat DIC in patients with ALF,205,310 the potential risks in these patients limit enthusiasm for this treatment. Gastrointestinal bleeding in a patient with ALF usually results from superficial gastric erosions and stress ulcers,204 which should be prevented by the use of H2 receptor antagonists or proton pump inhibitors. The effectiveness of acid suppression should be monitored by periodic assessment of intragastric pH, which should ideally be maintained above 4–5. Hemodynamically significant gastrointestinal bleeding should prompt diagnostic and therapeutic upper endoscopy, as for patients without ALF.
** **
300
*
200 100 0
Cardiovascular Derangements Maintaining peripheral tissue perfusion and oxygenation requires a normal blood volume as well as normal cardiac rhythm and contractility. Therefore, intravascular volume must be accurately assessed, and in all patients with but the mildest forms of ALF a central line should be inserted. In patients with more advanced ALF a Swan–Ganz catheter should be floated to measure the pulmonary capillary wedge pressure (PCWP), and fluids administered to keep the PCWP between 8 and 12 mmHg.263 Cardiac arrythmias can result from the electrolyte and acid–base disturbances associated with ALF311 and should be managed by correction of the underlying abnormality before the administration of antiarrhythmics. If the stroke volume and cardiac output remain low despite adequate PCWP values and normal cardiac rhythm, an inotropic agent (dobutamine or norepinephrine) may be considered,312 the latter being preferred because of its inotropic as well as its peripheral vasocontrictive effects. Whereas moderate sinus tachycardia does not require intervention, the cardiac output drops when the heart rate exceeds 180 beats/min,311 and tachyarrythmias exceeding this range should be treated appropriately. The development of sinus bradycardia should raise the possibility of severe intracranial hyper-tension and is specifically managed with atropine and adrenergic agonists. When blood pressure cannot be maintained with fluid administration and a normal cardiac rate and rhythm, systemic vascular resistance should be increased with norepinephrine.313 Despite improved mean arterial pressure, the use of vasopressors does not reliably improve tissue oxygenation because of constriction of peripheral arterioles,314 and so other agents have been used in this situation. Prostacyclin (PgI) decreases platelet aggregation and postcapillary venular tone,315 and the addition of PgI to epinephrine or norepinephrine improves tissue oxygenation, as evidenced by increased oxygen extraction ratio.314,316–318 N-acetyl cysteine (NAC) has also been shown to improve tissue oxygenation in ALF. As a potent antioxidant,319 NAC decreases the production of vasoactive substances by activated endothelial cells320 and thereby decreases peripheral arteriovenous shunting. Indeed, patients with acetaminophen-induced ALF who received NAC experienced improved hemodynamics compared to those who did not.321 Other studies have shown that NAC improves hemodynamics and oxygenation, and decreases renal failure and progression to deeper grades of coma.267,322 The effects of NAC administered with PgI appear to be additive, with improvement in cardiac index and mixed venous oxygen content (Figure 21-11).267 The use of NAC is therefore reasonable in patients with ALF, regardless of etiology.
398
Cardiac Index (liters · min-1 · m-2)
14 **
12 *
*
10 8 6 4 2 0 Base Line 1
Epoprostenol
Base Line 2
Acetylcysteine
Acetylcysteine and Epoprostenol
Figure 21-11. The effects of epoprostenol (prostacyclin) and N-acetylcysteine (NAC) on cardiac index and oxygen consumption in 12 patients with acetaminophen-induced ALF. (From Harrison PM, Wendon JA, Gimson AES, et al. Improvement by acetylcysteine of hemodynamics and oxygen transport in fulminant hepatic failure. New Engl J Med 1991;324:1852, with permission.)
Pulmonary Complications and Ventilatory Support A major decision in the management of a patient with ALF is the timing of endotracheal intubation. The indications for intubation include airway protection, provision of respiratory support, and management of intracranial hypertension.263 A less quantifiable indication includes extreme agitation, which risks exacerbating intracranial hypertension. In addition, paralytic agents can be employed after intubation which will decrease spontaneous hyperventilation and hence lactic acidosis due to excessive activity of respiratory muscles.323 Although there are no controlled clinical trials to support a specific form of respiratory support, volume-controlled ventilation is generally chosen. Although PEEP (2–4 mmHg) often improves oxygenation, higher pressures cause hepatic venous congestion and a decrease in hepatic arterial flow.324 High PEEP also compromises venous return to the heart, cardiac output, blood pressure, peripheral tissue perfusion, and above all may increase the propensity to develop cerebral edema.178
Prevention and Management of Infection Sepsis remains one of the major causes of death in patients with ALF, owing to its frequency and subtle clinical presentation (Figure 21-12).206 Bacteremia has been identified in up to 50–80% of patients with ALF,208,325,326 usually within the first few days after onset.207 Compared to other patients undergoing orthotopic liver transplantation (OLT), patients with ALF are particularly vulnerable to bacterial infections.327
Chapter 21 ACUTE LIVER FAILURE
25
Total White blood cell count >11× 109/liter and temperature > 38°C
Number of infections
20
High white blood cell count
Figure 21-12. Clinical indicators of infection in patients with ALF. (From Rolando N, PhilpottHoward J, Williams R. Management of infection in acute liver failure. In: Lee WM, Williams R, eds. Acute liver failure. Cambridge: Cambridge University Press, 1997, with permission.)
High temperature Neither
15
10
5
0 Bacteremia
Respiratory
Urinary
Urinary tract 22%
IV catheter 12%
Chest 50%
Blood only 16%
Figure 21-13. Sites of infection in patients with ALF. (From Rolando N, Philpott-Howard J, Williams R. Management of infection in acute liver failure. In: Lee WM, Williams R, eds. Acute liver failure. Cambridge: Cambridge University Press, 1997, with permission.)
The most common infections in patients with ALF are pneumonia followed by systemic bacteremia and urinary tract infections (Figure 21-13).207 Of these, pneumonia accounts for half of cases and is almost invariably accompanied by chest X-ray abnormalities. Grampositive organisms account for the majority of bacterial infections, with Staphylococcus aureus the most commonly isolated, and Escherichia coli the most commonly isolated Gram-negative organism.207 Fungal infections, most commonly Candida spp., have been found in up to 32% of cases328 and contributed to death in 13% in one study.326 Prevention of infection is thus an important objective of the medical management of ALF,206 and general guidelines to avoid nosocomial transmission of organisms should be strictly enforced (Figure 21-14). Whether potentially infectious colonizing microbes should be monitored for remains controversial. Some authorities advocate
I.V. cannulae
obtaining daily blood and urine cultures, especially early in the course of ALF,206 in order to obtain antibiotic sensitivities in case of future infection.207 Prophylactic antibiotics generally appear to reduce infections in patients with ALF. One study randomized patients with ALF, and no evidence of infection on admission to the intensive care unit, to a regimen of oral colistin, tobramycin and amphotericin B, and a 4-day course of intravenous cefuroxime, or to placebo.329 This significantly decreased the infection rate compared to patients who did not receive the antibiotics. Similarly, antibiotic prophylaxis was associated with an improved trend towards survival. Other regimens consisting of oral neomycin, colistin and norfloxacin have also been effective.330 Thus, based on current literature, patients with ALF should receive a third-generation cephalosporin parenterally in combination with amphotericin with or without additional antibiotics (e.g. norfloxacin) administered orally. Once infection is suspected or documented, antibiotic coverage should be empirically administered according to the results of surveillance cultures, or, in their absence, adjustments made after definitive identification of the organism and determination of its antibiotic sensitivities (Figure 21-14).206 Fungal infections should be suspected in patients with high fever unresponsive to antibiotics, profound leukocytosis, and acute renal failure. In such cases, amphotericin, fluconazole, or caspofungin should be administered. Recently, granulocyte colony-stimulating factor has also been advocated to improve neutrophil function in patients with ALF.331
Management of Renal Insufficiency The development of renal failure is a marker of poor prognosis and greatly complicates fluid and electrolyte, hemodynamic, and ventilatory management of the patient with ALF. The management of renal failure depends upon the underlying cause (see Table 2111).173,183,186 Once oliguria develops, continuous arteriovenous or venovenous hemofiltration should be considered. Hemofiltration techniques minimize hypotension, rapid fluid shifts, and plasma osmolality changes, and thereby decrease the risk of cerebral edema compared to standard hemodialysis.332–337 Plasma osmolality should
399
Section III. Clinical Consequences of Liver Disease
Figure 21-14. Algorithm for the prevention and treatment of infection in patients with ALF. (From Rolando N, Philpott-Howard J, Williams R. Management of infection in acute liver failure. In: Lee WM, Williams R, eds. Acute liver failure. Cambridge: Cambridge University Press, 1997, with permission.)
Admission to liver failure intensive care unit Microbiology from Referral Hospital
Perform clinical cultures & microbiological screening
Positive significant cultures
Commence prophylactic antibiotics IV* e.g.
Commence appropriate antibiotics
24–48 hours
Ceftazidime + Flucloxacillin Piperacillin + Tazobactam
Others Vaginal Clotrimazole
Cefotaxime + Flucloxacillin
Positive clinical cultures
Positive microbiological screening
MRSA
Oral/NG Amphotericin B suspension
Commence MRSA eradication therapy
Yes
No
Candida Enterococci, Assess patient for coliforms active infection Pseudomonas
Recovery Stop antibiotics after 5 days
Clinical deterioration / suspected sepsis Yes
No
Perform cultures
Listed for OLT
Previous colonization
Cover pathogenic organism
Significant cultures
Adjust therapy
Recovery
Unknown
Commence Amphotericin B & Imipenem + Vancomycin or Cipro + Amoxycillin
Yes
Yes
No No
Stop antibiotics Commence Amphotericin B IV OLT Continue antibiotics x 5 days Antifungals for 21 days
Change prophylactic antibiotics e.g. Ciprofloxacin + amoxycillin Imipenem + Van if MRSA (+) Ceftazidime + Flucioxacillin
be assessed prior to the institution of hemodialysis, as a rapid drop in plasma urea levels may exacerbate cerebral edema in a patient with low osmolality before dialysis. Such patients may be managed with mannitol infusions to raise serum osmolality before the institution of dialysis. Hepatorenal syndrome usually requires hemodialysis and liver transplantation, but the use of vasopressors (e.g.
400
norepinephrine) to improve effective renal blood flow may reverse this condition in cirrhotic patients338 and should be considered. Although terlipressin, another systemic vasoconstrictor, has been increasingly used to reverse hepatorenal syndrome in cirrhosis, this agent increases cerebral blood flow and intracranial pressure in patients with ALF, and should be avoided.339
Chapter 21 ACUTE LIVER FAILURE
Multiorgan failure The microcirculatory disturbances associated with ALF lead to multiorgan failure, characterized by non-cardiogenic pulmonary edema (ARDS), renal failure, gastrointestinal bleeding, ileus, and acidosis. Although sepsis usually precipitates this chain of events,340 a systemic inflammatory response syndrome (SIRS) commonly develops even in the absence of sepsis. In a retrospective analysis of 887 patients with ALF180 40% of individuals without sepsis developed SIRS and multiorgan failure. This portends worsening intracranial hypertension and death. Consequently, patients with severe peripheral microcirculatory dysfunction and acidosis should be considered for a trial of prostacyclin and/or NAC, with the recognition that these agents have not been shown to improve survival.341,342
Hepatic Encephalopathy The administration of lactulose has not been shown to improve hepatic encephalopathy in ALF and remains controversial.263 Furthermore, the injudicious use of lactulose may result in volume depletion, electrolyte disturbances, and ileus. In practice, however, a cautious trial of lactulose in a patient with ALF seems reasonable to treat the hyperammonemia; lactulose should be discontinued if no benefit is observed after two enemas. Reversible precipitating factors, particularly sepsis, should be ruled out, as recent analysis of the ALF Study Group Registry found that infection and the presence of SIRS predicted progression from early to late stages of hepatic encephalopathy.343
Cerebral Edema and Intracranial Hypertension (Table 21-18) Hepatic encephalopathy in the presence or absence of cerebral edema often cannot be clinically distinguished, even in some patients with marked elevations in intracranial pressure (ICP).251,344 Whereas
Table 21-18. Management of Cerebral Edema in ALF Methods to modulate cerebral blood flow Elevate head of bed by 20–30∞ Correct volume overload Maintain mean BP around 50–60 mmHg Hyperventilate to keep PCO2 25–30 mmHg Correct factors that increase ICP Minimize head turning Correct excessive PEEP Intratracheal lidocaine before respiratory suctioning Treat agitation with intubation and sedation (propofol); paralyze if necessary Treat hypertension Monitor for and treat seizures Correct hypoxemia Direct measures Correct hypo-osmolarity with mannitol before hemodialysis Mannitol (1–2 mg/kg IV bolus over 5 min) Pentobarbital coma Induce hypernatremia to 150–155 mmol/l Liver transplantation Desperate measures Hypothermia Total hepatectomy
patients with hepatic encephalopathy alone usually recover neurologically after liver transplantation or spontaneous hepatic regeneration, those with severe cerebral edema and intracranial hypertension often do not. The goals of treating intracranial hypertension are therefore not only to prevent herniation but also to optimize neurologic recovery. Although head CT scanning often reveals evidence of cerebral edema in patients with advanced hepatic encephalopathy (grade >II),345 the absence of radiologic evidence of cerebral edema does not rule out intracranial hypertension.346,347 Consequently, ICP monitors have become increasingly employed in patients with advanced encephalopathy,348 although this has not resulted in major improvements in patient survival.347,349 Insertion of an intracranial pressure monitor should be considered in any patient with ALF who appears to be a viable liver transplant candidate and breaches grade II encephalopathy or an arterial ammonia concentration of 150 g/dl.350 Epidural monitor placement carries a 5% risk of intracranial hemorrhage compared to about 20% with intraparenchymal or subdural monitors;351 accordingly, coagulopathy should be corrected with FFP, recombinant factor VII, and/or plasmapheresis before monitor placement.263 Insertion of an ICP monitor should not be delayed until the latest stages of hepatic encephalopathy, as the risk of insertion increases and the potential benefit decreases with advanced cerebral edema.259 Cerebral ischemia, the consequence of cerebral edema and intracranial hypertension, appears to determine long-term neurologic prognosis in patients with ALF who survive,244,352 and is determined by the cerebral perfusion pressure (CPP), the difference between mean arterial pressure (MAP) and ICP. Measurements predictive of poor neurologic outcome include sustained increases in ICP over 40 mmHg, and decreases in CPP under 40 mmHg for more than 2 hours, although more recent data have refuted the latter.259,353 The goals are therefore to maintain the ICP below 20 mmHg and the CPP between 50 and 65 mmHg.263,354–356 Initially, these goals can be accomplished by raising the patient’s head by 20–30∞, controlling agitation, minimizing head turning, avoiding the excessive use of PEEP, and moderately hyperventilating the patient (Table 21-18). The PCO2 should be maintained between 25 and 30 mmHg, which decreases cerebral blood flow by restoring autoregulation (Figure 21-15).357 The practical value of therapeutic hyperventilation is modest, however, because patients with ALF usually hyperventilate spontaneously to achieve this level of hypocapnia. Mean arterial pressure should be supported with pressors if intravenous volume is insufficient. Fever can increase intracranial blood flow and should be treated aggressively.358 Finally, a recent randomized placebo-controlled study explored the use of induced hypernatremia (using hypertonic saline) to prevent sustained increases in ICP in patients with grade III/IV encephalopathy.359 Both the incidence and the severity of intracranial hypertension were significantly lower in the hypernatremic patients (mean peak sodium concentration 153 mmol/l) than in those who received placebo, suggesting that induced hypernatremia may be effective prophylaxis against cerebral edema in patients with ALF. Indications for specific therapy of intracranial hypertension include the development of intracranial ‘pressure waves’, which presage the development of irreversible neurologic injury or herniation,251,360–362 or sustained ICP >25 mmHg. Mannitol should be administered as an intravenous bolus (1–2 mg/kg) after a baseline
401
Section III. Clinical Consequences of Liver Disease
125 Pa O2
CBF (ml/100 g/min)
Pa CO2
75
25 BP 0 0
25
75
125
Pressure (torr) Figure 21-15. Relationship of cerebral blood flow (CBF), a major determinant of intracranial pressure, to arterial PO2, PCO2, and mean systemic blood pressure. (From Muñoz, S. Difficult management problems in fulminant hepatic failure. Semin Liver Dis 1993;13:395, with permission.)
osmolality has been drawn. The infusion of mannitol decreases cerebral edema by increasing the osmolality of blood in brain capillaries, which draws water back into the intravascular space. The administration of mannitol can be repeated, and has been shown not only to resolve cerebral edema but also to improve survival in patients with ALF.363 If serum osmolality increases to >320 mmol/l, mannitol is ineffective in reducing intracranial pressure and should not be repeated. In approximately 20% of individuals the intracranial pressure rises paradoxically after mannitol administration,363 and therefore it should not be repeated. Mannitol is most effective for mild to moderate intracranial hypertension and relatively ineffective when the ICP exceeds 60 mmHg.263 In oliguric or anuric patients, mannitol administration should be accompanied by the institution of continuous arteriovenous or venovenous hemofiltration, which removes the drug from the circulation, allowing subsequent doses to work.364 With repeated doses, the efficacy of mannitol decreases.263,280 When intracranial hypertension becomes refractory to mannitol, the institution of a barbiturate coma should be considered. Barbiturates decrease brain metabolism and have been shown to protect the brain from further injury in ALF. A thiopental drip (10 mg/kg/min starting dose) should be titrated according to the ICP and CPP rather than its plasma level.263,365 It must be remembered that the EEG cannot be used to diagnose brain death in a patient in a barbiturate coma. Also, one must be cognisant of the ability of barbiturates to produce hypotension, which can adversely affect the CPP. When the intracranial hypertension fails to respond to mannitol and barbiturates, the clinician often turns to less well accepted (i.e. desperate) measures. Complete hepatectomy has been rarely used in uncontrollable intracerebral hypertension in order to bridge a patient with ALF to liver transplantation.366 Limited data also indicate that the induction of hypothermia may stabilize the ICP in patients with ALF.367–370 The largest series comes from University
402
College, London, where 20 patients fulfilling King’s College criteria for poor prognosis and with uncontrolled intracranial hypertension were cooled to 32∞C.350 Six patients who were deemed unsuitable for liver transplantation died after rewarming, and 13 of the remaining 14 were successfully bridged to transplant after a mean of 32 (range 10–118) hours of cooling. Mean ICP fell from 36.5 mmHg before cooling to a plateau of 16–17 mmHg between 4 and 24 hours after cooling. Induced hypothermia also decreased arterial ammonia by 30% and ammonia delivery to the brain by 66%. These promising results must be confirmed by larger randomized studies in which potential adverse effects of hypothermia (such as increased risk of infection and cardiovascular instability) are critically assessed. Liver transplantation remains the definitive treatment for severe intracranial hypertension associated with ALF. At surgery, handling the recipient liver can cause marked increases in intracranial pressure and hemodynamic instability. Accordingly, heterotopic liver transplantation, in which the native liver is left in situ and a right lobe graft is implanted caudal to the native liver, has been advocated in patients too unstable for orthotopic transplantation.371,372
Psychomotor Agitation Psychomotor agitation frequently accompanies the neurologic syndrome associated with ALF and often heralds the development of cerebral edema, progression to grade III–IV coma, and the onset of seizures;263,373 concomitantly, psychomotor agitation exacerbates intracranial hypertension.263,361 Therefore, patients with marked agitation should be sedated. Because all forms of sedation increase the risk of respiratory depression and aspiration, it seems prudent to intubate before sedating such patients. Propofol is the drug of choice for sedating the agitated patient with ALF as it can be titrated easily and may decrease cerebral blood flow;350 one small study has suggested that propofol (50 mg/kg/min) may reduce ICP.374
Seizures The development of cerebral edema lowers the seizure threshold and may precipitate focal or generalized seizures,373,375 which in turn may exacerbate cerebral edema. Subclinical seizure activity was documented in nearly a third of patients with ALF and grade III hepatic encephalopathy;375 therefore, an electroencephalogram (EEG) should be performed on patients with grade III or IV encephalopathy to rule out subclinical status epilepticus. An EEG should be repeated if the mental status does not improve, or worsens despite stable intracranial pressure and liver function. Prophylactic phenytoin (15 mg/kg IV loading dose followed by 100 mg every 8 hours) has been suggested to decrease subclinical seizures in patients with ALF and advanced grades of hepatic encephalopathy,375 but a more recent study has found no benefit of prophylactic phenytoin.376
THERAPIES TO IMPROVE LIVER FUNCTION Methods to Protect the Liver from Injury and Promote Regeneration A successful outcome of ALF depends on both cessation of liver injury and regeneration of the injured liver – the ‘Holy Grail’ of the management of ALF. Although numerous modalities have been tried to achieve these objectives, all have failed. These include infusions of insulin and glucagon,377–379 corticosteroids,380 human
Chapter 21 ACUTE LIVER FAILURE
cross-circulation,381 extracorporeal liver perfusion,382 exchange transfusions,383,384 hemodialysis,385 and charcoal hemoperfusion.90 In early studies,129,302,306 high-volume plasmapheresis improved cerebral hemodynamics and was associated with a 35% spontaneous survival rate in patients with ALF. Other investigators, however, have not demonstrated improved survival using plasmapheresis.386,387 Prostaglandins (PG) have also been investigated as adjuncts to the treatment of ALF because of their well-documented cytoprotective effects (Table 21-19).388,389 In one study,390 17 patients with ALF or subacute fulminant liver failure due to causes other than acetaminophen toxicity were given 0.2 mg/kg/h of PGE1 initially and then 0.6 mg/kg/h intravenously for 28 days. Of this cohort, 14 had grade III or IV coma at admission; only five died, a surprisingly good outcome. Another controlled clinical trial, however, did not clearly demonstrate a benefit for PGE1,391 and the use of this agent is generally not supported by controlled clinical data.
Orthotopic Liver Transplantation Liver transplantation remains the definitive treatment for patients with severe ALF, and clearly improves both short- and long-term survival in those with grade III or IV encephalopathy.348,392–394 The success of transplantation mandates that patients with ALF be transported to a transplant center whenever feasible. The challenge remains to identify patients with a high risk of mortality with medical management and a high probability of survival with transplant (see Assessment of Prognosis, below). It is equally important to decide when not to proceed with transplantation. Patients with a poor prognosis for neurologic recovery, such as a sustained increase in ICP >40 mmHg or a decrease in CPP <40 mmHg, may not benefit from transplantation even if technically successful. The presence of septicemia or advanced multiorgan failure is also a contraindication to transplantation. Indeed, the mortality in patients with more than two-system organ failure or an APACHE II score >30 is over 60%.395,396 A transplant psychologist and an ethicist should assist in decision making when a patient presents with an intentional acetaminophen overdose or history of drug or alcohol abuse. Many international liver transplant centers have reported average survival after transplantation for ALF of about 65%, which compares favorably with medical management.348,392,394,397–403 In one of the largest studies404 a number of static and dynamic variables were
Table 21-19. Effects of Prostaglandin E on Survival in ALF Author
n
Drug and route of administration
Sinclair390 Bernau432
17 22 (HBV) 22 (drug/ indet.) 22 20 41 13
PGE1 IV, followed by PGE2 p.o. PGE IV PGE IV
O’Brien433,434 Sheener435 Sterling391
PGE1 IV PGE1 IV PGE2 IV vs placebo PGE1 IV vs placebo
Survival (%) 71 13 13 72 100 40 vs. 38 60 vs. 50
HBV, ALF due to hepatitis B virus; drug/indet., ALF due to idiosyncratic drug reaction or indeterminate etiologies.
evaluated as predictors of outcome after transplantation in 100 patients with ALF. In patients with ALF unrelated to acetaminophen (n = 79) the etiology was an important predictor of survival, with 100% of those with Wilson’s disease surviving 2 months after transplant compared to only 12.5% of those with drug-induced ALF. Of the dynamic variables, an elevated serum creatinine predicted poor outcome, as did grade III/IV encephalopathy (80% survival for those with grade I/II vs 56% for grade III/IV).405–407 Survival after liver transplantation is also adversely influenced by the utilization of suboptimal organs (fatty liver, ABO-incompatible liver).394,405,408 The outcome of liver transplantation in children is similar to that in adults.348,409–411 Although experience worldwide with adult-to-adult living donor liver transplantation is limited, the procedure has been used in occasional patients with ALF.412,413 The major hindrance to its widespread use in ALF, however, will probably remain the amount of time required to evaluate a potential donor, which often requires several days. Auxiliary liver transplantation, in which a donor liver (whole or partial) is heterotopically implanted below the native liver to provide support while regeneration of the native liver occurs, has also been explored in patients with ALF.414–416 Withdrawal of immunosuppression after regeneration of the native liver causes rejection and atrophy of the donated liver, and obviates the need for long-term immunosuppression. In one recent series, complete regeneration of the native liver was noted in 68% of 22 patients with ALF who received an auxiliary transplant.83
PROGNOSIS AND NATURAL HISTORY Patients with ALF have one of three outcomes: spontaneous recovery, liver transplantation, or death. As of June 2004, the US Acute Liver Failure Study Group Registry of nearly 700 patients recorded that roughly 45% recovered spontaneously, 25% underwent liver transplantation (of whom 13% died), and 30% died without transplantation (WM Lee, personal communication). The overall survival, with or without liver transplantation and for all major etiologies, has steadily improved with time (Figure 21-16) and is currently 66% in the US Registry. The ability to predict which patient with ALF will recover spontaneously with medical management, and who will die without transplantation, remains of paramount importance. Although liver transplantation offers hope of survival from ALF, the decision to transplant introduces the need for lifelong immunosuppression, an operative mortality of up to 30%, and the use of a scarce resource.417 Thus, universal liver transplantation for ALF cannot be endorsed. Although mortality from all causes of ALF parallels the depth of hepatic encephalopathy (>80% mortality rates for grade III/IV encephalopathy),6 spontaneous recovery occasionally follows even deepest hepatic coma;418 thus, more accurate predictors of outcome are needed. Several groups have proposed guidelines with which to select a patient for liver transplantation (Table 21-20). The most widely accepted were proposed by O’Grady et al. in 1989,3 and have become known as the King’s College criteria. Based on a retrospective review of 588 patients with ALF who were managed medically between 1973 and 1985, these authors identified poor prognostic variables in patients with ALF due to acetaminophen overdose and
403
Section III. Clinical Consequences of Liver Disease
other etiologies by multivariate analysis, and then applied the variables to a test group of 175 patients who were seen between 1986 and 1987. For patients with acetaminophen overdose, acidosis (arterial pH <7.30) on admission or the combination of a peak prothrombin time of >100 seconds, serum creatinine >3.4 mg/dl, and grade III/IV hepatic encephalopathy was highly associated with mortality without liver transplantation. Fulfillment of one of these criteria predicted 77% of the total deaths in the test group. In the
Hepatitis A
Hepatitis B
Acetaminophen
Non A non B, indeterminate
Halothane
80 Transplantation
Survival (%)
60
40
20
0 ’73-’76 ’77-’79 ’80-’82 ’83-’85 ’86-’88 ’89-’91 ’91-’93 Year Figure 21-16. Improving survival of patients with ALF at the King’s College Hospital Liver Failure Unit according to etiology. Survival represents patients with grade III/IV hepatic encephalopathy who received liver transplants as well as those who spontaneously recovered. (From Williams R, Wendon, J. Indications for orthotopic liver transplantation in fulminant liver failure. Hepatology 1994; 20:S5, with permission.)
group with ALF from non-acetaminophen etiologies, three static variables obtained on admission (etiology, age, and duration of jaundice to onset of encephalopathy >7 days), and two dynamic variables obtained during the evolution of liver failure (peak bilirubin and prothrombin time), predicted poor prognosis. The presence of a prothrombin time of >100 seconds or, in patients with a prothrombin time <100 seconds, any three of the following: age <10 or >40 years, non-A, non-B viral or idiosyncratic drug etiology, evolution from jaundice to encephalopathy of >7 days, prothrombin time >50 seconds, bilirubin >17.4 g/dl, predicted over 96% of the fatalities in the test group. The predictive accuracy of the King’s College criteria has been substantiated by other groups.86,419–421 In a recent study from Pittsburgh, however, failure to fulfill the King’s College criteria did not predict survival,422 and in the ALF Study Group Registry, the King’s College criteria predicted patient mortality with a sensitivity of only 12%.8 Other indices have been developed to improve on the King’s College criteria. One scheme was developed from a retrospective analysis of survival from hepatitis B and non-A, non-B viral ALF, with subsequent prospective validation.423 The risk score was derived from bilirubin, white blood cell count, prothrombin time, ALT, and duration of illness before encephalopathy, and demonstrated a predictive accuracy of nearly 90%. More recently, the APACHE (Acute Physiology and Chronic Health Evaluation) II score was prospectively evaluated in patients with ALF due to acetaminophen overdose.424 Calculated after the first 24 hours following admission, this scoring system is widely used to estimate risk of hospital death, and is based upon easily obtained parameters.395 An APACHE II score of >15 was highly predictive of death or the need for liver transplantation, even though prothrombin time and bilirubin, two most important indicators of hepatic necrosis, are not part of the score. Coagulation parameters also have been recommended as the best prognostic indicators for mortality in patients with ALF (Table 2120). The prothrombin time is probably the most reliable, widely
Table 21-20. Schemes for Predicting Mortality and Need for Liver Transplantation in ALF Test
Etiology of ALF
Criteria for liver transplantation
Reference
King’s College Criteria
APAP
Arterial pH < 7.30 or all of the following: PT > 100 s, and Creatinine >3.4 mg/dl, and Grade 3/4 encephalopathy PT >100 s (INR > 6.5) or any three of the following: NANB/drug/halothane etiology, Jaundice to encephalopathy > 7 d, Age < 10 or > 40 yrs PT > 50 s Bilirubin > 17.4 mg/dl Age < 30 yrs: factor V < 20% or Any age: factor V < 30% and grade 3/4 encephalopathy Gc level < 34 g/ml (normal 280–560 g/ml) Hepatocyte necrosis > 70% See reference > 1.2 mmol/l > 3.5 mmol/l Score > 15
O’Grady, 19893
Non-APAP
Factor V (Clichy Criteria)
Viral
Unbound serum Gc protein Liver biopsy Severity index Arterial phosphate Serum lactate APACHE II score
Mixed Mixed HBV, NANB APAP APAP APAP
APAP, acetaminophen; HBV, hepatitis B virus; NANB, non-A, non-B viral hepatitis; mixed, mixed etiologies.
404
Bernuau, 1986, 199129,400 Lee, 1995421 Donaldson, 1993419 Takahashi, 1994423 Schmidt, 2002196 Bernal, 2002427 Mitchell, 1998424
Chapter 21 ACUTE LIVER FAILURE
80 Survived
Died
Factor V (%)
60
40
20
0 0
1
2
3
4
Days after admission Figure 21-17. Sequential factor V levels in 22 patients with acetaminopheninduced ALF according to prognosis. Factor V levels recovered to within a normal range (60–150%) within 4 days of admission in survivors, whereas patients who died had no significant recovery of factor V levels. (From Pereira L, Langley P, Hayllar K, et al. Coagulation factor V and VIII/V ratio as predictors of outcome in paracetamol induced fulminant hepatic failure: relation to other prognostic indicators. Gut 1992;33:98, with permission.)
available parameter.3,425 Bernuau et al.29 first proposed that factor V, a liver-derived coagulation factor with a short half-life (12–24 hours), may be a more accurate indicator of the need for liver transplantation than the prothrombin time, which tends to become markedly elevated relatively late in ALF, and is disproportionately elevated in certain etiologies of ALF.417 Based on experience with HBV-induced ALF, these ‘Clichy criteria’ were later refined in a prospective study of patients with viral ALF. In patients with stage III/IV hepatic encephalopathy a factor V level of <20% in patients aged less than 30 years, or a level of <30% in patients over 30 years, predicted very high mortality (90%) and thus the need for liver transplantation.400 Although factor V levels fared nearly as well as other prognostic indicators in two subsequently studied groups of patients with non-acetaminophen ALF,420,426 no level of factor V clearly discriminated patients with acetaminophen-induced ALF who lived or died.426 Other studies, however, have found low factor V levels useful in predicting mortality from acetaminophen-induced ALF, especially when levels do not recover over the first few days of hospitalization (Figure 21-17).154 Other readily available laboratory parameters also appear to have prognostic value. In a multivariate, retrospectively analyzed group of over 100 patients with acetaminophen-induced ALF, arterial lactate concentration was identified as a predictor of death (8.5 mmol/l in those who died versus 1.4 mmol/l in those who survived).427 In a subsequent prospectively studied patient sample, a threshold of arterial lactate before volume expansion of 3.5 mmol/l had a sensitivity of 67% and specificity of 95% in predicting early death from acetaminophen-induced ALF. Serum phosphate also may improve standard prognostic criteria.196 In patients with acetaminophen-induced ALF, serum phosphate concentrations 48–72 hours after the overdose were significantly higher in those who died than in survivors (mean 2.65 vs 0.68 mmol/l, respectively), and a
threshold value of 1.2 mmol/l provided 89% sensitivity and 100% specificity. The role of liver biopsy in determining prognosis remains controversial. Scotto et al.428 first advocated estimating liver cell volume by early liver biopsy for prognosis, but found that coagulation parameters reflected the likelihood of regeneration nearly as well, and that biopsy offered no prognostic information in many patients with ALF. Despite suspected sampling variation in this and subsequent studies, Donaldson et al. have shown prognosis to be dismal with >70% hepatocyte necrosis.419 However, others have not found that the extent of necrosis predicts outcome.429,430 Other proposed prognostic criteria include serum a-fetoprotein (AFP) levels, a marker of liver regeneration, which tend to be higher in patients who recover spontaneously.29,418 An increasing trend in AFP over days 1 and 3 of admission may better predict spontaneous survival than spot levels on admission:; in the Acute Liver Failure Study Group, 71% of patients in whom AFP levels increased survived without transplantation, whereas 80% in whom levels did not increase died or required transplantation.8 Levels of Gc globulin, a plasma protein synthesized by the liver and released into the circulation after massive hepatocyte necrosis, also correlate with prognosis,421,431 but the assay is cumbersome and not widely available.
REFERENCES 1. Gimson AE, O’Grady J, Ede RJ, et al. Late onset hepatic failure: clinical, serological and histological features. Hepatology 1986;6:288–294. 2. O’Grady JG, Williams R. Classification of acute liver failure. Lancet 1993;342:743. 3. O’Grady JG, Alexander GJ, Hayllar KM, et al. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989;97:439–445. 4. Ellis AJ, Saleh M, Smith H, et al. Late-onset hepatic failure: clinical features, serology and outcome following transplantation. J Hepatol 1995;23:363–372. 5. O’Grady JG, Schalm SW, Williams R. Acute liver failure: redefining the syndromes. Lancet 1993;342:273–275. 6. Trey C, Davidson C. The management of of fulminant hepatic failure. Prog Liver Dis 1970;282–298. 7. Williams R. Classification and clinical syndromes of acute liver failure. In: Lee WM, Williams R, eds. Acute liver failure. Cambridge: Cambridge University Press, 1997: 1–9. 8. Lee WM. Acute liver failure in the United States. Semin Liver Dis 2003;23:217–226. 9. Castells A, Salmeron JM, Navasa M, et al. Liver transplantation for acute liver failure: analysis of applicability. Gastroenterology 1993;105:532–538. 10. Detre K, Belle S, Beringer K, et al. Liver transplantation for fulminant hepatic failure in the United States: October 1987 through December 1991. Clin Transplant 1994;8:274–280. 11. Dodson SF, Dehara K, Iwatsuki S. Liver transplantation for fulminant hepatic failure. ASAIO J 1994;40:86–88. 12. Acharya SK, Dasarathy S, Kumer TL, et al. Fulminant hepatitis in a tropical population: clinical course, cause, and early predictors of outcome. Hepatology 1996;23:1448–1455. 13. Masada CT, Shaw BW Jr, Zetterman RK, et al. Fulminant hepatic failure with massive necrosis as a result of hepatitis A infection. J Clin Gastroenterol 1993;17:158–162. 14. Liaw YF, Yang CY, Chu CM, et al. Appearance and persistence of hepatitis A IgM antibody in acute clinical hepatitis A observed in an outbreak. Infection 1986;14:156–158.
405
Section III. Clinical Consequences of Liver Disease
15. Tandon BN, Joshi YK, Tandon M. Acute liver failure. Experience with 145 patients. J Clin Gastroenterol 1986;8:664–668. 16. Mathiesen LR, Skinoj P, Nielsen JO, et al. Hepatitis type A, B, and non-A non-B in fulminant hepatitis. Gut 1980;21:72–77. 17. Gimson AE, White YS, Eddleston AL, et al. Clinical and prognostic differences in fulminant hepatitis type A, B and non-A non-B. Gut 1983;24:1194–1198. 18. Williams R, Wendon J. Indications for orthotopic liver transplantation in fulminant liver failure. Hepatology 1994;20:S5–10S. 19. Lee WM. Acute liver failure. N Engl J Med 1993;329:1862–1872. 20. O’Grady J. Management of acute and fulminant hepatitis A. Vaccine 1992;10(Suppl 1):S21–S23. 21. Fagan E, Yousef G, Brahm J, et al. Persistence of hepatitis A virus in fulminant hepatitis and after liver transplantation. J Med Virol 1990;30:131–136. 22. O’Grady JG. Fulminant hepatitis in patients with chronic liver disease. J Viral Hepatol 2000;7(Suppl 1):9–10. 23. Keeffe EB. Is hepatitis A more severe in patients with chronic hepatitis B and other chronic liver diseases? Am J Gastroenterol 1995;90:201–205. 24. Vento S, Garofano T, Renzini C, et al. Fulminant hepatitis associated with hepatitis A virus superinfection in patients with chronic hepatitis C. N Engl J Med 1998;338:286–290. 25. Akriviadis EA, Redeker AG. Fulminant hepatitis A in intravenous drug users with chronic liver disease. Ann Intern Med 1989;110:838–839. 26. Gane E, Sallie R, Saleh M, et al. Clinical recurrence of hepatitis A following liver transplantation for acute liver failure. J Med Virol 1995;45:35–39. 27. Lettau LA, McCarthy JG, Smith MH, et al. Outbreak of severe hepatitis due to delta and hepatitis B viruses in parenteral drug abusers and their contacts. N Engl J Med 1987;317: 1256–1262. 28. Woolf IL, El Sheikh N, Cullens H, et al. Enhanced HBsAb production in pathogenesis of fulminant viral hepatitis type B. Br Med J 1976;2:669–671. 29. Bernuau J, Goudeau A, Poynard T, et al. Multivariate analysis of prognostic factors in fulminant hepatitis B. Hepatology 1986;6:648–651. 30. Wu JC, Chen CL, Hou MC, et al. Multiple viral infection as the most common cause of fulminant and subfulminant viral hepatitis in an area endemic for hepatitis B: application and limitations of the polymerase chain reaction. Hepatology 1994;19:836–840. 31. Papaevangelou G, Tassopoulos N, Roumeliotou-Karayannis A, et al. Etiology of fulminant viral hepatitis in Greece. Hepatology 1984;4:369–372. 32. Chu CM, Liaw YF. The incidence of fulminant hepatic failure in acute viral hepatitis in Taiwan: increased risk in patients with pre-existing HBsAg carrier state. Infection 1990;18:200–203. 33. Arora NK, Nanda SK, Gulati S, et al. Acute viral hepatitis types E, A, and B singly and in combination in acute liver failure in children in north India. J Med Virol 1996;48:215–221. 34. Aziz AB, Hamid S, Iqbal S, et al. Prevalence and severity of viral hepatitis in Pakistani pregnant women: a five year hospital based study. J Pak Med Assoc 1997;47:198–201. 35. Brechot C, Bernuau J, Thiers V, et al. Multiplication of hepatitis B virus in fulminant hepatitis B. Br Med J (Clin Res Ed) 1984;288:270–271. 36. Saracco G, Macagno S, Rosina F, et al. Serologic markers with fulminant hepatitis in persons positive for hepatitis B surface antigen. A worldwide epidemiologic and clinical survey. Ann Intern Med 1988;108:380–383. 37. Wright TL, Mamish D, Combs C, et al. Hepatitis B virus and apparent fulminant non-A, non-B hepatitis. Lancet 1992;339:952–955.
406
38. Flowers MA, Heathcote J, Wanless IR, et al. Fulminant hepatitis as a consequence of reactivation of hepatitis B virus infection after discontinuation of low-dose methotrexate therapy. Ann Intern Med 1990;112:381–382. 39. Lok AS, Liang RH, Chiu EK, et al. Reactivation of hepatitis B virus replication in patients receiving cytotoxic therapy. Report of a prospective study. Gastroenterology 1991;100:182–188. 40. Rizzetto M, Bonino F, Verme G. Hepatitis delta virus infection of the liver: progress in virology, pathobiology, and diagnosis. Semin Liver Dis 1988;8:350–356. 41. Smedile A, Farci P, Verme G, et al. Influence of delta infection on severity of hepatitis B. Lancet 1982;2:945–947. 42. Hoofnagle JH. Type D (delta) hepatitis. JAMA 1989;261:1321–1325. 43. Govindarajan S, Chin KP, Redeker AG, et al. Fulminant B viral hepatitis: role of delta agent. Gastroenterology 1984;86: 1417–1420. 44. Wright TL, Hsu H, Donegan E, et al. Hepatitis C virus not found in fulminant non-A, non-B hepatitis. Ann Intern Med 1991;115:111–112. 45. Feray C, Gigou M, Samuel D, et al. Hepatitis C virus RNA and hepatitis B virus DNA in serum and liver of patients with fulminant hepatitis. Gastroenterology 1993;104:549–555. 46. Sallie R, Silva AE, Purdy M, et al. Hepatitis C and E in non-A non-B fulminant hepatic failure: a polymerase chain reaction and serological study. J Hepatol 1994;20:580–588. 47. Villamil FG, Hu KQ, Yu CH, et al. Detection of hepatitis C virus with RNA polymerase chain reaction in fulminant hepatic failure. Hepatology 1995;22:1379–1386. 48. Yanagi M, Kaneko S, Unoura M, et al. Hepatitis C virus in fulminant hepatic failure. N Engl J Med 1991;324:1895–1896. 49. Yoshiba M, Dehara K, Inoue K, et al. Contribution of hepatitis C virus to non-A, non-B fulminant hepatitis in Japan. Hepatology 1994;19:829–835. 50. Jain A, Kar P, Madan K, et al. Hepatitis C virus infection in sporadic fulminant viral hepatitis in North India: cause or cofactor? Eur J Gastroenterol Hepatol 1999;11:1231–1237. 51. Reyes GR, Purdy MA, Kim JP, et al. Isolation of a cDNA from the virus responsible for enterically transmitted non-A, non-B hepatitis. Science 1990;247:1335–1339. 52. DeCock KM, Govindarajan S, Redeker AG. Serological response to hepatitis delta virus in hepatitis D. Lancet 1987;1:1438. 53. Liang TJ, Jeffers L, Reddy RK, et al. Fulminant or subfulminant non-A, non-B viral hepatitis: the role of hepatitis C and E viruses. Gastroenterology 1993;104:556–562. 54. Nanda SK, Yalcinkaya K, Panigrahi AK, et al. Etiological role of hepatitis E virus in sporadic fulminant hepatitis. J Med Virol 1994;42:133–137. 55. Krawczynski K. Hepatitis E. Hepatology 1993;17:932–941. 56. Tsega E, Hansson BG, Krawczynski K, et al. Acute sporadic viral hepatitis in Ethiopia: causes, risk factors, and effects on pregnancy. Clin Infect Dis 1992;14:961–965. 57. Fagan EA, Ellis DS, Tovey GM, et al. Toga virus-like particles in acute liver failure attributed to sporadic non-A, non-B hepatitis and recurrence after liver transplantation. J Med Virol 1992;38:71–77. 58. Phillips MJ, Blendis LM, Poucell S, et al. Syncytial giant-cell hepatitis. Sporadic hepatitis with distinctive pathological features, a severe clinical course, and paramyxoviral features. N Engl J Med 1991;324:455–460. 59. Pappo O, Yunis E, Jordan JA, et al. Recurrent and de novo giant cell hepatitis after orthotopic liver transplantation. Am J Surg Pathol 1994;18:804–813. 60. Zuckerman AJ. The new GB hepatitis viruses. Lancet 1995;345:1453–1454. 61. Naoumov NV, Petrova EP, Thomas MG, et al. Presence of a newly described human DNA virus (TTV) in patients with liver disease. Lancet 1998;352:195–197.
Chapter 21 ACUTE LIVER FAILURE
62. Lau JY, Koukoulis G, Mieli-Vergani G, et al. Syncytial giant-cell hepatitis – a specific disease entity? J Hepatol 1992;15:216–219. 63. Saiz JC, Sans M, Mas A, et al. Hepatitis G virus infection in fulminant hepatic failure. Gut 1997;41:696–699. 64. Hadziyannis SJ. Fulminant hepatitis and the new G/GBV-C flavivirus. J Viral Hepatol 1998;5:15–19. 65. Gimenez-Barcons M, Forns X, Ampurdanes S, et al. Infection with a novel human DNA virus (TTV) has no pathogenic significance in patients with liver diseases. J Hepatol 1999;30:1028–1034. 66. Ben Ari Z, Samuel D, Zemel R, et al. Fulminant non-A–G viral hepatitis leading to liver transplantation. Arch Intern Med 2000;160:388–392. 67. Taylor RJ, Saul SH, Dowling JN, et al. Primary disseminated herpes simplex infection with fulminant hepatitis following renal transplantation. Arch Intern Med 1981;141:1519–1521. 68. Chase RA, Pottage JC Jr, Haber MH, et al. Herpes simplex viral hepatitis in adults: two case reports and review of the literature. Rev Infect Dis 1987;9:329–333. 69. Dits H, Frans E, Wilmer A, et al. Varicella-zoster virus infection associated with acute liver failure. Clin Infect Dis 1998;27:209–210. 70. Sobue R, Miyazaki H, Okamoto M, et al. Fulminant hepatitis in primary human herpesvirus-6 infection. N Engl J Med 1991;324:1290. 71. Feranchak AP, Tyson RW, Narkewicz MR, et al. Fulminant Epstein–Barr viral hepatitis: orthotopic liver transplantation and review of the literature. Liver Transpl Surg 1998;4:469–476. 72. Mutimer DJ, Shaw J, O’Donnell K, et al. Enhanced (cytomegalovirus) viral replication after transplantation for fulminant hepatic failure. Liver Transpl Surg 1997;3:506–512. 73. Samuel D, Dussaix E. Cytomegalovirus infection, fulminant hepatitis, and liver transplantation: the sides of the triangle. Liver Transpl Surg 1997;3:547–551. 74. Morales JM. Successful acyclovir therapy of severe varicella hepatitis in an adult renal transplant recipient. Am J Med 1991;90:401. 75. Michaels MG, Green M, Wald ER, et al. Adenovirus infection in pediatric liver transplant recipients. J Infect Dis 1992;165: 170–174. 76. Archer JS. Acute liver failure in pregnancy. A case report. J Reprod Med 2001;46:137–140. 77. Sokal EM, Melchior M, Cornu C, et al. Acute parvovirus B19 infection associated with fulminant hepatitis of favourable prognosis in young children. Lancet 1998;352:1739–1741. 78. Zimmerman HJ. The spectrum of hepatotoxicity. In: Zimmerman HJ, ed. Hepatotoxicity. The adverse effects of drugs and other chemicals on the liver. Philadelphia: Lippincott Williams & Wilkins, 1999: 3–9. 79. Sze G, Kaplowitz N. Drug hepatotoxicity as a cause of acute liver failure. In: Lee WMaWR, ed. Acute liver failure. Cambridge: Cambridge University Press, 1997: 19–31. 80. McKenzie R, Fried MW, Sallie R, et al. Hepatic failure and lactic acidosis due to fialuridine (FIAU), an investigational nucleoside analogue for chronic hepatitis B. N Engl J Med 1995;333:1099–1105. 81. Pessayre D, Bentata M, Degott C, et al. Isoniazid–rifampin fulminant hepatitis. A possible consequence of the enhancement of isoniazid hepatotoxicity by enzyme induction. Gastroenterology 1977;72:284–289. 82. Ransohoff DF, Jacobs G. Terminal hepatic failure following a small dose of sulfamethoxazole–trimethoprim. Gastroenterology 1981;80:816–819. 83. Chenard-Neu MP, Boudjema K, Bernuau J, et al. Auxiliary liver transplantation: regeneration of the native liver and outcome in 30 patients with fulminant hepatic failure – a multicenter European study. Hepatology 1996;23:1119–1127.
84. Ellis AJ, Wendon JA, Portmann B, et al. Acute liver damage and Ecstasy ingestion. Gut 1996;38:454–458. 85. Andreu V, Mas A, Bruguera M, et al. Ecstasy: a common cause of severe acute hepatotoxicity. J Hepatol 1998;29:394–397. 86. Anand AC, Nightingale P, Neuberger JM. Early indicators of prognosis in fulminant hepatic failure: an assessment of the King’s criteria. J Hepatol 1997;26:62–68. 87. Schiodt FV, Rochling FA, Casey DL, et al. Acetaminophen toxicity in an urban county hospital. N Engl J Med 1997;337:1112–1117. 88. Lee WM. Acetaminophen and the US Acute Liver Failure Study Group: lowering the risks of hepatic failure. Hepatology 2004;40:6–9. 89. Zimmerman HJ, Maddrey WC. Acetaminophen (paracetamol) hepatotoxicity with regular intake of alcohol: analysis of instances of therapeutic misadventure. Hepatology 1995;22: 767–773. 90. O’Grady JG, Gimson AE, O’Brien CJ, et al. Controlled trials of charcoal hemoperfusion and prognostic factors in fulminant hepatic failure. Gastroenterology 1988;94:1186–1192. 91. Mitchell JR, Jollow DJ, Potter WZ, et al. Acetaminopheninduced hepatic necrosis. I. Role of drug metabolism. J Pharmacol Exp Ther 1973;187:185–194. 92. Jollow DJ, Mitchell JR, Potter WZ, et al. Acetaminopheninduced hepatic necrosis. II. Role of covalent binding in vivo. J Pharmacol Exp Ther 1973;187:195–202. 93. Mitchell JR, Jollow DJ, Potter WZ, et al. Acetaminopheninduced hepatic necrosis. IV. Protective role of glutathione. J Pharmacol Exp Ther 1973;187:211–217. 94. Potter WZ, Davis DC, Mitchell JR, et al. Acetaminopheninduced hepatic necrosis. 3. Cytochrome P450-mediated covalent binding in vitro. J Pharmacol Exp Ther 1973;187:203–210. 95. Manyike PT, Kharasch ED, Kalhorn TF, et al. Contribution of CYP2E1 and CYP3A to acetaminophen reactive metabolite formation. Clin Pharmacol Ther 2000;67:275–282. 96. Rumack BH. Acetaminophen misconceptions. Hepatology 2004;40:10–15. 97. Makin A, Williams R. Acetaminophen-induced acute liver failure. In: Lee WMaWR, ed. Acute liver failure. Cambridge: Cambridge University Press, 1997: 32–42. 98. Seeff LB, Cuccherini BA, Zimmerman HJ, et al. Acetaminophen hepatotoxicity in alcoholics. A therapeutic misadventure. Ann Intern Med 1986;104:399–404. 99. Klein AS, Hart J, Brems JJ, et al. Amanita poisoning: treatment and the role of liver transplantation. Am J Med 1989;86:187–193. 100. O’Brien BL, Khuu. A fatal Sunday brunch: Amanita mushroom poisoning in a Gulf Coast family. Am J Gastroenterol 1996;91:581–583. 101. Broussard CN, Aggarwal A, Lacey SR, et al. Mushroom poisoning – from diarrhea to liver transplantation. Am J Gastroenterol 2001;96:3195–3198. 102. Mahler H, Pasi A, Kramer JM, et al. Fulminant liver failure in association with the emetic toxin of Bacillus cereus. N Engl J Med 1997;336:1142–1148. 103. Kaplowitz N. Hepatotoxicity of herbal remedies: insights into the intricacies of plant–animal warfare and cell death. Gastroenterology 1997;113:1408–1412. 104. Stedman C. Herbal hepatotoxicity. Semin Liver Dis 2002;22:195–206. 105. McCullough AJ, Fleming CR, Thistle JL, et al. Diagnosis of Wilson’s disease presenting as fulminant hepatic failure. Gastroenterology 1983;84:161–167. 106. Berman DH, Leventhal RI, Gavaler JS, et al. Clinical differentiation of fulminant Wilsonian hepatitis from other causes of hepatic failure. Gastroenterology 1991;100:1129–1134.
407
Section III. Clinical Consequences of Liver Disease
107. Schiodt FV, Atillasoy E, Shakil AO, et al. Etiology and outcome for 295 patients with acute liver failure in the United States. Liver Transpl Surg 1999;5:29–34. 108. Belay ED, Bresee JS, Holman RC, et al. Reye’s syndrome in the United States from 1981 through 1997. N Engl J Med 1999;340:1377–1382. 109. Monto AS. The disappearance of Reye’s syndrome – a public health triumph. N Engl J Med 1999;340:1423–1424. 110. Treem WR. Inherited and acquired syndromes of hyperammonemia and encephalopathy in children. Semin Liver Dis 1994;14:236–258. 111. Bhaduri BR, Mieli-Vergani G. Fulminant hepatic failure: pediatric aspects. Semin Liver Dis 1996;16:349–355. 112. Hamid SS, Jafri SM, Khan H, et al. Fulminant hepatic failure in pregnant women: acute fatty liver or acute viral hepatitis? J Hepatol 1996;25:20–27. 113. Castro MA, Fassett MJ, Reynolds TB, et al. Reversible peripartum liver failure: a new perspective on the diagnosis, treatment, and cause of acute fatty liver of pregnancy, based on 28 consecutive cases. Am J Obstet Gynecol 1999;181:389–395. 114. Ibdah JA, Bennett MJ, Rinaldo P, et al. A fetal fatty-acid oxidation disorder as a cause of liver disease in pregnant women. N Engl J Med 1999;340:1723–1731. 115. Powell-Jackson PR, Ede RJ, Williams R. Budd–Chiari syndrome presenting as fulminant hepatic failure. Gut 1986;27:1101–1105. 116. Irie H, Mori W. Fatal hepatic necrosis after shock. Acta Pathol Jpn 1986;36:363–374. 117. Nouel O, Henrion J, Bernuau J, et al. Fulminant hepatic failure due to transient circulatory failure in patients with chronic heart disease. Dig Dis Sci 1980;25:49–52. 118. Banerjee S, Owen C, Chopra S. Sickle cell hepatopathy. Hepatology 2001;33:1021–1028. 119. Emre S, Kitibayashi K, Schwartz ME, et al. Liver transplantation in a patient with acute liver failure due to sickle cell intrahepatic cholestasis. Transplantation 2000;69:675–676. 120. Ichai P, Huguet E, Guettier C, et al. Fulminant hepatitis after grand mal seizures: mechanisms and role of liver transplantation. Hepatology 2003;38:443–451. 121. Harrison HB, Middleton HM III, Crosby JH, et al. Fulminant hepatic failure: an unusual presentation of metastatic liver disease. Gastroenterology 1981;80:820–825. 122. Rowbotham D, Wendon J, Williams R. Acute liver failure secondary to hepatic infiltration: a single centre experience of 18 cases. Gut 1998;42:576–580. 123. Oldhafer KJ, Bornscheuer A, Fruhauf NR, et al. Rescue hepatectomy for initial graft non-function after liver transplantation. Transplantation 1999;67:1024–1028. 124. Lucke B, Mallory T. The fulminant form of epidemic hepatitis. Am J Pathol 1946;22:867. 125. Bourliere M, Le Treut YP, Arnoux D, et al. Acute Budd–Chiari syndrome with hepatic failure and obstruction of the inferior vena cava as presenting manifestations of hereditary protein C deficiency. Gut 1990;31:949–952. 126. Klein AS, Cameron JL. Diagnosis and management of the Budd–Chiari syndrome. Am J Surg 1990;160:128–133. 127. Povzun SA. [The diagnosis of acute liver failure in the surgical clinic]. Vestn Khir Im I I Grek 1993;151:93–97. 128. Ahsan T, Rab SM, Shekhani MS. Falciparum malaria or fulminant hepatic failure? J Pak Med Assoc 1993;43:206–208. 129. Krummenerl T, Lohmann J, Kamanabroo D, et al. [Treatment of acute liver failure by plasmapheresis]. Dtsch Med Wochenschr 1983;108:261–266. 130. Huo TI, Wu JC, Chiu CF, et al. Severe hyperbilirubinemia due to acute hepatitis A superimposed on a chronic hepatitis B carrier with glucose-6-phosphate dehydrogenase deficiency. Am J Gastroenterol 1996;91:158–159.
408
131. Roche-Sicot J, Benhamou JP. Acute intravascular hemolysis and acute liver failure associated as a first manifestation of Wilson’s disease. Ann Intern Med 1977;86:301–303. 132. Rudman D, DiFulco TJ, Galambos JT, et al. Maximal rates of excretion and synthesis of urea in normal and cirrhotic subjects. J Clin Invest 1973;52:2241–2249. 133. Albrecht J, Jones EA. Hepatic encephalopathy: molecular mechanisms underlying the clinical syndrome. J Neurol Sci 1999;170:138–146. 134. Clemmesen JO, Kondrup J, Ott P. Splanchnic and leg exchange of amino acids and ammonia in acute liver failure. Gastroenterology 2000;118:1131–1139. 135. de Knegt RJ, Groeneweg M, Schalm SW, et al. Encephalopathy from acute liver failure and from acute hyperammonemia in the rabbit. A clinical and biochemical study. Liver 1994;14: 25–31. 136. Conn HO. Hyperammonemia and intracranial hypertension: lying in wait for patients with hepatic disorders? Am J Gastroenterol 2000;95:814–816. 137. Conn HO. Hyperammonia and cerebral herniation: is an abnormality of ammonia metabolism responsible? Am J Gastroenterol 1999;94:3646–3649. 138. Clemmesen JO, Larsen FS, Kondrup J, et al. Cerebral herniation in patients with acute liver failure is correlated with arterial ammonia concentration. Hepatology 1999;29:648–653. 139. Arns PA, Branch RA. Prescribing for patients with liver disease. Baillières Clin Gastroenterol 1989;3:109–130. 140. Bauer TM, Ritz R, Haberthur C, et al. Prolonged sedation due to accumulation of conjugated metabolites of midazolam. Lancet 1995;346:145–147. 141. Losser MR, Payen D. Mechanisms of liver damage. Semin Liver Dis 1996;16:357–367. 142. Blei AT. Pharmacokinetic–hemodynamic interactions in cirrhosis. Semin Liver Dis 1986;6:299–308. 143. Sherding RG. Acute hepatic failure. Vet Clin North Am Small Anim Pract 1985;15:119–133. 144. Matthew H. Acute acetaminophen poisoning. Clin Toxicol 1973;6:9–11. 145. Zimmerman HJ, Ishak KG. Valproate-induced hepatic injury: analyses of 23 fatal cases. Hepatology 1982;2:591–597. 146. Munoz SJ, Maddrey WC. Major complications of acute and chronic liver disease. Gastroenterol Clin North Am 1988;17:265–287. 147. Record CO, Iles RA, Cohen RD, Williams R. Acid–base and metabolic disturbances in fulminant hepatic failure. Gut 1975;16:144–149. 148. Saibara T, Onishi S, Maeda T, et al. Arterial blood ketone body ratio as a possible indicator for predicting fulminant hepatitis in patients with acute hepatitis. Liver 1992;12:392–396. 149. Furunaga A, Tsuboi H, Okada H, et al. [Significance of AKBR as an indicator of liver injury after cardiopulmonary bypass]. Kyobu Geka 1994;47:1055–1058. 150. Charlton MR. Protein metabolism and liver disease. Baillières Clin Endocrinol Metab 1996;10:617–635. 151. Matkowitz R, Hartig W, Junghans P, et al. [Protein turnover in liver failure – determination using the stable isotope 15N]. Leber Magen Darm 1984;14:83–89. 152. Coon WW, Iob VL. The liver and protein metabolism. Major Probl Clin Surg 1964;1:127–151. 153. Byrne J. Liver function studies, part III: Tests that measure protein metabolism. Nursing 1977;7:13–14. 154. Pereira LM, Langley PG, Hayllar KM, et al. Coagulation factor V and VIII/V ratio as predictors of outcome in paracetamol induced fulminant hepatic failure: relation to other prognostic indicators. Gut 1992;33:98–102. 155. Clark R, Rake MO, Flute PT, et al. Coagulation abnormalities in acute liver failure: pathogenetic and therapeutic implications. Scand J Gastroenterol 1973;19(Suppl):63–70.
Chapter 21 ACUTE LIVER FAILURE
156. Robert A, Chazouilleres O. Prothrombin time in liver failure: time, ratio, activity percentage, or international normalized ratio? Hepatology 1996;24:1392–1394. 157. Munoz SJ. Prothrombin time in fulminant hepatic failure. Gastroenterology 1991;100:1480–1481. 158. Ostapchenko DA, Shishkina EV, Moroz VV. [Oxygen transport and consumption in patients in critical states]. Anesteziol Reanimatol 2000;2:68–72. 159. Gutierrez G, Arfeen QA. Oxygen transport and utilization. In: Dantzker DR, Scharf SM, ed. Cardiopulmonary critical care. Philadelphia: WB Saunders, 1998: 173–209. 160. Sekiyama KD, Yoshiba M, Thomson AW. Circulating proinflammatory cytokines (IL-1 beta, TNF-alpha, and IL-6) and IL-1 receptor antagonist (IL-1Ra) in fulminant hepatic failure and acute hepatitis. Clin Exp Immunol 1994;98: 71–77. 161. Czaja MJ, Geerts A, Xu J, et al. Monocyte chemoattractant protein 1 (MCP-1) expression occurs in toxic rat liver injury and human liver disease. J Leukocyte Biol 1994;55:120–126. 162. Hughes RD, Nicolaou N, Langley PG, et al. Plasma cytokine levels and coagulation and complement activation during use of the extracorporeal liver assist device in acute liver failure. Artif Organs 1998;22:854–858. 163. Iwai H, Nagaki M, Naito T, et al. Removal of endotoxin and cytokines by plasma exchange in patients with acute hepatic failure. Crit Care Med 1998;26:873–876. 164. Mizoguchi Y, Kawada N, Tsutsui H, et al. Network of cytokine and arachidonic acid cascade in acute hepatic failure. Gastroenterol Jpn 1993;28(Suppl 4):30–32. 165. Lush CW, Kvietys PR. Microvascular dysfunction in sepsis. Microcirculation 2000;7:83–101. 166. Agostoni PG, Bussotti M, Nani F, et al. Oxygen consumption. Cardiologia 1999;44:987–992. 167. De La MM, Meager A, Rolando N, et al. Tumour necrosis factor production in fulminant hepatic failure: relation to aetiology and superimposed microbial infection. Clin Exp Immunol 1990;82:479–484. 168. Wagner PD. Diffusive resistance to O2 transport in muscle. Acta Physiol Scand 2000;168:609–614. 169. Pittman RN. Oxygen supply to contracting skeletal muscle at the microcirculatory level: diffusion vs. convection. Acta Physiol Scand 2000;168:593–602. 170. Ellis A, Wendon J. Circulatory, respiratory, cerebral, and renal derangements in acute liver failure: pathophysiology and management. Semin Liver Dis 1996;16:379–388. 171. Wendon JA, Harrison PM, Keays R, et al. Arterial–venous pH differences and tissue hypoxia in patients with fulminant hepatic failure. Crit Care Med 1991;19:1362–1364. 172. Bihari D, Gimson AE, Lindridge J, et al. Lactic acidosis in fulminant hepatic failure. Some aspects of pathogenesis and prognosis. J Hepatol 1985;1:405–416. 173. Bihari DJ, Gimson AE, Williams R. Cardiovascular, pulmonary and renal complications of fulminant hepatic failure. Semin Liver Dis 1986;6:119–128. 174. Shimazu M. [Cardiorespiratory dynamics following hepatic resection]. Nippon Geka Gakkai Zasshi 1984;85: 160–174. 175. Cohn JD, Greenspan M, Goldstein CR, et al. Arteriovenous shunting in high cardiac output shock syndromes. Surg Gynecol Obstet 1968;127:282–288. 176. Versteeg P, Palsma DM, Schalm L. [Cardiac output in hepatic disorders]. Rev Int Hepatol 1965;15:691–697. 177. Lenz K, Kleinberger G, Hortnagl H, et al. [Circulatory behavior of patients with liver insufficiency]. Wien Klin Wochenschr 1985;97:469–474. 178. Trewby PN, Warren R, Contini S, et al. Incidence and pathophysiology of pulmonary edema in fulminant hepatic failure. Gastroenterology 1978;74:859–865.
179. Laggner A, Kleinberger G, Haller J, et al. [Pulmonary complications in hepatic coma]. Leber Magen Darm 1982;12:208–212. 180. Rolando N, Wade J, Davalos M, et al. The systemic inflammatory response syndrome in acute liver failure. Hepatology 2000;32:734–739. 181. Baudouin SV, Howdle P, O’Grady JG, et al. Acute lung injury in fulminant hepatic failure following paracetamol poisoning. Thorax 1995;50:399–402. 182. O’Grady J. Acute liver failure. J Roy Coll Phys Lond 1997;31:603–607. 183. Bal C, Longkumer T, Patel C, et al. Renal function and structure in subacute hepatic failure. J Gastroenterol Hepatol 2000;15:1318–1324. 184. Hoofnagle JH, Carithers RL Jr, Shapiro C, et al. Fulminant hepatic failure: summary of a workshop. Hepatology 1995;21:240–252. 185. Jain S, Pendyala P, Varma S, et al. Effect of renal dysfunction in fulminant hepatic failure. Trop Gastroenterol 2000;21: 118–120. 186. Moore K. Renal failure in acute liver failure. Eur J Gastroenterol Hepatol 1999;11:967–975. 187. Kurten K. Metabolism and function of septic kidneys. Prog Clin Biol Res 1989;308:601–605. 188. Achparaki A, Kotzampassi K, Eleftheriadis E, et al. Ultrastructural changes of the renal cortex after septic shock in rats. Histol Histopathol 1988;3:133–146. 189. Marcussen H, Asnaes S. Renal cortical necrosis. An evaluation of the possible relation to the Shwartzman reaction. Acta Pathol Microbiol Scand A 1972;80:351–356. 190. Jalan R, Hayes PC. Review article: quantitative tests of liver function. Aliment Pharmacol Ther 1995;9:263–270. 191. Sural S, Sharma RK, Gupta A, et al. Acute renal failure associated with liver disease in India: etiology and outcome. Renal Failure 2000;22:623–634. 192. Vickers C, Neuberger J, Buckels J, et al. Transplantation of the liver in adults and children with fulminant hepatic failure. J Hepatol 1988;7:143–150. 193. Detroz B, Honore P, Monami B, et al. Combined treatment of liver failure and hepatorenal syndrome with orthotopic liver transplantation. Acta Gastroenterol Belg 1992;55:350–357. 194. Alam AN, Wilkinson SP, Poston L, et al. Intracellular electrolyte abnormalities in fulminant hepatic failure. Gastroenterology 1977;72:914–917. 195. Dawson DJ, Babbs C, Warnes TW, Neary RH. Hypophosphataemia in acute liver failure. Br Med J (Clin Res Ed) 1987;295:1312–1313. 196. Schmidt LE, Dalhoff K. Serum phosphate is an early predictor of outcome in severe acetaminophen-induced hepatotoxicity. Hepatology 2002;36:659–665. 197. Liehr H, Gruen M, Thiel H. Different stimuli to disseminated intravascular coagulation in liver diseases. Thromb Diath Haemorrh 1972;28:325–326. 198. Fukui H, Uemura M, Yamada T. [DIC associated with acute liver failure]. Nippon Rinsho 1982;40:938–943. 199. Breddin HK, Kirchmaier CM. Substitution of antithrombin III. Wien Klin Wochenschr 1984;96:875–879. 200. Fujiwara K, Okita K, Akamatsu K, et al. Antithrombin III concentrate in the treatment of fulminant hepatic failure. Gastroenterol Jpn 1988;23:423–427. 201. Weston MJ, Langley PG, Rubin MH, et al. Platelet function in fulminant hepatic failure and effect of charcoal haemoperfusion. Gut 1977;18:897–902. 202. Rubin MH, Weston MJ, Bullock G, et al. Abnormal platelet function and ultrastructure in fulminant hepatic failure. Q J Med 1977;46:339–352. 203. Schiodt FV, Balko J, Schilsky M, et al. Thrombopoietin in acute liver failure. Hepatology 2003;37:558–561.
409
Section III. Clinical Consequences of Liver Disease
204. Rueff B, Benhamou JP. Acute hepatic necrosis and fulminant hepatic failure. Gut 1973;14:805–815. 205. Gazzard BG, Clark R, Borirakchanyavat V, et al. A controlled trial of heparin therapy in the coagulation defect of paracetamolinduced hepatic necrosis. Gut 1974;15:89–93. 206. Rolando N, Philpott-Howard J, Williams R. Bacterial and fungal infection in acute liver failure. Semin Liver Dis 1996;16: 389–402. 207. Rolando N, Harvey F, Brahm J, et al. Prospective study of bacterial infection in acute liver failure: an analysis of fifty patients. Hepatology 1990;11:49–53. 208. Wyke RJ, Canalese JC, Gimson AE, et al. Bacteraemia in patients with fulminant hepatic failure. Liver 1982;2:45–52. 209. Goularte TA, Craven DE. Results of a survey of infection control practices for respiratory therapy equipment. Infect Control 1986;7:327–330. 210. Wyke RJ. Bacterial infections complicating liver disease. Baillières Clin Gastroenterol 1989;3:187–210. 211. Wyke RJ, Rajkovic IA, Eddleston AL, et al. Defective opsonisation and complement deficiency in serum from patients with fulminant hepatic failure. Gut 1980;21:643–649. 212. Altin M, Rajkovic IA, Hughes RD, et al. Neutrophil adherence in chronic liver disease and fulminant hepatic failure. Gut 1983;24:746–750. 213. Bailey RJ, Woolf IL, Cullens H, et al. Metabolic inhibition of polymorphonuclear leucocytes in fulminant hepatic failure. Lancet 1976;1:1162–1163. 214. Ede RJ, Moore KP, Marshall WJ, et al. Frequency of pancreatitis in fulminant hepatic failure using isoenzyme markers. Gut 1988;29:778–781. 215. Navasa M, Garcia-Pagan JC, Bosch J, et al. Portal hypertension in acute liver failure. Gut 1992;33:965–968. 216. Lebrec D, Nouel O, Bernuau J, et al. Portal hypertension in fulminant viral hepatitis. Gut 1980;21:962–964. 217. Gazzard BG, Portmann B, Murray-Lyon IM, et al. Causes of death in fulminant hepatic failure and relationship to quantitative histological assessment of parenchymal damage. Q J Med 1975;44:615–626. 218. Vaquero J, Chung C, Cahill ME, et al. Pathogenesis of hepatic encephalopathy in acute liver failure. Semin Liver Dis 2003;23:259–269. 219. Blei AT. Brain edema and portal–systemic encephalopathy. Liver Transpl 2000;6(4 Suppl 1):S14–S20. 220. Lockwood AH, Ginsberg MD, Rhoades HM, et al. Cerebral glucose metabolism after portacaval shunting in the rat. Patterns of metabolism and implications for the pathogenesis of hepatic encephalopathy. J Clin Invest 1986;78:86–95. 221. Cordoba J, Blei AT. Brain edema and hepatic encephalopathy. Semin Liver Dis 1996;16:271–280. 222. Zieve FJ, Zieve L, Doizaki WM, et al. Synergism between ammonia and fatty acids in the production of coma: implications for hepatic coma. J Pharmacol Exp Ther 1974;191:10–16. 223. Zieve L, Doizaki WM, Zieve J. Synergism between mercaptans and ammonia or fatty acids in the production of coma: a possible role for mercaptans in the pathogenesis of hepatic coma. J Lab Clin Med 1974;83:16–28. 224. Baker BL, Morrow AL, Vergalla J, et al. Gamma-aminobutyric acid (GABAA) receptor-function in a rat model of hepatic encephalopathy. Metab Brain Dis 1990;5:185–193. 225. Sillanpaa M, Makela AL, Koivikko A. Acute liver failure and encephalopathy (Reye’s syndrome?) during salicylate therapy. Acta Paediatr Scand 1975;64:877–880. 226. Blei A, Dux S, Grosskopf I, et al. Hyperammonemia complicating mesenteric vein thrombosis. Am J Med 1986;80:299–300. 227. Banister EW, Cameron BJ. Exercise-induced hyperammonemia: peripheral and central effects. Int J Sports Med 1990;11 Suppl 2:S129–S142.
410
228. Blei AT, Olafsson S, Therrien G, et al. Ammonia-induced brain edema and intracranial hypertension in rats after portacaval anastomosis. Hepatology 1994;19:1437–1444. 229. Bosman DK, Deutz NE, De Graaf AA, et al. Changes in brain metabolism during hyperammonemia and acute liver failure: results of a comparative 1H-NMR spectroscopy and biochemical investigation. Hepatology 1990;12:281–290. 230. Lockwood AH, McDonald JM, Reiman RE, et al. The dynamics of ammonia metabolism in man. Effects of liver disease and hyperammonemia. J Clin Invest 1979;63:449–460. 231. Lockwood AH, Yap EW, Rhoades HM, et al. Altered cerebral blood flow and glucose metabolism in patients with liver disease and minimal encephalopathy. J Cereb Blood Flow Metab 1991;11:331–336. 232. Lockwood AH. Acute and chronic hyperosmolality. Effects on cerebral amino acids and energy metabolism. Arch Neurol 1975;32:62–64. 233. Taylor-Robinson SD, Sargentoni J, Mallalieu RJ, et al. Cerebral phosphorus-31 magnetic resonance spectroscopy in patients with chronic hepatic encephalopathy. Hepatology 1994;20: 1173–1178. 234. Larsen FS, Ejlersen E, Clemmesen JO, et al. Preservation of cerebral oxidative metabolism in fulminant hepatic failure: an autoregulation study. Liver Transpl Surg 1996;2:348–353. 235. Hindfelt B, Plum F, Duffy TE. Effect of acute ammonia intoxication on cerebral metabolism in rats with portacaval shunts. J Clin Invest 1977;59:386–396. 236. Basile AS. The contribution of endogenous benzodiazepine receptor ligands to the pathogenesis of hepatic encephalopathy. Synapse 1991;7:141–150. 237. Baraldi M, Zeneroli ZL. Experimental hepatic encephalopathy: changes in the binding of gamma-aminobutyric acid. Science 1982;216:427–429. 238. Ferenci P, Jacobs R, Pappas SC, et al. Enzymes of cerebral GABA metabolism and synaptosomal GABA uptake in acute liver failure in the rabbit: evidence for decreased cerebral GABA–transaminase activity. J Neurochem 1984;42: 1487–1490. 239. Jones EA. Pathogenesis of hepatic encephalopathy. Clin Liver Dis 2000;4:467–485. 240. Bosman DK, Deutz NE, Maas MA, et al. Amino acid release from cerebral cortex in experimental acute liver failure, studied by in vivo cerebral cortex microdialysis. J Neurochem 1992;59:591–599. 241. Erdem G, Caglar M, Ceyhan M, et al. Hepatic mucormycosis in a child with fulminant hepatic failure. Turk J Pediatr 1996;38:511–514. 242. Lavoie J, Giguere JF, Layrargues GP, et al. Amino acid changes in autopsied brain tissue from cirrhotic patients with hepatic encephalopathy. J Neurochem 1987;49:692–697. 243. Larse FS, Gottstein J, Blei AT. Cerebral hyperemia and nitric oxide synthase in rats with ammonia-induced brain edema. J Hepatol 2001;34:548–554. 244. Aggarwal S, Kramer D, Yonas H, et al. Cerebral hemodynamic and metabolic changes in fulminant hepatic failure: a retrospective study. Hepatology 1994;19:80–87. 245. Blei AT, Larsen FS. Pathophysiology of cerebral edema in fulminant hepatic failure. J Hepatol 1999;31:771–776. 246. Larsen FS, Ejlersen E, Hansen BA, et al. Functional loss of cerebral blood flow autoregulation in patients with fulminant hepatic failure. J Hepatol 1995;23:212–217. 247. Wendon JA, Harrison PM, Keays R, et al. Cerebral blood flow and metabolism in fulminant liver failure. Hepatology 1994;19:1407–1413. 248. Cordoba J, Blei AT. Brain edema and fulminant hepatic failure. Trop Gastroenterol 1994;15:45–54. 249. Traber PG, Dal Canto M, Ganger DR, et al. Electron microscopic evaluation of brain edema in rabbits with galactosamine-induced fulminant hepatic failure: ultrastructure
Chapter 21 ACUTE LIVER FAILURE
250.
251.
252.
253. 254.
255.
256.
257.
258. 259. 260.
261.
262.
263. 264.
265.
266.
267.
268. 269.
270.
271.
and integrity of the blood–brain barrier. Hepatology 1987;7:1272–1277. Potvin M, Finlayson MH, Hinchey EJ, et al. Cerebral abnormalities in hepatectomized rats with acute hepatic coma. Lab Invest 1984;50:560–564. Blei AT. Cerebral edema and intracranial hypertension in acute liver failure: distinct aspects of the same problem. Hepatology 1991;13:376–379. Butterworth RF. Hepatic encephalopathy and brain edema in acute hepatic failure: does glutamate play a role? Hepatology 1997;25:1032–1034. Cordoba J. Glutamine, myo-inositol, and brain edema in acute liver failure. Hepatology 1996;23:1291–1292. Cordoba J, Gottstein J, Blei AT. Glutamine, myo-inositol, and organic brain osmolytes after portocaval anastomosis in the rat: implications for ammonia-induced brain edema. Hepatology 1996;24:919–923. McManus ML, Churchwell KB, Strange K. Regulation of cell volume in health and disease. N Engl J Med 1995;333:1260–1266. McConnell JR, Antonson DL, Ong CS, et al. Proton spectroscopy of brain glutamine in acute liver failure. Hepatology 1995;22:69–74. Haussinger D, Laubenberger J, vom Dahl S, et al. Proton magnetic resonance spectroscopy studies on human brain myoinositol in hypo-osmolarity and hepatic encephalopathy. Gastroenterology 1994;107:1475–1480. Vale JA, Proudfoot AT. Paracetamol (acetaminophen) poisoning. Lancet 1995;346:547–552. Aldersley MA, O’Grady JG. Hepatic disorders. Features and appropriate management. Drugs 1995;49:83–102. Mackay IR. Treatment of chronic active hepatitis and other liver diseases with corticosteroid agents. Med J Aust 1987;146:370–374. Giacchino R, Marazzi MG, Barabino A, et al. Syndromic variability of Wilson’s disease in children. Clinical study of 44 cases. Ital J Gastroenterol Hepatol 1997;29:155–161. Enomoto K, Ishibashi H, Irie K, et al. Fulminant hepatic failure without evidence of cirrhosis in a case of Wilson’s disease. Jpn J Med 1989;28:80–84. Munoz SJ. Difficult management problems in fulminant hepatic failure. Semin Liver Dis 1993;13:395–413. Ruprah M, Mant TG, Flanagan RJ. Acute carbon tetrachloride poisoning in 19 patients: implications for diagnosis and treatment. Lancet 1985;1:1027–1029. Smilkstein MJ, Knapp GL, Kulig KW, et al. Efficacy of oral N-acetylcysteine in the treatment of acetaminophen overdose. Analysis of the national multicenter study (1976 to 1985). N Engl J Med 1988;319:1557–1562. Mitchell JR, Thorgeirsson SS, Potter WZ, et al. Acetaminopheninduced hepatic injury: protective role of glutathione in man and rationale for therapy. Clin Pharmacol Ther 1974;16:676–684. Harrison PM, Wendon JA, Gimson AE, et al. Improvement by acetylcysteine of hemodynamics and oxygen transport in fulminant hepatic failure. N Engl J Med 1991;324: 1852–1857. Rumack BH, Matthew H. Acetaminophen poisoning and toxicity. Pediatrics 1975;55:871–876. Keays R, Harrison PM, Wendon JA, et al. Intravenous acetylcysteine in paracetamol induced fulminant hepatic failure: a prospective controlled trial. Br Med J 1991;303:1026–1029. Makin AJ, Wendon J, Williams R. A 7-year experience of severe acetaminophen-induced hepatotoxicity (1987–1993). Gastroenterology 1995;109:1907–1916. Makin AJ, Williams R. Acetaminophen-induced hepatotoxicity: predisposing factors and treatments. Adv Intern Med 1997;42:453–483.
272. Smilkstein MJ, Bronstein AC, Linden C, et al. Acetaminophen overdose: a 48-hour intravenous N-acetylcysteine treatment protocol. Ann Emerg Med 1991;20:1058–1063. 273. Perry HE, Shannon MW. Efficacy of oral versus intravenous Nacetylcysteine in acetaminophen overdose: results of an openlabel, clinical trial. J Pediatr 1998;132:149–152. 274. Floersheim GL, Eberhard M, Tschumi P, et al. Effects of penicillin and silymarin on liver enzymes and blood clotting factors in dogs given a boiled preparation of Amanita phalloides. Toxicol Appl Pharmacol 1978;46:455–462. 275. Koppel C. Clinical symptomatology and management of mushroom poisoning. Toxicon 1993;31:1513–1540. 276. Fahy RJ, Crouser E, Pacht ER. Herpes simplex type 2 causing fulminant hepatic failure. South Med J 2000;93:1212–1216. 277. Nzerue MC. Lassa fever: review of virology, immunopathogenesis, and algorithms for control and therapy. Cent Afr J Med 1992;38:247–252. 278. Holmes GP, McCormick JB, Trock SC, et al. Lassa fever in the United States. Investigation of a case and new guidelines for management. N Engl J Med 1990;323:1120–1123. 279. Srivastava A, Khanduri A, Lakhtakia S, et al. Falciparum malaria with acute liver failure. Trop Gastroenterol 1996;17: 172–174. 280. Oster JR, Singer I. Hyponatremia, hyposmolality, and hypotonicity: tables and fables. Arch Intern Med 1999;159:333–336. 281. Larner AJ, Vickers CR, Adu D, et al. Correction of severe hyponatraemia by continuous arteriovenous haemofiltration before liver transplantation. Br Med J 1988;297:1514–1515. 282. Knochel JP. Does hypophosphatemia play a role in acute liver failure? Hepatology 1989;9:504–505. 283. Ostapowicz G, Lee WM. Acute hepatic failure: a Western perspective. J Gastroenterol Hepatol 2000;15:480–488. 284. Oswiecimski P, Khan A, Salwan R, et al. Profound hypocalcemia in fulminant hepatic failure. Am J Gastroenterol 2000;95:824–825. 285. Meier-Kriesche HU, Finkel KW, Gitomer JJ, et al. Unexpected severe hypocalcemia during continuous venovenous hemodialysis with regional citrate anticoagulation. Am J Kidney Dis 1999;33:e8. 286. O’Keefe SJ, El Zayadi AR, Carraher TE, et al. Malnutrition and immuno-incompetence in patients with liver disease. Lancet 1980;2:615–617. 287. O’Keefe SJ, Williams R, Calne RY. ‘Catabolic’ loss of body protein after human liver transplantation. Br Med J 1980;280:1107–1108. 288. Chase RA, Davies M, Trewby PN, et al. Plasma amino acid profiles in patients with fulminant hepatic failure treated by repeated polyacrylonitrile membrane hemodialysis. Gastroenterology 1978;75:1033–1040. 289. Michalak A, Butterworth RF. Selective increases of extracellular brain concentrations of aromatic and branched-chain amino acids in relation to deterioration of neurological status in acute (ischemic) liver failure. Metab Brain Dis 1997;12:259–269. 290. Alexander WF, Spindel E, Harty RF, et al. The usefulness of branched chain amino acids in patients with acute or chronic hepatic encephalopathy. Am J Gastroenterol 1989;84:91–96. 291. Campollo O, Sprengers D, McIntyre N. The BCAA/AAA ratio of plasma amino acids in three different groups of cirrhotics. Rev Invest Clin 1992;44:513–518. 292. Fischer JE. Branched-chain-enriched amino acid solutions in patients with liver failure: an early example of nutritional pharmacology. J Parenter Enteral Nutr 1990;14(5 Suppl):249S–256S. 293. Nishizaki T, Takenaka K, Yanaga K, et al. Nutritional support after hepatic resection: a randomized prospective study. Hepatogastroenterology 1996;43:608–613.
411
Section III. Clinical Consequences of Liver Disease
294. Chin SE, Shepherd RW, Thomas BJ, et al. Nutritional support in children with end-stage liver disease: a randomized crossover trial of a branched-chain amino acid supplement. Am J Clin Nutr 1992;56:158–163. 295. Munoz SJ. Nutritional therapies in liver disease. Semin Liver Dis 1991;11:278–291. 296. Rossi-Fanelli F, Angelico M, Cangiano C, et al. Effect of glucose and/or branched chain amino acid infusion on plasma amino acid imbalance in chronic liver failure. J Parenter Enteral Nutr 1981;5:414–419. 297. O’Keefe SJ, Ogden J, Dicker J. Enteral and parenteral branched chain amino acid-supplemented nutritional support in patients with encephalopathy due to alcoholic liver disease. J Parenter Enteral Nutr 1987;11:447–453. 298. Nagayama M, Okuno M, Takai T, et al. Effect of fat emulsion in patients with liver disorders. Nutrition 1991;7:267–270. 299. Record CO, Chase RA, Alberti KG, et al. Disturbances in glucose metabolism in patients with liver damage due to paracetamol overdose. Clin Sci Mol Med 1975;49:473–479. 300. Pereira SP, Langley PG, Williams R. The management of abnormalities of hemostasis in acute liver failure. Semin Liver Dis 1996;16:403–414. 301. Kondrup J, Almdal T, Vilstrup H, et al. High volume plasma exchange in fulminant hepatic failure. Int J Artif Organs 1992;15:669–676. 302. Freeman JG, Matthewson K, Record CO. Plasmapheresis in acute liver failure. Int J Artif Organs 1986;9:433–438. 303. Hart GK. Plasmapheresis in intensive care. Part 2 of 2: indications for plasmapheresis and plasma exchange in the intensive care unit. Intens Care World 1990;7:80–84. 304. Clemmesen JO, Larsen FS, Ejlersen E, et al. Haemodynamic changes after high-volume plasmapheresis in patients with chronic and acute liver failure. Eur J Gastroenterol Hepatol 1997;9:55–60. 305. Clemmesen JO, Gerbes AL, Gulberg V, et al. Hepatic blood flow and splanchnic oxygen consumption in patients with liver failure. Effect of high-volume plasmapheresis. Hepatology 1999;29:347–355. 306. Larsen FS, Hansen BA, Jorgensen LG, et al. Cerebral blood flow velocity during high volume plasmapheresis in fulminant hepatic failure. Int J Artif Organs 1994;17:353–361. 307. Larsen FS, Hansen BA, Ejlersen E, et al. Cerebral blood flow, oxygen metabolism and transcranial Doppler sonography during high-volume plasmapheresis in fulminant hepatic failure. Eur J Gastroenterol Hepatol 1996;8:261–265. 308. Munoz SJ, Ballas SK, Moritz MJ, et al. Perioperative management of fulminant and subfulminant hepatic failure with therapeutic plasmapheresis. Transplant Proc 1989;21:3535–3536. 309. Larsen FS, Hansen BA, Jorgensen LG, et al. High-volume plasmapheresis and acute liver transplantation in fulminant hepatic failure. Transplant Proc 1994;26:1788. 310. Rake MO, Flute PT, Shilkin KB, et al. Early and intensive therapy of intravascular coagulation in acute liver failure. Lancet 1971;2:1215–1218. 311. Weston MJ, Talbot IC, Horoworth PJ, et al. Frequency of arrhythmias and other cardiac abnormalities in fulminant hepatic failure. Br Heart J 1976;38:1179–1188. 312. Vincent JL, Bihari DJ. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med 1994;331:1160–1161. 313. Clemmesen JO, Galatius S, Skak C, et al. The effect of increasing blood pressure with dopamine on systemic, splanchnic, and lower extremity hemodynamics in patients with acute liver failure. Scand J Gastroenterol 1999;34:921–927. 314. Wendon JA, Harrison PM, Keays R, et al. Effects of vasopressor agents and epoprostenol on systemic hemodynamics and oxygen
412
315. 316.
317.
318.
319.
320.
321.
322.
323. 324.
325.
326.
327.
328.
329.
330. 331.
332. 333.
334. 335.
transport in fulminant hepatic failure. Hepatology 1992;15:1067–1071. Davies SC, Machin SJ. Prostacyclin (PGI2). Intens Care Med 1983;9:49–52. Fujiwara K, Mochida S, Ohno A, et al. Use of prostaglandin I2 analog in treatment of massive hepatic necrosis associated with endothelial cell injury and diffuse sinusoidal fibrin deposition. Dig Dis Sci 1995;40:41–47. Sawada M. [The beneficial effects of PGI2 analogue (OP-41483alpha-CD) on hepatic dysfunction after warm ischemic damage]. Nippon Geka Gakkai Zasshi 1994;95:83–93. Guarner F, Boughton-Smith NK, Blackwell GJ, et al. Reduction by prostacyclin of acetaminophen-induced liver toxicity in the mouse. Hepatology 1988;8:248–253. Burgunder JM, Varriale A, Lauterburg BH. Effect of Nacetylcysteine on plasma cysteine and glutathione following paracetamol administration. Eur J Clin Pharmacol 1989;36:127–131. Devlin J, Ellis AE, McPeake J, et al. N-acetylcysteine improves indocyanine green extraction and oxygen transport during hepatic dysfunction. Crit Care Med 1997;25:236–242. Harrison PM, Keays R, Bray GP, et al. Improved outcome of paracetamol-induced fulminant hepatic failure by late administration of acetylcysteine. Lancet 1990;335: 1572–1573. Ben Ari Z, Vaknin H, Tur-Kaspa R. N-acetylcysteine in acute hepatic failure (non-paracetamol-induced). Hepatogastroenterology 2000;47:786–789. Aubier M, Trippenbach T, Roussos C. Respiratory muscle fatigue during cardiogenic shock. J Appl Physiol 1981;51:499–508. Bonnet F, Richard C, Glaser P, et al. Changes in hepatic flow induced by continuous positive pressure ventilation in critically ill patients. Crit Care Med 1982;10:703–705. Yousif-Kadaru AG, Wyke RJ, Rajkovic IA, et al. Defects of serum chemoattraction and polymorphonuclear leucocyte movement in patients with primary hepatocellular carcinoma. Clin Exp Immunol 1982;49:467–473. Larcher VF, Wyke RJ, Mowat AP, et al. Bacterial and fungal infection in children with fulminant hepatic failure: possible role of opsonisation and complement deficiency. Gut 1982;23:1037–1043. Wade JJ, Rolando N, Hayllar K, et al. Bacterial and fungal infections after liver transplantation: an analysis of 284 patients. Hepatology 1995;21:1328–1336. Rolando N, Harvey F, Brahm J, et al. Fungal infection: a common, unrecognised complication of acute liver failure. J Hepatol 1991;12:1–9. Rolando N, Gimson A, Wade J, et al. Prospective controlled trial of selective parenteral and enteral antimicrobial regimen in fulminant liver failure. Hepatology 1993;17:196–201. Salmeron JM, Tito L, Rimola A, et al. Selective intestinal decontamination in the prevention of bacterial infection in patients with acute liver failure. J Hepatol 1992;14:280–285. Rolando N, Clapperton M, Wade J, et al. Administering granulocyte colony-stimulating factor to acute liver failure patients corrects neutrophil defects. Eur J Gastroenterol Hepatol 2000;12:1323–1328. Davenport A. The management of renal failure in patients at risk of cerebral edema/hypoxia. New Horiz 1995;3:717–724. Davenport A, Will EJ, Davison AM. Effect of renal replacement therapy on patients with combined acute renal and fulminant hepatic failure. Kidney Int 1993;41(Suppl):S245–S251. Davenport A. Haemofiltration in patients with fulminant hepatic failure. Lancet 1991;338:1604. Davenport A, Will EJ, Davidson AM. Improved cardiovascular stability during continuous modes of renal replacement therapy in critically ill patients with acute hepatic and renal failure. Crit Care Med 1993;21:328–338.
Chapter 21 ACUTE LIVER FAILURE
336. Davenport A, Will EJ, Davison AM, et al. Changes in intracranial pressure during haemofiltration in oliguric patients with grade IV hepatic encephalopathy. Nephron 1989;53:142–146. 337. Davenport A, Will EJ, Davison AM. Early changes in intracranial pressure during haemofiltration treatment in patients with grade 4 hepatic encephalopathy and acute oliguric renal failure. Nephrol Dial Transplant 1990;5:192–198. 338. Duvoux C, Zanditenas D, Hezode C, et al. Effects of noradrenalin and albumin in patients with type I hepatorenal syndrome: a pilot study. Hepatology 2002;36:374–380. 339. Shawcross DL, Davies NA, Mookerjee RP, et al. Worsening of cerebral hyperemia by the administration of terlipressin in acute liver failure with severe encephalopathy. Hepatology 2004;39:471–475. 340. Selden C, Shariat A, McCloskey P, et al. Three-dimensional in vitro cell culture leads to a marked upregulation of cell function in human hepatocyte cell lines – an important tool for the development of a bioartificial liver machine. Ann NY Acad Sci 1999;875:353–363. 341. Bihari DJ, Smithies M, Pozniak A, et al. A comparison of direct and indirect measurements of oxygen delivery and consumption: the effects of prostacyclin in two human volunteers. Scand J Clin Lab Invest 1987;188(Suppl):37–45. 342. Bihari DJ, Tinker J. The therapeutic value of vasodilator prostaglandins in multiple organ failure associated with sepsis. Intens Care Med 1988;15:2–7. 343. Vaquero J, Polson J, Chung C, et al. Infection and the progression of hepatic encephalopathy in acute liver failure. Gastroenterology 2003;125:755–764. 344. Cordoba J, Blei AT. Cerebral edema and intracranial pressure monitoring. Liver Transpl Surg 1995;1:187–194. 345. Wijdicks EF, Plevak DJ, Rakela J, et al. Clinical and radiologic features of cerebral edema in fulminant hepatic failure. Mayo Clin Proc 1995;70:119–124. 346. Munoz SJ, Robinson M, Northrup B, et al. Elevated intracranial pressure and computed tomography of the brain in fulminant hepatocellular failure. Hepatology 1991;13:209–212. 347. Lidofsky SD, Bass NM, Prager MC, et al. Intracranial pressure monitoring and liver transplantation for fulminant hepatic failure. Hepatology 1992;16:1–7. 348. Ascher NL, Lake JR, Emond JC, et al. Liver transplantation for fulminant hepatic failure. Arch Surg 1993;128:677–682. 349. Keays RT, Alexander GJ, Williams R. The safety and value of extradural intracranial pressure monitors in fulminant hepatic failure. J Hepatol 1993;18:205–209. 350. Jalan R. Intracranial hypertension in acute liver failure: pathophysiological basis of rational management. Semin Liver Dis 2003;23:271–282. 351. Blei AT, Olafsson S, Webster S, et al. Complications of intracranial pressure monitoring in fulminant hepatic failure. Lancet 1993;341:157–158. 352. Nyberg SL, Cerra FB, Gruetter R. Brain lactate by magnetic resonance spectroscopy during fulminant hepatic failure in the dog. Liver Transpl Surg 1998;4:158–165. 353. Davies MH, Mutimer D, Lowes J, et al. Recovery despite impaired cerebral perfusion in fulminant hepatic failure. Lancet 1994;343:1329–1330. 354. Larsen FS, Hansen BA, Blei AT. Intensive care management of patients with acute liver failure with emphasis on systemic hemodynamic instability and cerebral edema: a critical appraisal of pathophysiology. Can J Gastroenterol 2000;14(Suppl D):105D–111D. 355. Donovan JP, Schafer DF, Shaw BW Jr, et al. Cerebral oedema and increased intracranial pressure in chronic liver disease. Lancet 1998;351:719–721. 356. Munoz SJ, Moritz MJ, Martin P, et al. Relationship between cerebral perfusion pressure and systemic hemodynamics in
357.
358.
359.
360.
361.
362.
363.
364.
365.
366.
367.
368.
369.
370.
371. 372.
373.
374.
375. 376.
377.
fulminant hepatic failure. Transplant Proc 1993;25: 1776–1778. Strauss G, Hansen BA, Knudsen GM, et al. Hyperventilation restores cerebral blood flow autoregulation in patients with acute liver failure. J Hepatol 1998;28:199–203. Procaccio F, Stocchetti N, Citerio G, et al. Guidelines for the treatment of adults with severe head trauma (part II). Criteria for medical treatment. J Neurosurg Sci 2000;44: 11–18. Murphy N, Auzinger G, Bernel W, et al. The effect of hypertonic sodium chloride on intracranial pressure in patients with acute liver failure. Hepatology 2004;39:464–470. Blei AT. Brain edema and intracranial hypertension: a focus for the use of liver support systems. Artif Organs 1997;21:1182–1184. Munoz SJ, Moritz MJ, Bell R, et al. Factors associated with severe intracranial hypertension in candidates for emergency liver transplantation. Transplantation 1993;55:1071–1074. Webster S, Gottstein J, Levy R, et al. Intracranial pressure waves and intracranial hypertension in rats with ischemic fulminant hepatic failure. Hepatology 1991;14(4 Pt 1): 715–720. Canalese J, Gimson AE, Davis C, et al. Controlled trial of dexamethasone and mannitol for the cerebral oedema of fulminant hepatic failure. Gut 1982;23:625–629. Davenport A, Will EJ, Davison AM, et al. Changes in intracranial pressure during machine and continuous haemofiltration. Int J Artif Organs 1989;12:439–444. Forbes A, Alexander GJ, O’Grady JG, et al. Thiopental infusion in the treatment of intracranial hypertension complicating fulminant hepatic failure. Hepatology 1989;10:306–310. Jalan R, Pollok A, Shah SH, et al. Liver derived proinflammatory cytokines may be important in producing intracranial hypertension in acute liver failure. J Hepatol 2002;37:536–538. Cordoba J, Crespin J, Gottstein J, et al. Mild hypothermia modifies ammonia-induced brain edema in rats after portacaval anastomosis. Gastroenterology 1999;116:686–693. Blei A. Hypothermia for fulminant hepatic failure: a cool approach to a burning problem. Liver Transpl 2000;6: 245–247. Roberts DR, Manas D. Induced hypothermia in the management of cerebral oedema secondary to fulminant liver failure. Clin Transplant 1999;13:545–547. Jalan R, Damink SW, Deutz NE, et al. Moderate hypothermia for uncontrolled intracranial hypertension in acute liver failure. Lancet 1999;354:1164–1168. Shaw BW Jr. Auxiliary liver transplantation for acute liver failure. Liver Transpl Surg 1995;1:194–200. Sudan DL, Shaw BW Jr, Fox IJ, et al. Long-term follow-up of auxiliary orthotopic liver transplantation for the treatment of fulminant hepatic failure. Surgery 1997;122:771–777. Decell MK, Gordon JB, Silver K, et al. Fulminant hepatic failure associated with status epilepticus in children: three cases and a review of potential mechanisms. Intens Care Med 1994;20:375–378. Wijdicks EF, Nyberg SL. Propofol to control intracranial pressure in fulminant hepatic failure. Transplant Proc 2002;34:1220–1222. Ellis AJ, Wendon JA, Williams R. Subclinical seizure activity and prophylactic phenytoin infusion in acute liver failure: a controlled clinical trial. Hepatology 2000;32:536–541. Bhatia V, Batra Y, Acharya SK. Prophylactic phenytoin does not improve cerebral edema or survival in acute liver failure – a controlled clinical trial. J Hepatol 2004;41:89–96. Harrison PM, Hughes RD, Forbes A, et al. Failure of insulin and glucagon infusion to stimulate liver regeneration in fulminant hepatic failure. J Hepatol 1990;10:332–336.
413
Section III. Clinical Consequences of Liver Disease
378. Woolf GM, Redeker AG. Treatment of fulminant hepatic failure with insulin and glucagon. A randomized, controlled trial. Dig Dis Sci 1991;36:92–96. 379. Zielinska W, Jasiel M, Wojciechowicz G. [Insulin and glucagon in the treatment of acute liver failure]. Pol Arch Med Wewn 1981;66:401–411. 380. European Association for the Study of the Liver. Randomised trial of steroid therapy in acute liver failure. Report from the European Association for the Study of the Liver (EASL). Gut 1979;20:620–623. 381. Sicot C, Frejaville JP, Roche J, et al. [Six cases of severe hepatitis treated by interhuman crossed circulation]. Ann Med Interne (Paris) 1971;122:381–387. 382. Parbhoo SP, Chalstrey LJ, Adjukiewicz AB, et al. Extracorporeal perfusion of pig liver in the treatment of acute liver failure. Br J Surg 1971;58:746–748. 383. Ayyub M, Barlas S, Lubbad E. Usefulness of exchange transfusion in acute liver failure due to severe falciparum malaria. Am J Gastroenterol 2000;95:802–804. 384. McGuire BM, Sielaff TD, Nyberg SL, et al. Review of support systems used in the management of fulminant hepatic failure. Dig Dis 1995;13:379–388. 385. Denis J, Opolon P, Nusinovici V, et al. Treatment of encephalopathy during fulminant hepatic failure by haemodialysis with high permeability membrane. Gut 1978;19:787–793. 386. Lepore MJ, Martel AJ. Plasmapheresis with plasma exchange in hepatic coma. Methods and results in five patients with acute fulminant hepatic necrosis. Ann Intern Med 1970;72:165–174. 387. Akamatsu K, Tada K. Plasma exchange therapy in patients with fulminant hepatic failure and limitations of effectiveness. Prog Clin Biol Res 1990;337:239–241. 388. Quiroga J, Prieto J. Liver cytoprotection by prostaglandins. Pharmacol Ther 1993;58:67–91. 389. Sinclair S, Levy G. Eicosanoids and the liver. Ital J Gastroenterol 1990;22:205–213. 390. Sinclair SB, Greig PD, Blendis LM, et al. Biochemical and clinical response of fulminant viral hepatitis to administration of prostaglandin E. A preliminary report. J Clin Invest 1989;84:1063–1069. 391. Sterling RK, Luketic VA, Sanyal AJ, et al. Treatment of fulminant hepatic failure with intravenous prostaglandin E1. Liver Transpl Surg 1998;4:424–431. 392. Alsina AE, Hull D, Bartus SA, et al. Liver transplantation for acute fulminant hepatic failure. Conn Med 1992;56:235–239. 393. Ansaldi-Balocco N, Cimadamore N, Barbera C. [Liver transplant in children. I]. Pediatr Med Chir 1989;11:379–383. 394. Bernal W, Wendon J, Rela M, et al. Use and outcome of liver transplantation in acetaminophen-induced acute liver failure. Hepatology 1998;27:1050–1055. 395. Knaus WA, Draper EA, Wagner DP, et al. APACHE II: a severity of disease classification system. Crit Care Med 1985;13:818–829. 396. Pitre J, Soubrane O, Dousset B, et al. How valid is emergency liver transplantation for acute liver necrosis in patients with multiple-organ failure? Liver Transpl Surg 1996;2:1–7. 397. Arnold JC, Kraus T, Otto G, et al. Liver transplantation for acute or chronic liver failure by hepatitis non-A, non-B and hepatitis C viral infections: a postoperative follow-up. Zeitschr Gastroenterol 1992;30:525–528. 398. Beckurts KT, Holscher AH, Heidecke CD, et al. [The role of liver transplantation in the treatment of acute liver failure following Amanita phalloides poisoning]. Dtsch Med Wochenschr 1997;122:351–355. 399. Bellary S, Hassanein T, Van Thiel DH. Liver transplantation for Wilson’s disease. J Hepatol 1995;23:373–381. 400. Bernuau J, Samuel D, Durand F. Criteria for emergency liver transplantation in patients with acute viral hepatitis and factor V
414
401.
402. 403.
404.
405.
406.
407.
408.
409. 410.
411.
412.
413.
414.
415.
416. 417.
418.
419.
420.
421.
422.
below 50% of normal: a prospective study. Hepatology 1991;14:Abstract 6. Bjoro K, Vatn M, Schrumpf E, et al. [Liver transplantation in fulminant hepatic failure]. Tidsskr Nor Laegeforen 1990;110:3372–3375. Buckels JA. Liver transplantation in acute fulminant hepatic failure. Transplant Proc 1987;19:4365–4366. Brandsaeter B, Hockerstedt K, Friman S, et al. Fulminant hepatic failure: outcome after listing for highly urgent liver transplantation – 12 years’ experience in the nordic countries. Liver Transpl 2002;8:1055–1062. Devlin J, Wendon J, Heaton N, et al. Pretransplantation clinical status and outcome of emergency transplantation for acute liver failure. Hepatology 1995;21:1018–1024. Bismuth H, Samuel D, Gugenheim J, et al. Emergency liver transplantation for fulminant hepatitis. Ann Intern Med 1987;107:337–341. Bismuth H, Samuel D, Castaing D, et al. Liver transplantation in Europe for patients with acute liver failure. Semin Liver Dis 1996;16:415–425. Bismuth H, Samuel D, Castaing D, et al. Orthotopic liver transplantation in fulminant and subfulminant hepatitis. The Paul Brousse experience. Ann Surg 1995;222:109–119. Bismuth H, Farges O, Castaing D, et al. [Evaluation of results of liver transplantation: experience based on a series of 1052 transplantations]. Presse Med 1995;24:1106–1114. Balistreri WF. Transplantation for childhood liver disease: an overview. Liver Transpl Surg 1998;4(5 Suppl 1):S18–S23. Bonatti H, Muiesan P, Connelly S, et al. Hepatic transplantation in children under 3 months of age: a single centre’s experience. J Pediatr Surg 1997;32:486–488. Bonatti H, Muiesan P, Connolly S, et al. Liver transplantation for acute liver failure in children under 1 year of age. Transplant Proc 1997;29:434–435. Miwa S, Hashikura Y, Mita A, et al. Living-related liver transplantation for patients with fulminant and subfulminant hepatic failure. Hepatology 1999;30:1521–1526. Marcos A, Ham JM, Fisher RA, et al. Emergency adult to adult living donor liver transplantation for fulminant hepatic failure. Transplantation 2000;69:2202–2205. van Hoek B, de Boer J, Boudjema K, et al. Auxiliary versus orthotopic liver transplantation for acute liver failure. EURALT Study Group. European Auxiliary Liver Transplant Registry. J Hepatol 1999;30:699–705. van Hoek B, Ringers J, Kroes AC, et al. Temporary heterotopic auxiliary liver transplantation for fulminant hepatitis B. J Hepatol 1995;23:109–118. Erhard J, Lange R, Giebler R, et al. [Auxiliary liver transplantation in urgent and emergency indications]. Langenbecks Arch Chir Suppl Kongressbd 1996;113:422–424. Lake JR, Sussman NL. Determining prognosis in patients with fulminant hepatic failure: when you absolutely, positively have to know the answer. Hepatology 1995;21:879–882. Karvountzis GG, Redeker AG, Peters RL. Long term follow-up studies of patients surviving fulminant viral hepatitis. Gastroenterology 1974;67:870–877. Donaldson BW, Gopinath R, Wanless IR, et al. The role of transjugular liver biopsy in fulminant liver failure: relation to other prognostic indicators. Hepatology 1993;18:1370–1376. Pauwels A, Mostefa-Kara N, Florent C, et al. Emergency liver transplantation for acute liver failure. Evaluation of London and Clichy criteria. J Hepatol 1993;17:124–127. Lee WM, Galbraith RM, Watt GH, et al. Predicting survival in fulminant hepatic failure using serum Gc protein concentrations. Hepatology 1995;21:101–105. Shakil AO, Kramer D, Mazariegos GV, et al. Acute liver failure: clinical features, outcome analysis, and applicability of prognostic criteria. Liver Transpl 2000;6:163–169.
Chapter 21 ACUTE LIVER FAILURE
423. Takahashi Y, Kumada H, Shimizu M, et al. A multicenter study on the prognosis of fulminant viral hepatitis: early prediction for liver transplantation. Hepatology 1994;19:1065–1071. 424. Mitchell I, Bihari D, Chang R, et al. Earlier identification of patients at risk from acetaminophen-induced acute liver failure. Crit Care Med 1998;26:279–284. 425. Harrison PM, O’Grady JG, Keays RT, et al. Serial prothrombin time as prognostic indicator in paracetamol induced fulminant hepatic failure. Br Med J 1990;301:964–966. 426. Izumi S, Langley PG, Wendon J, et al. Coagulation factor V levels as a prognostic indicator in fulminant hepatic failure. Hepatology 1996;23:1507–1511. 427. Bernal W, Donaldson N, Wyncoll D, et al. Blood lactate as an early predictor of outcome in paracetamol-induced acute liver failure: a cohort study. Lancet 2002;359:558–563. 428. Scotto J, Opolon P, Eteve J, et al. Liver biopsy and prognosis in acute liver failure. Gut 1973;14:927–933. 429. Horney JT, Galambos JT. The liver during and after fulminant hepatitis. Gastroenterology 1977;73(4 Pt 1):639–645. 430. Hanau C, Munoz SJ, Rubin R. Histopathological heterogeneity in fulminant hepatic failure. Hepatology 1995;21:345–351.
431. Goldschmidt-Clermont PJ, Van Baelen H, Bouillon R, et al. Role of group-specific component (vitamin D binding protein) in clearance of actin from the circulation in the rabbit. J Clin Invest 1988;81:1519–1527. 432. Bernuau J, Babany G, Pauwels A, et al. Prostaglandin E1 (PGE1) has no beneficial effects in patients with either severe or fulminant hepatitis due to drugs or of undetermined etiology. Hepatology 1990;12:931. 433. O’Brien CB, Henzel BS, Naji A, et al. Prostaglandin E1 infusion is effective in the late treatment of patients with acetaminophen induced acute hepatic failure referred for liver transplantation. Gastroenterology 1992;102:A862. 434. O’Brien C, Henzel BS, Naji A, et al. Prostaglandin E1 infusion improves survival of patients with fulminant hepatic failure. Gastroenterology 1992;102:A862. 435. Sheener P, Sinclair S, Greig P, et al. A randomized controlled trial of prostaglandin E2 (PGE2) in the treatment of fulminant hepatic failure. Hepatology 1992;16:88A. 436. Cain SM. Supply dependency of oxygen uptake in ARDS: myth or reality? Am J Med Sci 1984;288:119–124.
415
Section III: Clinical Consequences of Liver Disease
22
RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY Vicente Arroyo, Pere Ginès, Mónica Guevara, and Juan Rodés Abbreviations ANP atrial natriuretic peptide ATN acute tubular necrosis ATP adenosine triphosphate AVP arginine vasopressin BUN blood urea nitrogen cAMP cyclic adenosine monophosphate CGRP calcitonin gene-related peptide EABV effective arterial blood volume
GFR HRS NE NO NOS NSAIDs PG PRA
glomerular filtration rate hepatorenal syndrome norepinephrine nitric oxide nitric oxide synthase non-steroid antiinflammatory drugs prostaglandins plasma renin activity
INTRODUCTION The presence of kidney function abnormalities in patients with liver disease has long been recognized.1 More than a century ago, Frerichs in Europe and Flint in the USA reported the association between liver disease and kidney dysfunction. These reports described the development of oliguria in patients with chronic liver disease in the setting of normal kidney histology, and proposed the first pathophysiologic interpretation of kidney dysfunction in liver disease by linking the abnormalities in kidney function to the disturbances present in the systemic circulation. Since then, the relationship between the liver and kidney function has been the object of a considerable amount of research, and substantial progress has been made in the last two decades with regard to the pathophysiology and management of kidney function abnormalities in liver disease (Table 22-1). Most abnormalities of kidney function in liver disease occur in patients with cirrhosis and are pathophysiologically related to the development of ascites and/or edema. Therefore, this chapter deals with the pathophysiology and clinical features of functional renal abnormalities in cirrhosis, i.e. alterations in renal function that occur in the absence of significant changes in kidney histology.2,3 The nonfunctional renal abnormalities (i.e. glomerulonephritis, renal tubular acidosis, drug nephrotoxicity) that may occur in cirrhosis, as well as the abnormalities in kidney function that sometimes occur in other types of liver disease (i.e. acute liver failure) are not discussed here and may be found in several recent reviews.4,5 The treatment of ascites and edema in patients with cirrhosis is covered by another chapter in this book. In this chapter, however, the medical treatment of the renal function abnormalities is discussed.
RAAS RBF SBP SNS TIPS TNF
renin-algiotensin-algosterone system renal blood flow spontaneous bacterial peritonitis sympathetic nervous system transjugular intrahepatic portosystemic shunts tumor necrosis factor
FUNCTIONAL RENAL ABNORMALITIES The most common functional renal abnormalities in patients with cirrhosis are an impaired ability to excrete sodium and water and a reduction of renal blood flow (RBF) and glomerular filtration rate (GFR), the latter two being secondary to vasoconstriction of the renal circulation. Sodium retention is a key factor in the formation of ascites and edema, whereas water retention is responsible for the development of dilutional hyponatremia. Renal vasoconstriction, when severe, leads to hepatorenal syndrome (HRS). Chronologically, sodium retention is the earliest alteration of kidney function observed in patients with cirrhosis, whereas water retention leading to dilutional hyponatremia and the development of HRS are late findings, particularly the latter. In most patients, abnormalities of kidney function usually worsen with time as the liver disease progresses. However, in some patients a spontaneous improvement or even normalization of sodium and, less frequently, water excretion may occur during the course of their disease.6,7 The improvement of sodium excretion is associated with the disappearance of ascites and edema, which allows for the discontinuation of diuretic therapy and the reintroduction of regular dietary salt intake. The improvement of water excretion is associated with an increase or normalization of serum sodium concentration. This improvement in renal function abnormalities is seen after alcohol abstinence in some patients with alcoholic cirrhosis or alcoholic hepatitis. The frequency and mechanism(s) of this improved renal function are not known. Spontaneous improvement of renal function after the development of HRS is extremely unusual.
417
Section III. Clinical Consequences of Liver Disease
SODIUM RETENTION AND ASCITES Sodium retention is the most frequent abnormality of kidney function in patients with cirrhosis, the existence of which was first documented more than 40 years ago when methods to measure electrolyte concentration in organic fluids became available.8 Since
Table 22-1. Landmarks in Ascites and Renal Dysfunction in Liver Disease 1800s 1940s 1950s
1960s–1970s
1980s
1990s
First description of renal dysfunction in liver diseases Demonstration of water retention in cirrhosis with ascites Demonstration of sodium retention in cirrhosis with ascites and its relationship with increased mineralocorticoid activity Description of the hyperkinetic circulatory syndrome Clinical description of hepatorenal syndrome Demonstration of renal vasoconstriction as the cause of hepatorenal syndrome Introduction of aldosterone antagonists in clinical practice Demonstration of increased activity of vasoconstrictor systems Demonstration of the role of prostaglandins in the maintenance of renal function Proposal of the overflow theory of ascites formation Reintroduction of therapeutic paracentesis in clinical practice Proposal of the arterial vasodilatation theory of ascites formation New definition and diagnostic criteria of hepatorenal syndrome Description of functional renal failure after spontaneous bacterial peritonitis and its prevention by intravenous albumin Introduction of transjugular intrahepatic portosystemic shunt in the management of refractory ascites and hepatorenal syndrome Use of splanchnic vasoconstrictors in the management of hepatorenal syndrome
75 65
18
180
16
160
55 14
418
25 15 5 4 3
140
12 120 10 8 6 4
GFR (mL/min)
35
Free water clearance (mL/min)
Sodium excretion (mEq/day)
45
then, it has been well established that sodium retention plays a key role in the pathophysiology of ascites and edema formation in cirrhosis. The amount of sodium retained within the body is dependent on the balance between the sodium ingested in the diet and that excreted in the urine. As long as the amount of sodium excreted is lower than that ingested, patients accumulate extracellular fluid as ascites and/or edema. The important role of sodium retention in the pathogenesis of ascites formation is supported by the fact that ascites can disappear just by reducing the dietary sodium content in some patients, or by increasing the urinary sodium excretion with the administration of diuretics in others.8,9 Conversely, a highsodium diet or diuretic withdrawal leads to the reaccumulation of ascites. The achievement of a negative sodium balance (i.e. excretion higher than intake) is the essence of pharmacological therapy of ascites. Finally, studies in experimental animals have constantly shown that sodium retention precedes ascites formation, further emphasizing the important role of this abnormality of renal function in the pathogenesis of ascites in cirrhosis.10,11 The severity of sodium retention in cirrhosis with ascites varies considerably from patient to patient. Some have relatively high urinary sodium excretion, whereas in others urinary sodium concentrations are very low or even undetectable (Figure 22-1). The proportion of patients with ascites who have marked sodium retention depends on the population of cirrhotic patients considered. Most patients who require hospitalization because of large ascites have marked sodium retention, as they excrete less than 10 mEq/ day while on a low-sodium diet and without diuretic therapy.12 Among patients who require hospitalization, sodium retention is particularly intense in patients with ascites refractory to diuretic treatment. By contrast, in a population of cirrhotic patients with mild or moderate ascites, the proportion with marked sodium retention is low and most excrete more than 10 mEq/day of sodium spontaneously (without diuretic therapy). The response to diuretics is usually better in patients with moderate sodium retention than in those with marked retention.12
100 80 60 40
2
2
1
0
20
0
-2
0
Figure 22-1. Individual values of urinary sodium excretion (in conditions of a 50-mEq sodium diet), free water clearance after an intravenous water load (20 ml/kg body weight of 5% dextrose), and glomerular filtration rate measured by inulin clearance in a series of 179 cirrhotic patients admitted to hospital for treatment of ascites. Shadowed areas represent the normal range.
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
NEPHRON SITES OF SODIUM RETENTION In most instances, sodium retention in cirrhosis is due to increased tubular reabsorption of sodium because it occurs in the presence of normal or only moderately reduced GFR.9,10 The exact contribution of the different segments of the nephron to this increased sodium reabsorption is not completely known. Micropuncture studies in rats with cirrhosis and ascites have demonstrated an enhanced reabsorption of sodium in the proximal tubule.10–12 On the other hand, it has been shown that the development of a positive sodium balance and the formation of ascites in cirrhotic rats can be prevented by aldosterone antagonists, which suggests that the collecting ducts are important sites of the increased sodium reabsorption in experimental cirrhosis.11–13 Investigations in patients with cirrhosis have also provided discrepant findings. Results from earlier studies using sodium, water or phosphate clearances to estimate the tubular handling of sodium suggested that the distal nephron is the main site of sodium retention.14 Studies using lithium clearance, which estimates sodium reabsorption in the proximal tubule, suggest that cirrhotic patients with ascites show a marked increase in proximal sodium reabsorption.15 Nevertheless, distal sodium reabsorption is also increased, especially in patients with more avid sodium retention. Clinical studies using spironolactone to antagonize the mineralocorticoid receptor indicate that this agent induces natriuresis in a large proportion of cirrhotic patients with ascites without renal failure, which supports a major role for increased sodium reabsorption in distal sites of the nephron in these patients.16–18 Taken together, these results suggest that in patients with cirrhosis without renal failure, an enhanced reabsorption of sodium in both proximal and distal tubules contributes to sodium retention. In fact, recent studies have shown that membrane pumps and proteins involved in sodium and chloride reabsorption in the proximal, distal and collecting tubules are overexpressed in rats with experimental cirrhosis, sodium retention, and ascites.19–21 Potential mediators of this increased sodium reabsorption include changes in the hydrostatic and colloid osmotic pressures in the peritubular capillaries, and increased activity of the sympathetic nervous system and the renin–angiotensin–aldosterone system (see later). Sodium retention is usually more marked in patients with renal failure than in those without, owing to both a reduction in filtered sodium load and a more marked activation of sodium-retaining mechanisms.
CLINICAL CONSEQUENCES Because sodium is retained together with water iso-osmotically in the kidney, sodium retention is associated with fluid retention, leading to expansion of extracellular fluid volume and increased amounts of fluid in the interstitial tissue. In some patients with cirrhosis, the total extracellular fluid volume may increase up to 40 l or even more (compared to the average 14 l in a 70 kg healthy adult), which represents an approximate cumulative gain of 3400 mEq of sodium (26 l of excess extracellular fluid volume times 130 mEq/l). In most patients with advanced cirrhosis, sodium retention is manifested by the development of ascites. The most common clinical symptom of ascites is discomfort due to abdominal swelling. In cases with marked accumulation of fluid, physical activity and respiratory function may be impaired. Other clinical consequences related to the presence of intra-abdominal fluid are the appearance of abdom-
inal wall hernias and the spontaneous infection of ascitic fluid (also known as spontaneous bacterial peritonitis).22,23 Both complications, especially the infection, contribute markedly to the high morbidity and mortality associated with the presence of ascites. Accumulation of fluid in the subcutaneous tissue as edema is also common in patients with cirrhosis and sodium retention, and in most cases occurs concomitantly with the presence of ascites. Edema is most commonly observed in the lower extremities, but generalized edema may occur as well. In cases with generalized edema the existence of proteinuria due to associated nephropathy or myocardiopathy should be ruled out. Mild or moderate edema may decrease or even disappear during bed rest and reappear during the daytime, reflecting an increased natriuresis in the supine position compared with the upright position.22,24 Both hypoalbuminemia and increased venous pressure in the inferior vena cava, due either to constriction of the vena cava within the liver or to increased intraabdominal pressure caused by ascites, may contribute to edema in cirrhotic patients with ascites. Leg edema is common in patients with cirrhosis treated with either surgical portocaval shunts or transjugular intrahepatic portosystemic shunts (TIPS), presumably because of the increased pressure in the inferior vena cava secondary to these procedures. Other clinical manifestations of sodium retention in cirrhosis include pleural and/or pericardial effusions. Clinically significant pleural effusions occur in up to 10% of patients with cirrhosis.25,26 In most cases the effusion is mild or moderate, more frequent on the right side, and associated with the presence of ascites. Left-sided effusions are uncommon and usually occur in patients who have right-sided effusions as well. Occasionally, large right pleural effusions may exist in the absence of ascites and constitute the main manifestation of the disease.25 These latter cases usually recur after therapy and are due to the existence of anatomical defects in the diaphragm that cause a communication between the peritoneal and pleural cavities. The gradient between the positive intra-abdominal pressure and the negative intrathoracic pressure explains the passage of all fluid formed in the peritoneal cavity to the pleural cavity. Although less commonly than ascitic fluid, pleural fluid may also become infected spontaneously, a condition known as spontaneous bacterial empyema.26 Finally, between one- and two-thirds of cirrhotic patients with ascites have also mild or moderate pericardial effusions, as demonstrated by echocardiography, that are not associated with clinical symptoms and which disappear after the elimination of ascites.
ASSESSMENT OF RENAL SODIUM EXCRETION IN CLINICAL PRACTICE The assessment of urinary excretion of sodium is very useful in the clinical management of patients with cirrhosis and ascites because it allows the precise quantification of sodium retention. Urine must be collected under conditions of fixed and controlled sodium intake (usually a low-sodium diet of 40–80 mEq/day during the previous 5 days), as sodium intake may influence sodium excretion. The amount of sodium ingested does not affect the excretion of sodium in patients with marked sodium retention, who have very low urinary sodium regardless of the amount ingested, but may affect sodium excretion in patients with mild or moderate sodium retention. Diuretics should not be given during the period prior to urine
419
Section III. Clinical Consequences of Liver Disease
Table 22-2. Renal and Circulatory Abnormalities in Patients with Preascitic Cirrhosis
Urine sodium 1.0
Increased cardiac output and blood volume Portal hypertension and splanchnic arterial vasodilatation Inability to excrete an acute or chronic sodium load Lack of escape from the sodium-retaining effect of mineralocorticoids Reduced sodium excretion in upright posture and increased sodium excretion in recumbency Increased atrial natriuretic peptide levels Development of sodium retention and ascites and edema after treatment with vasodilators or non-steroidal anti-inflammatory drugs
Survival
0.8
0.6
0.4
>=10 mEq/day
0.2
<10 mEq/day
tory ascites (see below). Patients with a urinary sodium excretion lower than 50 mEq/8 hours usually respond either poorly or not at all to diuretics.29
0.0 0
1
2
3
4
5
6
7
8
9
10
Years Figure 22-2. Long-term survival of patients with cirrhosis and ascites according to baseline sodium excretion in conditions of low-sodium diet and without diuretics.
collection to avoid a pharmacological increase in sodium excretion. This 5-day period is particularly important in patients receiving spironolactone or other aldosterone antagonists that have a very prolonged half-life. Finally, although the measurement of sodium concentration in a spot of urine may provide a rough estimate of sodium excretion, the assessment of sodium excretion in a 24-hour period is preferable, because it is more representative of sodium excretion throughout the day and takes into account the urine output. In clinical practice, sodium excretion should ideally be measured under the conditions stated above when patients with ascites are first seen or when there are signs suggestive of disease progression (i.e. a marked increase in ascites or edema despite compliance with the sodium-restricted diet and diuretic therapy). On the other hand, the measurement of sodium excretion in patients on diuretic therapy is very useful to monitor the response to treatment. Baseline sodium excretion is one of the best predictors of the response to diuretic treatment.9 Therefore, the measurement of urine sodium is helpful in establishing the therapeutic schedule in cirrhotic patients with ascites. Patients with marked sodium retention in whom a positive sodium balance is anticipated despite a reduction in sodium intake should be started on moderately high doses of aldosterone antagonists (i.e. spironolactone 200 mg/day), either alone or in association with loop diuretics (i.e. furosemide 40 mg/day). Conversely, patients with moderate sodium retention would probably respond to low doses of aldosterone antagonists (i.e. spironolactone 50–100 mg/day). Finally, the intensity of sodium retention also provides prognostic information in patients with ascites. Patients with baseline urinary sodium lower than 10 mEq/day have a median survival time of less than 2 years, compared to 4–5 years in those with urinary sodium higher than 10 mEq/day (Figure 22-2).27,28 Also, sodium excretion after a single oral dose of furosemide (80 mg in a single dose in patients who have not received diuretics for 3 days) is a useful test to predict refrac-
420
ABNORMALITIES OF RENAL SODIUM HANDLING IN PREASCITIC CIRRHOSIS Patients with cirrhosis in the preascitic stage (without a past history of ascites or edema) do not exhibit overt sodium retention but may have subtle abnormalities in renal sodium handling (Table 22-2).30–33 The finding of increased blood volume in these patients strongly supports the existence of sodium retention sufficient to expand the intravascular volume but without causing ascites or edema. Patients with preascitic cirrhosis come into sodium balance as long as their sodium intake is maintained within normal limits. However, some may be unable to handle a sodium load and develop ascites and/or edema in conditions of high sodium intake (especially when they receive intravenous saline solutions) or when treated with nonsteroidal anti-inflammatory drugs or vasodilators.34–36 This abnormal renal sodium handling is also evidenced by the lack of escape from the sodium-retaining effect of mineralocorticoids.37,38 Finally, it has been shown that patients in the preascitic stage on a normal sodium diet retain sodium while they are in an upright posture, whereas they show an exaggerated natriuresis during bed rest compared to healthy subjects.30 This increased natriuresis during recumbency may be responsible for the maintenance of sodium balance and help prevent the formation of ascites or edema that would occur as a consequence of sodium retention that takes place during standing. These subtle abnormalities in renal sodium handling present in compensated preascitic cirrhotic patients that are responsible for the increased blood volume are probably a homeostatic mechanism to compensate for the increased vascular capacitance of the splanchnic vascular bed that develops in cirrhosis due to arterial vasodilatation. This interpretation is supported by the observation that a pharmacologically induced vasodilatation (with the administration of the a-adrenergic blocker prazosin or nitrates) is followed by sodium retention and increased plasma volume and ascites and edema formation in most preascitic cirrhotic patients,34,35 and also by the findings that patients who develop ascites or edema while on a highsodium diet, or who fail to escape from the sodium-retaining effect of mineralocorticoids, are those with more marked abnormalities in systemic hemodynamics, as indicated by higher cardiac output and lower total systemic vascular resistance.38 An alternative interpretation suggests that abnormal sodium handling in preascitic cirrhotic patients is pathogenetically related to the degree of liver dysfunc-
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
tion, as it has been shown that the impaired renal sodium excretion after a sodium load correlates directly with the impairment in liver function tests.31,39
Diuresis after water load 1.0
0.8
Since the pioneer studies by Papper et al. and Shear et al.,40,41 it is well known that a derangement in the renal capacity to regulate water balance occurs commonly in advanced cirrhosis. Cirrhotic patients without ascites, or those with ascites and mild sodium retention, usually have normal or only slightly impaired renal water handling compared to healthy subjects. Therefore, in these patients total body water, plasma osmolality, and serum sodium concentration are normal and hyponatremia does not develop, even in conditions of excessive water intake. By contrast, an impairment in the renal capacity to excrete solute-free water is common in patients with ascites and severe sodium retention.40,41 As with sodium retention, the impairment of water excretion is not uniform in all patients with ascites; rather, it varies markedly from patient to patient (see Figure 22-1).12 In some patients, water retention is moderate and can only be detected by measuring solutefree water excretion after a water load. These patients are able to eliminate water normally and maintain a normal serum sodium concentration as long as their water intake is kept within normal limits, but they may develop hyponatremia when water intake is increased. In other patients the severity of the disorder is such that they retain most of the water taken in the diet, leading to hyponatremia and hypoosmolality. Therefore, hyponatremia in cirrhosis with ascites is almost always dilutional in origin, as it occurs in the setting of an increased total body water. Hyponatremia is also paradoxical, in that it is associated with sodium retention and a marked increase in total body exchangeable sodium. The development of spontaneous dilutional hyponatremia requires a profound impairment in solute-free water excretion, as it usually occurs with a solute-free water clearance after a water load below 1 ml/min (normal: 6–12 ml/min). One-third of hospitalized cirrhotic patients with ascites have spontaneous dilutional hyponatremia.12 The existence of an impaired capacity to excrete solute-free water and/or dilutional hyponatremia in patients with cirrhosis and ascites is indicative of a poor prognosis (Figure 22-3).28,42 Several factors may aggravate the impairment of water excretion in cirrhotic patients with ascites and/or precipitate the appearance of hyponatremia, including the administration of hypotonic fluids in excess of the capacity to excrete solute-free water, treatment with diuretics or non-steroidal antiinflammatory drugs (NSAIDs), or large-volume paracentesis without plasma volume expansion.43,44
MECHANISMS OF IMPAIRED RENAL WATER HANDLING The pathogenesis of water retention in cirrhosis is complex and probably involves several factors, including a reduced delivery of filtrate to the ascending limb of the loop of Henle, reduced renal synthesis of prostaglandins, and non-osmotic hypersecretion of arginine vasopressin (AVP).28,43,44 Definitive data about the relative impor-
Survival
WATER RETENTION AND DILUTIONAL HYPONATREMIA
0.6
0.4 >=3 ml/min 0.2 <3 ml/min 0.0 0
1
2
3
4
5
6
7
8
9
10
Years Figure 22-3. Long-term survival of patients with cirrhosis and ascites according to diuresis after water load (20 ml/kg body weight of 5% dextrose) in conditions of low-sodium diet and without diuretics.
tance of these factors in the pathogenesis of water retention in patients with cirrhosis is lacking. However, from the results obtained in studies using specific antagonists of the tubular effect of AVP (V2 antagonists), it appears likely that AVP hypersecretion plays a major role in water retention in patients without renal failure (see later).45–53 In patients with renal failure it is likely that besides AVP, a reduced distal delivery of filtrate due to decreased filtered load and increased sodium and water reabsorption in the proximal tubule also play an important role in water retention.
CLINICAL CONSEQUENCES The clinical consequence of an impairment in solute-free water excretion is the development of dilutional hyponatremia. Dilutional hyponatremia is associated with sodium retention and increased total body sodium, and should be distinguished from hyponatremia, which, although less common, may develop in cirrhotic patients who are maintained on high doses of diuretics and sodium restriction after the complete disappearance of ascites and edema. In some patients dilutional hyponatremia is asymptomatic, but in others it is probably associated with clinical symptoms similar to those found in dilutional hyponatremia of other etiologies, including anorexia, headache, difficulty in mental concentration, sluggish thinking, lethargy, nausea, vomiting and, occasionally, seizures. However, in many instances it may be difficult or even impossible to establish whether these symptoms are due to hyponatremia itself, to the underlying liver disease, and/or to associated conditions. Cerebral edema plays a major role in the pathogenesis of cerebral dysfunction in fulminant hepatic failure. There is evidence that in cirrhosis there is also increased water content in the brain.54,55 The mechanism of cerebral edema in liver disease is thought to be related to hyperammonemia. The metabolization of ammonia (a freely diffusible substance) to glutamine by the astrocytes increases intracellular osmolality, induces water shift from the extracellular to the
421
Section III. Clinical Consequences of Liver Disease
intracellular space, and produces brain edema. In conditions of severe hyponatremia the reduction in extracellular osmolality also increases water shift from the extracellular to the intracelular space and may promote brain edema. In both conditions (hyperammonemia and hyponatremia) there are compensatory mechanisms, including the elimination of intracellular potassium and other osmolytes such as myoinositol.56 If intracellular osmolytes are reduced due to the simultaneous existence of hyperammonemia and hyponatremia, the buffer mechanisms can be considerably impaired. These patients are more susceptible to precipitating events of cerebral dysfunction and encephalopathy, such as intensive diuretic therapy, or bacterial infections.
ASSESSMENT OF RENAL WATER EXCRETION IN CLINICAL PRACTICE As stated previously, the assessment of the kidney’s capacity to excrete solute-free water in cirrhotic patients with ascites should ideally include not only the determination of serum sodium concentration but also the evaluation of the renal response to the administration of water. The capacity to excrete solute-free water is a major prognostic factor of cirrhosis with ascites,42 and its assessment may be of value when a thorough evaluation of the disease is required, especially in patients who are being considered for liver transplantation. Moreover, the assessment of solute-free water excretion may be of importance in the clinical management of patients, as it allows for the adjustment of fluid intake to the renal capacity to eliminate water, thus preventing the development of hyponatremia due to inappropriately high fluid intake. However, there are patients with significant hyponatremia and normal or only moderately reduced renal ability to excrete free water. The mechanism of hyponatremia in these latter patients is unknown, but clearly they do not need treatment to improve free water excretion. Without doubt, the recent introduction of the new oral specific antagonists of V2 vasopressin receptors into the diuretic armamentarium will lead to the determination of free water clearance after water load in the assessment of cirrhotic patients with ascites and dilutional hyponatremia.
RENAL VASOCONSTRICTION AND HEPATORENAL SYNDROME Investigations performed during the late 1960s and early 1970s provided conclusive evidence indicating that the renal failure of functional origin occurring in patients with cirrhosis – the so-called hepatorenal syndrome (HRS) – is due to a marked vasoconstriction of the renal circulation.57–59 Later studies showed that, besides the striking renal vasoconstriction present in patients with HRS, patients with cirrhosis and ascites and normal or slightly increased serum creatinines frequently have mild or moderate degrees of vasoconstriction in the renal circulation.12,60,61 When renal perfusion is estimated by sensitive clearance techniques, such as para-aminohippurate or inulin clearances, in a population of hospitalized patients with ascites, normal values are found in only one-fifth of cases. In another 15–20% renal hypoperfusion is very intense and meets the criteria for HRS.62 In the remaining patients, mild or moderate reductions in renal perfusion exist (Figure 22-1).12 These latter
422
patients show slightly increased serum creatinine and/or blood urea nitrogen (BUN) levels in baseline conditions (in the absence of diuretic therapy). This moderate renal vasoconstriction is clinically relevant for several reasons: first, it is frequently associated with marked sodium and water retention and the presence of refractory ascites;62 second, it predisposes to the development of HRS;63,64 and third, it is associated with a reduced survival.28,42,64
DEFINITION OF HEPATORENAL SYNDROME The definition of HRS proposed by the International Ascites Club, which is the most widely accepted, is as follows: ‘Hepatorenal syndrome is a clinical condition that occurs in patients with advanced chronic liver disease, liver failure, and portal hypertension characterized by impaired renal function and marked abnormalities in the arterial circulation and activity of the endogenous vasoactive systems. In the kidney there is marked renal vasoconstriction that results in low GFR, whereas in the extrarenal circulation there is predominance of arterial vasodilatation, which results in reduction of total systemic vascular resistance and arterial hypotension.’62 Although HRS occurs predominantly in advanced cirrhosis, it may also develop in other chronic liver diseases associated with severe liver failure and portal hypertension, such as alcoholic hepatitis, or in acute liver failure.
PATHOGENIC MECHANISMS The pathophysiologic hallmark of HRS is a vasoconstriction of the renal circulation.59,62,65,66 Studies of renal perfusion with renal arteriography, 133Xe washout technique, para-aminohippuric acid excretion or, more recently, duplex Doppler ultrasonography, have demonstrated the existence of marked vasoconstriction in the kidneys of patients with HRS, with a characteristic reduction in renal cortical perfusion.59,64,67–69 The functional nature of HRS has been conclusively demonstrated by the lack of significant morphological abnormalities in kidney histology2,3 and normalization of renal function after liver transplantation.70–72 The mechanism of this vasoconstriction is incompletely understood and possibly multifactorial, involving changes in systemic hemodynamics, increased pressure in the portal venous system, activation of vasoconstrictor factors, and suppression of vasodilator factors acting on the renal circulation (see later). Recent studies have shown that, in contrast to the previous belief, other vascular beds besides the renal circulation are also vasoconstricted in patients with HRS, including the brachial, femoral, cerebral and intrahepatic circulations.69,73,74 This indicates the existence of a generalized arterial vasoconstriction in the non-splanchnic vascular beds of patients with HRS, and suggests that the main vascular bed responsible for arterial vasodilatation and reduced total systemic vascular resistance in cirrhosis with HRS is the gastrointestinal and splenic circulation.
CLINICAL AND LABORATORY FINDINGS Hepatorenal syndrome is a common complication of patients with cirrhosis. In patients with ascites, the probability of developing HRS during the course of the disease is 18% at 1 year and increases up to 40% after 5 years of follow-up.63 The clinical manifestations include a combination of signs and symptoms related to renal, circulatory, and liver failure.
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
Renal failure may have been rapid or insidious in onset and is usually associated with marked sodium and water retention, which results in ascites and edema and dilutional hyponatremia, respectively.62,65,66 HRS may occur in two different clinical patterns, according to the intensity and form of onset of renal failure (Table 22-3).60 Type 1 HRS is the classic type and represents the end of the spectrum of changes in renal function in cirrhosis. The dominant clinical features of type 1 HRS are those of severe and rapidly progressive renal failure with oliguria or anuria and increased serum levels of urea and creatinine. Despite an important reduction of GFR in these patients, serum creatinine levels are commonly lower than values observed in non-cirrhotic patients with acute renal failure of similar intensity with respect to the reduction in GFR.65 This is probably due to the lower endogenous production of creatinine secondary to reduced muscle mass in patients with cirrhosis compared to patients without liver disease. This type of HRS is frequently seen in patients with alcoholic cirrhosis, especially when associated with alcoholic hepatitis, but it occurs in non-alcoholic cirrhosis as well. Type 1 HRS is associated with a very low survival expectancy, the median survival time of patients in old series being only 2 weeks (Figure 22-4).63 Type 2 HRS is characterized by a less severe and stable reduction in GFR that does not meet the criteria proposed for type 1. Patients are usually in a better clinical condition than those with type 1 HRS and their survival expectancy is longer. The dominant clinical feature of these patients is refractory
1.0
0.8
0.6
ascites due to the combination of intense sodium retention, reduced GFR, and marked stimulation of antinatriuretic systems.62 Severe spontaneous hyperkalemia is an uncommon feature of HRS. However, marked hyperkalemia may occur if patients are treated with aldosterone antagonists, especially those with type 1 HRS. Severe metabolic acidosis and pulmonary edema, which are frequent complications of acute renal failure in patients without liver disease, are uncommon findings in patients with HRS. Because HRS is a form of functional renal failure, the findings on analysis of the urine are those of prerenal azotemia with oliguria, i.e. low urine sodium concentration, and increased urine osmolality and urine-to-plasma osmolality ratio.2,62 Nevertheless, there are non-oliguric forms of the syndrome and in some cases urine sodium concentration is not extremely reduced.2 As shown in Table 22-4, urinary indices are not currently considered essential for the diagnosis of HRS according to the criteria established by the International Ascites Club. Circulatory failure in patients with HRS is characterized by severe arterial hypotension (most patients have a mean arterial pressure in the range of 60–80 mmHg) and low total systemic vascular resistance, despite the existence of severe vasoconstriction in several vascular beds, as discussed above. Because cardiac output is increased over normal values in most patients with HRS, circulatory failure associated with HRS has been traditionally considered to be due to the progression of the splanchnic arterial vasodilatation already present in non-azotemic patients with portal hypertension and ascites.75 However, recent studies have shown that type 1 HRS may develop in the setting of a significant reduction in cardiac output, and that cardiac output does not increase in patients with type 2 HRS despite the accentuation of the splanchnic arterial vasodilatation. These observations suggest that an impairment in cardiac function may also contribute to the development of the HRS (see below for further discussion). Finally, the third type of clinical manifestation of HRS is related to the existence of liver failure. Most patients show findings of advanced liver insufficiency, particularly jaundice, coagulopathy, poor nutri-
0.4 Table 22-4. Diagnostic Criteria for Hepatorenal Syndrome* 0.2
0.0 0
4
8
12
Weeks Figure 22-4. Probability of survival after development of type 1 hepatorenal syndrome.
Table 22-3. Clinical Types of Hepatorenal Syndrome Type 1 Rapid and progressive impairment of renal function as defined by a doubling of the initial serum creatinine to a level higher than 2.5 mg/dl or a 50% reduction of the initial 24-hour creatinine clearance to a level lower than 20 ml/min in less than 2 weeks Type 2 Impairment in renal function (serum creatinine > 1.5 mg/dl) that does not meet the criteria for type 1
Major criteria Low glomerular filtration rate, as indicated by serum creatinine > 1.5 mg/dl or 24-h creatinine clearance < 40 ml/min Absence of shock, ongoing bacterial infection, fluid losses and current treatment with nephrotoxic drugs No sustained improvement in renal function (decrease in serum creatinine to 1.5 mg/dl or less or increase in creatinine clearance to 40 ml/min or more) following diuretic withdrawal and expansion of plasma volume with 1.5 l of a plasma expander Proteinuria < 500 mg/day and no ultrasonographic evidence of obstructive uropathy or parenchymal renal disease Additional criteria Urine volume < 500 ml/day Urine sodium < 10 mEq/l Urine osmolality greater than plasma osmolality Urine red blood cells less than 50 per high-power field Serum sodium concentration < 130 mEq/l *All major criteria must be present for the diagnosis of hepatorenal syndrome. Additional criteria are not necessary for the diagnosis, but provide supportive evidence.
423
Section III. Clinical Consequences of Liver Disease
tional status and encephalopathy, although some patients with HRS may show only moderate liver failure. In general, patients with type 1 HRS have more severe liver failure than those with type 2 HRS.
In some patients HRS develops without any identifiable precipitating factor, whereas in others it occurs in close chronologic relationship with bacterial infections, particularly spontaneous bacterial peritonitis.12,62,76,77 Approximately one-third of patients with spontaneous bacterial peritonitis develop an impairment of renal function during treatment with non-nephrotoxic antibiotics and in the absence of shock.76 This impairment in renal function is of functional origin and occurs in the setting of a further decrease in effective arterial blood volume of patients with ascites, as indicated by a marked activation of vasoconstrictor systems, and increased serum and ascitic fluid levels of cytokines.76,77 In approximately one-third of patients developing renal failure after spontaneous bacterial peritonitis, the impairment in renal function is reversible after resolution of infection. However, in the remaining patients it is not, and meets the criteria for HRS (type 1 in most cases). Patients who develop type 1 HRS after spontaneous bacterial peritonitis have a dismal outcome, with an in-hospital mortality close to 100%.76,77 The administration of albumin at the time of diagnosis of the infection and 2 days later (1.5 g/kg and 1 g/kg body weight, respectively) prevents the development of HRS and improves survival in these patients.78 Spontaneous bacteremia also is associated with an increase in the prevalence of HRS comparable to that of spontaneous bacterial peritonitis. The prevalence of HRS is lower during other infections, such as pneumonia, urinary tract infections, and lymphangitis. Although uncommon, HRS has been reported after therapeutic paracentesis without plasma expansion.79 This is one of the reasons that the administration of IV albumin is recommended when largevolume paracentesis is performed.80 Gastrointestinal bleeding has been classically considered a precipitating factor of HRS.2 However, the development of renal failure after this complication is not very common in patients with cirrhosis (approximately 10%) and occurs mainly after hypovolemic shock, in most cases associated with ischemic hepatitis, which suggests that renal failure in this setting is frequently related to the development of acute tubular necrosis (ATN) and is not functional in origin.81 Diuretic treatment has also been classically described as a precipitating factor of HRS, but there is no clear evidence to support such a pathogenic relationship.
DIAGNOSIS The diagnosis of HRS is currently based on several diagnostic criteria (Table 22-4).62 The minimum level of serum creatinine required for the diagnosis is 1.5 mg/dl, because most patients with cirrhosis with a serum creatinine above 1.5 mg/dl have a GFR below 30 ml/min (Figure 22-5).65 In patients receiving diuretics, serum creatinine measurement should be repeated after diuretic withdrawal because in some patients serum creatinine may increase slightly during diuretic therapy due to volume depletion. Because no specific laboratory tests are available for the diagnosis of HRS and patients with advanced cirrhosis may develop renal failure of other etiologies (prerenal failure due to volume depletion, acute tubular necrosis (ATN), drug-induced nephrotoxicity, and glomerulonephritis), the most important step in the diagnosis of
424
Glomerular filtration rate (ml/min)
PRECIPITATING FACTORS
200
150
100
50
0 0
2
4
6
8
Serum creatinine (mg/dl) Figure 22-5. Correlation between glomerular filtration rate measured by inulin clearance and serum creatinine concentration in patients with cirrhosis and ascites.
HRS is to rule out renal failure secondary to volume depletion or intrinsic renal disease.62 Gastrointestinal fluid losses, due to vomiting or diarrhea, or renal fluid losses, due to excessive diuresis, should be sought in all patients with cirrhosis presenting with renal failure. If renal failure is secondary to volume depletion, renal function improves rapidly after volume repletion and treatment of the precipitating factor. Shock is another common condition in patients with cirrhosis that may lead to renal failure due to ATN. Whereas hypovolemic shock due to gastrointestinal bleeding is easily recognized, the presence of septic shock may be more difficult to diagnose because of the paucity of symptoms of bacterial infection in some patients with cirrhosis. Moreover, arterial hypotension due to the infection may be erroneously attributed to the underlying liver disease. In some patients with septic shock oliguria is the first sign of infection. These patients may be misdiagnosed as having HRS if signs of infection (cell blood count, examination of ascitic fluid) are not sought. On the other hand, as discussed earlier, patients with cirrhosis and spontaneous bacterial peritonitis may develop renal failure during the course of the infection, in the absence of septic shock.76–78 Renal failure in these patients may either improve with the antibiotic therapy or evolve into a true HRS, even after the infection has been resolved. The administration of NSAIDs is another common cause of acute renal failure in patients with cirrhosis and ascites, which is clinically indistinguishable from a true HRS.81–83 Therefore, use of these drugs should always be ruled out before the diagnosis of HRS is made. Likewise, patients with cirrhosis are at high risk of developing renal failure due to ATN when treated with aminoglycosides.84,85 Because of this high risk of nephrotoxicity and the existence of other effective antibiotics (i.e. third-generation cephalosporins) treatment with aminoglycosides should be avoided in patients with chronic liver disease. Finally, patients with cirrhosis may also develop renal failure due to glomerulonephritis.5 In these cases, proteinuria and/or hematuria are almost constant and provide a clue for the diagnosis, which may be confirmed by renal biopsy in selected cases.
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
FACTORS INVOLVED IN FUNCTIONAL RENAL ABNORMALITIES IN CIRRHOSIS CIRCULATORY ABNORMALITIES
in arterial blood flow. This phenomenon is thought to be a homeostatic mechanism to protect the intestine against edema formation. This protective mechanism is not operative in chronic portal hypertension, and arteriolar resistance is reduced and not increased. The resultant increases in capillary pressure and filtration may be important factors in the formation of ascites in cirrhosis. The mechanism(s) by which portal hypertension induces splanchnic arteriolar vasodilatation is not completely understood, although a number of vasoactive mediators have been proposed (see below).
HEPATIC AND SPLANCHNIC CIRCULATION The existence of cirrhosis causes marked structural abnormalities in the liver that result in severe disturbance of intrahepatic circulation, causing increased resistance to portal flow and subsequent hypertension in the portal venous system. Progressive collagen deposition and the formation of nodules alter the normal vascular architecture of the liver. Moreover, selective deposition of collagen in the space of Dissé may constrict the sinusoids, resulting in further mechanical obstruction to flow. In addition to this passive resistance to portal flow there is an active component of intrahepatic resistance, which is due to the contraction of hepatic stellate cells (myofibroblast-like cells) present in sinusoids and terminal hepatic venules.5,86,87 The contraction of these cells is affected by endogenous vasoconstrictors and can be modulated by vasodilators and drugs that antagonize the vasoconstrictor factors.5,88–90 Moreover, there is a strong body of evidence indicating that despite the overproduction of the vasodilator nitric oxide (NO) in the systemic circulation in cirrhosis, the production of NO in the intrahepatic circulation is reduced and contributes further to the increased intrahepatic resistance characteristic of cirrhotic livers. Interestingly, in vivo gene transfer of neuronal nitric oxide synthase (NOS) isoform to cirrhotic rat livers increases NO production from endothelial cells and hepatic stellate cells, and significantly decreases portal pressure. Portal hypertension induces profound changes in the splanchnic circulation. Classically, portal hypertension was believed to cause changes only in the venous side of the splanchnic circulation. However, studies in experimental animals indicate that portal hypertension also causes marked changes in the arterial side of the splanchnic vascular bed. In the venous side, the main changes consist of increased pressure and the formation/opening of a portocollateral circulation, which causes the appearance of gastroesophageal varices and shunting of blood from the portal venous system to the systemic circulation, which in turn are responsible for gastrointestinal bleeding and hepatic encephalopathy, respectively. On the arterial side there is marked arterial vasodilatation, which increases portal venous inflow. This high portal venous inflow plays an important role in the increased pressure in the portal circulation and may explain, at least in part, why portal pressure remains increased despite the development of portocollateral circulation. This arteriolar vasodilatation is also responsible for marked changes in splanchnic microcirculation that may predispose to increased filtration of fluid. It has been shown that the increases in intestinal capillary pressure and lymph flow that occur in chronic portal hypertension (i.e. cirrhosis) are much greater than those caused by an acute increase in portal pressure of the same magnitude (i.e. acute portal vein obstruction). This is probably due to a loss of the normal autoregulatory mechanism of the splanchnic microcirculation. The acute elevation of venous pressure in the intestine elicits a strong myogenic mechanism, which leads to a reduction
Several lines of evidence indicate that portal hypertension is a major factor in the pathogenesis of ascites. First, patients with early cirrhosis without portal hypertension do not develop ascites or edema. Moreover, a certain level of portal hypertension is required for ascites formation. Ascites rarely develops in patients with portal pressure below 12 mmHg, as assessed by the difference between wedged and free hepatic venous pressure (normal portal pressure: 5 mmHg).5,91,92 Second, cirrhotic patients treated with surgical portosystemic shunts for the management of bleeding gastroesophageal varices have a much lower risk of developing ascites than do patients treated with procedures that obliterate gastroesophageal varices but do not affect portal pressure (i.e. sclerotherapy, esophageal transection). Finally, reduction of portal pressure with side-to-side or end-to-side portocaval anastomosis or TIPS is associated with an improvement of renal function and suppression of antinatriuretic systems.93 The mechanism(s) by which portal hypertension contributes to renal functional abnormalities and ascites and edema formation is not completely understood, yet several pathogenic mechanisms have been proposed: 1. Alterations in the splanchnic and systemic arterial circulation that would result in activation of vasoconstrictor and antinatriuretic systems and subsequent renal sodium and water retention; 2. Hepatorenal reflex due to increased hepatic pressure which would cause renal sodium and water retention; 3. Putative antinatriuretic substances escaping from the splanchnic area through portosystemic collaterals that would have a sodium-retaining effect in the kidney. Most data from experimental and human cirrhosis supports the first of these three potential mechanisms (see below for further discussion).
SYSTEMIC CIRCULATION The development of portal hypertension is associated with marked hemodynamic changes not only in the hepatic and splanchnic circulation, as discussed in the previous section, but also in the systemic arterial circulation. These changes, which have been well characterized in human and experimental cirrhosis, consist of reduced systemic vascular resistance and arterial pressure, increased cardiac index, increased plasma volume, and activation of systemic vasoconstrictor and antinatriuretic factors. The decrease in arterial pressure and the stimulation of these vasoactive systems are more marked as the disease progresses.94 Numerous studies have presented evidences that all these hemodynamic changes of cirrhosis are related to an arterial vasodilatation located mainly in the splanchnic circulation.94,95 Whether or not arterial vasodilatation occurs also in non-splanchnic territories is still controversial. Some studies using
425
Section III. Clinical Consequences of Liver Disease
duplex Doppler have found arterial vasodilatation and reduced vascular resistance in the upper and lower limbs, whereas others have reported that blood flow in these arterial beds is normal or even reduced relative to the increased cardiac output.69,73 Whether located exclusively or predominantly in the splanchnic circulation, this arterial vasodilatation causes an abnormal distribution of blood volume, which results in a reduction of the blood volume in the central arterial tree that is sensed by baroreceptors (known as effective arterial blood volume).95,96 This may explain why in most patients with cirrhosis and ascites systemic vasoconstrictor factors remain activated despite an increased plasma volume that in normal conditions would suppress the activation of these systems. The reduction in central blood volume correlates directly with systemic vascular resistance and inversely with portal pressure, indicating that the greater the vasodilatation and the pressure in the portal system, the lower the central blood volume.95 The crucial role played by the reduced central blood volume in the activation of vasoconstrictor systems has been further corroborated by studies showing that improvement of central blood volume by the combination of expansion of plasma volume or head-out water immersion and administration of vasoconstrictor agents, suppresses the activation of vasoconstrictor systems.97–99 Despite extensive investigation, the mechanism(s) responsible for arterial vasodilatation in cirrhosis is not completely understood. Several explanations have been proposed, including opening of arteriovenous fistulas, reduced sensitivity to vasoconstrictors, dysfunction of the autonomic nervous system, and increased circulating levels of vasodilator substances.94 This latter mechanism seems to be the most important and has been extensively investigated. Proposed mediators of this vasodilatation include glucagon, vasoactive intestinal peptide, prostaglandins, natriuretic peptides, platelet-activating-factor, substance P, calcitonin gene-related peptide (CGRP), adrenomedullin, endocannabinoids, and carbon monoxide, but their role in the pathogenesis of vasodilatation is either minor or still unclear. At present, most available data, obtained mainly from experimental cirrhosis, indicate that nitric oxide (NO) is the main mediator of arterial vasodilatation in cirrhosis (Table 22-5)
Table 22-5. Evidence for a Role of an Increased Vascular Production of Nitric Oxide (NO) in the Pathogenesis of Arterial Vasodilatation and Subsequent Sodium and Water Retention in Cirrhosis Experimental cirrhosis Reversal of the impaired pressor response to vasoconstrictors of isolated aortic rings or splanchnic vascular preparations by NOS inhibition Enhanced vasodilator response to NO-dependent vasodilators Increased pressor effect of systemic NOS inhibition Increased NO synthesis in vascular tissue Normalization of the hyperdynamic circulation, activity of antinatriuretic systems and sodium and water retention by chronic NOS inhibition Increased expression of NOS isoenzymes in vascular tissue Human cirrhosis Correction of the arterial hyporesponsiveness to vasoconstrictors by NOS inhibition Enhanced vasodilatory response to NO-dependent vasodilators Increased plasma levels of NO and NO metabolites Increased NO in the exhaled air Increased NOS activity in polymorphonuclear cells and monocytes
426
(reviewed in 100). NO synthesis from arterial vessels of cirrhotic animals is markedly increased compared to that from control animals. This increased NO synthesis appears to occur in the whole vascular bed, except for the intrahepatic circulation, where NO synthesis is reduced (as discussed previously), but predominates in the splanchnic territory.101 Among the different isoforms of NO synthase, the constitutive form appears to be the one responsible for the increased NO synthesis, although a role for the inducible and neuronal isoforms has also been proposed.102 The observation that the normalization of NO synthesis in experimental cirrhosis by the administration of inhibitors of NO synthesis is associated with a marked improvement of splanchnic and systemic hemodynamics, suppression of the increased activity of the renin–angiotensin–aldosterone system and AVP concentration, increased sodium and water excretion, and a reduction in or the disappearance of ascites provides a strong argument in favor of the important role of NO overproduction in the pathogenesis of circulatory dysfunction and subsequent functional renal abnormalities in cirrhosis (Figure 22-6).103 The mechanism of the increased NO synthesis in cirrhosis is not well established. Because arterial vasodilatation occurs predominantly in the splanchnic circulation, a local factor acting in the mesenteric arterial vascular compartment is a very likely mechanism. Endotoxin or other products derived from the activity of the intestinal flora has been proposed as a possible mechanism, as intestinal bacterial orvergrowth, translocation of bacteria from the intestinal lumen into mesenteric lymphatics, or the absorption of bacterial products such as endotoxin, that stimulate the vascular synthesis of NO, are common in cirrhosis with portal hypertension.104 An alternative hypothesis is an activation of the nonadrenergic non-cholinergic nervous system. This is a sensitive system, present in the intestine and other organs and tissues, which responds to mechanical and chemical stimuli and releases neurotransmitters leading to vasodilatation by calcitonin gene-related peptide, substance P, and nitric oxide. During the last few years evidence has been presented indicating that the pathogenesis of the circulatory dysfunction in cirrhosis, particularly in the later phases of disease, is more complex than previously thought. The traditional concept is that the deterioration of systemic hemodynamics in cirrhosis is the consequence of a progression of the splanchnic arterial vasodilatation during the course of the disease as a result of the progression of portal hypertension and liver failure. However, this has not been confirmed in patients with HRS, the extreme expression of circulatory dysfunction in cirrhosis. Studies in these patients suggest that impairment in circulatory function in advanced cirrhosis is due to both a progression of arterial vasodilatation and an impairment in cardiac function.75
NEUROHUMORAL SYSTEMS The functional renal abnormalities that occur in cirrhosis are the result of a complex interplay between different systemic and local (intrarenal) neurohumoral systems. Some of these systems play a pathogenic role in the functional renal abnormalities seen in cirrhosis, whereas others represent defensive mechanisms designed to maintain renal function within normal levels. In this section, the former are referred to as effector mechanisms and the latter as
Chapter 22
3.0 p<0.05 2.5 2.0 1.5 1.0 0.5 0.0
p<0.01 60
40 20
0 Cirrhosis with ascites
Cirrhosis with ascites + LNMA
Cirrhosis with ascites
80
Cirrhosis with ascites + LNMA
60
40 20 0.0
Plasma arginine vasopresin (pg/ml)
6 p<0.05
p<0.001 5 4 3 2 1 0
Cirrhosis with ascites
Cirrhosis with ascites + LNMA
Cirrhosis with ascites
counterbalancing systems. Although the relative contribution of each particular effector and counterbalancing system in the alteration/maintenance of renal function in cirrhosis cannot be definitively established, an attempt has been made to define the role of the most important systems on the basis of available evidence obtained from both experimental and human cirrhosis.
EFFECTOR MECHANISMS Renin–Angiotensin–Aldosterone System The activity of the renin–angiotensin–aldosterone system (RAAS), as assessed by either measuring plasma renin activity (PRA) or angiotensin II or aldosterone plasma levels, is normal in patients without ascites and increased in most patients with ascites.60,105 The effectors of the RAAS are angiotensin II, which has a potent arterial vasoconstrictor effect, and aldosterone, which acts by increasing sodium reabsorption in the renal collecting ducts. Investigations using pharmacological agents that interrupt RAAS activity have provided evidence indicating that this system is activated as a homeostatic mechanism to counteract the reduction in total systemic vascular resistance due to the arterial vasodilatation present in cirrhosis. The administration of angiotensin II receptor antagonists or converting-enzyme inhibitors to cirrhotic patients with ascites and activated RAAS induces a marked reduction in arterial pressure and systemic vascular resistance, but has almost no effect in patients with compensated cirrhosis or healthy subjects, which indicates that the activation of RAAS is a response to maintain arterial pressure106–108 (Figure 22-7). Of all antinatriuretic factors potentially involved in the increased renal sodium retention in cirrhosis, aldosterone has been the most
Cirrhosis with ascites + LNMA
100
90 Mean arterial pressure (mmHg)
Water excretion (% after water load)
Figure 22-6. Effects of inhibition of nitric oxide synthesis with NG-nitro-L-arginine methyl ester on urine sodium, plasma aldosterone concentration, water excretion, and plasma arginine vasopressin concentration in rats with cirrhosis and ascites. (Reproduced with permission from Martin PY, Ohara M, Gines P, et al. Nitric oxide synthase (NOS) inhibition for one week improves renal sodium and water excretion in cirrhotic rats with ascites. J Clin Invest 1998;101:235–242.)
80 Plasma aldosterone (ng/dl)
Sodium excretion mmol/day)
RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
80
70
60
50
40 Pre
All antagonist
Post
Figure 22-7. Effect of infusion of the angiotensin II antagonist 1-sar-8-ala angiotensin II on arterial pressure in cirrhotic patients with (continuous line) and without ascites (discontinuous line). The shaded area represents the effect on normal subjects. (Reproduced with permission from Schroeder ET, Anderson GH, Goldman SH, Streeten DH. Effect of blockade of angiotensin II on blood pressure, renin and aldosterone in cirrhosis. Kidney Int 1976;9:511–519.)
427
Section III. Clinical Consequences of Liver Disease
RAAS.97,98,109 The strongest evidence presented so far supporting the relationship between renal failure and increased RAAS activity derives from a recent study assessing the effects of albumin in the prevention of renal failure in patients with cirrhosis and spontaneous bacterial peritonitis.78 Patients developing renal failure during the infection showed a striking activation of the RAAS compared with patients who did not develop renal failure, in whom RAAS activity remained unchanged throughout the infection (Figure 22-8). Albumin administration was able to prevent not only the activation of the RAAS, but also the development of renal failure. Nevertheless, a firm pathogenic relationship between the RAAS and renal vasoconstriction in HRS has not been possible to establish because the interruption of the RAAS in patients with increased RAAS activity is associated with marked arterial hypotension, as the vasoconstrictor effect of angiotensin II in the systemic arterial circulation cannot be dissociated from that in the renal circulation. Besides its vasoconstrictor effect in the renal circulation, recent studies suggest that the RAAS could also act as vasoconstrictor in the hepatic portal circulation and raise portal pressure by increasing intrahepatic vascular resistance. In fact, the administration of angiotensin II to animals with cirrhosis increases portal pressure, whereas the administration of losartan, an antagonist of the AT1 receptors of angiotensin II, has been shown to lower portal pressure in human cirrhosis. This effect on portal pressure may be due to a relaxation of hepatic stellate cells and myofibroblasts located in terminal hepatic venules which have abundant AT1 receptors. Although the results of these studies require confirmation, they support a role for the RAAS in the pathogenesis of portal hypertension. A summary of the effects of the RAAS on renal and circulatory function in cirrhosis with ascites is shown in Figure 22-9.
extensively studied. Plasma aldosterone levels are increased in most cirrhotic patients with ascites and marked sodium retention. By contrast, in ascitic patients with mild sodium retention plasma aldosterone levels are only slightly elevated or even normal. It should be pointed out, however, that these ‘normal’ concentrations occur in the presence of an increase in total body sodium and plasma volume of a degree that would suppress aldosterone concentration in normal subjects. Evidence supporting an important role for aldosterone in the pathogenesis of sodium retention in cirrhosis includes the following: (1) in patients with ascites there is an inverse correlation between plasma aldosterone levels and urinary sodium excretion;108 (2) in experimental cirrhosis a chronologic relationship between hyperaldosteronism and sodium retention has been demonstrated;11 and (3) the administration of spironolactone, a specific aldosterone antagonist, is able to reverse sodium retention in the great majority of patients with ascites without renal failure.16–18 The observation that sodium retention may occur in cirrhotic patients in the absence of increased plasma aldosterone levels has raised the suggestion that other factors in addition to aldosterone may contribute to the increased sodium retention in cirrhosis. Nevertheless, it has also been suggested that cirrhotic patients may have an increased tubular sensitivity to aldosterone. This may explain the natriuretic response to spironolactone observed in cirrhotic patients with ascites and normal plasma aldosterone concentrations. Thus, the possibility exists that aldosterone may participate in renal sodium retention in cirrhosis even in the presence of normal plasma concentrations of the hormone. In addition to aldosterone, increased intrarenal levels of angiotensin II may also contribute to sodium retention in patients with cirrhosis by a direct effect on proximal tubular sodium reabsorption. So far, however, no direct evidence exists to support this latter contention. Because the activation of RAAS is particularly intense in patients with HRS, and because angiotensin II is a powerful renal vasoconstrictor, a role for angiotensin II in the pathogenesis of renal vasoconstriction in HRS has been proposed.105 This role is further supported by studies showing that the improvement of renal function in patients with HRS achieved by the administration of the vasopressin analogs ornipressin or terlipressin, or the insertion of a TIPS, is associated with a marked suppression of the activity of the
Plasma renin activity (ng/ml/h)
Cefotaxime
Cefotaxime + albumin
25
Numerous studies have presented evidence indicating increased activity of the sympathetic nervous system (SNS) in cirrhosis. The plasma concentration of norepinephrine (NE) in the systemic circulation, an index of the activation of the SNS system, is increased in most patients with ascites and is normal or only slightly elevated in patients without ascites.110 Again, however, this ‘normal’ plasma
No renal failure
Renal failure *
25 *p<0.05 *p<0.05
20 15
20
*
*
15
*
*
*
10
10
5
5
0
*
0 0
3
6 Day
428
Sympathetic Nervous System
9
0
3
6 Day
9
Figure 22-8. Left, Plasma renin activity on days 0, 3, 6, and 9 in patients with cirrhosis and spontaneous bacterial peritonitis treated with cefotaxime plus albumin (1.5 g/kg of body weight at the time of diagnosis and 1 g/kg body weight on day 3) and in patients treated with cefotaxime alone. Right, Plasma renin activity on days 0, 3, 6, and 9 in patients with cirrhosis and spontaneous bacterial peritonitis, divided into those who developed renal failure during the infection and those who did not. (Reproduced with permission from Sort P, Navasa M, Arroyo V, et al. Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med 1999;341:403–409.)
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
Systemic vascular resistance
Figure 22-9. Proposed mechanism of activation and renal and circulatory effects of renin– angiotensin–aldosterone system in cirrhosis with ascites.
Effective arterial blood volume Angiotensinogen Renin Angiotensin I Hepatic portal circulation
Converting enzyme
Systemic circulation Angiotensin II
Vasoconstriction
Arterial vasoconstriction
Aldosterone
Kidney Renal vasoconstriction
Tubular sodium reabsorption
Portal hypertension
Arterial pressure
Sodium retention
Hepatorenal syndrome
Ascites and edema
NE concentration is relatively increased in the presence of plasma volume expansion and high total body sodium, which occurs in early cirrhosis. Investigations using tritiated NE, to provide a more accurate assessment of the SNS activity, have confirmed that the high plasma NE levels are due to an increased activity of the SNS and not to an impaired elimination of NE, as the total spillover of NE to plasma is markedly increased in cirrhotic patients with ascites, whereas the plasma clearance of NE is normal. Measurements of NE release and spillover in specific vascular beds have shown that the activity of the SNS is increased in many vascular territories, including kidneys, splanchnic organs, heart, muscle and skin, supporting the concept of a generalized activation of the SNS. Direct evidence of the overactivity of the SNS in cirrhosis has been provided by measuring the sympathetic nerve discharge rates from a peripheral muscular nerve. Muscular sympathetic nerve activity is markedly increased in patients with ascites and normal in patients without ascites, and correlates directly with plasma NE concentration.111 The cause of the increased activity of the SNS in cirrhosis with ascites is not completely understood. Two major explanations have been proposed: either a baroreceptor-mediated response to a decrease in effective arterial blood volume due to arterial vasodilatation, or a hepatorenal reflex resulting from activation of hepatic baroreceptors due to sinusoidal hypertension. The first explanation seems more likely, as the estimated central blood volume, i.e. the blood volume in the heart cavities, lungs, and central arterial tree, is reduced in cirrhotic patients and correlates inversely with SNS
activity, and the activity of the SNS can be suppressed by maneuvers that increase effective arterial blood volume, such as the administration of vasopressin analogs and albumin or the insertion of a peritoneovenous shunt or TIPS.74,109 Because the SNS has profound effects on renal function, it is reasonable to assume that the increased renal sympathetic nervous activity in cirrhosis may play a role in the pathogenesis of functional renal abnormalities. In fact, evidence suggests that the SNS is involved in sodium and water retention in cirrhosis. The activity of the SNS, either estimated by plasma NE or total NE spillover to plasma or measured from intraneural recordings, correlates inversely with sodium and water retention. In addition, bilateral renal denervation increases urine volume and sodium excretion in animals with experimental cirrhosis and ascites. Similarly, anesthetic blockade of the lumbar SNS, a maneuver that reduces the activity of the kidney SNS, improves sodium excretion in patients with cirrhosis and ascites. Nevertheless, the results of other studies have challenged the role of the SNS as mediator of the increased sodium reabsorption because no correlation was found between the activity of the SNS and sodium excretion; moreover, sodium excretion in patients with cirrhosis did not increase after SNS blockade with clonidine. By contrast, water excretion increased markedly after this latter maneuver.111,112 Because of its powerful renal vasoconstrictor action, a role for the SNS in the pathogenesis of HRS has also been proposed. Patients with HRS have significantly higher plasma levels of NE than do
429
Section III. Clinical Consequences of Liver Disease
patients without renal failure, and both venous peripheral and arterial and renal venous NE levels correlate inversely with renal blood flow, suggesting that the SNS may participate in the renal vasoconstriction observed in patients with HRS. Moreover, the improvement of renal blood flow and GFR after pharmacological treatment or TIPS placement in patients with HRS is paralleled by a marked suppression of the activity of the SNS.97,98,109 Unfortunately, as occurs with the RAAS, the inhibition of the SNS activity in patients with cirrhosis, using the central a2-agonist clonidine, is associated with a marked fall in arterial pressure which does not allow determination of the effects of SNS on the renal circulation. In spite of this reduction in arterial pressure, clonidine administration in patients with cirrhosis is associated with a reduction in renal vascular resistance and an increase in GFR and filtration fraction, suggesting that the activation of the SNS causes renal vasoconstriction by increasing arterial tone in the afferent arteriole. Besides norepinephrine, other neurotransmitters released upon SNS activation may also play a role in renal vasoconstriction. In fact, the circulating levels of neuropeptide Y, a neurotransmitter with a very potent vasoconstrictor action in the renal circulation that is released in the setting of a marked activation of the SNS, are increased in patients with HRS but not in those with ascites without renal failure.113 Besides the effects on renal function, two other potentially important effects of the increased SNS activity deserve a comment. First, the persistent overactivity of the SNS may have deleterious effects on heart function and contribute to the development of the so-called cirrhotic cardiomyopathy, an impairment in cardiac function that occurs in advanced cirrhosis yet is asymptomatic in most patients. In fact, in experimental cirrhosis a down-regulation of cardiac badrenergic receptors has been demonstrated, and in human cirrho-
sis a direct relationship between the increased SNS activity and a prolonged Q-T interval, another finding of the cirrhotic cardiomyopahty, has been shown.114,115 Second, the overactivity of the SNS probably contributes to portal hypertension by increasing intrahepatic vascular resistance, either in the sinusoids or in the terminal hepatic veins. This assumption is based on the observations that the infusion of norepinephrine causes a rise in intrahepatic vascular resistance in isolated and perfused cirrhotic livers, whereas infusion of the b-adrenergic agonist isoproterenol causes opposite findings, and the administration of the a-adrenergic blocker prazosin to patients with cirrhosis and ascites is associated with a marked reduction in portal pressure. A summary of the effects of the SNS on renal and circulatory function in cirrhosis with ascites is shown in Figure 22-10.
Arginine Vasopressin (Antidiuretic Hormone) The plasma levels of AVP are increased in a significant proportion of patients with cirrhosis and ascites. The presence of a normal systemic clearance of AVP in patients with cirrhosis suggests that the high plasma levels of AVP are due to an increased hypothalamic synthesis, an assumption that has been confirmed in experimental cirrhosis.43 The main mechanisms responsible for AVP release in pathophysiologic conditions are an increase in serum osmolality and an impairment in circulatory function. In cirrhosis with ascites, the increased plasma levels of AVP occur in the setting of hypoosmolality and hyponatremia of such a degree that would suppress AVP release in normal subjects. Moreover, AVP levels are frequently unresponsive to the administration of a water load. This indicates that AVP hypersecretion is not due to an osmotic stimulus. The mechanism of the
Systemic vascular resistance
Effective arterial blood volume
Sympathetic nervous system activity Hepatic portal circulation
Systemic circulation
Vasoconstriction
Arterial vasoconstriction
Portal hypertension
Arterial pressure
Kidney
Tubular sodium reabsorption
Renal vasoconstriction
Sodium retention
Hepatorenal syndrome
Heart
Renin release
Impaired heart function
Cirrhotic cardiomyopathy
Ascites and edema Figure 22-10. Proposed mechanism of activation and renal and circulatory effects of sympathetic nervous system in cirrhosis with ascites.
430
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
105
3000
Mean arterial pressure (mmHg)
* p < 0.001
Day –1
100
Day 1
2500
95 90
*
2000
*
1500
85 *
80
1000
75 500 70 V1 antagonist
Saralasin
V1 antagonist + saralasin
Figure 22-11. Arterial pressure in cirrhotic rats with ascites before (open bars) and after (shadowed bars) administration of V1 antagonist, saralasin (angiotensin II antagonist) and a combination of the two. (Reproduced with permission from Claria J, Jimenez W, Arroyo V, et al. Effect of V1-vasopressin receptor blockade on arterial pressure in conscious rats with cirrhosis and ascites. Gastroenterology 1991;100:494–501.)
0 Placebo
25 mg
50 mg
100 mg 200 mg 300 mg
Figure 22-12. Urine volume during 4 hours after the administration of VPA985 (selective V2 receptor antagonist). (Reproduced with permission from Gerbes AL, Gulberg V, Gines P, et al. Therapy of hyponatremia in cirrhosis with a vasopressin receptor antagonist: a randomized double-blind multicenter trial. Gastroenterology 2003;124:933–939.)
Systemic vascular resistance
non-osmotic hypersecretion of AVP in cirrhosis is probably hemodynamic, because plasma AVP levels correlate directly with the activity of the RAAS and SNS and are suppressed by maneuvers that improve circulatory function, such as head-out water immersion or peritoneovenous shunting in human cirrhosis, or inhibition of NO synthesis in experimental cirrhosis. This hemodynamic mechanism of AVP release in cirrhosis is also supported by the observation that the administration of a specific antagonist of the vascular effect of AVP (V1 antagonist) induces arterial hypotension in rats with experimental cirrhosis and ascites and water retention but not in control rats (Figure 22-11). This finding suggests that AVP hypersecretion in cirrhosis contributes to the maintenance of arterial pressure.43,116 Because of its major biological action promoting water reabsorption in the renal collecting tubules through the V2 receptors located in the apical membranes of principal collecting duct cells, AVP is thought to play a major role in the pathogenesis of impaired solutefree water excretion and dilutional hyponatremia present in advanced cirrhosis with ascites. This belief is supported by the following observations in human and experimental cirrhosis: (1) plasma AVP levels correlate closely with the reduction in solute-free water excretion, patients with higher plasma AVP levels being those with the more severe impairment in water excretion; (2) a chronologic relationship between AVP hypersecretion and impairment in water excretion has been demonstrated in rats with cirrhosis and ascites; (3) the impairment in water excretion does not develop in Brattleboro rats (rats lacking AVP because of a congenital deficiency) with cirrhosis; (4) an increased trafficking of aquaporin-2, the AVPregulated water channel which mediates the transport of water from the tubular side to the capillary side of the cell, has been demonstrated in the collecting duct cells from cirrhotic rats; increased aquaporin-2 m-RNA and protein has also been shown in some studies but not in others; and finally (5), the administration of specific antagonists of the tubular effect of AVP (V2 antagonists)
Effective arterial blood volume Systemic circulation
Arterial vasoconstriction
Arterial pressure
Non-osmotic arginine-vasopressin hypersecretion
Kidney
Tubular water reabsorption
Water retention
Dilutional hyponatremia Figure 22-13. Proposed mechanism of hypersecretion and renal and systemic effects of arginine vasopressin in cirrhosis with ascites.
restores the renal ability to excrete solute-free water in experimental as well as in human cirrhosis (Figure 22-12).48,49 A summary of the effects of AVP on renal and circulatory function in cirrhosis with ascites is shown in Figure 22-13.
Endothelins The endothelin family comprises three homologous peptides (ET-1, ET-2, and ET-3) with a very potent vasoconstrictor action, synthesized by a number of cell types, including endothelial cells from the systemic arterial and venous circulation, sinusoidal endothelial cells and hepatic stellate cells from the liver, and mesangial cells from the
431
Section III. Clinical Consequences of Liver Disease
kidney.117 The effects of ETs are mediated by two types of receptor, ETA and ETB, that exhibit distinct selectivity for ET isopeptides. The ETA receptor binds ET-1 and ET-2 with a higher affinity than ET-3, whereas ETB displays similar affinities for all three isopeptides. ETA is responsible for the vasoconstrictor effect of ETs, whereas the stimulation of ETB causes mainly vasodilatation through increased synthesis of NO and prostaglandins. ET synthesized by endothelial cells is thought to participate in the regulation of vascular tone by acting as a paracrine substance on the underlying vascular smooth muscle cells. Because of its marked vasoconstrictor effect, ET has been implicated in the pathogenesis of arterial hypertension and other conditions associated with increased vascular resistance and reduced organ perfusion.117 The role of ET in the pathogenesis of renal, hemodynamic, and neurohumoral abnormalities in cirrhosis has been explored intensively in the last 9 years (over 150 Pubmed citations from 1992 to 2000, including studies in both experimental and human cirrhosis). Despite this great research effort, the origin and pathogenic significance of increased ET levels in cirrhosis still remain incompletely understood. There is general agreement that the plasma levels of ET-1 are increased in patients with cirrhosis and ascites compared with those in healthy subjects.117,118 Most studies have also shown that ET-1 plasma levels are also increased in patients without ascites, albeit to a lesser extent. Plasma ET-3 levels parallel those of ET-1. The high plasma ET levels are not due to a generalized overproduction of ET in all organs and/or in the whole systemic circulation. Rather, they appear to derive from an increased synthesis of the peptide in the splanchnic and hepatic circulations. An increased release of ET from the splanchnic circulation exists in patients with cirrhosis compared to that in control subjects. High ET levels in plasma samples from the splenic vein have been found in patients with cirrhosis, together with increased preproET-mRNA and ET peptide expression in splenic endothelial cells and lymphocytes, suggesting that the splenic circulation is also a site for ET overproduction in cirrhosis. Finally, an increased concentration of ET and its receptors, ETA and ETB, has been demonstrated in experimental and human cirrhotic livers. This increased hepatic ET content appears to derive not only from ET synthesized in sinusoidal endothelial cells and activated hepatic stellate cells,288,293–295 but also from hepatocytes.119 The mechanism(s) leading to the increased hepatosplanchnic production of ET in cirrhosis remains elusive. As opposed to other vasoconstrictor factors the activity of which is increased in cirrhosis with ascites, such as angiotensin II or norepinephrine, it is unlikely that the increased plasma ET levels found in cirrhosis with ascites are a compensatory mechanism triggered by effective arterial hypovolemia, because ET levels are not suppressed by maneuvers that improve circulatory function, such as plasma volume expansion either alone or combined with the administration of splanchnic vasoconstrictors. In addition, the blockade of ET receptors in experimental cirrhosis is not followed by changes in arterial pressure and systemic hemodynamics in most studies. A role for endotoxemia in the increased ET levels in cirrhosis has been proposed based on findings of a correlation between ET and endotoxin levels. However, such correlation has not been demonstrated in all studies, and plasma ET levels do not increase during bacterial infections. Considering that the liver appears to be a major site for ET overpro-
432
duction in experimental cirrhosis and that experimental liver injury (even at very early stages) is associated with increased hepatic ET synthesis from several liver cell types, it is likely that chronic liver injury (whatever the cause) would lead to increased hepatic ET synthesis and high plasma levels of ET. This would explain why plasma and hepatic ET levels are higher in patients with ascites and/or advanced liver disease than those found in patients without ascites and/or preserved liver function. The mechanism of the increased ET synthesis in the splanchnic circulation remains unknown.119 Because plasma ET levels are increased mainly in patients with ascites, and because ET has major effects on kidney function, a role for ET in the pathogenesis of renal functional abnormalities has been proposed. Of all renal effects of ET, the vasoconstrictor effect has been the most extensively studied. Although possible, the potential role of ET as a pathogenic factor contributing to renal vasoconstriction in HRS has not been established convincingly. An inverse relationship between plasma ET levels and renal blood flow and/or GFR in patients with ascites has been demonstrated in some studies, but not in others. The correlation found between plasma ET levels and renal function was weak in most studies. Moreover, studies specifically assessing ET plasma levels in patients with HRS also show discrepant findings, with some studies showing very high values and others values similar to those found in patients with ascites without renal failure. Nevertheless, it should be noted that because of the paracrine action of ET, circulating ET levels do not necessarily reflect ET levels within the kidney, and that intrarenal ET content has not been assessed in HRS. Finally, an improvement of renal function after the administration of a selective antagonist of ETA receptors was reported in a small group of patients with HRS, but this has not been followed by studies in larger groups of patients.120 Some clinical and experimental data suggest that ET participates in the modulation of water excretion in cirrhosis. The plasma levels of ET correlate inversely with solute-free water excretion and serum sodium, and directly with plasma AVP levels in human cirrhosis, and ETA and ETB receptor blockade in cirrhotic rats is followed by an impairment in water excretion. These findings, together with the demonstration of increased ET-1 levels in the renal medulla and overexpression of ETB receptor in the inner medullary collecting duct cells of rats with cirrhosis, suggest that ET may act in cirrhosis as a paracrine system to maintain water excretion in the setting of increased AVP levels. A role for ET in the increased renal sodium reabsorption in cirrhosis is less likely because no correlation between plasma or urinary ET and sodium excretion in patients with cirrhosis has been found in most studies, and ET receptor blockade in cirrhotic rats is not associated with changes in sodium excretion.119 As stated previously, a role for ET in the maintenance of arterial pressure in patients with cirrhosis to compensate for the marked arterial vasodilatation is highly unlikely. By contrast, it is likely that the increased intrahepatic synthesis of ET in cirrhosis plays a significant role in the pathogenesis of portal hypertension by increasing intrahepatic vascular resistance through an autocrine or paracrine action of ET on contractile cells present in the sinusoids and terminal hepatic venules (activated hepatic stellate cells and myofibroblasts). A number of studies in different models of portal hypertension in rats (bile duct ligation, portal vein ligation, and carbon tetrachloride injury) have shown that ET receptor blockade
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
is followed by a reduction in portal pressure. Studies in patients with cirrhosis are awaited. Finally, recent studies also suggest a role for ET in the regulation of liver fibrogenesis by modulating collagen synthesis from hepatic stellate cells. However, the discussion of this effect is beyond the scope of this chapter. A summary of the origin and effects of ET on the kidney and liver in cirrhosis is shown in Figure 22-14.
Vasoconstrictor Eicosanoids In addition to the vasodilator prostaglandins, the kidney synthesizes a number of eicosanoids that have antinatriuretic and/or vasoconstrictor activities. The renal production of some of these eicosanoids has been investigated in patients with cirrhosis. It was first suggested that HRS could be the consequence of an imbalance between the renal synthesis of vasodilator and vasoconstrictor eicosanoids based on the observation of reduced urinary excretion of prostaglandin E2 (PGE2) and 6-keto-PGF1a (a metabolite of prostaglandin I2 – PGI2) and increased urinary excretion of thromboxanes (i.e. TXB2) in patients with HRS. These findings, however, were not confirmed by subsequent investigations, in which the urinary excretion of TXB2 in patients with HRS was found to be decreased.103,317,318 Moreover, the inhibition of thromboxane synthesis does not improve renal function in these patients.121–123 Little is known about the possible role of other eicosanoids in the pathogenesis of functional renal abnormalities in cirrhosis. The urinary excretion of leukotriene E4 and N-acetyl-leukotriene E4, compounds with a vasoconstrictor effect on the renal circulation, is increased in cirrhotic patients with HRS compared to that of healthy subjects and patients without ascites, suggesting that leukotrienes may participate in the pathogenesis of renal vasoconstriction. Moreover, it has also been shown that the concentration of the vasoconstrictor metabolite 20-hydroxyeicosatetraenoic acid is increased markedly in the urine of patients with cirrhosis and ascites, probably reflecting an increased renal synthesis of the compound compared with that of control subjects, and correlates inversely with renal blood flow, thus raising the possibility of a vasoconstrictor action of this compound in the cirrhotic kidney.121,124 The potential role of these as well as other vasoconstrictor eicosanoids in the
Liver injury
Liver
Increased intrahepatic Kidney synthesis of endothelin
Increased intrahepatic vascular resistance
Tubular water reabsorption
Renal vasoconstriction
Portal hypertension
Maintenance of water excretion
Hepatorenal syndrome
Figure 22-14. Proposed mechanism of hypersecretion and renal and hepatic effects of endothelin in cirrhosis with ascites.
pathogenesis of renal functional abnormalities in cirrhosis deserves further investigation.
Adenosine Adenosine is an endogenous nucleoside produced locally in most cells by the intracellular degradation of adenosine triphosphate, mainly in response to hypoxia. Adenosine is a potent vasodilator in most vascular beds except the kidneys, where it causes vasoconstriction. Adenosine-1 receptors are present on the afferent arteriole in the kidney and cells of the proximal tubules, whereas adenosine-2 receptors are found in the systemic vasculature. Stimulation of adenosine-1 receptors cause renal vasoconstriction and sodium and water retention, whereas that of adenosine-2 receptors causes vasodilatation. The possible role of adenosine in the pathogenesis of renal functional abnormalities in human cirrhosis has been assessed using several approaches. The acute administration of methylxanthines, which act as non-specific adenosine antagonists, to patients with cirrhosis and ascites is associated with an increase in RBF, GFR, and sodium and water excretion, whereas the acute administration of an adenosine-1 receptor antagonist to patients with cirrhosis and ascites induces a marked increase in sodium excretion and urine flow, without changes in renal hemodynamics.125 Conversely, the acute administration of dipyridamole, a drug that acts (at least in part) by increasing the levels of adenosine in the extracellular fluid due to inhibition of the cellular uptake of this substance, is associated with renal vasoconstriction and increased sodium and water retention and increased activity of the RAAS, particularly in patients with ascites.126
COUNTERBALANCING SYSTEMS Prostaglandins Prostaglandins (PG) are known to have a protective effect on renal circulation in pathophysiologic situations associated with increased activity of renal vasoconstrictor systems. According to this formulation, PGs appear to play a key role in the homeostasis of renal circulation and water excretion in cirrhotic patients with ascites. The urinary excretion of PGE2 and 6-keto-prostaglandin F1a, which estimate the renal synthesis of PGE2 and PGI2, respectively, is increased in patients with cirrhosis and ascites without renal failure compared to that of healthy subjects and patients without ascites. Studies in experimental cirrhosis have shown that this increased synthesis of PGs in the kidney is related to an increased activity and expression of cytosolic phospholipase A2 (cPLA2), the first enzyme of the metabolic cascade of eicosanoid synthesis.127 Evidence supporting a role for renal PGs in the maintenance of RBF and GFR in cirrhosis with ascites derives from studies using NSAIDs to inhibit PG synthesis. The non-selective inhibition of the cyclooxygenase activity (COX-1, constitutive, and COX-2, inducible) by the administration of classic NSAIDs, even in single doses, to cirrhotic patients with ascites causes a profound decrease in RBF and GFR in patients who have a marked activation of vasoconstrictor systems, but has no or little effect in patients without activation of these systems (Figure 22-15). By contrast, the administration of selective inhibitors of COX-2 activity neither decreases significantly the renal synthesis of PGE2 and PGI2 nor affects renal
433
Section III. Clinical Consequences of Liver Disease
12 Free water clearance (ml/min)
140
GFR (ml/min)
120 100 80 60 40
Membrane phospholipids
8
Angiotensin II Norepinephrine arginine vasopressin
Phospholipase A2
6 Arachidonic acid
4
Systemic and splanchnic circulation
2
Kidney Prostaglandins
0 Baseline
434
10
Sulindac
Baseline
Sulindac
Figure 22-15. Effects of sulindac, 400 mg/day for 3 days, on glomerular filtration rate (GFR) and free water clearance in cirrhotic patients with ascites.
Arterial vasodilation
function in rats with cirrhosis and ascites or in patients with decompensated cirrhosis and ascites.128 Besides contributing to the maintenance of glomerular filtration PGE2 also contributes to the maintenance of solute-free water excretion in non-azotemic cirrhotic patients with ascites, as the inhibition of PG synthesis by NSAIDs in these patients impairs solute-free water excretion independently of changes in renal hemodynamics. The relationship between the renal PG system and HRS is controversial. As pointed out earlier, several studies have reported that patients with HRS have lower urinary excretion of PGE2 and 6-ketoPGF1a than do patients with ascites without renal failure, which suggests that a reduced renal synthesis of vasodilator PGs may play a role in the pathogenesis of HRS.127,129 The finding of low renal levels of cyclooxygenase in patients with HRS is consistent with this hypothesis. Other studies, however, did not find reduced urinary excretion of vasodilator PGs in patients with HRS. Nevertheless, ‘normal’ synthesis of PGs may be low relative to the increased activity of vasoconstrictor systems in cirrhosis. Because patients with HRS have the greatest activation of renal vasoconstrictor systems, an imbalance between vasoconstrictor systems and the renal production of vasodilator PGs has been proposed to explain, at least in part, the marked reduction of RBF and GFR that occur in this condition.1,129 PGs synthesis in cirrhosis is also increased in extrarenal organs. Increased expression of COX-1, but not COX-2, in the splanchnic vasculature and thoracic aorta has been demonstrated in animals with prehepatic portal hypertension induced by portal vein ligation. Furthermore, it has been shown that the inhibition of PG synthesis by NSAIDs is associated with a reduction of splanchnic blood flow, which suggests a role for PGs in the pathogenesis of splanchnic arterial vasodilatation. Consistent with these findings is the observation that patients with cirrhosis have high urinary excretion of 2-3-dinor6-keto-PGF1a, a metabolite of PGI2 considered to be an index of systemic PGI2 production. Moreover, the inhibition of PG synthesis by the administration of indomethacin increases systemic vascular resistance and ameliorates the hyperdynamic circulation in cirrhotic patients. A summary of the effects of PGs in renal and circulatory function in cirrhosis is shown in Figure 22-16.
Arterial pressure
Renal vasodilation
Inhibition of arginine vasopressin effect
Maintenance of renal hemodynamics
Maintenance of water excretion
Figure 22-16. Proposed mechanism of activation and renal and circulatory effects of prostaglandins in cirrhosis with ascites.
Natriuretic Peptides Despite some controversial findings in early studies, most recent investigations show that the plasma concentration of atrial natriuretic peptide (ANP), the most important member of the natriuretic peptide family, is increased in patients with cirrhosis and ascites.130 In patients without ascites, plasma ANP levels have been reported to be either normal or increased. The high plasma levels of ANP in cirrhosis with ascites are due to increased cardiac secretion of the peptide and not reduced hepatic or systemic catabolism, as cardiac production of ANP is increased in cirrhotic patients with ascites and splanchnic and peripheral extraction are normal.131 Consistent with these observations is the finding of increased mRNA expression for ANP in ventricles from cirrhotic rats with ascites. The cardiac production and release of ANP in cirrhosis with ascites can be further increased by maneuvers that increase the central blood volume, such as insertion of a peritoneovenous shunt or TIPS. In contrast to other diseases showing increased cardiac ANP secretion, in cirrhosis with ascites this increased secretion occurs in the presence of normal atrial pressure and reduced estimated central blood volume. The mechanism(s) responsible for this increased cardiac secretion of ANP is not known. The existence of increased plasma levels of ANP in cirrhosis with ascites sufficient to have a natriuretic effect in healthy subjects, together with the presence of renal sodium retention, indicates a renal resistance to the effects of ANP. This has been confirmed in studies in human and experimental cirrhosis in which pharmacological doses of natriuretic peptides (ANP or brain natriuretic peptide, BNP) were administered. In these investigations, patients with activation of antinatriuretic systems (RAAS and SNS) had a blunted or no natriuretic response after ANP infusion. This blunted response can be reversed by maneuvers that increase distal sodium delivery in human cirrhosis, or by bilateral renal denervation in experimental cirrhosis, suggest-
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
ing that the renal resistance to ANP in cirrhosis is related to the increased activity of antinatriuretic systems.131,132 Limited information exists on other peptides of the natriuretic peptide family. As with ANP, the plasma concentration of BNP is increased in cirrhotic patients with ascites compared to that of healthy subjects.38 Recently, it has been reported that patients with cirrhosis and ascites and renal failure have increased urinary excretion of C-type natriuretic peptide, together with reduced plasma levels of the peptide compared to those of healthy subjects.133 Finally, the urinary excretion of urodilatin, a member of the natriuretic peptide family exclusively synthesized in the kidney, which probably reflects the renal production of the peptide, is normal in patients with cirrhosis and ascites.134 The role that natriuretic peptides play in cirrhosis is unknown. Because most of these peptides have vasodilator properties, a role in the pathogenesis of arterial vasodilatation in cirrhosis has been proposed but not proved. By contrast, data from experimental studies suggest that they play an important role in the maintenance of renal perfusion and modulation of RAAS activity, as the selective blockade of the natriuretic peptide A and B receptors causes renal vasoconstriction and increased PRA and aldosterone levels in experimental cirrhosis. It could be speculated, therefore, that the cardiac synthesis of ANP and BNP is increased in an attempt to maintain renal perfusion within normal levels and limit the activation of the RAAS. The mechanism(s) leading to this increased synthesis of natriuretic peptides remains unknown. During the last few years BNP has attracted increasing attention as an accurate marker of left ventricular dysfunction. BNP is an independent predictor of high left ventricular pressure, estimates left ventricular dysfunction, and closely correlates with the New York Heart Association classification. The increased release of BNP in left ventricular dysfunction is thought to be a compensatory response elicited by ventricular remodeling aimed at reducing systemic pressure load hypertrophy through sodium and water diuresis. Recent studies have proposed that the increased cardiac release and plasma levels of BNP in cirrhosis may also be a consequence of cirrhotic cardiomyopathy.135 The observation that the plasma levels of BNP in cirrhosis correlate with septal thickness, left ventricular diameter at the end of diastole, and QT interval rather than cardiac output and peripheral vascular resistance is in keeping with this belief. Distinct mechanisms may therefore influence the increased cardiac release of the different natriuretic peptides in cirrhosis.
Nitric Oxide Apart from its effects in the regulation of systemic hemodynamics and arterial pressure, NO also participates in the regulation of renal function in health and disease. Both constitutive and inducible nitric oxide synthase have been found in several cell types in the kidney, and NO participates in the modulation of arteriolar tone and mesangial cell contractility under normal circumstances. Moreover, NO facilitates natriuresis in response to changes in renal perfusion pressure and regulates renin release. Several lines of evidence indicate that the production of NO is increased in the kidney of rats with experimental cirrhosis. First, kidneys from cirrhotic rats show enhanced endothelium-dependent vasodilator response compared to those from control animals. Second, infusion of L-arginine, the precursor of NO, causes a greater
increase in renal perfusion in cirrhotic rats than in control rats.135a Finally, increased expression of nitric oxide synthase in kidney tissue from cirrhotic rats has been found in two studies, although discrepant findings were obtained with respect to the NOS isoform responsible for the increased NO synthesis.136,137 The role that the increased renal NO production found in experimental cirrhosis plays in the regulation of renal function is not known, because the inhibition of NO synthesis does not result in renal hypoperfusion. Nevertheless, because NO synthesis inhibition gives rise to a marked increase in urinary PG excretion and the simultaneous inhibition of NO and PG synthesis results in a marked renal vasoconstriction, it is likely that NO interacts with PGs to maintain renal perfusion in cirrhosis. No information exists on the possible role of renal NO synthesis in renal function abnormalities in human cirrhosis.
Renal Kallikrein–Kinin System Some studies have reported that urinary kallikrein excretion is increased in patients with ascites without renal failure and reduced in patients with HRS, and correlates directly with GFR, suggesting that the renal kallikrein–kinin system, an intrarenal system with vasodilating and natriuretic activities, may contribute to the maintenance of renal hemodynamics in cirrhosis. Other investigations, however, have found reduced urinary kallikrein excretion in patients with ascites.138,139 More specific methods to evaluate the activity of the kallikrein–kinin system are needed before the role of this system in the homeostasis of renal function in cirrhosis can be defined.
TIME-COURSE OF RENAL FUNCTION ABNORMALITIES IN CIRRHOSIS The course of renal function abnormalities in cirrhosis is usually progressive, although in some cases (mainly alcoholic cirrhosis) renal function may improve during follow-up. The main consequence of the reduced ability to excrete sodium in cirrhosis is the development of sodium retention and ascites, and this occurs when the renal sodium excretion falls below the sodium intake.9 This represents a marked impairment in renal sodium metabolism. The renal ability to excrete free water in healthy subjects is far in excess of that required to eliminate the water ingested in a regular diet. Under conditions of water overload the kidneys’ ability to excrete water approaches 10 ml/min (14 l/day) in healthy individuals, an amount that is taken only by patients with serious psychiatric conditions.140 Dilutional hyponatremia (arbitrarily defined as a serum sodium concentration of less than 130 mEq/l) is the clinical consequence of the impaired free water excretion, and this occurs when free water clearance is severely reduced (usually <1 ml/min).141 Finally, the main consequence of the impaired renal perfusion and GFR is hepatorenal syndrome (HRS), which has been arbitrarily defined as a GFR below 40 ml/min (normal GFR >120 ml/min) or a serum creatinine >1.5 mg/dl. Sodium retention, dilutional hyponatremia and HRS appear at different times during the evolution of the disease.62 Therefore, the clinical course of ascites in cirrhosis can be divided into phases according to the onset of each one of these complications.
435
Section III. Clinical Consequences of Liver Disease
PHASE 1: IMPAIRED RENAL SODIUM RETENTION IN COMPENSATED CIRRHOSIS Chronologically, the first renal function abnormality occurring in cirrhosis is an impairment of renal sodium metabolism, which can already be detected before the development of ascites. At this stage of the disease patients have normal renal perfusion, GFR and free water clearance, and are able to excrete the sodium ingested with the diet. However, they have subtle abnormalities in renal sodium excretion.142 For example, they have a reduced natriuretic response to the acute administration of sodium chloride (i.e. after the infusion of a saline solution) and may not be able to escape from the sodium-retaining effect of mineralocorticoids.37,38,143 Abnormal natriuretic responses to changes in posture is another relevant feature at this phase of the disease. Urinary sodium excretion is reduced in the upright and increased in the supine posture compared to normal subjects.30,144 Moreover, these patients show an increased plasma volume, suggesting sodium retention.145 It is interesting that some of these abnormalities develop in patients with higher portal pressure and lower peripheral vascular resistance, indicating a relationship with the deterioration of circulatory function.38 The term ‘preascitic cirrhosis’ has been used to define this phase of the disease, although no study has demonstrated that it represents a state of impending ascites formation. Nevertheless, it is possible that the renal ability to excrete sodium in some patients with compensated cirrhosis may be just at the limit of a normal sodium intake. In these patients the formation of ascites might be precipitated by increasing the intake of sodium or by impairing renal sodium excretion, for example after the administration of vasodilators such as nitrates146,147 or prazosin.34
PHASE 2: RENAL SODIUM RETENTION WITHOUT ACTIVATION OF THE RENIN–ANGIOTENSIN–ALDOSTERONE AND SYMPATHETIC NERVOUS SYSTEMS As a result of disease progression the patient becomes unable to excrete their regular sodium intake. Sodium is then retained together with water, and the fluid accumulates in the abdominal cavity as ascites. Urinary sodium excretion, although reduced, is usually higher than 10 mEq/day, and in some cases it is over 50–90 mEq/day. Hence, a negative sodium balance and therefore the loss of ascites, may be achieved only by reducing the sodium content of the diet.148,149 Renal perfusion, GFR, the renal ability to excrete free water, plasma renin activity and the plasma concentrations of antidiuretic hormone are normal.145,150,151 In this setting, sodium retention is unrelated to the renin–aldosterone system and the sympathetic nervous system, the two most important antinatriuretic systems so far identified.12 The plasma levels of atrial natriuretic peptide, brain natriuretic peptide, and natriuretic hormone are increased in these patients, indicating that sodium retention is not due to a reduced synthesis of endogenous natriuretic peptides.130,152 It has been suggested that circulatory dysfunction at this phase, although greater than in compensated cirrhosis without ascites, is not intense enough to stimulate the sympathetic nervous and renin–angiotensin–aldosterone systems. However, it would activate a still unknown, extremely sensitive sodium-retaining mechanism (renal or extrarenal).145,153 Alternatively, it has been proposed that
436
sodium retention at this phase of the disease is unrelated to circulatory function (i.e. increased renal tubular sensitivity to aldosterone or catecholamines,145,151 decreased synthesis of a putative hepatic natriuretic factor, or the existence of hepatorenal nervous reflexes promoting sodium retention154). However, this is unlikely, as sodium retention in the absence of an impaired circulatory function would be associated with arterial hypertension, a feature not observed in patients with decompensated cirrhosis, who are in fact hypotensive. Investigations into intrarenal sodium handling suggest that sodium retention in these patients occurs predominantly in the distal nephron.110
PHASE 3: STIMULATION OF THE ENDOGENOUS VASOCONSTRICTOR SYSTEMS WITH PRESERVED RENAL PERFUSION AND GFR When sodium retention is intense (urinary sodium excretion below 10 mEq/day), the plasma renin activity and the plasma concentrations of aldosterone and norepinephrine are invariably increased.14,155–157 Aldosterone increases sodium reabsorption in the distal and collecting tubules. In contrast, the renal sympathetic nervous activity stimulates sodium reabsorption in the proximal tubule, loop of Henle, and distal tubule.158,159 Thus, sodium retention in these patients is due to increased sodium reabsorption in the entire nephron. The plasma volume, cardiac output, and peripheral vascular resistance in these patients do not differ from the previous phase.157 Circulatory dysfunction, however, is more intense because increased activity of the sympathetic nervous system and renin–angiotensin system is needed to maintain circulatory homeostasis. Arterial pressure at this phase of the disease is critically dependent on the increased activity of the renin–angiotensin and sympathetic nervous systems and antidiuretic hormone. The administration of drugs that interfere with these systems (saralasin106), losartan,160 convertingenzyme inhibitors,107 clonidine,161 V1 vasopressin antagonists162 may precipitate arterial hypotension and renal failure. Although angiotensin II, norepinephrine, and antidiuretic hormone are powerful renal vasoconstrictors, renal perfusion and GFR in this phase are normal or only moderately reduced because their effects on the renal circulation are antagonized by intrarenal vasodilator mechanisms, particularly prostaglandins.129 Cirrhosis is a human condition in which renal perfusion and GFR are more dependent on the renal production prostaglandins, and renal failure may develop in this phase if renal prostaglandins are inhibited with NSAIDs.82,163–166 Other vasodilatory systems probably involved in the maintenance of renal function at this phase of the disease are nitric oxide167 and the natriuretic peptides.168,169 The kidney’s ability to excrete free water is reduced at this phase of the disease, owing to the high circulating plasma levels of antidiuretic hormone.170 However, only a few patients have significant hyponatremia141 because the effect of antidiuretic hormone is counteracted by an increased renal production of prostaglandin E2.171
PHASE 4: THE DEVELOPMENT OF TYPE 2 HEPATORENAL SYNDROME Type 2 HRS develops in very advanced phases of cirrhosis in the setting of worsening of circulatory function. Patients with type 2 HRS
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
have very high plasma levels of renin, aldosterone, norepinephrine, and antidiuretic hormone, significant arterial hypotension, and increased heart rate.65 The arterial vascular resistance in these patients is increased not only in the kidneys,73,172 but also in the brain,74 muscle,69 and skin, indicating a generalized arterial vasoconstriction to compensate for the intense splanchnic arterial vasodilatation.173 Type 2 HRS is probably due to the extreme overactivity of the endogenous vasoconstrictor systems, which overcomes the intrarenal vasodilatory mechanisms.61 There are studies suggesting that in these patients the cardiac output may be not as high as in the previous phase.1 However, further studies are needed to confirm this feature. The degree of sodium retention in type 2 HRS is very intense. These patients exhibit a reduced filtered sodium load and marked increased sodium reabsorption in the proximal tubule.174 The delivery of sodium to the distal nephron, the site of action of diuretics, is very low.175 Therefore, most of these patients do not respond to diuretics and present with refractory ascites.62 Free water clearance is also markedly reduced and most patients show significant hyponatremia.141 The prognosis of patients with type 2 HRS is very poor, with a survival rate of 50% and 20% at 5 months and 1 year after the onset of the renal failure, respectively (V. Arroyo, unpublished observations).
PHASE 5: THE DEVELOPMENT OF TYPE 1 HEPATORENAL SYNDROME Although type 1 HRS may arise spontaneously it frequently occurs in close chronological relationship with a precipitating factor, such as severe bacterial infection, acute hepatitis (ischemic, alcoholic, toxic, viral) superimposed on cirrhosis, a major surgical procedure, or massive gastrointestinal hemorrhage.62 Patients with type 2 HRS are especially predisposed to develop type 1 HRS,1 although type 1 HRS may develop in patients with normal serum creatinine concentrations. The prognosis of patients with type 1 HRS is extremely poor, with 80% of patients dying less than 2 weeks after the onset of HRS63 (see Figure 22-4). Patients die with progressive circulatory, hepatic and renal failure, and hepatic encephalopathy. Type 1 HRS has been of particular interest in patients with spontaneous bacterial peritonitis (SBP) as 30% of patients with SBP develop this type of renal failure.76 The two most important predictors of type 1 HRS development in SBP are an increased serum creatinine prior to the infection and an intense intra-abdominal inflammatory response, as suggested by high ascitic fluid concentration of polymorphonuclear leukocytes and cytokines (tumor necrosis factor (TNF)-a and interleukin (IL)-6) at the time of diagnosis of the infection.77 It is unknown whether these features represent a more severe infection or a late diagnosis of the infection. SBPinduced HRS develops in most patients despite a rapid resolution of the infection with antibiotics.176,177 Type 1 HRS after SBP occurs in the setting of a severe deterioration of circulatory function, as indicated by a reduction in arterial pressure and a marked increase in the plasma levels of renin and noradrenaline.63,178 Two recent studies have assessed systemic hemodynamics and renal function in a large series of patients with SBP at infection diagnosis and 1 week later.178,179 Resolution of the infection occurred in most cases. Development of HRS in these patients was associated with a significant decrease in cardiac output and in
the absence of changes in peripheral vascular resistance, suggesting that the impairment in circulatory function in patients with HRS is far more complex than previously considered. In addition to arterial vasodilatation in the splanchnic circulation, it is evident that a decreased cardiac output also participates in the impairment of the effective arterial blood volume during SBP. The demonstration that plasma volume expansion with albumin given at the time of diagnosis of the infection reduces by more than 60% the incidence of renal impairment and hospital mortality in patients with SBP supports the concept that the reduction in cardiac output associated with HRS is probably related to a decreased central blood volume and cardiac preload.78 The progressive nature of renal failure in type 1 HRS is related to the rapid deterioration of circulatory function observed in these patients, although changes in intrarenal vasoactive mechanisms are probably also of great importance. As previously mentioned, the kidney produces vasodilatory substances, such as prostaglandins and nitric oxide, that diminish the effect of the endogenous vasoconstrictor systems on renal perfusion and GFR.167,169,180 The moderate and steady course of type 1 HRS is probably related to a decreased production of the substances that antagonize the intense overactivity of the renin–angiotensin and sympathetic nervous systems and antidiuretic hormone. When there is intense reduction of renal perfusion the synthesis of these vasodilatory substances may be impaired.110 On the other hand, renal ischemia stimulates the intrarenal synthesis of vasoconstrictor substances such as angiotensin II181 and adenosine.182 Therefore it can be postulated that type 1 HRS is initiated by an acute deterioration of circulatory function promoted by a precipitating event in patients who already have severely compromised circulatory function. This would lead to renal ischemia, increased intrarenal production of vasoconstrictor systems, decreased synthesis of renal vasodilators and more renal ischemia, thus creating a vicious cycle within the kidney that accentuates and perpetuates the deterioration of renal function. Several features support this mechanism: (1) a syndrome comparable to type 1 HRS can be produced in cirrhotic patients with ascites and increased activity of the renin–angiotensin and sympathetic nervous systems, or in experimental animals with carbon tetrachlorideinduced cirrhosis and ascites following inhibition of prostaglandin synthesis61,82,129,163,164,166,183 and nitric oxide,167 or after the administration of dypiridamole, a drug that increases the circulating levels of adenosine;126 (2) the long-term (1–2 weeks) administration of intravenous albumin and vasoconstrictor substances (ornipressin,97 noradrenaline184) improves circulatory function and suppresses plasma renin activity and norepinephrine concentration to normal or near-normal levels within the first 2–3 days of treatment. However, an increase in GFR is not observed until 1–2 weeks later. Therefore, there is a clear lag between the normalization of systemic circulatory function and the improvement in renal perfusion and GFR, which may be the period required for the deactivation of the intrarenal mechanisms; (3) once HRS has been resolved following plasma volume expansion with albumin and the administration of vasoconstrictor agents it does not recur after stopping treatment, suggesting that the rapidly progressive renal failure is due more to the precipitating event than to the liver disease itself.98 The development of HRS in patients with SBP is associated not only with a deterioration of circulatory and renal function, but
437
Section III. Clinical Consequences of Liver Disease
also with an impairment in hepatic function leading to hepatic encephalopathy. A recent study has shown that in these patients there is an increase in the intrahepatic vascular resistance and portal pressure and a fall in hepatic blood flow that correlates closely with an increase in renin and norepinephrine levels.178 Circulatory dysfunction in HRS, therefore, also affects the liver.
PATHOPHYSIOLOGY OF FUNCTIONAL RENAL ABNORMALITIES IN CIRRHOSIS THE OVERFLOW THEORY The existence of a primary defect in renal sodium reabsorption in patients with cirrhosis with ascites was proposed in an attempt to explain the paradox of coexistence of sodium retention and increased plasma volume in patients with ascites. According to this theory, the expansion of plasma volume would result in an increased cardiac index and reduced systemic vascular resistance as an adaptive circulatory response to the increased intravascular volume. The existence of portal hypertension and circulating hypervolemia would lead to ‘overflow’ of fluid within the peritoneal cavity. It has been proposed that the primary signal for sodium retention would arise from the liver, either as a consequence of intrahepatic portal hypertension by means of hepatic low-pressure baroreceptors, or liver failure by means of decreased hepatic clearance of a sodiumretaining factor or reduced hepatic synthesis of a natriuretic factor12 (Figure 22-17). However, the hemodynamic pattern of cirrhotic patients with ascites does not correspond with that predicted by the overflow theory because the arterial vascular compartment is not overfilled, as arterial pressure is low in most patients despite the increased plasma volume and cardiac index (Table 22-6). Moreover, in most patients with ascites there is marked overactivity of vasoconstrictor mechanisms, which would be suppressed if there were overfilling in the systemic circulation.95 Because of the increasing evidence against the existence of vascular overfilling in cirrhosis with ascites, the overflow theory has recently been redefined to exclusively explain changes that occur in the preascitic stage of cirrhosis. Proponents of this theory suggest that in the preascitic stage subtle sodium retention leading to plasma
volume expansion would have two components: one related to the circulatory changes occurring in the splanchnic circulation aimed at maintaining the effective arterial blood volume (EABV), and one related to the existence of intrahepatic portal hypertension. Recent studies in patients with cirrhosis without ascites indicate that the existence of arterial vasodilatation is crucial in the development of sodium retention and ascites formation. In fact, preascitic cirrhotic patients with sinusoidal portal hypertension treated with mineralocorticoids do not show mineralocorticoid escape and develop ascites only when marked arterial vasodilatation is present. Moreover, vasodilatation induced pharmacologically in preascitic cirrhotic patients by means of the administration of prazosin, an a-adrenergic blocker, is associated with the development of ascites and/or edema in a significant proportion of cases. It is important to note that the development of sodium retention in these two studies was related neither to the degree of portal hypertension nor to the intensity of liver failure. In fact, in patients receiving prazosin, sodium retention occurred despite a marked reduction of portal pressure and improvement of liver perfusion.
THE ARTERIAL VASODILATATION THEORY The traditional concept of ascites formation in cirrhosis considers that the key event is a ‘backward’ increase in hydrostatic pressure
Hepatorenal reflex? Liver failure?
Cirrhosis
Hepatorenal reflex? Liver failure?
Sodium and water retention
Renal vasoconstriction
Plasma volume expansion
Functional renal failure
Hyperdynamic circulation
Ascites formation
Figure 22-17. Pathogenesis of functional renal abnormalities and ascites formation in cirrhosis according to the overflow theory.
Table 22-6. Hemodynamic Profile of Cirrhotic Patients in Different Stages of the Disease
Cardiac output Arterial pressure Systemic vascular resistance Plasma volume Portal pressure Vasoconstrictor system activity Renal vascular resistance Brachial or femoral vascular resistance Cerebral vascular resistance *Markedly reduced in hepatorenal syndrome type 1. **May be normal in 20–30% of patients.
438
Preascitic cirrhosis
Cirrhosis with ascites
Hepatorenal syndrome
Normal or increased Normal Normal or reduced Normal or increased Normal or increased Normal Normal Normal or reduced Normal
Increased Normal or reduced Reduced Increased Increased Increased** Normal or increased Normal or increased Increased
Increased, normal or decreased Reduced* Reduced Increased Increased Markedly increased Markedly increased Increased Increased
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
in the hepatic and splanchnic circulations owing to the increased resistance to portal flow. This would cause a disruption of the Starling equilibrium and an increased filtration of fluid into the interstitial space. Initially, this capillary hyperfiltration is compensated for by an increase in lymphatic flow, which returns the fluid to the systemic circulation via the thoracic duct. However, as portal hypertension increases, the lymphatic system is unable to drain the excess of interstitial fluid, which then accumulates in the peritoneal cavity as ascites. Loss of fluid from the intravascular compartment results in true hypovolemia, which is then sensed by cardiopulmonary and arterial receptors resulting in a compensatory renal sodium retention. The retained fluid cannot adequately fill the intravascular compartment and suppress the sodium-retaining signals to the kidney because fluid is continuously leaking into the peritoneal cavity, thereby creating a vicious cycle. In cases with extreme hypovolemia renal vasoconstriction develops, leading to HRS (Figure 22-18). This hypothesis is similar to the backward theory of edema formation in heart failure, which suggests that sodium retention and edema formation are secondary to the disruption of the Starling equilibrium in the microcirculation owing to the backward increase in capillary hydrostatic pressure. The classic underfilling theory of ascites formation, however, does not correspond with the systemic hemodynamic abnormalities associated with cirrhosis (see Table 22-6). If this theory were correct, changes in systemic circulation would consist of reduced plasma volume and cardiac index and increased systemic vascular resistance. However, findings in patients with cirrhosis and ascites are exactly the opposite, with increased plasma volume and cardiac index and reduced systemic vascular resistance. These traditional backward theories of edema formation in cirrhosis and heart failure have been substituted by new theories that fit more precisely with the modern concepts of regulation of extracellular fluid volume, which consider that a reduction in effective arterial blood volume (EABV) is the main determinant of sodium
retention in major edematous states. Arterial vasodilatation would be the triggering factor for sodium retention in cirrhosis, whereas a reduction in cardiac output would be the triggering factor in heart failure. The arterial vasodilatation theory considers that the reduction in EABV in cirrhosis with ascites is not due to true hypovolemia, as proposed by the traditional theory, but rather to a disproportionate enlargement of the arterial tree secondary to arterial vasodilatation (Figure 22-19).96,185 According to this theory, portal hypertension is the initial event, with resultant splanchnic arteriolar vasodilatation causing underfilling of the arterial circulation. The arterial receptors then sense the arterial underfilling and stimulate the SNS and the RAAS, causing non-osmotic hypersecretion of AVP. Renal sodium and water retention is the final consequence of this compensatory response to a reduction in EABV. In the early stages of cirrhosis, when splanchnic arteriolar vasodilatation is moderate and the lymphatic system is able to return the increased lymph production to the systemic circulation, the EABV is preserved by transient periods of sodium retention. The fluid retained by the kidneys increases plasma volume and suppresses the signals stimulating the antinatriuretic systems, and sodium retention terminates. Therefore, no ascites or edema is formed at this stage and the relationship between EABV and extracellular fluid ECF volume is maintained. As liver disease progresses, splanchnic arterial vasodilatation increases, resulting in a more intense arterial underfilling and more marked sodium and water retention. At this time the EABV can no longer be maintained by the increased plasma volume, probably because the retained fluid leaks from the splanchnic circulation into the peritoneal cavity as ascites and/or from the systemic circulation to the interstitial tissue as edema. A persistent stimulation of vasoconstrictor systems occurs in an attempt to maintain EABV. The activation of these systems perpetuates renal sodium and water retention, which accumulates as ascites. The correlation between EABV and extracellular fluid volume is no longer maintained, as
Figure 22-18. Pathogenesis of functional renal abnormalities and ascites formation in cirrhosis according to the classic underfilling hypothesis.
Portal hypertension Lymph formation > lymph removal Ascites formation Leak of fluid into the abdominal cavity
Loss of intravascular fluid Hypovolemia Low and high pressure baroreceptors Activation of sodium and water retaining systems
Sodium and water retention
Renal vasoconstriction
Functional renal failure
439
Section III. Clinical Consequences of Liver Disease
Figure 22-19. Pathogenesis of functional renal abnormalities and ascites formation in cirrhosis according to the arterial vasodilatation hypothesis.
Sinusoidal portal hypertension
Splanchnic arterial vasodilation
Effective arterial blood volume
High pressure baroreceptor mediated activation of renin angiotensin and sympathetic nervous system and arginine vasopressin
Sodium and water retention Adequate to normalize circulatory homeostasis
Inadequate to normalize circulatory homeostasis
Increase in plasma volume
Normalization of the activity of sodium and water retaining systems
Persistent activation of sodium and water retaining systems
Normal sodium and water excretion
Continuous sodium and water retenion
Renal vasoconstriction
No ascites
Ascites formation
Hepatorenal syndrome
EABV remains contracted despite progressive expansion of extracellular fluid volume. HRS probably represents the most extreme manifestation of the reduction in EABV. Studies in experimental models of portal hypertension aimed at investigating carefully the chronological relationship between abnormalities in the systemic circulation and sodium retention indicate that arterial vasodilatation with reduced systemic vascular resistance precedes sodium retention and subsequent plasma volume expansion.186 The arterial vasodilatation theory not only provides a reasonable explanation for the circulatory changes and activation of antinatriuretic systems observed in cirrhosis with ascites, but also for the preferential location of retained fluid in the peritoneal cavity. The existence of splanchnic arterial vasodilatation causes a ‘forward’ increase in splanchnic capillary pressure that enhances the effects of portal hypertension on the filtration coefficient in splanchnic capillaries, which facilitates the formation of ascites.
THE CARDIOCIRCULATORY THEORY If circulatory dysfunction in cirrhosis were solely due to the progression of splanchnic arterial vasodilatation and the hyperdynamic circulation a compensatory mechanism of this disorder, cardiac output should increase with the progression of the disease, as occurs with other homeostatic mechanism of effective arterial blood
440
volume, such as the overactivity of the renin–angiotensin and sympathetic nervous systems. However, this is not the case. Despite the progressive increase in the plasma levels of renin and norepinephrine during the course of cirrhosis, indicating an accentuation of arterial vasodilatation, cardiac output is similar in patients with compensated cirrhosis, non-azotemic patients with cirrhosis, patients with ascites and normal or increased plasma levels of renin and norepinephrine, and patients with type 2 HRS. The heart rate also does not increase despite the progressive stimulation of the sympathetic nervous system. This feature suggests that circulatory dysfunction in cirrhosis is related not only to a progression of arterial vasodilatation but also to an inability of the heart to increase the cardiac output in response to a decrease in cardiac preload. The recent demonstration in non-azotemic patients with cirrhosis and spontaneous bacterial peritonitis that the development of type 1 HRS occurs in the setting of a significant decrease in cardiac output further supports the suggestion that cardiac dysfunction is an important event in the pathogenesis of the impairment in circulatory and renal function seen in decompensated cirrhosis (Figure 22-20). The pathogenesis of cardiac dysfunction in cirrhosis is probably multifactorial. A decreased cardiac preload could be an important mechanism. In fact, circulatory function and the kidneys’ ability to excrete sodium and free water is normalized in non-azotemic cir-
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
Inability of the heart to compensate the decreased cardiac preload by increasing cardiac output
Progression of systemic arterial vasodilatation
Figure 22-20. Pathogenesis of functional renal abnormalities and ascites formation in cirrhosis according to the cardiocirculatory theory.
Effective arterial hypovolemia
Arterial hypotension
Activation of renin– angiotensin–system
Activation of sympathetic nervous system
Activation of arginine vasopressin
Sodium retention Ascites
Water retention
Hyponatremia
Renal vasoconstriction
Hepatorenal syndrome
rhosis with ascites by expansion of the plasma volume with saline and noradrenaline, but not by the administration of noradrenaline alone. Similarly, HRS can be reversed using IV albumin and terlipressin or an a-adrenergic agonist, but not with terlipressin alone.98,184,187,188 Noradrenaline and terlipressin improve circulatory and renal function by inducing vasoconstriction in splanchnic arterial circulation, and plasma volume expansion probably helps by increasing venous return and the cardiac output. An impaired cardiac chronotropic function is another significant mechanism, as heart rate does not increase during the course of the disease in patients with decompensated cirrhosis despite a progressive decrease in arterial pressure and an increase in the activity of the sympathetic nervous system. Finally, in cirrhosis there are echocardiographic and left ventricular function abnormalities compatible with the existence of a cirrhotic cardiomyopathy.189
MANAGEMENT OF RENAL DYSFUNCTION OF CIRRHOSIS The specific treatment of ascites (paracentesis, peritoneovenous shunt, and TIPS for refractory ascites, indication for liver transplantation in the patient with ascites) is covered by other chapters in this book. In this chapter only the specific therapeutic measures used to reverse the renal function abnormalities in cirrhosis are reviewed.
SODIUM RETENTION The assumption of an upright posture associated with moderate physical exercise in patients with cirrhosis and ascites induces an
intense stimulation of the renin–aldosterone and sympathetic nervous systems.190,191 Therefore, although there is no specific study, from a theoretical point of view bed rest may be useful for the management of sodium retention in patients with a poor response to diuretics. Because the natriuretic effect of furosemide starts soon after its administration and disappears in approximately 2–3 hours, bed rest should be adjusted to this feature.192 The effect of spironolactone lasts for more than 1 day and therefore is not important in planning bed rest. Mobilization of ascites occurs when a negative sodium balance is achieved. In 10% of patients, those with normal plasma aldosterone and norepinephrine concentration and relatively high urinary sodium excretion, this can be obtained simply by reducing the sodium intake to 60–90 mEq/day.149 A greater reduction in sodium intake interferes with nutrition and is not advisable. In the majority of cases, however, urinary sodium excretion is very low and a negative sodium balance cannot be achieved without diuretics.193 Even in these cases, sodium restriction is important because it reduces diuretic requirements.194,195 Nevertheless, in the face of the frequent dilemma ‘decrease sodium intake or increase diuretic dosage,’ it is better to increase diuretic dosage if the patient responds satisfactorily to these drugs. Sodium restriction is essential in patients responding poorly to diuretics. A frequent cause of ‘apparently’ refractory ascites is inadequate sodium restriction. This should be suspected when ascites does not decrease despite a good natriuretic response to diuretics.196 Furosemide and spironolactone are the diuretics more commonly used in the treatment of ascites in cirrhosis. Furosemide inhibits chloride and sodium reabsorption in the thick ascending limb of the loop of Henle but has no effect on the distal nephron (distal and
441
Section III. Clinical Consequences of Liver Disease
collecting tubules).197,198 It is rapidly absorbed from the gut, is highly bound to plasma proteins, and is actively secreted from the blood into the urine by the proximal tubular cells. Once in the luminal compartment, furosemide is carried out with the luminal fluid to the loop of Henle, where it inhibits the Na+ 2Cl- K+ co-transport system located in the luminal membrane. Because between 30 and 50% of the filtered sodium is reabsorbed in the loop of Henle using this transport system, furosemide has a high natriuretic potency. At high dosage it may increase sodium excretion up to 30% of the filtered sodium load in normal subjects. Furosemide also increases the synthesis of prostaglandin E2 by the ascending limb cells, and this effect is also related to its natriuretic effect, as prostaglandins inhibit sodium reabsorption in the loop of Henle and NSAIDs impair the diuretic and natriuretic effect of furosemide.163,199 The onset of the action of furosemide is very rapid (within 30 minutes after oral administration), with peak effect occurring within 1–2 hours; the diuretic effect finishes 3–4 hours after administration. Spironolactone undergoes extensive metabolism leading to numerous biologically active compounds, the most important quantitatively being canrenone.200,201 These aldosterone metabolites are bound to plasma proteins, from which they are released slowly to the kidney and other organs. In the kidney, spironolactone acts by competitively inhibiting the tubular effect of aldosterone in the distal nephron. Aldosterone interacts with a cytosolic receptor, is translocated into the nuclei, and stimulates the synthesis of sodium channels, which are inserted into the luminal membrane, and the transporter Na-K-ATPase, which activates the extrusion of sodium from the intracellular space into the peritubular interstitial space.202 This transporter and the activation of potassium channels in the luminal membrane are the mechanisms for the kaliuretic effect of aldosterone. Spironolactone and its metabolites enter the basolateral membrane in the collecting tubule and interact with the cytosolic receptor of aldosterone, but the complex spironolactone receptor, unlike the aldosterone receptor, is unable to interact with the DNA.203–205 The half-life of the aldosterone-induced proteins and of spironolactone metabolites is prolonged, there being a delay of 2–3 days between the onset or the discontinuation of treatment with spironolactone and the onset or the end of the natriuretic effect, respectively. The clearance of spironolactone metabolites is reduced in cirrhosis,200 so the natriuretic effect of spironolactone after discontinuation of the drug persists for a longer time in cirrhotic patients than in normal subjects. Because the amount of sodium reabsorption in the collecting tubule is relatively low (approximately 5% of the filtered sodium), the intrinsic natriuretic potency of spironolactone is lower than that of furosemide. In contrast to what occurs in healthy subjects, in whom furosemide is more potent than spironolactone, in cirrhotic patients with ascites spironolactone is more effective than furosemide. This was demonstrated in a randomized controlled trial 20 years ago.18 Cirrhotic patients with ascites and marked hyperaldosteronism (50% of the patients with ascites) do not respond to furosemide or other loop diuretics.206 In contrast, most cirrhotic patients with ascites respond to spironolactone. Patients with normal or slightly increased plasma aldosterone concentration respond to low doses of spironolactone (100–150 mg/day), but as much as 300–400 mg/day may be required in patients with marked hyperaldosteronism. Two mechanisms account for the resistance to furosemide in patients
442
with ascites and marked hyperaldosteronism:207 first, an increased proximal sodium reabsorption leading to a low sodium delivery to the ascending limb of the loop of Henle;208,209 and second, most of the sodium not reabsorbed in the loop of Henle owing to the action of furosemide is subsequently reabsorbed in the distal nephron by the effect of aldosterone.14 Therefore, spironolactone is the basic drug for the management of patients with cirrhosis and ascites.18 The simultaneous administration of furosemide and spironolactone increases the natriuretic effect of both agents and reduces the incidence of hypo- or hyperkalemia that may be observed when these drugs are given alone. There are two different approaches to the use of diuretics in cirrhotic patients with sodium retention and ascites. The ‘stepped care’ approach consists of the progressive implementation of the therapeutic measures currently available, starting with sodium restriction; if ascites does not decrease, spironolactone is given at increasing doses (100 mg/day as initial dose; if there is no response within 4 days, 200 mg/day; if no response 400 mg/day). When there is no response to the highest dose of spironolactone, furosemide is added at increasing doses every 2 days (40 mg/day to 160 mg/day).17,18,30,149,210 The second approach is the combined treatment, which is particularly indicated in patients with tense ascites and avid sodium retention. It begins with sodium restriction and the simultaneous administration of spironolactone 100 mg/day and furosemide 40 mg/day. If the diuretic response is insufficient after 4 days, the dose of furosemide and spironolactone is increased up to 160 mg/day and 400 mg/day, respectively.211 There is general agreement that patients not responding to these doses will not respond to higher diuretic dosage. In those receiving the combined treatment with an exaggerated response, diuretics should be adjusted by reducing the dose of furosemide. The goal of diuretic treatment should be to achieve a negative sodium balance with a weight loss of 0.3–0.5 kg/day in patients without edema and 0.5–1.0 kg/day in patients with peripheral edema.212 Diuretic treatment in cirrhosis is not free of complications, particularly in patients requiring high dosages. Approximately 20% of patients develop significant renal impairment (increase in blood urea and serum creatinine concentration), which is usually moderate and always reversible after diuretic withdrawal.213,214 The development of renal insufficiency is due to an imbalance between the fluid loss induced by the diuretic treatment and the rate of reabsorption of ascites, which varies greatly from patient to patient. Patients with ascites and peripheral edema less frequently develop diureticinduced renal failure because there is no limitation to the reabsorption of peripheral edema, and it therefore compensates for the slow rate of reabsorption of ascites. Hyponatremia secondary to a decrease in renal ability to excrete free water also occurs in approximately 20% of these patients.213 This is related to a reduction in intravascular volume leading to an increased secretion of antidiuretic hormone, and also to a reduction in the generation of free water in the loop of Henle by the effect of furosemide. Free water is formed within the kidney by the active reabsorption of sodium chloride in the water-impermeable loop of Henle, and this process is inhibited by the loop diuretics. The most severe complication related to diuretic treatment is hepatic encephalopathy, which occurs in approximately 25% of patients admitted to hospital with tense ascites requiring high diuretic doses.213 This complication is also
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
related to an impairment in circulating blood volume which increases renal production and decreases renal clearance of ammonia.215 Other complications include hyperkalemia or metabolic acidosis in patients with hepatorenal syndrome treated with high doses of spironolactone,216 hypokalemia in patients treated with high doses of furosemide and no or low doses of spironolactone,217 gynecomastia in patients receiving spironolactone, and muscle cramps. Gynecomastia is related to the antiandrogenic activity of most spironolactone metabolites.218,219 Canrenone, which apparently has lower antiandrogenic activity than spironolactone, is available for clinical use in some countries. Muscle cramps are clearly related to a reduction in intravascular volume, as they occur when there is marked activation of the renin–angiotensin system and may be prevented by plasma volume expansion with albumin.220 The oral administration of quinidine also reduces the frequency of diureticinduced muscle cramps.221
WATER RETENTION AND DILUTIONAL HYPONATREMIA Aquaretic drugs are agents that interfere with the renal effects of antidiuretic hormone, inhibit water reabsorption in the collecting tubules, and produce hypotonic polyuria without affecting solute excretion. These drugs are the ideal treatment for water retention and dilutional hyponatremia in cirrhosis and in other conditions associated with increased circulating levels of antidiuretic hormone, such as congestive heart failure and the syndrome of inappropriate ADH secretion.43 The hydroosmotic effect of antidiuretic hormone is mediated by the insertion of water channels (aquaporin 2) into the luminal membrane of the collecting tubular epithelial cells.222 In the unstimulated state this membrane is impermeable to water owing to the lack of water channels. In contrast, the basocellular membrane, which is very rich in aquaporin 3, is highly permeable to water. The hydroosmotic effect of antidiuretic hormone is initiated by the binding of the hormone to a V2 receptor on the basolateral membrane of the collecting duct epithelial cells.223 This receptor is coupled to adenylate cyclase, the stimulation of which releases cyclic adenosine monophosphate (cAMP) from adenosine triphosphate (ATP).224,225 cAMP activates a protein kinase that promotes the insertion of aquaporin 2 molecules into the luminal membrane. Water is then reabsorbed passively from the hypotonic tubular lumen to the hypertonic medullary interstitium.226 The vasoconstrictor effect of antidiuretic hormone227 depends on the interaction of the hormone with a different receptor, the V1 receptor, in the vascular smooth muscle cells, which leads to an increase in the cytosolic calcium concentration.43 Many aquaretic drugs were identified prior to 1990. Demeclocycline reduces the renal effects of antidiuretic hormone in humans by inhibiting adenylate cyclase.228 k Opioid agonists produce hypotonic polyuria in humans and experimental animals by inhibiting the release of antidiuretic hormone by the neurohypophysis.229–232 Finally, several selective peptide V2 antagonists were developed during the 1970s by modifying the desmopressin molecule.233,234 However, none of these substance could be used in cirrhosis. Demeclocycline induces renal failure in decompensated cirrhosis with ascites.235 The k opioid agonists have the risk of inducing hepatic encephalopathy.236 Finally, the peptide V2 antagonists were
aquaretic in rats and dogs but had agonistic antidiuretic hormone activity in humans.237 The field of the modern aquaretic drugs started in 1991, when an orally active non-peptide V1 antagonist was discovered using functional screening strategies.238 One year later, via a series of structural conversions of this molecule, the first selective non-peptide V2 receptor antagonist (OPC-31260) was obtained,239 and this became the basis for the synthesis of other V2 antagonists (VPA-985,240,241 SR-121463,242 OPC-41061,243 YM-087244). Several studies in animals with experimental cirrhosis and phase 1 and 2 studies in patients with cirrhosis and ascites have demonstrated that these agents are extremely effective in increasing free water clearance and normalizing serum sodium concentration in cirrhotic patients with ascites and dilutional hyponatremia.45,53,243,245–247 The increase in urine volume is dose dependent, and when appropriate doses are given serum sodium concentration is normalized within a few days after the onset of treatment. Many aspects concerning the use of aquaretic drugs in cirrhosis with ascites need to be investigated. The aquaretic agents have clear interactions with the natriuretic agents (i.e. diuretics) given to these patients, and this needs to be investigated. Also, it is essential to know whether these drugs affect sodium excretion in patients with cirrhosis and ascites. Although in normal conditions aquaretic drugs increase urine volume without affecting sodium excretion, this has not been the case in experimental animals and in patients with cirrhosis and ascites, in which, in addition to hypotonic polyuria, these agents increased urinary sodium excretion. Finally, indications for the use of aquaretic drugs for conditions other than spontaneous dilutional hyponatremia (i.e. diuretic-induced hyponatremia, hyponatremia prior to liver transplantation) should be investigated.
TREATMENT OF HRS Many vasoactive drugs (dopamine, fenoldopan, prostaglandins, misoprostol, saralasin, phentolamine, dazoxiben, norepinephrine, metaraminol, octapressin) have been used in patients with HRS, either to improve systemic hemodynamics or to vasodilate the intrarenal circulation. In no case, however, did renal function improve, leading to a general impression that HRS was an intractable terminal event of cirrhosis. It is important to remark that in these studies drugs were given for a few hours or days, and we now know that this is insufficient to reverse HRS. The intractability of HRS was reinforced after the demonstration that the LeVeen shunt, a prosthesis allowing the continuous passage of ascites to the circulation, also failed to improve renal function in patients with HRS despite a significant suppression of the renin–angiotensin and sympathetic nervous systems. The concept that the poor prognosis associated with HRS was due to liver failure rather than to renal failure, and that any improvement in renal function would have little impact on survival, was an additional feature supporting the theory that the only possible therapy for patients with HRS was liver transplantation. A better understanding of the pathogenesis of HRS, combined with the recent observation that its reversal requires a sustained improvement in circulatory function during a relatively long period (1–2 weeks) and, finally, the demonstration that renal failure is an important determinant of the poor prognosis of patients with HRS, has increased interest in the treatment of this complication of portal hypertension.248
443
Section III. Clinical Consequences of Liver Disease
VOLUME EXPANSION AND VASOCONSTRICTORS The first study showing that HRS can be reversed pharmacologically was performed by Guevara et al. using ornipressin.97 Eight patients with type 1 HRS were treated for 3 days (albumin was given at a dose of 1 g/kg on the first day and 20–60 g/day for the next 2 days; ornipressin was given as an intravenous stepped dose infusion of 2–6 IU/h). A normalization of the plasma levels of renin and norepinephrine was obtained, indicating a marked improvement in circulatory function. However, only a slight increase in GFR (from 15 ± 4 ml/min to 24 ± 4 ml/min; normal values over 100 ml/min) was observed. The remaining eight patients were treated for 15 days. Ornipressin was given at a dose of 2 IU/h. Albumin was given at a dose of 1 g/kg during the first day. The amount of albumin given during the following days was adjusted according to values of plasma renin activity. In four patients treatment was stopped between 4 and 9 days after initiation of therapy because of ischemic complications in three cases and bacteremia in one. In these four patients a marked decrease in serum creatinine during therapy and a progressive decline in renal function after treatment withdrawal were observed. In the remaining four patients who completed treatment, there was a significant elevation in mean arterial pressure, a normalization of plasma renin activity, a marked decrease in plasma norepinephrine concentration, an increase in GFR, and a normalization of serum creatinine concentration. These four patients died 12, 60, 62, and 133 days after treatment; HRS did nor recur in any of them during follow-up. In a subsequent study, the same group treated nine patients with HRS (six with type 1 and three with type 2 HRS) with terlipressin (0.5–2 mg/4 h IV) and IV albumin for 5–15 days.98 Reversal of HRS (normalization of serum creatinine) was observed in seven patients. No patient developed ischemic complications. HRS did not recur in any patient. Five cases were transplant candidates and three were transplanted 5, 12, and 99 days after treatment. The two other patients died 30 and 121 days after the completion of the study. The remaining four patients died 13–102 days after treatment. In both studies dilutional hyponatremia was corrected with the normalization of serum creatinine. These observations have been confirmed by other groups (see Table 22-2). Gülberg et al. treated seven patients with type 1 HRS using ornipressin (6 IU/h), dopamine (2–3 mg/kg/min) and IV albumin.249 HRS was reversed in four patients after 5–27 days of treatment. In one patient treatment had to be stopped because of intestinal ischemia. The remaining two patients did not respond. In two of the four patients responding to treatment, HRS recurred 2 and 8 months later and they were retreated, and the HRS was reversed in one patient. In the other, treatment had to be stopped because of ventricular tachyarrhythmia. In total, two patients reached liver transplantation and one was alive 1 year after two successful treatments. Mulkay et al.250 treated 12 patients with type 1 HRS with terlipressin (2 mg every 8–12 h) and albumin infusion (0.5–1 g/kg/day for 5 days) for 1–9 weeks. HRS was reversed in seven patients. In the remaining five cases serum creatinine also decreased, but did not reach normal levels. Withdrawal of terlipressin without recurrence of HRS was observed in six patients. No patient developed complications related to treatment. Three patients were transplanted 34, 36, and 111 days after inclusion. The remaining patients died after a median survival of 42 days. Finally, Moreau et al.,251 in a retrospective study, collected 99
444
patients with type 1 HRS from 24 centers who had been treated with terlipressin. Most of them also received albumin. Improvement in renal function was obtained in 58% of cases. The probability of survival was markedly improved in patients who responded to therapy. Catecholamines are also effective for the treatment of HRS. Angeli et al. used oral midodrine, an a-adrenergic agonist, IV albumin and subcutaneous octreotride (to suppress glucagon) in five patients with type 1 HRS.187 Midodrine dosage was adjusted to increase mean arterial pressure 15 mmHg or more. Patients received treatment for at least 20 days in hospital and subsequently at home. In all cases there was an improvement in renal perfusion, GFR, blood urea nitrogen, serum creatinine, and serum sodium concentration and a suppression of plasma renin, aldosterone and antidiuretic hormone to normal or near normal levels. Two patients were transplanted 20 and 64 days after inclusion while on therapy. One patient who was not a candidate for transplantation was alive without treatment 472 days after discharge from hospital. The remaining two patients died 29 and 75 days after inclusion in the study. These results were compared with those obtained in eight patients with type 1 HRS treated with IV albumin plus dopamine (2–4 mg/kg/min). In all eight patients a progressive worsening in renal function was observed. One patient was transplanted but died 15 days later because of a fungal infection. The remaining seven patients died within 2 weeks after the initiation of treatment. Duvoux et al.184 treated 12 patients with type 1 HRS with intravenous albumin (to maintain central venous pressure >7 mmHg) and noradrenaline (0.5–3 mg/h) for a minimum of 5 days. A significant improvement in serum creatinine in association with a marked suppression of plasma renin activity was observed in 10 patients. Transient myocardial ischemia was observed in one patient. Three patients were transplanted and three were still alive after 8 months of follow-up Finally, Ortega et al. have recently assessed whether albumin is necessary in the treatment of HRS with vasoconstrictors.188 Twentyone patients with HRS were studied. The first 13 were treated with terlipressin (0.5–2 g/4 h) and albumin (1 g/kg the first day; 20–40 g/day thereafter). The last eight patients received terlipressin alone. Treatment was given until normalization of serum creatinine or for a maximum of 15 days. In patients treated with terlipressin plus albumin there was a significant increase in mean arterial pressure, a marked suppression of plasma renin activity, and a decrease in serum creatinine. In contrast, no significant changes in these parameters were observed in patients treated with terlipressin alone. A complete response (normalization of serum creatinine) was achieved in 10 patients treated with terlipressin plus albumin, and in only two treated without albumin. Recurrence of HRS only occurred in two patients. One-month survival without transplantation was 87% in patients receiving terlipressin plus albumin and 13% in patients receiving terlipressin alone. These studies show that: (1) type 1 HRS is reversible following treatment with IV albumin and vasoconstrictors; (2) the two components of the treatment are important, as HRS does not reverse when vasoconstrictors or plasma volume expansion are given alone; (3) the constant infusion of vasoconstrictors (ornipressin or noradrenaline) is associated with ischemic complications, a feature not observed when they are given intermittently (terlipressin); (4) there is a delay of
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
several days between the improvement in circulatory function and the increase in GFR; (5) reversal of HRS improves survival and a significant number of patients may reach liver transplantation. There are few data on the effectiveness of vasoconstrictors in type 2 HRS. Alessandria et al.252 treated 11 patients with type 2 HRS with terlipressin and albumin. Normalization of serum creatinine was observed in eight cases. However, in all of them HRS recurred after discontinuation of therapy. It appears as if treatment with volume expansion and vasoconstrictors is effective only in type 1 HRS. Table 22-7 shows the results of several studies with vasocontrictors agents in type 1 and type 2 HRS.
TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT (TIPS) Because portal hypertension is the initial event of circulatory dysfunction in cirrhosis, the reduction of portal pressure by portocaval anastomosis is a rational approach for the treatment of HRS. There are several case reports showing reversal of HRS following surgical portocaval shunts. However, the applicability of major surgical procedures in patients with HRS is limited. The development of TIPS has reintroduced the idea of treating HRS by reducing portal pressure. Four studies assessing TIPS in the management of type 1 HRS have been reported109,253,253a,253b and recently reviewed by Brensing et al.253 In total, 30 patients were treated. In two series no liver transplantation was performed, whereas in the other two series three out of nine patients were transplanted 7, 13, and 35 days after TIPS. TIPS insertion was technically successful in all cases. Only one patient died as a consequence of the procedure. GFR improved markedly within 1–4 weeks after TIPS, and stabilized thereafter. In one study specifically investigating the neurohormonal systems, improvement in GFR and serum creatinine was related to a marked suppression of the plasma levels of renin activity and antidiuretic hormone. The suppression of plasma norepinephrine is less than that of renin, a feature also observed in refractory ascites treated by TIPS. Follow-up data concerning hepatic function were obtained from 21 patients. De novo hepatic encephalopathy or deterioration
of pre-existing hepatic encephalopathy occurred in nine patients, but in five it could be controlled with lactulose. Survival rates based on the 27 patients without early liver transplantation at 1 month, 3 months, and 6 months were 81%, 59%, and 44%, respectively. These studies strongly suggest that TIPS is useful in the management of type 1 HRS and improves survival.
SEQUENTIAL TREATMENT WITH VASOCONSTRICTORS AND TIPS One of the intriguing issues on the treatment of type 1 HRS with vasoconstrictors is the observation that, despite marked suppression of the renin–angiotensin and sympathetic nervous systems and normalization of serum creatinine, renal function does not reach normal levels in the majority of patients, and there is persistence of low GFR, which ranges between 30 and 50 ml/min in most cases (normal: 120 ml/min). The reason for this is unknown, but it could be due either to the existence of a component of renal failure unresponsive to changes in circulatory function or to the fact that the effective arterial blood volume, although improved, is not normalized with pharmacologic therapy. A recent study by Wong et al.255 is consistent with this latter hypothesis. Treatment with TIPS in patients responding to pharmacological treatment (midodrine, octreotide, and albumin) was associated with a normalization of GFR in most cases. Whether the effect of TIPS in the normalizing the GFR was due to the correction of the arterial vasodilatation, to an increase in cardiac preload and ventricular function, or both, remains to be investigated.
OTHER TREATMENTS Hemodialysis and arteriovenous or venovenous hemofiltration are frequently used in patients with HRS, but their efficacy has not been adequately assessed. Recently, extracorporeal albumin dialysis, a system that uses an albumin-containing dialysate that is recirculated and perfused through a charcoal and anion exchanger column, has been shown to improve systemic hemodynamics and reduce the plasma levels of renin in patients with type 1 HRS. In a small series of patients improved survival has been reported. Further studies are needed to confirm these findings.248
Table 22-7. Results of Several Studies with Vasoconstrictor Agents in Types 1 and 2 HRS: Response to Treatment and Outcome Patients
Study Treatment
(n)
Reversal of HRS
Patients surviving > 1 month
Patients undergoing OLT
Guevara et al.97 Gulberg et al.249 Uriz et al.92 Mulkay et al.250 Angeli et al.187 Duvoux et al.184 Ortega et al.88 Moreau et al.250 Alessandria et al.251 Wong et al.255
Ornipressin plus albumin Ornipressin plus albumin Terlipressin plus albumin Terlipressin plus albumin Midodrine plus octreotide and albumin Noradrenaline plus albumin Terlipressin with or without albumin Terlipressin with or without albumin Terlipressin Midodrine plus octreotide and albumin Followed by TIPS Without TIPS
8 7 9 12 5 12 13 99 11 14
4 4 7 7 4 10 10 58 8 10 5 5
5 4 5 4 4 6 9 36 8
— 2 3 2 2 2 5 13 NR
4 2
1 2
445
Section III. Clinical Consequences of Liver Disease
PREVENTION OF HEPATORENAL SYNDROME Two randomized controlled studies in large series of patients have shown that HRS can be prevented in specific clinical settings. In the first study,78 the administration of albumin (1.5 g/kg IV at infection diagnosis and 1 g/kg IV 48 hours later) together with cefotaxime to patients with cirrhosis and spontaneous bacterial peritonitis markedly reduced the incidence of impairment in circulatory function and the occurrence of type 1 HRS, compared to a control group of patients receiving cefotaxime alone (10% incidence of HRS in patients receiving albumin versus 33% in the control group). Moreover, the hospital mortality rate (10% vs 29%) and the 3-month mortality rate (22% vs 41%) were lower in patients receiving albumin. In a second study,256 the administration of the TNF inhibitor pentoxifylline (400 mg t.i.d.) to patients with severe acute alcoholic hepatitis reduced the occurrence of HRS (8% in the pentoxifylline group vs 35% in the placebo group) and hospital mortality (24% vs 46%, respectively). Because bacterial infections and acute alcoholic hepatitis are two important precipitating factors of type 1 HRS, these prophylactic measures may decrease the incidence of this complication.
REFERENCES 1. Arroyo V, Guevara M, Gines P. Hepatorenal syndrome in cirrhosis: pathogenesis and treatment. Gastroenterology 2002;122:1658–1676. 2. Papper S. Hepatorenal syndrome. In: Epstein M, ed. The Kidney in Liver Disease. New York: Elsevier Biomedical, 1978: 91–112. 3. Shear L, Kleinerman J, Gabuzda GJ. Renal failure in patients with cirrhosis of the liver. I. Clinical and pathologic characteristics. Am J Med 1965;39:184–198. 4. Ellis AJ, O’Grady JG. Clinical disorders or renal function in acute liver failure. In: Arroyo V, Gines P, Rodes J, Schrier RW, eds. Ascites and renal dysfunction in liver disease. Oxford: Blackwell Science, 1999:63–78. 5. Stehman-Breen C, Johnson RJ. Hepatitis C virus-associated glomerulonephritis. Adv Intern Med 1998;43:79–97. 6. Post J, Sicam L. The clinical course of Laennec’s cirrhosis under modern medical management. Med Clin North Am 1960;44:639–651. 7. Pecikyan R, Kanzaki G, Berger EY. Electrolyte excretion during the spontaneous recovery from the ascitic phase of cirrhosis of the liver. Am J Med 1967;42:359–367. 8. Eisenmenger WJ, Blondheim SH, Bongiovanni AM, et al. Electrolyte studies on patients with cirrhosis of the liver. J Clin Invest 1950;29:1491–1499. 9. Arroyo V, Rodes J. A rational approach to the treatment of ascites. Postgrad Med J 1975;51:558–562. 10. Levy M, Allotey JB. Temporal relationships between urinary salt retention and altered systemic hemodynamics in dogs with experimental cirrhosis. J Lab Clin Med 1978;92:560–569. 11. Jimenez W, Martinez-Pardo A, Arroyo V, et al. Temporal relationship between hyperaldosteronism, sodium retention and ascites formation in rats with experimental cirrhosis. Hepatology 1985;5:245–250. 12. Gines P, Fernandez-Esparrach G, Arroyo V, et al. Pathogenesis of ascites in cirrhosis. Semin Liver Dis 1997;17:175–189.
446
13. Claria J, Jimenez W. Renal dysfunction and ascites in carbon tetrachloride-induced cirrhosis in rats. In: Arroyo V, Gines P, Rodes J, Schrier RW, eds. Ascites and renal dysfunction in liver disease. Oxford: Blackwell Science, 1999:379–396. 14. Wilkinson SP, Jowett TP, Slater JD, et al. Renal sodium retention in cirrhosis: relation to aldosterone and nephron site. Clin Sci (Lond) 1979;56:169–177. 15. Angeli P, Gatta A, Caregaro L, et al. Tubular site of renal sodium retention in ascitic liver cirrhosis evaluated by lithium clearance. Eur J Clin Invest 1990;20:111–117. 16. Eggert RC. Spironolactone diuresis in patients with cirrhosis and ascites. Br Med J 1970;4:401–403. 17. Gatta A, Angeli P, Caregaro L, et al. A pathophysiological interpretation of unresponsiveness to spironolactone in a stepped-care approach to the diuretic treatment of ascites in nonazotemic cirrhotic patients. Hepatology 1991;14: 231–236. 18. Perez-Ayuso RM, Arroyo V, Planas R, et al. Randomized comparative study of efficacy of furosemide versus spironolactone in nonazotemic cirrhosis with ascites. Relationship between the diuretic response and the activity of the renin–aldosterone system. Gastroenterology 1983;84:961–968. 19. Jonassen TE, Heide AM, Janjua NR, et al. Collecting duct function in liver cirrhotic rats with early sodium retention. Acta Physiol Scand 2002;175:237–244. 20. Jonassen TE, Christensen S, Sorensen AM, et al. Effects of chronic octreotide treatment on renal changes during cirrhosis in rats. Hepatology 1999;29:1387–1395. 21. Jonassen TE, Petersen JS, Sorensen AM, et al. Aldosterone receptor blockade inhibits increased furosemide-sensitive sodium reabsorption in rats with liver cirrhosis. J Pharmacol Exp Ther 1998;287:931–936. 22. Ring-Larsen H, Henriksen JH, Wilken C, et al. Diuretic treatment in decompensated cirrhosis and congestive heart failure: effect of posture. Br Med J (Clin Res Ed) 1986;292:1351–1353. 23. Rimola A, Garcia-Tsao G, Navasa M, et al. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. International Ascites Club. J Hepatol 2000;32:142–153. 24. Moore KP, Wong F, Gines P, et al. The management of ascites in cirrhosis: report on the consensus conference of the International Ascites Club. Hepatology 2003;38:258–266. 25. Strauss RM, Boyer TD. Hepatic hydrothorax. Semin Liver Dis 1997;17:227–232. 26. Xiol X, Guardiola J. Hepatic hydrothorax. Curr Opin Pulm Med 1998;4:239–242. 27. Arroyo V, Bosch J, Gaya-Beltran J, et al. Plasma renin activity and urinary sodium excretion as prognostic indicators in nonazotemic cirrhosis with ascites. Ann Intern Med 1981;94:198–201. 28. Llach J, Gines P, Arroyo V, et al. Prognostic value of arterial pressure, endogenous vasoactive systems, and renal function in cirrhotic patients admitted to the hospital for the treatment of ascites. Gastroenterology 1988;94:482–487. 29. Spahr L, Villeneuve JP, Tran HK, et al. Furosemide-induced natriuresis as a test to identify cirrhotic patients with refractory ascites. Hepatology 2001;33:28–31. 30. Bernardi M, Di Marco C, Trevisani F, et al. Renal sodium retention during upright posture in preascitic cirrhosis. Gastroenterology 1993;105:188–193. 31. Bernardi M, Di Marco C, Trevisani F, et al. The hemodynamic status of preascitic cirrhosis: an evaluation under steady-state conditions and after postural change. Hepatology 1992;16:341–346. 32. Wong F, Liu P, Blendis L. Sodium homeostasis with chronic sodium loading in preascitic cirrhosis. Gut 2001;49:847–851.
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
33. Wong F, Liu P, Allidina Y, et al. Pattern of sodium handling and its consequences in patients with preascitic cirrhosis. Gastroenterology 1995;108:1820–1827. 34. Albillos A, Lledo JL, Rossi I, et al. Continuous prazosin administration in cirrhotic patients: effects on portal hemodynamics and on liver and renal function. Gastroenterology 1995;109:1257–1265. 35. Henriksen JH, Ring-Larsen H. Renal effects of drugs used in the treatment of portal hypertension. Hepatology 1993;18:688–695. 36. Wong F, Massie D, Hsu P, et al. Indomethacin-induced renal dysfunction in patients with well-compensated cirrhosis. Gastroenterology 1993;104:869–876. 37. Denison EK, Lieberman FL, Reynolds TB. 9-Fluorohydrocortisone induced ascites in alcoholic liver disease. Gastroenterology 1971;61:497–503. 38. La Villa G, Salmeron JM, Arroyo V, et al. Mineralocorticoid escape in patients with compensated cirrhosis and portal hypertension. Gastroenterology 1992;102:2114–2119. 39. Wong F, Massie D, Hsu P, et al. Renal response to a saline load in well-compensated alcoholic cirrhosis. Hepatology 1994;20:873–881. 40. Papper S, Saxon L. The diuretic response to administered water in patients with liver disease. II. Laennec’s cirrhosis of the liver. AMA Arch Intern Med 1959;103:750–757. 41. Shear L, Hall PW III, Gabuzda GJ. Renal failure in patients with cirrhosis of the liver. II. Factors influencing maximal urinary flow rate. Am J Med 1965;39:199–209. 42. Fernandez-Esparrach G, Sanchez-Fueyo A, Gines P, et al. A prognostic model for predicting survival in cirrhosis with ascites. J Hepatol 2001;34:46–52. 43. Gines P, Berl T, Bernardi M, et al. Hyponatremia in cirrhosis: from pathogenesis to treatment. Hepatology 1998;28:851–864. 44. Gines P, Abraham WT, Schrier RW. Vasopressin in pathophysiological states. Semin Nephrol 1994;14:384–397. 45. Jimenez W, Gal CS, Ros J, et al. Long-term aquaretic efficacy of a selective nonpeptide V(2)-vasopressin receptor antagonist, SR121463, in cirrhotic rats. J Pharmacol Exp Ther 2000;295:83–90. 46. Bosch-Marce M, Poo JL, Jimenez W, et al. Comparison of two aquaretic drugs (niravoline and OPC-31260) in cirrhotic rats with ascites and water retention. J Pharmacol Exp Ther 1999;289:194–201. 47. Fernandez-Varo G, Ros J, Cejudo-Martin P, et al. Effect of the V1a/V2-AVP receptor antagonist, Conivaptan, on renal water metabolism and systemic hemodynamics in rats with cirrhosis and ascites. J Hepatol 2003;38:755–761. 48. Gerbes AL, Gulberg V, Gines P, et al. Therapy of hyponatremia in cirrhosis with a vasopressin receptor antagonist: a randomized double-blind multicenter trial. Gastroenterology 2003;124:933–939. 49. Guyader D, Patat A, Ellis-Grosse EJ, et al. Pharmacodynamic effects of a nonpeptide antidiuretic hormone V2 antagonist in cirrhotic patients with ascites. Hepatology 2002;36:1197–1205. 50. Verbalis JG. Vasopressin V2 receptor antagonists. J Mol Endocrinol 2002;29:1–9. 51. Schrier RW, Cadnapaphornchai MA, Umenishi F. Water-losing and water-retaining states: role of water channels and vasopressin receptor antagonists. Heart Dis 2001;3:210–214. 52. Decaux G. Difference in solute excretion during correction of hyponatremic patients with cirrhosis or syndrome of inappropriate secretion of antidiuretic hormone by oral vasopressin V2 receptor antagonist VPA-985. J Lab Clin Med 2001;138:18–21. 53. Inoue T, Ohnishi A, Matsuo A, et al. Therapeutic and diagnostic potential of a vasopressin-2 antagonist for impaired water handling in cirrhosis. Clin Pharmacol Ther 1998;63:561–570.
54. Rovira A, Cordoba J, Raguer N, et al. Magnetic resonance imaging measurement of brain edema in patients with liver disease: resolution after transplantation. Curr Opin Neurol 2002;15:731–737. 55. Cordoba J, Alonso J, Rovira A, et al. The development of lowgrade cerebral edema in cirrhosis is supported by the evolution of (1)H-magnetic resonance abnormalities after liver transplantation. J Hepatol 2001;35:598–604. 56. Restuccia T, Gomez-Anson B, Guevara M, et al. Effects of dilutional hyponatremia on brain organic osmolytes and water content in patients with cirrhosis. Hepatology 2004;39:1613–1622. 57. Schroeder ET, Shear L, Sancetta SM, et al. Renal failure in patients with cirrhosis of the liver. Evaluation of intrarenal blood flow by para-aminohippurate extraction and response to angiotensin. Am J Med 1967;43:887–896. 58. Hecker R, Sherlock S. Electrolyte and circulatory changes in terminal liver failure. Lancet 1956;271:1121–1125. 59. Epstein M, Berk DP, Hollenberg NK, et al. Renal failure in the patient with cirrhosis. The role of active vasoconstriction. Am J Med 1970;49:175–185. 60. Arroyo V, Bosch J, Mauri M, et al. Renin, aldosterone and renal haemodynamics in cirrhosis with ascites. Eur J Clin Invest 1979;9:69–73. 61. Rimola A, Gines P, Arroyo V, et al. Urinary excretion of 6-keto-prostaglandin F1 alpha, thromboxane B2 and prostaglandin E2 in cirrhosis with ascites. Relationship to functional renal failure (hepatorenal syndrome). J Hepatol 1986;3:111–117. 62. Arroyo V, Gines P, Gerbes AL, et al. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. International Ascites Club. Hepatology 1996;23:164–176. 63. Gines A, Escorsell A, Gines P, et al. Incidence, predictive factors, and prognosis of the hepatorenal syndrome in cirrhosis with ascites. Gastroenterology 1993;105:229–236. 64. Platt JF, Marn CS, Baliga PK, et al. Renal dysfunction in hepatic disease: early identification with renal duplex Doppler US in patients who undergo liver transplantation. Radiology 1992;183:801–806. 65. Bataller R, Gines P, Guevara M, et al. Hepatorenal syndrome. Semin Liver Dis 1997;17:233–247. 66. Moore K. The hepatorenal syndrome. Clin Sci (Lond) 1997;92:433–443. 67. Kew MC, Brunt PW, Varma RR, et al. Renal and intrarenal blood-flow in cirrhosis of the liver. Lancet 1971;2:504– 510. 68. Lancestremere RG, Davidson PL, Earley LE, et al. Renal failure in Laennec’s cirrhosis. II. Simultaneous determination of cardiac output and renal hemodynamics. J Clin Invest 1962;41:1922–1927. 69. Maroto A, Gines P, Arroyo V, et al. Brachial and femoral artery blood flow in cirrhosis: relationship to kidney dysfunction. Hepatology 1993;17:788–793. 70. Iwatsuki S, Popovtzer MM, Corman JL, et al. Recovery from ‘hepatorenal syndrome’ after orthotopic liver transplantation. N Engl J Med 1973;289:1155–1159. 71. Gonwa TA, Morris CA, Goldstein RM, et al. Long-term survival and renal function following liver transplantation in patients with and without hepatorenal syndrome – experience in 300 patients. Transplantation 1991;51:428–430. 72. Rimola A, Gavaler JS, Schade RR, et al. Effects of renal impairment on liver transplantation. Gastroenterology 1987;93:148–156. 73. Fernandez-Seara J, Prieto J, Quiroga J, et al. Systemic and regional hemodynamics in patients with liver cirrhosis and ascites with and without functional renal failure. Gastroenterology 1989;97:1304–1312.
447
Section III. Clinical Consequences of Liver Disease
74. Guevara M, Bru C, Gines P, et al. Increased cerebrovascular resistance in cirrhotic patients with ascites. Hepatology 1998;28:39–44. 75. Ruiz-del-Arbol L, Urman J, Fernandez J, et al. Systemic, renal, and hepatic hemodynamic derangement in cirrhotic patients with spontaneous bacterial peritonitis. Hepatology 2003;38:1210–1218. 76. Follo A, Llovet JM, Navasa M, et al. Renal impairment after spontaneous bacterial peritonitis in cirrhosis: incidence, clinical course, predictive factors and prognosis. Hepatology 1994;20:1495–1501. 77. Navasa M, Follo A, Filella X, et al. Tumor necrosis factor and interleukin-6 in spontaneous bacterial peritonitis in cirrhosis: relationship with the development of renal impairment and mortality. Hepatology 1998;27:1227–1232. 78. Sort P, Navasa M, Arroyo V, et al. Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med 1999;341:403–409. 79. Gines P, Tito L, Arroyo V, et al. Randomized comparative study of therapeutic paracentesis with and without intravenous albumin in cirrhosis Gastroenterology 1988;94:1493–1502. 80. Gines P, Arroyo V, Rodes J. Ascites, hepatorenal syndrome, and spontaneous bacterial peritonitis. In: McDonald KM, Burroughs AK, Feagan B, eds. Evidence based gastroenterology and hepatology. London: BMJ Books, 1999: 427–442. 81. Cardenas A, Gines P, Uriz J, et al. Renal failure after upper gastrointestinal bleeding in cirrhosis: incidence, clinical course, predictive factors, and short-term prognosis. Hepatology 2001;34:671–676. 82. Boyer TD, Zia P, Reynolds TB. Effect of indomethacin and prostaglandin A1 on renal function and plasma renin activity in alcoholic liver disease. Gastroenterology 1979;77:215–222. 83. Salerno F, Badalamenti S. Drug-induced renal failure in cirrhosis. In: Arroyo V, Gines P, Rodes J, Schrier RW, eds. Ascites and renal sysfunction in liver disease. Pathogenesis, diagnosis and treatment. Philadelphia: Blackwell Science Malden, 1999:511–521. 84. Cabrera J, Arroyo V, Ballesta AM, et al. Aminoglycoside nephrotoxicity in cirrhosis. Value of urinary beta 2microglobulin to discriminate functional renal failure from acute tubular damage. Gastroenterology 1982;82:97–105. 85. McCormick PA, Greenslade L, Kibbler CC, et al. A prospective randomized trial of ceftazidime versus netilmicin plus mezlocillin in the empirical therapy of presumed sepsis in cirrhotic patients. Hepatology 1997;25:833–836. 86. Bhathal PS, Grossman HJ. Reduction of the increased portal vascular resistance of the isolated perfused cirrhotic rat liver by vasodilators. J Hepatol 1985;1:325–337. 87. Rockey DC. Hepatic blood flow regulation by stellate cells in normal and injured liver. Semin Liver Dis 2001;21:337–349. 88. Bataller R, Sancho-Bru P, Gines P, et al. Activated human hepatic stellate cells express the renin–angiotensin system and synthesize angiotensin II. Gastroenterology 2003;125: 117–125. 89. Gorbig MN, Gines P, Bataller R, et al. Human hepatic stellate cells secrete adrenomedullin: potential autocrine factor in the regulation of cell contractility. J Hepatol 2001;34:222–229. 90. Bataller R, Gines P, Nicolas JM, et al. Angiotensin II induces contraction and proliferation of human hepatic stellate cells. Gastroenterology 2000;118:1149–1156. 91. Morali GA, Sniderman KW, Deitel KM, et al. Is sinusoidal portal hypertension a necessary factor for the development of hepatic ascites? J Hepatol 1992;16:249–250. 92. Rector WG Jr. Portal hypertension: a permissive factor only in the development of ascites and variceal bleeding. Liver 1986;6:221–226.
448
93. Ochs A, Rossle M, Haag K, et al. The transjugular intrahepatic portosystemic stent–shunt procedure for refractory ascites. N Engl J Med 1995;332:1192–1197. 94. Groszmann RJ. Hyperdynamic circulation of liver disease 40 years later: pathophysiology and clinical consequences. Hepatology 1994;20:1359–1363. 95. Moller S, Henriksen JH. The systemic circulation in cirrhosis. In: Arroyo V, Gines P, Rodes J, Schrier RW, eds. Ascites and renal dysfunction in liver disease. Oxford: Blackwell Science, 1999: 307–329. 96. Schrier RW, Niederberger M, Weigert A, et al. Peripheral arterial vasodilatation: determinant of functional spectrum of cirrhosis. Semin Liver Dis 1994;14:14–22. 97. Guevara M, Gines P, Fernandez-Esparrach G, et al. Reversibility of hepatorenal syndrome by prolonged administration of ornipressin and plasma volume expansion. Hepatology 1998;27:35–41. 98. Uriz J, Gines P, Cardenas A, et al. Terlipressin plus albumin infusion: an effective and safe therapy of hepatorenal syndrome. J Hepatol 2000;33:43–48. 99. Nicholls KM, Shapiro MD, Kluge R, et al. Sodium excretion in advanced cirrhosis: effect of expansion of central blood volume and suppression of plasma aldosterone. Hepatology 1986;6:235–238. 100. Martin PY, Gines P, Schrier RW. Nitric oxide as a mediator of hemodynamic abnormalities and sodium and water retention in cirrhosis. N Engl J Med 1998;339:533–541. 101. Wiest R, Groszmann RJ. Nitric oxide and portal hypertension: its role in the regulation of intrahepatic and splanchnic vascular resistance. Semin Liver Dis 1999;19:411–426. 102. Morales-Ruiz M, Jimenez W, Perez-Sala D, et al. Increased nitric oxide synthase expression in arterial vessels of cirrhotic rats with ascites. Hepatology 1996;24:1481–1486. 103. Martin PY, Ohara M, Gines P, et al. Nitric oxide synthase (NOS) inhibition for one week improves renal sodium and water excretion in cirrhotic rats with ascites. J Clin Invest 1998;101:235–242. 104. Such J, Frances R, Perez-Mateo M. Nitric oxide in patients with cirrhosis and bacterial infections. Metab Brain Dis 2002;17:303–309. 105. Epstein M, Levinson R, Sancho J, et al. Characterization of the renin–aldosterone system in decompensated cirrhosis. Circ Res 1977;41:818–829. 106. Schroeder ET, Anderson GH, Goldman SH, et al. Effect of blockade of angiotensin II on blood pressure, renin and aldosterone in cirrhosis. Kidney Int 1976;9:511–519. 107. Pariente EA, Bataille C, Bercoff E, et al. Acute effects of captopril on systemic and renal hemodynamics and on renal function in cirrhotic patients with ascites. Gastroenterology 1985;88:1255–1259. 108. Arroyo V, Bosch J, Mauri M, et al. Effect of angiotensin-II blockade on systemic and hepatic haemodynamics and on the renin–angiotensin–aldosterone system in cirrhosis with ascites. Eur J Clin Invest 1981;11:221–229. 109. Guevara M, Gines P, Bandi JC, et al. Transjugular intrahepatic portosystemic shunt in hepatorenal syndrome: effects on renal function and vasoactive systems. Hepatology 1998;28: 416–422. 110. Arroyo V, Planas R, Gaya J, et al. Sympathetic nervous activity, renin–angiotensin system and renal excretion of prostaglandin E2 in cirrhosis. Relationship to functional renal failure and sodium and water excretion. Eur J Clin Invest 1983;13:271–278. 111. Dudley F, Esler M. The sympathetic nervous system in cirrhosis. In: Arroyo V, Gines P, Rodes J, Schrier RW, eds. Ascites and renal dysfunction in liver disease. Oxford: Blackwell Science, 1999: 198–219.
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
112. DiBona GF, Sawin LL, Jones SY. Characteristics of renal sympathetic nerve activity in sodium-retaining disorders. Am J Physiol 1996;271:R295–R302. 113. Uriz J, Gines P, Ortega R, et al. Increased plasma levels of neuropeptide Y in hepatorenal syndrome. J Hepatol 2002;36:349–355. 114. Liu H, Song D, Lee SS. Cirrhotic cardiomyopathy. Gastroenterol Clin Biol 2002;26:842–847. 115. Ma Z, Lee SS. Cirrhotic cardiomyopathy: getting to the heart of the matter. Hepatology 1996;24:451–459. 116. Cardenas A, Arroyo V. Mechanisms of water and sodium retention in cirrhosis and the pathogenesis of ascites. Best Pract Res Clin Endocrinol Metab 2003;17:607–622. 117. Cahill PA, Redmond EM, Sitzmann JV. Endothelial dysfunction in cirrhosis and portal hypertension. Pharmacol Ther 2001;89:273–293. 118. Epstein M, Goligorsky MS. Endothelin and nitric oxide in hepatorenal syndrome: a balance reset. J Nephrol 1997;10:120–135. 119. Jimenez W, Poo JL, Leivas A. Endothelin and systemic, renal and hepatic hemodynamics disturbances in cirrhosis. In: Arroyo V, Gines P, Rodes J, Schrier RW, eds. Ascites and renal dysfunction in liver disease. Oxford: Blackwell Science, 1999: 291–303. 120. Soper CP, Latif AB, Bending MR. Amelioration of hepatorenal syndrome with selective endothelin-A antagonist. Lancet 1996;347:1842–1843. 121. Huber M, Kastner S, Scholmerich J, et al. Analysis of cysteinyl leukotrienes in human urine: enhanced excretion in patients with liver cirrhosis and hepatorenal syndrome. Eur J Clin Invest 1989;19:53–60. 122. Zipser RD, Radvan GH, Kronborg IJ, et al. Urinary thromboxane B2 and prostaglandin E2 in the hepatorenal syndrome: evidence for increased vasoconstrictor and decreased vasodilator factors. Gastroenterology 1983;84: 697–703. 123. Zipser RD, Kronborg I, Rector W, et al. Therapeutic trial of thromboxane synthesis inhibition in the hepatorenal syndrome. Gastroenterology 1984;87:1228–1232. 124. Sacerdoti D, Balazy M, Angeli P, et al. Eicosanoid excretion in hepatic cirrhosis. Predominance of 20-HETE. J Clin Invest 1997;100:1264–1270. 125. Stanley AJ, Forrest EH, Dabos K, et al. Natriuretic effect of an adenosine-1 receptor antagonist in cirrhotic patients with ascites. Gastroenterology 1998;115:406–411. 126. Llach J, Gines P, Arroyo V, et al. Effect of dipyridamole on kidney function in cirrhosis. Hepatology 1993;17:59–64. 127. Moore K. Arachidonic acid metabolites and the kidney in cirrhosis. In: Arroyo V, Gines P, Rodes J, Schrier RW, eds. Ascites and renal dysfunction in liver disease. Oxford: Blackwell Science, 1999: 249–272. 128. Claria J, Kent JD, Lopez-Parra M, et al. Effects of celecoxib and naproxen on renal function in nonazotemic patients with cirrhosis and ascites. Hepatology 2005;41:579–587. 129. Laffi G, La Villa G, Pinzani M, et al. Arachidonic acid derivatives and renal function in liver cirrhosis. Semin Nephrol 1997;17:530–548. 130. Gines P, Jimenez W, Arroyo V, et al. Atrial natriuretic factor in cirrhosis with ascites: plasma levels, cardiac release and splanchnic extraction. Hepatology 1988;8:636–642. 131. Levy M. Atrial natriuretic peptide: renal effects in cirrhosis of the liver. Semin Nephrol 1997;17:520–529. 132. La Villa G, Laffi G, Gentilini P. Atrial natriuretic peptide and other natriuretic factors in cirrhosis. In: Arroyo V, Gines P, Rodes J, Schrier RW, eds. Ascites and renal dysfunction in liver disease. Oxford: Blackwell Science, 1999: 231–248. 133. Gulberg V, Moller S, Henriksen JH, et al. Increased renal production of C-type natriuretic peptide (CNP) in patients
134. 135.
135a.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148. 149. 150.
151.
152.
with cirrhosis and functional renal failure. Gut 2000;47:852–857. Salo J, Jimenez W, Kuhn M, et al. Urinary excretion of urodilatin in patients with cirrhosis. Hepatology 1996;24:1428–1432. Henriksen JH, Gotze JP, Fuglsang S, et al. Increased circulating pro-brain natriuretic peptide (proBNP) and brain natriuretic peptide (BNP) in patients with cirrhosis: relation to cardiovascular dysfunction and severity of disease. Gut 2003;52:1511–1517. Ros J, Clariz J, Jimenez W, et al. Role of nitric oxide and prost cyclin in the control of renal perfusion in experimental cirrhosis. Hepatology 1995;22:915–920. Criado M, Flores O, Ortiz MC, et al. Elevated glomerular and blood mononuclear lymphocyte nitric oxide production in rats with chronic bile duct ligation: role of inducible nitric oxide synthase activation. Hepatology 1997;26:268–276. Bosch-Marce M, Morales-Ruiz M, Jimenez W, et al. Increased renal expression of nitric oxide synthase type III in cirrhotic rats with ascites. Hepatology 1998;27:1191–1199. Hattori K, Hasumura Y, Takeuchi J. Role of renal kallikrein in the derangement of sodium and water excretion in cirrhotic patients. Scand J Gastroenterol 1984;19:844–848. Perez-Ayuso RM, Arroyo V, Camps J, et al. Renal kallikrein excretion in cirrhotics with ascites: relationship to renal hemodynamics. Hepatology 1984;4:247–252. Robertson GL, Berl T. Pathophysiology of water metabolism: water retaining disorders. In: Brenner BM, Recotr FC, eds. The kidney, 5th edn. Philadelphia: WB Saunders, 1996: 873–928. Arroyo V, Rodes J, Gutierrez-Lizarraga MA, et al. Prognostic value of spontaneous hyponatremia in cirrhosis with ascites. Am J Dig Dis 1976;21:249–256. Papper S, Rosenbaum JD. Abnormalities in the excretion of water and sodium in ‘compensated’ cirrhosis of the liver. J Lab Clin Med 1952;40:523–530. Caregaro L, Lauro S, Angeli P, et al. Renal water and sodium handling in compensated liver cirrhosis: mechanism of the impaired natriuresis after saline loading. Eur J Clin Invest 1985;15:360–364. Trevisani F, Bernardi M, Gasbarrini A, et al. Bed-rest-induced hypernatriuresis in cirrhotic patients without ascites: does it contribute to maintain ‘compensation’? J Hepatol 1992;16:190–196. Bernardi M, Trevisani F, Santini C, et al. Aldosterone related blood volume expansion in cirrhosis before and during the early phase of ascites formation. Gut 1983;24:761–766. Merkel C, Gatta A, Donada C, et al. Long-term effect of nadolol or nadolol plus isosorbide-5-mononitrate on renal function and ascites formation in patients with cirrhosis. GTIP Gruppo Triveneto per l’Ipertensione Portale. Hepatology 1995;22:808–813. Salmeron JM, Ruiz dA, Gines A, et al. Renal effects of acute isosorbide-5-mononitrate administration in cirrhosis. Hepatology 1993;17:800–806. Gines P, Arrovo V, Rodes J. Pharmacotherapy of ascites associated with cirrhosis. Drugs 1992;43:316–332. Gerbes AL. Medical treatment of ascites in cirrhosis. J Hepatol 1993;17(Suppl 2):S4–S9. Henriksen JH, Ring-Larsen H, Kanstrup IL, et al. Splanchnic and renal elimination and release of catecholamines in cirrhosis. Evidence of enhanced sympathetic nervous activity in patients with decompensated cirrhosis. Gut 1984;25:1034–1043. Ring-Larsen H, Hesse B, Henriksen JH, et al. Sympathetic nervous activity and renal and systemic hemodynamics in cirrhosis: plasma norepinephrine concentration, hepatic extraction, and renal release. Hepatology 1982;2:304–310. Fernandez-Rodriguez C, Prieto J, Quiroga J, et al. Atrial natriuretic factor in cirrhosis: relationship to renal function and hemodynamic changes. J Hepatol 1994;21:211–216.
449
Section III. Clinical Consequences of Liver Disease
153. Arroyo V, Jimenez W. Complications of cirrhosis. II. Renal and circulatory dysfunction. Lights and shadows in an important clinical problem. J Hepatol 2000;32(Suppl 1):157–170. 154. Kostreva DR, Castaner A, Kampine JP. Reflex effects of hepatic baroreceptors on renal and cardiac sympathetic nerve activity. Am J Physiol 1980;238:R390–R394. 155. Arroyo V, Bosch J, Mauri M, et al. Renin, aldosterone and renal haemodynamics in cirrhosis with ascites. Eur J Clin Invest 1979;9:69–73. 156. Bichet DG, Van Putten VJ, Schrier RW. Potential role of increased sympathetic activity in impaired sodium and water excretion in cirrhosis. N Engl J Med 1982;307:1552–1557. 157. Bosch J, Arroyo V, Betriu A, et al. Hepatic hemodynamics and the renin–angiotensin–aldosterone system in cirrhosis. Gastroenterology 1980;78:92–99. 158. DiBona GF, Sawin LL. Effect of renal nerve stimulation on NaCl and H2O transport in Henle’s loop of the rat. Am J Physiol 1982;243:F576–F580. 159. DiBona GF, Kopp UC. Neural control of renal function. Physiol Rev 1997;77:75–197. 160. Gonzalez-Abraldes J, Albillos A, Banares R, et al. Randomized comparison of long-term losartan versus propranolol in lowering portal pressure in cirrhosis. Gastroenterology 2001;121:382–388. 161. Albillos A, Banares R, Barrios C, et al. Oral administration of clonidine in patients with alcoholic cirrhosis. Hemodynamic and liver function effects. Gastroenterology 1992;102:248–254. 162. Claria J, Jimenez W, Arroyo V, et al. Effect of V1-vasopressin receptor blockade on arterial pressure in conscious rats with cirrhosis and ascites. Gastroenterology 1991;100:494–501. 163. Planas R, Arroyo V, Rimola A, et al. Acetylsalicylic acid suppresses the renal hemodynamic effect and reduces the diuretic action of furosemide in cirrhosis with ascites. Gastroenterology 1983;84:247–252. 164. Arroyo V, Gines P, Rimola A, et al. Renal function abnormalities, prostaglandins, and effects of nonsteroidal antiinflammatory drugs in cirrhosis with ascites. An overview with emphasis on pathogenesis. Am J Med 1986;81:104–122. 165. Arroyo V, Rimola A, Gaya J, et al. Prostaglandins and renal function in cirrhosis. Prog Liver Dis 1986;8:505–523. 166. Quintero E, Gines P, Arroyo V, et al. Sulindac reduces the urinary excretion of prostaglandins and impairs renal function in cirrhosis with ascites. Nephron 1986;42:298–303. 167. Ros J, Claria J, Jimenez W, et al. Role of nitric oxide and prostacyclin in the control of renal perfusion in experimental cirrhosis. Hepatology 1995;22:915–920. 168. Zhang PL, Jimenez W, Mackenzie HS, et al. HS-142-1, a potent antagonist of natriuretic peptides in vitro and in vivo. J Am Soc Nephrol 1994;5:1099–1105. 169. Angeli P, Jimenez W, Arroyo V, et al. Renal effects of natriuretic peptide receptor blockade in cirrhotic rats with ascites. Hepatology 1994;20:948–954. 170. Bichet D, Szatalowicz V, Chaimovitz C, et al. Role of vasopressin in abnormal water excretion in cirrhotic patients. Ann Intern Med 1982;96:413–417. 171. Perez-Ayuso RM, Arroyo V, Camps J, et al. Evidence that renal prostaglandins are involved in renal water metabolism in cirrhosis. Kidney Int 1984;26:72–80. 172. Maroto A, Gines A, Salo J, et al. Diagnosis of functional kidney failure of cirrhosis with Doppler sonography: prognostic value of resistive index. Hepatology 1994;20:839–844. 173. Sato S, Ohnishi K, Sugita S, et al. Splenic artery and superior mesenteric artery blood flow: nonsurgical Doppler US measurement in healthy subjects and patients with chronic liver disease. Radiology 1987;164:347–352. 174. Diez J, Simon MA, Anton F, et al. Tubular sodium handling in cirrhotic patients with ascites as analysed by the renal lithium clearance method. Eur J Clin Invest 1990;20:266–271.
450
175. Arroyo V, Bernardi M, Epstein M, et al. Pathophysiology of ascites and functional renal failure in cirrhosis. J Hepatol 1988;6:239–257. 176. Llovet JM, Planas R, Morillas R, et al. Short-term prognosis of cirrhotics with spontaneous bacterial peritonitis: multivariate study. Am J Gastroenterol 1993;88:388–392. 177. Llovet JM, Rodriguez-Iglesias P, Moitinho E, et al. Spontaneous bacterial peritonitis in patients with cirrhosis undergoing selective intestinal decontamination. A retrospective study of 229 spontaneous bacterial peritonitis episodes. J Hepatol 1997;26:88–95. 178. Fernandez J, Navasa M, Ruiz-del-Arbol L, et al. Effects of spontaneous bacterial peritonitis (SBP) on systemic haemodynamics in patients with cirrhosis. J Hepatol 2001;34:73. 179. Gonzalez-Garcia M, Monescillo A, Peña E, et al. Estudio prospectivo clínico-hemodinámico del síndrome hepatorenal (SHR) en la cirrosis hepática [abstr]. Rev Esp Enf Dig 2001;93(Suppl 1):11. 180. Claria J, Jimenez W, Ros J, et al. Pathogenesis of arterial hypotension in cirrhotic rats with ascites: role of endogenous nitric oxide. Hepatology 1992;15:343–349. 181. Bailie MD, Rector FC Jr, Seldin DW. Angiotensin II in arterial and renal venous plasma and renal lymph in the dog. J Clin Invest 1971;50:119–126. 182. Osswald H, Vallon V, Muhlbauer B. Role of adenosine in tubuloglomerular feedback and acute renal failure. J Autonom Pharmacol 1996;16:377–380. 183. Lopez-Parra M, Claria J, Planaguma A, et al. Cyclooxygenase-1 derived prostaglandins are involved in the maintenance of renal function in rats with cirrhosis and ascites. Br J Pharmacol 2002;135:891–900. 184. Duvoux C, Zanditenas D, Hezode C, et al. Effects of noradrenalin and albumin in patients with type I hepatorenal syndrome: a pilot study. Hepatology 2002;36:374–380. 185. Schrier RW, Arroyo V, Bernardi M, et al. Peripheral arterial vasodilatation hypothesis: a proposal for the initiation of renal sodium and water retention in cirrhosis. Hepatology 1988;8:1151–1157. 186. Colombato LA, Albillos A, Groszmann RJ. Temporal relationship of peripheral vasodilatation, plasma volume expansion and the hyperdynamic circulatory state in portalhypertensive rats. Hepatology 1992;15:323–328. 187. Angeli P, Volpin R, Gerunda G, et al. Reversal of type 1 hepatorenal syndrome with the administration of midodrine and octreotide. Hepatology 1999;29:1690–1697. 188. Ortega R, Gines P, Uriz J, et al. Terlipressin therapy with and without albumin for patients with hepatorenal syndrome: results of a prospective, nonrandomized study. Hepatology 2002;36:941–948. 189. Moller S, Henriksen JH. Cardiovascular dysfunction in cirrhosis. Pathophysiological evidence of a cirrhotic cardiomyopathy. Scand J Gastroenterol 2001;36: 785–794. 190. Salo J, Gines A, Anibarro L, et al. Effect of upright posture and physical exercise on endogenous neurohormonal systems in cirrhotic patients with sodium retention and normal supine plasma renin, aldosterone, and norepinephrine levels. Hepatology 1995;22:479–487. 191. Salo J, Guevara M, Fernandez-Esparrach G, et al. Impairment of renal function during moderate physical exercise in cirrhotic patients with ascites: relationship with the activity of neurohormonal systems. Hepatology 1997;25:1338– 1342. 192. Ring-Larsen H, Henriksen JH, Wilken C, et al. Diuretic treatment in decompensated cirrhosis and congestive heart failure: effect of posture. Br Med J (Clin Res Ed) 1986;292:1351–1353.
Chapter 22 RENAL DYSFUNCTION IN CIRRHOSIS: PATHOPHYSIOLOGY, CLINICAL FEATURES AND THERAPY
193. Reynolds TB, Lieberman FL, Goodman AR. Advantages of treatment of ascites without sodium restriction and without complete removal of excess fluid. Gut 1978;19:549–553. 194. Arroyo V, Gines P, Rodes J. Treatment of ascites in patients with cirrhosis of the liver. J Hepatol 1986;2:504–512. 195. Gauthier A, Levy VG, Quinton A, et al. Salt or no salt in the treatment of cirrhotic ascites: a randomised study. Gut 1986;27:705–709. 196. Garcia-Tsao G. Current management of the complications of cirrhosis and portal hypertension: variceal hemorrhage, ascites, and spontaneous bacterial peritonitis. Gastroenterology 2001;120:726–748. 197. Suki WN, Eknoyan G. Physiology of diuretic action. In: Seldin DW, Giebisch G, eds. The kidney: physiology and pathophysiology, 2nd edn. New York: Raven Press, 1992: 3629–3670. 198. Puschet JB, Winaver J. Effects of diurectics on renal function. In: Ewindhager E, ed. Handbook of physiology. Oxford: Oxford University Press, 1992: 2335–2407. 199. Attallah AA. Interaction of prostaglandins with diuretics. Prostaglandins 1979;18:369–375. 200. Sungaila I, Bartle WR, Walker SE, et al. Spironolactone pharmacokinetics and pharmacodynamics in patients with cirrhotic ascites. Gastroenterology 1992;102:1680–1685. 201. Karim A. Spironolactone metabolism in man revisited. In: Brunner HR, ed. Contemporary trends in diuretic therapy. Amsterdam: Excerpta Medica, 1986: 23–27. 202. Verrey F. Regulation of gene expression by aldosterone in tight epithelia. Semin Nephrol 1990;10:410–420. 203. Marver D, Stewart J, Funder JW, et al. Renal aldosterone receptors: studies with (3H)aldosterone and the antimineralocorticoid (3H)spirolactone (SC-26304). Proc Natl Acad Sci USA 1974;71:1431–1435. 204. Cheng SC, Suzuki K, Sadee W, et al. Effects of spironolactone, canrenone and canrenoate-K on cytochrome P450, and 11betaand 18-hydroxylation in bovine and human adrenal cortical mitochondria. Endocrinology 1976;99:1097–1106. 205. Corvol P, Claire M, Oblin ME, et al. Mechanism of the antimineralocorticoid effects of spirolactones. Kidney Int 1981;20:1–6. 206. Arroyo V, Bosch J, Casamitjana R, et al. Use of piretanide, a new loop diuretic, in cirrhosis with ascites: relationship between the diuretic response and the plasma aldosterone level. Gut 1980;21:855–859. 207. Campra JL, Reynolds TB. Effectiveness of high-dose spironolactone therapy in patients with chronic liver disease and relatively refractory ascites. Am J Dig Dis 1978;23:1025–1030. 208. Lopez-Novoa JM, Rengel MA. A micropuncture study of salt and water retention in chronic experimental cirrhosis. Am J Physiol 1977;232:F315–F318. 209. Chiandussi L, Bartoli E, Arras S. Reabsorption of sodium in the proximal renal tubule in cirrhosis of the liver. Gut 1978;19:497–503. 210. Takaya A, Fukui H, Matsumura M, et al. Stepped care medical treatment for cirrhotic ascites: analysis of factors influencing the response to treatment. J Gastroenterol Hepatol 1995;10:30–35. 211. Forns X, Gines A, Gines P, et al. Management of ascites and renal failure in cirrhosis. Semin Liver Dis 1994;14:82–96. 212. Shear L, Ching S, Gabuzda GJ. Compartmentalization of ascites and edema in patients with hepatic cirrhosis. N Engl J Med 1970;282:1391–1396. 213. Sherlock S, Senewiratne B, Scott A, et al. Complications of diuretic therapy in hepatic cirrhosis. Lancet 1966;1:1049–1052. 214. Rodes J, Bosch J, Arroyo V. Clinical types and drug therapy of renal impairment in cirrhosis. Postgrad Med J 1975;51:492–497.
215. Baertl JM, Sancetta SM, Gabuzda GJ. Relation of acute potassium depletion to renal ammonium metabolism in patients with cirrhosis. J Clin Pharmacol 1963;42:696–707. 216. Gussin RZ. Potassium-sparing diuretics. J Clin Pharmacol 1977;17:651–662. 217. Good DW, Wright FS. Luminal influences on potassium secretion: sodium concentration and fluid flow rate. Am J Physiol 1979;236:F192–F205. 218. Stripp B, Taylor AA, Bartter FC, et al. Effect of spironolactone on sex hormones in man. J Clin Endocrinol Metab 1975;41:777–781. 219. Menard RH, Loriaux DL, Bartter FC, et al. The effect of the administration of spironolactone on the concentration of plasma testosterone, estradiol and cortisol in male dogs (1). Steroids 1978;31:771–782. 220. Angeli P, Albino G, Carraro P, et al. Cirrhosis and muscle cramps: evidence of a causal relationship. Hepatology 1996;23:264–273. 221. Lee FY, Lee SD, Tsai YT, et al. A randomized controlled trial of quinidine in the treatment of cirrhotic patients with muscle cramps. J Hepatol 1991;12:236–240. 222. Nielsen S, Marples D, Frokiaer J, et al. The aquaporin family of water channels in kidney: an update on physiology and pathophysiology of aquaporin-2. Kidney Int 1996;49:1718–1723. 223. Lolait SJ, O’Carroll AM, McBride OW, et al. Cloning and characterization of a vasopressin V2 receptor and possible link to nephrogenic diabetes insipidus. Nature 1992;357:336–339. 224. Fushimi K, Uchida S, Hara Y, et al. Cloning and expression of apical membrane water channel of rat kidney collecting tubule. Nature 1993;361:549–552. 225. Knepper MA, Nielsen S, Chou CL, et al. Mechanism of vasopressin action in the renal collecting duct. Semin Nephrol 1994;14:302–321. 226. Knepper MA. Molecular physiology of urinary concentrating mechanism: regulation of aquaporin water channels by vasopressin. Am J Physiol 1997;272:F3–12. 227. Hofbauer KG, Studer W, Mah SC, et al. The significance of vasopressin as a pressor agent. J Cardiovasc Pharmacol 1984;6(Suppl 2):S429–438. 228. Arroyo V, Claria J, Salo J, et al. Antidiuretic hormone and the pathogenesis of water retention in cirrhosis with ascites. Semin Liver Dis 1994;14:44–58. 229. Leander JD. Further study of kappa opioids on increased urination. J Pharmacol Exp Ther 1983;227:35–41. 230. Leander JD, Zimmerman DM. Effects of picenadol and its agonist and antagonist isomers on schedule-controlled behavior. J Pharmacol Exp Ther 1983;227:671–675. 231. Peters GR, Ward NJ, Antal EG, et al. Diuretic actions in man of a selective kappa opioid agonist: U-62,066E. J Pharmacol Exp Ther 1987;240:128–131. 232. Ohnishi A, Mihara M, Yasuda S, et al. Aquaretic effect of the stable dynorphin-A analog E2078 in the human. J Pharmacol Exp Ther 1994;270:342–347. 233. Kinter LB, Huffman WF, Stassen FL. Antagonists of the antidiuretic activity of vasopressin. Am J Physiol 1988;254:F165–F177. 234. Laszlo FA, Laszlo F Jr, De Wied D. Pharmacology and clinical perspectives of vasopressin antagonists. Pharmacol Rev 1991;43:73–108. 235. Perez-Ayuso RM, Arroyo V, Camps J, et al. Effect of demeclocycline on renal function and urinary prostaglandin E2 and kallikrein in hyponatremic cirrhotics. Nephron 1984;36:30–37. 236. Gadano A, Moreau R, Pessione F, et al. Aquaretic effects of niravoline, a kappa-opioid agonist, in patients with cirrhosis. J Hepatol 2000;32:38–42.
451
Section III. Clinical Consequences of Liver Disease
237. Sawyer WH, Manning M. The development of potent and specific vasopressin antagonists. Kidney Int Suppl 1988;26:S34–37. 238. Yamamura Y, Ogawa H, Chihara T, et al. OPC-21268, an orally effective, nonpeptide vasopressin V1 receptor antagonist. Science 1991;252:572–574. 239. Yamamura Y, Ogawa H, Yamashita H, et al. Characterization of a novel aquaretic agent, OPC-31260, as an orally effective, nonpeptide vasopressin V2 receptor antagonist. Br J Pharmacol 1992;105:787–791. 240. Albright JD, Reich MF, Delos Santos EG, et al. 5-Fluoro-2methyl-N-[4-(5H-pyrrolo[2,1-c]-[1, 4]benzodiazepin-10(11H)ylcarbonyl)-3-chlorophenyl]benzamide (VPA-985): an orally active arginine vasopressin antagonist with selectivity for V2 receptors. J Med Chem 1998;41:2442–2444. 241. Chan PS, Coupet J, Park HC, et al. VPA-985, a nonpeptide orally active and selective vasopressin V2 receptor antagonist. Adv Exp Med Biol 1998;449:439–43.:439–443. 242. Serradeil-Le Gal C, Lacour C, Valette G, et al. Characterization of SR 121463A, a highly potent and selective, orally active vasopressin V2 receptor antagonist. J Clin Invest 1996;98:2729–2738. 243. Yamamura Y, Nakamura S, Itoh S, et al. OPC-41061, a highly potent human vasopressin V2-receptor antagonist: pharmacological profile and aquaretic effect by single and multiple oral dosing in rats. J Pharmacol Exp Ther 1998;287:860–867. 244. Tahara A, Saito M, Sugimoto T, et al. Pharmacological characterization of YM087, a potent, nonpeptide human vasopressin V1A and V2 receptor antagonist. Naunyn Schmiedebergs Arch Pharmacol 1998;357: 63–69. 245. Tsuboi Y, Ishikawa S, Fujisawa G, et al. Therapeutic efficacy of the non-peptide AVP antagonist OPC-31260 in cirrhotic rats. Kidney Int 1994;46:237–244. 246. Ohnishi A, Orita Y, Okahara R, et al. Potent aquaretic agent. A novel nonpeptide selective vasopressin 2 antagonist (OPC31260) in men. J Clin Invest 1993;92:2653–2659.
452
247. Ohnishi A, Orita Y, Takagi N, et al. Aquaretic effect of a potent, orally active, nonpeptide V2 antagonist in men. J Pharmacol Exp Ther 1995;272:546–551. 248. Gines P, Guevara M, Arroyo V, et al. Hepatorenal syndrome. Lancet 2003;362:1819–1827. 249. Gulberg V, Bilzer M, Gerbes AL. Long-term therapy and retreatment of hepatorenal syndrome type 1 with ornipressin and dopamine. Hepatology 1999;30:870–875. 250. Mulkay JP, Louis H, Donckier V, et al. Long-term terlipressin administration improves renal function in cirrhotic patients with type 1 hepatorenal syndrome: a pilot study. Acta Gastroenterol Belg 2001;64:15–19. 251. Moreau R, Durand F, Poynard T, et al. Terlipressin in patients with cirrhosis and type 1 hepatorenal syndrome: a retrospective multicenter study. Gastroenterology 2002;122:923–930. 252. Alessandria C, Venon WD, Marzano A, et al. Renal failure in cirrhotic patients: role of terlipressin in clinical approach to hepatorenal syndrome type 2. Eur J Gastroenterol Hepatol 2002;14:1363–1368. 253. Brensing KA, Textor J, Perz J, et al. Long term outcome after transjugular intrahepatic portosystemic stent–shunt in nontransplant cirrhotics with hepatorenal syndrome: a phase II study. Gut 2000;47:288–295. 253a. Alam L, Bass NM, La Berge JM, et al. Treatment of hepatorenal syndrome with the transjugular intrahepatic shunt (TIPS). Gastroenterology 1995;108:1024A. 253b. Ochs A, Rösle M, Haag K, et al. TIPS for hepatorenal syndrome. Hepatology 1994;20:114A. 254. Brensing KA PJST. TIPS in hepatorenal syndrome. In: Arroyo V, ed. Treatment of liver diseases. Barcelona: Masson, 1999: 53–59. 255. Wong F, Pantea L, Sniderman K. Midodrine, octreotide, albumin, and TIPS in selected patients with cirrhosis and type 1 hepatorenal syndrome. Hepatology 2004;40:55–64. 256. Akriviadis E, Botla R, Briggs W, et al. Pentoxifylline improves short-term survival in severe acute alcoholic hepatitis: a doubleblind, placebo-controlled trial. Gastroenterology 2000;119:1637–1648.
Section III: Clinical Consequences of Liver Disease
23
CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR Samuel S. Lee and Soon Koo Baik Abbreviations ANP atrial natriuretic peptide BDL bile duct-ligated rat BNP brain or B-type natriuretic peptide CNS central nervous system CO carbon monoxide dP/dt first time derivative of peak ventricular pressure rise
E/A ratio early diastolic filling wave velocity/late diastolic filling wave velocity ratio G-protein guanosine triphosphate-binding protein HO heme oxygenase LV left ventricle NO nitric oxide
You raise and gather the threads and the grams of life, the final distillate, the intimate essences. Submerged viscus, measurer of the blood. Pablo Neruda, Ode to the Liver
NOS NTS SVR TIPS
nitric oxide synthase nucleus of the solitary tract systemic vascular resistance transjugular intrahepatic portosystemic stent-shunt
as an obscure scientific curiosity, has evolved into a clinically important entity. Nowadays it is recognized that these cardiovascular derangements contribute to the pathogenesis of several complications of liver disease, including variceal bleeding, ascites and hepatorenal syndrome, hepatopulmonary syndrome, and increased susceptibility to peri- and postoperative complications. In the sections that follow, all aspects of cardiovascular function and dysfunction in liver disease will be reviewed, including the peripheral vasculature, the heart, and the regulatory mechanisms controlling both.
INTRODUCTION
HYPERDYNAMIC CIRCULATION
Until relatively recently in history, the liver was thought to play a major role in blood circulation. For example, it was not until 1628 that William Harvey demonstrated that the blood circulation is a continuous circuit controlled by the heart, thus dispelling the belief that the liver controlled a separate venous circulation. In the latter half of the 20th century, with the development of techniques for accurately measuring cardiovascular phenomena, Kowalski and Abelmann1 confirmed that hepatic cirrhosis is associated with cardiovascular disturbances. Specifically, the cardiac output was increased, and systemic vascular resistance and blood pressure decreased in their patients with alcoholic cirrhosis. They ascribed this hyperdynamic circulation to the presence of a circulating vasodilator produced by the diseased liver. This work was followed by many others that confirmed the presence of the hyperdynamic circulation in cirrhosis (reviewed in 2–5). Almost as an afterthought, Kowalski and Abelmann also reported that the electrocardiographic QT interval was prolonged in some patients. This phenomenon went essentially unnoticed and was ‘rediscovered’ more than four decades later. This seminal paper, dating back some 50 years, launched the modern era of ‘cardiohepatology,’ the study of cardiovascular disturbances in liver disease. This area of inquiry, from its early days
Since the initial description of the hyperdynamic circulation in cirrhosis, much progress has been made in understanding the pathophysiology and pathogenesis of this phenomenon. The hyperdynamic circulation is characterized by increased cardiac output and heart rate, and decreased systemic vascular resistance (SVR) with low arterial blood pressure.1–5 These hemodynamic alterations arise as a complication of portal hypertension, although the exact mechanisms that lead to a hyperdynamic circulation remain unclear. However, it is clear that many aspects of the cardiovascular system are abnormal in portal hypertension or cirrhosis. The heart itself functions abnormally in several respects, which will be detailed in the next section. Abnormalities in several regional vascular beds, including the hepatic, mesenteric/splanchnic, renal, pulmonary, skeletal muscle and cerebral circulations, have been documented (Table 23-1).
CLINICAL FEATURES OF HYPERDYNAMIC CIRCULATION The hyperdynamic circulation is readily evident when well established. The patient with tachycardia, hypotension, and bounding
453
Section III. Clinical Consequences of Liver Disease
Table 23-1. Clinical Consequences of Hyperdynamic Circulation and Cirrhotic Cardiomyopathy Tissue or vascular bed
Clinical syndrome
Putative mechanism
Clinical significance
HDC, lung HDC, gut HDC, kidney
Hepatopulmonary syndrome Gastroesophageal varices Sodium and water retention
Dyspnea, limited exercise tolerance Increased risk of variceal bleeding Peripheral edema and ascites
HDC, brain
Encephalopathy?
HDC, skeletal muscle Heart, endocardium
Muscle wasting? Infective endocarditis
Heart, coronary arteries Heart, pericardium Cardiomyocyte repolarization CCM, liver after TIPS
Decreased atherosclerosis?
Pulmonary vasodilatation causes hypoxemia Mesenteric vasodilatation increases portal venous flow Peripheral vasodilatation produces decreased effective volume? Decreased cerebral blood flow; regional redistribution of flow in CNS Hyperemia in early stages? Hypoperfusion in late stages? Defective immune cell phagocytosis, opsonization, bacterial killing Decreased afterload? Protective effects of alcohol? Decreased cholesterol? Manifestation of generalized systemic fluid overload Abnormal potassium-channel function causing delayed repolarization Abrupt increase in preload severely aggravates diastolic dysfunction Afterload normalization stresses ventricular function
CCM, liver after transplantation CCM, kidney CCM, kidney
Pericardial effusions Prolonged QT interval Precipitate overt heart failure Precipitate acute heart failure Precipitate hepatorenal syndrome in SBP? Sodium and water retention?
Blunted cardiac response to sepsis reduces renal perfusion? Inadequate LV function decreases effective circulating volume?
Unknown Unknown Triples risk of endocarditis compared to controls; still rare Unclear whether this clinical suspicion is correct None Probably none. Torsades de pointes arrhythmia very rare Uncommon and usually transient Usually transient but more common than suspected in first 3 days Strong circumstantial evidence Only circumstantial evidence to date
HDC, hyperdynamic circulation; CCM, cirrhotic cardiomyopathy; TIPS, transjugular intrahepatic portosystemic shunt; LV, left ventricle; SBP, spontaneous bacterial peritonitis.
pulses is easily recognized. Although hyperdynamic circulation per se is not distressing to the patient, the clinical relevance of this phenomenon is its propensity to aggravate or precipitate some of the complications of portal hypertension. The severity of the hyperdynamic circulation correlates with advancing liver failure, i.e. patients with end-stage liver failure generally show the greatest extent of peripheral vasodilatation and increased cardiac output.2–6 Thus patients with decompensated cirrhosis virtually all show evidence of a hyperdynamic circulation. However, the presence of portal hypertension, rather than liver failure, is a sine qua non for the development of a hyperdynamic circulation. Patients with prehepatic portal hypertension due to portal vein obstruction, hepatic schistosomiasis, and non-cirrhotic portal fibrosis have normal or near-normal hepatocellular function and yet manifest a hyperdynamic circulation.2–5 Animal models of portal vein stenosis or ligation also confirm this finding.7,8 The hyperdynamic splanchnic circulation is problematic. The gut and liver receive a third of the entire cardiac output, and abnormalities in this vascular bed directly or indirectly contribute to two of the most troublesome complications of cirrhosis: ascites and variceal bleeding. In concert with the increased total cardiac output, mesenteric blood flow also increases.7,8 Moreover, studies in both humans and animal models of cirrhosis or portal hypertension confirm that mesenteric hyperemia is due not only to a passive increase in blood flow as part of the increased cardiac output, but also to mesenteric vasodilatation, i.e. the percentage of overall cardiac output perfusing the mesenteric organs also increases.3–5,7,8 As portal venous pressure is the product of resistance times flow, this mesenteric hyperemia contributes to portal hypertension (the ‘forward flow’ mechanism). Moreover, the portal hypertension induces the creation of portosystemic collaterals (gastroesophageal
454
varices), which may not only bleed but may also affect the local and systemic circulations. Recently, bacterial infection has been recognized as a risk factor for precipitating variceal bleeding.9 The underlying mechanism of this curious observation remains unknown, but it has been suggested that humoral substances released during the course of sepsis, including endotoxins and cytokines such as TNF-a, intensify the hyperdynamic circulation, and thus increase blood flow through varices. Complications and management of the hyperdynamic circulation in different organs or vascular beds, such as the lung, kidney, and portal circulation, are discussed in detail elsewhere (Chapters 19, 20, 22, 24). For example, peripheral vasodilatation has been proposed to be the main pathogenic determinant leading to renal sodium and water retention, the ‘primary peripheral vasodilatation’ hypothesis.10 Vasodilatation in the pulmonary vasculature produces the hepatopulmonary syndrome.
PATHOGENESIS The exact pathogenic mechanisms leading to the hyperdynamic circulation remain to be definitively determined. Several factors to date have been postulated to play a role, including humoral substances, central neural activation, tissue hypoxia and hypervolemia. A proposed schema is outlined in Figure 23-1. Because portal venous hypertension is a prerequisite for developing the hyperdynamic circulation, consideration should begin there. Portal hypertension per se, or an associated problem such as mesenteric congestion or portosystemic collateralization, starts a chain of events that eventually leads to the hyperdynamic circulation. The congested gut favors translocation of bacteria or endotoxin to the portal circulation, and these access the systemic circulation by intrahepatic or extrahepatic portosystemic collaterals. Moreover, these
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR Figure 23-1. Pathogenesis of hyperdynamic circulation. Central neural activation
SNS AVP
Cardiac output
Sensory afferent nerves?
Afterload
Blood/plasma volume
RAAS
Effective circulating volume
Systemic vasodilation
?
Na+/H2O retention
? NO Tissue hypoxia?
Liver disease ?
Gut Portal pressure
Humoral factors from gut NO, CO? Endocannabinoids? Endotoxins? Oxidant stress?
Mesenteric venous congestion Mesenteric venous pooling
factors stimulate the release of other vasoactive substances, such as cytokines, nitric oxide (NO), carbon monoxide, and endocannabinoids. Portal or mesenteric venous hypertension also leads to neuronal activation in the brainstem and hypothalamic central cardiovascular regulatory areas such as the ventrolateral medulla and nucleus of the solitary tract (NTS). These subcortical regions then activate the circulation by neurohormonal mechanisms such as the sympathetic nervous system and arginine vasopressin. Intrahepatic portal hypertension also leads to expansion of total blood and plasma volume, by a direct signal either from the liver/gut to the kidney, or to the systemic vessels to produce peripheral vasodilatation. The vasodilatation decreases effective circulating volume, thereby inducing renal sodium and water retention, leading to an increase in the extracellular fluid volume. The expanded extracellular volume remains ineffective in improving organ perfusion due to pooling in the mesenteric venous reservoir. The decreased effective circulating volume, in concert with humoral factors such as NO and cytokines, induces tissue hypoxia. Hypoxia, along with other humoral substances, stimulates oxidative stress, i.e. reactive oxygen and nitroxy intermediates. Moreover, the tissue hypoxia and decreased effective volume are sensed by the central neural centers, which stimulate further activating signals to the heart and vasculature.
Therefore, in this schema, the hyperdynamic circulation starts as a reactive compensatory attempt by the CNS to relieve mesenteric congestion and/or tissue hypoxia. With progression of liver failure, hyperdynamic alterations become more profound and are associated with hyporesponsiveness to vasoconstrictors, increased portosystemic shunt formation, and autonomic neuropathy. In other words, several positive feedback loops involving humoral factors and neurohormonal substances accentuate the cardiovascular changes. The evidence on which this schema is based is briefly discussed in the following sections.
Nitric Oxide NO is a powerful endogenous vasodilator released from many different sources, including vascular endothelial cells. It is synthesized from L-arginine by three distinct isoforms of NO synthase (NOS), neuronal or brain NOS (nNOS, bNOS, NOS1), inducible NOS (iNOS, NOS2), and endothelial NOS (eNOS, NOS3). In 1991, Vallance and Moncada11 proposed that in cirrhosis, gut-derived endotoxemia up-regulates iNOS and the overproduction of NO causes the hyperdynamic circulation. This hypothesis was subsequently tested in numerous studies. At present, there is abundant evidence that NO contributes to the development or maintenance of hyperdynamic circulation.12–15 For example, many studies have demon-
455
Section III. Clinical Consequences of Liver Disease
strated an increased synthesis of NO in human and animal models of cirrhosis, and that inhibition of NOS suppresses the hyperdynamic circulation.16–18 Increased levels of NO in cirrhosis also blunt the activity of vasoconstrictors such as endothelin and vasopressin. Still unsettled is the exact origin of the increased NO. Most evidence indicates that iNOS is not pathogenetically involved in the hyperdynamic circulation.19–22 An alternative hypothesis of eNOS up-regulation is supported by much evidence. Rats with prehepatic portal hypertension show enhanced eNOS activity and protein expression.20–22 A major mechanism of eNOS up-regulation is increased shear stress in vascular endothelial cells. However, this raises a chicken-and-egg paradox, as the shear stress is due to peripheral vasodilation and increased blood flow. Proponents of the eNOS hypothesis point to other upregulatory mechanisms, including heat shock protein-90, endocannabinoids, and bacterial translocation with subsequent increased TNF-a production and elevated tetrahydrobiopterin levels.23–25 Studies in gene-knockout mice lacking the eNOS gene have produced conflicting results. Theodorakis et al.26 reported that such mice do not develop most features of hyperdynamic circulation after portal vein stenosis, whereas Iwakiri and co-workers showed that portal-stenosed knockout mice lacking eNOS as well as other mice lacking both iNOS and eNOS still developed the hyperdynamic circulation27 (Figure 23-2). It is possible that nNOS compensates for eNOS in situations where eNOS is inhibited or down-regulated. In the aortae of eNOS-knockout mice, increased nNOS levels partially
P=NS
CI (ml/min/kg bw)
MAP (mmHg)
The psychotropic properties of exogenous cannabinoids derived from the plant Cannabis sativa were discovered centuries ago, but only relatively recently has it been recognized that endogenous cannabinoids, or endocannabinoids, are found in mammals, including humans. Endocannabinoids exert a vasodilatory effect by means of a ligand-specific G protein-coupled receptor, the CB1 receptor. In cirrhosis, endocannabinoid levels are increased and vascular CB1 receptors activated.30–32 Additionally, the CB1 receptor antagonist SR141716A increases the arterial pressure and reduces the elevated mesenteric blood flow and portal pressure in a cirrhotic rat model.30,31 An increase in monocyte-derived endocannabinoids compared with controls was demonstrated in human and rat cirrhosis.30,31 Therefore, increased endogenous cannabinoids and activation of vascular cannabinoid CB1 receptors in mesenteric vasculature may well contribute to splanchnic vasodilation and
500
120
60
*
400 300 200 100 0
0 Sham
Sham
PVL
eNOS/iNOS–/–knockout mice
PVL
eNOS/iNOS–/–knockout mice
P<0.05
P<0.05 0.7
* 5
0
SVR (mmHg.min.kg bw/ml)
10
PP (mmHg)
Endocannabinoids
P<0.01
180
0.6 0.5 *
0.4 0.3 0.2 0.1 0
Sham
PVL
eNOS/iNOS–/–knockout mice
456
compensated for the absence of eNOS.28 A pathogenic role for upregulated nNOS in cirrhosis is also supported by other studies.5,29 Xu and co-workers treated rats with CCl4-induced cirrhosis with the selective nNOS inhibitor 7-nitroindazole for 1 week.29 This treatment reversed the hyperdynamic circulation in these rats, without affecting renal function (Figure 23-3). All these results indicate that in portal hypertension and cirrhosis the source of the increased NO levels is probably eNOS, but nNOS may also contribute, especially in situations where eNOS is inhibited.
Sham
PVL
eNOS/iNOS–/–knockout mice
Figure 23-2. Hyperdynamic circulation persists after portal vein ligation in gene knockout mice lacking eNOS and iNOS. MAP, mean arterial pressure; CI, cardiac index; PP, portal pressure; SVR, systematic vascular resistance. (Reproduced from reference 27, with permission.)
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
by stimulating guanylate cyclase to generate cGMP. Some circumstantial evidence implicates a role for the HO–CO pathway in the modulation of vascular tone in portal hypertension. HO-1 upregulation is observed in cirrhosis with portal hypertension.33–35 In portal-hypertensive rats NOS inhibition does not completely abolish relaxation of isolated arterial smooth muscle cells and hyporeactivity to vasoconstrictors, suggesting a contribution by a NO-independent local mechanism. HO inhibition partially restores the depressed reactivity to KCl-induced vasoconstriction of isolated perfused mesentery in portal-hypertensive rats.33
Cardiac Index (ml/min/100 g)
80
* 60
40
20
Other Humoral Factors
0
Evidence of a possible pathogenic role for several other vasoactive humoral substances has been published. The list includes adrenomedullin, glucagon, adenosine, prostaglandins, substance P, calcitonin gene-related peptide, and bile acids.4,5,36–39 Patients or animal models of cirrhosis or portal hypertension have elevated levels of all these substances. However, whether this represents cause or effect in terms of the hyperdynamic circulation remains unsettled.
A
SVR (mmHg/ml/min/100 g)
5 4 3
*
2
Central Neural Mechanisms 1 0
B
120 #
MAP (mmHg)
100 80 60 40 20 0 C
Control (n=7)
Cirrhosis (n=3)
Cirrhosis + 7-NI (n=3)
Figure 23-3. Effect of 1-week treatment with the nNOS inhibitor 7-nitroindazole in rats with CCl4-induced cirrhosis. Cirrhotic rats treated with 7-NI normalized cardiac index, systemic vascular resistance and mean arterial pressure. (Reproduced from reference 29, with permission.)
portal hypertension. However, not all studies show similar results: one recent study found elevated endocannabinoid levels in patients with cirrhosis, but no correlation between these levels and any hemodynamic measurement.32 This result weighs against a pathogenic role of endocannabinoids.
Heme Oxygenase and Carbon Monoxide The enzyme heme oxygenase (HO), which exists in an inducible (HO-1) and constitutive (HO-2) isoform, catalyzes the oxidation of heme to biologically active molecules: iron, biliverdin and carbon monoxide (CO). Like NO, CO is a short-lived gas that vasorelaxes
Normally, the cardiovascular system is regulated by specific nuclei in the brainstem and hypothalamus, including the nucleus of the solitary tract (NTS), the ventrolateral medulla, the supraoptic nucleus and the paraventricular nucleus. These areas serve as processing and integration centers for the afferent and efferent nervous traffic that directly controls cardiac and peripheral vascular tone. cfos is an immediate-early gene which has important roles in cellular signal transduction and transcriptional regulation.40 Immunohistochemical detection of the c-fos protein product, Fos, has been well established as a marker to identify neurons and neuronal pathways that have been activated by physiologic and pathophysiologic stimuli.40 Recent studies show that central neural activation through a c-fos-mediated pathway is a crucial initiating factor in the genesis of the hyperdynamic circulation in cirrhotic and portal hypertensive animal models.41–44 Fos immunoreactivity in several central nuclei precedes the development of the hyperdynamic circulation in a portal-hypertensive rat model41 (Figures 23-4, 23-5). Even more compelling is the finding that inhibition of Fos-mediated neuronal activation in the NTS by local microinjections of c-fos antisense oligonucleotides completely eliminates the hyperdynamic circulation in portal-hypertensive rats.41 Therefore, it appears that the activation of central cardiovascular regulatory nuclei through a c-fos-dependent pathway is necessary for development of the hyperdynamic circulation in portal-hypertensive rats. Several questions remain unanswered, including which central neurons are activated, as the obvious ones – sympathetic catecholaminergic neurons – proved to be only modestly activated43,46 (Figure 23-6). Perhaps more importantly, what is the cause of this central neural activation? The administration of capsaicin, which denervates sensory afferent nerves, significantly reduces the baseline central Fos expression in portal-hypertensive and cirrhotic rats,44 and also blocks the development of both ascites and the hyperdynamic circulation.45 In other words, the interruption of peripheral-to-central afferent signaling, presumably from the gut/liver to the cardiovascular regulatory nuclei, blocks central neu-
457
Number of Fos immunoreactive nuclei/ section
Section III. Clinical Consequences of Liver Disease
SHAM
180
Figure 23-4. Central neural activation and hemodynamics following portal vein stenosis (PVS) or sham operation in rats. Top panel: numbers of Fospositive neurons in paraventricular nucleus (PVN), supraoptic nucleus (SON), solitary tract nucleus (NTS) and ventrolateral medulla (VLM) at postoperative days 1, 3, 5, 10. Fos-positive neurons are significantly increased in all areas at all time points in the PVS group. A, Cardiac output; B, mean arterial pressure; C, systemic vascular resistance. *Significantly different from corresponding sham-control value. Note that central neuronal activation precedes development of hyperdynamic circulation. (Reproduced from reference 41, with permission.)
PVS
150 120 90 60 30 0
PVN SON NTS VLM
PVN SON NTS VLM
PVN SON NTS VLM
PVN SON NTS VLM
Day 1
Day 3
Day 5
Day 10
60 A
PVS
SHAM
*
ml/min/100 g
* 40
20
0 Day 1
Day 3
Day 5
Day 10
*
*
Day 5
Day 10
*
*
Day 5
Day 10
B
mmHg
120
80
40 Day 1
Day 3
8 mmHg/min/ml/100 g
C 6
4
2
0 Day 1
458
Day 3
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
A
B
C
D
Figure 23-5. Fos staining in paraventricular nucleus (PVN) and supraoptic nucleus (SON) of rats at day 1 after portal vein stenosis or sham operation. A, PVN of sham-control shows virtually no Fos staining neurons (small oval dark spots). B, PVN of portal hypertensive rat shows increased Fos staining. C, SON of shamcontrol shows scant Fos staining. D, SON of portal hypertensive rat shows intense Fos staining.
ronal activation and also normalizes the deranged hemodynamics. The exact origin of the peripheral signal remains unknown, but we speculate that it may be mesenteric venous congestion or distention due to portal hypertension. Hence, central cardiovascular dysregulation plays a crucial initiating role in the genesis of the hyperdynamic circulation in portal hypertension and cirrhosis.
Increased Blood and Plasma Volume That plasma and total blood volume are expanded due to renal sodium and water retention in cirrhosis is indisputable. However, many other issues about hypervolemia remain highly controversial, in particular the mechanism underlying the renal dysfunction. The debate about the ‘overload’ theory versus the ‘underfill’ or ‘peripheral vasodilatation’ theories has raged for more than three decades. Depending on the theory, the central blood volume – defined as the volume in the heart, aorta, and lungs – should be either expanded or low. Studies have reported both high and low central volume, and this point also remains very controversial.2,47 Further discussion of this topic (see Chapter 22) is beyond the scope of this chapter, but some aspects of hypervolemia merit discussion here.
In patients with cirrhosis there is evidence of volume-dependent effects on the systemic and splanchnic circulations. During acute volume expansion by head-out water immersion, hypervolemia and consequent natriuresis are more pronounced in well-compensated cirrhotics than in controls.48 In contrast, depletion of circulating volume after furosemide administration reduces the portal pressure and improves the hyperdynamic circulation.49 These observations indicate that significant volume expansion or depletion may regulate the hyperdynamic circulation in well-compensated cirrhosis. In other words, hypervolemia might be involved in the development of the hyperdynamic circulation in early cirrhosis. In this regard, a two-phase pathogenesis of the hyperdynamic circulation has been proposed.2 First, the initial event in early cirrhosis without ascites may be extracellular volume expansion due to subclinical sodium retention and subsequent passive relaxation of the vasculature in order to accommodate the increased volume. Sodium retention could be either due to a direct hepatorenal interaction acting through sinusoidal portal hypertension and/or hepatic dysfunction (overload theory), or secondary to peripheral vasodilation and hence decreased effective volume (peripheral vasodilatation theory). In
459
Section III. Clinical Consequences of Liver Disease
Figure 23-6. Dual-labeled immunofluorescence staining for tyrosine hydroxylase-containing neurons (green) and Fos-positive neurons (orange) in NTS of portal hypertensive rat. Cytoplasmic TH stain indicates a catecholaminergic neuron. Nuclear Fos stain indicates activated neuron. Note that most activated neurons are not TH-positive. Day 18 after portal vein stenosis surgery. Calibration bar = 25 mm.
the second phase, worsening vasodilatation secondary to hyporesponsiveness to vasoconstrictors, activation of neurohormonal systems such as the renin–angiotensin–aldosterone, sympathetic and endothelin systems, and increased portosystemic shunting may explain the greater severity of the hyperdynamic circulation in advanced stages of cirrhosis.2 If blood volume is increased, why are SVR and arterial pressure so low? Three possible reasons are: (1) tremendous arterial dilatation, (2) pooling of blood volume in the circulation, and (3) inadequate pump function (discussed in the next section). There is ample evidence for the first, as detailed earlier. The second also seems correct. The obvious site for volume pooling is the capacity reservoir of the body, the veins, especially in the splanchnic region. A Danish study using whole-body scintigraphy with indicator dilution methods confirmed that the expanded volume is indeed sequestered in the mesenteric veins.47,50 Whether an impaired ability to mobilize this venous reservoir contributes to circulatory dysfunction has not been formally examined, but seems plausible.
Tissue Hypoxia and Oxidative Stress Another adverse effect of the three factors above, or perhaps other mechanisms, is latent or overt tissue hypoxia. A significant number of patients have the hepatopulmonary syndrome with hypoxemia. Furthermore, even those without overt hypoxemia may suffer from a latent type of tissue hypoxia. Moreau and colleagues noted that patients with cirrhosis show a supply-dependent oxygen consumption.51,52 In patients with a relatively normal mean PaO2 of 92 mmHg, whole body oxygen delivery was changed by vasoactive drugs.52 When oxygen delivery was decreased, oxygen extraction increased to preserve oxygen uptake. However, when oxygen delivery was increased, oxygen consumption increased linearly without an apparent plateau.51 A normal non-hypoxic individual will not
460
consume any superfluous oxygen, whereas those in a hypoxic state, such as patients with septic shock, avidly consume extra delivered oxygen. Interestingly, patients with cirrhosis show the same delivery-dependent oxygen uptake as those with hypoxic shock syndromes. This suggests the existence of a latent or subclinical hypoxic state in these patients. The intriguing speculation that the hyperdynamic circulation is a compensatory reaction to this latent tissue hypoxia53 was tested by Moller and colleagues in 1996.54 The administration of 100% oxygen for 20 minutes to cirrhotic patients significantly decreased cardiac output by 11% and increased SVR by 15%. These results suggest that tissue hypoxia does not entirely explain the hyperdynamic circulation, but may contribute in part.54 A related issue is oxidative stress in cirrhosis. In addition to hypoxia, many other factors, such as bile acids, endotoxin, NO, sympathetic activation, angiotensin, inflammation, ischemia–reperfusion and cellular necrosis, can generate reactive oxygen and nitroxy intermediates such as O2- and ONOO-.55 There is little doubt about the presence of oxidative stress in liver disease, as many markers such as isoprostanes are elevated in serum and most tissues in cirrhosis.55 The key question is whether oxidative stress contributes to the genesis of the circulatory dysfunction. Moore and co-workers have shown that treatment of portal hypertensive rats with the antioxidants N-acetylcysteine and lipoic acid abrogates the hyperdynamic circulation.56,57 Although these data are suggestive, a definitive pathogenic role for oxygen and nitroxy intermediates is not yet confirmed.
COMPLICATIONS OF HYPERDYNAMIC CIRCULATION Autonomic Dysfunction The autonomic nervous system plays a critical role in modulating cardiac performance and vasomotor activity. Patients with cirrhosis
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
show an impairment of autonomic cardiovascular reflexes.58–60 The prevalence and severity of autonomic dysfunction increase with advancing hepatic dysfunction in cirrhosis.61 Up to 80% of patients with advanced cirrhosis show some evidence of autonomic dysfunction.61–63 The presence of vagal neuropathy has been reported to be a predictor of reduced survival in patients with compensated cirrhosis.59–61 An equal prevalence of autonomic neuropathy in alcoholic and non-alcoholic cirrhosis is observed, which has important implications with regard to pathogenesis.62 In all types of cirrhosis, the parasympathetic system seems to be more commonly affected than the sympathetic system. Vagal dysfunction is reported in 30–60% of patients, but only 10–20% show sympathetic dysfunction.59–61 Autonomic reflex function can be evaluated by standardized cardiovascular tests that quantify the changes in heart rate and blood pressure during a Valsalva maneuver, deep breathing, posture change, sustained hand grip and pharmacological stimulation.58–63 Baroreceptor function can also be quantified by using intravenous phenylephrine to rapidly increase blood pressure. The abrupt hypertension stimulates baroreceptors to transmit afferent signals to the brainstem cardiovascular centers, which then increase parasympathetic traffic via vagal efferents to reduce heart rate and contractility. This baroreflex function is known to be depressed in cirrhotic patients awaiting liver transplantation.59,60 It has been shown that autonomic dysfunction is associated with the hyperdynamic circulation.61,63 Because some patients without autonomic dysfunction have the hyperdynamic circulation, such dysfunction is not a prerequisite for the appearance of circulatory abnormalities. However, autonomic dysfunction may amplify the hyperdynamic circulatory abnormalities.2,63 Furthermore, the severity of the hyperdynamic circulation is correlated with the degree of autonomic dysfunction.2,3,63 Thus, autonomic dysfunction can contribute to circulatory abnormalities in several ways. Parasympathetic dysfunction reduces inhibitory vagal tone, thus allowing unfettered cardioacceleration. Moreover, defective sympathetic function may contribute to the blunted cardiovascular response to vasoconstrictors.2–5 Clinical consequences of baroreflex dysfunction include postural hypotension and diminished ability to maintain circulatory homeostasis following a cardiovascular insult. This may partially explain the relative fragility of patients with advanced cirrhosis to infection, hemorrhage, and volume depletion.
Cerebral Circulation Cerebral blood flow, measured by various methods, appears to be decreased in patients with cirrhosis.3,64–66 Moreover, virtually all studies have also documented regional redistribution of blood flow, generally from cortical to subcortical regions. Although the exact subcortical regions of such redistribution remain to be definitively determined, most studies have found that the basal ganglia, thalamus, and cerebellum enjoy more blood flow relative to cortical areas such as the frontal and temporal lobes.3,64–66 In part, disparities in the literature may reflect heterogeneity in patient populations. Alcohol is a well-known neurotoxin, and patients with alcoholic cirrhosis generally have more frontal cortical atrophy and hence decreased blood flow in that region than do non-alcoholics. The cerebral circulation is normally protected from changes in systemic hemodynamics, owing to a high degree of vascular autoregulation. For example, cerebral perfusion is unaffected by arterial
hypertension or hypotension within a broad range. Such autoregulation appears to be generally intact in patients with cirrhosis, but some studies have found defective autoregulation in those with severe liver failure or with hepatic encephalopathy.65,66 Even more controversial is whether the cerebral blood flow changes are related to functional neurological problems such as encephalopathy, abnormal cerebral ammonia metabolism or altered permeability of the blood–brain barrier. Evidence for and against these notions has been reported, and no clear consensus view is currently available.3,66
MANAGEMENT Many attempts to alter hyperdynamic circulation have been reported, although the aim of most such studies is to investigate pathogenesis, rather than treatment. Current pharmacotherapy focused on the hyperdynamic circulation includes drugs such as vasopressin, terlipressin, somatostatin, octreotide, and b-blockers such as propranolol and nadolol, all of which exert systemic and splanchnic effects.67,68 Because these drugs are administered for specific effects in the splanchnic circulation, such as control of variceal bleeding (Chapter 20), or to treat hepatorenal syndrome (Chapter 22), discussion here will be limited to other treatments relevant to the hyperdynamic circulation. As NO is a key mediator of circulatory dysfunction, this approach has been studied. Acute administration of the non-specific NOS inhibitor, N(G)-monomethyl-L-arginine (L-NMMA) corrected the altered systemic hemodynamics and improved renal function in one study,18 whereas in another arterial pressure increased but renal function remained unchanged.69 All studies with NOS inhibitors have been acute, and so chronic effects remain unknown. A theoretical concern of chronic NOS inhibitors is worsening of portal hypertension by aggravating the deficiency of NO in the hepatic microcirculation. However, a study by the Edinburgh group showed no effect of an acute L-NMMA dose on portal pressure in cirrhotic patients.70 The ideal therapeutic mode of manipulating NO would be a selective NOS blockade of a specific vascular bed or selective delivery of NO to hepatic microcirculation. NO-selective delivery by NOS gene transfer to the liver has been demonstrated in animal models, with promising results.71 The oral antibiotic norfloxacin has been used in cirrhotic patients in two recent studies to reduce bacterial endotoxin levels. In a Spanish study,72 42% of 71 patients showed high levels of lipopolysaccharide-binding protein (LBP), putatively a marker of endotoxin activation (although 60% of those with high LBP levels had normal endotoxin levels). Norfloxacin treatment for 4 weeks improved SVR and cardiac output only in the high-LBP subgroup (Figure 23-7). An Australian study treated 14 patients with alcoholic cirrhosis with norfloxacin, also for 4 weeks.73 These patients showed a decreased forearm blood flow and increased SVR and arterial pressure, associated with a significant decrease in endotoxin levels. In neither study did renal function change. These studies suggest an important role of gut bacteria in the genesis of the hyperdynamic circulation, and also highlight the need to confirm these results by larger randomized studies.
THE HEART IN CIRRHOSIS Despite all the attention focused on the peripheral circulation, the heart itself was relatively ignored until 1969. This was probably
461
Section III. Clinical Consequences of Liver Disease
Systemic vascular resistance 3500
3500
20
20
3000
3000
16
16
2500
2500
12
12
8
8
4
4
Norfloxacin
1000 p<0.01
500 Basal
2000
1500
1000 NS
0 Basal
2000
1500
p<0.01 0
dyn.sec.cm–5
24
dyn.sec.cm–5
24
μg/ml
μg/ml
LBP plasma levels
Placebo
NS 500
Basal
Norfloxacin
Basal
Placebo
Figure 23-7. Effect of norfloxacin treatment or placebo on LBP levels and systemic vascular resistance in patients with ascites. LBP, lipopolysaccharide-binding protein. (Reproduced from reference 72, with permission.)
because the increased basal cardiac output as part of the hyperdynamic circulation seemed to be undisputable evidence of normal contractile function. However, in that year, two pioneering studies found similar results. Regan and colleagues carefully studied cardiac contractile function under angiotensin stress in patients with alcoholic liver disease.74 With angiotensin infusion, left ventricular enddiastolic pressure increased dramatically, almost tripling from the basal value (Figure 23-8). This markedly increased pressure indicated a large increment in ventricular diastolic filling. A normal response to increased diastolic filling would be a significant augmentation of stroke volume. However, on average, stroke volume only rose slightly and even decreased in some patients, thus indicating a significantly attenuated ventricular contractile response. A few months later, Gould and colleagues subjected patients with alcoholic cirrhosis to exercise stress and noted a similar result: ventricular end-diastolic pressure rose but stroke work index remained unchanged.75 Five years later, Limas et al. again showed that patients with alcoholic cirrhosis had blunted contractile responsiveness to stimuli, and the short-acting cardiac glycoside ouabain was ineffective in increasing contractility.76 Over the next decade or so, many further studies confirmed attenuated cardiac responsiveness to physiological and pharmacological challenges in cirrhotic patients, a phenomenon ascribed by every study to the presence of mild or latent alcoholic cardiomyopathy.53,77–81 The idea that these blunted contractile responses were due to cirrhosis rather than alcohol remained unimaginable until the late 1980s. In 1986, Caramelo and colleagues infused physiological saline into rats with CCl4-induced cirrhosis and found an astounding 50% decrease in cardiac output.82 This landmark study was completely ignored by the scientific community at the time, probably because there was no plausible explanation for this unexpected result. In
462
1989, Lee53 proposed that the blunted cardiac responsiveness was not due to alcohol but to cirrhosis per se, an idea subsequently confirmed by numerous studies over the next decade.77–81 The term ‘cirrhotic cardiomyopathy’ was coined to describe this phenomenon. The dominant feature of this syndrome is normal or increased ventricular contractility with an attenuated cardiac response to stress. These stresses can include pharmacological stress, exercise, volume expansion and contraction, and procedures such as placement of transjugular intrahepatic portosystemic stent–shunts (TIPS) and liver transplantation.77–81 The following sections describe the clinical features and pathogenesis of this syndrome and other features of the heart in cirrhosis.
DIAGNOSIS OF CIRRHOTIC CARDIOMYOPATHY Because no widely accepted, specific diagnostic criteria yet exist, the exact prevalence of the syndrome remains unknown. An expert working group is formulating consensus recommendations about diagnostic criteria, with a preliminary report expected in 2006. In the absence of these consensus specific criteria, cirrhotic cardiomyopathy can be defined by the following general criteria: (1) normal or increased left ventricular systolic contractility at rest, but attenuated systolic or diastolic responses to stress stimuli; (2) electrophysiological abnormalities such as prolonged electrocardiographic QT interval; (3) structural or histologic changes in cardiac chambers; and (4) serum markers suggestive of cardiac stress. Not all features need be present to suspect the presence of the syndrome.
Systolic and Diastolic Function Ventricular contraction during systole can be quantified in several ways. These include measurement of stroke volume, end-diastolic and end-systolic volumes, and ejection fraction, which is calculated
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
55 Preangeotensin Values 50
Values during Angiotensin After-load Test
Figure 23-8. Effect of angiotensin infusion in patients with alcoholic cirrhosis. Small inset displays mean values for stroke work index and LVEDP. Thicker line is means of stroke volume index and LVEDP. Modest increase in SVI and SWI despite tripling of LVEDP indicates ventricular dysfunction. (Reproduced from reference 74, with permission.)
Stroke Volume Index ml/m²
45
40
35 75 30 SWI
50 25
25 20
0 0
5
0 0
5
10
15
20
10 15 20 LVEDP (mmHg)
25
Left Ventricular End-Diastolic Pressure (mmHg)
by the formula EF (%) = (stroke volume/end-diastolic volume) ¥ 100. Maximal or submaximal exercise provides the ‘purest’ stress test of cardiac function. Pharmacological manipulations such as angiotensin to ‘normalize’ the afterload have the inevitable drawback of possible direct drug effects on the heart. Gould et al.75 were the first to use exercise stress in patients with alcoholic cirrhosis. These authors found that with exercise, indices of left ventricular systolic function, such as the stroke index, mean systolic ejection rate, stroke-work and stroke-power, showed a markedly blunted response. Many years later, with more sophisticated methods available to study ventricular function, patients with alcoholic cirrhosis were examined by echocardiography and radionuclide angiocardiography.83 Whereas baseline ejection fractions in cirrhotic patients were similar to controls, with submaximal exercise these parameters increased 14% in controls but only 6% in cirrhotics. In addition, echocardiography demonstrated reduced ventricular wall compliance in the patients with cirrhosis, along with enlarged left atrial dimensions. In 1995, the maximal exercise responses in patients with both alcoholic and non-alcoholic cirrhosis was examined.84 Healthy controls increased their left ventricular ejection fractions and tripled their cardiac output from resting values, whereas the cirrhotic patients showed no change in ejection fraction and only doubled their cardiac output. Alcoholic and non-alcoholic patients showed virtually identical exercise responses. More recently, patients with all types of cirrhosis exhibited impaired systolic function with exercise.85 Again in this study, increments in maximal ejection fractions and cardiac outputs during exercise were limited in cirrhotic patients, with ascitic patients showing greater dysfunction than those without ascites (Figure 23-9).
Despite the aforementioned potential drawbacks of studying vasoactive drug responses, such studies have provided much useful information. The pioneering study of Regan and co-workers used angiotensin to increase afterload stress.74 Others normalized the peripheral vascular resistance by infusing angiotensin.76 This maneuver did not change cardiac output, despite a doubling of pulmonary capillary wedge pressure (a measurement of left ventricular enddiastolic pressure), indicating ventricular dysfunction. Since this study, many others have also shown similar systolic dysfunction under pharmacological stress, equally in alcoholic and non-alcoholic cirrhosis.53,77–81 Because the major function of diastole is to quickly refill the ventricle with blood, the factors that determine diastolic function are pressure, volumes, and mechanical forces in the ventricle. Normal diastolic refilling has both passive and active components. Hypertrophy causes the ventricle to become stiff and non-compliant, thus hindering passive elastic recoil and active relaxation in early diastole. Incomplete or slow refilling places added importance on the ‘atrial kick’ of late diastole, which contributes relatively little in the normal ventricle. Echocardiographically, these components can be measured as the E wave, representing the velocity of blood movement in early diastolic filling, and the A wave, the velocity of atrial contraction in late diastole. The E/A ratio is a commonly used parameter of diastolic function. The deceleration time of the E wave also is a measure of the ventricular compliance during the early rapid phase. In non-cirrhotic types of heart failure, such as ischemic or hypertensive cardiomyopathy, diastolic dysfunction usually precedes the more clinically overt systolic dysfunction. A similar chronology appears to be present in cirrhotic cardiomyopathy. In virtually every
463
Section III. Clinical Consequences of Liver Disease
Normal
900
*†
**
**
**
*
*
20
mm/sec
40
700
500 Controls
Pre-ascitics
Ascitics A
Ejection fraction
A
700
500
0
300
16
B
300
E/A ratio
14 12
1.5
DEC time 300
* *
** ** **
10 8
* **
6 4
**
1.0
2 0 Controls
Pre-ascitics
msec
Increase in EF (%)
**
900
60 mm/sec
HR at peak exercise (% of predicted values)
100 80
E wave
A wave
Heart rate
200
Ascitics
B
Cardiac Index
Increase in CI (%)
400
C
300
**
**
200 100 0
C
Controls
Pre-ascitics
Ascitics
Figure 23-9. Cardiac function with maximal exercise in controls and cirrhotic patients with and without ascites. A, Heart rate (HR) expressed as percentage of predicted value; B, increase in ejection fraction; C, increase in cardiac index. Significantly different from controls, *p <0.05, **p <0.01. †significant difference between preascitic and ascitic groups, p <0.05. (Reproduced from reference 85, with permission.)
study to examine diastolic function in cirrhosis, evidence of a stiff, hypertrophic left ventricle has been found, even at rest.77–81 The E/A ratio is invariably reduced, although some studies reported that this is due to increased A-wave velocity only, whereas others found a combination of both reduced E and increased A velocities78–81,85–87 (Figure 23-10). E-wave deceleration time is generally prolonged. In many of these studies indices of systolic function are unimpaired, even increased, and a strong stress is required to show some blunted systolic response. These results suggest that some element of diastolic dysfunction is present in most – if not all – patients with cirrhosis, whereas systolic dysfunction only manifests under significant stress, and then only in some individuals. Insertion of transjugular intrahepatic portosystemic stent–shunts (TIPS) represents a significant cardiac stressor as it shunts portal
464
0.5
D
100
Figure 23-10. Left ventricular function measured by echocardiography in patients with cirrhosis and controls. A, A wave, velocity of blood filling in late diastole; B, E wave, velocity of blood filling in early diastole; C, E/A ratio, ratio of early:late filling velocities; D, DEC, deceleration time of E-wave velocity. Green bars = controls; violet = all cirrhotic patients; orange = ascitic patients before paracentesis; turquoise = ascitic patients after paracentesis. Significantly different, * p <0.05, **p <0.01. (Reproduced from reference 86, with permission.)
venous volume to the right heart. Not unexpectedly, ventricular contractility, particularly diastolic function, can deteriorate following TIPS insertion.88–90 Overt left ventricular failure has been reported after TIPS insertion. Indeed, in a large randomized trial of TIPS versus large-volume paracentesis to treat ascites, overt congestive heart failure was precipitated in 12% of the TIPS patients compared to none of the paracentesis group.91 By a similar mechanism, the creation of a surgical portosystemic shunt in patients with cirrhosis also stresses ventricular function, allowing indices of contractile dysfunction to manifest.92
Cardiac Chamber Dimensions and Histology Cardiac chamber sizes can be increased volumetrically and by wall thickness. The previous concept that volume overload induces dilatation whereas pressure overload induces hypertrophy is now considered an oversimplification. An example is the cardiac chamber dimensions in cirrhosis, with its hyperdynamic circulation, a lowpressure, volume-overloaded state. There is universal agreement that the left atrium is enlarged, consistent with volume overload and the hyperdynamic circulation. Most studies in both cirrhotic
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
Electrophysiological Abnormalities The three manifestations of electrophysiological abnormalities in cirrhosis are QT prolongation, chronotropic incompetence, and asynchrony of electromechanical coupling. Since the almost incidental finding of prolonged electrocardiographic QT interval by Kowalski and Abelmann,1 case reports and series of cardiac arrhythmias in cirrhosis have appeared sporadically.53,77–81 Cases of atrial fibrillation and flutter, ectopic atrial and ventricular beats, and ventricular arrhythmias have been reported. As with much of the older literature in cirrhosis, interpretation of these results is enormously complicated because most cases had alcoholic cirrhosis, and so potentially arrhythmogenic effects of alcohol cannot be excluded. In other words, are these arrhythmias due to alcohol or to cirrhosis per se? The answer to this question remains elusive. QT prolongation was investigated further in 1993.95 These authors suggested that QT prolongation is a manifestation of autonomic dysfunction in patients with primary biliary cirrhosis. Subsequently, several more studies conclusively showed that the rate-corrected QT interval (QTc; normal up to 440 ms) is prolonged in approximately 30–60% of patients with cirrhosis of all etiologies.96–98 Current consensus opinion is that the prevalence of QT prolongation increases with advancing degrees of liver failure96–98 (Figure 23-11), and is reversible after liver transplantation.96–98 Delayed repolarization evidenced by QTc prolongation is thought to be a risk factor for a type of ventricular tachycardias called ‘torsades de pointes.’ However, whether patients with cirrhosis and QT prolongation are truly at risk of this tachycardia remains unclear. Although cases of torsades de pointes have been reported in cirrhotic patients, in every case other potential causes of torsades, such as electrolyte disturbance, shock, and drugs known to cause torsades, were present. Moreover, sudden
550
500
Q-Tc (msec)
patients and animal models have documented slight but significant left ventricular hypertrophy. Echocardiography in cirrhotic patients reveals increased wall thickness in the left ventricle and interventricular septum, as well as increased relative wall thickness, defined as LV wall plus septal thickness relative to the LV internal dimensions.84–87 Right heart chamber sizes remain controversial. The right ventricle has been reported to be smaller, normal or increased in size.78–81 Animal data are similar. In rats with biliary cirrhosis due to chronic bile duct ligation, the left ventricle shows eccentric hypertrophy whereas the right ventricle is normal.93 Antemortem cardiac tissue is very difficult to obtain in the cirrhotic patient. Although there are several large histological series in patients with cirrhosis, all are autopsy studies and almost all only included patients with alcoholic cirrhosis.53,78–80 These studies showed myocardial hypertrophy, subendocardial and myocyte edema, patchy fibrosis, exudation, nuclear vacuolation, and unusual pigmentation. Whether these changes truly reflect cirrhotic cardiomyopathy or alcoholic cardiotoxicity remains uncertain. The only study to date to examine antemortem endomyocardial and liver biopsies from the same patient was recently published.94 This retrospective series is limited by a very small sample size and possible referral/selection bias. Notwithstanding these drawbacks, however, the cardiac samples demonstrated patchy fibrosis and hypertrophy, generally similar to the autopsy studies. Further research is needed before firm conclusions can be drawn.
450
400
350
0 A
B
C
Controls
Figure 23-11. Electrocardiographic QTc prolongation in patients with cirrhosis (Child–Pugh classes A, B, C) and controls. Dotted line represents the accepted value for the normal limit (440 ms). (Reproduced from reference 96, with permission.)
cardiac death is not thought to be increased in patients with cirrhosis. Chronotropic incompetence refers to an inability of the cirrhotic heart to respond to stimuli with the appropriate degree of tachycardia. The clearest demonstration of this is the blunted tachycardiac response to exercise83,85 (see Figure 23-9), or vasoactive drug infusion in patients and animal models of cirrhosis. The most obvious clinical consequence of this phenomenon would be limited exercise capacity. Finally, investigators showed that in cirrhotic patients the time interval between the electrocardiographic and the actual mechanical onset of systole is significantly more variable than in controls99 (Figure 23-12). In other words, the normally tightly regulated time between the QRS complex and the actual onset of ventricular systole becomes abnormally shortened or prolonged. Most patients with prolonged QTc in their study also had prolonged electromechanical time intervals.99 The clinical significance of this curious phenomenon remains unknown.
Serum Markers Several serum markers, such as cardiac troponin I and the family of natriuretic peptides, appear to reflect myocardial strain or distress. These differ from traditional markers of necrosis/injury such as CKMB, in that cardiac muscle stretching or strain without necrosis is sufficient to increase levels. Atrial natriuretic peptide (ANP) is released mainly by the atria in response to stretch, and brain or Btype natriuretic peptide (BNP) by the ventricles. They are released as preprohormones and cleaved in the circulation. A cardiology consensus committee agreed that BNP and pro-BNP are excellent markers of a ventricle distressed by pressure/volume overload.100 Many disease states resulting in volume expansion increase levels of these natriuretic peptides, including renal and heart failure, and
465
Section III. Clinical Consequences of Liver Disease
Control 12 10
Number
8 6 4 2 0 –125 –100 –75 –50 –25
0
25
50
70 100 125 150 175
t (msec) Cirrhosis 12 10
Number
8 6 4 2 0 –125 –100 –75 –50 –25
0
25
50
70 100 125 150 175
t (msec) Figure 23-12. Increased dispersal of Dt (time between onset of electrical and mechanical systole) in patients with cirrhosis. (Reproduced from reference 99, with permission.)
cirrhosis. For troponin I, conditions leading to ventricular hypertrophy or dilatation increase levels. Therefore, some suggest that these markers in cirrhosis merely reflect the hyperdynamic circulation and volume overload rather than true cardiac distress. The available evidence, however, does not support that view. A French study documented increased serum troponin I levels in about one-third of cirrhotic patients.101 Elevated levels correlated with decreased ventricular stroke volume index, rather than any hemodynamic measure of the hyperdynamic circulation, portal hypertension, or stage of cirrhosis. In cirrhotic patients BNP levels correlated with left ventricular size and diastolic dysfunction,102 and both BNP and pro-BNP correlated with QT interval and degree of liver failure, rather than measures of hyperdynamic circulation such as cardiac output and peripheral vascular resistance.103,104 These results indicate the potential usefulness of these serum markers to diagnose myocardial distress in cirrhosis, but specific criteria, such as exact diagnostic cut-off values, remain to be determined.
COMPLICATIONS OF CIRRHOTIC CARDIOMYOPATHY A possible relationship between ventricular dysfunction, peripheral vasodilatation and sodium retention in cirrhosis has been postulated.
466
Regardless of whether one believes in the ‘primary peripheral vasodilatation’ or the ‘overload’ theory of salt and fluid retention, renal blood perfusion is dependent on the cardiovascular system. No matter what the status of the vessels or the fluid inside them, it is intuitive that the pump function must be adequate for proper renal perfusion. However, definitively ‘proving’ that depressed pump function contributes to the pathogenesis of renal sodium retention in cirrhosis is difficult. A recent study subjected patients with preascitic cirrhosis to a high-sodium diet for 7 days.105 As expected, these patients started retaining sodium and fluid during this challenge. Whereas healthy controls showed the normal cardiac response, increasing the slope of the peak systolic pressure to end-systolic volume relationship, about half the preascitic cirrhotic patients showed an abnormal inverse relationship, i.e. a reduction of systolic pressure with increasing end-systolic volume. In other words, cardiac contractility was already abnormal at baseline and worsened under a sodium load challenge, in conjunction with renal sodium retention.105 Circumstantially this suggests that inadequate ventricular contractile function may contribute to the renal salt and water retention in cirrhosis. Another recent intriguing study examined cardiovascular and renal function in 23 patients admitted with spontaneous bacterial peritonitis.106 All successfully resolved their infection with appropriate antibiotics. However, during the admission, eight patients subsequently developed hepatorenal syndrome (HRS), whereas 15 showed unimpaired renal function. In those who developed HRS, compared to those with normal renal function, cardiac output at admission was lower and declined even further with infection resolution, along with a drop in arterial pressure. In contrast, cardiac output did not change in those who had unimpaired kidneys. SVR, a measure of peripheral vascular tone, remained unchanged, indicating that the reduced cardiac output translated directly to decreased renal perfusion. These observations clearly imply, as detailed in an editorial,107 that inadequate ventricular contractility may contribute to the pathogenesis of hepatorenal syndrome associated with spontaneous bacterial peritonitis.
PATHOGENESIS OF CIRRHOTIC CARDIOMYOPATHY Owing to the difficulty of obtaining antemortem cardiac tissue or performing invasive research protocols in humans, most pathogenic studies have been done in animal models, particularly the bile ductligated (BDL) rat, which develops biliary cirrhosis 3–4 weeks after surgery. BDL rats show almost all the characteristics of human cirrhosis, including portal hypertension, ascites, jaundice, encephalopathy, and cardiovascular disturbances, including hyperdynamic circulation, hepatopulmonary syndrome and cirrhotic cardiomyopathy108–110 (Figure 23-13).
Cardiomyocyte Membrane Mechanisms Myocardial contractility is regulated primarily by the stimulatory b-adrenergic receptor system (Figure 23-14). When the badrenoceptor is occupied by a catecholamine ligand, it undergoes conformational change to couple with guanosine triphosphate (GTP)-binding protein (G protein) to activate adenylate cyclase to produce the second messenger cAMP. cAMP then stimulates protein
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
Contractile Response (% of basal)
150
120
90
60
30
0 10–10
10–9
10–8
10–7
10–6
[Isoproterenol] (mol/L)
A
cAMP (pmoles/mg protein/min)
30 DDL
SHAM
25
*
*
4
3
*
20 *
15 10 5 0 8
B
7
6
5
-LOG [Isoproterenol] (M)
Figure 23-13. Isoproterenol-stimulated dose–response effects. A, Isolated left ventricular papillary muscles. Squares = controls, circles = bile duct-ligated (BDL) rats. Maximal contractility (Rmax) in BDL muscles is significantly decreased. B, cAMP generation in cardiomyocyte plasma membranes from BDL rats and controls. *Significantly different from corresponding BDL value. These data indicate the presence of ventricular hyporesponsiveness to b-adrenergic stimulation. (A Reproduced from reference 112, with permission. B Reproduced from reference 109, with permission.)
kinase A-catalyzed phosphorylation of intracellular calciumregulatory proteins that eventually lead to calcium fluxes and cell contraction by actin-myosin cross-linking. The major G-protein type associated with the b-adrenergic receptor, Gs (G-stimulatory), is a heterotrimer with a, b and g subunits. The b-adrenoceptor–ligand complex couples with oligomeric Gs and eventually dissociates off, leaving the Gsa-GTP subunit, which then stimulates adenylate cyclase to generate cAMP. A physiological counterbalance is the muscarinic M2 receptor coupled to Gi, which inhibits adenylate cyclase. No direct human heart data are available, but Gerbes and colleagues examined lymphocyte b2-adrenoceptors in patients with cirrhosis.111 Generally, lymphocyte b2-adrenoceptors reflect the status of cardiac b-receptors (which are predominantly b1 subtype). They reported that lymphocyte b-adrenoceptor density is signifi-
cantly reduced only in a subgroup of patients with decompensated cirrhosis. In myocardial membranes from BDL and other rat models of cirrhosis, several defects in b-adrenergic receptor signaling leading to reduced cAMP-generating ability are present.108–110,112–114 (reviewed in 80–83). Compared to controls, b-adrenoceptor density,108,112 and the content and function of Gs protein112 are significantly reduced. The presence of jaundice impairs the activity of the adenylate cyclase enzyme and abnormal membrane biophysical properties affect badrenoceptor signaling.109,110,113 Cellular plasma membranes are composed of a lipid bilayer. Rather than a static arrangement of phospholipids, cholesterol, and other lipids lined up in a neat bilayer, both the lipids and the various protein receptors embedded in the membrane are in constant motion. Such motions, including lateral diffusion, rotation and wobbling, can be quantified by measuring the movement signals generated by fluorescent lipid probes inserted in the membrane. The biophysical term for this movement ability is membrane fluidity, and decreased fluidity implies restricted motional ability. Receptor proteins, when occupied by the appropriate ligand, undergo conformational change, hence normal membrane fluidity is required for such receptors to function properly. Membrane fluidity is determined predominantly by the lipid composition of the bilayer, especially the cholesterol:phospholipid ratio, with higher ratios becoming less fluid. Cardiomyocyte plasma membranes from BDL rats show increased cholesterol content and hence decreased fluidity109 (Figure 23-15). An in vitro membrane-fluidizing fatty acid analog, A2C, when incubated with isolated BDL cardiomyocyte membrane preparations, restores fluidity to values seen in normal membranes, and concomitantly restores the cAMP-generating ability.109 Ma et al. used A2C and other drugs in cirrhotic membranes to directly stimulate the b-adrenergic signaling pathway at different levels (receptor, G protein and adenylate cyclase) to identify the mechanism whereby decreased fluidity impairs b-adrenergic signaling function.113 The results suggest that in the cirrhotic cardiomyocyte decreased fluidity suppresses cAMP generation by hindering the coupling between the b-adrenoceptor–ligand complex and G protein, which is undoubtedly a conformation-dependent process. The inhibitory muscarinic cholinergic system has also been examined. In the cirrhotic rat heart, muscarinic M2 receptor density and binding affinity are unchanged, but overall muscarinic function is blunted.115 Membrane Gi protein content is dramatically reduced, explaining the attenuated muscarinic function. This attenuation is very likely a compensatory effort to balance the blunted stimulatory b-adrenergic system.
Cellular Calcium Kinetics Intracellular free calcium is the crucial element in contractility. The sources of Ca2+ include entry through plasma membrane calcium channels and release from intracellular stores in the sarcoplasmic reticulum (SR). SR stores of Ca2+ are rapidly released by signals such as depolarization of adjacent transverse tubules (invaginations of the sarcolemmal plasma membrane), free Ca2+ that has entered the cell through Ca2+ channels, or drugs such as caffeine or ryanodine (the SR calcium release channel is sometimes termed the ryanodine receptor). A significant reduction of the elicited inward current of the plasma membrane L-type calcium channel in BDL cardio-
467
Section III. Clinical Consequences of Liver Disease
HO bAR
AC NO
G protein a b g
Ca channel
+
+
+
-
+
cAMP (PKA)
-
CO
+
Figure 23-14. Regulation of cardiomyocyte contractility. BAR, b-adrenergic receptor; AC, adenylate cyclase; NO, nitric oxide; HO, heme oxygenase; CO, carbon monoxide; PKA, protein kinase A; PKG, protein kinase G; SR, sarcoplasmic reticulum; iCa, intracellular free calcium. + denotes stimulatory influence; - denotes inhibitory influence. (Reproduced from reference 117, with permission.)
cGMP
(PKG)
+
-
SR + iCa
Myocyte Contraction
intact.114 These results indicate that defects in calcium kinetics are not inside the cell, but are concentrated in the plasma membrane.
0.175 Anisotropy Parameter (rs)
BDL *
0.150
SHAM
*
cGMP-Mediated Mechanisms
0.125
0.100
0.075 0.05 0
5
10
15
20
Depth in Bilayer Figure 23-15. Dynamic component of membrane fluidity in cardiac sarcolemmal plasma membranes from BDL cirrhotic rat ventricles and shamoperated controls. Fluidity expressed as anisotropy parameter, rs is significantly decreased (higher rs) in BDL membranes at the superficial depths of the bilayer. (Reproduced from reference 109, with permission.)
myocytes, at baseline and also after isoproterenol stimulation, has been shown.114 Western blotting revealed that L-type calcium channel protein expression is decreased in BDL ventricles compared to controls.114 On the other hand, intracellular SR calcium handling is intact. Several methods to examine SR function and the structure of its component proteins showed no differences between BDL and control hearts. For example, Ca2+ reuptake and binding, and ryanodine receptor-binding characteristics are all unimpaired in the cirrhotic heart.114 mRNA and protein expression of the proteins controlling SR calcium reuptake and release, the SR Ca2+-ATPase pump (SERCA2) and ryanodine receptor, respectively, are similarly
468
NO helps regulate contractility by inhibiting certain b-adrenergic functions or calcium kinetics.116 The exact sites of NO action are still not settled, but evidence indicates that it may directly inhibit adenylate cyclase, or its intracellular messenger cGMP might inhibit L-type Ca2+ channels or the ryanodine receptor.116 Systemic or local cardiac overproduction of NO may also play a role in the cardiodepression of cirrhosis.117–119 Two studies have examined this issue. One group treated isolated working hearts from BDL rats with the NOS inhibitor NG-monomethyl-L-arginine and noted increased left ventricular systolic pressure and dP/dt, whereas control hearts were unaffected.118 Others documented increased iNOS mRNA and protein expression in BDL ventricles, but no change in eNOS mRNA and protein.119 Immunohistochemical staining localized the iNOS mainly to cardiomyocytes (Figure 23-16). In control hearts, the NO donor S-nitroso-acetyl-penicillamine reduced isolated LV papillary muscle contractility. Administration of the NOS inhibitor L-NAME normalized the attenuated papillary muscle contractility in cirrhotic rats, but did not affect control muscles.119 These studies therefore strongly suggest that NO generated by local cardiac iNOS up-regulation plays a role in the pathogenesis of contractile dysfunction in cirrhosis. It is interesting that eNOS, rather than iNOS, seems to be involved in the peripheral vasculature, whereas the opposite is found in the heart. This highlights the tremendous variability in pathophysiological mechanisms of NO in different organs and tissues. CO has been much less studied in cardiac physiology and pathophysiology. One group demonstrated increased HO-1 mRNA and protein expression in BDL hearts compared to controls.120 BDL left ventricular cGMP levels were also elevated, but these levels normalized by treatment with the HO inhibitor zinc protoporphyrin IX (ZnPP). Incubation of cirrhotic papillary muscles with ZnPP also
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR Figure 23-16. Immunohistochemical staining for iNOS (NOS2) in left ventricles of (A) sham-operated controls and (B) bile duct-ligated cirrhotic rats. No immunoprecipitate is visible in the control, whereas some cardiomyocytes stain positively in the cirrhotic ventricle. Magnification ¥ 50. (Reproduced from reference 119, with permission.)
A
B
469
Section III. Clinical Consequences of Liver Disease
restored the blunted contractile response to isoproterenol, but had no effect in control muscles.120 This data suggests a pathogenic role for the HO-CO-cGMP system, but further studies are needed before accepting this notion. A critical unanswered question is the possible interaction of the NO and CO systems in the cirrhotic heart. The relative magnitude of the cGMP generated by each gas also needs clarification, as NO is generally a significantly stronger generator of cGMP than of CO. Other possible pathogenic mechanisms, such as the TNF-a superfamily, endotoxins, endocannabinoids, and reactive oxygen and nitrogen intermediates, are currently under study.
Mechanism of Delayed Repolarization The underlying mechanism of the QT prolongation is suggested by animal studies. Ward et al. examined BDL rat ventricular myocytes and observed significant reduction of two types of K+ current, the Ca2+-independent transient outward K+ current, It, and the sustained delayed rectifier current (Isus).121 Such a reduction in K+ currents should prolong the action potential and hence the QT interval. However, this rat data should be extrapolated to humans only with caution, because of species variability in subtypes of myocardial K+ channel.
CORONARY ARTERY DISEASE IN CIRRHOSIS A pervasive view dating back to the early 20th century is that patients with cirrhosis are somehow protected from coronary atherosclerosis. Autopsy studies from the first half of the last century generally support this idea. Several theoretical considerations underpin this notion. Serum cholesterol levels are low in non-cholestatic cirrhosis. Moderate chronic alcohol intake may somehow decrease atherogenesis. Finally, a major risk factor for coronary disease, arterial hypertension, is distinctly uncommon once advanced cirrhosis supervenes. On the other hand, moderate or excessive alcohol consumption is a well-known risk for precipitating or aggravating hypertension, and type II diabetes, another coronary risk factor, is prevalent in cirrhosis. The literature suggests that in the early to middle stages of cirrhosis/liver dysfunction, very low-density lipoprotein levels are normal, whereas cardioprotective high-density lipoprotein levels are high; as liver failure progresses, LDL increases and HDL levels decline. Large epidemiological surveys indicate a U-shaped mortality curve for alcohol consumption: more coronary events in teetotallers than in moderate (10–30 g ethanol daily) drinkers, whereas heavy drinkers show excess mortality from accidents, liver disease, and certain cancers.122 Four studies have used the gold-standard test, coronary angiography, in patients with end-stage cirrhosis being evaluated for transplantation (reviewed in 123). Three studies showed a 3–6% prevalence of coronary atherosclerosis, whereas one reported a 27% prevalence of moderate or severe lesions. Limitations of these studies include a lack of non-cirrhotic control groups, referral and selection bias, and variability in the definitions of significant atherosclerosis. However, it is clear that coronary lesions are indeed present in a minority of patients with end-stage cirrhosis, dispelling the myth that cirrhosis is uniformly protective against ischemic
470
heart disease. The larger question of whether cirrhosis decreases coronary atherogenesis remains unsettled.
ENDOCARDITIS AND PERICARDITIS Patients with cirrhosis are immunosuppressed. Defective opsonization, phagocytosis, and bacterial killing in their immune cells have been demonstrated. A population-based Scandinavian study revealed a 10-fold increased risk of bacterial infections in patients with cirrhosis compared to healthy controls.124 It is unsurprising, then, to find a modestly increased risk of bacterial endocarditis in cirrhosis. One group reviewed the records of more than 41 000 autopsies in the Yale-affiliated hospitals, and found a threefold increase in endocarditis (0.1% in non-cirrhotic patients vs 0.34% in cirrhotics).125 Small pericardial effusions are very common in patients with ascites. These effusions are almost never clinically symptomatic, and presumably reflect the generalized edematous state of the patient with advanced cirrhosis.53
CARDIOVASCULAR FUNCTION AND LIVER TRANSPLANTATION Liver transplantation is arguably the lengthiest and most strenuous (to both patient and surgeon) routine surgical procedure. Accordingly, it places great stress on the cardiovascular system. Approximately 7–15% of early and late mortality after liver transplantation is due to cardiac causes.126–128 The peripheral vasodilatation of advanced cirrhosis usually improves quickly, increasing the ventricular afterload. In one study,129 by the third postoperative day systemic vascular resistance increased by 28% and cardiac output decreased 26%. Given the prevalence of diastolic dysfunction in end-stage cirrhosis, it is not surprising that pulmonary congestion and edema are common in the immediate postoperative period, affecting 47–56% of patients.126,127 Fortunately these episodes are often subclinical and transient. Clinically overt severe heart failure is much less common, afflicting about 1–3% of patients in the early postoperative period. In part, this relatively small percentage may be due to careful selection: patients with significant coronary disease or overt heart failure are excluded from transplantation. Later complications potentially include ischemic heart disease due to the antirejection drugs that induce hypertension and hyperlipidemia. Several studies of the incidence of ischemic heart disease after transplantation show conflicting results,128 so this question remains unresolved. Another controversial question is whether the hyperdynamic systemic and splanchnic circulations completely reverse after transplantation. Several studies have examined cardiac output, SVR and arterial pressure, as well as splanchnic measurements such as portal pressure and azygos venous blood flow (index of portosystemic collaterals). Some have found no or only minimal changes post transplantation, whereas others find complete or near-complete resolution. The contention that reversal is a gradual process, and therefore the discrepancy is due to the timing of the follow-up hemodynamic study, does not entirely explain the controversy. For example, Henderson and colleagues followed patients for up to 2 years after transplantation who remained hyperdynamic,130 whereas
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
others131,132 reported essentially complete normalization by 6 months. A regression analysis of factors affecting hyperdynamic circulation post transplantation demonstrated that infection, anemia, rejection, time after transplant and persistence of portosystemic shunts are significant factors.133
MANAGEMENT OF CIRRHOTIC CARDIOMYOPATHY It is difficult at present to make specific management recommendations in view of the lack of definitive diagnostic criteria for cirrhotic cardiomyopathy. Moreover, there are very few human studies on the treatment of any manifestations of this syndrome.134,135 Fortunately, in the absence of coexisting alcoholic cardiomyopathy, severe overt heart failure is uncommon. This is probably due to the peripheral vasodilatation of cirrhosis, which ‘unloads’ the left ventricle, as well as to a compensatory reduction in myocardial inhibitory systems such as the muscarinic system. Nevertheless, some management points should be noted. First, because a hallmark of this syndrome is attenuated ventricular response to stimuli, situations that significantly stress the heart should elicit extra vigilance by treating physicians. Such conditions include major surgery, hemorrhage, infection, rapid volume expansion, paracentesis, TIPS insertion, and vasoactive drug administration. Second, if the presence of cardiac dysfunction is uncertain, imaging studies such as MRI, radionuclides or echocardiography in the resting unstressed patient may not reveal any defects; a stress test such as exercise or drug challenge is needed. Third, if overt severe congestive failure appears, then the same general treatment principles of non-cirrhotic CHF still apply, with two important caveats. The first is that digitalis compounds may not be effective. The evidence for this is a study from 1974 that showed no effect of ouabain in patients with alcoholic cirrhosis.76 However, this conclusion is still tentative, as some patients in that study may have had alcoholic cardiomyopathy. The second caveat is that afterload reduction, a mainstay of CHF treatment in non-cirrhotic patients, should be done cautiously in cirrhotics. Many already have significant hypotension, and aggressively administering large doses of vasodilators may precipitate vascular collapse and renal shutdown. The treatment of severe CHF should include standard measures such as bed rest, salt restriction, diuretics, preload reduction, oxygen supplementation, and cautious use of vasodilators to reduce afterload. Animal studies in cirrhosis imply that inotropic-stimulating drugs based on b-adrenergic stimulation may not be effective, because of the multiple defects in the signaling pathway of this system. Scant available human data support this idea. Patients with cirrhosis show attenuated chronotropic or vascular responses to dobutamine and isoproterenol.53,77 Phosphodiesterase antagonists such as amrinone inhibit cAMP degradation, and thus might be useful because many of the b-adrenergic signaling defects are upstream of adenylate cyclase. Unfortunately, the only available study with this type of drug did not show efficacy. Cirrhotic patients undergoing partial hepatectomy for hepatocellar carcinoma were given amrinone in an attempt to decrease perioperative ischemia–reperfusion injury.136 This drug did not affect cardiac output or arterial pressure in these patients.
In non-cirrhotic heart failure, current consensus suggests blocking the potentially cardiotoxic effects of the sympathetic cardiac overdrive by drugs such as b-blockers. This somewhat paradoxical idea has been examined in one acute study in patients with cirrhosis. The Copenhagen group administered propranolol to patients with cirrhosis, some of whom had QT prolongation.137 This drug significantly reduced the prolonged QTc from a mean value of 460 ms to 440 ms. Whether chronic b-blockade helps remodel or protect the cirrhotic heart remains unstudied. The cardiac fibrogenic influence of the angiotensin–aldosterone system has recently been recognized. In this regard, Pozzi and colleagues reported that 24 weeks of chronic treatment with the aldosterone receptor antagonist potassium canrenoate in cirrhotic patients significantly reduced left ventricular wall thickness and peripheral sympathetic activation, and showed a non-significant tendency to improve the E/A ratio.138 It is possible that a longer period of drug administration might eventually have improved the diastolic dysfunction. This promising lead needs to be investigated further. Liver transplantation of course is the ultimate curative therapy for cirrhosis and virtually all its complications. In this regard, unlike the controversial reversal of the hyperdynamic circulation, transplantation eventually appears to completely reverse all aspects of cirrhotic cardiomyopathy (discussed in Natural history, below). But transplantation is not universally available, and some patients are not suitable candidates. Thus there is a pressing need for further research on the treatment of cirrhotic cardiomyopathy.
NATURAL HISTORY Even though each of the myriad cardiovascular abnormalities in cirrhosis is unique in many ways, they also share a few common features. One such commonality relates to the natural history. Specifically, the magnitude of the cardiovascular disorder is correlated to the degree of liver failure. In other words, patients with mild cirrhosis tend to have the mildest perturbation, whereas those with severe end-stage disease show the greatest derangement of cardiovascular function.2–6,53,77–81 So far this relationship appears to hold in every organ or tissue examined, including the brain, kidneys, lungs, gut and heart, as well as the overall hyperdynamic circulation. For example, cerebral blood flow is generally normal or only slightly decreased in the well-compensated patient, but decreases as liver function deteriorates. The extent of cirrhotic cardiomyopathy also tends to increase with decreasing liver function: ventricular contractile response to exercise or other stress progressively diminishes as cirrhosis worsens, and QT prolongation is accentuated as liver function deteriorates. Like other biological systems, however, the relationship between these parameters is not infinitely linear, i.e. a plateau effect is observed for cardiovascular changes. For example, the extent of peripheral vasodilatation reaches a plateau when the mean arterial pressure drops to the 55–60 mmHg range. No matter how severe the liver failure, mean arterial pressures below this range are rare in the cirrhotic patient (in the absence of acute problems such as bleeding or sepsis). Presumably the plateau effect represents the point at which the cardiovascular disturbance triggers new or previously inadequate compensatory mechanisms, thus allowing continued existence of the organism.
471
Section III. Clinical Consequences of Liver Disease
The relationship between circulatory change and liver failure continues to hold when liver function is improved. An example is the decrease in indices of hyperdynamic circulation in patients with alcoholic cirrhosis who stop drinking and gradually improve their liver function. However, the alternative explanation could be that this is due to relief from the many direct toxic effects of alcohol per se, rather than to any improvement in liver function. The strongest evidence in favor of the notion comes from studies of liver transplantation, the ultimate way to improve liver function. Many studies have examined various types of cardiovascular abnormality before and after transplantation, and although there are a few contrary reports, the weight of evidence indicates that most circulatory disturbances revert to normal or improve dramatically after transplantation. Hepatopulmonary syndrome, cerebral hypoperfusion, hepatorenal syndrome, splanchnic hyperemia and cirrhotic cardiomyopathy132 all normalize or disappear at some point after successful liver transplantation. Thus, transplantation appears to be the ultimate treatment for cardiovascular complications of cirrhosis.
CONCLUSION The hyperdynamic circulation is characterized by an increased cardiac output and decreased peripheral vascular resistance with low arterial pressure. Despite the increased baseline cardiac output, the ventricular systolic and diastolic response to stimuli is blunted, a condition termed cirrhotic cardiomyopathy. Other features include abnormal left heart chamber dimensions, electrophysiological abnormalities, and serum markers suggestive of ventricular strain. Both conditions exert widespread effects in different organs and vascular beds. Moreover, a complex interplay of multifactorial pathogenic mechanisms probably underlies these phenomena, including some factors common to both the hyperdynamic circulation and cirrhotic cardiomyopathy, such as NO and neurohumoral activation. Extensive research has improved our understanding of pathogenic mechanisms, thus allowing the possibility of novel treatment modalities. Future studies should focus on pharmacologic and genetic approaches to modulate cardiovascular regulatory systems, and thereby ameliorate complications related to the hyperdynamic circulation and cirrhotic cardiomyopathy.
ACKNOWLEDGMENTS Dr Lee is supported by a Senior Scholarship award from the Alberta Heritage Foundation for Medical Research. Dr Baik was supported by a sabbatical leave award from the Yonsei University Wonju College of Medicine, and Visiting Scholarship awards from the Korean Association for Study of Liver and GlaxoSmithKline Korea.
REFERENCES 1. Kowalski HJ, Abelmann WH. The cardiac output at rest in Laennec’s cirrhosis. J Clin Invest 1953;32:1025–1033. 2. Wong F, Blendis L. The hyperdynamic circulation in cirrhosis: an overview. Pharmacol Ther 2001;89:221–231. 3. Menon KV, Kamath PS. Regional and systemic hemodynamic disturbances in cirrhosis. Clin Liver Dis 2001;5:617–627.
472
4. Lebrec D, Moreau R. Pathogenesis of portal hypertension. Eur J Gastroenterol Hepatol 2001;13:309–311. 5. Garcia-Tsao G. Portal hypertension. Curr Opin Gastroenterol 2003;19:250–258. 6. Braillon A, Cales P, Valla D, et al. Influence of the degree of liver failure on systemic and splanchnic haemodynamics and on response to propranolol in patients with cirrhosis. Gut 1986;27:1204–1219. 7. Vorobioff J, Bredfeldt JE, Groszmann RJ. Hyperdynamic circulation in portal-hypertensive rat model: a primary factor for maintenance of chronic portal hypertension. Am J Physiol 1983;244:G52–G57. 8. Lee SS, Girod C, Valla D, et al. Effects of pentobarbital sodium on the splanchnic hemodynamics of normal and portal hypertensive rats. Am J Physiol 1985;249:G528–G532. 9. Goulis J, Armonis A, Patch D, et al. Bacterial infection is independently associated with failure to control bleeding in cirrhotic patients with gastrointestinal hemorrhage. Hepatology 1998;27:1207–1212. 10. Schrier RW, Arroyo V, Bernardi M, et al. Peripheral arterial vasodilation hypothesis: a proposal for the initiation of renal sodium and water retention in cirrhosis. Hepatology 1988;8:1151–1157. 11. Vallance P, Moncada S. Hyperdynamic circulation in cirrhosis: a role for nitric oxide? Lancet 1991;337:776–778. 12. Gonzalez-Abraldes J, Garcia-Pagan JC, Bosch J. Nitric oxide and portal hypertension. Metab Brain Dis 2002;17:311–324. 13. Wiest R, Groszmann RJ. The paradox of nitric oxide in cirrhosis and portal hypertension: too much, not enough. Hepatology 2002;35:478–491. 14. Chu CJ, Lee FY, Wang SS, et al. Hyperdynamic circulation of cirrhotic rats with ascites: role of endotoxin, tumour necrosis factor-a and nitric oxide. Clin Sci 1997;93:219–225. 15. Sarela AI, Mihaimeed FM, Batten JJ, et al. Hepatic and splanchnic nitric oxide activity in patients with cirrhosis. Gut 1999;44:749–753. 16. Niederberger M, Martin PY, Gines P, et al. Normalization of nitric oxide production corrects arterial vasodilation and hyperdynamic circulation in cirrhotic rats. Gastroenterology 1995;109:1624–1630. 17. Martin PY, Ohara M, Gines P, et al. Nitric oxide synthase (NOS) inhibition for one week improves renal sodium and water excretion in cirrhotic rats with ascites. J Clin Invest 1998;101:235–242. 18. La Villa G, Barletta G, Pantaleo P, et al. Hemodynamic, renal, and endocrine effects of acute inhibition of nitric oxide synthase in compensated cirrhosis. Hepatology 2001;34:19–27. 19. Fernandez M, Garcia-Pagan JC, Casadevall M, et al. Evidence against a role for inducible nitric oxide synthase in the hyperdynamic circulation of portal-hypertensive rats. Gastroenterology 1995;108:1487–1495. 20. Cahill PA, Redmond EM, Hodges R, et al. Increased endothelial nitric oxide synthase activity in the hyperemic vessels of portal hypertensive rats. J Hepatol 1996;25:370–378. 21. Shah V, Wiest R, Garcia-Cardena G, et al. Hsp90 regulation of endothelial nitric oxide synthase contributes to vascular control in portal hypertension. Am J Physiol 1999;277: G463–G468. 22. Pateron D, Tazi KA, Sogni P, et al. Role of aortic nitric oxide synthase 3 (eNOS) in the systemic vasodilation of portal hypertension. Gastroenterology 2000;119:196–200. 23. Shah V, Cao S, Hendrickson H, et al. Regulation of hepatic eNOS by caveolin and calmodulin after bile duct ligation in rats. Am J Physiol 2001;280:G1209–G1216. 24. Wiest R, Das S, Cadelina G, et al. Bacterial translocation in cirrhotic rats stimulates eNOS-derived NO production and impairs mesenteric vascular contractility. J Clin Invest 1999;104:1223–1233.
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
25. Garcia-Tsao G, Wiest R. Gut microflora in the pathogenesis of the complications of cirrhosis. Best Pract Res Clin Gastroenterol 2004;18:353–372. 26. Theodorakis NG, Wang YN, Skill NJ, et al. The role of nitric oxide synthase isoforms in extrahepatic portal hypertension: studies in gene-knockout mice. Gastroenterology 2003;124:1500–1508. 27. Iwakiri Y, Cadelina G, Sessa WC, Groszmann RJ. Mice with targeted deletion of eNOS develop hyperdynamic circulation associated with portal hypertension. Am J Physiol 2002;283:G1074–G1081. 28. Biecker E, Neef M, Sagesser H, et al. Nitric oxide synthase 1 is partly compensating for nitric oxide synthase 3 deficiency in nitric oxide synthase 3 knock-out mice and is elevated in murine and human cirrhosis. Liver Int 2004;24:345–353. 29. Xu L, Carter EP, Ohara M, et al. Neuronal nitric oxide synthase and systemic vasodilation in rats with cirrhosis. Am J Physiol 2000;279:F1110–F1115. 30. Batkai S, Jarai Z, Wagner JA, et al. Endocannabinoids acting at vascular CB1 receptors mediate the vasodilated state in advanced liver cirrhosis. Nature Med 2001;7:827–832. 31. Ros J, Claria J, To-Figueras J, et al. Endogenous cannabinoids: a new system involved in the homeostasis of arterial pressure in experimental cirrhosis in the rat. Gastroenterology 2002;122:85–93. 32. Fernandez-Rodriguez CM, Romero J, Petros TJ, et al. Circulating endogenous cannabinoid anandamide and portal, systemic and renal hemodynamics in cirrhosis. Liver Int 2004;24:477–485. 33. Fernandez M, Lambrecht RW, Bonkovsky HL. Increased heme oxygenase activity in splanchnic organs from portal hypertensive rats: role in modulating mesenteric vascular reactivity. J Hepatol 2001;34:812–817. 34. Moreau R. Heme oxygenase: protective enzyme or portal hypertensive molecule? J Hepatol 2001;34:936–939. 35. Makino N, Suematsu M, Sugiura Y, et al. Altered expression of heme oxygenase-1 in the livers of patients with portal hypertensive diseases. Hepatology 2001;33:32–42. 36. Guevara M, Gines P, Jimenez W, et al. Increased adrenomedullin levels in cirrhosis: relationship with hemodynamic abnormalities and vasoconstrictor systems. Gastroenterology 1998;114:336–343. 37. Lee SS, Chilton EL, Pak JM. Adenosine receptor blockade reduces splanchnic hyperemia in cirrhotic rats. Hepatology 1992;15:1107–1111. 38. Lee FY, Lin HC, Tsai YT, et al. Plasma substance P levels in patients with liver cirrhosis: relationship to systemic and portal hemodynamics. Am J Gastroenterol 1997;92:2080–2084. 39. Pak JM, Lee SS. Vasoactive effects of bile salts in cirrhotic rats: in vivo and in vitro studies. Hepatology 1993;18: 1175–1181. 40. Sheng M, Greenberg ME. The regulation and function of c-fos and other immediate early genes in the nervous system. Neuron 1990;4:477–485. 41. Song D, Liu H, Sharkey KA, Lee SS. Hyperdynamic circulation in portal-hypertensive rats is dependent on central c-fos gene expression. Hepatology 2002;35:159–166. 42. Breitman DR, Lee SS. Blunted responsiveness of the neuronal activation marker Fos in brainstem cardiovascular nuclei of cirrhotic rats. Hepatology 1997;26:1380–1385. 43. Song D, Sharkey KA, Breitman DR, et al. Disordered central cardiovascular regulation in portal hypertensive and cirrhotic rats. Am J Physiol 2001;280:G420–G430. 44. Li Y, Song D, Zhang Y, Lee SS. Effect of neonatal capsaicin treatment on haemodynamics and renal function in cirrhotic rats. Gut 2003;52: 293–299. 45. Lee SS, Sharkey KA. Capsaicin treatment blocks development of hyperkinetic circulation in portal hypertensive and cirrhotic rats. Am J Physiol 1993;264:G868–G873.
46. Fernandez B, Vatta M, Rodriguez-Fermepin M, et al. Norepinephrine uptake modifications in circumventricular organs, pons and myelencephalic areas and nuclei in prehepatic portal hypertensive rats. Neurochem Int 1999;34:177–181. 47. Henriksen JH. Volume adaptation in chronic liver disease: on the static and dynamic location of water, salt, protein and red cells in cirrhosis. Scand J Clin Lab Invest 2004;64:523–533. 48. Campbell PJ, Leung WM, Logan AG, et al. Hyperresponsiveness to water immersion in sodium retaining cirrhotics: the role of atrial natriuretic factor. Clin Invest Med 1988;11:392–395. 49. Cereda JM, Roulot D, Braillon A, et al. Reduction of portal pressure by acute administration of furosemide in patients with alcoholic cirrhosis. J Hepatol 1989;9:246–251. 50. Kiszka-Kanowitz M, Henriksen JH, Moller S, Bendtsen F. Blood volume distribution in patients with cirrhosis: aspects of the dual-head gamma-camera technique. J Hepatol 2001;35:605–612. 51. Moreau R, Lee SS, Soupison T, et al. Abnormal tissue oxygenation in patients with cirrhosis and liver failure. J Hepatol 1988;7:98–105. 52. Moreau R, Lee SS, Hadengue A, et al. Relation between oxygen transport and consumption during vasoactive drug administration in patients with cirrhosis. Hepatology 1989;9:427–432. 53. Lee SS. Cardiac abnormalities in liver cirrhosis. West J Med 1989;151:530–535. 54. Moller S, Becker U, Schifter S, et al. Effect of oxygen inhalation on systemic, central, and splanchnic haemodynamics in cirrhosis. J Hepatol 1996;5:316–328. 55. Bomzon A, Ljubuncic P. Oxidative stress and vascular smooth muscle cell function in liver disease. Pharmacol Ther 2001;89:295–308. 56. Fernando B, Marley R, Holt S, et al. N-acetylcysteine prevents development of the hyperdynamic circulation in the portal hypertensive rat. Hepatology 1998;28:689–694. 57. Marley R, Holt S, Fernando B, et al. Lipoic acid prevents development of the hyperdynamic circulation in anesthetized rats with biliary cirrhosis. Hepatology 1999;29:1358–1363. 58. Lunzer MR, Newman SP, Bernard AG, et al. Impaired cardiovascular responsiveness in liver disease. Lancet 1975;2:382–385. 59. Barron HV, Alam I, Lesh MD, et al. Autonomic nervous system tone measured by baroreflex sensitivity is depressed in patients with end-stage liver disease. Am J Gastroenterol 1999;94:986–989. 60. Trevisani F, Sica G, Mainqua P, et al. Autonomic dysfunction and hyperdynamic circulation in cirrhosis with ascites. Hepatology 1999;30:1387–1392. 61. Hendrickse MT, Thuluvath PJ, Triger DR. Natural history of autonomic neuropathy in chronic liver disease. Lancet 1992;339:1462–1464. 62. Oliver MI, Miralles R, Rubies-Prat J, et al. Autonomic dysfunction in patients with non-alcoholic chronic liver disease. J Hepatol 1997;26:1242–1248. 63. Bernardi M, Fornale L, Di Marco C, et al. Hyperdynamic circulation of advanced cirrhosis: a re-appraisal based on postureinduced changes in hemodynamics. J Hepatol 1995;22:309–318. 64. Spahr L, Burkhard PR, Grotzsch H, Hadengue A. Clinical significance of basal ganglia alterations at brain MRI and 1H MRS in cirrhosis and role in the pathogenesis of hepatic encephalopathy. Metab Brain Dis 2002;17:399–41. 65. Strauss GI, Hansen BA, Herzog T, Larsen FS. Cerebral autoregulation in patients with end-stage liver disease. Eur J Gastroenterol Hepatol 2000;12:767–771 66. Butterworth RF. Pathogenesis of hepatic encephalopathy: new insights from neuroimaging and molecular studies. J Hepatol 2003;39:278–285. 67. Reynaert H, Geerts A. Pharmacological rationale for the use of somatostatin and analogues in portal hypertension. Aliment Pharmacol Ther 2003;18:375–386.
473
Section III. Clinical Consequences of Liver Disease
68. Baik SK, Park DH, Kim MY, et al. Acute hemodynamic effects of octreotide and terlipressin in patients with cirrhosis: a randomized comparison. Am J Gastroenterol 2005:100:631–635. 69. Thiesson HC, Skott O, Jespersen B, et al. Nitric oxide synthase inhibition does not improve renal function in cirrhotic patients with ascites. Am J Gastroenterol 2003;98:180–186. 70. Forrest EH, Jones AL, Dillon JF, et al. The effect of nitric oxide synthase inhibition on portal pressure and azygos blood flow in patients with cirrhosis. J Hepatol 1995;23:254–258. 71. Yu Q, Shao R, Qian HS, et al. Gene transfer of the neuronal NO synthase isoform to cirrhotic rat liver ameliorates portal hypertension. J Clin Invest 2000;105:741–748. 72. Albillos A, de la Hera A, Gonzalez M, et al. Increased lipopolysaccharide binding protein in cirrhotic patients with marked immune and hemodynamic derangement. Hepatology 2003;37:208–213 73. Rasaratnam B, Kaye D, Jennings G, et al. The effect of selective intestinal decontamination on the hyperdynamic circulatory state in cirrhosis. A randomized trial. Ann Intern Med. 2003;139:186–193. 74. Regan TJ, Levinson GE, Oldewurtel HA, et al. Ventricular function in noncardiacs with alcoholic fatty liver: role of ethanol in the production of cardiomyopathy. J Clin Invest 1969;48:397–407. 75. Gould L, Shariff M, Zahir M, DiLieto M. Cardiac hemodynamics in alcoholic patients with chronic liver disease and a presystolic gallop. J Clin Invest 1969;48:860–868. 76. Limas CJ, Guiha NH, Lekagul O, Cohn JN. Impaired left ventricular function in alcoholic cirrhosis with ascites. Ineffectiveness of ouabain. Circulation 1974;49:754–760. 77. Ma Z, Lee SS. Cirrhotic cardiomyopathy: getting to the heart of the matter. Hepatology 1996;24:451–459. 78. Liu H, Lee SS. Cardiopulmonary dysfunction in cirrhosis. J Gastroenterol Hepatol 1999;14:600–608. 79. Blendis L, Wong F. Is there a cirrhotic cardiomyopathy? Am J Gastroenterol 2000;95:3026–3028. 80. Moller S, Henriksen JH. Cirrhotic cardiomyopathy: a pathophysiological review of circulatory dysfunction in liver disease. Heart 2002;87:9–15. 81. Baik SK, Lee SS. Cirrhotic cardiomyopathy: causes and consequences. J Gastroenterol Hepatol 2004;19(Suppl 1):S185–S190. 82. Caramelo C, Fernandes-Munoz D, Santos JC, et al. Effect of volume expansion on hemodynamics, capillary permeability, and renal function in conscious, cirrhotic rats. Hepatology 1986;6:129–134. 83. Kelbaek H, Eriksen J, Brynjolf I, et al. Cardiac performance in patients with asymptomatic alcoholic cirrhosis of the liver. Am J Cardiol 1984;54:852–855. 84. Grose RD, Nolan J, Dillon JF, et al. Exercise-induced left ventricular dysfunction in alcoholic and non-alcoholic cirrhosis. J Hepatol 1995;22:326–332. 85. Wong F, Girgrah N, Graba J, et al. The cardiac response to exercise in cirrhosis. Gut 2001;49:268–275. 86. Pozzi M, Carugo S, Boari G, et al. Evidence of functional and structural cardiac abnormalities in cirrhotic patients with and without ascites. Hepatology 1997;26:1131–1137. 87. Finucci G, Desideri A, Sacerdoti D, et al. Left ventricular diastolic function in liver cirrhosis. Scand J Gastroenterol 1996;31:279–284. 88. Wong F, Sniderman K, Liu P, et al. Transjugular intrahepatic portosystemic stent shunt: effects on hemodynamics and sodium homeostasis in cirrhosis and refractory ascites. Ann Intern Med 1995;122:816–822. 89. Huonker M, Schumacher YO, Ochs A, et al. Cardiac function and haemodynamics in alcoholic cirrhosis and effects of the transjugular intrahepatic portosystemic stent shunt. Gut 1999;44:743–748.
474
90. Merli M, Valeriano V, Funaro S, et al. Modifications of cardiac function in cirrhotic patients treated with transjugular intrahepatic portosystemic shunt (TIPS). Am J Gastroenterol 2002;97:142–148. 91. Gines P, Uriz J, Calahorra B, et al. Transjugular intrahepatic portosystemic shunting versus paracentesis plus albumin for refractory ascites in cirrhosis. Gastroenterology 2002;123:1839–1847. 92. Keller H, Bezjak V, Stegaru B, et al. Ventricular function in cirrhosis and portasystemic shunt: a two-dimensional echocardiographic study. Hepatology 1988;8:658–662. 93. Inserte J, Perello A, Agullo L, et al. Left ventricular hypertrophy in rats with biliary cirrhosis. Hepatology 2003;38:589–598. 94. Ocel JJ, Edwards WD, Tazelaar HD, et al. Heart and liver disease in 32 patients undergoing biopsy of both organs, with implications for heart or liver transplantation. Mayo Clin Proc 2004;79:492–501. 95. Kempler P, Szalay F, Varadi A, et al. Prolongation of the QTcinterval reflects the severity of autonomic neuropathy in primary biliary cirrhosis and in other non-alcoholic liver diseases. Zeitschr Gastroenterol 1993;31(Suppl 2):96–98. 96. Bernardi M, Calandra S, Colantoni A, et al. Q-T interval prolongation in cirrhosis: prevalence, relationship with severity, and etiology of the disease and possible pathogenetic factors. Hepatology 1998;27:28–34. 97. Finucci G, Lunardi F, Sacerdoti D, et al. Q-T interval prolongation in liver cirrhosis. Reversibility after orthotopic liver transplantation. Jpn Heart J 1998;39:321–329. 98. Bal JS, Thuluvath PJ. Prolongation of QTc interval: relationship with etiology and severity of liver disease, mortality and liver transplantation. Liver Int 2003;23:243–248. 99. Henriksen JH, Fuglsang S, Bendtsen F, et al. Dyssynchronous electrical and mechanical systole in patients with cirrhosis. J Hepatol 2002;36:513–520. 100. Silver MA, Maisel A, Yancy CW, et al. BNP Consensus Panel 2004: A clinical approach for the diagnostic, prognostic, screening, treatment monitoring, and therapeutic roles of natriuretic peptides in cardiovascular diseases. Congest Heart Fail 2004;10(Suppl 3):1–30. 101. Pateron D, Beyne P, Laperche T, et al. Elevated circulating cardiac troponin I in patients with cirrhosis. Hepatology. 1999;29:640–643. 102. Wong F, Siu S, Liu P, Blendis LM. Brain natriuretic peptide: is it a predictor of cardiomyopathy in cirrhosis? Clin Sci 2001;101:621–628. 103. Jimenez W, Arroyo V. Origins of cardiac dysfunction in cirrhosis. Gut 2003;52:1392–1394. 104. Henriksen JH, Gotze JP, Fuglsang S, et al. Increased circulating pro-brain natriuretic peptide (proBNP) and brain natriuretic peptide (BNP) in patients with cirrhosis: relation to cardiovascular dysfunction and severity of disease. Gut 2003;52:1511–1517. 105. Wong F, Liu P, Lilly L, et al. Role of cardiac structural and functional abnormalities in the pathogenesis of hyperdynamic circulation and renal sodium retention in cirrhosis. Clin Sci 1999;97:259–267. 106. Ruiz-del-Arbol L, Urman J, Fernandez J, et al. Systemic, renal, and hepatic hemodynamic derangement in cirrhotic patients with spontaneous bacterial peritonitis. Hepatology 2003;38:1210–1218. 107. Lee SS. Cardiac dysfunction in spontaneous bacterial peritonitis: a manifestation of cirrhotic cardiomyopathy? Hepatology 2003;38:1089–1091. 108. Lee SS, Marty J, Mantz J, et al. Desensitization of myocardial badrenergic receptors in cirrhotic rats. Hepatology 1990;12:481–485. 109. Ma Z, Meddings JB, Lee SS. Membrane physical properties determine cardiac beta-adrenergic receptor function in cirrhotic rats. Am J Physiol 1994;267:G87–93.
Chapter 23 CLINICAL CONSEQUENCES OF LIVER DISEASE: CARDIOVASCULAR
110. Ma Z, Zhang Y, Huet PM, Lee SS. Differential effects of jaundice and cirrhosis on b-adrenoceptor signaling in three rat models of cirrhotic cardiomyopathy. J Hepatol 1999;30:485–491. 111. Gerbes AL, Remien J, Jungst D, et al. Evidence for downregulation of beta-2-adrenoceptors in cirrhotic patients with severe ascites. Lancet 1986;1:1409–1411. 112. Ma Z, Miyamoto A, Lee SS. Role of altered b-adrenoceptor signal transduction in the pathogenesis of cirrhotic cardiomyopathy in rats. Gastroenterology 1996;110:1191–1198. 113. Ma Z, Lee SS, Meddings JB. Effects of altered cardiac membrane fluidity on b-adrenergic receptor signalling in rats with cirrhotic cardiomyopathy. J Hepatol 1997;26:904–912. 114. Ward CA, Liu H, Lee SS. Altered cellular calcium regulatory systems in a rat model of cirrhotic cardiomyopathy. Gastroenterology 2001;121:1209–1218. 115. Jaue DN, Ma Z, Lee SS. Cardiac muscarinic receptor function in rats with cirrhotic cardiomyopathy. Hepatology 1997;25:1361–1365. 116. Hare JM, Colucci WS. Role of nitric oxide in the regulation of myocardial function. Prog Cardiovasc Dis 1995;38:155–166. 117. Garcia-Estan J, Ortiz MC, Lee SS. Nitric oxide and renal and cardiac dysfunction in cirrhosis. Clin Sci 2002;102:213–222. 118. van Obbergh L, Vallieres Y, Blaise G. Cardiac modifications occurring in the ascitic rat with biliary cirrhosis are nitric oxide related. J Hepatol 1996;24:747–752. 119. Liu H, Ma Z, Lee SS. Contribution of nitric oxide to the pathogenesis of cirrhotic cardiomyopathy in bile duct-ligated rats. Gastroenterology 2000;118:937–944. 120. Liu H, Song D, Lee SS. Role of heme oxygenase–carbon monoxide pathway in pathogenesis of cirrhotic cardiomyopathy in the rat. Am J Physiol 2001;280:G68–74. 121. Ward CA, Ma Z, Lee SS, Giles WR. Potassium currents in atrial and ventricular myocytes from a rat model of cirrhosis. Am J Physiol 1997;273:G537–G544. 122. Sesso HD. Alcohol and cardiovascular health: recent findings. Am J Cardiovasc Drugs 2001;1:167–172 123. Plotkin JS, Johnson LB, Rustgi V, Kuo PC. Coronary artery disease and liver transplantation: the state of the art. Liver Transpl 2000;6(Suppl 1):S53–S56. 124. Thulstrup AM, Sorensen HT, Schonheyder HC, et al. Population-based study of the risk and short-term prognosis for bacteremia in patients with liver cirrhosis. Clin Infect Dis 2000;31:1357–1361.
125. Snyder N, Atterbury CE, Pinto Correia J, Conn HO. Increased concurrence of cirrhosis and bacterial endocarditis. A clinical and postmortem study. Gastroenterology 1977;73:1107–1113. 126. Myers RP, Lee SS. Cirrhotic cardiomyopathy and liver transplantation. Liver Transpl 2000;6(Suppl 1):S44–S52. 127. Therapondos G, Flapan AD, Plevris JN, Hayes PC. Cardiac morbidity and mortality related to orthotopic liver transplantation. Liver Transpl 2004;10:1441–1453. 128. Johnston SD, Morris JK, Cramb R, et al. Cardiovascular morbidity and mortality after orthotopic liver transplantation. Transplantation. 2002;73:901–906. 129. Nasraway SA, Klein RD, Spanier TB, et al. Hemodynamic correlates of outcome in patients undergoing orthotopic liver transplantation. Evidence for early postoperative myocardial depression. Chest 1995;107:218–224. 130. Henderson JM, Mackay GJ, Hooks M, et al. High cardiac output of advanced liver disease persists after orthotopic liver transplantation. Hepatology 1992;15:258–262. 131. Park SC, Beerman LB, Gartner JC, et al. Echocardiographic findings before and after liver transplantation. Am J Cardiol 1985;55:1373–1378. 132. Torregrosa M, Aguade S, Dos L, et al. Cardiac alterations in cirrhosis: reversibility after liver transplantation. J Hepatol 2005;42:68–74. 133. Gadano A, Hadengue A, Widmann JJ, et al. Hemodynamics after orthotopic liver transplantation: study of associated factors and long-term effects. Hepatology 1995;22:458–465. 134. Ma Z, Lee SS. Management of cirrhotic cardiomyopathy. In: Krawitt EL, ed. Medical management of liver disease. New York: Marcel Dekker, 1999: 583–589. 135. Liu H, Song D, Lee SS. Cirrhotic cardiomyopathy. Gastroenterol Clin Biol 2002;26:842–847. 136. Orii R, Sugawara Y, Hayashida M, et al. Effects of amrinone on ischaemia–reperfusion injury in cirrhotic patients undergoing hepatectomy: a comparative study with prostaglandin E1. Br J Anaesth 2000;85:389–395 137. Henriksen JH, Bendtsen F, Hansen EF, Moller S. Acute nonselective b-adrenergic blockade reduces prolonged frequencyadjusted Q-T interval (QTc) in patients with cirrhosis. J Hepatol 2004;40:239–246. 138. Pozzi M, Grassi G, Ratti L, et al. Cardiac, neuroadrenergic and portal hemodynamic effects of prolonged aldosterone blockade in postviral Child A cirrhosis. Am J Gastroenterol 2005;100:1110–1116.
475
Section III: Clinical Consequences of Liver Disease
24
PULMONARY COMPLICATIONS IN PATIENTS WITH LIVER DISEASE Michael B. Fallon and Miguel R. Arguedas Abbreviations CBDL common bile duct ligation CO carbon monoxide CO cardiac output CT computerized tomography CXR chest radiography DLCO diffusing capacity for carbon monoxide ET-1 endothelin-1
HO-1 HPS iNOS L-NAME MPAP PFT
heme oxygenase-1 hepatopulmonary syndrome inducible nitric oxide synthase NG-nitro-L-arginine methyl ester mean pulmonary artery pressure pulmonary function tests
INTRODUCTION Pulmonary symptoms and gas exchange abnormalities occur commonly in patients with chronic liver disease, and a variety of causes have been identified. These include intrinsic cardiopulmonary disorders as well as unique problems associated with the presence of liver disease and/or portal hypertension. Over the last two decades, two distinct pulmonary vascular disorders have emerged as important clinical complications in patients with liver disease: hepatopulmonary syndrome (HPS) and portopulmonary hypertension (POPH). This chapter will highlight the epidemiology, clinical features and management of these unique pulmonary vascular complications of liver disease.
THE SPECTRUM OF PULMONARY ABNORMALITIES IN LIVER DISEASE As many as 70% of patients with cirrhosis undergoing evaluation for liver transplantation complain of dyspnea if questioned.1 In addition, gas exchange abnormalities and abnormal pulmonary function test results occur in as many as 45–50% of patients.2 The common causes of pulmonary abnormalities in liver disease are outlined in Table 241. The most common causes of pulmonary abnormalities are chronic obstructive pulmonary disease and congestive heart failure, as in patients without liver disease. In patients with cirrhosis and portal hypertension, the presence of ascites and/or hepatic hydrothorax can result in pulmonary restriction (Chapter 19). In addition, the deconditioning and muscle wasting associated with advanced liver disease may also cause dyspnea. In a small subset of patients specific liver diseases are associated with unique pulmonary parenchymal abnormalities, including pulmonary granulomas or fibrosing alveolitis in primary biliary cirrhosis (Chapter 41) and panacinar emphysema in a1-antitrypsin deficiency (Chapter 68). Finally, unique pulmonary vascular complications, including microvascular
POPH PVR TIPS TNF
portopulmonary hypertension pulmonary syndrome pulmonary vascular resistance transjugular intrahepatic portosystemic shunt tumor necrosis factor
dilatation in HPS and vasoconstriction and remodeling in resistance vessels in POPH, may develop in as many as 20–30% of patients with liver disease.
HEPATOPULMONARY SYNDROME DEFINITION HPS results from intrapulmonary microvascular dilatation that develops in a subgroup of patients with liver disease and/or portal hypertension. It is commonly defined by the presence of hepatic dysfunction or portal hypertension, a widened age-corrected alveolar–arterial oxygen gradient on room air with or without hypoxemia, and intrapulmonary vasodilatation.3,4 Although the association between pulmonary dysfunction and liver disease has been recognized for over 100 years,5 the term ‘hepatopulmonary syndrome’ was not used until 19776 as the concept emerged that intrapulmonary vasodilatation caused the gas exchange abnormalities in these patients. Currently, studies demonstrate that as many as 40% of cirrhotic patients have detectable intrapulmonary vasodilatation,7 and that up to 10–20% will develop impaired oxygenation leading to significant functional limitations.3 Early definitions emphasized the need to exclude intrinsic cardiopulmonary disease or hepatic hydrothorax in order to make the diagnosis of HPS.4 However, it is now clear that HPS may occur in the setting of other cardiopulmonary abnormalities8,9 and contribute significantly to gas exchange abnormalities in these patients.
EPIDEMIOLOGY HPS is most commonly diagnosed in subjects with cirrhosis and portal hypertension. However, no specific etiology of cirrhosis has been found to increase the risk of developing HPS. Also, conflicting reports regarding whether the presence or severity of HPS correlate with the degree of hepatic synthetic dysfunction and portal hypertension have been published.7,8,10–13 Recently, the spectrum of
477
Section III. Clinical Consequences of Liver Disease
Table 24-1. Pulmonary Abnormalities in Chronic Liver Disease Intrinsic cardiopulmonary disease Chronic obstructive pulmonary disease Interstitial lung disease Congestive heart failure Pneumonia Asthma Related to liver disease Associated with specific liver diseases Panacinar emphysema (a1-antitrypsin deficiency) Fibrosing alveolitis, pulmonary granulomas (primary biliary cirrhosis Associated with complications from cirrhosis and/or portal hypertension Ascites Hepatic hydrothorax Muscular wasting/debilitation Specific pulmonary vascular abnormalities Hepatopulmonary syndrome Portopulmonary hypertension
hepatic abnormalities associated with the development of HPS has broadened to include portal hypertension without cirrhosis (prehepatic portal hypertension, nodular regenerative hyperplasia, congenital hepatic fibrosis, and hepatic venous outflow obstruction14–17) and hepatic dysfunction in the absence of established portal hypertension (acute and chronic hepatitis18,19). A single case of HPS has been reported in a patient with metastatic carcinoid, normal liver function and no evidence of portal hypertension,20 suggesting that tumor-derived vasoactive substances may have triggered intrapulmonary vasodilatation. Finally, a syndrome similar to HPS has been found in children with congenital cardiovascular abnormalities that result in altered hepatic venous drainage to the lungs.21,22 This observation supports the belief that factors normally produced or metabolized in the liver modulate pulmonary vascular tone.
PATHOLOGY AND PATHOGENESIS The key underlying structural alteration in HPS is dilatation of the pre- and postcapillary pulmonary vasculature,23 which leads to impaired oxygenation of venous blood as it passes through the lung.24,25 The pathogenesis of these vascular changes is an area of active investigation. One fundamental question regarding the pathogenesis of HPS is whether the mechanisms of pulmonary vasodilatation are similar to those of the splanchnic and systemic vasodilatation that develops in cirrhosis. The recent recognition that HPS may develop in the absence of cirrhosis, and possibly in the absence of portal hypertension, supports the suggestion that alterations in the pulmonary microvasculature may be distinct from those occurring in the systemic and splanchnic vasculature in cirrhosis. In humans, increased pulmonary production of nitric oxide (NO) is one contributor to intrapulmonary vasodilatation. Exhaled NO levels are increased in cirrhotic HPS patients and normalize after OLT26–28 as HPS resolves. In addition, acute inhibition of NO production or action with NG-nitro-L-arginine methyl ester (L-NAME) or methylene blue, respectively, transiently improve HPS.29–31 However, the mechanisms of increased endogenous NO production in human HPS and its relationship to the presence of portal hypertension, the
478
hyperdynamic circulation, and the degree of liver injury, remain uncertain. In addition, whether other mediators contribute to intrapulmonary vasodilatation has not yet been studied. The only established experimental model of HPS is chronic common bile duct ligation (CBDL) in the rat.32,33 CBDL appears to be unique among rodent models of cirrhosis and/or portal hypertension in that other commonly used models, such as thioacetamideinduced cirrhosis and partial portal vein ligation, do not result in the development of HPS.34 Increased pulmonary production of NO also appears to be a central event in the development of experimental HPS.35–38 In this situation, increased pulmonary vascular expression and the activity of endothelial nitric oxide synthase (eNOS) is a major source of pulmonary NO production.35,36 The increase in eNOS appears to derive from a combination of enhanced hepatic production and release of endothelin-1 (ET-1) and a shear-mediated increase in pulmonary endothelial endothelin B receptor (ETB) expression.39–41 These events lead to enhanced ET-1 activation of eNOS through the ETB receptor. As experimental HPS progresses, intravascular macrophages accumulate and produce inducible nitric oxide synthase (iNOS)35,37,38 and heme oxygenase-1 (HO-1).37,42 These events contribute to vasodilatation through the production of iNOS-derived NO and HO-1-derived carbon monoxide (CO). Both increased tumor necrosis factor-a (TNF-a) production and ET-1 itself appear to contribute to endothelial alterations and macrophage accumulation and activation.41,43 Accordingly, in CBDL animals both TNF-a inhibition with intestinal decontamination and pentoxifylline administration and the use of selective ETB receptor antagonists ameliorate HPS.41,43,44 Figure 24-1 summarizes the current understanding of mechanisms of pulmonary microvascular dilatation in experimental HPS. However, whether similar mechanisms are operative in human disease is unknown.
CLINICAL FEATURES The majority of patients with HPS are recognized to have the insidious onset of dyspnea (Table 24-2). In preliminary work, 95% of patients in one cohort, found to have HPS during screening, complained of dyspnea.1 Classically, an increase in dyspnea with standing (platypnea) has been described in HPS, attributed to the predominance of vasodilatation in the lung bases and the increased blood flow through these regions when upright.45 However, the frequency or usefulness of this observation in the diagnosis of HPS remains undefined. Several other clinical signs, including spider angiomata, digital clubbing and cyanosis, are also commonly described in subjects with HPS but also have not been prospectively evaluated as diagnostic indicators. In addition, because respiratory symptoms are common and may coexist with poor physical condition, smoking, ascites and/or intrinsic lung disease in cirrhosis, the diagnosis of HPS may be delayed and identified only after severe arterial hypoxemia has ensued. Chest radiography (CXR) and pulmonary function tests (PFT) are often performed to evaluate dyspnea. In HPS the CXR is most commonly normal, but may reveal lower lobe interstitial changes that may be confused with pulmonary fibrosis.46 PFTs typically demonstrate well-preserved spirometry and lung volumes in HPS. However, the diffusing capacity for carbon monoxide (DLCO) is often significantly reduced and may suggest the diagnosis. Unfortunately, the DLCO is also commonly decreased
Chapter 24 PULMONARY COMPLICATIONS IN PATIENTS WITH LIVER DISEASE
Hepatic injury Portal hypertension
Liver
Endothelin-1 Bacterial translocation/TNF a Sheer stress Other
Lung
HPS
POPH Lumen ? Inflammation ? Genetic factors
iNOS HO-1
ETB receptor eNOS
Injury NO CO
Macrophage
Figure 24-1. Potential mechanisms of pulmonary vascular alterations in HPS and POPH. Hepatic injury and/or portal hypertension influence the production and release of vasoactive mediators and cytokines and modulate vascular shear stress. In experimental HPS, pulmonary vasodilatation results when endothelin-1, produced in the liver, is released into the circulation and stimulates pulmonary vascular endothelial nitric oxide synthasederived nitric oxide (NO) production through an increased number of endothelin B (ETB) receptors. Macrophages also accumulate in the vascular lumen and produce NO from inducible nitric oxide synthase (iNOS) and carbon monoxide (CO) from heme oxygenase-1 (HO-1), contributing to vasodilatation. In POPH, similar events, possibly modified by genetic factors and the inflammatory response, may cause endothelial injury, resulting in smooth muscle proliferation and vascular remodeling.
Endothelial cell Smooth muscle cell
NO
Vasoconstriction Remodeling
Vasodilation
Table 24-2. Characteristics of HPS and POPH Hepatopulmonary syndrome
Portopulmonary hypertension
Key features: Intrapulmonary vasodilatation Develops with hepatic synthetic dysfunction and/or portal hypertension Present in 8–20% of patients with cirrhosis Liver transplantation generally curative Symptoms: May be asymptomatic Dyspnea Platypnea Signs: Spider angiomata Digital clubbing Cyanosis Arterial blood gases: Widened AaPO2 common Hypoxemia common Pulmonary function tests: Decreased DLCO Chest radiography: Normal or basilar interstitial changes
Key features: Intrapulmonary vasoconstriction and arterial remodeling Develops in the setting of portal hypertension Present in 3–12% of patients with advanced liver disease Liver transplantation generally contraindicated Symptoms: Often asymptomatic Dyspnea (most common) Chest pain Syncope Signs: Jugular distention Accentuated P2 Tricuspid regurgitation murmur Anasarca Arterial blood gases: Widened AaPO2 uncommon Hypoxemia uncommon Pulmonary function tests: Normal Chest radiography: Normal or prominent PA/right heart chambers
in cirrhosis in the absence of HPS, and the diagnostic utility of a reduced value is not established.47
DIAGNOSIS A high index of suspicion is important to identify HPS in patients with chronic liver disease and/or portal hypertension. An overview of the evaluation for HPS and POPH is outlined in Figure 24-2. The threshold for pursuing the diagnosis may be influenced by a number of factors, including the presence of specific signs and symptoms, risk factors for intrinsic cardiopulmonary disease, and whether liver transplantation is being considered. Specifically, evaluation is appropriate in all patients complaining of dyspnea and/or displaying clubbing or cyanosis. In patients with specific risk factors (smoking and other cardiovascular risk factors, occupational exposure, liver diseases associated with intrinsic lung disease) assessment of the presence and severity of intrinsic cardiopulmonary disease is crucial. In patients being considered for liver transplantation, regardless of the presence of symptoms, screening is important. In this latter group it is particularly important to diagnose and differentiate HPS and POPH, given that the presence of these disorders may influence transplant candidacy and priority. The diagnosis of HPS rests on documenting the presence of arterial gas exchange abnormalities resulting from intrapulmonary vasodilatation in the appropriate clinical setting. Gas exchange abnormalities are generally detected by arterial blood gas measurements and have been defined as a widened alveolar–arterial oxygen gradient (AaPO2, >5–20 mmHg) with or without hypoxemia (PaO2 <70 mmHg) in HPS.4 Although including mild gas exchange abnor-
479
Section III. Clinical Consequences of Liver Disease
Liver Disease Portal Hypertension
OLT Evaluation Specific signs and symptoms Risk factors for cardiopulmonary disease SpO2 <97%
History/Physical Exam Arterial Blood Gases Pulmonary Function Tests Imaging
Contrast Doppler Echocardiography (CE)
+ CE Abnl ABG
NL CardioPulmonary eval
Abnl CardioPulmonary eval
+ CE NL ABG
– CE Abnl ABG
– CE NL ABG
PAS >40–45 +/– Abnl RV
LV Dysfunction
PAS <40
? Preclinical HPS
Intrinsic CardioPulmonary Dz
Observe
RHC
Treat
Observe
MAA
+
-
HPS
Intrinsic Lung Dz
Normal
MPAP NL PCWP PVR
MPAP Abnl PCWP NL PVR
Observe
POPH
Volume Overload
Figure 24-2. Diagnostic approach to HPS and POPH. See text for details. OLT, orthotopic liver transplantation; ABG, arterial blood gases; SpO2, pulse oximetry oxygen saturation; PAS, pulmonary artery systolic pressure; RV, right ventricle; LV, left ventricle; RHC, right heart catheterization; MAA, technetium-labeled macroaggregated albumin scan; MPAP, mean pulmonary arterial pressure; PCWP, pulmonary capillary wedge pressure; PVR, pulmonary vascular resistance.
malities in the definition of HPS maximizes the potential for detecting ‘early’ disease, the clinical importance and specificity of mild arterial blood gas changes are not well defined.48 In addition, the AaPO2 normally widens with age. Therefore, the AaPO2 should be corrected for age (normal = 10 + 0.43 (age – 20)) to avoid an overestimation of the prevalence HPS. Obtaining arterial blood gases in the sitting position may enhance the detection of arterial deoxygenation in HPS owing to the predominance of vasodilatation in the lower lung fields. Pulse oximetry is an alternative non-invasive screening modality that indirectly measures oxygen saturation and can screen for arterial hypoxemia. It may be useful to target the diagnostic evaluation for HPS to patients who have a higher likelihood of relatively advanced disease (PaO2 < 70 mmHg).49 However, the threshold oximetry value for detecting hypoxemia in cirrhosis (£ 97%) is higher than typically expected, and oximetry alone would not detect less advanced disease (normal PaO2, widened AaPO2).49,50 Intrapulmonary vasodilatation can be detected using contrast echocardiography, lung perfusion scanning, pulmonary angiography,
480
and high-resolution chest CT scanning. Two-dimensional transthoracic contrast echocardiography is the most sensitive and most commonly employed screening technique. Typically, agitated saline is used as a contrast agent because the microbubbles are visible on echocardiography. A positive test for intrapulmonary vasodilatation occurs when late visualization (third heartbeat after injection) of intravenously administered microbubbles is observed in the left cardiac chambers51,52 (Figure 24-3). Immediate visualization of injected contrast in the left heart indicates intracardiac shunting. Transesophageal contrast echocardiography may increase the sensitivity of detecting intrapulmonary vasodilatation compared to transthoracic echocardiography, but is invasive and more expensive.51,52 Echocardiography can also assess cardiac function and estimate pulmonary arterial systolic pressure, and is useful for screening for cardiac dysfunction and POPH. As many as 40% of patients with cirrhosis and normal arterial blood gases may have a positive contrast echocardiogram, suggesting that mild intrapulmonary vasodilatation insufficient to alter gas exchange is common in cirrhotics.7
Chapter 24 PULMONARY COMPLICATIONS IN PATIENTS WITH LIVER DISEASE
A
B
Also, a positive result on contrast echocardiography in a hypoxemic patient with a concomitant pulmonary disorder (pleural effusion, chronic obstructive pulmonary disease) is insufficient to establish the diagnosis of HPS, because either intrapulmonary vasodilatation or the underlying pulmonary process could be responsible for gas exchange abnormalities. Radionuclide lung perfusion scanning using technetium-labeled macroaggregated albumin particles (99mTcMAA scan) is another method for detecting intrapulmonary vasodilatation (Figure 24-4). In this test, macroaggregated albumin particles 50–100 mm in size are injected intravenously. Normally, all particles are trapped in the lung microvasculature. In HPS, some particles escape through dilated capillaries and lodge in downstream capillary beds. Quantitative imaging of the lung and brain using a standardized methodology allows the calculation of a shunt fraction.8 The 99mTcMAA scan offers one significant advantage over contrast echocardiography: a positive scan (shunt fraction >6%) is specific for the presence of HPS even in the setting of coexisting intrinsic lung disease.8 In addition, it can be used to quantify intrapulmonary shunting and is useful for following the progression and/or resolution of disease prospectively. However, as a screening test, 99mTcMAA scanning is less sensitive than contrast echocardiography in detecting intrapulmonary vasodilatation, and cannot evaluate cardiac function or intracardiac shunting, or estimate pulmonary artery pressure. Pulmonary angiography is an invasive and insensitive diagnostic modality for detecting intrapulmonary vasodilatation in HPS53 and is not useful as a screening test. Two types of angiographic finding have been reported: type 1, a diffuse ‘spongiform’ appearance of pulmonary vessels during the arterial phase; and type 2, small discrete arteriovenous communications. However, many patients have a normal angiogram in the setting of clinically significant HPS. Angiography may have a therapeutic role in the rare HPS patient with a poor response to 100% oxygen (arbitrarily defined as a PO2 <150 mmHg) and anatomic shunting due to type 2 lesions that may be amenable to embolization.54 A recent study has demonstrated that high-resolution chest computerized tomography (CT) may be a less invasive radiologic method to detect dilated pulmonary vessels in the lung in HPS.55 The degree of dilatation observed on CT correlated with the severity of gas exchange abnormalities in these patients with HPS, suggesting that quantification of intrapulmonary vasodilatation was possible. Further studies are warranted to define whether chest CT will be useful in assessing the presence and severity of HPS.
C
Figure 24-3. Contrast echocardiogram for detecting intrapulmonary vasodilatation. (A) Parasternal four-chamber view of the heart (RA, right atrium; RV, right ventricle; LA, left atrium; LV, left ventricle) prior to the administration of agitated saline contrast. (B) Four-chamber view immediately after the administration of contrast, demonstrating the presence of echogenic microbubbles in the right atrium and ventricle immediately after injection of agitated saline into the antecubital vein. (C) Visualization of echogenic microbubbles in the left atrium and ventricle three cardiac cycles after visualization on the right due to intrapulmonary vasodilatation in a patient with hepatopulmonary syndrome. Intracardiac shunting results in the immediate passage of microbubbles from the right to left chambers without a three-cycle delay and can be excluded using this technique.
THERAPY No clearly effective medical therapy for HPS is available. Somatostatin, almitrine, indomethacin, intravenous L-NAME and plasma exchange have all been tried unsuccessfully.15 Aspirin increased arterial oxygenation in two children with HPS,56 and a case report57 and a subsequent open-label trial58 using garlic also suggested a beneficial effect. In the later trial, garlic powder was administered for a minimum of 6 months; 6 of 15 (40%) patients had a significant improvement in PaO2 (>10 mmHg) and one had resolution of hypoxemia (PaO2: 46 mmHg to 80 mmHg) over a 1.5-year period. Acute infusion of methylene blue, a dye that inhibits the effect of NO on soluble guanylate cyclase, also transiently improved oxy-
481
Section III. Clinical Consequences of Liver Disease
Figure 24-4. Technetium-labeled macroaggregated albumin (MAA) scanning to detect and quantify intrapulmonary shunting in HPS. (A) Normal MAA scan with regions of interest drawn around the lungs and cerebrum. In the absence of intrapulmonary vasodilatation, little of the intravenously administered labeled albumin passes through the lungs and signal intensity is low in the cerebrum. Shunting is quantified by comparing the relative signal intensity in the lung and the brain. (B) An abnormal MAA scan in HPS demonstrates significant cerebral uptake owing to the passage of labeled albumin through the dilated pulmonary microvasculature. (Reproduced from Abrams GA et al. Use of macroaggregated albumin lung perfusion scan to diagnose hepatopulmonary syndrome: a new approach. Gastroenterology 1998;114:308, with permission.)
A
B
genation in two reports comprising a total of eight patients.30,31 In addition, the acute inhibition of nitric oxide production with inhaled L-NAME also transiently improved oxygenation in one patient.29 Finally, a single case report suggests that norfloxacin may have also contributed to improvement in oxygen saturation in HPS.59 These reports highlight the need to target likely pathogenetic mechanisms in randomized multicenter trials in order to recruit sufficient numbers of patients to achieve adequate statistical power. Seven case reports have evaluated the effects of transjugular intrahepatic portosystemic shunt (TIPS) on HPS in cirrhosis. Five have found a degree of improvement in oxygenation. However, the short duration of follow-up in two60 and the presence of coexistent hepatic hydrothorax in another61 limit evaluation of the utility of TIPS in these cases. In a fourth report arterial oxygenation clearly improved by 20 mmHg 6 months after TIPS placement.62 However, based on radionuclide lung perfusion scanning, significant intrapulmonary shunting persisted and the cardiac output increased after TIPS. These findings suggest that improved oxygenation may not have been due to reversal of intrapulmonary vasodilatation. A fifth and a sixth report in an 11-year-old girl with biliary atresia and in a 46-year-old woman with alcoholic liver disease revealed significantly improved oxygenation and decreased intrapulmonary shunting sustained over respectively 8 months and 3 years after TIPS placement.63 There is a seventh report of failure of TIPS to improve oxygenation in one patient, and identification of two patients where HPS developed in the setting of a functioning TIPS.64 Together, these findings document the considerable uncertainty regarding the utility of TIPS for HPS. Currently, TIPS should be considered an experimental treatment and its use confined to the setting of clinical trials, so that its efficacy may be judged. Liver transplantation is the only proven therapy for HPS based on the total resolution of or a significant improvement in postoperative
482
gas exchange in more than 85% of reported patients.65 However, the length of time for arterial hypoxemia to normalize after transplantation varies, and may be more than 1 year.66 In addition, mortality is increased after transplantation in HPS patients compared to subjects without HPS,65,67 and unique postoperative complications, including pulmonary hypertension,68 cerebral embolic hemorrhage,69 and immediate postoperative deoxygenation requiring prolonged mechanical ventilation,70 have been reported. Innovative approaches such as frequent body positioning71 or inhaled nitric oxide72,73 have been used to improve postoperative gas exchange. Further research focused on the perioperative medical management of HPS patients is needed to optimize survival.
PROGNOSIS AND NATURAL HISTORY The natural history of HPS is incompletely characterized, but available data indicate that quality of life and survival are adversely affected by its presence. Over time, most patients appear to develop progressive intrapulmonary vasodilatation and worsening gas exchange,74 and spontaneous improvement is rare.75 Mortality is significant in patients with HPS74 and may be due in part to causes directly related to intrapulmonary vasodilatation. In addition, many patients with moderate to severe HPS have comparatively well preserved hepatic synthetic function (i.e. Child–Pugh class A), making it likely that the presence of HPS will impair their quality of life.7,12,13 A recent prospective study has evaluated the natural history of HPS in a cohort of 111 patients with cirrhosis, of whom 27 (24%) had HPS.10 The median survival among patients with HPS was significantly shorter (10.6 months) than in those without HPS (40.8 months). Mortality remained higher in those with HPS after adjusting for the severity of the underlying liver disease and after excluding patients who underwent liver transplantation during follow-up.
Chapter 24 PULMONARY COMPLICATIONS IN PATIENTS WITH LIVER DISEASE
The causes of death in patients with HPS were mainly the complications of hepatocellular dysfunction and portal hypertension, and correlated with the severity of hypoxemia. Currently, liver transplantation is the only effective treatment for patients with HPS.65,76 Arguedas et al.67 prospectively evaluated the utility of the severity of HPS as a predictor of outcome after liver transplantation in a cohort of 24 patients with cirrhosis and HPS. They found that mortality after liver transplantation was markedly increased in severe HPS, in part owing to the development of unique postoperative complications recognized in HPS patients.77,78 A preoperative PaO2 of £50 mmHg, either alone or in combination with a macroaggregated albumin shunt fraction ≥20%, was the strongest predictor of postoperative mortality. Hence, the presence of HPS may adversely affect survival in patients with cirrhosis, and the outcome of transplantation worsens as HPS progresses.
PORTOPULMONARY HYPERTENSION DEFINITION The association between pulmonary arterial hypertension and portal hypertension was initially described in 1951.79 Over the last 10–15 years, portal hypertension has been recognized as a relatively common underlying condition associated with the development of pulmonary arterial hypertension.80 POPH is defined by the NIH Patient Registry for the Characterization of Primary Pulmonary Hypertension as a mean pulmonary artery pressure >25 mmHg and a pulmonary capillary wedge pressure <15 mmHg occurring in the setting of portal hypertension.81 An elevated transpulmonary gradient (mean pulmonary artery pressure - pulmonary capillary wedge pressure >10 mmHg) and/or pulmonary vascular resistance (>240 dyne/s/cm-5) are ancillary criteria included in the definition of this syndrome.
trophy and hyperplasia, concentric intimal fibrosis, plexogenic arteriopathy, and necrotizing vasculitis.23,82,89–91 The underlying mechanisms in POPH remain incompletely understood and no animal models have been developed. To date, all patients with POPH have been found to have portal hypertension, suggesting that some consequence of elevated portal pressures is critical for the development of pulmonary hypertension.83 Accordingly, the hyperdynamic circulatory state, causing increased vascular shear stress and portosystemic shunting leading to altered production or metabolism of vasoactive substances, has been hypothesized to contribute to the vascular changes present in POPH.25 A number of specific endothelial and circulating factors (prostacyclin, thromboxane, serotonin, endothelin-1) as well as genetic polymorphisms in genes regulating vascular proliferative responses (serotonin, TGF-b receptor superfamily) might contribute to POPH, but have not been directly evaluated. In addition, the finding that either HPS or POPH may occur in the same clinical setting suggests that these two entities may share underlying pathogenetic mechanisms. One emerging hypothesis suggests that the degree of endothelial dysfunction or injury may determine whether vasoproliferation and inflammation (more endothelial dysfunction/injury, POPH) or vasodilatation (less endothelial dysfunction/injury, HPS) is the predominant alteration (see Figure 24-1).
CLINICAL FEATURES
POPH has been described most commonly in patients with cirrhosis and portal hypertension. It has also been observed in disorders characterized by portal hypertension without cirrhosis, supporting the suggestion that portal hypertension is an important predisposing condition.25 In an autopsy series of 17 901 specimens, pathological changes consistent with pulmonary hypertension were found in 0.73% of patients with cirrhosis, compared to a prevalence of 0.13% in subjects without chronic liver disease.82 A subsequent prospective study of 507 patients with portal hypertension who underwent right heart catheterization revealed a 2% prevalence of POPH.83 More recently, retrospective studies in patients referred for liver transplantation have found an even higher prevalence of this disorder, with reported values ranging from 3.5% to 16%.84–87 The prevalence and severity of POPH do not appear to correlate with the degree of hepatic synthetic dysfunction or the severity of portal hypertension.83 However, the severity of cardiopulmonary symptoms worsens with increasing pulmonary hypertension.88
Symptoms are common in POPH, but are non-specific. A number of patients may be asymptomatic.87 The most common symptom, as in HPS, is dyspnea on exertion. As the disease advances, progressive fatigue, dyspnea at rest, peripheral edema, syncope, and chest pain may develop.92 Edema, syncope and chest pain are not characteristic features of HPS (Table 24-2). On physical examination elevated jugular pressure, a loud pulmonary component of the second heart sound, a systolic murmur resulting from tricuspid regurgitation, and lower extremity edema are commonly noted in POPH. Electrocardiographic abnormalities are present in the majority of patients and consist of right atrial enlargement, right ventricular hypertrophy, right axis deviation, and/or right bundle branch block. Radiographic findings are generally subtle, but in advanced cases a prominent main pulmonary artery or cardiomegaly due to prominent right cardiac chambers may be appreciated. Gas exchange abnormalities are generally mild and less severe than in HPS. An increased AaPO2 with mild hypoxemia and hypocarbia may be seen, particularly in more severe disease.25,93 Other causes of dyspnea in patients with cirrhosis and portal hypertension, including intrinsic lung disease, deconditioning, muscle wasting, ascites, hepatic hydrothorax, and HPS, should be considered when the diagnosis of POPH is entertained. In addition, other causes of elevated pulmonary pressures and/or right heart failure, including left ventricular dysfunction, volume overload, and chronic obstructive lung disease, may present with clinical features similar to those of POPH.
PATHOLOGY AND PATHOGENESIS
DIAGNOSIS
The pulmonary histologic abnormalities of POPH occur in the resistance arterial vessels and mimic those in primary pulmonary hypertension. These abnormalities include smooth muscle hyper-
Because a number of patients with POPH may be asymptomatic and the diagnostic utility of various clinical features (systemic hypertension, accentuated P2, electrocardiographic and chest
EPIDEMIOLOGY
483
Section III. Clinical Consequences of Liver Disease
radiographic abnormalities) is low,85,94 the diagnosis requires a high index of suspicion (see Figure 24-2). In general, in patients not being evaluated for liver transplantation, the presence of ‘compatible’ symptoms and signs, and/or the exclusion of other cardiopulmonary diseases signals the need for screening for POPH. In all patients being evaluated for liver transplantation, regardless of signs or symptoms, screening is warranted because the presence of POPH may influence their transplant candidacy.95 Transthoracic Doppler echocardiography is the best non-invasive screening study to detect POPH. If combined with intravenous contrast injection, screening for HPS and POPH can be accomplished at the same time. The presence of pulmonary hypertension is suggested by an increased estimated pulmonary artery (PA) systolic pressure (derived from measuring the velocity of the tricuspid regurgitant jet), pulmonary valve insufficiency, right atrial enlargement, and/or right ventricular hypertrophy or dilatation. Several recent studies have evaluated the utility of estimated PA systolic pressure measurements in the diagnosis of POPH.87,94,96 In these studies, estimated PA systolic pressures used to define an elevated value ranged from 30 to 50 mmHg. In each study, between 10 and 15% of patients had elevated estimated PA systolic pressures by echocardiography, and roughly half of these were confirmed to have POPH on subsequent testing. In the most recent prospective study, Doppler echocardiography had positive and negative predictive values of 59% and 100%, respectively, in detecting POPH.87 However, the precise methods for estimating PA systolic pressures have not been standardized between studies and may have influenced the operating characteristics of echocardiographic screening. From a practical perspective, finding an estimated PA systolic pressure of >40–45 mmHg, particularly if right atrial and/or right ventricular abnormalities are also present, should trigger further evaluation, and this approach is likely to detect almost all patients with POPH. The most common causes for a false positive echocardiogram are elevated pulmonary venous pressures due to the hyperdynamic circulatory state and volume overload.87 Patients with suggestive echocardiographic findings should undergo right heart catheterization to confirm elevated mean PA pressure and to exclude pulmonary venous hypertension. Direct measurement of PA pressures, pulmonary capillary wedge pressure and cardiac output, and calculation of systemic and pulmonary vascular resistance should be done. Responsiveness to a number of vasodilator agents, most frequently nitric oxide and/or epoprostenol, is often measured in those with confirmed POPH in an effort to predict a favorable response to long-term vasodilator therapy.25 However, the utility of vasodilator testing in the management of POPH has not been studied.
TREATMENT Medical treatment for POPH is palliative and is based largely on experience in primary pulmonary hypertension. Treatment with vasodilators is the mainstay of therapy and can reverse the vasoconstriction associated with POPH, but has little or no effect on the fibrotic and proliferative remodeling changes. In primary pulmonary hypertension the administration of calcium channel blockers prolongs survival,97 but these agents are not recommended in POPH owing to the possibility of increasing portal pressures.98 A single case has found short- and long-term beneficial pulmonary hemodynamic
484
effects from the use of isosorbide-5¢ mononitrate.99 The use of badrenergic blockers in patients with POPH is controversial, based on the potential risk for cardiac depression. Diuretics are often required to control fluid retention in cirrhosis and portal hypertension, and this requirement may be significantly increased in the setting of right heart failure due to POPH. However, diuretics should be used with particular caution in POPH, as intravascular volume depletion may critically reduce the cardiac output by decreasing right ventricle preload. Oral anticoagulation is not recommended in POPH because of the increased risk of bleeding in the presence of thrombocytopenia, coagulopathy, and varices. Prostacyclin PGI2 (epoprostenol) is a potent vasodilator and platelet aggregation inhibitor that results in clinical improvement and increased survival in primary pulmonary hypertension, and is useful as a bridge to lung transplantation.100–102 Although randomized controlled trials have not been performed in POPH, two small series and a case report have also demonstrated improved mean pulmonary artery pressures, pulmonary vascular resistance, cardiac output and 6-minute walking test with the chronic use of intravenous epoprostenol.103,104 However, epoprostenol does not appear to improve survival in POPH, and whether it might provide a bridge to liver transplantation has not been clearly defined.105 In addition, there have been reports of worsening splenomegaly and hypersplenism and concerns related to worsening ascites with the chronic use of epoprostenol.106 Preliminary reports of the use of other prostacyclin analogs, including treprostinil (subcutaneous injection) and iloprost (inhaled), suggest that these agents may also be useful to improve pulmonary hemodynamics in POPH. Newer agents developed or under study for the treatment of primary pulmonary hypertension may also be useful in patients with POPH. These include endothelin receptor antagonists, inhaled nitric oxide, phosphodiesterase inhibitors, and L-arginine. Bosentan is an orally available dual ET receptor antagonist (A and B) that improves pulmonary hemodynamics in primary pulmonary hypertension. However, this agent has been associated with increases in hepatic enzymes,107 possibly owing to inhibition of hepatocyte bile acid transport,108 and may lower systemic blood pressure. The safety of this agent in cirrhosis, particularly if advanced, is unknown. Sitaxsentan, a selective ETA receptor blocker, has been associated with severe cases of acute hepatitis and should be avoided in POPH.109 Finally, a recent report has shown that sildenafil significantly reduced mean PA pressure in a patient with POPH without detrimental effects on systemic hemodynamics.110 The safety and efficacy of newer agents for POPH needs to be established. The efficacy of liver transplantation as a treatment for POPH also remains controversial. Based on retrospective data and clinical experience, severe POPH (mean pulmonary artery pressure >50 mmHg) is a contraindication to transplantation owing to a perioperative mortality of approximately 40% and lack of reversibility of pulmonary hypertension111,112 (Table 24-3). Patients with mild POPH (mean pulmonary artery pressure <35 mmHg) appear to have no increase in perioperative cardiopulmonary mortality after liver transplantation, although no results of either long-term follow-up or resolution of pulmonary hypertension have been documented.95 The outcome after liver transplantation in patients with moderate POPH (mean pulmonary artery pressure 35–50 mmHg) and in those who have improvement in pulmonary artery pressures on long-term medical
Chapter 24 PULMONARY COMPLICATIONS IN PATIENTS WITH LIVER DISEASE
Table 24-3. Pulmonary Hemodynamics and Liver Transplant Survival in Portopulmonary Hypertension Parameter**
n
Cardiopulmonary mortality (%)
p value
MPAP ≥ 35 mmHg MPAP £ 35 mmHg PVR ≥ 250 PVR £ 250 CO > 8.0 CO < 8.0
29 14 20 18 7 31
14 (48) 0 (0) 11 (55) 1 (6) 1 (14) 11 (35)
0.003 0.003 0.52
** MPAP ≥ 50 mortality 6/6 (100%). MPAP, mean pulmonary artery pressure; PVR, pulmonary vascular resistance (dynes/s/ cm5); CO, cardiac output. Adapted from Krowka MJ et al. Pulmonary hemodynamics and perioperative cardiopulmonary-related mortality in patients undergoing liver transplantation. Liver Transpl 2000;6:443–450.
therapy is less well defined and requires further evaluation.111 Although case reports have demonstrated successful outcomes after combination lung–liver or heart–lung–liver transplantation, limited organ availability and the technical challenges limit the feasibility of such approaches for POPH.113
PROGNOSIS AND NATURAL HISTORY The major complication of POPH is the development of progressive right ventricular dysfunction and cor pulmonale. Survival in pulmonary hypertension correlates with the severity of right-sided cardiac dysfunction, as assessed by the degree of elevation in the right-sided cardiac pressures and the degree of decline in cardiac output.114 Compared to primary pulmonary hypertension, survival appears to be prolonged in POPH (5-year survival 25% vs 50%, respectively), possibly related to the beneficial effects of the hyperdynamic circulatory state.24 However, in a cohort of patients with severe POPH (mean pulmonary artery pressure 59 mmHg) the median survival was 6 months, with a 5-year survival of less than 10%.92 To date, no studies have demonstrated that medical therapy for POPH improves survival.
SUMMARY AND CONCLUSIONS HPS and POPH are increasingly recognized complications of liver disease and/or portal hypertension that may cause significant morbidity and influence survival and liver transplant candidacy. HPS occurs in approximately 20% and POPH occurs in approximately 6% of patients with cirrhosis being evaluated for liver transplantation. HPS results from pre- and postcapillary dilatation in the pulmonary microvasculature and frequently causes symptoms and hypoxemia. POPH results from vasoconstriction and remodeling in resistance vessels, and may cause fewer symptoms and is less frequently associated with hypoxemia. The pathogenesis of pulmonary vascular abnormalities in HPS and POPH is an area of ongoing investigation, and similar mechanisms may play a role in each syndrome. There are no effective medical therapies for HPS, but liver transplantation can reverse the syndrome in the majority of patients. In contrast, there are palliative medical therapies for POPH, but for many patients liver transplantation is currently contraindicated or controversial. Transplantation carries increased mortality in both
severe HPS and POPH, emphasizing the importance of screening for these disorders in patients undergoing liver transplant evaluation. Understanding pathogenesis and developing effective medical therapies are also priorities for both disorders, so that the use of liver transplantation can be targeted to patients with the greatest potential for long-term benefit at an acceptable risk level.
REFERENCES 1. Sood G, Fallon MB, Niwas S, et al. Utility of a dyspnea–fatigue index for screening liver transplant candidates for hepatopulmonary syndrome. Hepatology 1998;28:2319 [abstract]. 2. Hourani J, Bellamy P, Tashkin D, et al. Pulmonary dysfunction in advanced liver disease: frequent occurrence of an abnormal diffusion capacity. Am J Med 1991;90:693–700. 3. Lange PA, Stoller JK. The hepatopulmonary syndrome. Ann Intern Med 1995;122:521–529. 4. Rodriguez-Roison R, Agusti AG, Roca J. The hepatopulmonary syndrome: new name, old complexities. Thorax 1992;47:897–902. 5. Fluckiger M. Vorkommen von trommelschagel formigen fingerendphalangen ohne chronische veranderungen an der lungen oder am herzen. Wien Med Wochenschr 1884;34:1457. 6. Kennedy TC, Knudson RJ. Exercise aggregated hypoxemia and orthodeoxia in cirrhosis. Chest 1977;72:305. 7. Abrams GA, Jaffe CC, Hoffer PB, et al. Diagnostic utility of contrast echocardiography and lung perfusion scan in patients with hepatopulmonary syndrome. Gastroenterology 1995;109:1283–1288. 8. Abrams G, Nanda N, Dubovsky E, et al. Use of macroaggregated albumin lung perfusion scan to diagnose hepatopulmonary syndrome: a new approach. Gastroenterology 1998;114:305–310. 9. Martinez G, Barbera J, Navasa M, et al. Hepatopulmonary syndrome associated with cardiorespiratory disease. J Hepatol 1999;30:882–889. 10. Schenk P, Schoniger-Hekele M, Fuhrmann V, et al. Prognostic significance of the hepatopulmonary syndrome in patients with cirrhosis. Gastroenterology 2003;125:1042–1052. 11. Martinez G, Barbera J, Visa J, et al. Hepatopulmonary syndrome in candidates for liver transplantation. J Hepatol 2001;34:756–758. 12. Krowka M, Wiseman G, Burnett O, et al. Hepatopulmonary syndrome: A prospective study of relationships between severity of liver disease, PaO2 response to 100% oxygen, and brain uptake after 99mTc MAA lung scanning. Chest 2000;118:615–624. 13. Whyte M, Hughes J, Peters A, et al. Analysis of intrapulmonary right to left shunt in hepatopulmonary syndrome. J Hepatol 1998;29:85–93. 14. Dimand RJ, Heyman MB, Bass NM, et al. Hepatopulmonary syndrome: response to hepatic transplantation. Hepatology 1991;141:55A [Abstract]. 15. Abrams G, Fallon M. The hepatopulmonary syndrome. Clin Liver Dis 1997;1:185–200. 16. Binay K, Sen S, Biswas PK, et al. Hepatopulmonary syndrome in inferior vena cava obstruction responding to cavoplasty. Gastroenterology 2000;118:192–196. 17. Gupta D, Vijaya DR, Gupta R, et al. Prevalence of hepatopulmonary syndrome in cirrhosis and extrahepatic portal venous obstruction. Am J Gastroenterol 2001;96:3395–3399. 18. Regev A, Yeshurun M, Rodriguez M, et al. Transient hepatopulmonary syndrome in a patient with acute hepatitis. Am J Viral Hepatol 2001;8:83–86. 19. Teuber G, Teupe C, Dietrich C, et al. Pulmonary dysfunction in non-cirrhotic patients with chronic viral hepatitis. Eur J Intern Med 2002;13:311–318.
485
Section III. Clinical Consequences of Liver Disease
20. Lee D, Lepler L. Severe intrapulmonary shunting associated with metastatic carcinoid. Chest 1999;115:1203–1207. 21. McFaul R, Tajik A, Mair D, et al. Development of pulmonary arteriovenous shunt after superior vena cava–right pulmonary artery (Glenn) anastamosis. Report of four cases. Circulation 1977;55:212–216. 22. Srivastava D, Preminger T, Lock JE, et al. Hepatic venous blood and the development of pulmonary arteriovenous malformations in congenital heart disease. Circulation 1995;92: 1217–1222. 23. Schraufnagel D, Kay J. Structural and pathologic changes in the lung vasculature in chronic liver disease. Clin Chest Med 1996;17:1–15. 24. Herve P, Lebrec D, Brenot F, et al. Pulmonary vascular disorders in portal hypertension. Eur Respir J 1998;11:1153–1166. 25. Budhiraja R, Hassoun PM. Portopulmonary hypertension: a tale of two circulations. Chest 2003;123:562–576. 26. Rolla G, Brussino L, Colagrande P, et al. Exhaled nitric oxide and oxygenation abnormalities in hepatic cirrhosis. Hepatology 1997;26:842–847. 27. Rolla G, Brussino L, Colagrande P. Exhaled nitric oxide and impaired oxygenation in cirrhotic patients before and after liver transplantation. Ann Intern Med 1998;129:375–378. 28. Cremona G, Higenbottam TW, Mayoral V, et al. Elevated exhaled nitric oxide in patients with hepatopulmonary syndrome. Eur Respir J 1995;8:1883–1885. 29. Brussino L, Bucca C, Morello M, et al. Effect on dyspnoea and hypoxaemia of inhaled NG-nitro-L-arginine methyl ester in hepatopulmonary syndrome. Lancet 2003;362:43–44. 30. Rolla G, Bucca C, Brussino L. Methylene blue in the hepatopulmonary syndrome. N Engl J Med 1994;331:1098. 31. Schenk P, Madl C, Rezale-Majd S, et al. Methylene blue improves the hepatopulmonary syndrome. Ann Intern Med 2000;133:701–706. 32. Chang S-W, O’Hara N. Pulmonary circulatory dysfunction in rats with biliary cirrhosis. Am Rev Respir Dis 1992;148: 798–805. 33. Fallon MB, Abrams GA, McGrath JW, et al. Common bile duct ligation in the rat: a model of intrapulmonary vasodilatation and hepatopulmonary syndrome. Am J Physiol 1997;272:G779–G784. 34. Luo B, Liu L, Tang L, et al. ET-1 and TNF-a in HPS: analysis in prehepatic portal hypertension and biliary and nonbiliary cirrhosis in rats. Am J Physiol Gastrointest Liver Physiol 2004;286:G294–303. 35. Fallon MB, Abrams GA, Luo B, et al. The role of endothelial nitric oxide synthase in the pathogenesis of a rat model of hepatopulmonary syndrome. Gastroenterology 1997;113:606–614. 36. Luo B, Abrams GA, Fallon MB. Endothelin-1 in the rat bile duct ligation model of hepatopulmonary syndrome: correlation with pulmonary dysfunction. J Hepatol 1998;29:571–578. 37. Zhang J, Ling Y, Luo B, et al. Analysis of pulmonary heme oxygenase-1 and nitric oxide synthase alterations in experimental hepatopulmonary syndrome. Gastroenterology 2003;125:1441–1451. 38. Nunes H, Lebrec D, Mazmanian M, et al. Role of nitric oxide in hepatopulmonary syndrome in cirrhotic rats. Am J Respir Crit Care Med 2001;164:879–885. 39. Zhang M, Luo B, Chen SJ, et al. Endothelin-1 stimulation of endothelial nitric oxide synthase in the pathogenesis of hepatopulmonary syndrome. Am J Physiol 1999;277:G944–G952. 40. Luo B, Liu L, Tang L, et al. Increased pulmonary vascular endothelin B receptor expression and responsiveness to endothelin-1 in cirrhotic and portal hypertensive rats: a potential mechanism in experimental hepatopulmonary syndrome. J Hepatol 2003;38:556–563.
486
41. Ling Y, Zhang J, Luo B, et al. The role of endothelin-1 and the endothelin B receptor in the pathogenesis of experimental hepatopulmonary syndrome. Hepatology 2004;39:1593–1602. 42. Carter EP, Hartsfield CL, Miyazono M, et al. Regulation of heme oxygenase-1 by nitric oxide during hepatopulmonary syndrome. Am J Physiol – Lung Cell Mol Physiol 2002;283:L346–353. 43. Rabiller A, Nunes H, Lebrec D, et al. Prevention of gramnegative translocation reduces the severity of hepatopulmonary syndrome. Am J Respir Crit Care Med 2002;166:514–517. 44. Sztrymf B, Rabiller A, Nunes H, et al. Prevention of hepatopulmonary syndrome by pentoxifylline in cirrhotic rats. Eur Respir J 2004;23:752–758. 45. Robin ED, Laman D, Horn BR, et al. Platypnea related to orthodeoxia caused by true vascular lung shunts. N Engl J Med 1976;294:941–943. 46. McAdams HP, Erasmus J, Crockett R, et al. The hepatopulmonary syndrome: radiologic findings in 10 patients. Am J Roentgenol 1996;166:1379–1385. 47. Lima B, Franca A, Pazin-Filho A, et al. Frequency, clinical characteristics, and respiratory parameters of hepatopulmonary syndrome. Mayo Clin Proc 2004;79:42–48. 48. Schenk P, Fuhrmann V, Madl C, et al. Hepatopulmonary syndrome: prevalence and predictive value of various cut offs for arterial oxygenation and their clinical consequences. Gut 2002;51:853–859. 49. Fallon M, Abrams G. Pulmonary dysfunction in chronic liver disease. Hepatology 2000;32:859–865. 50. Abrams GA, Fallon MB, Nail C, et al. Pulse oximetry: Utility in detecting arterial hypoxemia in subjects with cirrhosis. Hepatology 1999;30:A1099 [abstract]. 51. Vedrinne JM, Duperret S, Bizollon T, et al. Comparison of transesophageal and transthoracic contrast echocardiography for detection of an intrapulmonary shunt in liver disease. Chest 1997;111:1236–1240. 52. Aller R, Moya JL, Moreira V, et al. Diagnosis of hepatopulmonary syndrome with contrast transesophageal echocardiography: advantages over contrast transthoracic echocardiography. Dig Dis Sci 1999;44:1243–1248. 53. MacNee W, Buist T, Finlayson N, et al. Multiple microscopic pulmonary arteriovenous connections in the lungs presenting as cyanosis. Thorax 1985;40:316–318. 54. Poterucha JJ, Krowka MJ, Dickson ER, et al. Failure of hepatopulmonary syndrome to resolve after liver transplantation and successful treatment with embolotherapy. Hepatology 1995;21:96–100. 55. Lee KN, Lee HJ, Shin WW, et al. Hypoxemia and liver cirrhosis (hepatopulmonary syndrome) in eight patients: comparison of the central and peripheral pulmonary vasculature. Radiology 1999;211:549–553. 56. Song JY, Choi JY, Ko JT, et al. Long-term aspirin therapy for hepatopulmonary syndrome. Pediatrics 1996;97:917–920. 57. Caldwell SH, Jeffers LJ, Narula OS, et al. Ancient remedies revisited: does Allium sativum (garlic) palliate the hepatopulmonary syndrome? J Clin Gastroenterol 1992;15:248–250. 58. Abrams GA, Fallon MB. Treatment of hepatopulmonary syndrome with Allium sativum L. (garlic): A pilot trial. J Clin Gastroenterol 1998;27:232–235. 59. Anel RM, Sheagren JN. Novel presentation and approach to management of hepatopulmonary syndrome with use of antimicrobial agents. Clin Infect Dis 2001;32:E131–136. 60. Reigler JL, Lang KA, Johnson SP, et al. Transjugular intrahepatic portosystemic shunt improves oxygenation in hepatopulmonary syndrome. Gastroenterology 1995;109:978–983. 61. Selim KM, Akriviadis EA, Zuckerman E, et al. Transjugular intrahepatic portosystemic shunt: a successful treatment for hepatopulmonary syndrome. Am J Gastroenterol 1998;93:455–458.
Chapter 24 PULMONARY COMPLICATIONS IN PATIENTS WITH LIVER DISEASE
62. Allgaier HP, Haag K, Ochs A, et al. Hepatopulmonary syndrome: Successful treatment by transjugular intrahepatic portosystemic stent–shunt (TIPS). J Hepatol 1995;23:102–105. 63. Paramesh A, Hussain S, Shneider B, et al. Improvement of hepatopulmonary syndrome after transjugular intrahepatic portasystemic shunting: Case report and review of literature. Pediatr Transpl 2003;7:157–162. 64. Corley DA, Scharschmidt B, Bass N, et al. Lack of efficacy of TIPS for hepatopulmonary syndrome. Gastroenterology 1997;113:728–731. 65. Lange PA, Stoller JK. The hepatopulmonary syndrome: effect of liver transplantation. Clin Chest Med 1996;17:115–123. 66. Philit F, Wiesendanger T, Gille D, et al. Late resolution of hepatopulmonary syndrome after liver transplantation. Respiration 1997;64:173–175. 67. Arguedas M, Abrams GA, Krowka MJ, et al. Prospective evaluation of outcomes and predictors of mortality in patients with hepatopulmonary syndrome undergoing liver transplantation. Hepatology 2003;37:192–197. 68. Mandell MS, Groves BM, Duke J. Progressive plexogenic pulmonary hypertension following liver transplantation. Transplantation 1995;59:1488–1490. 69. Abrams GA, Rose K, Fallon MB, et al. Hepatopulmonary syndrome and venous emboli causing intracerebral hemorrhages after liver transplantation: a case report. Transplantation 1999;68:1–3. 70. Scott V, Mira A, Kang Y, et al. Reversibility of the hepatopulmonary syndrome by orthotopic liver transplantation. Transplant Proc 1993;25:1787–1788. 71. Meyers C, Low L, Kaufman L, et al. Trendelenburg positioning and continuous lateral rotation improve oxygenation in hepatopulmonary syndrome after liver transplantation. Liver Transpl Surg 1998;6:510–512. 72. Alexander J, Greenough A, Rela M, et al. Nitric oxide treatment of severe hypoxemia after liver transplantation in hepatopulmonary syndrome. Liver Transpl Surg 1997;1:54–55. 73. Durand P, Baujard C, Grosse AL. Reversal of hypoxemia by inhaled nitric oxide in children with severe hepatopulmonary syndrome, type 1, during and after liver transplantation. Transplantation 1998;65:437–439. 74. Krowka MJ, Dickson ER, Cortese DA. Hepatopulmonary syndrome. Clinical observations and lack of therapeutic response to somatostatin analogue. Chest 1993;104:515–521. 75. Saunders KB, Fernando SSD, Dalton HR, et al. Spontaneous improvement in a patient with the hepatopulmonary syndrome assessed by serial exercise tests. Thorax 1994;49:725–727. 76. Krowka MJ, Mandell S, Ramsay MA, et al. Hepatopulmonary syndrome and portopulmonary hypertension: A report of the multicenter liver transplant database. Liver Transpl 2004;10:174–182. 77. Kaspar MD, Ramsay MA, Shuey CB Jr, et al. Severe pulmonary hypertension and amelioration of hepatopulmonary syndrome after liver transplantation. Liver Transpl Surg 1998;4:177–179. 78. Martinez-Palli G, Barbera JA, Taura P, et al. Severe portopulmonary hypertension after liver transplantation in a patient with preexisting hepatopulmonary syndrome. J Hepatol 1999;31:1075–1079. 79. Mantz F, Craige E. Portal axis thrombosis with spontaneous portocaval shunt and resultant cor pulmonale. AMA Arch Pathol 1951;52:91–97. 80. Rubin L. Primary pulmonary hypertension: ACCP consensus statement. Chest 1993;104:236–250. 81. Rich S, Dantzker D, Ayres S, et al. Primary pulmonary hypertension: A national prospective study. Ann Intern Med 1987;107:216–223. 82. McDonnell P, Toye P, Hutchins G. Primary pulmonary hypertension and cirrhosis: are they related? Am Rev Respir Dis 1983;127:437–441.
83. Hadengue A, Benhayoun M, Lebrec D, et al. Pulmonary hypertension complicating portal hypertension: prevalence and relation to splanchnic hemodynamics. Gastroenterology 1991;100:520–528. 84. Benjaminov FS, Prentice M, Sniderman KW, et al. Portopulmonary hypertension in decompensated cirrhosis with refractory ascites. Gut 2003;52:1355–1362. 85. Kuo P, Plotkin J, Johnson L, et al. Distinctive clinical features of portopulmonary hypertension. Chest 1997;112:980–986. 86. Plevak D, Krowka M, Rettke S, et al. Successful liver transplantation in patients with mild to moderate pulmonary hypertension. Transplant Proc 1993;25:1840. 87. Colle I, Moreau R, Godinho E, et al. Portopulmonary hypertension in candidates for liver transplantation: diagnosis at evaluation comparing Doppler echocardiography with cardiac catheterization and incidence on the waiting list. Hepatology 2003;37:401–409. 88. Rubin L. Primary pulmonary hypertension. N Engl J Med 1997;336:111–117. 89. Edwards B, Weir E, Edwards W, et al. Coexistent pulmonary and portal hypertension: morphologic and clinical features. J Am Coll Cardiol 1987;10:1233–1238. 90. Pietra G. Histopathology of primary pulmonary hypertension. Chest 1994;105:2S–6S. 91. Matsubara O, Nakamura T, Uehara T, et al. Histometrical investigation of the pulmonary artery in severe hepatic disease. J Pathol 1984;143:31–37. 92. Robalino B, Moodie D. Association between primary pulmonary hypertension and portal hypertension: analysis of its pathophysiology and clinical, laboratory and hemodynamic manifestations. J Am Coll Cardiol 1991;17:492–498. 93. Hoeper MM, Krowka MJ, Strassburg CP. Portopulmonary hypertension and hepatopulmonary syndrome. Lancet 2004;363:1461–1468. 94. Pilatis N, Jacobs L, Rerkpattanapipat P, et al. Clinical predictors of pulmonary hypertension in patients undergoing liver transplant evaluation. Liver Transpl 2000;6:85–91. 95. Krowka M, Plevak D, Findlay J, et al. Pulmonary hemodynamics and perioperative cardiopulmonary-related mortality in patients with portopulmonary hypertension undergoing liver transplantation. Liver Transpl 2000;6:443–450. 96. Torregrosa M, Genesca J, Gonzalez A, et al. Role of Doppler echocardiography in the assessment of portopulmonary hypertension in liver transplantation candidates. Transplantation 2001;71:572–574. 97. Rich S, Kaufmann E, Levey P. The effect of high doses of calcium channel blockers on survival in primary pulmonary hypertension. N Engl J Med 1992;327:76–81. 98. Ota K, Shijo H, Kokawa H, et al. Effects of nifedipine on hepatic venous pressure gradient and portal vein blood flow in patients with cirrhosis. J Gastroenterol Hepatol 1995;10:198–204. 99. Ribas J, Angrill J, Barbera J, et al. Isosorbide-5-mononitrate in the treatment of pulmonary hypertension associated with portal hypertension. Eur Respir J 1999;13:210–212. 100. Higenbottam T, Wheeldon D, Wells F, et al. Long-term treatment of primary pulmonary hypertension with continuous intravenous epoprostenol (prostacyclin). Lancet 1984;1:1046–1047. 101. Higenbottam T, Spiegelhalter D, Scott J, et al. Prostacyclin (epoprostenol) and heart–lung transplantation as treatments for severe pulmonary hypertension. Br Heart J 1993;70: 336–370. 102. Barst R, Rubin L, Long W, et al. A comparison of continuous intravenous epoprostenol (prostacyclin) with conventional therapy for primary pulmonary hypertension. The Primary Pulmonary Hypertension Study Group. N Engl J Med 1996;334:296–302.
487
Section III. Clinical Consequences of Liver Disease
103. Kuo P, Johnson L, Plotkin J, et al. Continuous intravenous epoprostenol for the treatment of portopulmonary hypertension. Transplantation 1997;63:604–606. 104. Krowka M, Frantz R, McGoon M, et al. Improvement in pulmonary hemodynamics during intravenous epoprostenol (prostacyclin): a study of 15 patients with moderate to severe portopulmonary hypertension. Hepatology 1999;30:641–648. 105. Plotkin J, Kuo P, Rubin L, et al. Successful use of chronic epoprostenol as a bridge to liver transplantation in severe portopulmonary hypertension. Transplantation 1998;65:457–459. 106. Findlay J, Plevak D, Krowka M, et al. Progressive splenomegaly after epoprostenol therapy in portopulmonary hypertension. Liver Transpl Surg 1999;5:362–365. 107. Sitbon O, Badesch DB, Channick RN, et al. Effects of the dual endothelin receptor antagonist bosentan in patients with pulmonary arterial hypertension: a 1-year follow-up study. Chest 2003;124:247–254. 108. Fattinger K, Funk C, Pantze M, et al. The endothelin antagonist bosentan inhibits the canalicular bile salt export pump: A
488
109.
110.
111. 112. 113.
114.
potential mechanism for hepatic adverse reactions. Clin Pharmacol Ther 2001;69:223–231. Barst RJ, Rich S, Widlitz A, et al. Clinical efficacy of sitaxsentan, an endothelin-A receptor antagonist, in patients with pulmonary arterial hypertension: open-label pilot study. Chest 2002;121:1860–1868. Makisalo H, Koivusalo A, Vakkuri A, et al. Sildenafil for portopulmonary hypertension in a patient undergoing liver transplantation. Liver Transpl 2004;10:945–950. Mandell M, Groves B. Pulmonary hypertension in chronic liver disease. Clin Chest Med 1996;17:17–33. Krowka M. Hepatopulmonary syndromes. Gut 2000; 46:1–4. Wallwork J, Calne R, Williams R. Transplantation of liver, heart and lungs for primary biliary cirrhosis and primary pulmonary hypertension. Lancet 1987;2:182–185. D’Alonzo G, Barst R, Ayres S, et al. Survival in patients with primary pulmonary hypertension: results from a national prospective registry. Ann Intern Med 1991;115:343–349.
Section III: Clinical Consequences of Liver Disease
25
HEMATOPOIETIC ABNORMALITIES AND HEMOSTASIS Eric Esrailian and Sammy Saab Abbreviations ADP adenosine 5¢-diphosphate ATG antithymocyte globulin ATIII antithrombin III CFU-GM colony-forming units DIC disseminated intravascular coagulation ELT euglobulin clot lysis time EPO erythropoietin FDP fibrin degradation products FFP fresh frozen plasma FHF fulminant hepatic failure
G-CSF GM-CSF GVHD HCV HLA INR ITP MELD pRBC
colony-stimulating factor granulocyte-macrophage colonystimulating factor graft-versus-host disease hepatitis C virus human leukocyte antigen international normalized ratio immune thrombocytopenic purpura Model for End-Stage Liver Disease packed red blood cell
INTRODUCTION From the anemia encountered in the treatment of viral hepatitis to gastrointestinal bleeding, the hematopoietic system and hemostasis are clinically relevant areas when treating patients with liver disease. Whereas abnormalities in these fields are often managed by internists and hematologists, hepatologists typically take a lead role when these conditions are a result of liver disease. The intricacies of bone marrow function and the coagulation cascade can be found in hematology textbooks, and this chapter will focus on the issues of practical importance to the hepatologist and provide only a brief description of the coagulation pathways.
EPIDEMIOLOGY AND PATHOGENESIS Abnormalities of hematopoiesis and hemostasis in patients with liver disease have been described since the early part of the 20th century.1–3 Although anemia in patients with cirrhosis is believed to be common, the prevalence has not been well studied. Anemia in this setting is multifactorial and may be related to decreased erythrocyte production and survival, renal insufficiency, splenic sequestration, medications, and blood loss.4 Defining incidence and prevalence data is difficult, and existing studies describe a prevalence range of 4.3–28.2%.5 This range is so broad because studies in children and adults that include such data vary with respect to their definitions of anemia and the contribution of immunosuppression to the anemia when liver transplantation patients are included.6,7
PSC PT rhFVIIa TAFI TF TFPI tPA TPO vWF Xa
primary sclerosing cholangitis prothrombin time recombinant human factor VIIa thrombin-activatable fibrinolysis inhibitor tissue factor tissue factor pathway inhibitor tissue plasminogen activator thrombopoitein von Willebrand factor activated factor X
PATHOGENESIS Hematopoiesis Hematopoiesis refers to the production of peripheral blood cells by the bone marrow. Growth factors stimulate hematopoietic stem cells to give rise to progenitor cells (Figure 25-1). These progenitor cells further differentiate into mature circulating cells. The blood islands of the yolk sac are the first developmental site of hematopoiesis.8 However, by the end of the first trimester the liver becomes the primary site of hematopoiesis in the embryo.9,10 As the embryonic and fetal bone marrow develops, the hematopoietic responsibility gradually shifts from the liver to the bone marrow. After birth, the bone marrow is the primary site of hematopoiesis. Whereas in adults erythropoietin (EPO) production becomes predominantly a function of the kidneys, thrombopoietin (TPO) is produced in the liver.11,12 EPO production is stimulated by tissue hypoxia and regulates the differentiation of erythroid precursor cells. Finally, hematopoietic-specific proteins called granulocyte colony-stimulating factor (G-CSF) and granulocyte–macrophage colony-stimulating factor (GM-CSF) drive granulopoiesis.13 These growth factors stimulate granulocyte and macrophage colony-forming units (CFU-GM) and, ultimately, granulocyte differentiation.14
Platelet Plug Formation and the Coagulation Cascade Other than von Willebrand factor (vWF), a significant portion of the crucial proteins involved in the coagulation cascade, and ultimately hemostasis, are produced in the liver.15 Therefore, it should not be
489
Section III. Clinical Consequences of Liver Disease
matrix, platelet aggregation, and the ultimate formation of the platelet plug.17 vWF, a protein product of the endothelium, facilitates the adherence of platelets to subendothelial components. This adherence is responsible for the commencement of hemostasis. Prior to platelet activation, glycoprotein Ib on the platelet surface facilitates the binding to vWF. In this manner, platelets bind to exposed subendothelial collagen until a sheet of platelets is in place. Platelets then become activated by substances such as adenosine 5¢diphosphate (ADP) and serotonin. These activated platelets then bind fibrinogen via an integrin assembly of glycoproteins IIb and IIIa (GP IIb/IIIa).18 Thromboxane A2 released from activated platelets stimulates platelet aggregation until, ultimately, the platelet plug is formed.3 Concurrent activation of the coagulation cascade results in fibrin formation, cross-linking, and stabilization of the platelet plug. The same endothelial cell injury that causes platelet adhesion allows tissue factor (TF) to be exposed and bind factor VII to form an activated complex (TF-VIIa).19 Figure 25-2 summarizes the consequent pathway towards the activation of factors IX and X, leading to thrombin formation (IIa). Thrombin further activates platelets, activates factors VIII, V, and XI, and cleaves fibrinogen to fibrin in order to stabilize the platelet plug.2 Although a small amount of thrombin is produced initially, clotting factor activation accelerates thrombin formation, leading to further hemostatic plug stabilization.20 Endogenous inhibition mechanisms are also activated after initiation of the hemostasis process to focus clot formation at the site of injury and prevent diffuse, catastrophic thrombosis. Tissue factor pathway inhibitor (TFPI) contributes to this process by binding activated factor X (Xa) and the TF–VIIa complex and inhibiting their aforementioned roles in the coagulation cascade.21 Protein C, protein S, and antithrombin III (ATIII) are also produced in the liver and make important intrinsic contributions to anticoagulation. Thrombin also contributes to anticoagulation by binding a protein on endothelial cells called thrombomodulin. This thrombin–thrombomodulin complex activates protein C. Activated protein C and the cofactor protein S mutually inhibit activated factors Va and VIIIa, thus inhibiting key stimulators of thrombin production.22
surprising that patients with compromised liver function often manifest a clinically significant coagulopathy. A delicate balance of platelet aggregation, fibrin clot formation, and fibrinolysis is part of the normal hemostatic mechanism.16 Endothelial cell damage is the initial key stimulus for platelet adherence to the subendothelial
Granulocyte macrophage lineage
Common myeloid progenitor
Myeloerythroid lineage Hematopoietic stem cell
Natural killer cell Common lymphoid progenitor
T Cell
B Cell
Figure 25-1. Hematopoiesis. Normal hematopoiesis involves the differentiation of pluripotential hematopoietic stem cells expressing the CD34 surface protein (CD34+) into two main committed progenitors: common myeloid and common lymphoid progenitors. Common myeloid progenitors can differentiate into all cells of the myeloerythroid lineage (granulocytes, macrophages, megakaryocytes, erythroid cells), and common lymphoid progenitors can differentiate into T cells, B cells, and natural killer cells. (Reproduced from McCormack MP, Rabbitts TH. Activation of the T-cell oncogene LMO2 after gene therapy for X-linked severe combined immunodeficiency. N Engl J Med 2004;350:913, with permission.)
Intrinsic Pathway
XII
XIIa Extrinsic Pathway XI
XIa
VII TF
IX
IXa + VIIIa VIIa-TF
Common Pathway
X
II (Prothrombin) Fibrogen
490
Xa + Va
X
IIa (Thrombin)
Fibrin
Figure 25-2. Simplified coagulation cascade. After endothelial cell injury, platelets adhere to the subendothelial matrix and the coagulation cascade is initiated. Tissue factor (TF) binds factor VII to form the activated TF–VIIa complex. Thrombin eventually cleaves fibrinogen to fibrin, leading to stabilization of the platelet plug. (Reproduced from Roberts HR, Monroe DM, Escobar MA. Current concepts of hemostasis: implications for therapy. Anesthesiology 2004;100:723, with permission.)
Chapter 25 HEMATOPOIETIC ABNORMALITIES AND HEMOSTASIS
ATIII is a circulating protein synthesized by the liver that inhibits thrombin formation and inactivates factors IXa, Xa, and XIa.23 Heparin and glycosaminoglycans on the endothelium stimulate this ATIII anticoagulation activity.24 A system of fibrinolysis acts in concert with coagulation and the aforementioned prothrombotic system. This interplay is also regulated by feedback mechanisms to prevent excessive activity of any one system. During fibrinolysis, plasminogen binds fibrin. Plasminogen is subsequently cleaved and activated by tissue plasminogen activator (tPA) to form plasmin. Plasmin, the powerful enzyme that drives this process, then lyses fibrin to form fibrin degradation products (FDP).25 Plasminogen activator inhibitor and a2antiplasmin serve as regulatory agents limiting the activity of this system.26,27
CLINICAL FEATURES AND DISEASE COMPLICATIONS ANEMIA IN PATIENTS WITH LIVER DISEASE Anemia in the setting of liver disease is usually multifactorial because of different liver disease etiologies, treatment regimens, and comorbidities. The causes of anemia are often stratified according to the mean corpuscular volume (MCV): microcytic, normocytic, or macrocytic.
Microcytic Anemia In patient with cirrhosis, microcytic anemia due to iron deficiency is common. Transferrin and ferritin are important components of iron metabolism that are synthesized in the liver, making the liver a critical organ in iron transport and storage.28,29 Transferrin is a protein that binds iron after absorption in the proximal small bowel to enable transport to tissues, and ferritin is a key protein for iron storage in hepatocytes, bone marrow, spleen, and other tissues. Iron deficiency from gastrointestinal blood loss is usually due to complications of portal hypertension. However, patient with cirrhosis may also have intestinal blood loss from non-portal hypertensive causes. Whereas bleeding from esophageal and gastric varices is striking, subtle blood loss may go unnoticed. Less rapid bleeding from portal hypertensive gastropathy can be occult and contribute to iron deficiency.30,31 Daily minute amounts of blood loss from gingival bleeding after tooth brushing and epistaxis may also have eventual cumulative effects on body iron stores. This gradual blood loss may be due to thrombocytopenia or to the coagulopathy.32,33 Patients with cirrhosis and elevated serum ferritin levels often spark work-ups for hereditary hemochromatosis. Ferritin is an acutephase reactant, and its specificity with respect to functioning as a marker of iron overload is low.34 Ferritin levels can be elevated in a variety of inflammatory disorders, infections, malignancies, and liver-specific injuries ranging from viral hepatitis to true iron overload states. Our increasing understanding of the genetic features of hereditary hemochromatosis and the development of testing strategies can assist the clinician in make this differentiation.35,36 Often, when the clinical scenario is not clear, a liver biopsy is not only helpful to assess the presence or absence of excess iron, but it can also illuminate the etiology and extent of liver disease. Transferrin
is made in the liver, and in advanced disease the levels of this protein may be low. Thus, given the problems with both ferritin and transferrin, a bone marrow aspiration may be the only method to diagnose iron deficiency in the cirrhotic patient. Sideroblastic anemia is another common cause of microcytic anemia in patients with alcoholic liver disease. However, the anemia may also be normocytic or macrocytic if it is multifactorial. This anemia falls into a subset called acquired sideroblast anemia and is usually due to ongoing alcohol use.37 Alcohol acts to inhibit steps in the heme biosynthetic pathway resulting in mitochondrial iron deposition, and these findings often resolve after the toxic effects of alcohol are no longer present.38
Macrocytic Anemia The most common causes of macrocytic anemias in patients with liver disease are megaloblastic anemia related to cobalamin (vitamin B12) and folic acid (folate) deficiency, and the normoblastic anemias related to erythrocyte membrane pathology.39 Both deficiencies contribute to impaired DNA synthesis and macrocytic erythrocytes (macrocytes) with excessive nuclear:cytoplasmic ratios and often hypersegmented neutrophils.40 However, even in the presence of these deficiencies the classic peripheral smear findings may not always be present.41 Cobalamin is synthesized exclusively by bacteria and is found primarily in meat, eggs, and dairy products. Although the stomach and small bowel are classically involved in the binding and absorption, respectively, of cobalamin, the liver has an important storage role. Transcobalamin I and III (R-binders) are present in gastric secretions and bind cobalamin. Cobalamin eventually undergoes absorption in the terminal ileum, transport to the liver, secretion into bile, and reabsorption in the terminal ileum.42 Nutritional cobalamin deficiency is less common in industrialized countries than in the developing world, but patients with alcoholism are particularly susceptible to this deficiency, making this issue clinically relevant in hepatology.43–45 Hepatic dysfunction negatively affects the enterohepatic pathway of cobalamin metabolism. Excess cobalamin can be released from the liver after hepatic injury of various etiologies. Cirrhosis, viral hepatitis, and hepatocellular carcinoma can all be associated with decreased cobalamin stores and increased circulating levels.46 Thus, serum cobalamin levels may overestimate true body stores and may not be the best method to screen at-risk patients. Measuring methylmalonic acid levels in the serum is more reliable in this setting.47 Folate is a water-soluble vitamin that is predominantly found in green, leafy vegetables. It is also found in meats and a variety of other foods. Folate plays a critical role in single-carbon metabolism, methionine metabolism, and other important biosynthetic pathways.48–50 As with cobalamin, the liver serves a storage role with folate, and the folate metabolism depends on the enterohepatic circulation. Patients with cirrhosis and even diminished hepatic function due to non-cirrhotic liver disease can exhibit disturbances in this circulation. The poor nutritional state of the alcoholic contributes to folate deficiency and other global nutritional deficiencies.51,52 In the setting of alcoholic liver disease, folate metabolism is affected by the impaired liver function and the toxic effects of alcohol itself.53 The enterohepatic circulation of folate depends on its excretion into bile by hepatocytes, and alcohol directly plays a
491
Section III. Clinical Consequences of Liver Disease
role in inhibiting the normal workings of this process and also affects folate metabolism in general. 54–59 Non-megaloblastic, or normoblastic, anemia and macrocytosis without anemia are also common among patients with liver disease.60 The exact etiology is unclear, but in patients with alcoholic cirrhosis a direct effect of alcohol on the red cell is also a potential contributing factor.61,62 This effect may be related to a modification of the lipid composition in erythrocyte membranes.63,64 In addition, anemia of chronic disease (also known as the anemia of acute and chronic inflammation) and renal disease often are present in patients with liver disease. The confluence of these factors may drop the mean corpuscular volume from a baseline high value to a normal or low value. Table 25-1 lists common etiologies of macrocytosis in patients with liver disease. Despite the strong association between anemia and liver disease, erythrocytosis can also occur in these patients. Erythrocytosis is most frequently associated with hepatocellular carcinoma (HCC), but it can rarely be seen in other forms of liver disease.65–67 A constellation of factors such as hypoxia, anemia, and thrombocytopenia may stimulate dysregulated EPO synthesis in the diseased liver.68
Hemolytic Anemia Structural abnormalities on erythrocyte membranes in patients with liver disease can also contribute to hemolysis and hemolytic anemia.69,70 Because of the crucial role of the liver in lipid metabolism, erythrocytes in liver disease may have abnormalities in lipid membrane composition that may contribute to hemolysis.71,72 Spur cell anemia is an important form of hemolytic anemia that is associated with both alcoholic and non-alcoholic liver disease.73 Acanthocytes are large erythrocytes with a spiculed appearance secondary to an abnormal free cholesterol-to-phospholipid ratio in the outer leaflet of the erythrocyte lipid bilayer. These cells have decreased membrane fluidity, irregular osmotic fragility, and a predisposition to splenic sequestration and destruction. Acanthocytes are not exclusive to spur cell anemia in liver disease, and have also been associated with other disorders such as abetalipoproteinemia; however, the pathophysiology differs depending on the underlying disorder.74 Several reports document improvement of acanthocytosis and spur cell anemia with liver transplantation, suggesting a relationship between liver function and the erythrocyte abnormality.75,76 Furthermore, erythrocytes transfused to a patient with liver disease and spur cell anemia can become acanthocytes, and acanthocytes may revert to a normal morphology when incubated with normal serum.77,78 Wilson’s disease is associated with hemolytic anemia, and should be considered in patients who present with liver disease and hemol-
Table 25-1. Causes of Macrocytosis Seen in Patients with Liver Disease Cobalamin deficiency Folate deficiency Alcohol use Cirrhosis Aplastic anemia Hemolytic anemias Drugs (such as antiviral agents)
492
ysis.79,80 In fulminant disease, patients can experience intravascular hemolysis. In this acute setting, plasma exchange can be an effective bridge treatment to subsequent chelation therapy or liver transplantation.81,82 Finally, Zieve’s syndrome is a rare entity which includes alcoholic hepatitis, hyperlipidemia, hemolytic anemia, jaundice, and abdominal pain.83 Although there is no specific treatment, alcohol cessation can result in resolution of the hemolytic anemia that is a hallmark feature of the syndrome.
Anemia Associated with Viral Hepatitis and its Treatment Hematologic complications may arise from the different forms of viral hepatitis as well as their treatments. Aplastic anemia is a condition in which hematopoiesis is compromised, and there is a paucity of normal bone marrow components. Although congenital forms of this hypocellular bone marrow state exist, a majority of cases are acquired. It is thought to have an immune-mediated pathophysiology and has also been associated with viral hepatitis. The precise mechanism of this association is the subject of ongoing study.84 A T cell-mediated destruction of important hemopoietic cells has been described.85,86 In hepatitis-associated aplastic anemia a genderspecific factor may be involved, as most cases have been described in young males, and patients typically present with pancytopenia within 3 months after the onset of hepatitis.87,88 This condition is potentially fatal, but combinations of immunosuppressive regimens have been successful. Approximately half of patients with severe cases treated with antithymocyte globulin (ATG) and ciclosporin have a sustained recovery.89 Despite the effectiveness of immunosuppressive therapy, corticosteroid use has not been effective and is not recommended.75 If human leukocyte antigen (HLA)-matched donors exist, survival rates are even higher with bone marrow transplantation, and results are generally better in younger patients.90 Viral hepatitis has also been associated with abnormalities in individual hematopoietic cell lines. Descriptions exist of different forms of viral hepatitis causing pure red cell aplasia, but this is a rare clinical association.91,92 Viral hepatitis in general has also been associated with both neutropenia and thrombocytopenia.93,94 Hepatitis C (HCV) has also been suggested to cause hemolytic anemia, neutropenia, and thrombocytopenia in the absence of other clear etiologies.95,96 Although the mechanism of the direct virologic effect of hepatitis viruses on hematopoietic cell lines requires further study, adverse hematologic effects have been described with antiviral therapy. Pegylated interferon-a and ribavirin therapy for HCV is associated with significant decreases in hemoglobin concentrations.97,98 Both drugs have specific mechanisms by which they cause these side effects. Interferon-a has a direct myelosuppressive effect, whereas ribavirin is associated with a dose-dependent hemolytic anemia.99,100 This hemolysis can be severe, and clinically evident hemoglobinuria may occur.101 A recent report also implicates ribavirin itself in causing a pure red cell aplasia.102
Hematologic Issues in Liver Transplantation Patients awaiting liver transplantation have complex medical problems, and anemia in this setting is usually multifactorial.4 The aforementioned hematologic complications and etiologies in liver
Chapter 25 HEMATOPOIETIC ABNORMALITIES AND HEMOSTASIS
disease can all be combined in the transplant patient. Hemolysis, gastrointestinal bleeding, drug side effects, renal insufficiency, and hematopoietic abnormalities may all play a role. Although values vary depending on the definition of anemia, post-transplant anemia has been described in up to 28.2% of these patients, and this anemia is also thought to be multifactorial.5,103 Post-transplant immunosuppressive regimens are clear contributors to these abnormalities.104,105 As previously mentioned, aplastic anemia refers to markedly reduced numbers or the absence of precursors of all cell lines in the bone marrow. Although this is classically associated with parvovirus B19 infection, this entity has also been reported after liver transplantation, and a higher incidence is associated with patients who underwent liver transplantation for fulminant viral hepatitis.106,107 Although rare, with a described incidence of approximately 1%, graft-versus-host disease (GVHD) should also be considered in patients who develop pancytopenia following both cadaveric and living donor transplant.108,109 Increased fibrinolysis may also be seen in the perioperative period, and antifibrinolytic therapy with aprotinin and tranexamic acid has been used with success in reducing blood product requirements.110–112 However, more evidence and well-designed studies are needed before widespread use can be advocated. Similar to the treatment of other complications of cirrhosis, orthotopic liver transplantation is considered to be the ultimate treatment for some disorders of hemopoiesis and hemostasis in patients with cirrhosis. Liver transplantation is discussed extensively elsewhere in this book, but from hemophilia to fibrinolysis, evidence exists that derangements in the coagulation cascade and their complications improve after transplantation of a healthy liver.113,114 Recent evidence cites improvement in hypersplenism with living donor transplantation as well.115
Alcohol and Anemia Alcohol is an important factor in the hematopoietic abnormalities seen in patients with liver disease. Alcohol has a broad bone marrow suppressive effect and can result in anemia, leukopenia, and thrombocytopenia.116,117 Pancytopenia is not common in alcoholics, but production of all three cell lines can be impaired, and this impairment is reversible.118 The aforementioned nutritional deficiencies that are common in alcoholics can also contribute to anemia in these patients. Furthermore, if patients develop alcoholic cirrhosis, portal hypertension, and splenic sequestration of platelets, thrombocytopenia may also be present. Alcohol is also a toxin that has a selective influence on the suppression of megakaryocytes.119–121 It has also been suggested that platelet function is impaired by alcohol in vitro.122
ment to sites of infection may be an important component of this reduced leukocyte function.125,126 Evidence also exists that compromised phagocytic function of neutrophil granulocytes may contribute to this susceptibility to bacterial infections.127,128 Furthermore, any bone marrow suppressive effects of toxins, drugs, and viral infections at play in the patient with liver disease will negatively affect this cell line as well.
THROMBOCYTOPENIA AND HYPERSPLENISM Thrombocytopenia is one of the most common laboratory abnormalities in patients with cirrhosis. It has been found to be an independent predictor of complications of portal hypertension, such as esophageal varices and splenomegaly.129 Thrombopoietin (TPO) is produced predominantly in the liver and is the primary regulator of platelet production.130 Acquired TPO deficiency may be a contributing factor to the thrombocytopenia seen in cirrhotic patients. Evidence suggests that patients with cirrhosis have lower TPO levels than those with non-cirrhotic liver disease.131 Moreover, TPO production appears to be related to the amount of functional liver cell mass, and some describe decreased TPO levels in proportion to the severity of the liver disease.132,133 After orthotopic liver transplantation, patients experience an increase in TPO production and a subsequent increase in bone marrow production of platelets.134 However, other studies refute these findings by not showing differences in TPO levels between patients with liver disease and controls, and study in this field is ongoing.135 Autoantibodies to GP IIb/IIIa are also implicated in immune thrombocytopenic purpura (ITP) seen in patients with cirrhosis.136 ITP is seen in association with viral hepatitis, and specifically with HCV infection.137,138 Further evidence for an immune-mediated component to the thrombocytopenia in liver disease is seen by increased levels of plateletassociated immunoglobulins in these patients.119 Hypersplenism refers to cytopenia, such as thrombocytopenia or leukopenia, as a result of platelet and leukocyte destruction and/or sequestration in the spleen.139 Hypersplenism has also been described as an independent risk factor for complications of portal hypertension, such as variceal bleeding.140 The prevalence of hypersplenism in cirrhosis ranges from 2% to 61%, depending on the definitions of leukopenia and thrombocytopenia.141 Decompression of portal hypertension, such as with transjugular intrahepatic portosystemic shunt (TIPS) or liver transplantation, can improve the consequences of hypersplenism, although not predictably. Table 25-2 lists common causes of hypersplenism in patients with liver disease.
COAGULOPATHY Abnormal coagulation in patients with liver disease often facilitates bleeding because of the accompanying thrombocytopenia and other
Leukocyte Abnormalities Patients with liver disease commonly demonstrate leukopenia and atypical leukocyte function. As with other cell lines, these findings are also likely to be multifactorial. Leukopenia may result from a combination of decreased leukocyte production, splenic sequestration, and impaired survival. Neutrophil survival appears to be shortened via apoptosis.123,124 In the setting of both decreased leukocyte counts and suboptimal leukocyte function, patients with cirrhosis are susceptible to bacterial infections. Ineffective neutrophil recruit-
Table 25-2. Causes of Hypersplenism Related to Liver Disease Cirrhosis Congestive hepatopathy Budd–Chiari syndrome Acute viral hepatitis Amyloidosis Portal vein thrombosis Hepatic schistosomiasis
493
Section III. Clinical Consequences of Liver Disease
potential contributing factors, such as disseminated intravascular coagulation (DIC) and fibrinolysis. Coagulation disorders are possible but uncommon in uncomplicated acute hepatitis. Furthermore, tests of coagulation are usually normal until levels of clotting factors fall to 30–40% of normal values.15 In sepsis, not only may coagulopathy be related to impaired synthesis of clotting factors, but consumption of clotting factors may also be present.142 Vitamin K is fat-soluble vitamin and a required cofactor for gcarboxylation of clotting factors II, VII, IX, and X. Bile salts are an important component of fat-soluble vitamin metabolism, and biliary obstruction and small bowel bacterial overgrowth can contribute to vitamin K deficiency.143 Factor VII has the shortest half-life of all of the clotting factors, and recombinant factor VII, as discussed below, may be a useful adjunct to conventional therapies when managing the coagulopathy of liver disease. The prothrombin time (PT) and international normalized ratio (INR) are the most commonly used tests used to assess the production of coagulation factors by the liver and to diagnose vitamin K deficiency. The PT/INR is also included in both the Model for End-Stage Liver Disease (MELD) and Child–Pugh classifications to assess liver disease severity.144 PT/INR values are also used in the King’s College criteria for predicting prognosis in fulminant hepatic failure (FHF).145 Administering vitamin K may gradually improve the PT and INR in the setting of vitamin K deficiency, but this treatment is not likely to be helpful during acute bleeding in the patient with liver disease and a coagulopathy.146
THROMBOTIC DISORDERS Although patients with liver disease commonly have the aforementioned derangement of the coagulation cascade leading to coagulopathy, some are actually at risk for developing thrombotic disorders.147 Endogenous anticoagulation proteins C, S, and antithrombin III are also produced in the liver, and synthetic dysfunction impairs normal production.148 Although the levels of these anticoagulant proteins in patients with liver disease may be as low as those seen with inherited deficiencies, thrombosis in inherited deficiencies is much more frequent. Low levels of these proteins may contribute to intrahepatic vein thrombosis and inflammation, which will accelerate fibrosis in patients with liver disease.149 Patients with cholestatic liver disease, such as primary sclerosing cholangitis (PSC), also demonstrate a potential for hypercoagulability. A difference in platelet function compared to noncholestatic liver disease may contribute to this association.150 Patients with HCC have hypercoagulable potential similar to that of patients with other solid organ malignancies. Furthermore, HCC is associated with a higher incidence of portal and hepatic vein thrombosis.151 Finally, other gene mutations, such as mutation 20210 of the prothrombin gene, cause thrombophilia and are described as an independent risk factor for portal vein thrombosis.152,153 Multiple prothrombotic factors may cumulatively increase the risk of venous thrombosis when present in the same patient with liver disease.
be life-threatening if urgent therapy is not administered.154 Although varices are covered extensively elsewhere in this book, abnormalities in hemopoiesis and hemostasis are intimately involved in the clinical scenarios involving esophageal, gastric, small bowel, and rectal varices and other related bleeding etiologies, such as portal hypertensive gastropathy.155 The acutely bleeding patient is often coagulopathic and thrombocytopenic, and resuscitation with blood products is often a necessary prerequisite to endoscopic management.
DISSEMINATED INTRAVASCULAR COAGULATION AND PRIMARY FIBRINOLYSIS Disseminated intravascular coagulation (DIC) refers to intravascular fibrin deposition in small and medium-sized vessels secondary to systemic activation of the coagulation cascade.156 No specific set of coagulation studies is pathognomonic for DIC, but patients typically have a prolonged PT, PTT, elevated fibrin degradation products (FDP), and low fibrinogen level.157 Hyperbilirubinemia in this setting may be related to hemolysis, underlying liver disease, or a combination of these factors. Patients with cirrhosis frequently have the same laboratory abnormalities as are present in DIC, leading to a debate as to whether or not DIC is part of the natural history of cirrhotic coagulopathy.158 However, patients with cirrhosis are certainly at risk for developing the known causes of DIC, such as sepsis and shock, and as with DIC in other patients, treatment of the underlying cause is the prevailing recommendation. Although an infusion of fresh frozen plasma (FFP) is helpful in transiently improving PT and PTT in this setting, fibrinogen levels usually stay low. Cryoprecipitate transfusion is useful in increasing the serum fibrinogen level when it is less than 100 mg/dl.159,160 However, cryoprecipitate can be misused, and it should not be transfused to treat a coagulopathy secondary to liver disease in the absence of hypofibrinogenemia.161 Tissue plasminogen activator (tPA) and a2-antiplasmin play an important role in the endogenous fibrinolytic system. Accelerated fibrinolysis has long been described in patients with cirrhosis. Fibrinolysis is suggested by a shortened whole blood euglobulin clot lysis time (ELT) and a low fibrinogen level. When severe, fibrinolysis can also cause increased levels of D-dimers and FDP. One study described hyperfibrinolysis in 31% of patients with cirrhosis and correlated the degree of abnormality with the severity of liver disease.162 Decreased hepatic clearance of tPA in cirrhosis, combined with decreased levels of regulatory agents such as a2-antiplasmin and thrombin-activatable fibrinolysis inhibitor (TAFI) contribute to the fibrinolysis, although not everyone agrees with their importance.163–170 e-Aminocaproic acid has been used for fibrinolysis associated with bleeding in cirrhosis, but unlike its use in cardiac surgery there is insufficient evidence for its widespread use.
TREATMENT OPTIONS TRANSFUSION OF BLOOD PRODUCTS
GASTROINTESTINAL BLEEDING Gastrointestinal bleeding, and particularly variceal bleeding, is one of the most serious complications of cirrhosis. Variceal bleeding can
494
Packed red blood cell (pRBC), FFP, cryoprecipitate, and platelet transfusions are important aspects of treatment when combating hemopoietic and hemostatic abnormalities. From performing per-
Chapter 25 HEMATOPOIETIC ABNORMALITIES AND HEMOSTASIS
cutaneous liver biopsy to gastrointestinal endoscopy, clinicians often have concern about bleeding in patients with liver disease. Because patients who require paracentesis often have cirrhosis as well as abnormal coagulation studies, excluding these patients because of the coagulopathy would mean a number of important diagnostic procedures would not be performed. The information obtained by these procedures may be critical to patient management, and the use of alternative approaches, such as transjugular liver biopsy or resuscitation with blood products to temporarily improve the coagulopathy, may be necessary.164 A recent study supports increasing the number of FFP units used in the initial management of coagulopathy in patients with chronic liver disease.165 However, in the absence of bleeding, pending invasive procedures, or other indications, clinicians must remember that adverse effects are possible with any transfusion, and these resources should be used judiciously.166 The effect of FFP on coagulation studies is rapid but transient and usually lasts 12–24 hours.15 Numerous reports exist detailing the use of recombinant human factor VIIa (rhFVIIa) to treat bleeding or correct coagulopathy prior to performing invasive procedures, but controlled trials are lacking.167–177 Owing to an increased risk of complications, clinical judgment should be exercised when considering invasive procedures such as surgery and percutaneous liver biopsy in the setting of a coagulopathy and/or thrombocytopenia. Administration of subcutaneous vitamin K is helpful in improving the PT and INR in patients with vitamin K deficiency, and this improvement may be all that is necessary prior to an elective outpatient liver biopsy. However, in the bleeding patient and in the absence of vitamin K deficiency, FFP transfusion should be the treatment of choice to correct this coagulopathy. Resuscitation with platelet transfusions and FFP should be considered to increase the platelet count to more than 50 000 and to lower the INR to less than 1.5 prior to performing invasive procedures.168,169 A transjugular liver biopsy by an interventional radiologist can be considered as an alternative when the INR cannot be easily corrected and liver biopsy is indicated.170,171
OTHER THERAPIES Because of practical issues regarding bone marrow suppression with interferon-based antiviral regimens and hemolysis with ribavirin, underlying anemia makes the treatment of viral hepatitis extremely challenging. Hematopoietic growth factors may have a role in the management of the hematologic abnormalities associated with the IFN-a and ribavirin treatments for HCV.172 Dose reductions of ribavirin are often necessary in the face of ribavirin-induced hemolytic anemia.173 Dose reductions of ribavirin may hinder efforts to obtain a sustained virologic response, and use of epoetin-a appears to be effective in combating the anemia associated with ribavirin therapy.174,175 Once-weekly dosing of epoetin-a may be an alternative to blood transfusions or ribavirin dose reductions, but future study is needed in this area.176 A recent prospective double-blind randomized controlled trial has demonstrated quality-of-life improvements in patients receiving epoetin-a compared with placebo controls.177 Although it is helpful in the hemolytic anemia related to ribavirin therapy, further study is needed to evaluate the use of epoetin-a in
post-transplant anemia. Prior to transplantation, cirrhotic patients demonstrate an inappropriate erythropoietin response to anemia that may improve after transplantation.178 With respect to thrombocytopenia not related to hypersplenism, a role might exist for antiviral therapy as well. Although controversial, published case reports suggest that interferon-a may be effective in HCV-related thrombocytopenia.179 Recombinant human thrombopoietin may also be a useful option when treating thrombocytopenia in patients with bleeding or impending invasive procedures. However, evidence in this field is in the early stages, and future study is required.180,181 DDAVP (1deamino-8-D-arginine vasopressin), also known as desmopressin, is also used to facilitate hemostasis in cirrhotic patients with coagulopathy. Although the mechanism is still not well defined, desmopressin administration can increase levels of vWF and factor VIII and shorten the bleeding time.182 Despite these changes, there are conflicting case series on the clinical impact of DDAVP on bleeding in cirrhosis, and more rigorous studies are needed.183
REFERENCES 1. Wintrobe MM, Schumacker HS Jr. Occurrence of macrocytic anemia in association with disorder of the liver. Bull Johns Hopkins Hosp 1933;52:387–404. 2. Wintrobe MM. Relation of disease of liver to anemia. Arch Intern Med 1936;57:289–302. 3. Berman L, Axelrod AR, Horan TN. The blood and bone marrow in patients with cirrhosis of the liver. Blood 1949;4:511–533. 4. Patt CH, Fairbanks KD, Thuluwath PJ. Renal insufficiency may partly explain chronic anemia in patients awaiting liver transplantation. Dig Dis Sci 2004;49:629–632. 5. Maheshwari A, Mishra R, Thuluvath PJ. Post-liver-transplant anemia: etiology and management. Liver Transpl 2004;10:165–173. 6. Misra S, Moore TB, Ament ME, et al. Profile of anemia in children after liver transplantation. Transplantation 2000;70:1459–1463. 7. Wiesner R, Rabkin J, Klintmalm G, et al. A randomized doubleblind comparative study of mycophenolate mofetil and azathioprine in combination with cyclosporine and corticosteroids in primary liver transplant recipients. Liver Transpl 2001;7:442–450. 8. Palis J, Yoder MC. Yolk-sac hematopoiesis: the first blood cells of mouse and man. Exp Hematol 2001;29:927–936. 9. Palis J, Segal GB. Developmental biology of erythropoiesis. Blood Rev 1998;12:106–114. 10. Chagraoui J, Lepage-Noll A, Anjo A, et al. Fetal liver stroma consists of cells in epithelial-to-mesenchymal transition. Blood 2003;101:2973–2982. 11. Lacombe C, Mayeux P. Biology of erythropoietin. Haematologica 1998;83:724–732. 12. Jelkmann W. The role of the liver in the production of thrombopoietin compared with erythropoietin. Eur J Gastroenterol Hepatol 2001;13:791–801. 13. Basu S, Dunn A, Ward A. G-CSF: function and modes of action. Int J Mol Med 2002;10:3–10. 14. Quesenberry PJ, Colvin GA. Hematopoiesis. In: Braunwald E, Fauci FS, Kasper DL, et al., eds. Harrison’s principles of internal medicine, 15th edn. New York: McGraw-Hill, 2001: 653–660. 15. Amitrano L, Guardascione MA, Brancaccio V, et al. Coagulation disorders in liver disease. Semin Liver Dis 2002;22:83–96.
495
Section III. Clinical Consequences of Liver Disease
16. Roberts HR, Monroe DM, Escobar MA. Current concepts of hemostasis. Anesthesiology 2004;100:722–730. 17. Lisman T, Leebeek FWG, de Groot PG. Haemostatic abnormalities in patients with liver disease. J Hepatol 2002;37:280–287. 18. Savage B, Shattil SJ, Ruggeri ZM. Modulation of platelet function through adhesion receptors: a dual role for glycoprotein IIb–IIIa (integrin alpha IIb beta 3) mediated by fibrinogen and glycoprotein Ib-von Willebrand factor. J Biol Chem 1992;267:11300–11306. 19. Giesen PLA, Rauch U, Bohrmann B, et al. Blood-borne tissue factor: another view of thrombosis. Proc Natl Acad Sci USA 1999;96:2311–2315. 20. Rauch U, Nemerson Y. Circulating tissue factor and thrombosis. Curr Opin Hematol 2000;7:273–277. 21. Bajzar L, Manuel R, Nesheim ME. Purification and characterization of TAFI, a thrombin-activatable fibrinolysis inhibitor. J Biol Chem 1995;270:14477–14484. 22. Caprini JA, Glase CJ, Anderson CB, et al. Laboratory markers in the diagnosis of venous thromboembolism. Circulation 2004;109:I4–8. 23. Topol E. Recent advances in anticoagulant therapy for acute coronary syndromes. Am Heart J 2001;142:S22–9. 24. Roemisch J, Gray E, Hoffman JN, et al. Antithrombin: a new look at the actions of a serine protease inhibitor. Blood Coag Fibrinol 2002;13:657–670. 25. Collen D. The plasminogen (fibrinolytic) system. Thromb Haemost 1999;82:259–270. 26. Lijnen HR. Elements of the fibrinolytic system. Ann NY Acad Sci 2001;936:226–236. 27. Favier R, Aoki N, de Moerloose P. Congenital alpha(2)-plasmin inhibitor deficiencies: a review. Br J Haematol 2001;114:4–10. 28. Brittenham GM. Disorders of iron metabolism: iron deficiency and overload. In: Hoffman R, Benz EJ, Shattil SJ, et al., eds. Hematology: basic principles and practice, 3rd edn. New York: Churchill Livingstone, 2000: 397–428. 29. Andrews NC. Disorders of iron metabolism. N Engl J Med 1999;341:1986–1995. 30. Thuluvath PJ, Yoo HY. Portal hypertensive gastropathy. Am J Gastroenterol 2002;97:2973–2978. 31. Burak KW, Lee SS, Beck PL. Portal hypertensive gastropathy and gastric antral vascular ectasia (GAVE) syndrome. Gut 2001;49:866–872. 32. Jones KA, Abramowicz JS, Anissi D, et al. Severe HELLP syndrome presenting with acute gum bleeding following toothbrushing at 38 weeks’ gestation. Am J Crit Care 1993;2:395–396. 33. Sandoval C, Dong S, Visintainer P, et al. Clinical and laboratory features of 178 children with recurrent epistaxis. J Pediatr Hematol Oncol 2002;24:47–49. 34. Tran TN, Eubanks SK, Schaffer KJ, et al. Secretion of ferritin by rat hepatoma cells and its regulation by inflammatory cytokines and iron. Blood 1997;90:4979–86. 35. Pietrangelo A. Hereditary hemochromatosis––a new look at an old disease. N Engl J Med 2004;350:2383–2397. 36. Tavill AS. Diagnosis and management of hemochromatosis. Hepatology 2001;33:1321–1328. 37. Alcindor T, Bridges KR. Sideroblastic anaemias. Br J Haematol 2002;116:733–743. 38. Larkin EC, Watson-Williams EJ. Alcohol and the blood. Med Clin North Am 1984;68:105–120. 39. Maruyama S, Hirayama C, Yamamoto S, et al. Red blood cell status in alcoholic and non-alcoholic liver disease. J Lab Clin Med 2001;138:332–337. 40. Wickramasinghe SN. The wide spectrum and unresolved issues of megaloblastic anemia. Semin Hematol 1999;36:3–18. 41. Zittoun J, Zittoun R. Modern clinical testing strategies in cobalamin and folate deficiency. Semin Hematol 1999;36:35–46.
496
42. Babior BM. Metabolic aspects of folic acid and cobalamin. In: Beutler E, Lichtman MA, Coller BS, et al., eds. Williams’ hematology, 6th edn. New York: McGraw-Hill, 2001: 305–318. 43. Bergheim I, Parlesak A, Dierks C, et al. Nutritional deficiencies in German middle-class male alcohol consumers: relation to dietary intake and severity of liver disease. Eur J Clin Nutr 2003;57:431–438. 44. Baker H, Leevy CB, DeAngelis B, et al. Cobalamin (vitamin B12) and holotranscobalamin changes in plasma and liver tissue in alcoholics with liver disease. J Am Coll Nutr 1998;17:235–238. 45. Antony AC. Vegetarianism and vitamin B-12 (cobalamin) deficiency. Am J Clin Nutr 2003;78:3–6. 46. Ermens AA, Vlasveld LT, Lindemans J. Significance of elevated cobalamin (vitamin B12) levels in blood. Clin Biochem 2003;36:585–590. 47. Hagelskjaer L, Rasmussen K. Methylmalonic acid concentration in serum not affected in hepatic disease. Clin Chem 1992;38:493–495. 48. Halsted CH, Villanueva JA, Devlin AM. Folate deficiency, methionine metabolism, and alcoholic liver disease. Alcohol 2002;27:169–172. 49. Donnelly JG. Folic acid. Crit Rev Clin Lab Sci. 2001;38:183–223. 50. Fowler B. The folate cycle and disease in humans. Kidney Int 2001;78(Suppl):S221–229. 51. Lindenbaum J, Roman MJ. Nutritional anemia in alcoholism. Am J Clin Nutr 1980;33:2727–2735. 52. Lindenbaum J. Hematologic complications of alcohol abuse. Semin Liver Dis 1987;7:169–181. 53. Halsted CH, Villanueva JA, Devlin AM, et al. Folate deficiency disturbs hepatic methionine metabolism and promotes liver injury in the ethanol-fed micropig. Proc Natl Acad Sci USA 2002;99:10072–10077. 54. Steinberg SE. Mechanisms of folate homeostasis. Am J Physiol 1984;246:G319–324. 55. Shin HC, Takakuwa F, Shimoda M, et al. Enterohepatic circulation kinetics of bile-active folate derivatives and folate homeostasis in rats. Am J Physiol 1995;269:R421–425. 56. Collins TD, Eisenga BH, Bhandari SD, et al. Effects of ethanol on tissue folate incorporation and recovery from folate deficiency in rats. Alcohol Clin Exp Res 1992;16:757–763. 57. Hidiroglou N, Camilo ME, Beckenhauer HC, et al. Effect of chronic alcohol ingestion on hepatic folate distribution in the rat. Biochem Pharmacol 1994;47:1561–1566. 58. Halsted CH, Villanueva JA, Devlin AM, et al. Metabolic interactions of alcohol and folate. J Nutr 2002;132(Suppl): 2367S–2372S. 59. Villanueva JA, Devlin AM, Halsted CH. Reduced folate carrier: tissue distribution and effects of chronic ethanol intake in the micropig. Alcohol Clin Exp Res 2001;25:415–420. 60. Morgan MY, Camilo ME, Luck W, et al. Macrocytosis in alcoholrelated liver disease: its value for screening. Clin Lab Haematol 1981;3:35–44. 61. Fink A, Hays RD, Moore AA, et al. Alcohol-related problems in older persons. Determinants, consequences, and screening. Arch Intern Med 1996;156:1150–1156. 62. Savage DG, Ogundipe A, Allen RH, et al. Etiology and diagnostic evaluation of macrocytosis. Am J Med Sci 2000;319:343–352. 63. Halsall S, Peters TJ. Erythrocyte and leukocyte lipids in alcoholic macrocytosis. Clin Chim Acta 1985;149:165–173. 64. Benedetti A, Birarelli AM, Brunelli E, et al. Modification of lipid composition of erythrocyte membranes in chronic alcoholism. Pharmacol Res Commun 1987;19:651–662. 65. Luo JC, Hwang SJ, Wu JC, et al. Clinical characteristics and prognosis of hepatocellular carcinoma patients with paraneoplastic syndromes. Hepatogastroenterology 2002;49:1315–1319.
Chapter 25 HEMATOPOIETIC ABNORMALITIES AND HEMOSTASIS
66. Trotter JF, Cohn A, Grant R. Erythrocytosis in a patient with hepatocellular carcinoma. J Clin Gastroenterol 2002;35:365–366. 67. Cheng TI, Tsou MH, Yang PS, et al. Dermatomyositis and erythrocytosis associated with hepatocellular carcinoma. J Gastroenterol Hepatol 2002;17:1239–1240. 68. Tacke F, Schoffski P, Luedde T, et al. Analysis of factors contributing to higher erythropoietin levels in patients with chronic liver disease. Scand J Gastroenterol 2004;39:259–266. 69. Cooper RA. Hemolytic syndromes and red cell membrane abnormalities in liver disease. Semin Hematol 1980;17:103–112. 70. Morse EE. Mechanisms of hemolysis in liver disease. Ann Clin Lab Sci 1990;20:169–174. 71. Shohet SB, Ness PM. Hemolytic anemias. Failure of the red cell membrane. Med Clin North Am 1976;60:913–932. 72. Shiraishi K, Matsuzaki S, Ishida H, et al. Impaired erythrocyte deformability and membrane fluidity in alcoholic liver disease: participation in disturbed hepatic microcirculation. Alcohol Alcoholism 1993;1A(Suppl):59–64. 73. Douglass CC, McCall MS, Frenel EP. The acanthocyte in cirrhosis with hemolytic anemia. Ann Intern Med 1968;68:390–397. 74. Biemer JJ. Acanthocytosis – biochemical and physiological considerations. Ann Clin Lab Sci 1980;10:238–249. 75. Chitale AA, Sterling RK, Post AB, et al. Resolution of spur cell anemia with liver transplantation: a case report and review of the literature. Transplantation 1998;65:993–995. 76. Malik P, Bogetti D, Sileri P, et al. Spur cell anemia in alcoholic cirrhosis: cure by orthotopic liver transplantation and recurrence after liver graft failure. Int Surg 2002;87:201–204. 77. Silber R, Amorosi E, Lhowe J, et al. Spur-shaped erythrocytes in Laennec’s cirrhosis. N Engl J Med 1966;275:639–643. 78. Cooper RA. Anemia with spur cells: a red cell defect acquired in serum and modified in the circulation. J Clin Invest 1969;48:1820–1831. 79. Gow PJ, Smallwood RA, Angus PW, et al. Diagnosis of Wilson’s disease: an experience over three decades. Gut 2000;46:415–419. 80. Loudianos G, Gitlin JD. Wilson’s disease. Semin Liver Dis 2000;20:353–364. 81. Kiss JE, Berman D, Van Thiel D. Effective removal of copper by plasma exchange in fulminant Wilson’s disease. Transfusion 1998;38:327–331. 82. Matsumura A, Hiraishi H, Terano A. Plasma exchange for hemolytic crisis in Wilson disease. Ann Intern Med 1999;131:866. 83. Zieve L, Jaundice, hyperlipidemia and hemolytic anemia: A heretofore unrecognized syndrome associated with alcoholic fatty liver and cirrhosis. Ann Intern Med 1958;48:471–496. 84. Young NS. Acquired aplastic anemia. Ann Intern Med 2002;136:534–546. 85. Young NS, Maciejewski J. The pathophysiology of acquired aplastic anemia. N Engl J Med 1997;336:1365–1372. 86. Lu J, Basu A, Melenhorst JJ, et al. Analysis of T-cell repertoire in hepatitis-associated aplastic anemia. Blood 2004;103:4588–4593. 87. Hagler L, Pastore RA, Bergin JJ, et al. Aplastic anemia following viral hepatitis: report of two fatal cases and literature review. Medicine (Baltimore) 1975;54:139–164. 88. Brown KE, Tisdale J, Barrett AJ, et al. Hepatitis-associated aplastic anemia. N Engl J Med 1997;336:1059–1064. 89. Rosenfeld S, Follmann D, Nunez O, et al. Antithymocyte globulin and cyclosporine for severe aplastic anemia: association between hematologic response and long-term outcome. JAMA 2003;289:1130–1135. 90. Ades L, Mary JY, Robin M, et al. Long-term outcome after bone marrow transplantation for severe aplastic anemia. Blood 2004;103:2490–2497.
91. Tomida S, Matsuzaki Y, Nishi M, et al. Severe acute hepatitis A associated with acute pure red cell aplasia. J Gastroenterol 1996;31:612–617. 92. al-Awami Y, Sears DA, Carrum G, et al. Pure red cell aplasia associated with hepatitis C infection. Am J Med Sci 1997;314:113–117. 93. Boxer LA, Yokoyama M, Wiebe RA. Autoimmune neutropenia associated with chronic active hepatitis. Am J Med 1972;52:279–282. 94. Copley JB, Waterfield WC. Thrombocytopenia with increased bone marrow megakaryocytes in two patients with viral hepatitis. South Med J 1982;75:1146–1148. 95. Young NS. Flaviviruses and bone marrow failure. JAMA 1990;263:3065–3068. 96. Ramos-Casals M, Garcia-Carrasco M, Lopez-Medrano F, et al. Severe autoimmune cytopenias in treatment-naive hepatitis C virus infection: clinical description of 35 cases. Medicine (Baltimore) 2003;82:87–96. 97. Gaeta GB, Precone DF, Felaco FM, et al. Premature discontinuation of interferon plus ribavirin for adverse effects: a multicentre survey in ‘real world’ patients with chronic hepatitis C. Aliment Pharmacol Ther 2002;16:1633–1639. 98. Sulkowski MS, Wasserman R, Brooks L, et al. Changes in haemoglobin during interferon alpha-2b plus ribavirin combination therapy for chronic hepatitis C virus infection. J Viral Hepatol 2004;11:243–250. 99. Soza A, Everhart JE, Ghany MG, et al. Neutropenia during combination therapy of interferon alfa and ribavirin for chronic hepatitis C. Hepatology 2002;36:1273–1279. 100. Grattagliano I, Russmann S, Palmieri VO, et al. Low membrane protein sulfhydrils but not G6PD deficiency predict ribavirininduced hemolysis in hepatitis C. Hepatology 2004;39:1248–1255. 101. Massoud OI, Yousef WI, Mullen KD. Hemoglobinuria with ribavirin treatment. J Clin Gastroenterol 2003;36: 367–368. 102. Tanaka N, Ishida F, Tanaka E. Ribavirin-induced pure red-cell aplasia during treatment of chronic hepatitis C. N Engl J Med 2004;350:1264–1265. 103. Iglesias-Berengue J, Lopez-Espinosa JA, Ortega-Lopez J, et al. Hematologic abnormalities in liver-transplanted children during medium- to long-term follow-up. Transplant Proc 2003;35:1904–1906. 104. Nosari A, Marbello L, De Carlis LG, et al. Bone marrow hypoplasia complicating tacrolimus (FK506) therapy. Int J Hematol 2004;79:130–132. 105. Danesi R, Del Tacca M. Hematologic toxicity of immunosuppressive treatment. Transplant Proc 2004;36:703–704. 106. Stock PG, Steiner ME, Freese D, et al. Hepatitis-associated aplastic anemia after liver transplantation. Transplantation 1987;43:595–597. 107. Itterbeek P, Vandenberghe P, Nevens F, et al. Aplastic anemia after transplantation for non-A, non-B, non-C fulminant hepatic failure: case report and review of the literature. Transpl Int 2002;15:117–123. 108. Smith DM, Agura E, Netto G, et al. Liver transplant-associated graft-versus-host disease. Transplantation 2003;75:118–126. 109. Soejima Y, Shimada M, Suehiro T, et al. Graft-versus-host disease following living donor liver transplantation. Liver Transpl 2004;10:460–464. 110. Wells PS. Safety and efficacy of methods for reducing perioperative allogeneic transfusion: a critical review of the literature. Am J Ther 2002;9:377–388. 111. Dalmau A, Sabate A, Koo M, et al. The prophylactic use of tranexamic acid and aprotinin in orthotopic liver transplantation: a comparative study. Liver Transpl 2004;10:279–284.
497
Section III. Clinical Consequences of Liver Disease
112. Porte RJ. Antifibrinolytics in liver transplantation: they are effective, but what about the risk–benefit ratio? Liver Transpl 2004;10:285–288. 113. Gordon FH, Mistry PK, Sabin CA, et al. Outcome of orthotopic liver transplantation in patients with haemophilia. Gut 1998;42:744–749. 114. Kahl BS, Schwartz BS, Mosher DF. Profound imbalance of profibrinolytic and anti-fibrinolytic factors (tissue plasminogen activator and plasminogen activator inhibitor type 1) and severe bleeding diathesis in a patient with cirrhosis: correction by liver transplantation. Blood Coag Fibrinol 2003;14:741–744. 115. Kaneko J, Sugawara Y, Akamatsu N, et al. Spleen volume and platelet number changes after living donor liver transplantation in adults. Hepatogastroenterology 2004;51:262–263. 116. Ballard HS. Alcohol-associated pancytopenia with hypocellular bone marrow. Am J Clin Pathol 1980;73:830–834. 117. Latvala J, Parkkila S, Niemela O. Excess alcohol consumption is common in patients with cytopenia: studies in blood and bone marrow cells. Alcohol Clin Exp Res 2004;28:619–624. 118. Nakao S, Harada M, Kondo K, et al. Reversible bone marrow hypoplasia induced by alcohol. Am J Hematol 1991;37: 120–123. 119. Gewirtz AM, Hoffman R. Transitory hypomegakaryocytic thrombocytopenia: aetiological association with ethanol abuse and implications regarding regulation of human megakaryocytopoiesis. Br J Haematol 1986;62:333–344. 120. Levine RF, Spivak JL, Meagher RC, et al. Effect of ethanol on thrombopoiesis. Br J Haematol 1986;62:345–354. 121. Smith CM 2nd, Tobin JD Jr, Burris SM, et al. Alcohol consumption in the guinea pig is associated with reduced megakaryocyte deformability and platelet size. J Lab Clin Med 1992;120:699–706. 122. Rand ML, Jakubowski JA, Fisher MJ, et al. Ethanol enhances the inhibitory effect of an oral GPIIb/IIIa antagonist on human platelet function. J Lab Clin Med 2002;140:391–397. 123. Kusaba N, Kumashiro R, Ogata H, et al. In vitro study of neutrophil apoptosis in liver cirrhosis. Intern Med 1998;37:11–17. 124. Aref SE, Mahmoud LA, El Refie MF, et al. Assessment of neutrophil apoptosis ex vivo in hepatosplenic patients with neutropenia pre and post splenectomy. Hematology 2003;8:265–272. 125. Fiuza C, Salcedo M, Clemente G, et al. In vivo neutrophil dysfunction in cirrhotic patients with advanced liver disease. J Infect Dis 2000;182:526–533. 126. Fiuza C, Salcedo M, Clemente G, et al. Granulocyte colonystimulating factor improves deficient in vitro neutrophil transendothelial migration in patients with advanced liver disease. Clin Diagn Lab Immunol 2002;9:433–439. 127. Trevisani F, Castelli E, Foschi FG, et al. Impaired tuftsin activity in cirrhosis: relationship with splenic function and clinical outcome. Gut 2002;50:707–712. 128. Garfia C, Garcia-Ruiz I, Solis-Herruzo JA. Deficient phospholipase C activity in blood polymorphonuclear neutrophils from patients with liver cirrhosis. J Hepatol 2004;40:749–756. 129. Thomopoulos KC, Labropoulou-Karatza C, Mimidis KP, et al. Non-invasive predictors of the presence of large oesophageal varices in patients with cirrhosis. Dig Liver Dis 2003;35:473–478. 130. Kaushansky K. Thrombopoietin: a tool for understanding thrombopoiesis. J Thromb Haemost 2003;1:1587–1592. 131. Giannini E, Botta F, Borro P, et al. Relationship between thrombopoietin serum levels and liver function in patients with chronic liver disease related to hepatitis C virus infection. Am J Gastroenterol 2003;98:2516–2520. 132. Peck-Radosavljevic M. Thrombocytopenia in liver disease. Can J Gastroenterol 2000;14(Suppl D):60D–66D.
498
133. Giannini E, Borro P, Botta F, et al. Serum thrombopoietin levels are linked to liver function in untreated patients with hepatitis C virus-related chronic hepatitis. J Hepatol 2002;37:572–577. 134. Peck-Radosavljevic M, Wichlas M, Zacherl J, et al. Thrombopoietin induces rapid resolution of thrombocytopenia after orthotopic liver transplantation through increased platelet production. Blood 2000;95:795–801. 135. Sanjo A, Satoi J, Ohnishi A, et al. Role of elevated plateletassociated immunoglobulin G and hypersplenism in thrombocytopenia of chronic liver diseases. J Gastroenterol Hepatol 2003;18:638–644. 136. Kajihara M, Kato S, Okazaki Y, et al. A role of autoantibodymediated platelet destruction in thrombocytopenia in patients with cirrhosis. Hepatology 2003;37:1267–1276. 137. Pockros PJ, Duchini A, McMillan R, et al. Immune thrombocytopenic purpura in patients with chronic hepatitis C virus infection. Am J Gastroenterol 2002;97:2040–2045. 138. Sakuraya M, Murakami H, Uchiumi H, et al. Steroid-refractory chronic idiopathic thrombocytopenic purpura associated with hepatitis C virus infection. Eur J Haematol 2002;68:49–53. 139. Peck-Radosavljevic M. Hypersplenism. Eur J Gastroenterol Hepatol 2001;13:317–323. 140. Liangpunsakul S, Ulmer BJ, Chalasani N. Predictors and implications of severe hypersplenism in patients with cirrhosis. Am J Med Sci 2003;326:111–116. 141. Bashour FN, Teran JC, Mullen KD. Prevalence of peripheral blood cytopenias (hypersplenism) in patients with nonalcoholic chronic liver disease. Am J Gastroenterol 2000;95:2936–2939. 142. Plessier A, Denninger MH, Consigny Y, et al. Coagulation disorders in patients with cirrhosis and severe sepsis. Liver Int 2003;23:440–448. 143. Kaul V V, Munoz SJ. Coagulopathy of liver disease. Curr Treat Options Gastroenterol 2000;3:433–438. 144. Kamath PS, Wiesner RH, Malinchoc M, et al. A model to predict survival in patients with end-stage liver disease. Hepatology 2001;33:464–470. 145. O’Grady JG, Alexander GJ, Hayllar KM, et al. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989;97:439–445. 146. Hanslik T, Prinseau J. The use of vitamin K in patients on anticoagulant therapy: a practical guide. Am J Cardiovasc Drugs 2004;4:43–55. 147. Denninger MH, Chait Y, Casadevall N, et al. Cause of portal or hepatic venous thrombosis in adults: the role of multiple concurrent factors. Hepatology 2000;31:587–591. 148. Poujol-Robert A, Rosmorduc O, Serfaty L, et al. Genetic and acquired thrombotic factors in chronic hepatitis C. Am J Gastroenterol 2004;99:527–531. 149. Papatheodoridis GV, Papakonstantinou E, Andrioti E, et al. Thrombotic risk factors and extent of liver fibrosis in chronic viral hepatitis. Gut 2003;52:404–409. 150. Pihusch R, Rank A, Gohring P, et al. Platelet function rather than plasmatic coagulation explains hypercoagulable state in cholestatic liver disease. J Hepatol 2002;37:548–555. 151. Sobhonslidsuk A, Reddy KR. Portal vein thrombosis: a concise review. Am J Gastroenterol 2002;97:535–541. 152. Amitrano L, Guardascione MA, Brancaccio V, et al. Risk factors and clinical presentation of portal vein thrombosis in patients with liver cirrhosis. J Hepatol 2004;40:736–741. 153. Samonakis DN, Koutroubakis IE, Sfiridaki A, et al. Hypercoagulable states in patients with hepatocellular carcinoma. Dig Dis Sci 2004;49:854–858. 154. Kovacs TO, Jensen DM. Recent advances in the endoscopic diagnosis and therapy of upper gastrointestinal, small intestinal, and colonic bleeding. Med Clin North Am 2002;86: 1319–1356. 155. Comar KM, Sanyal AJ. Portal hypertensive bleeding. Gastroenterol Clin North Am 2003;32:1079–1105.
Chapter 25 HEMATOPOIETIC ABNORMALITIES AND HEMOSTASIS
156. Levi M, de Jonge E, van der Poll T. New treatment strategies for disseminated intravascular coagulation based on current understanding of the pathophysiology. Ann Med 2004;36:41–49. 157. Slofstra SH, Spek CA, ten Cate H. Disseminated intravascular coagulation. Hematol J 2003;4:295–302. 158. Ben-Ari Z, Osman E, Hutton RA, et al. Disseminated intravascular coagulation in liver cirrhosis: fact or fiction? Am J Gastroenterol 1999;94:2977–2982. 159. Mammen EF. Disseminated intravascular coagulation (DIC). Clin Lab Sci 2000;13:239–245. 160. French CJ, Bellomo R, Angus P. Cryoprecipitate for the correction of coagulopathy associated with liver disease. Anaesth Intens Care 2003;31:357–361. 161. Pantanowitz L, Kruskall MS, Uhl L. Cryoprecipitate. Patterns of use. Am J Clin Pathol 2003;119:874–881. 162. Hu KQ, Yu AS, Tiyyagura L, et al. Hyperfibrinolytic activity in hospitalized cirrhotic patients in a referral liver unit. Am J Gastroenterol 2001;96:1581–1586. 163. Hersch SL, Kunelis T, Francis RB Jr. The pathogenesis of accelerated fibrinolysis in liver cirrhosis: a critical role for tissue plasminogen activator inhibitor. Blood 1987;69:1315–1319. 164. Bosch J, Abraldes JG. Management of gastrointestinal bleeding in patients with cirrhosis of the liver. Semin Hematol 2004;41(Suppl 1):8–12. 165. Youssef WI, Salazar F, Dasarathy S, et al. Role of fresh frozen plasma infusion in correction of coagulopathy of chronic liver disease: a dual phase study. Am J Gastroenterol 2003;98:1391–1394. 166. O’Shaughnessy DF, Atterbury C, Bolton Maggs P, et al. and the British Committee for Standards in Haematology, Blood Transfusion Task Force. Guidelines for the use of fresh-frozen plasma, cryoprecipitate and cryosupernatant. Br J Haematol 2004;126:11–28. 167. Anantharaju A, Mehta K, Mindikoglu AL, et al. Use of activated recombinant human factor VII (rhFVIIa) for colonic polypectomies in patients with cirrhosis and coagulopathy. Dig Dis Sci 2003;48:1414–1424. 168. Friedman LS. The risk of surgery in patients with liver disease. Hepatology 1999;29:1617–1623. 169. Bravo AA, Sheth SG, Chopra S. Liver biopsy. N Engl J Med 2001;344:495–500. 170. Bass NM, Yao FY. The role of the interventional radiologist. Transjugular procedures. Gastrointest Endosc Clin North Am 2001;11:131–161.
171. Stieltjes N, Ounnoughene N, Sava E, et al. Interest of transjugular liver biopsy in adult patients with haemophilia or other congenital bleeding disorders infected with hepatitis C virus. Br J Haematol 2004;125:769–776. 172. Dieterich DT, Spivak JL. Hematologic disorders associated with hepatitis C virus infection and their management. Clin Infect Dis 2003;37:533–541. 173. Jain AB, Eghtesad B, Venkataramanan R, et al. Ribavirin dose modification based on renal function is necessary to reduce hemolysis in liver transplant patients with hepatitis C virus infection. Liver Transpl 2002;8:1007–1013. 174. Brau N. Epoetin alfa treatment for acute anaemia during interferon plus ribavirin combination therapy for chronic hepatitis C. J Viral Hepatol 2004;11:191–197. 175. Sulkowski MS. Anemia in the treatment of hepatitis C virus infection. Clin Infect Dis 2003;37(Suppl 4):S315–322. 176. Dieterich DT, Wasserman R, Brau N, et al. Once-weekly epoetin alfa improves anemia and facilitates maintenance of ribavirin dosing in hepatitis C virus-infected patients receiving ribavirin plus interferon alfa. Am J Gastroenterol 2003;98:2491–2499. 177. Afdhal NH, Dieterich DT, Pockros PJ, et al. Proactive Study Group. Epoetin alfa maintains ribavirin dose in HCV-infected patients: a prospective, double-blind, randomized controlled study. Gastroenterology 2004;126:1302–1311. 178. Vasilopoulos S, Hally R, Caro J, et al. Erythropoietin response to post-liver transplantation anemia. Liver Transpl 2000;6:349–355. 179. Rajan S, Liebman HA. Treatment of hepatitis C related thrombocytopenia with interferon alpha. Am J Hematol 2001;68:202–209. 180. Kuter DJ, Begley CG. Recombinant human thrombopoietin: basic biology and evaluation of clinical studies. Blood 2002;100:3457–3469. 181. Basser R. The impact of thrombopoietin on clinical practice. Curr Pharm Des 2002;8:369–377. 182. Levi MM, Vink R, de Jonge E. Management of bleeding disorders by prohemostatic therapy. Int J Hematol 2002;76(Suppl 2):139–144. 183. Wong AY, Irwin MG, Hui TW, et al. Desmopressin does not decrease blood loss and transfusion requirements in patients undergoing hepatectomy. Can J Anaesth 2003;50:14–20.
499
Section IV. Toxin Mediated Liver Injury
26
DRUG-INDUCED LIVER INJURY Herbert L. Bonkovsky, Dean P. Jones, Douglas R. LaBrecque, and Steven I. Shedlofsky Abbreviations 6-MP 6-mercaptopurine 5-FU 5-fluorouracil ADR adverse drug reactions AERS adverse events reporting system ALF acute (or subacute) liver failure AP alkaline phosphatase ARE antioxidant response element ANT adenine nucleotide translocase CAM complementary and alternative medications CDS clinical diagnostic scale CYP cytochrome P450 DILI drug-induced liver injury ER endoplasmic reticulum ERCP endoscopic retrograde cholangiopancreatography
FDA GSH HAART HIPAA MMPT MTX NAPQI NIDDK NNRTI NRTI PAPS
food and drug administration glutathione highly active antiretroviral therapy health insurance privacy and portability act mitochondrial membrane permeability transition methotrexate N-acetyl-p-benzoquinoneimine national institutes of health non-nucleoside reverse transcriptase inhibitors nucleoside analogue reverse transcriptase inhibitors 3¢-phospho-adenosine-5¢-phosphosulfate
INTRODUCTION The liver is the principal site for the metabolism of foreign compounds in humans and other higher organisms. It has the function of taking up and disposing of an astonishing array of chemical substances, both from outside the organism (xenobiotics) and compounds made within the organism, including many made within the liver itself, as well as other organs and tissues. In general, the liver and kidneys are chiefly responsible for maintenance of the internal milieu of chemicals within narrow normal ranges. These organs also function to remove potentially toxic compounds from organisms. In general, such compounds of lower molecular weight and higher water solubility are removed chiefly by the kidneys through glomerular filtration and/or tubular secretion. In contrast, larger, more lipophilic substances must be taken up and undergo initial metabolism by the liver, prior to their excretion either in the bile and feces or in the urine. The multistep process for the handling of drugs and chemicals is summarized in Figure 26-1. Most such chemicals are ingested orally and absorbed, chiefly in the proximal small intestine. Some of them undergo initial metabolism within the gastrointestinal tract. Parent compounds and/or metabolites then enter the splanchnic blood, whence they are eventually delivered by the portal circulation to the liver. Depending upon the xenobiotic in question, cells within the liver take up a variable proportion of the compound. This initial removal of compounds from the portal blood, which is the chief blood supply to the liver, is called the ‘first-pass effect’ of the liver. The uptake into liver cells occurs primarily, but not solely, into hepatocytes. During the past several years a number of transporters of cations and anions have been described and have been identified
PI PPAR PXR RNS ROS RUCAM SLE TNF-a VDAC UDPGA ULN WHO
protease inhibitors peroxisome proliferator-activated receptor-g pregnane X receptor reactive nitrogen species reactive oxygen species Roussel–UCLAF causality assessment method systemic lupus erythematosus tumor necrosis factor-a voltage-dependent anion channel uridine diphosphoglucuronic acid upper limit of normal World Health Organization
as important in the uptake of endogenous chemicals and xenobiotics (drugs, foreign chemicals) by liver cells. Once inside hepatocytes, these chemicals undergo further intracellular binding and transport. The intracellular mechanisms responsible for such transport are less well understood. It is likely that highly lipophilic compounds dissolve readily into the membranes of cells and diffuse widely and quickly within and across such membranes. In contrast, more hydrophilic compounds require protein binding and other means for transport. As shown in Figure 26-1, many (but certainly not all) drugs and chemicals require an initial hydroxylation reaction, catalyzed by one of the 57 varieties of cytochrome P450 (CYP). Many of these CYPs are found in hepatocytes, and they carry out hydroxylation reactions in concert with NADPH as a source of electrons, cytochrome P450 reductase, and, for some chemicals, cooperation with another hemecontaining protein called cytochrome b5. These enzymes and reactions occur principally in the smooth endoplasmic reticulum. Following the initial hydroxylation reaction, one of several additional reactions leads to the addition of more water-soluble moieties to the initial hydroxylated product. The enzymes responsible for this so-called phase II metabolism are chiefly glucuronosyl transferases, sulfotransferases, and enzymes that add glutathione or products from glutathione-S (GSH transferases). The key substrates for these conjugation reactions are uridine diphosphoglucuronic acid (UDPGA), 3¢-phospho-adenosine-5¢-phosphosulfate (PAPS), and reduced glutathione (GSH, the tripeptide L-g-glutamyl-L-cysteinylglycine). The third phase of hepatic drug metabolism involves the transport of the parent drug and/or its metabolites out of hepatocytes. This may occur across the plasma membrane, with eventual excretion of water-soluble hydrophilic products in the urine and/or trans-
503
Section IV. Toxin Mediated Liver Injury
Oral ingestion
φ II metabolism Conjugation reactions
Esophagus
φ III transport into bile or blood φ I metabolism CYP’s OH-ation
Drug Class Dose Duration
Host Age Gender BMI Genetic factors Immunol factors Other diseases
Liver Common bile duct
Kidney
Portal vein
Environment Diet Other toxins, chemicals Antioxidants, probiotics
Pancreas Water soluble low MW filtered at glomerulus and/or secreted by tubular cells
Enterohepatic circulation Small intestine
Colon
Ureter
Excretion in urine more hydrophilic smaller compounds
Excretion in feces more lipophilic, higher MW larger compounds
Figure 26-1. Disposition of xenobiotics in humans.
port of drugs and metabolites across the apical membrane domain of hepatocytes into bile canaliculi (secretion into the bile), with eventual excretion in the feces. Drug-induced liver injury (DILI) is the major toxic effect of drugs. It is the principal reason for abandonment of the development of possible new drugs, for the failure of new drugs to achieve approval by the Food and Drug Administration (FDA), and for their withdrawal from the market after initial approval. Adverse drug reactions, which are detected in preclinical studies in cells or experimental animals, usually lead to abandonment of the drug as a candidate for further development, unless there is a unique and highly important desirable effect of such compounds. Those new drugs that survive this initial level of scrutiny then go through phases 1–3 of clinical development and testing. In these phases, only a limited number of highly selected subjects, typically between 2000 and 10 000, receive the medication. As a result, clinically significant but relatively rare adverse drug reactions (ADR) usually are not detected until after drugs have been approved and are in use by a larger number of persons, often with other underlying problems or conditions, which may increase the risk of development of ADR. Thus, continued monitoring of new drugs during phase 4 (post-
504
Figure 26-2. Factors believed to be involved in pathogenesis of DILI.
approval surveillance) is now recognized by the US FDA and by the pharmaceutical industry as of paramount importance. As described above, DILI is a rare event, occurring only in a small percentage of persons who take drugs. The reasons for this are not fully understood. However, based on a combination of experimental results in cell and whole animal models and careful clinical observations, and by analogy to numerous other disorders, an interplay of at least three factors – namely the drug, the host (person ingesting the drug), and the environment of the host – causes and modulates DILI. This is shown in Figure 26-2. It is difficult to establish a diagnosis of DILI. This is because there are numerous other potential causes of liver injury, and because there is no single pathognomonic test to establish that a given drug in a given subject is the cause of liver injury. Furthermore, the clinical and laboratory manifestations of DILI vary markedly, from asymptomatic laboratory abnormalities, which may resolve even though the drug is continued (so-called ‘adaptation’), to severe, lifethreatening acute (or subacute) liver failure ((s)ALF). Because of these factors, a diagnosis of DILI is frequently delayed or may be missed entirely. Similar considerations also apply to non-ethical chemicals, especially so-called ‘herbal remedies’ and complementary and alternative medications (CAM). The scope of this chapter is to provide a review and update of drug-induced liver injury due to ethical drugs, both over-the-counter and prescription. The hepatic effects and toxicities of herbal remedies and CAM are considered elsewhere (see Chapter 27).
EPIDEMIOLOGY: SCOPE OF THE PROBLEM Although the occurrence of drug toxicities, and in particular hepatotoxicities, should be an expected consequence of the widespread use of complex and potent medications, the US public generally expects the FDA to keep toxic drugs off the market. Regulatory agencies in Europe, Australia, and other developed countries serve a similar function to protect their populations. However, there are
Chapter 26 DRUG-INDUCED LIVER INJURY
many drugs with known, well described hepatotoxicities, including acetaminophen, isoniazid, valproate, phenytoin, and propylthiouracil, that remain available because their medical benefit is deemed to outweigh the risk. Nevertheless, between 1990 and 2002, acute liver failure from these five drugs accounted for 15% of all liver transplantations in the US, and 51% of the cases were deemed not to be due to suicide attempts from acetaminophen overdose but to be ADRs to drugs taken with therapeutic intent.1 ADRs were estimated to be the fourth to sixth most common cause of US inhospital mortality in the 1990s.2 For the large number of new therapeutic drugs that have come to market since the FDA began stringent regulatory control in 1962, the required animal toxicity studies and preclinical human trials eliminated most new chemical entities that were likely to cause predictable and dose-related organ toxicity.3 However, some drugs that appeared safe were withdrawn in the postmarketing period because of serious ADRs that became apparent only after many more patients were exposed to the drug (Table 26-1). These withdrawals represented huge economic setbacks for the pharmaceutical companies that had developed the drugs, and they sometimes frustrated patients and their providers who were satisfied with, and depended on the availability of, the withdrawn drugs. Hepatotoxicity was the reason for withdrawal in almost half the cases. Worthy of note is that drug withdrawals due to hepatotoxicity were the most common reasons before 1990, and have occurred less frequently since 1990, perhaps as the result of improved preclinical assessment of risk. If a new medication causes a serious ADR with risk of mortality in as few as 1 in 10 000 exposed persons, society deems this unacceptable, especially if there are alternative drugs available. Phase 1–3 trials generally include fewer than 10 000 carefully selected subjects. Although this is a large number, it is not large enough to reliably identify serious idiosyncratic ADRs that occur less frequently than 1–2 in 10 000. Therefore, the postmarketing surveillance process has
Table 26-1. Drugs Approved by the US FDA but withdrawn in the Postmarketing Setting Drug
Toxicity
Year withdrawn
Iproniazid (Marsalid) Thalidomide (Thalomid) Ibufenac** (in Europe) Ticrynafen (Selacryn) Benoxaprofen (Oraflex) Perihexilene** (in France) Dilevalol** (in Ireland, Portugal) Encainide (Enkaid) Temafloxacin (Omniflox) Flosequinan (Manoplex) Bromfenac (Duract) Mibefradil (Posicor) Terfenadine (Seldane) Troglitazone (Rezulin) Cisapride (Propulsid) Alosetron (Lotronex) Cerivistatin (Baycol) Rofecoxib (Vioxx)
Hepatotoxicity Limb deformities Hepatotoxicity Hepatotoxicity Hepatotoxicity Hepatotoxicity Hepatotoxicity
1956 1961* 1975 1979 1982 1985 1990
Excessive mortality Hemolytic anemia Excessive mortality Hepatotoxicity Multiple drug interactions Cardiac arrhythmias Hepatotoxicity Cardiac arrhythmias Ischemic colitis Rhabdomyolysis Cardiac mortality
1991 1992 1993 1998 1998 1998 2000 2000† 2001†† 2001 2004
Made available again on a severely restricted basis in *1998, †2000, and ††2002. ** Drugs not approved in USA and withdrawn in other countries.
become critically important in identifying ADRs. In January 2002 the FDA organized an Office of Drug Safety, which oversees a large database of ADRs entered into its Adverse Events Reporting System (AERS) at its Medwatch website (www.fda.gov/medwatch/safety/ 3500.pdf) or by fax or post. Similar pharmacovigilance efforts have been mounted in Europe. The World Health Organization (WHO) maintains the Uppsala Monitoring Center (www.who-umc.org) that receives and analyzes ADRs from around the world, including reports from the US FDA.4 With these efforts, it would seem possible to accurately report epidemiologic trends in ADRs and identify offending medications quickly. However, most of the reporting systems rely on voluntary submission of information. And, even though submission of personal health data to Medwatch is exempt from Health Insurance Privacy and Portability Act (HIPAA) privacy restrictions,5 very few healthcare providers take the time to submit reports. It is estimated that only about 1–10% of ADRs are reported.5 Furthermore, the databases consist of large numbers of reports (184 702 were submitted to the FDA’s AERS in 2002), varying in both quality and detail, submitted by pharmaceutical companies (as required by law) or by physicians.4 Because of the potential economic impact on pharmaceutical companies, and because of complex legal liabilities, the raw data from all of these ADRs are not easily available in the public domain. The FDA allows public access to extracts of AERS data through the National Technical Information Service report. The WHO’s Uppsala Monitoring Center has contracted with Vigibase Services (www.umc-products.com), which can provide customized data but at a substantial cost. This service is directed towards regulatory agency professionals and pharmaceutical companies, who study drug safety and utilize data mining techniques. Currently, the incidence and impact of ADRs from specific agents remain largely unknown. Many healthcare providers now use one or more of the available web-based drug information services such as the Physician’s Desk Reference (www.PDR.net) or Epocrates (www.epocrates.com) to review drug toxicity data. However, such sources do not provide specific information regarding the incidence and types of hepatotoxicity caused. We therefore rely upon regulatory agencies such as the FDA to monitor ADRs and either issue ‘black box’ warnings or recommend voluntary withdrawal of the drug. Because of medicolegal liabilities, most pharmaceutical companies voluntarily withdraw their products when evidence of hepato- or other toxicities begins to mount. However, it is likely that, if physicians and other healthcare providers were more conscientious in suspecting and reporting ADRs, the collective data would lead to better patient care and fewer ADRs. With current electronic web-based reporting systems, such reports are now relatively easy and quick to submit.
CAUSALITY ASSESSMENT A continuing challenge in the study of adverse effects of drugs and chemicals on the liver (and other organs) is that of causality assessment, the process whereby the likelihood of the diagnosis DILI is determined. This process of deduction involves analysis of the relevant data and should include an assessment of the temporal relationship, clinical features, laboratory data, histologic data if
505
Section IV. Toxin Mediated Liver Injury
available, and what is currently known about the drug in question. From a statistical standpoint the most scientifically sound approach is to use Bayes’ theorem. Using this technique, an attempt is made to estimate the overall probability of a particular adverse event occurring in a particular individual in a particular situation (posterior probability), given the probability of this event occurring in a group of individuals similarly exposed (prior probability). Individual and situational details considered include the subject’s clinical history, temporal relationships, histologic pattern of injury, resolution with discontinuation of the agent, and whether rechallenge with the agent resulted in the adverse event recurring. These details are used to develop a likelihood ratio; the product of this ratio and the prior probability is a measure of the posterior probability. The major problems that limit the application of Bayes’ theorem in practice are that it is time-consuming, and data needed to compute the likelihood ratio, e.g. background incidence, are often unavailable. Most causality assessment is done using causality schemes. Subjectivity can result in significant intra- and interobserver variations when the same person or different people analyze a case of presumed DILI. To obviate this problem various algorithms and schemes have been developed. Based on the degree of certainty of a causal interaction between drug intake and hepatic injury, different terms are used to describe the strength of the relationship. ‘Definite’ is usually supported by a ‘signature’ clinical pattern, a strong temporal correlation, including a positive rechallenge, and exclusion of all other potential causes. Weaker relationships are termed ‘probable’ or ‘possible’ in descending order of strength; these terms are used when the relevant evidence is judged to be less compelling. The advantages of this approach are that it should allow more consistency in the evaluation of individual cases, and different cases are more likely to be evaluated in the same way. This should in theory lead to better communication between groups evaluating the same problem. This is not always the case, however, and critics argue that enough subjectivity is allowed in some of these processes to significantly alter the outcomes.6 Different causality assessment schemes do not always agree. A recent study used two such scales to look at the same clinical data and demonstrated a low (18%) absolute agreement. A possible explanation was that too much was left to interpretation and judgment in one of the scales.7 The most widely used causality instruments are those of CIOMS (also called RUCAM – Roussel–UCLAF causality assessment method)8,9 and the Clinical Diagnostic Scale (CDS).10,11 These scales add or subtract points for important elements from the history, such as (i) the time of onset of the reaction from the time that a suspect drug was started; (ii) the time course after cessation of the suspect drug; (iii) the presence or absence of risk factors, such as age >55 years, or concomitant pregnancy or the use of ethanol. Additional points are given if non-drug causes have been excluded, including viral infections, biliary obstruction, alcoholism, acute recent hypotension, or hypoxia. These scales do not consider histopathologic findings, but they do consider precedent (listed in the package insert or published reports). A useful website, especially for recent reports of DILI, is the Medline-Pub Med database of the National Library of Medicine (http://www.nchi.nlm.nih.gov/entre2/query.fcgi). One thorny issue is whether the subject in question has DILI or autoimmune hepatitis, because the former may trigger the latter,
506
and because subjects with autoimmune–allergic diatheses are probably more prone to develop DILI of the immune–allergic type. In all cases of suspected cholestatic DILI it is important to exclude biliary tract disease and/or biliary obstruction.
MECHANISMS OF HEPATIC INJURY DUE TO DRUGS AND CHEMICALS APOPTOSIS AND NECROSIS ‘Necrosis’ and ‘apoptosis’ are terms to describe patterns of morphologic changes associated with cell death. Both morphologies include a spectrum of biochemical processes, can occur concomitantly, and often vary in appearance depending on the unique properties of a toxic agent, the time course and dose dependence of exposure, and interactions with other host and environmental factors (Figure 26-2). The most common recognized morphologic form of cell death in liver is necrosis, which is characterized by cellular and organellar swelling and membranal lysis with release of cytoplasmic contents. After such changes the outlines of cells are often indistinct and cells have an amorphous or coarsely granular appearance. Detailed morphologic studies of pathologic as well as normal tissues revealed a second, distinct morphology of cell death, originally termed ‘shrinkage necrosis’12 and now termed ‘apoptosis’.13 These distinct morphologies are now recognized to represent two general cellular phenomena that can occur concomitantly. In apoptosis, typical changes include cell shrinkage, organellar compaction, nuclear condensation, fragmentation of cells into smaller ‘apoptotic bodies’, and the appearance of phagocytosis signals on the cell surface.14 Apoptotic cells are rapidly removed by phagocytosis. Necrosis generally represents a loss of osmotic regulation and cell lysis, whereas apoptosis represents activation of an enzymemediated autolytic cell disposal system. Both are often linked to bioenergetic changes, mitochondrial failure, and oxidative stress. Thus, consideration of the distinctions between necrosis and apoptosis are important not only for pathologic evaluation of hepatotoxicity, but also for investigation of underlying mechanisms of cell injury. Although apoptosis is defined morphologically, it is generally used to refer specifically to a series of cellular changes that result from the activation of a family of highly conserved enzymes, termed caspases. These enzymes are proteases, which cleave specific target amino acid sequences, resulting in characteristic morphology and leading to cell elimination by phagocytosis.15 Activation occurs through plasma membrane-associated death receptor activation of caspase 8,16 through mitochondria-mediated activation of caspase 9,17 and through endoplasmic reticulum-mediated activation of caspase 12.18 These activation mechanisms normally function in homeostatic control of the liver and represent a mechanism to eliminate damaged cells and allow replacement by mitosis. Chemicals that alter the expression and function of the death receptor components, disrupt mitochondrial function, or disrupt the secretory pathway can therefore be expected to activate apoptosis (Figure 26-3). In addition, disruption of the cell cycle and inhibition of proteosomes also activate apoptosis. Thus, many agents previously thought to kill cells by disruption of homeostatic processes
Chapter 26 DRUG-INDUCED LIVER INJURY
Conditioning Events FAS
Bcl-2
Induced expression of death receptors on plasma membranes
Triggering events Ligand binding to death receptors Signal transduction Critical increase in mitochondrial permeability (MMPT) Effector cascade Commitment Leakage of cytochrome C Activation of caspases
Progression
Apoptotic (Acidophilic) body
Shrinkage of cells with recognisable organelles DNA cleaved into oligonucleosomal lengths (”laddering”) Phosphatidyl serine exposed on cell surface
Resolution (clean-up) Apoptotic bodies engulfed by macrophages and digested Figure 26-3. The apoptotic pathway for cell death.
are now believed to do so by activation of apoptosis. Of critical importance is that an increase in apoptosis is often a more sensitive indicator of tissue injury than necrosis. However, because the liver is always undergoing renewal and this process involves apoptosis, definition of the lower limit of toxicity in terms of an increase in apoptosis can be difficult without detailed examination of a large number of cells. An extension of this concept to higher doses and increased activation of apoptosis reveals that a true distinction between necrosis and apoptosis as causative mechanisms in liver toxicity may not be possible. If apoptosis occurs at a rate that exceeds the capacity of phagocytic cells (itself variable) to remove the apoptotic cells, large fields of contiguous cells will swell and lyze, producing characteristics of necrosis. On the other hand, if the toxic insult causes rapid loss of ionic homeostasis, cells may rapidly swell and lyze (i.e. undergo necrosis), even though the apoptotic cascade has been activated. Many toxicants show a dose-dependent switch between apoptosis and necrosis owing to differential effects on mitochondrial function and energy metabolism. Disruption of only a fraction of mitochondria can result in sufficient cytochrome c release to activate the caspase 9/caspase 3 pathway without disrupting cellular energetics. Under these conditions, cells maintain osmotic regula-
tion and undergo apoptosis. However, with greater disruption of mitochondria, cellular energetics are impaired, osmotic regulation is lost, and cells undergo swelling and lysis. Because of this, the mechanisms involved in activation of caspases and/or loss of osmotic regulation are key to understanding chemical-induced liver disease. Mitochondria are a common target of toxicity in liver and play a central role in both apoptosis and necrosis. A recent model integrates major features of energy metabolism, oxidative stress, and apoptosis through central regulatory functions of complex III and aconitase. In oxidative phosphorylation, essentially all electrons flow from coenzyme Q to cytochrome c through complex III to drive the production of ATP. The electron flow through this complex is partitioned between reduction of cytochrome c and 1-electron reduction of O2 to produce the reactive oxygen species superoxide ion. The superoxide ion provides an oxidant source for the regulation of aconitase, a key enzyme in the citric acid cycle, for oxidation of cardiolipin, a key step in the release of cytochrome c and activation of the caspase 9/caspase 3 cascade, and opening of the mitochondrial membrane permeability transition (MMPT) pore, a generalized trigger for both cytochrome c release and loss of ATP production. This complex can therefore serve as a sensor for energetic and redox homeostasis, integrating ATP supply requirements, the efficacy of GSH and other antioxidant systems, calcium homeostasis, and nutritional supply of oxidizable substrates. Consequently, physiologic variations such as food supply and hepatic O2 supply (e.g. hypoxia) can function as important modulators of the biochemical mechanisms and morphology associated with specific toxicant exposures. During the past decade, rapidly developing knowledge about the mechanisms of apoptosis has dramatically changed the perception of how chemicals induce hepatic injury. In the past, injury was considered to be due to failure of critical cell machinery, especially that controlling Ca2+ homeostasis. Now, however, attention has shifted to mechanisms of apoptosis, which is executed by the caspase proteolytic cascade that is activated by specific signaling involving death receptors or disruption of mitochondria, endoplasmic reticulum (ER), nuclei, etc. Central features of chemical-induced liver injury remain the same, but death is now viewed as occurring through targeted cleavage of specific proteins rather than generalized failure. Bioactivation of organic compounds to reactive electrophiles occurs prominently in the liver owing to the presence of high concentrations of enzyme systems designed to aid in the elimination of foreign compounds (see Figure 26-1). Electrophilic agents covalently modify macromolecules, disrupting normal functions, including protein–protein interactions and protein degradation by proteosomes. Oxidants alter the expression of death receptor machinery, enhancing death receptor-mediated apoptosis, and target the MMPT pore, triggering mitochondria-mediated apoptosis. Protection against reactive electrophiles and oxidants occurs through systems that depend upon GSH, and maintenance of GSH is a key mechanism for protection against chemical-induced liver injury. As mass spectrometry and proteomic techniques become widely used in toxicologic research, the possibility of applying systems biology approaches to define toxicity improves. By incorporating a broad spectrum of potential targets into the toxicologic models, this approach is likely to yield a more nearly complete and accurate understanding of the mechanisms of hepatoxicity.
507
Section IV. Toxin Mediated Liver Injury
BIOACTIVATION OF XENOBIOTIC AGENTS Some compounds are not toxic to liver in the parent form but are bioactivated to reactive species. One of the more common mechanisms of bioactivation involves conversion to compounds with electron-seeking properties (i.e. to electrophiles). In most cases, these electrophiles are the result of phase I metabolism by CYP-dependent reactions. Epoxides are an important class of toxic electrophile. Bromobenzene and aflatoxin B1 are metabolized by hepatic mixedfunction oxidases to the epoxide intermediates bromobenzene-3,4oxide19,20 and aflatoxin B1-8,9-oxide,21 respectively. Other electrophilic species include alkyl and aryl halides, carbonium and diazonium ion intermediates, aldehydes, esters, a,b-unsaturated carbon compounds, and compounds containing doubly bound nitrogen (e.g. isothiocyanates, isocyanates, quinazolines).22 Phase II metabolism also may result ultimately in toxic electrophiles, exemplified by toxic GSH S-conjugates, glucuronides, and sulfates;23,24 these metabolites may be toxic to the liver as well as to extrahepatic organs. Reactive electrophiles generated during bioactivation react with specific macromolecules or sites to cause toxicity.25 For instance, earlier research showed that calcium transport systems in the plasma membrane26 and endoplasmic reticulum27,28 contain oxidizable cysteine thiol groups that are critical for function. More recent studies show that molecular chaperones, proteolytic systems, and transcription factors are susceptible to redox modifications.29–31 DNA may be a target of electrophiles, in which case the lesion may cause acute hepatocellular cell death or lead to carcinogenicity. The epoxide of aflatoxin B1 formed during hepatic biotransformation binds guanine residues in DNA at the N-7 position, which may ultimately result in hepatocarcinogenesis.21,32 Covalent modification of proteins may result in the formation of a neoantigen against which an immune response is mounted, giving rise to immunoallergic DILI. Metabolites of halothane, phenytoin, and numerous other drugs may cause liver injury by this type of idiosyncratic mechanism.33,34
ROLE OF GLUTATHIONE (GSH) IN CHEMICAL DETOXIFICATION OF REACTIVE ELECTROPHILES GSH is a major low-molecular weight thiol compound that makes up over 90% of the acid-soluble thiol pool in hepatocytes and accounts for about 30% of the total thiol groups in liver.35 GSH serves many important functions, including detoxification of peroxides and electrophiles (see below), maintenance of protein thiols in a reduced state, serving as a non-toxic storage form of cysteine, synthesis of leukotrienes and prostaglandins, and reduction of ribonucleotides to deoxyribonucleotides.35 The liver is very active in synthesis of GSH, not only for detoxification functions but also to provide a reservoir of cysteine through an interorgan transport mechanism.36 Effective function of this system in sulfur amino acid homeostasis depends on the presence of the cystathionine pathway in the liver,37 and cirrhotic changes in the liver therefore not only affect the sensitivity of the liver to toxicity, but also affect the sensitivity of other organ systems to toxicity due to impaired cysteine and glutathione regulation. Of great relevance to chemical toxicity is that GSH is a key compound involved in the detoxification of electrophiles. The thiol
508
group of GSH is a nucleophilic center that undergoes S-conjugation with electrophiles; in most cases this leads to detoxification. Many electrophiles form GSH S-conjugates non-enzymatically to some extent, which is a function of charge localization of both electrophile and nucleophile.38 Hepatocytes and other cells do not rely on non-enzymatic conjugation of electrophiles; instead, they contain enzymes termed GSH S-transferases that catalyze S-conjugation of GSH to electrophiles. Four major classes of cytosolic GSH Stransferase (i.e. a, m, p, d) and one microsomal enzyme have been characterized in mammalian tissues.39 The cytosolic GSH Stransferases have been studied in the greatest detail and have been shown to be a multigene family of enzymes. Each cytosolic GSH S-transferase is a dimer, a discrete gene codes for each subunit, only subunits of the same class form dimers, and dimers may contain identical subunits (homodimers) or different subunits (heterodimers). The cytosolic enzymes are expressed to various extents in different tissues and are important in detoxification of several groups of xenobiotics, including polycyclic aromatic hydrocarbons, aflatoxins, aromatic amines, and alkylating agents.40 The liver is most active in GSH-dependent detoxification of electrophiles, and human liver is particularly rich in GSH S-transferases of the a class; in other species, such as rat, the m class of GSH S-transferases also is abundant in liver. The generation of large quantities of electrophiles in liver ultimately depletes cellular GSH pools, resulting in enhanced covalent binding to critical macromolecules and cell death. Because GSH and the GSH S-transferases play such an integral role in detoxification of electrophilic species in liver, physiologic or pathophysiologic conditions that either decrease or elevate levels or activities in liver would be expected to affect chemical detoxification in the anticipated direction. Experimentally, this has been demonstrated for a variety of electrophiles. For example, depletion of hepatocellular GSH exacerbates hepatotoxicity associated with electrophiles, including metabolites of acetaminophen41 and bromobenzene.20 Fasting for 1 or 2 days decreases hepatic GSH content by 30–50%42 and enhances the liver injury caused by many electrophilic agents. Diurnal variation in hepatic GSH stores of about 25–30%43 may influence hepatotoxicity as a result of the availability of GSH for detoxification. Cysteine prodrugs (N-acetylcysteine or oxothiazolidine-4-carboxylate) and GSH esters44 can increase hepatic GSH levels and protect against the hepatotoxicity caused by acetaminophen overdose.45 Similar to the CYPs, the GSH S-transferases have relatively broad and overlapping substrate specificities and their activities may be increased following exposure to certain drugs, environmental chemicals, and dietary components. This occurs chiefly through a well characterized transcription enhancer system, consisting of an antioxidant response element (ARE)-binding sequence in the DNA and the Nrf-2/Maf transcription factor system.46,47 Nrf-2 is normally present as an inactive complex with Keap-1, bound to cytoskeletal components in the cytoplasm. Keap-1 has several cysteine thiols that are sensitive to oxidation and alkylation. Modification of these thiols results in the release of Nrf-2, which translocates to the nucleus, interacts with small Maf proteins and binds to the ARE, and activates transcription of a broad range of phase 2 detoxification systems (Figure 26-4) Dietary inducers contained in cruciferous vegetables induce GSH S-transferases and other protective phase 2 enzymes
Chapter 26 DRUG-INDUCED LIVER INJURY HNCOCH3 UDPGA
PAPS Acet-SO4 Non-toxic
Sulfotransferases
Glucuronosyl transferases
Acet-glucuronide Non-toxic
OH CYPs 2E1 1A2 3A4
Bioactivation
Detoxification
NCOCH3
GSH
Acet-Mercapturates Non-toxic
Oxidative stress covalent binding to critical macromolecules mitochondrial toxicity
GST’s O
Hepatocytic necrosis
Reactive, toxic intermediate NAPQI Adaptation/Protection Keap
Nrf2
Nrf2 GSH depletion oxidative stress
Actin
Translocation to nucleus
+ Maf
Nrf2
Maf
ARE’s
Defense gene transcription GCS, GST’s UDPGT’s, HO-1, ....
Figure 26-4. Hepatic metabolism and effects of acetaminophen and the Nfr2, Maf, ARE system for cytoprotection.
without having a significant effect on the CYPs.48–51 These results suggest that increased intake of foods containing these agents may provide a simple and effective means to prevent toxicity as well as cancer caused by toxicants. A prototypic example of bioactivation and covalent binding in hepatotoxicity is provided by the analgesic acetaminophen (Nacetyl-p-aminophenol; paracetamol). In overdose, acetaminophen results in severe hepatocellular necrosis in zone 3 of the hepatic acini (centrilobular necrosis) (Figure 26-5).52 Bioactivation occurs via hepatic CYP to the electrophilic intermediate N-acetyl-pbenzoquinoneimine (NAPQI)53 (Figure 26-4). In humans, CYP 2E1 and 1A2 account for the largest fraction of this conversion.54 At low rates of production, NAPQI is detoxified by S-conjugation with GSH; however, at higher rates of NAPQI production, hepatocellular GSH pools are depleted and extensive covalent binding to cellular macromolecules occurs.55,56 Ultrastructural and functional studies have shown that mitochondria are an early target in the hepatocellular necrosis caused by acetaminophen;57–60 however, acetaminophen metabolites also form adducts to hepatic proteins in the cytosol, microsomes, nuclei, and plasma membranes.34 Arylation of specific proteins has been reported,61–63 but large numbers of adducted proteins have also been detected by mass, and other mechanisms of toxicity have also been suggested.64–70
Figure 26-5. Submassive (zone 3) coagulative necrosis due to an overdose of acetaminophen. Viable hepatocytes surround the portal areas, while the terminal hepatic venules are surrounded by necrotic tissue. Inset shows the necrotic hepatocytes at high magnification.
509
Section IV. Toxin Mediated Liver Injury
OXIDATIVE STRESS AND FREE RADICAL REACTIONS IN HEPATOTOXICITY Oxidative stress is an imbalance between pro-oxidants and antioxidants in which the former predominate, but increasing knowledge of redox signaling mechanisms indicates that toxicants can also disrupt redox signaling and control without necessarily altering major systems that control redox balance. In either case, key mechanisms involve reactive oxygen species (ROS), reactive nitrogen species (RNS), and a range of free radicals generated by bioactivation of xenobiotics. All of the classes of cellular macromolecules can be the target of free radical-induced liver injury. As discussed above for covalent modification, proteins are most often considered the critical targets in acute necrosis, but free radicals also are genotoxic. Free radicals can be generated in the liver in several ways. CYP enzymes generate radicals from xenobiotics by three different mechanisms, a 1-electron oxidation to form a cation radical (RÆ•R+ + e-), a 1-electron reduction to yield an anion radical (R + e- ‡•R-), or homolytic bond scission to yield a neutral radical (R-R ‡•R + •R).71 Hepatotoxicants of occupational/ environmental (e.g. CCl4) as well as of clinical importance (e.g. halothane) are bioactivated in liver to free radical species. CCl4 toxicity provides a useful model and is representative of a large number of halogenated hydrocarbons that can be similarly activated. CCl4 is a prototypical hepatotoxin that causes centrilobular necrosis and associated fatty liver. Caspase 3 is activated and released into the plasma with a time course suggesting initial activation of apoptosis followed by secondary necrosis.72 A primary event in the pathogenesis is the reductive dehalogenation of CCl4 to the trichloromethyl free radical (•CCl3) by hepatic mixed function oxidases (Figure 26-6). •CCl3 can initiate lipid peroxidation, and in the
CCl4 Cytochrome P450
CCl3 Low oxygen
Covalent binding to cellular macromolecules Abstraction of hydrogen from PUFA to form CHCl3 and a lipid radical Reaction with GS-
High oxygen
Formation of CCl3OO• with subsequent lipid peroxidation and termination reactions Decomposition of CCl3OOto CCl2O, which can react with GSH or bind covalently to cellular macromolecules
Termination reactions with itself or lipid radicals Figure 26-6. Hepatic metabolism and oxygen-dependent effects of carbon tetrachloride (CCl4). •CC13, trichloromethyl radical; GS–, glutathione thiolate anion; GSH, glutathione; PUFA, polyunsaturated fatty acid.
510
presence of oxygen it forms the more reactive trichloromethylperoxy free radical (CCl3OO•), which also decomposes to phosgene (CCl2O). The lipid peroxidation in liver associated with CCl4 has been viewed as a critical event because it occurs early and is associated with reductions of enzyme activities73and inactivation of the Ca2+-sequestering capacity of endoplasmic reticulum.74,75 Elevation of cellular Ca2+ establishes conditions for activation of the mitochondrial permeability transition, with associated cytochrome c release and caspase activation (see Figure 26-3). In addition to free radicals of the parent compound, reactive oxygen and reactive nitrogen species are often involved in hepatotoxicity. Reactive oxygen species such as the hydroxyl radical (•OH) are generated during redox cycling of several xenobiotics, following activation of the respiratory burst in host phagocytic cells, and during exposure to ionizing radiation. Reactive nitrogen species can also be involved in these toxic processes due to formation of the free radical nitric oxide (NO•), an important signaling agent that reacts with superoxide anion (•O2–) to generate peroxynitrite.76–78 Redox cycling refers to a pathway whereby a compound undergoes a series of cyclic 1-electron reductions and oxidations with concomitant generation of toxic oxygen species. A variety of flavoproteins catalyze 1-electron reductions. In the presence of oxygen, the reduced product can spontaneously oxidize back to the parent compounds, and this oxidation is coupled to the reduction of molecular oxygen to the superoxide radical ion •O2– (Figure 26-7). Many redox cycling agents cause toxicity to hepatocytes in vitro, but also usually cause toxicity to other organ systems when administered in vivo. Examples are the lung injury associated with paraquat79 and the cardiotoxicity caused by adriamycin.59 This probably reflects a relative resistance of liver parenchyma, owing to the large capacity of the liver to detoxify reactive oxygen species. Menadione (2-methyl-l,4-naphthoquinone; vitamin K3) is a therapeutic quinone compound that can cause liver injury due to redox cycling. Menadione undergoes 1-electron reduction to the semiquinone free radical, which is catalyzed by many flavoenzymes, including NADPH:cytochrome P450 reductase.71 In hepatocytes and other cells, it can also undergo a 2-electron reduction to the hydroquinone in a reaction catalyzed by NADPH:quinone reductase, a cytosolic enzyme also known as DT-diaphorase (see Figure 26-7). This 2-electron reduction provides a detoxication reaction because the hydroquinone can be conjugated by sulfotransferases or uridine diphosphoglucuronic acid (UDPGA)-glucuronosyltransferases. DT-diaphorase knockout mice are more sensitive to the hepatotoxicity of menadione.80 The oxidative stress caused by redox cycling of menadione leads to irreversible cell injury via a complex interplay between oxidation of soluble thiols (e.g. GSH) and that of protein thiols, causing a sustained rise in Ca2+ that is critical to the activation of mitochondriamediated apoptosis. Oxidation of critical protein thiols decreases microsomal Ca2+ sequestering capacity27,28 and plasma membrane extrusion of Ca2+ from cells.26 Oxidation of soluble thiols precedes this and is a contributing factor in inhibition of the microsomal Ca2+ pump because GSH keeps the protein thiols in a reduced and functional form. In the presence of elevated Ca2+ mitochondria load Ca2+, and this loading sets conditions appropriate for activation of the mitochondrial permeability transition.81
Chapter 26 DRUG-INDUCED LIVER INJURY
O CH3
O Menadione O2
NAD(P)+
O2•
NAD(P)H O CH3
NAD(P)H Quinone reductase (DT-diaphorase)
2NAD(P)+ OH Semiquinone O2
NAD(P)+ 2 electron reduction
•
O2
NAD(P)H O CH3
Lipid peroxidation can occur as a consequence of activation of any of the above free radical processes. Lipid peroxidation decreases membrane fluidity and is associated with inactivation of membranebound receptors and enzymes, increased permeability of membranes, and the generation of toxic degradation products of lipid peroxidation.73,85 GSH plays a central role in protection against lipid peroxidation through enzyme-catalyzed reactions and through nonenzymatic reduction of other antioxidants (vitamins C and E). GSH is required for degradation of lipid hydroperoxides and other hydroperoxides in reactions catalyzed by the selenium-dependent GSH peroxidase. For this reduction, the fatty acid hydroperoxides must be released first from the bulk lipid by the action of phospholipase A2.86,87 However, a separate selenium-dependent phospholipid hydroperoxide GSH peroxidase that directly detoxifies phospholipid hydroperoxides without a requirement for phospholipase A2 has been characterized.88,89 A selenium-independent GSH peroxidase, which has been ascribed to GSH S- transferases of the a class,90 also detoxifies lipid hydroperoxides; like the seleniumdependent GSH peroxidase, it requires release of the fatty acid peroxide from the membrane. The selenium-independent form also is active in detoxication of cumene hydroperoxide and nucleic acid hydroperoxides.91 GSH also detoxifies toxic degradation products of lipid hydroperoxides, most notably the 4-hydroxyalkenals (e.g. 4-hydroxynonenal) via S-conjugation. 4-Hydroxynonenal is an extremely toxic product of lipid peroxidation, with submicromolar concentrations causing genotoxic lesions to cultured rat hepatocytes.92 Its conjugation with GSH is catalyzed by a specific form of glutathione S-transferase,93,94 suggesting that this reaction may be fundamentally important in the prevention of free radical-mediated liver injury.
OH Hydroquinone Figure 26-7. Redox cycling of menadione, an example of an hepatotoxic drug that produces superoxide (O2. ). NAD+, nicotinamide adenine dinucleotide; NADH, nicotinamide adenine dinucleotide (reduced form). NADP+, nicotinamide adenine dinucleotide phosphate; NADPH, nicotinamide adenine dinucleotide phosphate (reduced form).
The MMPT occurs in response to the opening of a high-conductance channel in the mitochondrial inner membrane.81 Ordinarily, the inner membrane is highly impermeable to solutes. However, in the presence of matrix Ca2+, certain agents trigger the opening of the high-conductance MMPT pore. The prevailing interpretation is that the pore is a protein complex containing adenine nucleotide translocase (ANT, inner membrane), voltage-dependent anion channel (VDAC, outer membrane), cyclophilin D (associated with ANT) and peripheral benzodiazepine receptor (associated with VDAC). Sensitivity to oxidants and thiol reagents, especially arsenicals, indicates that the pore contains thiols, probably vicinal thiols, which control opening. Thus, in the presence of elevated Ca2+, oxidants trigger opening of the MMPT pore, with resulting swelling and release of cytochrome c82 and other proapoptotic components.66,83,84 Cytochrome c binds to APAF-1, an assembly protein that allows the recruitment and activation successively of procaspase 9 and procaspase 3 (see Figure 26-3).17
CLINICOPATHOLOGIC PATTERNS OF DRUG-INDUCED LIVER INJURY Drugs and foreign chemicals produce manifold and varied damage and changes to the liver. Although not absolute, drugs typically produce patterns of injury that are characteristic for each individual drug. It is been found useful to categorize these patterns as hepatocellular (or hepatitic), cholestatic, mixed, or steatotic. Some key features of these four patterns are summarized in Table 26-2.
HEPATOCELLULAR (HEPATITIC) The majority of drugs that cause DILI produce principally a pattern of hepatocellular-type injury.1,47,95,96 Most of these instances are asymptomatic and mild. When they are unusually severe, patients typically have symptoms similar to what would be expected in those with acute viral hepatitis. They lose their appetites, especially for smoking; they become nauseated. When very severe, they develop vomiting, which can be intractable. They usually complain of vague upper abdominal pain, generally primarily in the epigastrium and right upper quadrant. The laboratory features of this type of injury typically include a normal complete blood count, although a mild increase in white count may occur, and a minority of patients with immunoallergic-type reactions will show peripheral eosinophilia. Because the
511
512
Normal liver or diffuse, homogeneous hepatomegaly, perhaps with changes compatible with diffuse fatty change or phospholipidosis No biliary dilation Pancreatic swelling may be present Acute viral hepatitis Ischemic hepatitis Acute congestive hepatitis Budd–Chiari syndrome Hepatic decompensation due to Wilson’ s disease Autoimmune hepatitis
Acute zone 3 necrosis Findings indistinguishable from acute viral hepatitis Increased eosinophils and/or acute granulomas may be present; Fat (usually mainly in zone 3) may be present; steatohepatitis may be present
Typical hepatobiliary– pancreatic imaging findings
Typical findings on liver biopsy
Cholestasis without acute cholangitis or pericholangitis Zone 3 hepatocytic swelling No bile lakes or other features typical of extrahepatic obstruction
Biliary obstruction due to gallstones, tumors, strictures, pancreatic diseases Primary biliary cirrhosis Primary sclerosing cholangitis “Overlap” syndromes of autoimmune cholangitis/ hepatitis
Serum ALT, AST < 5 ¥ ULN Serum AP > 2 ¥ ULN Serum TBR, DBR > 2 ¥ ULN (often > 5 ¥ ULN) Resembles biliary obstruction or cholestatic phase of acute hepatitis A Normal liver or diffuse, homogeneous hepatomegaly No biliary dilation No pancreatic abnormalities No changes to suggest chronic liver disease or cholecystitis
Serum ALT, AST > 5 ¥ ULN Serum AP < 2 ¥ ULN Serum TBR, DBR variable May resemble acute ischemic or viral hepatitis
Typical laboratory findings
Major considerations for differential diagnosis
Jaundice; itching; nausea, anorexia, when very severe
Cholestatic
Nausea, anorexia (vomiting) Loss of taste for food, smoking Upper abdominal pain
Hepatocellular (hepatitic)
Typical clinical presentation
Features
Table 26-2. Clinicopathologic Patterns of Drug-Induced Liver Injury
Biliary obstruction due to gallstones, tumors, strictures, pancreatic diseases Primary biliary cirrhosis Primary sclerosing cholangitis “Overlap” syndromes of autoimmune cholangitis/ hepatitis Cholestasis without acute cholangitis or pericholangitis Zone 3 hepatocytic swelling No bile lakes or other features typical of extrahepatic obstruction Acute zone 3 necrosis Findings indistinguishable from acute viral hepatitis Increased eosinophils and/or acute granulomas may be present Fat (usually mainly in zone 3) may be present; steatohepatitis may be present
Normal liver or diffuse, homogeneous hepatomegaly No biliary dilation No pancreatic abnormalities No changes to suggest chronic liver disease or cholecystitis
Serum ALT, AST > 3 ¥ ULN Serum AP > 2 ¥ ULN Serum TBR, DBR > 2 ¥ ULN Features of both hepatocellular and cholestatic patterns
Jaundice; itching; nausea, anorexia, when very severe
Mixed
Patterns of injury
Hepatocytic “swelling” with foamy appearance of cytoplasm and centrally located nuclei Apoptotic bodies— hepatocyte dropout with minimal inflammation No or minimal fibrosis
Reye’ s syndrome Acute fatty liver of pregnancy Inborn or other acquired defects in mitochondrial function—fatty acid oxidation and/or ATP production
Normal liver No biliary dilatation Normal pancreas Normal spleen No PHT No changes to suggest chronic liver disease or cholecystitis
Nausea, anorexia Vomiting Confusion Somnolence (hepatic encephalopathy) Serum ALT, AST 5–25 ¥ ULN Serum AP 1–3 ¥ ULN Serum TBR, DBR variable, often normal Resembles acute viral hepatitis
Microvesicular
Variable amounts of neutral fat accumulation in hepatocytes, usually mainly in zones 3 and 2 Hepatocyte nuclei pushed to periphery of cells by macrovesicular steatosis Apoptotic bodies, hepatocyte dropout Variable inflammation Variable fibrosis, usually pericellular Lipogranulomas common in zone 3
Alcoholic liver disease Liver disease associated with metabolic syndrome: NAFL, NASH Inborn or other acquired defects in normal hepatic lipid metabolism
Diffuse, generalized hepatomegaly Increased echogenicity (US) Decreased attenuation (CT) No biliary dilatation Normal or “fatty” pancreas No changes to suggest chronic liver disease or cholecystitis
Serum ALT, AST 1–5 ¥ ULN Serum AP 1–3 ¥ ULN Serum TBR, DBR variable, usually normal Resembles alcoholic hepatitis
Asymptomatic Upper abdominal discomfort, heaviness Nausea, anorexia
Mixed micro/macrovesicular
Steatosis
Section IV. Toxin Mediated Liver Injury
Stop offending drug N-acetylcysteine for acetaminophen, Prednisolone, 20–30 mg/d, azathioprine, 1–2 mg/kg/d, for severe immunoallergic disease
Follow “Hy’ s rule”: ~10% develop jaundice ~10% of those which develop jaundice die If FHF develops, case-fatality rate for non-acetaminophen cases is ~75%; For acetaminophen cases is ~25% A minority (perhaps ~15–30%) with smoldering presentations may develop bridging fibrosis or cirrhosis Triggering of ongoing AI hepatitis by drugs is very rare (<0.5%)
Usual therapy
Course and long-term prognosis
Protracted cholestatic syndrome lasting weeks to months Severity often increases even after offending drug has been stopped. Great majority of patients recover, apparently nearly completely (although few follow-up biopsies are done) Small minority (perhaps ~ 1%) develop vanishing bile duct syndrome or course that resembles sclerosing cholangitis or biliary cirrhosis
Protracted course with symptoms, signs, and labs worsening or remaining for 30–60 days Gradual improvement, thereafter, but may require > 180 days to resolve Stop offending drug Ursodeoxycholic acid, 20– 30 mg/kg/d Cholestyramine, phenobarbital, (rifampicin) for severe itching Stop offending drug Prednisolone, 20–30 mg/d, Azathioprine, 1–2 mg/kg/d, for severe immunoallergic disease Ursodeoxycholic acid, 20– 30 mg/kg/d Cholestyramine, phenobarbital (rifampicin) for severe itching A mixture of the prognoses listed to the left.
Variable course Usually more protracted than hepatocellular, but less than cholestatic.
Full recovery No progression to chronic liver disease
Stop offending drug Supportive care, nutrition Urgent liver transplant for severe disease with grade 3–4 encephalopathy
Rapid improvement in symptoms, signs, and labs with >50% decreases within 8–30 days
Variable, depending upon underlying conditions and duration and nature of prior injury
Stop offending drug Supportive care, nutrition Consider prednisone (20–40 mg/d), pentoxfylline (400 mg tid) for severe disease (DF > 32 or renal insufficiency)
Variable, depending upon drug accumulation, half life Often, underlying alcohol or metabolic syndrome effects persist
AI, autoimmune; ALT, alanine aminotransferase; AP, alkaline phosphatase; AST, aspartate aminotransferase; DBR, direct-reacting bilirubin; FHF, fulminant hepatic failure; TBR, total bilirubin; ULN< upper limit of normal.
Rapid improvement in symptoms, signs, and lab tests, with >50% decreases within 8–30 days
Typical course after inciting agent stopped
Chapter 26
DRUG-INDUCED LIVER INJURY
513
Section IV. Toxin Mediated Liver Injury
underlying pathogenesis involves primarily apoptosis and/or necrosis of hepatocytes, serum ALT and AST levels are markedly elevated. In the case of acute poisoning of intrinsic hepatotoxins, such as acetaminophen, carbon tetrachloride, or other halogenated hydrocarbons, the elevations of serum aminotransferases may be extreme (more than 100 times the upper limit of normal (ULN). For the larger number of drugs that produce idiosyncratic, unpredictable, non-dose dependent DILI, the degree of elevation of serum ALT and AST generally is less marked (10–25 times the ULN). The serum alkaline phosphatase (AP) is generally normal or mildly elevated (less than twice the ULN). Serum total and direct bilirubin are variable. They may remain normal, although with the more severe forms of injury they are invariably increased. The degree of increase in serum bilirubin may be extreme, and it is one of the negative prognostic factors for hepatocellular-type injury. Hepatobiliary–pancreatic imaging in such injury shows a normal liver or diffuse homogeneous hepatomegaly. For some drugs, changes compatible with diffuse fatty change (Table 26-3) or phospholipidosis (Table 26-4) may be present. Of particular importance, especially when patients are jaundiced, is the lack of evidence of dilatation of the biliary tree or cholecystitis. Of course, pre-existing gallstones may be present, making it somewhat more difficult to arrive at a correct diagnosis. Some drugs, such as acetaminophen, can also cause acute pancreatic, myocardial, or renal injury. If pancreatitis occurs, the pancreas on imaging studies generally shows diffuse enlargement or edema. Typically, changes suggestive of chronic underlying liver disease are absent, although there is nothing about pre-existing liver disease that prevents patients from developing DILI. Therefore, such changes may be present. The major considerations for the differential diagnosis of acute hepatocellular injury due to drugs include acute ischemic liver injury, acute viral hepatitis, which may be due to any of the agents that are capable of causing this syndrome (see Chapters 30–35), acute congestive hepatitis, including Budd–Chiari syndrome, autoimmune hepatitis, or hepatic decompensation due to Wilson’s disease.
Table 26-3. Some Drugs and Chemicals that Produce Hepatic Steatosis Microvesicular
Macrovesicular or mixed micro-/macrovesicular
Aflatoxin b1 Amiodarone L-Asparaginase Aspirin Chloroform Cocaine Coumadin Deferoxamine Didanosine Ethanol
FIAU Halothane Methotrexate Minocycline Mitomycin Tamoxifen Tetra-, trichloroethylene Tetracyclines Valproic acid
CHOLESTATIC PATTERN OF INJURY
Table 26-4. Some Drugs and Chemicals that may Produce Mallory Bodies Amiodarone Diethylstilbestrol 4,4¢-Diethylaminoethoxyhexestrol Ethanol
514
The typical findings on liver biopsy, when performed during acute hepatocellular injury due to drugs, are variable and highly dependent upon the offending agent. The most common hepatotoxic drug, namely acetaminophen, causes acute necrosis first and foremost in zone 3 of the hepatic acinus. When very severe, necrosis extends into and through zone 2 as well (see Figure 26-5). Other common findings include variable inflammation of the portal tracts, often with a considerable number of polymorphonuclear or eosinophilic forms. Acute granulomas may also occur. Indeed, drug-induced liver injury is one of the common causes of granulomas in the liver.97–99 Some drugs and chemicals are well known to produce fatty change in the liver. Usually this is primarily in zone 3, although it is certainly not restricted to this zone. All of the features of steatohepatitis may sometimes be present. In most instances of hepatocellular injury, particularly when it has been sudden and acute in onset, there is a rapid improvement in symptoms, signs, and laboratory features when the offending agent is discontinued. This does not always occur, however, and in rare individuals drugs appear to be capable of triggering the development of self-perpetuating autoimmune hepatitis.96,100 The short- and long-term prognosis of hepatocellular type injury follows ‘Hy’s rule’. This was popularized by Hyman Zimmerman, a clinical hepatologist with special interest in drug-induced liver injury.95 Hy’s rule states that about 10% of patients with druginduced liver injury of the hepatocellular type develop jaundice, and that, among those who do, about 10% will die of drug-induced liver injury. The case fatality rates for persons who develop fulminant hepatic failure due to drugs is very high (around 75%) for drugs other than acetaminophen. In contrast, the case fatality rate for acetaminophen-induced fulminant hepatic failure is much lower, with only about 25% of patients dying and/or requiring liver transplant. For the most part there is no specific therapy for drug-induced liver injury, beyond identifying the offending agent and stopping its use. It is clear that acute acetaminophen overdose should be treated immediately with N-acetylcysteine. For adults with acetaminophen ingestion less than 24 hours before presentation, a loading dose of 140 mg/kg/body weight should be given, followed by 70 mg/kg every 4 hours for 17 doses, starting 4 hours after the loading dose. It has been suggested that N-acetylcysteine may be of benefit in other forms of fulminant hepatic failure, and indeed there seems little to be lost by administering it in other forms. Particularly when hepatocellular type injury is severe, and/or when it is accompanied by evidence of immunoallergic features, a corticosteroid, such as prednisolone (20–30 mg/day), and azathioprine (1–2 mg/kg body weight per day) often are given as well.
Glucocorticoids Griseofulvin Nifedipine Tamoxifen
The typical presentation of cholestatic hepatitis due to drugs is jaundice and itching. Nausea, anorexia, or vomiting typically occur only when the reaction is very severe. The typical laboratory features are those of any cholestatic syndrome, with elevations primarily in serum AP, which is more than twice the ULN, and serum total and direct bilirubin, which also are at least twice the ULN. In the pure cholestatic case, serum aminotransferases are normal or only mildly elevated, and certainly less than three times the ULN.
Chapter 26 DRUG-INDUCED LIVER INJURY
The typical hepatobiliary–pancreatic imaging findings in cholestatic DILI are chiefly important by showing no evidence of biliary dilatation and no pancreatic abnormalities. The liver is usually normal or nearly normal, and there is nothing to suggest chronic liver disease or cholecystitis (see Table 26-2). The major differential diagnosis for cholestatic DILI includes biliary obstruction due to gallstones, tumors, strictures, or pancreatic diseases, and autoimmune disorders that affect chiefly the bile ducts, such as primary biliary cirrhosis or primary sclerosing cholangitis. There are also ‘overlap’ syndromes of autoimmune cholangitis and autoimmune hepatitis. These are considered in greater detail in Section VI (Immune diseases). Typical findings on liver biopsy in cholestatic DILI are the presence of bile in hepatocytes, bile plugs in canaliculi, and hepatocyte swelling in zone 3 (Figure 26-8). Bile lakes or other features of extrahepatic obstruction are absent, and as a rule there are no findings of acute cholangitis or pericholangitis, such as one would expect to see in bacterial ascending cholangitis. The typical course of cholestatic hepatitis is quite different from that of hepatocellular DILI in being more protracted. In fact, it is not uncommon for signs and laboratory worsening to continue after the drug has been stopped, sometimes for as long as 30–60 days. There is gradual improvement thereafter, unless the offending agent or another like it is readministered. However, this can require as long as 180 days. There are rare instances in which the disease does not resolve but rather goes on to produce the adult vanishing bile duct syndrome, sometimes with progression to secondary biliary cirrhosis.99 The usual therapy of cholestatic DILI is to stop the offending drug and administer ursodeoxycholic acid. In light of growing evidence that higher doses of this agent are more effective in chronic cholestatic disorders such as primarily biliary cirrhosis or primary sclerosing cholangitis, it is our recommendation that the drug be given at a dose of 20–30 mg/kg/day in two divided doses. If the itching is severe the usual treatment is cholestyramine, but this must be given at times other than when ursodeoxycholic acid or other drugs are administered, because it will bind the drugs and prevent their absorption. We generally recommend that the cholestyramine be administered in the morning, when there is maximal turnover of the biliary pool. Phenobarbital and/or rifampicin can be helpful for severe itching, although both of these drugs, especially rifampicin may to cause hepatotoxicity on their own.
‘MIXED’ PATTERN OF INJURY This pattern, as the name implies, involves features both of hepatocellular and cholestatic injury (see Table 26-2). The typical clinical presentation is nausea, anorexia, and vomiting when severe. There is also jaundice and itching. The typical laboratory findings are for serum aminotransferase levels to be greater than three times the ULN and for serum AP and total and direct bilirubin to be more than twice the ULN. The biopsy features are also a mixture of features described above for the two types of injury. The considerations for differential diagnosis must include ischemic hepatitis, acute congestive hepatitis, acute viral hepatitis, autoimmune hepatitis, or overlap syndromes of autoimmune cholangitis and hepatitis, hepatic decompensation due to Wilson’s
A
B
Figure 26-8. (A) Cholestatic injury. This biopsy from a patient who became jaundiced while taking the NSAID nabumetone has relatively ‘bland’ cholestasis with numerous canalicular bile plugs (arrows), but relatively little hepatocellular injury. The peak serum bilirubin in the patient was 110 mg/dL. (B) Cholestatic injury. This biopsy from a patient who became jaundiced after a course of amoxicillin shows a combined hepatocellular and cholestatic injury with canalicular bile plugs (arrows) as well as hepatocyte injury, apoptosis and dropout with Kupffer cell hypertrophy and lymphocytic inflammation, producing disarray of the liver cell plates.
disease, primary biliary cirrhosis, primarily sclerosing cholangitis, and biliary obstruction due to gallstones, tumors, strictures, or primary pancreatic diseases. The typical treatments are the same as already described for hepatocellular and cholestatic injuries. The typical course is somewhat longer than for hepatocellular injury, but somewhat shorter than for typical cases of pure cholestatic DILI.
STEATOSIS (FATTY LIVER) As shown in Table 26-2, there are two major types of disease that produce primarily fatty change in the liver, namely, pure small droplet fat (microvesicular) and fewer large droplet fat (macrovesicular), although the latter is usually associated with at least a mild degree of microvesicular steatosis as well.
515
Section IV. Toxin Mediated Liver Injury
Microvesicular steatosis is due principally to mitochondrial toxicity, leading both to a deficiency in mitochondrial b-oxidation of free fatty acids and to critical compromise of mitochondrial ATP production. Patients with these defects commonly present with nausea, anorexia, vomiting, confusion, or coma, the latter due to hepatic encephalopathy with prominent and severe hyperammonemia. They often have significant lactic acidosis, owing to the critical defect in mitochondrial respiration and oxidative phosphorylation. The typical laboratory features are moderate to marked increases in serum aminotransferases; serum AP is normal or only slightly increased, and serum bilirubin levels are variable, depending on the severity of the injury. Typical hepatobiliary pancreatic imaging studies in patients with microvesicular steatosis show a normal liver, no biliary dilatation, normal pancreas and spleen, and nothing to suggest portal hypertension or chronic liver disease. The major differential diagnosis is Reye’s syndrome or inborn or other acquired defects in mitochondrial function, particularly fatty acid metabolism or ATP production. The findings on liver biopsy are remarkably mild (Figure 26-9). In order to see the lipid clearly, it may be necessary to perform oil red O staining on frozen sections. The reason is that there is diffuse lipid accumulation in very small droplets, often smaller than the limit of resolution by light microscopy. There is no displacement of hepatocytic nuclei, such that the lipid may not be apparent in formalin-fixed tissue stained in the routine way. There is minimal inflammation, although apoptotic bodies and evidence of hepatocytic dropout may be present, and there is usually no fibrosis. The usual course is one of rapid improvement if the inciting agent is stopped. However, some patients have such severe defects that they may be unable to recover unless they receive urgent liver transplantation. Certainly, all such patients who might be transplant candidates and who develop higher grades of hepatic encephalopathy should rapidly be transferred to a transplant center. If patients can be nursed successfully through the acute phase of disease, complete
recovery with no progression to chronic liver disease will ensue. Examples of drugs that produce microvesicular steatosis are summarized in Table 26-3.
MACROVESICULAR OR MIXED MICRO- AND MACROVESICULAR STEATOSIS The accumulation of fat is probably the most common liver abnormality. Potential causes of fatty liver are manifold and discussed in greater detail in Chapter 55. Drugs and chemicals are among the important causes of fatty liver. Indeed, if one considers ethanol as a drug, they are probably the most common causes. Most people with fatty liver due to alcohol or other conditions that produce macrovesicular steatosis are asymptomatic. When the fatty deposition is severe hepatomegaly ensues, and patients may have upper abdominal discomfort and a sense of heaviness. It is rare for more severe symptoms, such as nausea, anorexia, vomiting, jaundice, etc., to occur. Laboratory studies may be entirely normal or may show mild increases in serum aminotransferases. Serum AP may be slightly increased; g-glutamyl transpeptidase is usually elevated more. Typical findings on hepatobiliary–pancreatic imaging are diffuse, generalized hepatomegaly. Ultrasound shows evidence of increased echogenicity, whereas CT scanning shows a decrease in hepatic attenuation. There is generally no biliary dilation and the pancreas is normal, or may show increased echogenicity indicative of a fatty deposition in the pancreas. In addition to heavy alcohol use, macrovesicular steatosis is commonly caused by liver disease associated with the metabolic syndrome (non-alcoholic fatty liver and non-alcoholic steatohepatitis, see Chapter 55). The typical findings on liver biopsy in drug-induced macrovesicular steatosis are indistinguishable from those caused by alcohol or by non-alcoholic fatty liver. It is common for patients to have these changes owing to an element of alcohol and nonalcoholic fatty liver plus one or more drugs. Mallory bodies may develop as a result of alcoholic or non-alcoholic steatohepatitis, and have been associated with several drugs (see Table 26-4). The usual therapy is to stop the offending drug. However, if the fatty change is mild and asymptomatic, and if the drug is essential for other reasons, such as methotrexate for the management of rheumatoid arthritis or psoriasis, the decision may be made to continue the drug with careful monitoring. In addition to the histopathological features already described, drugs can cause the accumulation of phospholipids in hepatocytes and other cells (Table 26-5), vascular lesions in the liver, including peliosis hepatis (Table 26-6), sinusoidal obstruction or veno-occlusive disease, and arterial vascular compromise, which is manifest as a syndrome that resembles sclerosing cholangitis.
PREDICTABLE VS UNPREDICTABLE DILI
Figure 26-9. Microvesicular steatosis in a child taking valproic acid. Most of the hepatocytes have small vacuoles of fat, and there has been liver cell dropout with Kupffer cell hypertrophy and a mild lymphocytic infiltrate.
516
Another useful way to categorize DILI is as predictable or ‘intrinsic’ injury versus unpredictable or ‘idiosyncratic’. By far the most important example of the former is acetaminophen, which, by mechanisms already described, will produce liver injury in virtually everyone who takes a sufficient dose. Examples of other drugs or toxins that act similarly are listed in Table 26-7. Most drugs, however, cause DILI unpredictably and in only a small percentage of subjects. Such reactions are called idiosyncratic
Chapter 26 DRUG-INDUCED LIVER INJURY
reactions. These are further subdivided according to whether they are accompanied by immunoallergic manifestations or not. Such manifestations include fever, peripheral eosinophilia, skin rash, arthralgia, arthritis etc. As shown in Table 26-7, many drugs are recognized to be capable of causing idiosyncratic DILI either with or without an immunoallergic phenotype, stressing the importance of genetic host factors in modulating the response to injury (see Figure 26-2). The mechanisms that probably give rise to immunoallergic injury are summarized in Figure 26-10. According to this view, drugs may give rise to antigens, by binding to host proteins (perhaps altering them), which are recognized as foreign and against which the host’s immune system mounts a T- or B-lymphocyte response. Because such neoantigens are displayed on hepatocytes, where most drug metabolism occurs, the net effect may resemble autoimmune hepatitis. Indeed, ingestion of drugs appears to trigger autoimmune hepatitis in rare individuals.96,100
Table 26-5. Some Drugs that Produce Phospholipidosis [All amphiphilic drugs] Amantadine Amikacin Amiodarone Amitryptiline Chloramphenicol Chlorcyclizine Chloripramine Chloroquine Chlorpheniramine Chlorpromazine Desipramine
Gentamicin Imipramine Iprindole Ketoconazole Mepacrine Promethazine Propranolol Sulfamethoxazole-trimethoprin Thioridazine Trimipramine Trippelennamine
Table 26-6. Some Drugs and Chemicals that may Produce Peliosis Hepatis Anabolic steroids Arsenic Azathioprine Contraceptive steroids Danazol Diethylstilbestrol Estrone
Glucocorticoids Medroxyprogesterone Tamoxifen Thioguanine Thorotrast Vinyl chloride Vitamin A excess
Parent drug Bioactivation
DILI DUE TO SPECIFIC AGENTS ANESTHETICS Of the agents currently in use to induce and maintain anesthesia, it is only the halogenated volatile agents that have clinically significant
Modified protein Kupffer cell
Native protein
Ag-processing APC
Modified protein
Reactive metabolite
Dendritic cell
Helper T cell MHC class 2
APC Processed Ag T cell activation + proliferation
Cytotoxic T cells
B cells producing antibodies
T cell-mediated killing of hepatocytes
Figure 26-10. Likely mechanisms for pathogenesis of drug-induced immunoallergic hepatitis.
517
Section IV. Toxin Mediated Liver Injury
Table 26-7. Classification of DILI: Comparison of Intrinsic (Predictable) vs Idiosyncratic (Unpredictable) DILI Variable
Type of drug-induced liver injury Intrinsic
Idiosyncratic With immunoallergic features
Without immunoallergic features
Predictability/dose dependence
High/yes All subjects given high doses will develop hepatotoxicity Regularly produced in experimental animals
Low/slight or nil Most subjects will not develop hepatotoxicity, regardless of dose Not reproducible in experimental animals
Low/slight or nil Most subjects will not develop hepatoxicity, regardless of dose Not reproducible in experimental animals
Associated features
Toxic damage to other tissues also occurs Drug-induced renal, pancreatic injury common
Fever, skin rash, peripheral adenopathy, eosinophilia Development of autoantibodies (ANA, ASMA), hyperglobulinemia
No extrahepatic manifestations of immunoallergic responses
Underlying risk factors
Induction (without inhibition) of enzymes that increase formation of toxic intermediates Conditions that decrease metabolism, detoxification, and removal of toxic intermediates
Allergic diathesis Other host genetic factors presumed to play a role such as presence of certain HLA types, factors that influence Th1 vs Th2 phenotypes, etc. Women more susceptible than men (as for most AI diseases)
Other host genetic factors presumed to play a role, such as genetic variations that influence expression of phase 1–3 enzymes of drug metabolism Presence of underlying liver disease, especially chronic viral hepatitis in subjects receiving HAART and fatty liver in subjects receiving methotrexate or dugs implicated in producing steatohepatitis.
Typical pattern of injury
Hepatocellular, acute
Hepatocellular, acute Less often, cholestatic or mixed
Hepatocellular, acute Less often, cholestatic or mixed
Response to rechallenge
Reproduced promptly and dependably
Very rapid recurrence (1–3 doses)
Variable—may be delayed for several weeks, usually more rapid than the initial episode
Examples of inciting agents
Acetaminophen Amanita toxins Bromobenzene Carbon tetrachloride Chloroform Halothane White phosphorus
Amoxicillin/clavulanic acid Alphamethyldopa Diclofenac Doxycycline Fenofibrate Halothane Hydralazines Minocycline Nitrofurantoin Penicillins Phenlbutazone Phenytoin Quinidine Statins (very rarely)
Amoxicillin/clavulanic acid Chlorpromazine (other phenothiazines) Enflurane Fluroxene Glitazones (rosi-, pio, troglitazone) Isoniazid Nifedipine Penicillins Phenelzine Phenylbutazone Propylthiouracil Statins (rarely) Sulfonylureas Quinidine
Typical duration of exposure, prior to onset
Very brief (<1 week), e.g. acetaminophen overdose in attempted suicide
Brief (1–5 weeks), e.g., development of allergic reaction to phenytoin
Valproic acid Highly variable, depending upon poorly defined host susceptibility factors (1–100 weeks), e.g. variable onset of glitazone-induced DILI
hepatotoxicity. The halogenated anesthetics, beginning with halothane in the 1950s, replaced the routine use of ether and chloroform. Halothane, besides being non-flammable, had much better pharmacokinetics than ether, and had fewer respiratory and cardiac side effects than chloroform. However, it was soon recognized that severe postoperative liver injury sometimes occurred, especially in patients re-exposed to halothane.95,101,102 Halothane was also an occupational hazard for those administering the anesthetic.103 The subsequent development of other halogenated agents, such as enflurane, isoflurane, desflurane, and sevoflurane, was associated with less hepatotoxicity, most likely because of the lesser hepatic metabolism
518
of these agents. However, the latter agents104–107 have also all been reported to cause liver injury, although sevoflurane may be the safest.108 Methoxyflurane, no longer available in the US, was known to cause hepatotoxicity109 and is metabolized to a greater extent than halothane. The incidence of hepatotoxic reactions after halothane has been estimated between 1:3000–30 000.110 The incidence after enflurane has been estimated to be ~1:800 000.111 The newer agents have an even lower incidence. Studies of pathogenesis support an immune mechanism for liver injury,96,112 with identification of halothane metabolite-modified neoantigens in injured livers113 and IgG antibodies to neoantigens
Chapter 26 DRUG-INDUCED LIVER INJURY
in sera of patients.114 CYP2E1 appears to be the cytochrome P450 responsible for the oxidative generation of the reactive metabolite.115 When halothane, isoflurane, and sevoflurane were administered to dogs,116 all caused aminotransferase elevations postoperatively, but halothane caused toxicity much earlier and to a greater degree than the other agents. The clinical features of halogenated anesthestic liver injury were reviewed years ago,95,101,103 and recently.117,118 Most individuals had undergone anesthesia on multiple occasions and there can be crosssensitization between agents. Non-specific symptoms, including fever and malaise, occur days to weeks postoperatively and are followed by marked aminotransferase elevations and then jaundice. Onset is variable, with jaundice occasionally taking over a month to develop. The usual histologic pattern of liver injury is centrilobular necrosis,119 but cholestatic features can also occur.120 The most frequently affected patients are obese women between 40 and 60 years old. A small percentage of patients go on to develop fatal fulminant liver failure, with some rescued by transplantation. However, complete resolution without residual liver dysfunction occurs in the majority of cases. Patients who have survived such reactions should not be re-exposed to any halogenated anesthetics. Anesthetic agents not reported to cause DILI include the barbituates [thiopental (Pentotnal), methohexital (Brevitol)], ketamine (Ketalar), propofol (Diprivan), the opioids [Alfentanil (Alfenta), fentanyl (Uublimaze), remifentanyl (Ultiva), sufentanyl (Sufenta)], etoridate (Amidate), dexmedetomidine (Precedex), and all the topical “–caines’.
ANTICONVULSANTS Many anticonvulsants have potential hepatotoxic effects (Table 268), including agents that have been in use for decades (e.g. phenytoin, valproic acid, carbamazepine), as well as newer ones such as felbamate and lamotrigine. Because of the clinical importance of seizure control, all of these agents remain in use despite their potential toxicities. Felbamate is considered an ‘adjunct’ second-line agent, to be used only if other agents are ineffective. One agent (-vinyl-GABA (vigabatrin), was withheld from the US market because its long-term use caused visual field defects,121,122 but the drug may be useful for treating methamphetamine and cocaine addiction.123 Many other anticonvulsant agents will probably become available in the future,122 and careful postmarketing monitoring will be required to elucidate their toxicities.
Phenytoin Phenytoin has been known for decades to cause hepatotoxicity in association with hypersensitivity reactions.124–126 The majority of patients on phenytoin develop mild elevations of AP and GGTP within the first few months of therapy that normalize with continued use of the drug. These changes are part of hepatic adaptation and generally do not require therapy to be stopped. It is the hypersensitivity reaction that is most worrisome.127 The formation of the reactive arene-oxide metabolite by CYP2C9, followed by formation of the orthoquinone by CYPs 2C9, 2C19, and 3A4, leads to haptens and immune activation. The incidence of this idiosyncratic non-dose related hepatotoxicity is estimated to be <1:10 000, and 56% of phenytoin hypersensitivity is associated with some hepatotoxicity.127 Liver failure necessitating transplantation still occurs.1
Table 26-8. Anticonvulsants and DILI Reported to cause DILI Drug
Comments
Phenytoin (Dilantin) Fosphenytoin (Cerebyx) Valproic acid, divalproex sodium (Depakote) Clonazepam (Klonopin)
Immunoallergic; hepatic > cholestatic Mitochondrial
Carbamazepine (Tegretol, Carbatrol) Oxcarbazepine (Trileptal) Felbamate (Felbatol) Lamotrigine (Lamictal) Topiramate (Topamax)
Very rare idiosyncratic (one case report147) Idiosyncratic Hepatitic > cholestatic Cholestatic > hepatitic
Not reported to cause hepatotoxicity Ethosuximide (Zarontin), gabapentin (Neurontin), levetriacetam (Keppra), phenobarbital* primidone (Mysoline), tiagabine (Gabitril), zonisamide (Zonegram) *Phenobarbital activates orphan nuclear receptor CAR and exerts well known proliferative effects on hepatocytes.
The clinical symptoms usually manifest within 1–8 weeks of drug exposure and include fever, malaise, lymphadenopathy, splenomegaly, and rash. Serum aminotransferases are elevated 2–100-fold (ALT>AST) and AP two- to eightfold.125,127,128 Leukocytosis and atypical lymphocytes suggesting mononucleosis and eosinophilia are common, with a lupus-like syndrome and pseudolymphoma reported occasionally. Other organ system toxicities can include interstitial nephritis, myositis and rhabdomyolysis, pneumonitis, and marrow suppression. The clinical presentation can also simulate viral hepatitis. When liver biopsies are performed, the histology shows a panlobular mixed mononuclear and polymorphonuclear infiltrate with prominent eosinophilia. In 10% of cases cholestasis is the predominant finding. The findings are not specific, however. Therapy is discontinuation of the drug, which in most cases leads to resolution of toxicity. However, once liver failure develops, the case:fatality ratio can be as high as 40%. Because of cross-reactivity with carbamazepine and oxcarbamazepine,126,129 these latter agents should not be used to replace phenytoin for seizure control in subjects who have experienced symptomatic phenytoin toxicity. A phosphate ester prodrug of phenytoin, fosphenytoin, developed for parenteral administration,130 should also be avoided.
Carbamazepine Like phenytoin, carbamazepine can also cause asymptomatic mild elevations in serum GGTP (64%) and AP (14%) that do not require discontinuation of therapy.127 However, aminotransferase elevations, seen in 22% of patients, may indicate susceptibility to develop the more serious idiosyncratic hypersensitivity reaction. A Swedish analysis131 estimated the risk to be about 1 in 6000, which is more common than with phenytoin. The hypersensitivity reaction is also due to formation of a reactive metabolite, probably an unstable
519
Section IV. Toxin Mediated Liver Injury
epoxide formed by CYP3A4.132 Drug toxicity usually occurs within 8–16 weeks of therapy and presents with fever, rash, and peripheral eosinophilia. Marrow suppression, nephritis, and pneumonitis can also occur. Carbamazepine is more likely than phenytoin to cause a pure cholestatic pattern of hepatotoxicity, which occurs in 30% of reactions. A mixed pattern of liver injury with AP, bilirubin, and aminotransferase elevations occurs in 50% of cases. A predominantly hepatocellular injury may have a worse prognosis.127 Consistent with the cholestatic clinical picture, histopathology commonly demonstrates a granulomatous reaction with eosinophilia.133 Resolution of injury takes several weeks after drug withdrawal. Because the injury is immune mediated rechallenge is not recommended, and both phenytoin and oxcarbamazepine should also be avoided.129,134
serum aminotransferases. Valproate therapy has also been reported to decrease serum albumin concentrations by up to 30% without apparent toxicity in a small study of children with severe neurologic disabilities.143 Because of the potential of valproate to damage mitochondrial function, L-carnitine was suggested as a protective agent144 and may improve survival with severe valproate hepatotoxicity,145 especially if given intravenously. Oral supplementation of 100 mg/kg/day is recommended for infants and young children taking valproate, and for patients with symptomatic hyperammonemia or multiple risk factors for hepatotoxicity.146 Prophylactic use of L-carnitine decreases the risk of valproate hepatotoxicity and is recommended.
Clonazepam Oxcarbazepine This keto analog of carbamazepine, first introduced in 1990 in Denmark, has recently become available in most countries, including the USA. It is considered to be a safe and useful anticonvulsant135 with fewer P450-related drug interactions than carbamazepine.122 However, it has also been reported to cause acute liver failure due to hypersensitivity,136 with a similar clinical presentation to that of carbamazepine and phenytoin.134
Valproic Acid This may be the most widely prescribed anticonvulsant worldwide,137 and in general is considered very safe, with the incidence of hepatotoxicity in adults and children older than 2 years being approximately 1 in 35 000.138 However, in children under 2, especially if taking other anticonvulsants, the incidence may be as frequent as 1 in 600. It is also clear that patients with genetic mitochondrial enzyme defects are at greater risk,139 most likely because of its depletion of coenzyme A levels and its metabolism via mitochondrial oxidation. Hepatotoxicity is the most common serious toxicity of the drug, and usually occurs within the first 3–6 months of therapy.127 Valproate’s hepatotoxicity is most likely dose related,139 although epidemiologic studies have suggested that other host factors and polypharmacy may be more important.138 Up to 40% of patients have transient asymptomatic ALT elevations that improve with dose reduction.127 High doses of drug, in addition to young age and polypharmacy, are significantly associated with higher excretion of thiol conjugates of the toxic valproate metabolite (E)-2,4-diene VPA.140 Therefore, ALT monitoring is recommended for the first 6 months of therapy and after dose increases. Patients taking olanzepine in addition to valproate had higher ALT elevations than with either drug alone.141 Although no specific degree of ALT elevation has been identified as an indicator of impending hepatic failure, a greater than threefold elevation should prompt drug cessation. If fever, nausea, vomiting, and abdominal pain accompany laboratory evidence of developing hepatic failure and poor seizure control, then liver failure will probably become irreversible. The characteristic histopathology of valproate hepatoxicity is that of microvesicular steatosis, similar to Reye’s syndrome, seen mainly in zones 2 and 3.127 These changes may occur without toxicity. In one recent report 61% of patients on long-term valproate had sonographic evidence of fatty liver,142 with the majority having normal
520
Although the medication database www.epocrates.com lists hepatotoxicity as a serious adverse effect of clonazepam, no hepatotoxicity is mentioned as an adverse event in the manufacturer’s drug insert (www.pdr.net), and only one case report could be found in the literature.147 Therefore, it is unlikely that clonazepam has any significant hepatotoxicity.
Felbamate When this aromatic compound was approved for use in 1993, it was the first new anticonvulsant approved in the US since the introduction of valproate in 1978. However, during its first year of use an incidence of hepatic failure of 1 in 6000 (and an aplastic anemia incidence of 1 in 3000) prompted restriction of its use to severe epilepsy not responding to other agents.122 However, felbamate is still considered an important drug, with over 8000 patients treated annually in the USA. By 1996, 36 cases of hepatotoxicity had been collected by the FDA, with five deaths.148 However, since then no further cases of hepatotoxicity have been reported. The mechanism of toxicity appears to involve the formation of an aldehyde monocarbamate149 that is activated to atropaldehyde.127,150,151 Because felbamate is usually given with other anticonvulsants and is a CYP3A4 substrate, studies of drug interactions have been carried out. A recent in vivo study152 suggests that felbamate may heteroactivate CYP3A4 to promote the formation of carbamazepine-10,11epoxide when these agents are used together. The paucity of reported cases makes it difficult to describe the clinical characteristics and histopathology of felbamate hepatotoxicity. However, presentation occurs between 3 weeks and 6 months, with a possible female preponderance.127
Lamotrigine This chlorinated phenyltriazine anticonvulsant has been in use for over a decade, and the first case of fulminant hepatic failure due to it was reported in 1995.153 Another severe case of an 8-year-old boy who recovered was reported in 1998.154 Despite over 2 million prescriptions written,122 only nine cases of hepatotoxicity have been reported so far in the literature155 and most were on polytherapy. The most common adverse event of lamotrigine is skin rash,156,157 which occurs in 3–10% of patients127 and can be a severe Stevens–Johnson syndrome. Whether the metabolism of lamotrigine to a reactive arene oxide158 is responsible for both the cutaneous toxicity and the rare hepatotoxicity is not yet clear.
Chapter 26 DRUG-INDUCED LIVER INJURY
Topirimate
Pergolide
This anticonvulsant, also marketed to prevent migraines, is related to carbonic anhydrase inhibitors and can cause a metabolic acidosis. It has been considered very safe with few side effects (the majority being CNS), especially if started at low doses and increased slowly (<50 mg/week).159 Only one case of acute liver failure necessitating a transplant has been reported, in a woman also on carbamazepine.160 Five additional cases of reversible hepatotoxicity in three children and two adults, all of whom were also on valproate, have since been reported.161–163 With over 3 million patients having received this drug worldwide, with monitoring by pharmacovigilance databases,122 it is not likely that topiramate will be deemed to have serious hepatotoxicity.
This dopamine agonist is approved for adjunctive treatment of Parkinson’s disease, and is also useful for restless leg syndrome170 and prolactinomas.171 Both the Epocrates database (www.epocrates.com) and the package insert (www.PDR.net) mention abnormal ALTs and hepatitis. However, no literature reports of liver toxicity could be found.
ANTIPARKINSON’S/ANTIMIGRAINE/ ANTI-ALZHEIMER’S/OTHER NEUROLOGIC AGENTS (Table 26-9) Tolcapone In 1998 this catechol-O-methyltransferase (COMT) inhibitor was the first of its drug class to be approved for adjunctive therapy with levodopa in Parkinson’s disease.164 Although no DILI was reported prior to approval, subsequent clinical trials noted ALT elevations more than three times normal in about 1% of patients on 100 mg/day, and 3% of those on 200 mg/day.165 In 1998 three fatal cases of acute liver failure occurred out of 60 000 patients exposed, which led to the drug’s withdrawal in Europe and Canada, and restrictions on its use in the USA.164,165 More than 90% of cases of serious toxicity occurred within 6 months of starting therapy. Tolcapone, but not a subsequently approved COMT inhibitor entacapone, was shown to uncouple oxidative phosphorylation in rat liver mitochondria.166 Oxidative metabolism by CYPs 1A2 and 2E1 may activate a tolcapone amine or acetylamine metabolite to a reactive species.167 In 2000 an expert panel suggested that tolcapone was safe to use with frequent ALT monitoring during the first 6 months of therapy, and that the drug should be stopped if ALT was two- to threefold elevated,168 although a more recent Cochrane Review has questioned the adequacy of this recommendation.169
Methazolamide A carbonic anhydrase inhibitor used in treatment of glaucoma, this drug can also be used for therapy of essential tremor. The Epocrates database mentions hepatic necrosis and dysfunction. However, only one possible case of hepatitis in association with red blood cell aplasia was reported in 1981.172
Tacrine This acetylcholinesterase inhibitor has a mild beneficial effect on Alzheimer’s disease, but can cause severe hepatocellular injury.173,174 Analysis of multicenter clinical trials in the USA, France, and Canada175,176 demonstrated that 25% of patients who take tacrine develop asymptomatic ALT elevations more than three times normal, usually within the first 6–12 weeks of therapy. Hepatic abnormalities are almost always reversible, and only one fatal case of hepatotoxicity has been reported in a 75-year-old woman who had been on the drug for 14 months.174 The mechanism of toxicity has been suggested to involve a hypoxia–reoxygenation injury mediated via the sympathetic nervous system,177 and an alteration in membrane fluidity that is not related to lipid peroxidation.178
Glatiramer This injectable preparation of synthetic polypeptides related to myelin basic protein may have some efficacy in slowing relapsing–remitting multiple sclerosis,179 although a recent Cochrane Review has not supported this claim.180 The Epocrates database mentions hepatotoxicity in its long list of adverse events. However, no citation could be found, and liver injury was not mentioned in the Cochrane Review.
Riluzole Table 26-9. Anti-Parkinson’s, Antimigraine, Anti-Alzheimer’s, Other Neurologic Drugs and DILI Drugs
Comments
Tolcapone (Tasmar) Methazolamide Pergolide (Permax) Tacrine (Cognex)
≠ALT, acute liver failure (4 cases) only one 1981 case report ‘hepatitis’ in database, but no citations ≠ALT common and must d/c or acute liver failure ‘hepatotoxicity’ in database, but no citations ≠ALT, 3 cases of acute hepatitis ≠ALT in database, but no citations
Glatiramer (Copaxone) Riluzole (Rilutek) Modafinil (Provigil)
Not reported to cause liver injury Amantidine (Symmetrel), benztropine (Cogentin), biperidan (Akinetan), bromocriptine (Parlodel), carbidoba/levodopa (Sinemet), donepezil (Aricept), entacapone (Contan), galantamine (Reminyl), memantine (Nameda), pramipexole (Mirapex), rivastigmine (Exelon), ropinirole (Requip), selegiline (Eldepryl), trihexyphenidyl (Artane), All the ‘ -triptans’
This centrally acting glutamate antagonist, used in Europe since 1995, is the first drug to be approved by the FDA for amyotrophic lateral sclerosis. It appears to prolong survival in ALS patients by approximately 1 month.181 It is regarded as relatively safe, although three cases of acute hepatitis have been reported182,183 and biopsies showed inflammation and microvesicular steatosis. No deaths due to hepatic failure have been reported. Recent reviews181,184 list ALT elevation as one of the most frequent adverse events, and serum ALTs more than three times normal are observed in 10–15% of patients.185 This last report recommends strict ALT monitoring of patients, and avoiding the drug if ALTs are already elevated.
Modafinil Approved for narcolepsy and sleep apnea, this drug is listed in the Epocrates database as causing elevated aminotransferases. However, no citation could be found, and a recent study of 151 patients did not report any instances of liver test abnormalities.186
521
Section IV. Toxin Mediated Liver Injury
ANTIDIABETIC AGENTS Sulfonylureas As a group, the sulfonylureas have a very low incidence of hepatotoxicity. The Epocrates database lists rare cholestatic events as occurring after glipizide, glyburide, and chlorpropamide, which is most likely due to a hypersensitivity reaction.187 Although tolbutamide, glimepiride, and tolazamide are not listed in the database, reports of acute cholestatic hepatitis, chronic vanishing bile duct syndrome, and granulomas have been reported for all of these agents.128,188
Repaglinide This benzoic acid derivative is listed in the Epocrates database as causing rare hepatic dysfunction. However, a recent review of five large long-term trials did not mention liver toxicity189 (and in addition found no liver toxicity with glyburide). Despite this track record, there has been a recent report of acute hepatotoxicity.190 Because repaglinide is metabolized by CYP2C8, caution is advised when also using inhibitors of this CYP, such as trimethoprim.191
Metformin This is the only remaining biguanide medication available for use. The other two agents in this class, buformin and phenformin, were withdrawn in the 1950s because of a high incidence of lactic acidosis.192 Metformin’s risk of lactic acidosis is lower, and the drug is considered safe as long as dose adjustments are made for renal impairment, liver impairment, surgery, and the use of radiologic contrast.193 Despite the Epocrates database not mentioning any hepatotoxicity with metformin, several cases of acute cholestatic hepatitis have been reported.194,195 The mechanism of toxicity is not known. Although metformin is felt to have added risks when used in patients with liver disease,187,192 because it increases insulin sensitivity at the receptor level,196 it has been suggested as possible therapy for patients with non-alcoholic steatohepatitis who are insulin resistant. However, a pilot open-label trial197 demonstrated only transient beneficial effects.
Thiazolidinediones These agents are peroxisome proliferator-activated receptor-g (PPAR) agonists that have complex metabolic effects on adipose cells, myocytes, and hepatocytes to improve overall insulin sensitivity.198 Troglitazone was the first of this class to reach the market, in 1997, but was withdrawn in 2000 after a large number of cases of serious hepatotoxicity, a number of which were fatal or required liver transplantation.202 Analysis of postmarketing and other data by FDA reviewers estimated a high risk of liver failure (1:600–1:1500) at 26 months of drug use.202 The other two agents, pioglitazone and rosiglitazone, have a lesser tendency to cause hepatotoxicity, although cases have been reported.198,199 In an analysis of 13 clinical trials,201 the frequency with which troglitazone caused increases in serum ALT more than three times the ULN was 1.91%, compared with only 0.26% for pioglitazone and 0.17% for rosiglitazone. Furthermore, 0.68% of the subjects on troglitazone developed serum ALT levels more than 10 times ULN, whereas no one developed such increases while taking the other agents. Analysis of postmar-
522
keting and other data by FDA reviewers estimated a high risk of liver failure (1:600–1:1500) at 26 months of drug use.202 The mechanism of liver injury for troglitazone is unknown. The drug does undergo metabolism, probably by CYP3A4 and to a lesser extent CYP2C8, to a reactive sulfonium ion that covalently binds to microsomal protein and GSH. The reactive metabolite may also adversely affect basolateral organic anion transporters.203 However, the parent drug is also hepatotoxic. The clinical presentation of hepatotoxicity in the reported cases was generally delayed, with a mean presentation after 4 months of use.202 Therefore, current recommendations relating to glitazone use include avoiding starting the drug if the baseline serum ALT is more than 2.5 times the ULN, and monitoring the ALT every 2 months for the first year. In addition, the drug should be stopped and immediate medical care sought if unexplained nausea, vomiting, abdominal pain, fatigue, anorexia, or dark urine develop, or if ALT becomes more than three times the ULN. The value of regular monitoring of serum ALT has not been established, and for troglitazone19 of 94 patients progressed from normal tests to irreversible liver injury within 1 month.202
ANTIMICROBIAL AGENTS ANTIMICROBIALS – ANTIFUNGAL/ANTIPARASITIC/ANTIMALARIAL ANTITUBERCULAR Amphotericin B (Table 26-10) Significant hepatotoxicity is very unlikely in view of this drug’s widespread use for so many years and the paucity of reports (a total of three documented cases204) describing DILI. Nephrotoxicity remains its most important toxic side effect.
Ketoconazole and Other Azoles Ketoconazole is an imidazole oral antifungal drug and a potent competitive inhibitor of hepatic CYP3A.205,206 This latter property has probably led to more adverse drug reactions than direct hepatotoxicity. However, shortly after its introduction in 1981, ketoconazole was noted to cause an acute hepatitic or mixed form of acute liver injury, often with jaundice, which was more common in women and elderly patients.207–209 Pure cholestatic injury occurred in 10% of cases. Evidence of allergy was rare. Injury was noted on average after 8 weeks of therapy. Fatalities were uncommon, and usually associated with continuation of the drug. A more recent prospective cohort study210 showed that 18% of patients have transient asymptomatic elevations of serum ALT that normalize with continued therapy (adaptation), but 3% develop clinical hepatitis. Another recent cohort study211 tracked acute liver injury in over 69 000 patients on various oral antifungal drugs and found that ketoconazole’s incidence was 1 in 750 person-months, itraconzaole’s was 1 in 10 000, and terbinafine’s was 1 in 40 000. Although ketoconazole’s idiosyncratic injury is not considered immunemediated, a recent report of a woman developing severe hepatitis just 48 hours after an unintentional rechallenge suggests that an immune reaction can occur.212 Another case report suggests that cirrhosis may develop despite withdrawal of the drug and resolution of acute liver injury.213
Chapter 26 DRUG-INDUCED LIVER INJURY
Table 26-10. Antifungal, Antiparasitic, Antitubercular Drugs and DILI Drugs
Comments
Antifungals Ketoconazole (Nizoral) Fluconazole (Diflucan) Itraconazole (Sporanox) Voriconazole (Vfend) Terbinafine (Lamisil) Griseofulvin Caspofungin (Cancidas) iv Flucytosine (Ancobon)
hepatitic > cholestatic hepatitic, ?Cholestatic cholestatic, hepatitic too new, incidence unknown cholestatic > hepatitic very low incidence too new, incidence unknown hepatitic
Antiparisitics Thiabendazole (Mintezol) Mebendazole (Vermox) Albendazole (Albenza)
cholestatic>hepatitic ≠alts ≠alts
Antimalarials Pyrimethamine/sulfadoxine (Fansidar) Amodiaquine (not available in U.S.)
very rare very rare
Antitubercular Isoniazid Rifampicin Pyrazinamide Ethambutol (Myambutol) Dapsone Rifapentine (Priftin) Ethionamide (Trecator-SC)
Idiosyncratic hepatitic Idiosyncratic hepatitic Idiosyncratic hepatitic doubtful ≠alts Idiosyncratic hepatitic Idiosyncratic hepatitic
Not reported to cause hepatotoxicity Antifungals: amphotericin (only mild ≠ALTs), clotrimazole (Mycelex), miconzaole (Monistat), nystatin (Mycostatin) Antimalarials: chloroquine (Aralen), hydroxychloroquine (Plaquenil), primaquine, mefloquine (Lariam +/-≠LFTs), atovaquone/proguanil (Malarone +/-≠LFTs), pyrimethamine (Daraprin) Antiparasitics: pentamidine (Pentam), atovaquone (Mepron), praziquantel (Biltricide), pyrantel (Antiminth), ivermectin (Stromectol +/-≠LFTs), nitzoxanide (Alina) Antitubercular: streptomycin, rifabutin (Mycobutin), cycloserine (Seromycin +/-≠LFTs)
The mechanism of ketoconazole liver injury appears to involve formation of an N-deacetyl metabolite that is converted to a toxic dialdehyde by the flavin-containing mono-oxygenases.214,215 Treatment is drug cessation, and ursodeoxycholate may help prevent progressive cholestatic injury.216 Itraconazole, a less potent CYP3A inhibitor than ketoconazole,206 also causes less hepatotoxicity.211 In a pharmaceutical database study of over 54 000 itraconazole and fluconazole users,217 ‘serious adverse liver events’ were reported in only 1 in 30 000 prescriptions for either drug. If itraconazole was given as ‘pulse’ therapy (1 week/month ¥ 3), then no serious hepatotoxicity was found.218 However, three cases of cholestatic liver injury were reported in patients taking itraconazole long term. These patients presented with jaundice, and ductopenia was noted in two of the three biopsies.219 Focal nodular hyperplasia has been linked to itraconazole in one patient who had been taking the drug for 4 months.220 Fluconazole is considered very safe, with only 5% of 562 children developing transient elevations of serum ALT.221 In some reports of DILI, fluconazole was found to have been administered with nitrofurantoin222 or amphotericin B.223 Voriconazole is probably still too
new for its incidence of hepatotoxicity to be known, and no literature citations have been found. However, recent safety reviews suggest monitoring liver tests as well as visual change when using voriconazole.224,225
Terbinafine This allylamine antifungal has replaced pulse intraconazole for the treatment of onychomycosis, and is widely advertised. The incidence of hepatobiliary dysfunction in postmarketing surveillance has been reported to be as low as 1 in 40 000.211,226 However, recent case reports,227–229 and many others too numerous to cite, suggest a higher incidence with a predominance of cholestatic reactions, including one liver transplant patient who was initially thought to have acute rejection 5 years after his transplant.230 Toxicity can be seen as early as 1 week after starting the drug.229 The mechanism of injury might be the formation of an N-dealkylated allylic aldehyde that is conjugated with GSH and transported across the canalicular membrane.228,231
Griseofulvin This older antifungal agent has been the mainstay of therapy for tinea capitis, but may now be supplanted by the newer antifungals that require only 2–3 weeks of therapy instead of 6 weeks.232 Although GI side effects of the drug are common, only one case report from 1976 describes griseofulvin hepatotoxicity.233 A more recent prospective study210 showed no liver test abnormalities in 74 patients treated for 3 months.
Caspofungin This new echinocandin antifungal is the first of its kind to be approved and is only available for intravenous use. It can be used by itself or with liposomal amphotericin B, or with voriconazole for refractory invasive aspergillosis and candidiasis in immunosuppressed patients,224,234–236 but is not effective for cryptococcus. The drug inhibits fungal cell wall b-(1,3)-glucan synthesis. It appears to be metabolized by hepatic P450s and may inhibit CYP3A4.237 Caution is therefore required when it is used with ciclosporin A238 and other calcineurin inhibitors. However, nelfinavir did not alter its pharmacokinetics.239 Because of a paradoxical loss of efficacy against Candida spp. at high concentrations,237 it may best be used for complicated infections in combination. Because phase 1 and 2 trials commonly reported elevated liver enzymes234, its long-term safety and incidence of hepatotoxicity are still to be determined.
Flucytosine This oral antifungal, available since the 1970s, is known to cause elevations of serum ALT in 5–15% of patients.204 Its mechanism of hepatic injury is unknown, but appears to be dose-related.240 Its main use currently is for the treatment of severe fungal infections. It has been used successfully in combination with fluconazole241 for cryptococcosis in a liver transplant patient. Because its use is so limited, it is doubtful that its mechanism of hepatotoxicity will ever be known.
Antimalarials The toxicities of most antimalarials, such as chloroquine and hydroxychloroquine, are chiefly neurologic and hematologic.
523
Section IV. Toxin Mediated Liver Injury
However, pyrimethamine/sulfadoxine242,243 and amodiaquine242,244 have been associated with DILI and ALF when the drugs have been continued after the onset of jaundice. The incidence of serious hepatotoxicity is estimated to be 1 in 11 000–15 000.242 Amodiaquine is not available in the USA, but is widely used in countries with endemic malaria where drug-resistant strains are a problem.245,246
Benzimidazole Antiparasitics Thiabendazole, mebendazole, and albendazole all seem to cause elevations of serum ALT at times,247 but only thiabendazole, available since 1964, has been reported to cause a cholestatic hepatitis that has led to ductopenia and cirrhosis.248–251 The incidence appears to be low, but has not been determined, and no recent case reports have appeared.
Isoniazid, Rifampicin, and Pyrazinamide Isoniazid (INH), available since the 1960s, is well known to be a hepatotoxic drug that causes an idiosyncratic hepatitic reaction leading to overt clinical hepatitis in 0.3–1.0% of patients when used as monotherapy.252,253 It is the second most common drug responsible for ALF requiring liver transplantation in the US.1 Hepatotoxicity increases when used in combination with other agents, such as rifampicin253 and pyrazinamide.254 Despite the risks, INH continues to be used because it is still the most effective agent for Mycobacterium tuberculosis therapy.252 For patients with latent TB infection the use of INH monotherapy for 6 or 9 months is considered the therapy of choice,255 but compliance is often low256 (<65%). A 2month course of INH, rifampicin, and pyrazinamide, given twice weekly, was shown to improve compliance, but was three times more likely to cause serious hepatotoxicity than INH alone.257–260 Therefore, this drug combination is unlikely to supplant INH. Of patients on INH alone, 10% develop elevations of serum ALT, most of which normalize with continued therapy, but 0.6% progress to overt clinical hepatitis that requires drug cessation.253 With INH and rifampicin, serum ALT increases occur in 35% of patients and 2.73% develop overt, clinical hepatitis. Overt hepatotoxicity usually presents within 6 weeks of INH monotherapy, and with INH and rifampicin usually in the first 2 weeks,261 but can be as early as a few days.253 When INH is used with rifampicin and pyrazinamide, hepatotoxicity beginning after 4 weeks is more ominous than early toxicity.261 Histologic changes resemble those of acute hepatitis A or B, and include diffuse lobular inflammation with ballooning or confluent necrosis, and occasionally macrovesicular fat.262 The mechanism of INH hepatotoxicity involves the formation of acetylisoniazid that is hydrolyzed to monoacetylhydrazine. It is hypothesized that CYP2E1 then activates the monoacetylhydrazine to a toxic metabolite. The exact mechanism of toxicity remains unclear, and is complicated by the fact that INH is often used with other agents. INH itself appears to inhibit a number of human CYPs, including 1A2, 2A6, 2C19, and 3A4,263,264 whereas rifampicin is a potent inducer of CYPs 2B6,265 2C8,266 2C9,267 3A4, and 3A5,268 as well as some phase II enzymes, all via activating the pregnane X receptor (PXR). Pyrazinamide does not appear to inhibit CYPs;263 how it might enhance toxicity with INH or rifampicin also remains unknown.
524
A major controversy has been whether slow or rapid acetylators of INH are at greater risk of hepatotoxicity. The most recent study to address this question was from Taiwan269 and used NAT2 genotyping, rather than phenotyping. The authors studied 224 patients, all on INH, rifampicin, and ethambutol. Low-dose pyrazinamide was used for the first 2 months of therapy. Toxicity was defined as serum ALT increased more than twofold above ULN that were proved to be causally related to INH by withdrawal and rechallenge. The incidence of hepatotoxicity was 26% in the slow acetylators that made up a quarter of their patients, and 11% in fast acetylators (p = 0.003). Besides slow acetylator status, age is the next most important predictor of hepatotoxicity. Although DILI due to INH is rare below the age of 20 and increases to 2.3% above age 50,252 pediatric Japanese patients under 5 years old had an unexpectedly high rate of severe hepatotoxicity270 (>8%), to which the use of pyrazinamide may have contributed. A study from India also implicated the use of pyrazinamide.254 Other variables, including female gender, underlying liver disease related to alcohol, hepatitis B and C, and malnutrition have all been variably implicated in increasing the incidence of severe hepatotoxicity.252,254 A case of hyperacute liver failure in a young patient also receiving carbamazepine was recently reported.271 In an effort to avoid overt hepatitis, complex algorithms have been developed255 that include baseline laboratory tests for all but the most healthy young (<35 years) non-HIV infected adults, and monitoring at monthly intervals if on INH alone. If the patient is also on rifampicin and/or pyrazinamide, more intense monitoring is recommended 261 (e.g. twice weekly for 2 weeks, then every 2 weeks up to 2 months, and then monthly). INH should be stopped if ALTs become more than three times the ULN with symptoms, or more than five times the ULN without symptoms. Patients must be advised to look for side effects and report them immediately. If overt clinical hepatitis develops, all therapy must be stopped. As antitubercular therapy must usually be resumed, some experts272 recommend a stepwise reintroduction of drugs, starting with INH at a low dose, then pyrazinamide, and finally rifampicin, with careful monitoring. Only 7% of patients rechallenged in this way developed subsequent hepatotoxicity.
Ethambutol This agent is added to INH, rifampicin, and pyrazinamide to treat active TB when INH resistance is suspected. Although fatal hepatotoxicity is listed as a rare adverse event in the drug databases, no specific citations could be found to suggest this agent has significant hepatotoxicity. As it is almost always given as part of a multidrug regimen with known hepatotoxic agents, liver injury ascribed to ethambutol would be difficult to identify and characterize.
Dapsone Dapsone is a second-line agent for TB and is used for other disorders such as dermatitis herpetiformis, leprosy, malaria, and Pneumocystis carinii pneumonia in HIV-infected patients. It is metabolized by CYPs 2E1 and 2C isoforms273 to a hydroxylamine that can cause hemolysis, methemoglobinemia, and agranulocytosis.274 A cutaneous syndrome is also described275 that is often associated with mild hepa-
Chapter 26 DRUG-INDUCED LIVER INJURY
totoxicity. When dapsone is given with pyrimethamine, the rate of serious hepatotoxicity has been estimated to be low, only 1 in 75 000.242
Table 26-11. Antiviral Drugs and DILI
DRUG
Rifapentine It is hoped that this long-acting cyclopental derivative of rifampicin, which can be given once weekly, will improve adherence to antitubercular treatment regimens. It has been shown to be safe and effective in HIV-negative pulmonary TB patients,276,277 and no dosage adjustments are required in patients with cirrhosis.278 However, in one study of its use with INH, adverse events were twice as high with rifapentine as with rifampicin.279 Because rifapentine is also an inducer of CYPs, it has significant drug interactions. Hepatotoxicity is listed in the drug databases, with an incidence of serum ALT elevations during combination TB therapy being slightly less than with rifampicin.
Ethionamide This agent is rarely used, but hepatotoxicity was recognized and reported in the 1960s.280 Because it is related to INH, its toxicity is probably due to an idiosyncratic hepatitis.
ANTIMICROBIALS–ANTIVIRALS Anti-HIV Agents (Table 26-11) With the development of effective antiviral agents for HIV and the dramatic decrease in mortality from AIDS in the 1990s, owing to highly active antiretroviral therapy (HAART), liver disease in HIV patients has emerged as an important cause of morbidity and mortality. The main culprits in promoting liver disease in this population are thought to be sequelae of co-infection with HIV and hepatitis viruses B and/or C,281–284 and different algorithms have been proposed for monitoring patients on HAART, depending on whether there is pre-existing liver disease, or co-infection with hepatitis B or C.285 How and why co-infection increases DILI and mortality is controversial. Some studies suggest that hepatitis C hampers the effectiveness of HAART.286,287 Others suggest that hepatitis C has no impact on survival.288 Liver damage by hepatitis B was probably controlled with lamivudine, which was a component of most HAART regimens, and now other effective agents are available for lamivudine-resistant strains. Therefore, the increasing incidence of liver failure in HIV patients could be drug related.289,290 Serious or severe drug-related ADRs occurred in 16% of patients in one large study,291 but very few of these were liver related, despite 20% of patients being noted to have asymptomatic increases in serum ALT. The complexity of this patient population on multiple therapeutic drugs and herbal remedies, with a high incidence of alcoholism (in those patients with a history of IV drug abuse), and a high incidence of hepatitis C co-infection (32–68%)281,287,288 and with frequent opportunistic infections, makes characterizing DILI difficult. Several recent comprehensive reviews of the hepatic injury from these agents have been published.292–295 Regular monitoring of liver tests is considered mandatory for patients started on HAART,285,290 with monthly testing for the first 3 months and then every 3 months thereafter. Avoidance of alcohol should always be recommended. Asymptomatic increases in serum ALT that are less than five times the ULN are considered relatively
REPORTED TO CAUSE DILI COMMENTS
Anti-HIV, NRTI Lamivudine (Epivir) Zidovudine (Retrovir) Didanosine (Videx) Stavudine (Zerit) Emtricitabine (Emtriva) Abacavir (Ziagen) Zalcitabine (Hivid) Anti-HIV, protease inhibitors Amprenavir (Agenerase) Lopinavir (Kaletra – with ritonavir) Saquinavir (Fortovase, Invirase) Indinavir (Crixivan) Atazanavir (Reyataz) Nelfinavir (Viracept) Ritonavir (Norvir) Fosamprenavir (Lexiva)
Mitochondrial (low incidence) Mitochondrial (low incidence) Mitochondrial (higher incidence) Mitochondrial (higher incidence) Mitochondrial (low incidence) Mitochondrial (low incidence) Mitochondrial (higher incidence) ≠ ALT ≠ ALT ≠ ALT ≠ ALT and inhibits UDPGT ≠ ALT ≠ ALT ≠ ALT and most potent CYP3A4 inhibitor ≠ ALT
Anti-HIV, Non-nucleoside RTI Efavirenz (Sustiva) Nevirapine (Viramune) Delavirdine (Rescriptor)
≠ ALT ≠ ALT ≠ ALT, ?immune features
Anti-HIV, HBV nucleotide RTI Tenofovir (Viread) Adefovir (Hepsea)
Very rare Very rare
Anti-CMV Cidofovir (Vistide)
Very rare
Not reported to cause DILI Anti-CMV: ganciclovir (Cytovene, +/- LFTs), foscarnet (Foscavir), valganciclovir (Valcyte) Anti-herpes: acyclovir (Zovirax), famciclovir (Famvir), valacyclovir (Valtrex) Anti-influenza: amantidine (Symmetrel), rimantidine, (Flumadine) zanamivir (Relenza), oseltamivir (Tamiflu)
‘safe’, but must be checked monthly until resolved, and other causes of hepatitis should be sought. If serum ALTs are more than five times ULN without symptoms, monitoring should be every 2 weeks. However, if clinical symptoms exist, or if serum ALTs are >10 times ULN, especially with lactic acidosis, HAART and other potentially hepatotoxic drugs must be stopped.
Nucleoside Analogue Reverse Transcriptase Inhibitors (NRTI) The most ominous toxic effect of NRTIs is the development of mitochondrial toxicity that leads to lactic acidosis and liver failure, similar to Reye’s syndrome.294 It is thought that depletion of mitochondrial DNA by NRTIs causes the mitochondrial dysfunction. This supposition was strengthened by findings that stavudine, didanosine, or zalcitabidine, which are known to deplete mtDNA in cultured hepatocytes,296 are associated with lower mtDNA in liver biopsy samples and higher serum lactate levels in HIV-HCV coinfected patients, compared with patients on zidovudine, lamivudine, and abacavir, which do not deplete mtDNA.297 Hepatitis C infection by itself might cause mitochondrial dysfunction,298 but it
525
Section IV. Toxin Mediated Liver Injury
is not yet clear whether co-infection increases the risk of mitochondrial toxicity. Stavudine is associated with more cases of lactic acidosis than other NRTIs,299 but all have been associated with cases of microvesicular steatosis and liver failure.292 Some experts281,294,299,300 advise the use of coenzyme-Q (30–60 mg tid), carnitine (1–3 g/day), riboflavin (50 mg/day), and/or thiamine (100 mg/day) in the event of lipoatrophy or severe lactic acidosis, but most of these interventions have not been studied in the setting of liver failure.
Protease Inhibitors (PI) The introduction of PIs in the 1990s marked the beginning of highly active HAART and control of HIV. All of the PIs (see Table 26-11) are used in combination with other antivirals, and increased serum ALT or AST more than five times ULN occurred in 1–9.5% of patients during the original registration trials.293 Ritonavir used in higher doses caused liver injury more frequently than the other agents, and is now given in lower doses in combination with other PIs. A characteristic of all PIs is that they are metabolized by and are inhibitors of hepatic CYPs, primarily 3A4, and ritonavir is the most potent.301 Furthermore, all except possibly indinavir, appear to be mechanism-based inhibitors of the P450, which means they irreversibly inactivate the enzyme. Nelfinavir may also affect CYP2C19.301a Indinavir also causes a reversible inhibition of UDPGT that leads to benign unconjugated hyperbilirubinemia in 12% of patients. Because ritonavir inhibits CYP3A so well, it is given with lopinavir in one of the more commonly used PI combinations, to prevent lopinavir’s metabolism. Liver toxicity with this combination is not higher than with nelfinavir-based HAART.302 The mechanism whereby PIs cause liver injury is not yet known,293 nor whether asymptomatic ALT elevations promote more rapid progression of fibrosis in patients co-infected with hepatitis B or C. Because PIs are always given in combination with other agents to patients with other causes of liver injury, sorting this out will be complex. However, inhibition of CYPs by PIs causes the most important clinically relevant ADRs. One example is the effect of lopinavir/ritonavir therapy for HIV in liver transplant patients who are receiving tacrolimus and must have profound dose reductions in that immunosuppressive agent.303
Non-nucleoside Reverse Transcriptase Inhibitors (NNRTI) Of the three available NNRTIs, nevirapine and efavirenz have been utilized to a much greater extent than delavirdine, possibly because of the latter’s being an inhibitor of several CYPs.304 A number of large cohort studies have shown that, when given with NRTIs and/or PIs, nevirapine is two to three times as likely as efavirenz to cause increases in serum ALT more than five times ULN.291,305–310 The incidence across multiple studies with nevirapine was 10%,307 with clinical symptoms in 4.9%. The hepatotoxicity with both nevirapine and efavirenz was often delayed, being recognized 3–9 months (median 5.5) after therapy began,310 and was more common in patients coinfected with hepatitis B and C. Another study has suggested that HAART regimens that contain nevirapine are associated with more rapid progression of hepatic fibrosis in hepatitis C patients,311 and that co-infected patients on PIs do better than those on NNRTIs.
526
Nevirapine has also been reported to cause severe hepatotoxicity and cutaneous reactions, both when used alone and when used with other antivirals for postexposure prophylaxis in non-HIV patients.312 These authors recommended avoiding nevirapine in prophylaxis regimens. The mechanism of hepatotoxicity for the NNRTIs may involve an idiosyncratic response with immune features, and this may be why the hepatotoxicity is worse in non-HIV subjects and is delayed in HIV patients.
Nucleotide Reverse Transcriptase Inhibitors Tenofovir is effective against HIV and does not appear to deplete mitochondrial DNA.313 Both tenofovir and adefovir are effective against hepatitis B.314,315 Although the Epocrates database lists hepatotoxicity as a serious reaction for both drugs, no citations could be found and adverse liver effects are not mentioned in the package inserts for either drug.
Cidofovir This anti-CMV agent, approved to treat CMV retinitis in HIV patients, causes nephrotoxicity and neutropenia as its major side effects (www.pdr.net). Hepatotoxicity is mentioned in the Epocrates database, but no citations could be found. Almost all of the anti-CMV agents appear to be devoid of hepatotoxicity.
ANTIMICROBIALS – ANTIBACTERIALS (Table 26-12)
Penicillins and Cephalosporins In general, the b-lactam antibiotics and structurally related cephalosporins have a very low incidence of hepatotoxicity, most of which is idiosyncratic with immune features that can manifest as either primarily hepatitic, primarily cholestatic, mixed, and/or granulomatous.204,247 That the same agent can present in several ways suggests that host factors must dictate the type of allergic injury. b-Lactamase-resistant agents, and penicillins with b-lactamase inhibitors, such as sulbactam and clavulanate, more often cause a cholestatic picture,316a–d occasionally with a prolonged clinical course.316e The incidence of DILI with amoxicillin alone is 1 in 30 000, but with clavulanate is 1 in 6000.316d If older patients are given repeated courses, the incidence may be as high as 1 in 1000. Piperacillin with another b-lactamase inhibitor, tazobactam, was noted to cause only mild increases in serum ALT in clinical trials,317 and no reports of cholestatic injury have yet been reported. Flucloxacillin, one of the earlier b-lactamase-resistant agents, had a very high incidence of chronic cholestasis318 and is no longer used in the USA. Only oxacillin and dicloxacillin are available in the US, with no serious hepatotoxicity listed for them in the Epocrates database. Yet, recent reports318,319 oxacillin of used intravenously in children suggest that hepatotoxicity cannot be discounted. For the cephalosporins, most of the reports of hepatotoxicity are from the 1980s.204 Ceftriaxone, which is excreted into bile, has been associated with the formation of biliary sludge and stones.320 In general, the cephalosporins have only a minor risk of hepatotoxicity. Although cefaclor and cefdinir are listed in the Epocrates database as causing cholestatic jaundice and hepatitis, no literature citations were found.
Chapter 26 DRUG-INDUCED LIVER INJURY
Table 26-12. Antibacterial Drugs and DILI Reported to cause hepatotoxicity Penicillins Ampicillin + sulbactam (Unasyn) Amoxicillin +clavulanic acid (Augmentin) Oxacillin Ticarcillin (Ticar) +clavulanic acid (Timentin) Piperacillin +tazobactam (Zosyn)
hepatitic > cholestatic cholestatic > hepatitic hepatitic > cholestatic cholestatic, granulomatous cholestatic, granulomatous unknown cholestatic ≠hepatitic or mixed
Cephalosporins Cefaclor (Ceclor) Cefdinir (Omnicef ) Ceftriaxone
possibly cholestatic possibly hepatitic biliary sludge
Macrolides Erythromycin Clarithromycin (Biaxin) Azithromycin (Zithromax) Telithromycin (Ketek)
cholestatic cholestatic cholestatic cholestatic
Quinolones Trovafloxacin (trovan) Probably all others
immune mediated FHF hepatitic or mixed
Sulfonamides Trimethoprim/sulfamethoxazole Sulfisoxazole (Gantrisin) Sulfadiazine
Immune hepatitic (esp HIV) cholestatic > hepatitic cholestatic > hepatitic
Tetracyclines Tetracycline Doxycycline Demeclocycline (Declomycin) Minocycline (Minocin, Vectrin)
mitochondrial in high doses hepatitic none reported autoimmune hepatitis, hepatitic
Other antimicrobials Nitrofurantoin Nalidixic acid Quinupristin/dalfopristin (Synercid) Fosfomycin (Monurol)
Acute hepatitic & chronic fibrosis possible cholestasis (one case) possible cholestasis ≠ALT
Not reported to cause significant hepatotoxicity Penicillins Pen V-K, ampicillin, nafcillin, mezlocillin (Mezlin) Cephalosporins Almost all first, second, and third-generation agents Other antimicrobials Chloramphenicol, aztreonam (Azactam, ±≠ALT), ertapenem (Invanz, ±≠ALT), meropenem (Merrem), clindamycin, metronidazole, tinidazole (Tindamax), furazolidine (Furoxone-not avail in US), vancomycin, daptomycin (Cubicin, ±≠ALT), imipenem/cilastin (Primaxin, ±≠ALT), linezolid (Zyvox), iodoquinol (Yodoxin), rifaximin (Xifaxan)
Macrolides Hepatotoxicity from erythromycin has been known for decades and can occur with either erythromycin base or any of the salts.204 Erythromycin toxicity is predominately cholestatic, owing to an idiosyncratic immunoallergic reaction. Often the clinical presentation will occur well after treatment has ended. Fever, jaundice, right upper quadrant pain, and nausea can present like acute cholecysti-
tis. Eosinophilia is often present. Fortunately the incidence is low (1 in 30 000).321a Although recovery can take many weeks, rarely is it fatal. Similar cholestatic presentations, including occasional fatalities, have been reported recently for clarithromycin,321b,322,323 azithromycin,324a–b and roxithromycin.324c However, the incidence of hepatotoxicity appears to be lower with the newer macrolides.325 Telithromycin, a new ketolide antibiotic that is a structural analog of erythromycin, was FDA approved in April 2004 to treat resistant Streptococcus pneumoniae respiratory infections and sinusitis.326,327 No reports of hepatotoxicity have appeared, although increased serum ALTs and hepatic dysfunction are listed as adverse reactions in the Epocrates database, and two instances have been entered into the US drug-induced liver injury network registry. Erythromycin and troleandomycin, a related macrolide no longer available, are well known to be potent inhibitors of CYP3A species and the p-glycoprotein transporter. Clarithromycin is also such an inhibitor.328,329 In fact, erythromycin and clarithromycin cause adverse drug interactions much more frequently than they cause cholestatic liver injury, especially with immunosuppressive agents such as ciclosporin A330,331 and tacrolimus,332 which require CYP3A metabolism. Azithromycin, roxithromycin, and dirithromycin are much weaker inhibitors of CYP3A.333 A potential link between CYP3A inhibition and cholestatic liver injury was identified when erythromycin and troleandomycin were found to block canalicular bile acid efflux in human hepatocytes much more effectively than the newer macrolides.334
Quinolones The fluoroquinolones are considered to be relatively safe antibiotics,335 with only trovafloxacin identified as having an incidence of hepatotoxicity (1 in 7000)204 appreciable enough to limit its use to serious infections in hospitalized patients. With trovafloxacin, the patients who developed ALF appeared to have a hypersensitivity reaction and were on medication for more than 14 days. However, case reports of hepatic failure have been published for most of the fluoroquinolones,336–339 and the Epocrates database lists increased serum ALTs as occurring with all of them.
Sulfonamides All the sulfonamides have been associated with reports of hepatotoxicity, usually thought to be idiosyncratic with immunoallergic features.204 A cholestatic clinical presentation is most common, usually with rash, fever, and eosinophilia. One case of intrahepatic cholestasis with phospholipidosis has been reported.340 Trimethoprim/sulfamethoxazole (TMP/SMX) is one of the oldest and most widely prescribed antibiotic combinations. The incidence of hepatotoxicity in the general population must be very low, because only occasional reports of hepatitis, ALF, and cholestatic disease appeared in the literature in the 1970s and 1980s.204 However, several recent case reports of liver failure341–343 serve to remind us of this potential. In HIV patients, the use of TMP/SMX has been noted to lead to a much higher incidence of allergic reactions (~20%) than in nonHIV patients.204,344–346 Because TMP/SMX is considered the best therapy for the treatment and prevention of Pneumocystis jiroveci pneumonia, desensitization protocols were developed347 and efforts
527
Section IV. Toxin Mediated Liver Injury
made to determine the cause of hypersensitivity. A slow acetylator status may contribute348 by allowing more drug to be activated by CYPs to sulfamethoxazole hydroxylamine.204 Co-treatment with CYP3A inducers leads to more measurable hydroxylamine, and inhibitors decrease this metabolite.349 No consensus has yet been reached to explain HIV-induced sensitivity. Interestingly, TMP/ SMX is widely used in developing countries as chronic prophylaxis against opportunistic infections, with some success,350–352 and no mention is made of severe allergic or hepatotoxic reactions with the use of the drug in these populations.
Tetracyclines The original descriptions of hepatotoxicity due to tetracyclines were in patients receiving high doses intravenously.204,247 The clinical presentation was similar to Reye’s syndrome, with ALF, renal failure, and acidosis. Serum ALTs were generally not very high (<1000 IU/mL), and histologic findings were of microvesicular steatosis with minimal necrosis. The mechanism of the steatosis and toxicity appeared to be inhibition of mitochrondrial fatty acid oxidation,353 and was probably dose related. Intravenous tetracycline is rarely used now, and the incidence of hepatotoxicity from oral low-dose tetracyclines is extremely low.354 However, reversible hepatic failure355 and prolonged cholestasis with paucity of bile ducts356 have been reported. Minocycline, which has been widely used as chronic therapy for acne in adolescents, was reported to cause both an acute hepatitis357 and chronic autoimmune hepatitis, with positive ANA and anti-smooth muscle antibodies.358,359 No reports of liver injury have been reported for demeclocycline, used mainly for SIADH, but the Epocrates database lists hepatotoxicity as a serious reaction.
Other Antibiotics Nitrofurantoin, still used as a chronic urinary antimicrobial, was reported in 1980 to cause acute and chronic liver disease.360 It is considered relatively safe to use as prophylaxis in children,361 but authors continue to report the development of DILI that resembles an autoimmune hepatitis in elderly women that may or may not improve with drug withdrawal.362,363 It is still a common cause of increased serum ALT that prompts referral to hepatologists,364 and caution with its use is warranted. Nalidixic acid is also used as a chronic urinary antimicrobial and is listed as possibly causing a cholestatic hepatitis, based on one case report from 1974.365 It is also photosensitizing. Whether it is safer than nitrofurantoin to use as a chronic medication is unclear. Quinupristin/dalfopristin is a relatively new streptogramin antibiotic for vancomycin-resistant enterococcus and resistant staphylococcal infections366 that caused hyperbilirubinemia when given to liver transplant recipients.367 Although that study suggested that the cholestatic changes were not drug related, another study368 suggested that elevated AP values might be drug related. The drug’s main side effect appears to be a myalgia/arthralgia syndrome. Fosfomycin, an epoxide low molecular weight antibiotic introduced in the early 1990s also for VRE and MRSA,369 was noted to frequently cause increases in serum ALT and Clostridium difficile colitis.370 One case of acute hepatotoxicity with repeated use in a cystic fibrosis patient has been reported.371
528
CARDIOVASCULAR AGENTS (Table 26-13) Antiplatelet/Anticoagulant/Thrombolytics A fixed dose of dipyridamole 200 mg and aspirin 25 mg (Aggrenox) taken twice daily has been shown to be effective in the secondary prevention of strokes,372,373 although there is still controversy as to its benefit over aspirin alone. Hepatic dysfunction is listed as an adverse reaction in the Epocrates database, but no literature citations could be found. On the other hand, ticlopidine, a thienopyridine inhibitor of platelet ADP-induced aggregation with thrombolytic effects,374 is more effective than aspirin in preventing stroke.375 Ticlopidine was first reported in 1993376 to cause hepatitis, and subsequent reports from many countries377–381 have confirmed that the drug causes a primarily cholestatic injury. Presentation was between 2 and 13 weeks and was not correlated with the degree of platelet inhibition.381 The hepatotoxicity can be associated with red cell aplasia,378 and an immune mechanism is therefore likely. Because of these side effects, another antiplatelet drug, clopidogrel, has become more popular382 and has been used successfully in one patient who had had ticlopidine hepatotoxicity.383 This was despite the fact that clopidogrel is also a thienopyridine. However, clopidogrel requires hepatic activation,384 and variable resistance to its efficacy has been found due to lack of activation by CYP3A4.385 Besides monitoring for antiplatelet effects, it is important to review medications that might inhibit CYP3A4 in patients who are on clopidogrel therapy. So far, only one case report of a mixed hepatocellular and cholestatic injury has appeared, in 2000.386 Therefore, clopidogrel probably has low hepatotoxicity. Hepatotoxicity due to warfarin is exceedingly rare, with only two case reports in the past decade,387,388 and in both cases the patients had been on phenprocoumon prior to warfarin. Cross-reactivity and rechallenge confirmed the sensitivity. Phenprocoumon, not available in the US, led to at least eight cases of severe hepatitis in Germany between 1992 and 2002, with one death and two patients requiring liver transplantation.389 Although no significant hepatotoxicity is reported with heparin, a number of the low-molecular weight heparins are listed in the Epocrates database as causing increased serum ALTs. For tinzaparin, cholestatic hepatitis is listed. However, only one citation390 could be found that reported increased ALTs for tinzaparin, and a more recent report 391 suggested that attribution of such increases is often erroneous (what was thought to be a dalteparin-induced reaction was judged on more careful investigation to be due to ranitidine). Therefore, the low-molecular weight heparins appear to be devoid of significant hepatotoxicity. No literature citation for liver toxicity could be found for the direct thrombin inhibitor lepirudin, and a recent review including postmarketing surveillance did not mention hepatotoxicity.200,387,391a
Angiotensin-Converting Enzyme (ACE) Inhibitors All of these drugs have an excellent safety profile, although hepatotoxicity and increased serum ALTs are listed in the Epocrates database. There is evidence that ACE inhibition might actually decrease fibrosis in the CCl4 animal model of liver injury,392,393 although another study found a beneficial effect of the angiotensin II receptor blocker candesartan, but not captopril.394 Many of the ACE
Chapter 26 DRUG-INDUCED LIVER INJURY
Table 26-13. Cardiovascular Drugs and DILI Reported to cause liver injury Antiplatelet/Anticoagulant/Thrombolytics Dipyridamole/ASA (Aggrenox) +/- hepatic dysfunction Ticlopidine (Ticlid) definite cholestatic Warfarin (Coumadin) +/- hepatitis/cholestatic jaundice Dalteparin (Fragmin) +/- ≠ALT Tinzaparin (Innohep) +/-cholestatic hepatitis Enoxaparin (Lovenox) +/- ≠ALT Lepirudin (Refludan) +/- ≠ALT ACE inhibitors All the ‘ -prils’
reports of cholestatic jaundice
Angiotensin receptor blockers All the ‘ -sartans’
hepatotoxicity, ≠ALT
Antiarrhythmics Amiodarone (Cordarone, Pacerone) Quinidine (Quinidex) Procainamide Propafenone (Rythmol) b-Blockers Labetatolol (Trandate, Normodyne)
Angiotensin II Receptor Blockers ≠ALT, phospholipidosis immune mixed with granulomas immune cholestasis with granulomas rare cholestasis
Antiarrhythmic Drugs rare hepatitic – immune rare hepatitic, rare granulomas rare hepatitic or cholestatic
Cholesterol lowering All statins Fenofibrate Niacin (slow release>crystalline)
≠ALT, rare hepatitic, cholestatic rare autoimmune picture rare acute hepatitic/cholestatic
Bosentan (Tracleer)
Although these drugs are relatively new, there have already been a number of reports suggesting that losartan can cause a mainly idiosyncratic hepatitic reaction,400,401 and candesartan,402 irbesartan,403 and valsartan404 have been reported to cause cholestatic hepatitis. Onset of clinical illness was always just a few weeks after starting therapy and resolution was relatively rapid after discontinuing the drug. The incidence of this reaction appears to be low.
≠ALT, hepatitic, non-immune
Ca2+ channel blockers Nifedipine Diltiazem (Cardizem) Verapamil
Other antihypertensive Hydralazine
inhibitors have been shown to have a low incidence of causing cholestatic hepatitis,128,187 including captopril,395,396 lisinopril,397 fosinopril,398 and ramipril.399 Whether the other congeners can cause this picture is not known. Patients who developed cholestasis were generally middle-aged and had been taking the drug for between 4 and 8 weeks. Most cases displayed a long recovery time. The sole fatality was a patient who had been continued on lisinopril for 3 weeks after developing jaundice, whose death was attributed to a perforated ulcer while his cholestasis was improving.397 Currently there are no known risk factors for developing cholestasis.
very rare immune hepatitic, granulomas ≠ALT, too new to characterize
Not reported to cause significant liver injury Antiplatelet/Anticoagulant/Thrombolytics Atteplase (Activase), clopidogrel (Plavix), anagrelide (Agrylin), dypridamole (Persantine), eptifibatide (Integrilin), cilostazol (Pletal), tirofiban (Aggrastat), abciximab (Reopro), fondaparinux (Arixta), bivalirudin (Angiomax), argatroban, antithrombin III (Atnativ), anistreplase (Eminase) streptokinase (Kabikinase), urokinase (Abbokinase), reteplase (Retevase), tenecteplase (TNKase) Antiarrhythmics Adenosine, bretylium, sotalol (Betapace), ibutilide (Corvert), moricizine (Ethmozine), mexiletine (Mexitil), disopyramide (Norpace), flecanide (Tambocor), dofetilide (Tikosyn) b-Blockers Almost all except labetolol appear without significant toxicity Ca2+ channel blockers All but verapamil and diltiazem appear safe Cholesterol lowering Gemfibrazole (Lopid), ezetimibe (Zetia) Diuretics All loop and thiazide appear without significant hepatotoxicity Other antihypertensive Minoxidil (Loniten), eplerenone (Inspra), treprostinil (Remodulin), epoprostenol (Flolan), fenoldapam (Corlopam), doxazosin (Cardura) clonidine (Catapres), terazosin (Hytrin), guanabenz (Wytensin), prazosin (Minipress), nesirtide (Natrecor)
The highly effective and widely used iodinated benzofuran antiarrhythmic amiodarone has long been known to cause liver injury.187 Its pulmonary toxicity is more serious, but elevations in serum aminotransferases or AP are common.405,406 Amiodarone was the drug most commonly associated with liver injury in one tertiary hepatology referral center.364 The spectrum of amiodarone liver injury is broad. An acute hepatitis can occur within 24 hours of starting parenteral therapy,407 but the incidence of this is difficult to assess because the drug is usually used during cardiac arrests and the majority of patients do not survive. Asymptomatic liver enzyme elevations occur in about 25% of patients on oral therapy, usually detected 10 months after exposure, with mean ALT (104 IU/mL) greater than mean AST (aspartate aminotransferase) (89 IU/ml), and generally normal AP and bilirubins. Although adaptation may occur, with normalization of values on continued use, the drug is often stopped because of toxicities to other organs and death from heart disease.405 Between 1 and 3% of patients develop symptomatic hepatitis, with hepatomegaly that resolves relatively quickly on drug withdrawal. The drug and its metabolite remain in liver and plasma for long periods and can cause persistent abnormalities for many months after cessation of therapy.405 The most ominous form of liver injury with amiodarone is the development of cirrhosis that has been termed pseudo-alcoholic based on the findings of Mallory’s hyaline, PMNs, and steatosis,408 and this can occur even with low doses of the drug.409 Regular monitoring of serum ALT is recommended, especially if doses greater than 400 mg/day are used. Decreasing the dose or stopping if ALTs are more than three times the ULN, and performing a liver biopsy if elevations persist, are also recommended. A recent prospective study of serum amiodarone levels in 125 patients suggested that only 6% will have serum ALTs more than three times the ULN if amiodarone levels are <2.5 mg/l, and there should be no ALT elevations if amiodarone levels are <1.5 mg/l.410
529
Section IV. Toxin Mediated Liver Injury
Amiodarone is metabolized by CYP3A4 and CYP2C8411 to Ndesethylamiodarone. The reason why drug and metabolite accumulate in the liver is related to their being amphiphilic with trapping in lysosomes and subsequent inhibition of phospholipases.187 Even in patients without liver injury, intralysosomal inclusions are found with a characteristic lamellar structure on electron microscopy that is the hallmark of phospholipidosis.412 Because of amiodarone’s iodine content, hepatic accumulation can be demonstrated on noncontrast CT scans.413 The drug can also accumulate in mitochondria and cause lipid peroxidation and steatohepatitis in animal models.414 However, whether this is how amiodarone (or a metabolite) causes hepatic injury is still unclear. The drug is also an inhibitor of the organic anion transporter oatp2 in rats.415 Quinidine remains an important antiarrhythmic despite many case reports from the 1970s describing idiosyncratic hypersensitivity hepatotoxic reactions that occurred within a month of therapy,187 often with granulomas on histology.416 No recent case reports have appeared, and the incidence of hepatotoxicity is not really known. Another important effect of quinidine recently recognized is its ability to heteroactivate CYP3A4 activity for other substrates.417,418 Therefore, complex drug interactions should be expected when using quinidine. Although only mild abnormalities of liver tests are listed in the Epocrates database for procainamide, the drug has been reported to cause intrahepatic cholestasis in a number of cases,187 often with granulomas. The drug is much more likely to cause a systemic lupuslike reaction than hepatoxicity. Propafenone is not listed to have any hepatotoxicity, but seven cases of cholestatic jaundice have been reported in the literature since 1980.419
b-Blockers All of these widely used agents have a very low incidence of hepatotoxicity, with the possible exception of labetalol.420,421 Labetalol causes mild asymptomatic ALT increases in 8% of patients, usually within the first few weeks of therapy, which usually normalize with continued therapy (adaptation?), but which worsen in 2%, requiring drug withdrawal. Three fatalities have been reported,420 and the RR-stereoisomer of labetalol, dilevalol, was withdrawn in postmarketing surveillance outside the US in 1990 because of hepatotoxicity (see Table 26-1). The mechanism of injury appears to be idiosyncratic without immune features. Only one case report (of an acute hepatitis) was found for DILI ascribed to metoprolol.422 More recently a case of severe cholestasis was reported for carvedilol,422a which recurred 1 year later when the patient was started on metoprolol. Considering the wide use of these agents, it appears that hepatotoxicity is very rare.
Ca2+ Channel Blockers These agents appear to have a very low incidence of hepatotoxicity, and only verapamil and diltiazem are listed in the Epocrates database as having any at all. Still, nifedipine has been reported to cause acute hepatitis with immune features,423 with the last report published in 1992.424 The cases of acute hepatic injury described after diltiazem425 were both in patients who had been on nifedipine prior to their exposure to diltiazem. Diltiazem has been reported to cause
530
granulomatous hepatitis.426 Verapamil has been reported to cause a few cases of both hepatitic and cholestatic injury.427
CHOLESTEROL-LOWERING AGENTS HMG-CoA Reductase Inhibitors (Statins) The introduction of the statins has had a major impact on the therapy of hypercholesterolemia and heart disease, making these agents among the most widely prescribed medications. However, because their use would be chronic, there was early concern about ocular, muscle, and liver toxicities that had been seen with previous inhibitors of cholesterol synthesis.187,428 Ocular toxicity did not occur, but asymptomatic increases in serum ALT more than twice the ULN were found in 1–3% of subjects in early studies, which was 10–30 times more frequent than rhabdomyolysis.428 The increases occurred within 3 months and there was a dose dependence. There were also a number of severe cases of acute hepatitis and cholestatic injury reported.128 Therefore, recommendations were to avoid the use of statins in patients with almost any liver disease. Unfortunately, following these recommendations would deny many with high cholesterols the benefit of the drugs, because these patients often have increased serum aminotransferases as part of the metabolic syndrome. Subsequent postmarketing surveillance studies of the statins have shown that asymptomatic ALT increases occur in 0.2–1.14% of patients,429,430 and that the majority have adaptation and normalization of the ALTs with continued therapy. Studies in specific groups, including children with familial hypercholesterolemia,431 obese patients,432 elderly patients,433 and patients with increases prior to therapy,434 showed that all tolerated statin therapy with a low incidence of side effects. It is still recommended that liver tests and CK levels be monitored at baseline, 3 months, and then every 6 months,435 although it has not been demonstrated that this will identify those patients at risk of, nor reduce the occurrence of, severe liver or muscle toxicity.128 Serum ALTs more than three times the ULN should be monitored every 2–4 weeks, and the drug discontinued if adaptation fails to occur or clinical symptoms develop. The dose should be adjusted as low as possible to control cholesterol. If using a statin that is metabolized by CYP3A4 (simvastatin, lovastatin, and atorvastatin), caution is necessary if prescribing other medications that inhibit this CYP (e.g. ketoconazole, erythromycin etc.) Fluvastatin is metabolized by CYP2C9, and pravastatin and rosuvastatin are not metabolized by the P450 system. There are also cautions in combining statins with gemfibrozil, niacin, amiodarone, and verapamil.435 Cerivistatin was withdrawn from the market in 1991 because of a high frequency of rhabdomyolysis and not because of liver injury. Fenofibrate has been used for decades and is effective for the management of hypertriglyceridemia. It has been reported to rarely cause an autoimmune hepatitis with ductopenia and fibrosis.128 Mild ALT increases occur rarely, and routine monitoring of liver tests is not considered necessary.436 Although the fibrates are potent peroxisomal proliferator agents in rodents, this effect is not seen in humans based on analysis of liver histology.187 The fibrates can also increase the lithogenicity of bile, but this effect does not appear to be clinically significant.436 Niacin has also been shown to cause an acute hepatitis187 that occurs more frequently with the sustained-release form, usually
Chapter 26 DRUG-INDUCED LIVER INJURY
after a relatively short time (2 days to 7 weeks),437 and often immediately after changing from the crystalline formulation that was well tolerated.438,439 A cholestatic picture has also been described.440 Overall, the incidence of hepatotoxicity must be low. There is some indication that toxicity may be dose related,187 but the mechanism of liver injury has never been clarified.
Diuretics Although rare hepatotoxicity is listed in the Epocrates database for hydrochlorthiazide, ethacrynic acid, and spironolactone, no citations could be found. Considering the widespread use of these agents in relatively sick patients, including those with liver disease, it is clear that they have very little hepatotoxicity. The uricosuric diuretic tienilic acid was withdrawn from the market in 1979 because of a large number of cases of acute and chronic hepatitis, most likely the result of an immune-mediated process.441
Other Antihypertensive Agents Hydralazine is not listed in the Epocrates database to cause any form of hepatotoxicity. However, its congener, dihydralazine (no longer available in the US), is a classic drug that causes immune-mediated drug-induced toxicity96 due to mechanism-based inactivation of CYPs 1A2 and 3A4,442 which then creates a neoantigen and the development of antimicrosomal antibodies. Most case reports of hepatotoxicity from dihydralazine443 and hydralazine444 are from the 1980s and show classic centrilobular necrosis. Granulomatous hepatitis due to hydralazine has also been reported.445 Apparently, hydralazine must not inactivate CYPs to the extent of dihydralazine, and the incidence of hepatotoxicity is low. From a historical perspective, a-methyldopa (Aldomet) was one of the first drugs in widespread use that was noted to have hepatotoxicity, but with a low enough incidence that it was not withdrawn from marketing. It was introduced in 1960 and became unavailable in the US only recently, undoubtedly because there now are so many other more effective and less risky antihypertensives. All forms of liver injury, including acute hepatitis, chronic hepatitis, cholestasis, fulminant liver failure, and cirrhosis, have been associated with its use.187 Bosentan is an orally available benzenesulfonamide designed to potently inhibit both endothelin receptors A and B.446 It has been FDA approved to treat pulmonary hypertension447,448 and will probably supplant epoprostanol as the first-choice therapy for these patients. However, bosentan’s major toxicity is liver injury, which occurs in 2–18% of patients, is dose related, and is reversible with drug withdrawal.449 It is interesting that blocking endothelin receptors with the drug was initially shown to protect the livers of experimental animals from warm ischemia,450 and it was beneficial in portal-hypertensive rats.451 When the drug was studied in humans, it became apparent that it was extensively metabolized by CYPs 2C9 and 3A4, and its induction of these same enzymes caused a drop in initial steady-state concentrations for as long as 3–5 days.448 It was also found to have important interactions with 3A4 and 2C9 substrates such as ciclosporin and warfarin.448 Although the mechanism of hepatotoxicity is still not certain, bosentan’s ability to inhibit the canalicular bile salt export pump may cause intracellular accumulation of cytotoxic bile salts.449 Despite its potential for hepato-
toxicity, there is great interest in using the drug for the treatment of portopulmonary hypertension, with several successful cases reported.452–454
NON-STEROIDAL ANTI-INFLAMMATORY DRUGS (NSAIDS) NSAIDs as a class are important causative agents of hepatotoxicity. Although the incidence of hepatotoxicity differs for different agents, the overall incidence of overt dysfunction is low (less than 0.1%).455 However, because of the large number of people who use NSAIDs on a regular basis (it is estimated that there are 20 million such subjects in the United States) the actual number of cases that occur is ultimately substantial.455 There are no large series reported in the US that can be used to glean the actual incidence of hepatotoxicity caused by NSAIDs. It is estimated, from Medicaid billing data, that acute hepatitis, probably related to NSAID use, results in about 2.2 hospitalizations per 100 000 people.456 In a retrospective Canadian study looking at 228 392 adult patients who contributed 645 456 person-years, the age- and sex-matched risk ratio for hospitalization for acute liver injury related to NSAID use was 1.7, or about five episodes per 100 000 patient-years.457 Risk factors for the development of hepatotoxicity from NSAIDs include advanced age, renal insufficiency, multiple drug use, and high doses.458 Different NSAIDs have different propensities to cause hepatotoxicity. Benoxaprofen proved to be an agent that caused liver injury with a high incidence and fatalities, prompting its withdrawal from medical use. Other agents have minimal potential for hepatotoxicity. NSAIDs that are capable of causing hepatotoxicity are listed in Table 26-14. The most common mechanism of injury appears to be idiosyncratic, probably as a result of metabolic abnormalities, and the most common type of injury appears to be hepatocellular. Cross-reactivity between different classes of NSAIDs may occur.459
Aspirin The primary risk factor for aspirin-induced hepatotoxicity appears to be the dose of the drug, and hence the serum salicylate levels.455 Most patients with hepatotoxicity as evidenced by elevated aminotransferase levels are taking 2–6 g of aspirin daily and have serum levels of salicylates exceeding 25 mg/dl, although toxicity has been seen at levels as low as 10 mg/dl.460 Predisposing conditions have also been suggested as being risk factors. These include the connective tissue disorders rheumatoid arthritis, SLE, and juvenile rheumatoid arthritis. However, the increased incidence in this subgroup is probably explained largely by the higher doses used in treating these conditions, rather than an intrinsic susceptibility, although the cytokine milieu in systemic inflammatory diseases may predispose to hepatotoxicity.461 The same explanations are likely for the increased incidence seen in cases of rheumatic fever. In patients who develop Reye’s syndrome aspirin intake appears to be one of the – probably the most common – triggers for the development of the characteristic features, namely a microvesicular hepatic steatosis and acute encephalopathy. This occurs in the setting of a febrile illness in children, most commonly induced by a viral infection. The underlying predisposing condition is as yet unclear, but may involve congenital mitochondrial enzyme defects or deficiencies, the effect of which is exacerbated by the use of aspirin.455,462 In experimental
531
Section IV. Toxin Mediated Liver Injury
Table 26-14. Non-steroidal Anti-inflammatory Drugs (NSAIDS) and DILI Class/agent
Type of injury
Proposed mechanism
Salicylates Aspirin Diflunisal Benorilate Salicylates
Hepatocellular Cholestatic Hepatocellular Hepatocellular
Toxic Idiosyncratic-metabolic Toxic Toxic
Acetic acid derivatives Amfenac Hepatocellular Clometacin Hepatocellular Diciofenac Hepatocellular Etodolac Hepatocellular Fenclofenac Hepatocellular Fenclofenamic acid Hepatocellular Fenclozic acid Cholestatic Fentiazac Hepatocellular Indomethacin Hepatocellular Isoxepac Hepatocellular Nabumetone Cholestatic Sulindac Cholestatic Tolmetin Hepatocellular
Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic Idiosyncratic-metabolic Idiosyncratic Idiosyncratic-immune Idiosyncratic
Propionic acid derivatives Benoxaprofen Cholestatic Carpofen Hepatocellular Fenbufen Hepatocellular Fenoprofen Hepatocellular/cholestatic Flurbiprofen Hepatocellular Ibufenac Hepatocellular Ibuprofen Hepatocellular Ketoproprofen Hepatocellular Naprosyn Hepatocellular/cholestatic Oxaprozm Hepatocellular Pirprofen Hepatocellular
Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-immune Idiosyncratic-immune Idiosyncratic-immune Idiosyncratic-metabolic
Fenamates Cinchophen Glafenine Meclofenamic acid Mefenamic acid Niflumic acid Tolfenamic acid
Hepatocellular Hepatocellular Hepatocellular Hepatocellular Hepatocellular Hepatocellular
Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-metabolic Idiosyncratic-immune Idiosyncratic-metabolic
Oxicams Droxicam lsoxicam Piroxicam Sudoxicam
Hepatocellular/cholestatic Cholestatic Hepatocellular/cholestatic Hepatocellular
Idiosyncratic Idiosyncratic-metabolic Idiosyncratic Idiosyncratic-metabolic
animals, salicylic acid inhibits mitochondrial b-oxidation of longchain fatty acids,463 and up to one-third of children who develop Reye’s syndrome have inborn errors of metabolism in this very pathway.464 The incidence of Reye’s syndrome is decreasing, mirroring the decline in use of aspirin for childhood viral illnesses.455 In this dose-dependent type of hepatotoxicity the mechanism is probably related to an intrinsic ability to injure the hepatocyte. Based on the ultrastructural histology, the site of injury appears to be the mitochondrion. Other mechanisms that have been postulated include lipid peroxidation, hydroxyl radical scavenging, and injury to the hepatocyte membrane.455 Apart from the features of the disease condition necessitating the use of aspirin, findings are minimal. Tender hepatomegaly may occur.
532
Liver injury is most often recognized by finding elevated serum AST and ALT levels, and, less commonly, ammonia and bilirubin levels. Up to 50% of individuals with serum levels of salicylate greater than 15 mg/dl have elevated AST and ALT.460 Acute liver failure, characterized by coagulation abnormalities and hepatic encephalopathy, is rare.465 The classic histologic description of liver injury from aspirin is a non-specific focal hepatitis. Ballooning degeneration that is more prominent in zone 3 is a typical finding. Hepatocyte necrosis is also seen, and inflammatory cell infiltration is minimal. Steatosis is unusual in the hepatotoxicity associated with high doses of aspirin. However, in Reye’s syndrome microvesicular steatosis is the hallmark. Aspirin hepatotoxicity is rapidly reversible when the drug is discontinued. Fatalities are very rare but have been reported.466 There is no conclusive evidence that aspirin can cause chronic hepatitis. Aspirin overdose is managed by discontinuation of the drug, with supportive care in the rare individual who has severe hepatotoxicity. If aspirin is absolutely essential in the individual’s management, an attempt may be made to restart the drug at a lower dose after the liver tests have returned to normal. Close monitoring of the liver tests in this rechallenge is necessary.
Salicylates other than Aspirin – Diflunisal (Dolobid) This is a difluorophenyl derivative of salicylic acid that has been reported to cause a cholestatic and mixed hepatocellular type injury.467,468
Acetic Acid Derivatives – Indomethacin (Indocin) This indole-containing drug is probably the most-used of the NSAIDS derived from acetic acid. There are relatively few reports of indomethacin-related hepatic injury compared to other organ toxicities caused by this drug.455,469 In one series, although indomethacin accounted for relatively fewer instances of hepatotoxicity (compared with other NSAIDS), the incidence of fatalities was higher.469 Case fatalities have been reported.469–471 Children may be more susceptible to severe injury, and so the drug is not recommended for use in the pediatric age group.455 Based on the few case reports available, the mechanism of toxicity appears to be metabolic idiosyncrasy. Features are usually nonspecific, with laboratory values suggesting hepatocellular injury, and much less often a mixture with cholestasis. Massive hepatocellular necrosis, primarily centrally located,470 is typical. Microvesicular steatosis and cholestasis may occur. Discontinuation of the drug and supportive measures should be instituted. A good outcome is expected with early detection and withdrawal. However, case fatalities have been reported.
Sulindac (Clinoril) Sulindac is also an indole derivative of acetic acid and therefore has some structural similarities to indomethacin. There are many reported cases of hepatotoxicity related to this drug, which is a potent analgesic and has relatively fewer gastrointestinal side effects than other NSAIDs. However, it is still considered one of the most likely NSAIDs to produce hepatic injury.455 In an analysis of 91 cases
Chapter 26 DRUG-INDUCED LIVER INJURY
reported to the FDA the ratio of females to males was 3:5.472 Based on the reported cases, the mechanism for most appeared to be a generalized hypersensitivity reaction (immune mediated), which included liver involvement. Metabolic idiosyncrasy may account for a minor subset.472 Patients present with features of a hypersensitivity reaction, including fever, skin rash, pruritus, and tender hepatomegaly. Stevens–Johnson syndrome may occur. The onset is usually within 4 weeks of starting the drug.473–475 Jaundice may occur in about two-thirds of cases.472 Laboratory tests often reveal significant hepatocellular damage. Eosinophilia tends to be more common when the pattern of injury is cholestatic than when it is primarily hepatocellular.472 Pancreatitis may occur in some cases.476 Cholestasis is prominent in most cases, with only about 25% showing hepatocellular injury.472 Five per cent (4 cases of 91) of patients with sulindac-associated jaundice died.472 Although the cause of death in most cases was attributable to systemic hypersensitivity, death from liver failure secondary to massive hepatocellular necrosis can occur.472 Rechallenge with the drug may result in the reappearance of the hypersensitivity reaction after only a few doses.474
Table 26-15. Antineoplastic and Immunosuppressive Agents and DILI Manifestation of hepatotoxicity
Agent
Sinusoidal obstruction syndrome
Mitomycin 6-Thioguanine Azathioprine Cytardbine Dacarbazine Indicine-N-oxide Daunorubicin Combination chemotherapy Radiation therapy plus Cyclophosphamide Busulfan Carmustine Mitomycin C Other regimens Common Mithramycin L-Asparaginase Streptozocin Methotrexate (high dose) Rare Nitrosoureas 6-Thioguanines Cytarabine Adriamycin 5-Fluorouracil Cyclophosphamide Etoposide Vinca alkaloids L-asparaginase Actinomycin-D Mitomycin C Bleomycin Methotrexate 6-Mercaptopurine Azathioprine Busulfan Amsacrine Methotrexate Azathioprine Floxuridine Androgens Hydroxyprogesterone Azathioprine Hydroxyurea Tamoxifen Azathioprine 6-Thioguanine Androgens Estrogens Estrogens Androgens Methotrexate
Hepatocellular necrosis
Diclofenac (Arthrotec, Voltaren) Diclofenac is a phenylacetic acid derivative that has been in use for some time. Although the most common manifestation of hepatotoxicity is asymptomatic elevations in the liver tests, there are numerous reports in the literature of significant hepatotoxicity and even fatalities attributable to the use of this drug.477,478 It is estimated that up to five of 100 000 individuals have significant hepatotoxicity. The onset is from 3 weeks to 12 months after starting the drug.477,478 Elderly women with osteoarthritis seem to be more susceptible.477 Data from the reports suggest that in most cases the cause is immunologic idiosyncrasy. However, metabolic idiosyncrasy has seemed to be the more logical explanation in other cases. Symptoms are non-specific in most cases, with nausea, vomiting, abdominal discomfort, and jaundice being hallmarks of more severe hepatitis. Rash and fever occur in a minority of cases. The liver test abnormalities favor hepatocellular damage. In rare cases the ANA titers may be elevated and care should be taken to rule out autoimmune chronic hepatitis.478,479 Zone 3 or spotty acute hepatocellular necrosis is the most common histologic finding. Other features may include granulomas, cholestasis, hepatic eosinophilia, and chronic hepatitis. Overdose is managed by withdrawal of the agent and supportive care. With early withdrawal the prognosis is good, even with severe hepatitis. Rarely, the use of diclofenac has been thought to trigger development of ongoing autoimmune hepatitis.479
Hepatic steatosis
Cholestasis
Fibrosis Sclerosing cholangitis Peliosis hepatis
Nodular regenerative hyperplasia
Hepatic neoplasms
ANTINEOPLASTIC AND IMMUNOSUPPRESSIVE AGENTS These agents produce a large number of hepatic abnormalities (Table 26-15). Causality assessment of hepatotoxicity in the setting of cancer chemotherapy is often difficult. Among the reasons for this are: 1. Abnormal liver tests may result from metastasis or infiltration of the liver parenchyma or biliary tree by tumor. A Budd–Chiarilike picture may resemble sinusoidal obstruction syndrome and
may occur as a result of the procoagulant state caused by many tumors. 2. Immunosuppression may result in sepsis and shock, with its attendant cytokine-induced effects on the liver, such as cholestasis. Occasionally the liver itself may be opportunistically infected, or transfusion may result in viral hepatitis. 3. Multiple drugs are often used in overlapping schedules, making it difficult to assign causality of DILI to a single drug.
533
Section IV. Toxin Mediated Liver Injury
Table 26-16. Roenigk Histopathological Classification of Methotrexate Hepatotoxicity
Grade I Grade II Grade IIIa Grade IIIb Grade IV
Fatty infiltration
Nuclear variability
Portal inflammation and necrosis
Fibrosis
Mild or none Moderate to severe May or may not be present May or may not be present May or may not be present
Mild or none Moderate to severe May or may not be present May or may not be present May or may not be present
Mild or none Moderate to severe May or may not be present May or may not be present May or may not be present
None None Mild Moderate to severe Cirrhosis
4. Other modalities of treatment (i.e. non-chemotherapy) may also lead to hepatotoxicity. Examples include the direct effects of radiation and graft-versus-host disease in patients undergoing bone marrow or stem cell transplants. 5. Drugs that have minimal hepatotoxic potential when the drug is used alone may produce severe liver disease when it is used in combination with other chemotherapeutic agents or with radiation therapy. 6. Liver biopsy, which might help in differential diagnosis, is often contraindicated because of thrombocytopenia and coagulation abnormalities caused by treatment. 7. Toxicity in other organ systems may result in abnormal liver tests (e.g. adriamycin-induced cardiac failure may result in hepatic congestion and its resultant liver test abnormalities).
ANTIMETABOLITES Methotrexate (MTX) Methotrexate, a derivative of aminopterin, is a folate analog that inhibits dihydrofolate reductase, which causes the arrest of rapidly dividing cells in the S-phase of the cell cycle. This property has been used in the treatment of leukemias and, more recently, as a diseasemodifying agent in several chronic inflammatory conditions, including psoriasis, rheumatoid arthritis, and chronic idiopathic inflammatory bowel disease. Hepatotoxicity has been recognized as a potential major adverse reaction that can occur with long-term use. Case reports describing cirrhosis as a result of MTX use first appeared in the 1960s.480 The pathogenesis of MTX hepatotoxicity is poorly understood. It has been hypothesized that the drug can activate hepatic stellate (Ito) cells, which leads to increased collagen deposition. Others have speculated that the drug itself, and its metabolites (polyglutamates), may accumulate, leading to prolonged folate inhibition with resultant impairment of nucleotide and methionine synthesis that in turn lead to hepatocyte injury.480 Patients with pre-existing liver disease seem to be more susceptible to toxicity.481 Factors associated with increased risk of MTX toxicity include heavy alcohol use, pre-existing liver disease, daily dosing, duration of therapy more than 2 years, cumulative dose >1500 mg, and obesity with diabetes mellitus.480 Acute symptoms are rare. With advanced toxicity and cirrhosis, clinical features will reflect these changes and are therefore non-specific. Minor elevations in liver tests may occur in many who take methotrexate (20–50%), but this does not necessarily imply significant toxicity.482 Conversely, liver tests may be normal in the setting of even severe fibrosis. With advanced disease, the laboratory findings reflect those associated with cirrhosis and its complications. In 1982 the Psoriasis Task Force (Roenigk and colleagues) devised a classification scheme for the liver biopsy findings in methotrexate
534
hepatotoxicity (Table 26-16). This is probably the most popular classification scheme, despite its subjective nature. Ultrastructural changes precede microscopic changes and include the deposition of fibrous tissue in the space of Disse and an increase in the size and number of hepatic stellate (Ito) cells in the perisinusoidal space.480 Microscopic changes include macrovesicular steatosis; nuclear variability; infiltration with chronic inflammatory cells; focal liver cell necrosis; fibrosis in the perivenular, pericellular, and portal regions; and eventually cirrhosis. Many of these findings may also be a result of other underlying conditions, many of which have been identified as being risk factors for hepatotoxicity. There is no antidote for MTX toxicity. Patients with cirrhosis have sometimes required transplantation. Prevention of significant toxicity requires close monitoring of patients who are on the drug long term. In cases where a pre-existing liver disease is strongly suspected, a baseline liver biopsy should be performed. In those with appreciable pre-existing liver disease, liver tests should be followed every 4–8 weeks for the first year and every 3–6 months thereafter for as long as the patient is on MTX. Subjects should avoid alcohol and should be placed on folic acid supplements. Patients with diabetes mellitus should be strictly controlled and obese patients strongly advised to lose weight. Patients with no history of liver disease but who develop abnormal liver tests early after starting MTX (within a few months) should have a liver biopsy before continuing with further treatment. Otherwise, when the patient has received a predetermined cumulative dose (the American College of Rheumatology recommends after 1.5 g initially, and thereafter every 1–1.5 g), liver biopsies should be performed. If the Roenigk classification is used (see Table 26-12), the finding of grade IIIb or grade IV fibrosis is grounds to discontinue the drug.
6-Mercaptopurine (6-MP) and Azathioprine As with MTX, these agents probably are used more as immunosuppressive agents in the treatment of chronic inflammatory disorders and in the post-transplant setting than as antineoplastic agents. 6-MP has been in use for the last 60 years. It is a thiopurine analog of the natural purine bases. Its potential to cause hepatotoxicity is well recognized. Cholestatic liver injury appears to be the most common manifestation of this toxicity and may occur in 6–40% of recipients.483–485 The effect appears to be dose dependent. Doses exceeding 2.5 mg/kg have the highest likelihood of toxicity.484,486,487 The latent period between commencement of the drug and the onset of toxicity is anywhere from 1 to 18 months. Adults appear to be more susceptible to injury than children.488 6-MP is metabolized extensively in the liver, and this probably relates to its hepatotoxic potential. The mechanism of toxicity appears to be intrinsic. Evidence in support of this conclusion includes the paucity
Chapter 26 DRUG-INDUCED LIVER INJURY
of evidence that would suggest a hypersensitivity mechanism (i.e. the long lag time before the onset of symptoms, and the lack of hypersensitivity features such as rash and eosinophilia). Other supporting facts include the relatively high incidence and dose dependence. Jaundice and pruritus are the main presenting features. The laboratory studies reflect a mixed hepatocellular and cholestatic injury with moderate elevations in serum AST, ALT, ALP, and serum bilirubin.486,487 Hepatic histopathology shows a mixed picture, with features of both cholestasis and hepatocellular necrosis. Management consists of discontinuing the agent. Cases of fatal hepatic necrosis, in the setting of continued use despite evidence of toxicity, have been described. Rechallenge has led to recurrent hepatotoxicity in some cases.487 Azathioprine is a prodrug of 6-MP and appears to be less hepatotoxic than its parent compound. This notwithstanding, it is capable of causing DILI with a spectrum of toxicity that is wider than that of 6-MP. In addition to the cholestatic injury seen with 6-MP,489,490 other patterns have been recognized. Predominant cholestasis, with evidence of a hypersensitivity reaction, has been reported,491 as has a primarily hepatocellular type of injury, especially in post-renal transplant patients.492 More recently, several other lesions with a common pathogenesis, in that they involve an insult to the vascular endothelium, have been appreciated. These conditions include striking sinusoidal dilation, peliosis hepatis,493 nodular regenerative hyperplasia,494,495 hepatoportal sclerosis,494 and VOD.496–498 These observations were all made in the post-renal transplant setting and, in one report about such patients, the incidence of VOD was estimated to be 2.5%.498 The onset of VOD occurs from 2 months to as long as 9 months after transplantation. There is a male preponderance. In one series co-infection with a hepatotrophic virus was suggested as a probable contributing factor in the pathogenesis of VOD.498 Clinically, signs of portal hypertension with minimal elevations of the liver tests in a non-specific pattern are noted. Portal hypertension may progress, which may have an effect on future morbidity and mortality.498 The hepatotoxicity appears to be primarily an idiosyncratic reaction, although azathioprine is converted to 6-MP in vivo and direct toxicity may also play a role. As alluded to previously, patients may display features of a hypersensitivity reaction in some instances of toxicity due to azathioprine. Histology is usually classic for VOD or the other pathologies described.
6-Thioguanine 6-Thioguanine is also a purine analog, used primarily in the treatment of acute and chronic leukemia. As with azathioprine, this agent also appears to result in endothelial dysfunction leading to manifestations of VOD, nodular regenerative hyperplasia, and hepatoportal sclerosis.484,499,500 In one study the incidence of portal hypertension in patients with chronic myeloid leukemia who were treated with busulfan alone or busulfan with 6-thioguanine was determined. In the latter group, 18 of 675 patients, compared to none in the busulfan-only group, developed portal hypertension. Histologically, idiopathic portal hypertension with minimal morphologic abnormalities or nodular regenerative hyperplasia was the major finding; three patients developed cirrhosis and its attendant complications.499 Other studies have described VOD in patients treated with 6thioguanine and cytosine arabinoside.501,502
5-Fluorouracil and Floxuridine 5-Fluorouracil (5-FU) is used in the treatment of malignancies of the digestive system, breast, and ovary. It is a pyrimidine-based analog that is metabolized by the liver with little hepatotoxicity when used orally. Floxuridine, a derivative of 5-FU, is administered by continuous intravenous infusion or directly into the hepatic artery for the treatment of hepatic metastasis from colon cancer.503,504 This leads to higher remission rates and improved survival, but at the cost of increased hepatic injury. Damage appears to be more common with direct hepatic artery infusions,484 which cause a chemical hepatitis in more than half of the patients treated.505 Liver tenderness and elevation in the serum AST, ALT, ALP, and bilirubin characterize the reaction. In a smaller subset of patients sclerosing cholangitis may develop.505–508 This is usually heralded by the onset of jaundice and marked elevations in serum ALP. In one study of intraarterial infusion 35 of 35 patients developed a predominantly cholestatic pattern of liver tests. Seven patients receiving intra-arterial therapy were studied with cholangiography, which in all cases demonstrated sclerosis of the intrahepatic or extrahepatic bile ducts. In addition, liver biopsies showed cholestasis and pericholangitis, with minimal hepatocytic damage. It was suggested that biliary sclerosis is probably more common than the often-described chemical hepatitis.505 Chemical hepatitis usually resolves after therapy is complete or discontinued. Fatal cirrhosis has been reported to result from the more serious sclerosing cholangitis.509 Cases with sclerosing cholangitis are managed with endoscopic retrograde cholangiopancreatography (ERCP) and stenting versus surgical therapy if complications develop or are impending. The biliary tree is highly dependent on the hepatic arterial supply for oxygenation and delivery of nutrients, and it is thought that damage or dysfunction of the arteries caused by the chemotherapeutic agent leads to the biliary sclerosis.
Cytosine Arabinoside Cytosine arabinoside is also a pyrimidine analog. Hepatotoxicity appears to be dose related and ranges from mild increases in the serum AST, ALT, and ALP to more significant elevations with frank jaundice.510–513 These changes are usually reversible. L-Asparaginase This is an enzyme that catalyses the hydrolysis of L-asparagine to aspartic acid and ammonia. Because leukemic cells cannot produce L-asparagine, whereas normal cells can, L-asparaginase is used to treat acute lymphocytic leukemia and T-cell lymphoblastic lymphoma. Abnormal liver tests have been reported in up to 75% of recipients.514 Hypersensitivity-type reactions, especially after repeated doses, are common and have been reported in 43% of recipients, although anaphylactic-like reactions occur in only about 10%.515,516 Steatosis, a finding more typical of a metabolic aberration, is common, occurring in 50–90% of recipients.517 This is probably a result of impaired mitochondrial protein synthesis. Given the frequency of its occurrence, DILI due to L-asparaginase is likely to be a direct toxic effect of the drug itself (rather than metabolic idiosyncrasy). The clinical features of reactions to L-asparaginase usually develop within 1 hour after administration and include pruritus, dyspnea, urticaria, swelling at the injection site, angioedema,
535
Section IV. Toxin Mediated Liver Injury
rash, abdominal pain, laryngospasm, nasal stuffiness, bronchospasm, and hypotension.515 The liver test abnormalities include modestly elevated serum AST, ALT, bilirubin, and ALP. The serum albumin and several other proteins that are synthesized by the liver are decreased. These proteins include factors I, II, VII, IX, and X; ceruloplasmin; haptoglobin; transferrin; and lipoproteins.514 Coagulopathy may be a prominent feature. Elevated ammonia levels may occur (in keeping with the mechanism of action). The most prominent histologic finding is microvesicular steatosis.517 In cases of significant toxicity it is necessary to stop the drug. However, abnormal liver tests are common, and it is often difficult to differentiate hepatotoxicity from other toxic effects of the drug. Fatal outcomes can occur. A less immunogenic form of the drug (pegaspargase) has been developed and is reportedly less likely to result in hypersensitivity reactions.515
Mithramycin (Plicamycin) This is an antibiotic that can intercalate with DNA and thus inhibit RNA synthesis. In addition to its use as an anticancer agent, it is sometimes used in the treatment of hypercalcemia and Paget’s disease. Abnormal liver tests occur in almost all patients treated.518,519 Serum aminotransferases may be quite elevated, and the levels correlate with dose. Depression of coagulation factor production and thrombocytopenia may result in a bleeding diathesis. Hepatocellular necrosis (zone 3) and steatosis have been observed in liver biopsies.519 The lower doses used to treat hypercalcemia and Paget’s disease are reportedly associated with less frequent hepatotoxicity.493 All of the above features imply that mithramycin is an intrinsic hepatotoxin.
Adriamycin (Doxorubicin) Adriamycin is also an antibiotic. It has rarely been implicated as the cause of hepatic injury. In six cases of acute lymphoblastic leukemia it was thought to have caused acute or chronic hepatitis.520 It has also been postulated that adriamycin potentiates the hepatotoxicity of 6-MP. It may increase the incidence of radiation-induced injury when used before radiation therapy.521 Adriamycin has a high propensity to produce cardiomyopathy that can result in congestive heart failure. The resultant liver congestion can sometimes be misleading, but reverses with appropriate treatment of the heart failure.
Dactinomycin (Actinomycin D) This antibiotic has been used for many years without much evidence of hepatotoxicity. A few cases of severe hepatic injury have been described when the agent is used alone or with vincristine.522–524 Cases of VOD have also been described, especially in the setting of concomitant irradiation for treatment of Wilms’ tumor.525–528
Ciclosporin A This peptide is extracted from Tolypocladium inflatum. It is used primarily as an immune suppressant in transplant medicine and has a narrow therapeutic window. It has been reported to cause a mild, dose-dependent cholestatic injury,529,530 with a highly variable frequency.530 In many patients ciclosporin A toxicity is subclinical.529,530 The liver tests reveal a mild, often transient increase in AP levels, occasionally accompanied by slight elevations in serum bilirubin and
536
aminotransferase levels. Ciclosporin levels tend to be on the high side of the therapeutic range.529 Dose reduction is recommended. In the settings in which this drug is used there are often multiple other potential causes of liver dysfunction (e.g. infection, organ rejection); confounding conditions should therefore be ruled out, especially graft rejection. Liver biopsy may be helpful. It appears that hepatotoxicity is of little prognostic significance. Nephrotoxicity and neurotoxicity are more important from this standpoint.
Vinca Alkaloids These alkaloids are derived from the periwinkle plant. Their antitumor effects are dependent on their ability to disrupt cellular microtubule function. Vincristine appears to increase liver toxicity when used with radiation therapy.484 Rarely it may result in a mild increase in serum aminotransferase levels outside the setting of radiation.531 Otherwise, these agents do not appear to be significant hepatotoxins.
Etoposide (VP-16) Also an alkaloid, etoposide is a derivative of podophyllotoxin. The drug disrupts the formation of the mitotic spindle. Acute hepatocellular necrosis has been reported.532 In combination with ifosfamide severe hepatotoxicity has been described.533
ALKYLATING AGENTS Cyclophosphamide This alkylating agent is commonly used in regimens for leukemia, lymphoma, and solid tumors. It is also used in the treatment of a few chronic inflammatory conditions, such as Wegener’s granulomatosis, and systemic lupus erythematosus (SLE). It is metabolized to its active form by CYPs. The alkylating species usually are formed only in cells with high turnover rates. However, hepatotoxicity may result from a metabolic idiosyncrasy in some individuals who form toxic amounts of these species in the hepatocytes. Hepatotoxicity appears to be a rare complication of therapy. There are case reports of hepatocellular necrosis that were possibly related to the use of cyclophosphamide.534 In patients with collagen vascular diseases cyclophosphamide rarely has caused hepatic injury, including mild hepatitis to massive hepatocellular necrosis.535 A convincing case of toxicity with resolution after withdrawal and recurrence on rechallenge was seen in a patient with SLE. There was jaundice and marked elevation in the serum ALT.536 There are increasing reports of VOD in patients undergoing bone marrow transplant who receive a conditioning regimen containing cyclophosphamide and busulfan.537–540 There is a report of VOD in the non-transplant setting related to the use of cyclophosphamide.541
Busulfan Busulfan appears to be a contributing factor in the causation of VOD, as described previously.
Ifosfamide A cholestatic injury has been reported when this agent is used in combination with etoposide (VP-16).533
Chapter 26 DRUG-INDUCED LIVER INJURY
Chlorambucil This drug rarely causes hepatotoxicity, but there is a recent report of an acute cholestatic hepatitis with its use. Older, sparse reports focus more on hepatocellular injury.542
Nitrosoureas (Carmustine [BCNU], Lomustine, Semustine, Streptozocin) These compounds can all cause what appears to be reversible hepatic dysfunction, with jaundice and an elevated AST in up to 25% of cases. Higher doses have been noted to increase the AST in up to 40% of patients.543,544 With high doses of BCNU, fatal hepatic necrosis can occur. Pericholangitis and intrahepatic cholestasis accompany mild hepatic necrosis in most cases. Recent animal studies provide evidence that lipid peroxidation and alterations in the antioxidant system may significantly contribute to BCNU-induced hepatotoxicity, and some antioxidant agents may be of benefit in reducing the incidence of cholestasis.545
Dacarbazine This agent is used primarily to treat malignant melanoma and some lymphomas. Recent case reports implicate this drug in the causation of acute hepatocellular necrosis secondary to VOD.546–548 This seems to occur within a few days of the second dose, and eosinophilia may be a feature, raising the possibility of an immunologically mediated process. Massive elevations in AST and ALT occur, and histology is consistent with VOD. There tends to be minimal inflammatory infiltration.546 Management is supportive. It has been recommended that if eosinophilia develops after the first dose of dacarbazine, subsequent doses should be avoided.546
BIOLOGIC RESPONSE MODULATORS Interferons These agents are used in the treatment of chronic viral hepatitis (C and B), some solid tumors (e.g. Kaposi’s sarcomas in patients with HIV disease), melanoma, and certain leukemias. Hepatotoxicity is extremely rare with the low percutaneous doses used to treat hepatitis. However, a few cases that probably represent induction of autoimmune hepatitis by interferon a-induced enhancement of the immune system have been described.549 In addition, a small subset of patients being treated for chronic hepatitis C often have mild elevations in AST and ALT, despite a good virologic response. These correct when the interferon is withdrawn, suggesting that interferon-a has a role to play in their cause. The incidence of liver enzyme abnormalities seems somewhat more common with the pegylated than with the standard interferons.550 Elevations of serum aminotransferases are more frequent with administration of the higher doses of interferon-a used in therapy of malignancies.551,552 Rare cases of jaundice and hepatic failure have been reported with interferon-a2b.552
Tumor Necrosis Factor-a (TNF-a) This is a biologic agent that is produced in response to several types of injury, such as alcoholic liver disease and chronic inflammatory bowel disease. It has been implicated in the pathogenesis of cholestasis,553 and therefore it is not surprising that it has been found
to cause profound cholestasis when it is used as treatment in advanced colorectal cancer.554
Anti-TNF-a Drugs Drugs of this class, including infliximab and etanercept, have been found to reactivate latent infections, such as chronic hepatitis B or tuberculosis, and to lead to enhanced rates of replication of the hepatitis C virus. Infliximab use has been associated with severe hepatotoxicity, leading to death or the need for liver transplant. The mechanism of injury is uncertain, but an immunoallergic reaction may be involved.
Interleukin-2 Interleukin-2 (IL-2) immunotherapy is associated with the development of profound reversible cholestasis and hyperbilirubinemia in a large proportion of patients (up to 85%) who receive it.555–558 There is evidence to suggest that this reversible cholestasis is a direct result of IL-2-dependent reduced excretion of bile.555 Clinical features include jaundice, right upper quadrant pain and tenderness, nausea, pruritus, and hepatomegaly. Surprisingly, the administration of total parenteral nutrition has been noted to reduce the incidence of this phenomenon.556
SUMMARY AND CONCLUSIONS DILI is underdiagnosed and underappreciated as a cause or contributor to liver injury. Drugs and toxins should be considered in virtually all types of liver injury occurring in subjects of all ages, although the risks are higher in older subjects (and probably increase progressively with age [it is not clear whether this is due to increased intrinsic risk or to the fact that older people take more drugs and therefore have more opportunities to experience ADRs]). A goal that now appears attainable within the next generation is to define the environmental and host factors that underlie the development of idiosyncratic DILI. However, this will require the establishment of a national registry of subjects with bona fide, well characterized DILI and the discovery of the genetic polymorphisms and other factors that set them apart from those with similar demographics and drug exposure who do not develop DILI. Such an effort has begun in the USA, thanks to funding provided by the US National Institutes of Health (NIDDK). There is a National DILI Network, comprising five clinical centers and regional consortia, based at the University of Connecticut Health Center, the University of North Carolina, Indiana University, the University of Michigan, and the University of California, San Francisco. More information is at http://dilin.dcri.duke.edu/. Healthcare providers are encouraged to contact one of these centers for advice or to refer subjects to the National DILIN Registry of patients. It is only through careful phenotype–genotype correlation in DILI subjects versus suitable controls that we will realize the promise implicit in the sequencing of the human genome and the analytical advances that have made metabolomics and metabonomics emerging fields of science.
Acknowledgments Supported by the following grants and contracts from the US PHS (NIH): DK38825, DK092326, DK065193, and RR06192. The
537
Section IV. Toxin Mediated Liver Injury
opinions expressed herein are those of the authors. They do not necessarily reflect the official views of the United States Public Health Service or the Universities of Connecticut, Iowa, or Kentucky, nor of Emory University. We thank Jean Clark for much help in preparing the manuscript.
REFERENCES 1. Russo MW, Galanko JA, Shrestha R, Fried MW, Watkins P. Liver transplantation for acute liver failure from drug induced liver injury in the United States. Liver Transpl 2004;10:1018–1023. 2. Lazarou J, Pomeranz BH, Corey PN. Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies. JAMA 1998;279:1200–1205. 3. Senior JR. Regulatory Perspectives. In: Kaplowitz N, Deleve LD, eds: Drug induced liver disease. New York: Marcel Dekker, 2003; 739–754. 4. Anonymous. US Government Printing Office, Washington DC Office of Drug Safety Annual Report FY 2002. 4-9-2004. 5. Lasser KE, Allen PD, Woolhandler SJ, et al. Timing of new black box warnings and withdrawals for prescription medications. JAMA 2002;287:2215–2220. 6. Hutchinson TA, Lane DA. Assessing methods for causality assessment of suspected adverse drug reactions. J Clin Epidemiol 1989;42:5–16. 7. Lucena MI, Camargo R, Andrade RJ, Perez-Sanchez CJ, Sanchez De La CF. Comparison of two clinical scales for causality assessment in hepatotoxicity. Hepatology 2001;33:123–130. 8. Benichou C. Criteria of drug-induced liver disorders. Report of an international consensus meeting. J Hepatol 1990;11:272–276. 9. Benichou C, Danan G, Flahault A. Causality assessment of adverse reactions to drugs – II. An original model for validation of drug causality assessment methods: case reports with positive rechallenge. J Clin Epidemiol 1993;46:1331–1336. 10. Maria VA, Victorino RM. Development and validation of a clinical scale for the diagnosis of drug-induced hepatitis. Hepatology 1997;26:664–669. 11. Aithal GP, Rawlins MD, Day CP. Clinical diagnostic scale: a useful tool in the evaluation of suspected hepatotoxic adverse drug reactions. J Hepatol 2000;33:949–952. 12. Kerr JF. Shrinkage necrosis: a distinct mode of cellular death. J Pathol 1971;105:13–20. 13. Kerr JF, Wyllie AH, Currie AR. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972;26:239–257. 14. Arends MJ, Wyllie AH. Apoptosis: mechanisms and roles in pathology. Int Rev Exp Pathol 1991;32:223–254. 15. Samali A, Zhivotovsky B, Jones D, Nagata S, Orrenius S. Apoptosis: cell death defined by caspase activation. Cell Death Differ 1999;6:495–496. 16. Ashkenazi A, Dixit VM. Death receptors: signaling and modulation. Science 1998;281:1305–1308. 17. Li P, Nijhawan D, Budihardjo I, et al. Cytochrome c and dATPdependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 1997;91:479–489. 18. Nakagawa T, Zhu H, Morishima N, et al. Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-beta. Nature 2000;403:98–103. 19. Brodie BB, Reid WD, Cho AK, et al. Possible mechanism of liver necrosis caused by aromatic organic compounds. Proc Natl Acad Sci USA 1971;68:160–164. 20. Jollow DJ, Mitchell JR, Zampaglione N, Gillette JR. Bromobenzene-induced liver necrosis. Protective role of
538
21. 22.
23. 24.
25.
26.
27.
28.
29.
30.
31. 32.
33. 34.
35. 36.
37.
38.
39. 40.
41.
42. 43.
glutathione and evidence for 3,4-bromobenzene oxide as the hepatotoxic metabolite. Pharmacology 1974;11:151–169. Garner RC, Martin CN. Fungal toxins, aflatoxins and nucleic acids. Boca Raton, FL: CRC Press; 1979. Douglas KT. Reactivity of glutathione in model systems for glutathione S-transferases and related enzymes. London: Academic Press, 1988. Monks TJ, Lau SS. Reactive intermediates and their toxicological significance. Toxicology 1988;52:1–53. Anders MW, Lash L, Dekant W, Elfarra AA, Dohn DR. Biosynthesis and biotransformation of glutathione S-conjugates to toxic metabolites. Crit Rev Toxicol 1988;18:311–341. Miller EC, Miller JA. Mechanisms of chemical carcinogenesis: nature of proximate carcinogens and interactions with macromolecules. Pharmacol Rev 1966;18:805–838. Nicotera P, Moore M, Mirabelli F, Bellomo G, Orrenius S. Inhibition of hepatocyte plasma membrane Ca2+-ATPase activity by menadione metabolism and its restoration by thiols. FEBS Lett 1985;181:149–153. Jones DP, Thor H, Smith MT, Jewell SA, Orrenius S. Inhibition of ATP-dependent microsomal Ca2+ sequestration during oxidative stress and its prevention by glutathione. J Biol Chem 1983;258:6390–6393. Thor H, Hartzell P, Svensson SA, et al. On the role of thiol groups in the inhibition of liver microsomal Ca2+ sequestration by toxic agents. Biochem Pharmacol 1985;34.3717–3723. Freeman ML, Borrelli MJ, Meredith MJ, Lepock JR. On the path to the heat shock response: destabilization and formation of partially folded protein intermediates, a consequence of protein thiol modification. Free Radic Biol Med 1999;26:737–745. Ueda S, Nakamura H, Masutani H, et al. Redox regulation of caspase-3(-like) protease activity: regulatory roles of thioredoxin and cytochrome c. J Immunol 1998;161:6689–6695. Moran LK, Gutteridge JM, Quinlan GJ. Thiols in cellular redox signalling and control. Curr Med Chem 2001;8:763–772. Essigmann JM, Croy RG, Nadzan AM, et al. Structural identification of the major DNA adduct formed by aflatoxin B1 in vitro. Proc Natl Acad Sci USA 1977;74:1870–1874. Pohl LR. Drug-induced allergic hepatitis. Semin Liver Dis 1990;10:305–315. Pumford NR, Halmes NC. Protein targets of xenobiotic reactive intermediates. Annu Rev Pharmacol Toxicol 1997;37:91–117. DeLeve LD, Kaplowitz N. Importance and regulation of hepatic glutathione. Semin Liver Dis 1990;10:251–266. Ookhtens M, Kaplowitz N. Role of the liver in interorgan homeostasis of glutathione and cyst(e)ine. Semin Liver Dis 1998;18:313–329. Beatty PW, Reed DJ. Involvement of the cystathionine pathway in the biosynthesis of glutathione by isolated rat hepatocytes. Arch Biochem Biophys 1980;204:80–87. Coles B. Effects of modifying structure on electrophilic reactions with biological nucleophiles. Drug Metab Rev 1984;15:1307–1334. Mannervik B, Awasthi YC, Board PG, et al. Nomenclature for human glutathione transferases. Biochem J 1992; 282:305–306. Coles B, Ketterer B. The role of glutathione and glutathione transferases in chemical carcinogenesis. Crit Rev Biochem Mol Biol 1990;25:47–70. Mitchell JR, Jollow DJ, Potter WZ, Gillette JR, Brodie BB. Acetaminophen-induced hepatic necrosis. IV. Protective role of glutathione. J Pharmacol Exp Ther 1973;187:211–217. Reed DJ. Cellular defense mechanisms against reactive metabolites. Orlando, FL: Academic Press, 1985. Beck LV, Rieks VD, Duncan B. Diurnal variation in mouse and rat liver sulfhydryl. Proc Soc Exp Biol Med 1958;97:224.
Chapter 26 DRUG-INDUCED LIVER INJURY
44. Meister A. Selective modification of glutathione metabolism. Science 1983;220:472–477. 45. Prescott LF, Critchley JA. The treatment of acetaminophen poisoning. Annu Rev Pharmacol Toxicol 1983;23:87–101. 46. Kwak MK, Wakabayashi N, Kensler TW. Chemoprevention through the Keap1–Nrf2 signaling pathway by phase 2 enzyme inducers. Mutat Res 2004;555:133–148. 47. Lee JM, Johnson JA. An important role of Nrf2-ARE pathway in the cellular defense mechanism. J Biochem Mol Biol 2004;37:139–143. 48. Prochaska HJ, Santamaria AB, Talalay P. Rapid detection of inducers of enzymes that protect against carcinogens. Proc Natl Acad Sci USA 1992;89:2394–2398. 49. Zhang Y, Talalay P, Cho CG, Posner GH. A major inducer of anticarcinogenic protective enzymes from broccoli: isolation and elucidation of structure. Proc Natl Acad Sci USA 1992;89:2399–2403. 50. Dinkova-Kostova AT, Talalay P. Persuasive evidence that quinone reductase type 1 (DT diaphorase) protects cells against the toxicity of electrophiles and reactive forms of oxygen. Free Radic Biol Med 2000;29:231–240. 51. Kensler TW, Curphey TJ, Maxiutenko Y, Roebuck BD. Chemoprotection by organosulfur inducers of phase 2 enzymes: dithiolethiones and dithiins. Drug Metabol Drug Interact 2000;17:3–22. 52. McClain CJ, Price S, Barve S, Devalarja R, Shedlofsky S. Acetaminophen hepatotoxicity: An update. Curr Gastroenterol Rep 1999;1:42–49. 53. Dahlin DC, Miwa GT, Lu AY, Nelson SD. N-acetyl-pbenzoquinone imine: a cytochrome P-450-mediated oxidation product of acetaminophen. Proc Natl Acad Sci USA 1984;81:1327–1331. 54. Manyike PT, Kharasch ED, Kalhorn TF, et al. Contribution of CYP2E1 and CYP3A to acetaminophen reactive metabolite formation. Clin Pharmacol Ther. 2000;67:275–282. 55. Jollow DJ, Mitchell JR, Potter WZ, et al. Acetaminopheninduced hepatic necrosis. II. Role of covalent binding in vivo. J Pharmacol Exp Ther 1973;187:195–202. 56. Potter WZ, Thorgeirsson SS, Jollow DJ, Mitchell JR. Acetaminophen-induced hepatic necrosis. V. Correlation of hepatic necrosis, covalent binding and glutathione depletion in hamsters. Pharmacology 1974;12:129–143. 57. Walker RM, Racz WJ, McElligott TF. Acetaminophen-induced hepatotoxicity in mice. Lab Invest 1980;42:181–189. 58. Placke ME, Ginsberg GL, Wyand DS, Cohen SD. Ultrastructural changes during acute acetaminophen-induced hepatotoxicity in the mouse: a time and dose study. Toxicol Pathol 1987;15:431–438. 59. Myers CE, McGuire WP, Liss RH, et al. Adriamycin: the role of lipid peroxidation in cardiac toxicity and tumor response. Science 1977;197:165–167. 60. Burcham PC, Harman AW. Mitochondrial dysfunction in paracetamol hepatotoxicity: in vitro studies in isolated mouse hepatocytes. Toxicol Lett 1990;50.37–48. 61. Bartolone JB, Birge RB, Bulera SJ, et al. Purification, antibody production, and partial amino acid sequence of the 58-kDa acetaminophen-binding liver proteins. Toxicol Appl Pharmacol 1992;113:19–29. 62. Pumford NR, Martin BM, Hinson JA. A metabolite of acetaminophen covalently binds to the 56 kDa selenium binding protein. Biochem Biophys Res Commun 1992;182:1348–1355. 63. Birge RB, Bartolone JB, Hart SG, et al. Acetaminophen hepatotoxicity: correspondence of selective protein arylation in human and mouse liver in vitro, in culture, and in vivo. Toxicol Appl Pharmacol 1990;105:472–482. 64. Labadarios D, Davis M, Portmann B, Williams R. Paracetamolinduced hepatic necrosis in the mouse-relationship between
65.
66.
67.
68.
69.
70.
71. 72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
covalent binding, hepatic glutathione depletion and the protective effect of alpha-mercaptopropionylglycine. Biochem Pharmacol 1977;26:31–35. Devalia JL, Ogilvie RC, McLean AE. Dissociation of cell death from covalent binding of paracetamol by flavones in a hepatocyte system. Biochem Pharmacol 1982;31: 3745–3749. Albano E, Rundgren M, Harvison PJ, Nelson SD, Moldeus P. Mechanisms of N-acetyl-p–benzoquinone imine cytotoxicity. Mol Pharmacol 1985;28:306–311. Tee LB, Boobis AR, Huggett AC, Davies DS. Reversal of acetaminophen toxicity in isolated hamster hepatocytes by dithiothreitol. Toxicol Appl Pharmacol 1986;83:294–314. Moore M, Thor H, Moore G, et al. The toxicity of acetaminophen and N-acetyl-p-benzoquinone imine in isolated hepatocytes is associated with thiol depletion and increased cytosolic Ca2+. J Biol Chem 1985;260:13035–13040. Coles B, Wilson I, Wardman P, et al. The spontaneous and enzymatic reaction of N-acetyl-p-benzoquinonimine with glutathione: a stopped-flow kinetic study. Arch Biochem Biophys 1988;264:253–260. Michael SL, Pumford NR, Mayeux PR, Niesman MR, Hinson JA. Pretreatment of mice with macrophage inactivators decreases acetaminophen hepatotoxicity and the formation of reactive oxygen and nitrogen species. Hepatology 1999;30:186–195. Anders MW. Bioactivation mechanisms and hepatocellular damage. New York: Raven Press, 1988. Sun F, Hamagawa E, Tsutsui C, et al. Evaluation of oxidative stress during apoptosis and necrosis caused by carbon tetrachloride in rat liver. Biochim Biophys Acta 2001;1535:186–191. Recknagel RO, Glende EA Jr, Dolak JA, Waller RL. Mechanisms of carbon tetrachloride toxicity. Pharmacol Ther 1989;43:139–154. Moore L, Rodman DG, Landon EJ. Calcium uptake of a rat liver microsomal subcellular fraction in response to in vivo administration of carbon tetrachloride. J Biol Chem 1976;251:1197–1201. Long RM, Moore L. Elevated cytosolic calcium in rat hepatocytes exposed to carbon tetrachloride. J Pharmacol Exp Ther 1986;238:186–191. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 1990;87:1620–1624. Radi R, Beckman JS, Bush KM, Freeman BA. Peroxynitriteinduced membrane lipid peroxidation: the cytotoxic potential of superoxide and nitric oxide. Arch Biochem Biophys 1991;288:481–487. Beckman JS, Crow JP. Pathological implications of nitric oxide, superoxide and peroxynitrite formation. Biochem Soc Trans 1993;21:330–334. Bus JS, Aust SD, Gibson JE. Lipid peroxidation: a possible mechanism for paraquat toxicity. Res Commun Chem Pathol Pharmacol 1975;11:31–38. Radjendirane V, Joseph P, Lee YH, et al. Disruption of the DT diaphorase (NQO1) gene in mice leads to increased menadione toxicity. J Biol Chem 1998;273:7382–7389. Bernardi P. Mitochondrial transport of cations: channels, exchangers, and permeability transition. Physiol Rev 1999;79:1127–1155. Yang J, Liu X, Bhalla K, et al. Prevention of apoptosis by Bcl-2: release of cytochrome c from mitochondria blocked. Science 1997;275:1129–1132. Susin SA, Lorenzo HK, Zamzami N, et al. Molecular characterization of mitochondrial apoptosis-inducing factor. Nature 1999;397:441–446.
539
Section IV. Toxin Mediated Liver Injury
84. Li LY, Luo X, Wang X. Endonuclease G is an apoptotic DNase when released from mitochondria. Nature 2001;412: 95–99. 85. Halliwell B, Chirico S. Lipid peroxidation: its mechanism, measurement, and significance. Am J Clin Nutr 1993;57:715S–724S. 86. van Kuijk FJ, Handelman GJ, Dratz EA. Consecutive action of phospholipase A2 and glutathione peroxidase is required for reduction of phospholipid hydroperoxides and provides a convenient method to determine peroxide values in membranes. J Free Radic Biol Med 1985;1:421–427. 87. van Kuijk FJ, Sevanian A, Handelman GJ. A new role for phospholipase A2: Protection of membranes from lipid peroxidation damage. Trends Biochem Sci 1987;12:31. 88. Ursini F, Maiorino M, Gregolin C. The selenoenzyme phospholipid hydroperoxide glutathione peroxidase. Biochim Biophys Acta 1985;839:62–70. 89. Ursini F, Bindoli A. The role of selenium peroxidases in the protection against oxidative damage of membranes. Chem Phys Lipids 1987;44:255–276. 90. Prohaska JR, Ganther HE. Glutathione peroxidase activity of glutathione-s-transferases purified from rat liver. Biochem Biophys Res Commun 1976;76:437–445. 91. Ketterer B, Meyer DJ, Dark AG. Soluble glutathione transferase isozymes. London: Academic Press, 1988. 92. Esterbauer H, Schaur RJ, Zollner H. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Biol Med 1991;11:81–128. 93. Jensson H, Guthenberg C, Alin P, Mannervik B. Rat glutathione transferase 8-8, an enzyme efficiently detoxifying 4-hydroxyalk2-enals. FEBS Lett 1986;203:207–209. 94. Danielson UH, Esterbauer H, Mannervik B. Structure-activity relationships of 4-hydroxyalkenals in the conjugation catalysed by mammalian glutathione transferases. Biochem J 1987;247:707–713. 95. Zimmerman HJ. Anesthetic agents in hepatotoxicity. New York: Appleton Century Croft, 1978. 96. Liu ZX, Kaplowitz N. Immune-mediated drug-induced liver disease. Clin Liver Dis 2002;6:467–486. 97. Bonkovsky HL. Granulomatous hepatitis and hepatic granulomas. In: Hurst JW, ed. Medicine for the practising physician. London: Butterworths, 1991: 1446–1449. 98. Klatskin G. Histopathology of the Liver. New York: Oxford University Press, 1993. 99. Goodman ZD. Drug hepatotoxicity. Clin Liver Dis 2002;6:381–397. 100. Pelli N, Setti M, Ceppa P, Toncini C, Indiveri F. Autoimmune hepatitis revealed by atorvastatin. Eur J Gastroenterol Hepatol 2003;15:921–924. 101. Klion FM, Schaffner F, Popper H. Hepatitis after exposure to halothane. Ann Intern Med 1969;71:467–477. 102. Inman WH, Mushin WW. Jaundice after repeated exposure to halothane: an analysis of Reports to the Committee on Safety of Medicines. Br Med J 1974;1:5–10. 103. Klatskin G, Kimberg DV. Recurrent hepatitis attributable to halothane sensitization in an anesthetist. N Engl J Med 1969;280:515–522. 104. Lewis JH, Zimmerman HJ, Ishak KG, Mullick FG. Enflurane hepatotoxicity. A clinicopathologic study of 24 cases. Ann Intern Med 1983;98:984–992. 105. Carrigan TW, Straughen WJ. A report of hepatic necrosis and death following isoflurane anesthesia. Anesthesiology 1987;67:581–583. 106. Martin JL, Plevak DJ, Flannery KD, et al. Hepatotoxicity after desflurane anesthesia. Anesthesiology 1995;83:1125–1129. 107. Ogawa M, Doi K, Mitsufuji T, Satoh K, Takatori T. Drug induced hepatitis following sevoflurane anesthesia in a child. Masui 1991;40:1542–1545.
540
108. Sato E, Nagano K, Akasaka N, Manaka S, Tateda T. The effect of repeated sevoflurane anesthesia on hepatic function and immunological system. Masui 2003;52:1062–1065. 109. Joshi PH, Conn HO. The syndrome of methoxyfluraneassociated hepatitis. Ann Intern Med 1974;80:395–401. 110. Bunker JP. Final Report of the National Halothane Study. Anesthesiology 1968;29:231–232. 111. Brown BR Jr, Gandolfi AJ. Adverse effects of volatile anaesthetics. Br J Anaesth 1987;59:14–23. 112. Kenna JG, Neugereger J. Immunopathogenesis and treatment of halothane hepatitis. Clin Immunother 1995;3:108–124. 113. Vergani D, Tsantoulas D, Eddleston AL, Davis M, Williams R. Sensitisation to halothane-altered liver components in severe hepatic necrosis after halothane anaesthesia. Lancet 1978;2:801–803. 114. Vergani D, Mieli-Vergani G, Alberti A, et al. Antibodies to the surface of halothane-altered rabbit hepatocytes in patients with severe halothane-associated hepatitis. N Engl J Med 1980;303:66–71. 115. Eliasson E, Kenna JG. Cytochrome P450 2E1 is a cell surface autoantigen in halothane hepatitis. Mol Pharmacol 1996;50:573–582. 116. Topal A, Gul N, Ilcol Y, Gorgul OS. Hepatic effects of halothane, isoflurane or sevoflurane anaesthesia in dogs. J Vet Med A Physiol Pathol Clin Med 2003;50:530–533. 117. Reichle FM, Conzen PF. Halogenated inhalational anaesthetics. Best Pract Res Clin Anaesthesiol 2003;17:29–46. 118. Kenna JG. Mechanism, pathology, and clinical presentation of hepatotoxicity of anesthetic agents. New York: Marcel Dekker, 2003. 119. Benjamin SB, Goodman ZD, Ishak KG, Zimmerman HJ, Irey NS. The morphologic spectrum of halothane-induced hepatic injury: analysis of 77 cases. Hepatology 1985;5:1163–1171. 120. Malnick SD, Mahlab K, Borchardt J, Sokolowski N, Attali M. Acute cholestatic hepatitis after exposure to isoflurane. Ann Pharmacother 2002;36:261–263. 121. Wilton LV, Stephens MD, Mann RD. Interim report on the incidence of visual field defects in patients on long term vigabatrin therapy. Pharmacoepidemiol Drug Saf 1999;1:S9–14. 122. Bialer M, Johannessen SI, Kupferberg HJ, et al. Progress report on new antiepileptic drugs: a summary of the Seventh Eilat Conference (EILAT VII). Epilepsy Res 2004;61:1–48. 123. Brodie JD, Figueroa E, Laska EM, Dewey SL. Safety and efficacy of gamma-vinyl GABA (GVG) for the treatment of methamphetamine and/or cocaine addiction. Synapse 2005;55:122–125. 124. Haruda F. Phenytoin hypersensitivity: 38 cases. Neurology 1979;29:1480–1485. 125. Mullick FG, Ishak KG. Hepatic injury associated with diphenylhydantoin therapy. A clinicopathologic study of 20 cases. Am J Clin Pathol 1980;74:442–452. 126. Vittorio CC, Muglia JJ. Anticonvulsant hypersensitivity syndrome. Arch Intern Med 1995;155:2285–2290. 127. Leeder JS, Pirmohamed M. Anticonvulsant agents. In: Kaplowitz N, DeLeve LD, eds. Drug induced liver disease. New York: Marcel Dekker, 2003: 425–446. 128. Chitturi S, George J. Hepatotoxicity of commonly used drugs: nonsteroidal anti-inflammatory drugs, antihypertensives, antidiabetic agents, anticonvulsants, lipid-lowering agents, psychotropic drugs. Semin Liver Dis 2002;22:169–183. 129. Misra UK, Kalita J, Rathore C. Phenytoin and carbamazepine cross reactivity: report of a case and review of literature. Postgrad Med J 2003;79:703–704. 130. Browne TR. Fosphenytoin (Cerebyx). Clin Neuropharmacol 1997;20:1–12. 131. Askmark H, Wiholm BE. Epidemiology of adverse reactions to carbamazepine as seen in a spontaneous reporting system. Acta Neurol Scand 1990;81:131–140.
Chapter 26 DRUG-INDUCED LIVER INJURY
132. Pirmohamed M, Kitteringham NR, Guenthner TM, Breckenridge AM, Park BK. An investigation of the formation of cytotoxic, protein-reactive and stable metabolites from carbamazepine in vitro. Biochem Pharmacol 1992;43:1675–1682. 133. Farrell GC. Drug-induced liver disease. Edinburgh: Churchill Livingstone, 1994. 134. Isojarvi JI, Pakarinen AJ, Rautio A, Pelkonen O, Myllyla VV. Liver enzyme induction and serum lipid levels after replacement of carbamazepine with oxcarbazepine. Epilepsia 1994;35:1217–1220. 135. Schmidt D, Arroyo S, Baulac M, et al. Recommendations on the clinical use of oxcarbazepine in the treatment of epilepsy: a consensus view. Acta Neurol Scand 2001;104:167–170. 136. Bosdure E, Cano A, Roquelaure B, et al. Oxcarbazepine and DRESS syndrome: a paediatric cause of acute liver failure. Arch Pediatr 2004;11:1073–1077. 137. Perucca E. Pharmacological and therapeutic properties of valproate: a summary after 35 years of clinical experience. CNS Drugs 2002;16:695–714. 138. Bryant AE, III, Dreifuss FE. Valproic acid hepatic fatalities. III. US experience since 1986. Neurology 1996;46:465–469. 139. Anderson GD, Acheampong AA, Wilensky AJ, Levy RH. Effect of valproate dose on formation of hepatotoxic metabolites. Epilepsia 1992;33:736–742. 140. Gopaul S, Farrell K, Abbott F. Effects of age and polytherapy, risk factors of valproic acid (VPA) hepatotoxicity, on the excretion of thiol conjugates of (E)-2,4-diene VPA in people with epilepsy taking VPA. Epilepsia 2003;44:322–328. 141. Gonzalez-Heydrich J, Raches D, Wilens TE, Leichtner A, Mezzacappa E. Retrospective study of hepatic enzyme elevations in children treated with olanzapine, divalproex, and their combination. J Am Acad Child Adolesc Psychiatry 2003;42:1227–1233. 142. Luef GJ, Waldmann M, Sturm W, et al. Valproate therapy and nonalcoholic fatty liver disease. Ann Neurol 2004;55:729–732. 143. Rugino TA, Janvier YM, Baunach JM, Bilat CA. Hypoalbuminemia with valproic acid administration. Pediatr Neurol 2003;29:440–444. 144. Coulter DL. Carnitine, valproate, and toxicity. J Child Neurol 1991;6:7–14. 145. Bohan TP, Helton E, McDonald I, et al. Effect of L-carnitine treatment for valproate-induced hepatotoxicity. Neurology 2001;56:1405–1409. 146. De V, Bohan TP, Coulter DL, et al. L-carnitine supplementation in childhood epilepsy: current perspectives. Epilepsia 1998;39:1216–1225. 147. Olsson R, Zettergren L. Anticonvulsant-induced liver damage. Am J Gastroenterol 1988;83:576–577. 148. O’Neil MG, Perdun CS, Wilson MB, McGown ST, Patel S. Felbamate-associated fatal acute hepatic necrosis. Neurology 1996;46:1457–1459. 149. Kapetanovic IM, Torchin CD, Thompson CD, et al. Potentially reactive cyclic carbamate metabolite of the antiepileptic drug felbamate produced by human liver tissue in vitro. Drug Metab Dispos 1998;26:1089–1095. 150. Kapetanovic IM, Torchin CD, Strong JM, et al. Reactivity of atropaldehyde, a felbamate metabolite in human liver tissue in vitro. Chem Biol Interact 2002;142:119–134. 151. Dieckhaus CM, Thompson CD, Roller SG, Macdonald TL. Mechanisms of idiosyncratic drug reactions: the case of felbamate. Chem Biol Interact 2002;142:99–117. 152. Egnell AC, Houston B, Boyer S. In vivo CYP3A4 heteroactivation is a possible mechanism for the drug interaction between felbamate and carbamazepine. J Pharmacol Exp Ther 2003;305:1251–1262. 153. Makin AJ, Fitt S, Williams R, Duncan JS. Fulminant hepatic failure induced by lamotrigine. Br Med J 1995;311:292.
154. Arnon R, DeVivo D, Defelice AR, Kazlow PG. Acute hepatic failure in a child treated with lamotrigine. Pediatr Neurol 1998;18:251–252. 155. Fayad M, Choueiri R, Mikati M. Potential hepatotoxicity of lamotrigine. Pediatr Neurol 2000;22:49–52. 156. Galindo PA, Borja J, Gomez E, et al. Anticonvulsant drug hypersensitivity. J Invest Allergol Clin Immunol 2002;12:299–304. 157. Dooley J, Camfield P, Gordon K, et al. Lamotrigine-induced rash in children. Neurology 1996;46:240–242. 158. Maggs JL, Naisbitt DJ, Tettey JN, Pirmohamed M, Park BK. Metabolism of lamotrigine to a reactive arene oxide intermediate. Chem Res Toxicol 2000;13:1075–1081. 159. Faught E. Clinical studies of topiramate. Drugs Today (Barcelona) 1999;35:49–57. 160. Bjoro K, Gjerstad L, Bentdal O, Osnes S, Schrumpf E. Topiramate and fulminant liver failure. Lancet 1998; 352:1119. 161. Bumb A, Diederich N, Beyenburg S. Adding topiramate to valproate therapy may cause reversible hepatic failure. Epileptic Disord 2003;5:157–159. 162. Doan RJ, Clendenning M. Topiramate and hepatotoxicity. Can J Psychiatry 2000;45:937–938. 163. Longin E, Teich M, Koelfen W, Konig S. Topiramate enhances the risk of valproate-associated side effects in three children. Epilepsia 2002;43:451–454. 164. Benabou R, Waters C. Hepatotoxic profile of catechol-Omethyltransferase inhibitors in Parkinson’s disease. Expert Opin Drug Saf 2003;2:263–267. 165. Watkins P. COMT inhibitors and liver toxicity. Neurology 2000;55:S51–S52. 166. Haasio K, Koponen A, Penttila KE, Nissinen E. Effects of entacapone and tolcapone on mitochondrial membrane potential. Eur J Pharmacol 2002;453:21–26. 167. Smith KS, Smith PL, Heady TN, et al. In vitro metabolism of tolcapone to reactive intermediates: relevance to tolcapone liver toxicity. Chem Res Toxicol 2003;16:123–128. 168. Olanow CW. Tolcapone and hepatotoxic effects. Tasmar Advisory Panel. Arch Neurol 2000;57:263–267. 169. Deane K, Spieker S, Clarke C. Catechol-O-methyltransferase inhibitors for levodopa-induced complications in Parkinson’s disease. Cochrane Database Syst Rev 2004;CD004554. 170. Trenkwalder C, Hundemer HP, Lledo A, et al. Efficacy of pergolide in treatment of restless legs syndrome: the PEARLS Study. Neurology 2004;62:1391–1397. 171. Orrego JJ, Chandler WF, Barkan AL. Pergolide as primary therapy for macroprolactinomas. Pituitary 2000;3:251–256. 172. Krivoy N, Ben-Arieh Y, Carter A, Alroy G. Methazolamideinduced hepatitis and pure RBC aplasia. Arch Intern Med 1981;141:1229–1230. 173. Hammel P, Larrey D, Bernuau J, et al. Acute hepatitis after tetrahydroaminoacridine administration for Alzheimer’s disease. J Clin Gastroenterol 1990;12:329–331. 174. Blackard WG Jr, Sood GK, Crowe DR, Fallon MB. Tacrine. A cause of fatal hepatotoxicity? J Clin Gastroenterol 1998;26:57–59. 175. Watkins PB, Zimmerman HJ, Knapp MJ, Gracon SI, Lewis KW. Hepatotoxic effects of tacrine administration in patients with Alzheimer ’s disease. JAMA 1994;271: 992–998. 176. Gracon SI, Knapp MJ, Berghoff WG, et al. Safety of tacrine: clinical trials, treatment IND, and postmarketing experience. Alzheimer Dis Assoc Disord 1998;12:93–101. 177. Stachlewitz RF, Arteel GE, Raleigh JA, et al. Development and characterization of a new model of tacrine-induced hepatotoxicity. Role of the sympathetic nervous system and hypoxia–reoxygenation. J Pharmacol Exp Ther 1997;282:1591–1599.
541
Section IV. Toxin Mediated Liver Injury
178. Galisteo M, Rissel M, Sergent O, et al. Hepatotoxicity of tacrine: occurrence of membrane fluidity alterations without involvement of lipid peroxidation. J Pharmacol Exp Ther 2000;294:160–167. 179. Wolinsky JS. Glatiramer acetate for the treatment of multiple sclerosis. Expert Opin Pharmacother 2004;5:875–891. 180. Munari L, Lovati R, Boiko A. Therapy with glatiramer acetate for multiple sclerosis. Cochrane Database Syst Rev 2004;CD004678. 181. Miller RG, Mitchell JD, Lyon M, Moore DH. Riluzole for amyotrophic lateral sclerosis (ALS)/motor neuron disease (MND). Cochrane Database Syst Rev 2002;CD001447. 182. Remy AJ, Camu W, Ramos J, Blanc P, Larrey D. Acute hepatitis after riluzole administration. J Hepatol 1999;30:527–530. 183. Castells LI, Gamez J, Cervera C, Guardia J. Icteric toxic hepatitis associated with riluzole. Lancet 1998;351:648. 184. Lacomblez L, Dib M, Doppler V, et al. Tolerance of riluzole in a phase IIIb clinical trial. Therapie 2002;57:65–71. 185. Bensimon G, Doble A. The tolerability of riluzole in the treatment of patients with amyotrophic lateral sclerosis. Expert Opin Drug Saf 2004;3:525–534. 186. Schwartz JR, Feldman NT, Fry JM, Harsh J. Efficacy and safety of modafinil for improving daytime wakefulness in patients treated previously with psychostimulants. Sleep Med 2003;4:43–49. 187. de la lglesia FA, Haskins JR, Feuer G. Hepatotoxicity of cardiovascular and antidiabetic drugs. In: Kaplowitz N, DeLeve LD, eds. Drug induced liver disease. New York: Marcel Dekker, 2003: 549–592. 188. Saw D, Pitman E, Maung M, et al. Granulomatous hepatitis associated with glyburide. Dig Dis Sci 1996;41:322–325. 189. Schatz H. Preclinical and clinical studies on safety and tolerability of repaglinide. Exp Clin Endocrinol Diabetes 1999;107(Suppl 4):S144–S148. 190. Nan DN, Hernandez JL, Fernandez-Ayala M, Carrascosa M. Acute hepatotoxicity caused by repaglinide. Ann Intern Med 2004;141:823. 191. Niemi M, Kajosaari LI, Neuvonen M, Backman JT, Neuvonen PJ. The CYP2C8 inhibitor trimethoprim increases the plasma concentrations of repaglinide in healthy subjects. Br J Clin Pharmacol 2004;57:441–447. 192. Lu HC, Parikh PP, Lorber DL. Phenformin-associated lactic acidosis due to imported phenformin. Diabetes Care 1996; 19:1449–1450. 193. Millican S, Cottrell N, Green B. Do risk factors for lactic acidosis influence dosing of metformin? J Clin Pharmacol Ther 2004;29:449–454. 194. Nammour FE, Fayad NF, Peikin SR. Metformin-induced cholestatic hepatitis. Endocr Pract 2003;9:307–309. 195. Babich MM, Pike I, Shiffman ML. Metformin-induced acute hepatitis. Am J Med 1998;104:490–492. 196. Holland W, Morrison T, Chang Y, Wiernsperger N, Stith BJ. Metformin (Glucophage) inhibits tyrosine phosphatase activity to stimulate the insulin receptor tyrosine kinase. Biochem Pharmacol 2004;67:2081–2091. 197. Nair S, Diehl AM, Wiseman M, Farr GH Jr, Perrillo RP. Metformin in the treatment of non-alcoholic steatohepatitis: a pilot open label trial. Aliment Pharmacol Ther 2004;20:23–28. 198. Yki-Jarvinen H. Thiazolidinediones. N Engl J Med 2004;351:1106–1118. 199. Tolman KG, Chandramouli J. Hepatotoxicity of the thiazolidinediones. Clin Liver Dis 2003;7:369–379. 200. Chitturi S, George J. Hepatotoxicity of commonly used drugs: nonsteroidal anti-inflammatory drugs, antihypertensives, antidiabetic agents, anticonvulsants, lipid lowering agents, psychotropic drugs. Sem Liver Dis 2002;22:169–183.
542
201. Lebovitz HE, Kreider M, Freed MI. Evaluation of liver function in type 2 diabetic patients during clinical trials: evidence that rosiglitazone does not cause hepatic dysfunction. Diabetes Care 2002;25:815–821. 202. Graham DJ, Green L, Senior JR, Nourjah P. Troglitazoneinduced liver failure: a case study. Am J Med 2003;114:299–306. 203. Nozawa T, Sugiura S, Nakajima M, et al. Involvement of organic anion transporting polypeptides in the transport of troglitazone sulfate: implications for understanding troglitazone hepatotoxicity. Drug Metab Dispos 2004;32:291–294. 204. Westphal JF, Brogard JM. Antibacterials and antifungal agents. In: Kaplowitz N, DeLeve LD, eds. Drug induced liver disease. New York: Marcel Dekker, 2003: 471–504. 205. Sheets JJ, Mason JI. Ketoconazole: a potent inhibitor of cytochrome P-450-dependent drug metabolism in rat liver. Drug Metab Dispos 1984;12:603–606. 206. Wang JS, Wen X, Backman JT, et al. Midazolam alphahydroxylation by human liver microsomes in vitro: inhibition by calcium channel blockers, itraconazole and ketoconazole. Pharmacol Toxicol 1999;85:157–161. 207. Lewis JH, Zimmerman HJ, Benson GD, Ishak KG. Hepatic injury associated with ketoconazole therapy. Analysis of 33 cases. Gastroenterology 1984;86:503–513. 208. Stricker BH, Blok AP, Bronkhorst FB, Van Parys GE, Desmet VJ. Ketoconazole-associated hepatic injury. A clinicopathological study of 55 cases. J Hepatol 1986;3:399–406. 209. Lake-Bakaar G, Scheuer PJ, Sherlock S. Hepatic reactions associated with ketoconazole in the United Kingdom. Br Med J (Clin Res Ed) 1987;294:419–422. 210. Chien RN, Yang LJ, Lin PY, Liaw YF. Hepatic injury during ketoconazole therapy in patients with onychomycosis: a controlled cohort study. Hepatology 1997;25:103–107. 211. Garcia Rodriguez LA, Duque A, Castellsague J, PerezGutthann S, Stricker BH. A cohort study on the risk of acute liver injury among users of ketoconazole and other antifungal drugs. Br J Clin Pharmacol 1999;48:847–852. 212. Chien RN, Sheen IS, Liaw YF. Unintentional rechallenge resulting in a causative relationship between ketoconazole and acute liver injury. Int J Clin Pract 2003;57:829–830. 213. Kim TH, Kim BH, Kim YW, et al. Liver cirrhosis developed after ketoconazole-induced acute hepatic injury. J Gastroenterol Hepatol 2003;18:1426–1429. 214. Rodriguez RJ, Acosta D, Jr. N-deacetyl ketoconazole-induced hepatotoxicity in a primary culture system of rat hepatocytes. Toxicology 1997;117:123–131. 215. Rodriguez RJ, Buckholz CJ. Hepatotoxicity of ketoconazole in Sprague–Dawley rats: glutathione depletion, flavin-containing monooxygenases-mediated bioactivation and hepatic covalent binding. Xenobiotica 2003;33:429–441. 216. Velayudham LS, Farrell GC. Drug-induced cholestasis. Expert Opin Drug Saf 2003;2:287–304. 217. Bradbury BD, Jick SS. Itraconazole and fluconazole and certain rare, serious adverse events. Pharmacotherapy 2002;22:697–700. 218. Gupta AK, Chwetzoff E, Del RJ, Baran R. Hepatic safety of itraconazole. J Cutan Med Surg 2002;6:210–213. 219. Adriaenssens B, Roskams T, Steger P, Van SW. Hepatotoxicity related to itraconazole. report of three cases. Acta Clin Belg 2001;56:364–369. 220. Wolf R, Wolf D, Kuperman S. Focal nodular hyperplasia of the liver after intraconazole treatment. J Clin Gastroenterol 2001;33:418–420. 221. Novelli V, Holzel H. Safety and tolerability of fluconazole in children. Antimicrob Agents Chemother 1999;43:1955–1960. 222. Linnebur SA, Parnes BL. Pulmonary and hepatic toxicity due to nitrofurantoin and fluconazole treatment. Ann Pharmacother 2004;38:612–616.
Chapter 26 DRUG-INDUCED LIVER INJURY
223. Crerar-Gilbert A, Boots R, Fraenkel D, MacDonald GA. Survival following fulminant hepatic failure from fluconazole induced hepatitis. Anaesth Intens Care 1999;27:650–652. 224. Ullmann AJ. Review of the safety, tolerability, and drug interactions of the new antifungal agents caspofungin and voriconazole. Curr Med Res Opin 2003;19:263–271. 225. Gothard P, Rogers TR. Voriconazole for serious fungal infections. Int J Clin Pract 2004;58:74–80. 226. Hall M, Monka C, Krupp P, O’Sullivan D. Safety of oral terbinafine: results of a postmarketing surveillance study in 25,884 patients. Arch Dermatol 1997;133:1213–1219. 227. Chambers WM, Millar A, Jain S, Burroughs AK. Terbinafineinduced hepatic dysfunction. Eur J Gastroenterol Hepatol 2001;13:1115–1118. 228. Ajit C, Suvannasankha A, Zaeri N, Munoz SJ. Terbinafineassociated hepatotoxicity. Am J Med Sci 2003;325:292–295. 229. Zapata Garrido AJ, Romo AC, Padilla FB. Terbinafine hepatotoxicity. A case report and review of literature. Ann Hepatol 2003;2:47–51. 230. Lovell MO, Speeg KV, Havranek RD, Sharkey FE. Histologic changes resembling acute rejection in a liver transplant patient treated with terbinafine. Hum Pathol 2003;34:187–189. 231. Iverson SL, Uetrecht JP. Identification of a reactive metabolite of terbinafine: insights into terbinafine-induced hepatotoxicity. Chem Res Toxicol 2001;14:175–181. 232. Gupta AK, Adam P, Dlova N, et al. Therapeutic options for the treatment of tinea capitis caused by Trichophyton species: griseofulvin versus the new oral antifungal agents, terbinafine, itraconazole, and fluconazole. Pediatr Dermatol 2001;18:433–438. 233. Chiprut RO, Viteri A, Jamroz C, Dyck WP. Intrahepatic cholestasis after griseofulvin administration. Gastroenterology 1976;70:1141–1143. 234. Johnson MD, Perfect JR. Caspofungin: first approved agent in a new class of antifungals. Expert Opin Pharmacother 2003;4:807–823. 235. Walsh TJ, Teppler H, Donowitz GR, et al. Caspofungin versus liposomal amphotericin B for empirical antifungal therapy in patients with persistent fever and neutropenia. N Engl J Med 2004;351:1391–1402. 236. Denning DW. Echinocandin antifungal drugs. Lancet 2003;362:1142–1151. 237. Stevens DA, Espiritu M, Parmar R. Paradoxical effect of caspofungin: reduced activity against Candida albicans at high drug concentrations. Antimicrob Agents Chemother 2004;48:3407–3411. 238. Marr KA, Hachem R, Papanicolaou G, et al. Retrospective study of the hepatic safety profile of patients concomitantly treated with caspofungin and cyclosporin A. Transpl Infect Dis 2004;6:110–116. 239. Stone JA, Migoya EM, Hickey L, et al. Potential for interactions between caspofungin and nelfinavir or rifampin. Antimicrob Agents Chemother 2004;48:4306–4314. 240. Vermes A, van der Sijs H, Guchelaar HJ. Flucytosine: correlation between toxicity and pharmacokinetic parameters. Chemotherapy 2000;46:86–94. 241. Singh N, Gayowski T, Marino IR. Successful treatment of disseminated cryptococcosis in a liver transplant recipient with fluconazole and flucytosine, an all oral regimen. Transpl Int 1998;11:63–65. 242. Phillips-Howard PA, West LJ. Serious adverse drug reactions to pyrimethamine–sulphadoxine, pyrimethamine–dapsone and to amodiaquine in Britain. J Roy Soc Med 1990;83:82–85. 243. Zitelli BJ, Alexander J, Taylor S, et al. Fatal hepatic necrosis due to pyrimethamine–sulfadoxine (Fansidar). Ann Intern Med 1987;106:393–395. 244. Bernuau J, Larrey D, Campillo B, et al. Amodiaquine-induced fulminant hepatitis. J Hepatol 1988;6:109–112.
245. Vreugdenhil CJ, Scheper FY, Hoogstraatte SR, et al. Comparison of the parasitologic efficacy of amodiaquine and sulfadoxine-pyrimethamine in the treatment of Plasmodium falciparum malaria in the Bungoma District of western Kenya. Am J Trop Med Hyg 2004;71:537–541. 246. Staedke SG, Mpimbaza A, Kamya MR, et al. Combination treatments for uncomplicated falciparum malaria in Kampala, Uganda: randomised clinical trial. Lancet 2004;364:1950–1957. 247. Brown SJ, Desmond PV. Hepatotoxicity of antimicrobial agents. Semin Liver Dis 2002;22:157–167. 248. Eland IA, Kerkhof SC, Overbosch D, Wismans PJ, Stricker BH. Cholestatic hepatitis ascribed to the use of thiabendazole. Ned Tijdschr Geneeskd 1998;142:1331–1334. 249. Skandrani K, Richardet JP, Duvoux C, et al. Hepatic transplantation for severe ductopenia related to ingestion of thiabendazole. Gastroenterol Clin Biol 1997;21:623–625. 250. Roy MA, Nugent FW, Aretz HT. Micronodular cirrhosis after thiabendazole. Dig Dis Sci 1989;34:938–941. 251. Manivel JC, Bloomer JR, Snover DC. Progressive bile duct injury after thiabendazole administration. Gastroenterology 1987;93:245–249. 252. Garg PK, Tandon RK. Antituberculous agents-induced liver injury. In: Kaplowitz N, DeLeve LD, eds. Drug induced liver disease. New York: Marcel Dekker, 2003: 505–517. 253. Steele MA, Burk RF, DesPrez RM. Toxic hepatitis with isoniazid and rifampin. A meta-analysis. Chest 1991;99:465–471. 254. Singh J, Arora A, Garg PK, et al. Antituberculosis treatmentinduced hepatotoxicity: role of predictive factors. Postgrad Med J 1995;71:359–362. 255. Targeted tuberculin testing and treatment of latent tuberculosis infection. [Official statement of the American Thoracic Society (ATS) and the Centers for Disease Control and Prevention (CDC). This statement was endorsed by the Council of the Infectious Diseases Society of America. (IDSA), September 1999.] Am J Respir Crit Care Med 2000;161:S221–S247. 256. LoBue PA, Moser KS. Use of isoniazid for latent tuberculosis infection in a public health clinic. Am J Respir Crit Care Med 2003;168:443–447. 257. Update: Fatal and severe liver injuries associated with rifampin and pyrazinamide for latent tuberculosis infection. 733–755. 2001. MMWR Morb Mortal Wkly Rep. American Thoracic Society/CDC recommendations – United States. 258. McNeill L, Allen M, Estrada C, Cook P. Pyrazinamide and rifampin vs isoniazid for the treatment of latent tuberculosis: improved completion rates but more hepatotoxicity. Chest 2003;123:102–106. 259. Jasmer RM, Saukkonen JJ, Blumberg HM, et al. Short-course rifampin and pyrazinamide compared with isoniazid for latent tuberculosis infection: a multicenter clinical trial. Ann Intern Med 2002;137:640–647. 260. van Hest R, Baars H, Kik S, et al. Hepatotoxicity of rifampinpyrazinamide and isoniazid preventive therapy and tuberculosis treatment. Clin Infect Dis 2004;39:488–496. 261. Durand F, Jebrak G, Pessayre D, Fournier M, Bernuau J. Hepatotoxicity of antitubercular treatments. Rationale for monitoring liver status. Drug Saf 1996;15:394–405. 262. Kanel GC. Histopathology of drug-induced liver disease. In: Kaplowitz N, DeLeve LD, eds. Drug induced liver disease. New York: Marcel Dekker, 2003: 243–286. 263. Nishimura Y, Kurata N, Sakurai E, Yasuhara H. Inhibitory effect of antituberculosis drugs on human cytochrome P450-mediated activities. J Pharmacol Sci 2004;96: 293–300. 264. Zhou S, Chan E, Lim LY, et al. Therapeutic drugs that behave as mechanism-based inhibitors of cytochrome P450 3A4. Curr Drug Metab 2004;5:415–442.
543
Section IV. Toxin Mediated Liver Injury
265. Faucette SR, Wang H, Hamilton GA, et al. Regulation of CYP2B6 in primary human hepatocytes by prototypical inducers. Drug Metab Dispos 2004;32:348–358. 266. Gerbal-Chaloin S, Pascussi JM, Pichard-Garcia L, et al. Induction of CYP2C genes in human hepatocytes in primary culture. Drug Metab Dispos 2001;29:242–251. 267. Chen Y, Ferguson SS, Negishi M, Goldstein JA. Induction of human CYP2C9 by rifampicin, hyperforin, and phenobarbital is mediated by the pregnane X receptor. J Pharmacol Exp Ther 2004;308:495–501. 268. Burk O, Koch I, Raucy J, et al. The induction of cytochrome P450 3A5 (CYP3A5) in the human liver and intestine is mediated by the xenobiotic sensors pregnane X receptor (PXR) and constitutively activated receptor (CAR). J Biol Chem 2004;279:38379–38385. 269. Huang YS, Chern HD, Su WJ, et al. Polymorphism of the N-acetyltransferase 2 gene as a susceptibility risk factor for antituberculosis drug-induced hepatitis. Hepatology 2002;35:883–889. 270. Ohkawa K, Hashiguchi M, Ohno K, et al. Risk factors for antituberculous chemotherapy-induced hepatotoxicity in Japanese pediatric patients. Clin Pharmacol Ther 2002;72:220–226. 271. Campos-Franco J, Gonzalez-Quintela A, Alende-Sixto MR. Isoniazid-induced hyperacute liver failure in a young patient receiving carbamazepine. Eur J Intern Med 2004;15:396–397. 272. Tahaoglu K, Atac G, Sevim T, et al. The management of antituberculosis drug-induced hepatotoxicity. Int J Tuberc Lung Dis 2001;5:65–69. 273. Mitra AK, Thummel KE, Kalhorn TF, et al. Metabolism of dapsone to its hydroxylamine by CYP2E1 in vitro and in vivo. Clin Pharmacol Ther 1995;58:556–566. 274. Coleman MD. Dapsone toxicity: some current perspectives. Gen Pharmacol 1995;26:1461–1467. 275. Kumar RH, Kumar MV, Thappa DM. Dapsone syndrome – a five year retrospective analysis. Indian J Leprosy 1998;70:271–276. 276. Weiner M, Bock N, Peloquin CA, et al. Pharmacokinetics of rifapentine at 600, 900, and 1,200 mg during once-weekly tuberculosis therapy. Am J Respir Crit Care Med 2004;169:1191–1197. 277. Benator D, Bhattacharya M, Bozeman L, et al. Rifapentine and isoniazid once a week versus rifampicin and isoniazid twice a week for treatment of drug-susceptible pulmonary tuberculosis in HIV-negative patients. A randomised clinical trial. Lancet 2002;360:528–534. 278. Keung AC, Eller MG, Weir SJ. Pharmacokinetics of rifapentine in patients with varying degrees of hepatic dysfunction. J Clin Pharmacol 1998;38:517–524. 279. Tam CM, Chan SL, Kam KM, Goodall RL, Mitchison DA. Rifapentine and isoniazid in the continuation phase of a 6month regimen. Final report at 5 years. Prognostic value of various measures. Int J Tuberc Lung Dis 2002;6:3–10. 280. Conn HO, Binder HJ, Orr HD. Ethionamide-induced hepatitis. A review with a report of an additional case. Am Rev Respir Dis 1964;90:542–552. 281. Bonacini M, Louie S, Bzowej N, Wohl AR. Survival in patients with HIV infection and viral hepatitis B or C: a cohort study. AIDS 2004;18:2039–2045. 282. Copeland L, Budd J, Robertson JR, Elton RA. Changing patterns in causes of death in a cohort of injecting drug users, 1980–2001. Arch Intern Med 2004;164:1214–1220. 283. Sheng WH, Chen MY, Hsieh SM, et al. Impact of chronic hepatitis B virus (HBV) infection on outcomes of patients infected with HIV in an area where HBV infection is hyperendemic. Clin Infect Dis 2004;38:1471–1477. 284. Lewden C, Salmon D, Morlat P, et al. Causes of death among human immunodeficiency virus (HIV)-infected adults in the
544
285.
286.
287.
288.
289. 290. 291.
292.
293.
294.
295.
296.
297.
298.
299.
300.
301. 301a.
302.
era of potent antiretroviral therapy: emerging role of hepatitis and cancers, persistent role of AIDS. Int J Epidemiol 2005;34:121–130. Dieterich DT. Human immunodeficiency virus and the liver: lessons learned and still to be learned. Semin Liver Dis 2003;23:107–114. Greub G, Ledergerber B, Battegay M, et al. Clinical progression, survival, and immune recovery during antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet 2000;356:1800–1805. Anderson KB, Guest JL, Rimland D. Hepatitis C virus coinfection increases mortality in HIV-infected patients in the highly active antiretroviral therapy era: data from the HIV Atlanta VA Cohort Study. Clin Infect Dis 2004;39:1507–1513. Macias J, Melguizo I, Fernandez-Rivera FJ, et al. Mortality due to liver failure and impact on survival of hepatitis virus infections in HIV-infected patients receiving potent antiretroviral therapy. Eur J Clin Microbiol Infect Dis 2002;21:775–781. Ogedegbe AO, Sulkowski MS. Antiretroviral-associated liver injury. Clin Liver Dis 2003;7:475–499. Kontorinis N, Dieterich D. Hepatotoxicity of antiretroviral therapy. AIDS Rev 2003;5:36–43. Fellay J, Boubaker K, Ledergerber B, et al. Prevalence of adverse events associated with potent antiretroviral treatment: Swiss HIV Cohort Study. Lancet 2001;358:1322–1327. Bonacini M, Louie S, Weissman K. Hepatic injury from antiviral agents. Kaplowitz N, DeLeve LD, eds. Drug induced liver disease. New York: Marcel Dekker, 2003: 519–548. Sulkowski MS. Hepatotoxicity associated with antiretroviral therapy containing HIV-1 protease inhibitors. Semin Liver Dis 2003;23:183–194. Montessori V, Harris M, Montaner JS. Hepatotoxicity of nucleoside reverse transcriptase inhibitors. Semin Liver Dis 2003;23:167–172. Kontorinis N, Dieterich DT. Toxicity of non-nucleoside analogue reverse transcriptase inhibitors. Semin Liver Dis 2003;23:173–182. Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clin Ther 2000;22:685–708. Walker UA, Bauerle J, Laguno M, et al. Depletion of mitochondrial DNA in liver under antiretroviral therapy with didanosine, stavudine, or zalcitabine. Hepatology 2004;39:311–317. Barbaro G, Di LG, Asti A, et al. Hepatocellular mitochondrial alterations in patients with chronic hepatitis C: ultrastructural and biochemical findings. Am J Gastroenterol 1999;94:2198–2205. Falco V, Crespo M, Ribera E. Lactic acidosis related to nucleoside therapy in HIV-infected patients. Expert Opin Pharmacother 2003;4:1321–1329. Rabing CE, Stegger M, Jensen-Fangel S, Laursen AL, Ostergaard L. Mitochondrial DNA levels in fat and blood cells from patients with lipodystrophy or peripheral neuropathy and the effect of 90 days of high-dose coenzyme Q treatment: a randomized, double-blind, placebo-controlled pilot study. Clin Infect Dis 2004;39:1371–1379. Ernest CS, Hall SD, Jones DR. Mechanism-based inactivation of CYP3A by HIV protease inhibitors. J Pharmacol Exp Ther 2005;312:583–591. Hirani VN, Raucy JL, Lasker JM. Conversion of the HIV protease inhibitor nelfinavir to a bioactive metabolite by human liver CYP2C19. Drug Metab Dispos 2004;32:1462–1467. Sulkowski MS, Mehta SH, Chaisson RE, Thomas DL, Moore RD. Hepatotoxicity associated with protease inhibitor-based
Chapter 26 DRUG-INDUCED LIVER INJURY
303.
304.
305.
306.
307.
308.
309.
310.
311.
312.
313.
314.
315.
316a.
316b.
316c. 316d.
316e.
317.
antiretroviral regimens with or without concurrent ritonavir. AIDS 2004;18:2277–2284. Jain AB, Venkataramanan R, Eghtesad B, et al. Effect of coadministered lopinavir and ritonavir (Kaletra) on tacrolimus blood concentration in liver transplantation patients. Liver Transpl 2003;9:954–960. Voorman RL, Payne NA, Wienkers LC, Hauer MJ, Sanders PE. Interaction of delavirdine with human liver microsomal cytochrome P450: inhibition of CYP2C9, CYP2C19, and CYP2D6. Drug Metab Dispos 2001;29:41–47. Sulkowski MS, Thomas DL, Mehta SH, Chaisson RE, Moore RD. Hepatotoxicity associated with nevirapine or efavirenzcontaining antiretroviral therapy: role of hepatitis C and B infections. Hepatology 2002;35:182–189. Manfredi R, Calza L, Chiodo F. Efavirenz versus nevirapine in current clinical practice: a prospective, open-label observational study. J AIDS 2004;35:492–502. Dieterich DT, Robinson PA, Love J, Stern JO. Drug-induced liver injury associated with the use of nonnucleoside reversetranscriptase inhibitors. Clin Infect Dis 2004;38(Suppl 2):S80–S89. Ena J, Amador C, Benito C, Fenoll V, Pasquau F. Risk and determinants of developing severe liver toxicity during therapy with nevirapine- and efavirenz-containing regimens in HIVinfected patients. Int J STD AIDS 2003;14:776–781. Law WP, Dore GJ, Duncombe CJ, et al. Risk of severe hepatotoxicity associated with antiretroviral therapy in the HIV-NAT Cohort, Thailand, 1996–2001. AIDS 2003;17:2191–2199. Martin-Carbonero L, Nunez M, Gonzalez-Lahoz J, Soriano V. Incidence of liver injury after beginning antiretroviral therapy with efavirenz or nevirapine. HIV Clin Trials 2003;4:115–120. Macias J, Castellano V, Merchante N, et al. Effect of antiretroviral drugs on liver fibrosis in HIV-infected patients with chronic hepatitis C. Harmful impact of nevirapine. AIDS 2004;18:767–774. Patel SM, Johnson S, Belknap SM, et al. Serious adverse cutaneous and hepatic oxicities associated with nevirapine use by non-HIV-infected individuals. J AIDS 2004;35:120–125. Biesecker G, Karimi S, Desjardins J, Meyer D, Abbott B, Bendele R, Richardson F. Evaluation of mitochondrial DNA content and enzyme levels in tenofovir DF-treated rats rhesus monkeys and woodchucks. Antiviral Res 2003;58:217–225. Kuo A, Dienstag JL, Chung RT. Tenofovir disoproxil fumarate for the treatment of lamivudine-resistant hepatitis B. Clin Gastroenterol Hepatol 2004;2:266–272. Yuen MF, Lai CL. Adefovir dipivoxil in the treatment of chronic hepatitis B virus infection. Expert Rev Anti Infect Ther 2004;2:475–483. Koklu S, Yuksel O, Filik L, Uskudar O, Altundag K, Altiparmak E. Recurrent cholestasis due to ampicillin. Ann Pharmacother 2003;37:395–397. Gresser U. Amoxicillin-clavulanic therapy may be associated with severe side effects – review of the literature. Eur J Med Res 2001;6:139–149. Berg P, Hahn EG. Hepatotoxic reactions induced by betalactamase inhibitors. Eur J Med Res 2001;6:535–542. Garcia Rodriguez LA, Stricker BH, Zimmerman HJ. Risk of acute liver injury associated with the combination of amoxicillin and clavulanic acid. Arch Intern Med 1996;156:1327–1332. Koklu S, Koksal AS, Asil M, Kiyici H, Coban S, Arhan M. Probable sulbactam/ampicillin-associated prolonged cholestasis. Ann Pharmacother 2004;38:2055–2058. Kuye O, Teal J, DeVries VG, Morrow CA, Tally FP. Safety profile of piperacillin/tazobactam in phase I and III clinical studies. J Antimicrob Chemother 1993;31(Suppl A): 113–124.
318. Al-Homaidhi H, Abdel-Haq NM, El-Baba M, Asmar BI. Severe hepatitis associated with oxacillin therapy. South Med J 2002;95:650–652. 319. Maraqa NF, Gomez MM, Rathore MH, Alvarez AM. Higher occurrence of hepatotoxicity and rash in patients treated with oxacillin, compared with those treated with nafcillin and other commonly used antimicrobials. Clin Infect Dis 2002;34:50–54. 320. Shiffman ML, Keith FB, Moore EW. Pathogenesis of ceftriaxone-associated biliary sludge. In vitro studies of calcium–ceftriaxone binding and solubility. Gastroenterology 1990;99:1772–1778. 321. Derby LE, Jick H, Henry DA, Dean AD. Erythromycinassociated cholestatic hepatitis. Med J Aust 1993;158:600–602. 321a. Tietz A, Heim HM, Eriksson U, et al. Fulminant liver failure associated with clarithromycin. Ann Pharmacother 2003;37:57–60. 322. Masia M, Gutierrez F, Jimeno A, et al. Fulminant hepatitis and fatal toxic epidermal necrolysis (Lyell disease) coincident with clarithromycin administration in an alcoholic patient receiving disulfiram therapy. Arch Intern Med 2002;162:474–476. 323. Fox JC, Szyjkowski RS, Sanderson SO, Levine RA. Progressive cholestatic liver disease associated with clarithromycin treatment. J Clin Pharmacol 2002;42:676–680. 324. Suriawinata A, Min AD. A 33-year-old woman with jaundice after azithromycin use. Semin Liver Dis 2002;22:207–210. 324a. Chandrupatla S, Demetris AJ, Rabinovitz M. Azithromycininduced intrahepatic cholestasis. Dig Dis Sci 2002;47:2186–2188. 324b. Easton-Carter KL, Hardikar W, Smith AL. Possible roxithromycin-induced fulminant hepatic failure in a child. Pharmacotherapy 2001;21:867–870. 325. Periti P, Mazzei T, Mini E, Novelli A. Adverse effects of macrolide antibacterials. Drug Saf 1993;9:346–364. 326. Zuckerman JM. Macrolides and ketolides: azithromycin, clarithromycin, telithromycin. Infect Dis Clin North Am 2004;18:621–649. 327. Wellington K, Noble S. Telithromycin. Drugs 2004;64:1683–1694. 328. Gorski JC, Jones DR, Haehner-Daniels BD, et al. The contribution of intestinal and hepatic CYP3A to the interaction between midazolam and clarithromycin. Clin Pharmacol Ther 1998;64:133–143. 329. Bruce MA, Hall SD, Haehner-Daniels BD, Gorski JC. In vivo effect of clarithromycin on multiple cytochrome P450s. Drug Metab Dispos 2001;29:1023–1028. 330. Campana C, Regazzi MB, Buggia I, Molinaro M. Clinically significant drug interactions with cyclosporin. An update. Clin Pharmacokinet 1996;30:141–179. 331. Spicer ST, Liddle C, Chapman JR, et al. The mechanism of cyclosporine toxicity induced by clarithromycin. Br J Clin Pharmacol 1997;43:194–196. 332. Paterson DL, Singh N. Interactions between tacrolimus and antimicrobial agents. Clin Infect Dis 1997;25:1430–1440. 333. Goldberg MJ, Ring B, DeSante K, et al. Effect of dirithromycin on human CYP3A in vitro and on pharmacokinetics and pharmacodynamics of terfenadine in vivo. J Clin Pharmacol 1996;36:1154–1160. 334. Kostrubsky VE, Strom SC, Hanson J, et al. Evaluation of hepatotoxic potential of drugs by inhibition of bile-acid transport in cultured primary human hepatocytes and intact rats. Toxicol Sci 2003;76:220–228. 335. Bertino J Jr, Fish D. The safety profile of the fluoroquinolones. Clin Ther 2000;22:798–817. 336. Coleman CI, Spencer JV, Chung JO, Reddy P. Possible gatifloxacin-induced fulminant hepatic failure. Ann Pharmacother 2002;36:1162–1167. 337. Bataille L, Rahier J, Geubel A. Delayed and prolonged cholestatic hepatitis with ductopenia after long-term
545
Section IV. Toxin Mediated Liver Injury
338.
339. 340.
341.
342.
343.
344.
345.
346.
347.
348.
349.
350.
351.
352.
353.
354. 355. 356.
546
ciprofloxacin therapy for Crohn’s disease. J Hepatol 2002;37:696–699. Airey K, Koller E. Acute hepatitis associated with levofloxacin in a patient with renal insufficiency. Can Med Assoc J 2003;69:755. Blum A. Ofloxacin-induced acute severe hepatitis. South Med J 1991;84:1158. Munoz SJ, Martinez-Hernandez A, Maddrey WC. Intrahepatic cholestasis and phospholipidosis associated with the use of trimethoprim–sulfamethoxazole. Hepatology 1990;12:342–347. Mainra RR, Card SE. Trimethoprim–sulfamethoxazoleassociated hepatotoxicity – part of a hypersensitivity syndrome. Can J Clin Pharmacol 2003;10:175–178. Zaman F, Ye G, Abreo KD, Latif S, Zibari GB. Successful orthotopic liver transplantation after trimethoprim–sulfamethoxazole associated fulminant liver failure. Clin Transplant 2003;17:461–464. Ilario MJ, Ruiz JE, Axiotis CA. Acute fulminant hepatic failure in a woman treated with phenytoin and trimethoprim–sulfamethoxazole. Arch Pathol Lab Med 2000;124:1800–1803. Jaffe HS, Abrams DI, Ammann AJ, Lewis BJ, Golden JA. Complications of co-trimoxazole in treatment of AIDSassociated Pneumocystis carinii pneumonia in homosexual men. Lancet 1983;2:1109–1111. Johnson MP, Goodwin SD, Shands JW Jr. Trimethoprimsulfamethoxazole anaphylactoid reactions in patients with AIDS: case reports and literature review. Pharmacotherapy 1990;10:413–416. Floris-Moore MA, Amodio-Groton MI, Catalano MT. Adverse reactions to trimethoprim/sulfamethoxazole in AIDS. Ann Pharmacother 2003;37:1810–1813. Gompels MM, Simpson N, Snow M, Spickett G, Ong E. Desensitization to co-trimoxazole (trimethoprim–sulphamethoxazole) in HIV-infected patients: is patch testing a useful predictor of reaction? J Infect 1999;38:111–115. Kaufmann GR, Wenk M, Taeschner W, et al. Nacetyltransferase 2 polymorphism in patients infected with human immunodeficiency virus. Clin Pharmacol Ther 1996;60:62–67. Winter HR, Trapnell CB, Slattery JT, et al. The effect of clarithromycin, fluconazole, and rifabutin on sulfamethoxazole hydroxylamine formation in individuals with human immunodeficiency virus infection (AACTG 283). Clin Pharmacol Ther 2004;76:313–322. Chintu C, Bhat GJ, Walker AS, et al. Co-trimoxazole as prophylaxis against opportunistic infections in HIV-infected Zambian children (CHAP). A double-blind randomised placebo-controlled trial. Lancet 2004;364:1865–1871. Mermin J, Lule J, Ekwaru JP, et al. Effect of co-trimoxazole prophylaxis on morbidity, mortality, CD4-cell count, and viral load in HIV infection in rural Uganda. Lancet 2004;364:1428–1434. Grimwade K, Swingler G. Cotrimoxazole prophylaxis for opportunistic infections in adults with HIV. Cochrane Database Syst Rev 2003;CD003108. Freneaux E, Labbe G, Letteron P, et al. Inhibition of the mitochondrial oxidation of fatty acids by tetracycline in mice and in man: possible role in microvesicular steatosis induced by this antibiotic. Hepatology 1988;8:1056–1062. Bjornsson E, Lindberg J, Olsson R. Liver reactions to oral lowdose tetracyclines. Scand J Gastroenterol 1997;32:390–395. Lienart F, Morissens M, Jacobs P, Ducobu J. Doxycycline and hepatotoxicity. Acta Clin Belg 1992;47:205–208. Hunt CM, Washington K. Tetracycline-induced bile duct paucity and prolonged cholestasis. Gastroenterology 1994;107:1844–1847.
357. Malcolm A, Heap TR, Eckstein RP, Lunzer MR. Minocyclineinduced liver injury. Am J Gastroenterol 1996;91:1641–1643. 358. Gough A, Chapman S, Wagstaff K, Emery P, Elias E. Minocycline induced autoimmune hepatitis and systemic lupus erythematosus-like syndrome. Br Med J 1996;312: 169–172. 359. Teitelbaum JE, Perez–Atayde AR, Cohen M, Bousvaros A, Jonas MM. Minocycline-related autoimmune hepatitis: case series and literature review. Arch Pediatr Adolesc Med 1998;152:1132–1136. 360. Sharp JR, Ishak KG, Zimmerman HJ. Chronic active hepatitis and severe hepatic necrosis associated with nitrofurantoin. Ann Intern Med 1980;92:14–19. 361. Karpman E, Kurzrock EA. Adverse reactions of nitrofurantoin, trimethoprim and sulfamethoxazole in children. J Urol 2004;172:448–453. 362. Volbeda F, Jonker AM, Vecht J, Groeneveld PH. Liver cirrhosis due to chronic use of nitrofurantoin. Ned Tijdschr Geneeskd 2004;148:235–238. 363. Amit G, Cohen P, Ackerman Z. Nitrofurantoin-induced chronic active hepatitis. Israel Med Assoc J 2002;4:184–186. 364. Galan MV, Potts JA, Silverman AL, Gordon SC. Hepatitis in a United States tertiary referral center. J Clin Gastroenterol 2005;39:64–67. 365. Rached-Mohassel MA, Hanjani AA, Nik-Akhtar B, Chariat C, Tabibi M. Cholestatic hepatitis induced by the prolonged administration of nalidixic acid (case report). Acta Med Iran 1974;17:47–56. 366. Yanagihara K, Okada M, Fukuda Y, et al. Efficacy of quinupristin-dalfopristin against methicillin-resistant Staphylococcus aureus and vancomycin-insensitive S. aureus in a model of hematogenous pulmonary infection. Chemotherapy 2004;50:260–264. 367. Linden PK, Bompart F, Gray S, Talbot GH. Hyperbilirubinemia during quinupristin–dalfopristin therapy in liver transplant recipients: correlation with available liver biopsy results. Pharmacotherapy 2001;21:661–668. 368. Raad I, Hachem R, Hanna H. Relationship between myalgias/arthralgias occurring in patients receiving quinupristin/dalfopristin and biliary dysfunction. J Antimicrob Chemother 2004;53:1105–1108. 369. Gobernado M. Fosfomycin. Rev Esp Quimioter 2003;16: 15–40. 370. Mayama T, Yokota M, Shimatani I, Ohyagi H. Analysis of oral fosfomycin calcium (Fosmicin) side-effects after marketing. Int J Clin Pharmacol Ther Toxicol 1993;31:77–82. 371. Durupt S, Josserand RN, Sibille M, Durieu I. Acute, recurrent fosfomycin-induced liver toxicity in an adult patient with cystic fibrosis. Scand J Infect Dis 2001;33:391–392. 372. Marissal JP, Selke B, Amarenco P. Economic assessment of the secondary prevention of ischaemic stroke with dipyridamole plus aspirin (Aggrenox/Asasantin) in France. Pharmacoeconomics 2004;22:661–670. 373. Lenz TL, Hilleman DE. Aggrenox: a fixed-dose combination of aspirin and dipyridamole. Ann Pharmacother 2000;34:1283–1290. 374. Gryglewski RJ, Korbut R, Swies J, et al. Thrombolytic action of ticlopidine: possible mechanisms. Eur J Pharmacol 1996;308:61–67. 375. Noble S, Goa KL. Ticlopidine. A review of its pharmacology, clinical efficacy and tolerability in the prevention of cerebral ischaemia and stroke. Drugs Aging 1996;8:214–232. 376. Sondag D, Bader R, Claude P, Schreiber M. Hepatitis due to ticlopidine: a new case. Ann Gastroenterol Hepatol (Paris) 1993;29:40–41. 377. Gandolfi A, Mengoli M, Rota E, et al. Ticlopidine-induced acute cholestatic hepatitis. A case report. Recenti Prog Med 2004;95:96–99.
Chapter 26 DRUG-INDUCED LIVER INJURY
378. Waguri N, Yamamoto M. Intrahepatic cholestasis and pure red cell aplasia associated with ticlopidine. Eur J Intern Med 2003;14:192–195. 379. Skurnik YD, Tcherniak A, Edlan K, Sthoeger Z. Ticlopidineinduced cholestatic hepatitis. Ann Pharmacother 2003;37:371–375. 380. Varvello L, Arno C, Coggiola M, Ramassotto E, Pascale C. Cholestasis from ticlopidine: two clinical cases. Ann Ital Med Int 2002;17:252–255. 381. Pizarro AE, Andrade RJ, Garcia-Cortes M, et al. Acute hepatitis due to ticlopidine. A report of 12 cases and review of the literature. Rev Neurol 2001;33:1014–1020. 382. Rajagopal V, Bhatt DL. Controversies of oral antiplatelet therapy in acute coronary syndromes and percutaneous coronary intervention. Semin Thromb Hemost 2004;30:649–655. 383. Zeolla MM, Carson JJ. Successful use of clopidogrel for cerebrovascular accident in a patient with suspected ticlopidine-induced hepatotoxicity. Ann Pharmacother 1999;33:939–941. 384. Savi P, Herbert JM, Pflieger AM, et al. Importance of hepatic metabolism in the antiaggregating activity of the thienopyridine clopidogrel. Biochem Pharmacol 1992;44:527–532. 385. Lau WC, Gurbel PA, Watkins PB, et al. Contribution of hepatic cytochrome P450 3A4 metabolic activity to the phenomenon of clopidogrel resistance. Circulation 2004;109:166–171. 386. Willens HJ. Clopidogrel-induced mixed hepatocellular and cholestatic liver injury. Am J Ther 2000;7:317–318. 387. Ehrenforth S, Schenk JF, Scharrer I. Liver damage induced by coumarin anticoagulants. Semin Thromb Hemost 1999;25:79–83. 388. Hohler T, Schnutgen M, Helmreich-Becker I, Mayet WJ, Mayer zum Buschenfelde KH. Drug-induced hepatitis: a rare complication of oral anticoagulants. J Hepatol 1994;21:447–449. 389. Schimanski CC, Burg J, Mohler M, et al. Phenprocoumoninduced liver disease ranges from mild acute hepatitis to (sub-) acute liver failure. J Hepatol 2004;41:67–74. 390. Friedel HA, Balfour JA. Tinzaparin. A review of its pharmacology and clinical potential in the prevention and treatment of thromboembolic disorders. Drugs 1994;48:638–660. 391. Valois M, Cooper MA, Shear NH. Clinical pharmacology consultations: consultation requests may be misleading – an organized approach to drug-induced hepatitis. Can J Clin Pharmacol 2003;10:59–62. 391a. Greinacher A. Lepiradin: a bivalent direct thrombin inhibitor for anticoagulation theraphy. Expert Rev Cardiovasc Ther 2004;2:339–357. 392. El-Khatib AS, Mansour MA. Prior treatment with captopril attenuates carbon tetrachloride-induced liver injury in mice. Res Commun Mol Pathol Pharmacol 2001;110:3–16. 393. Jonsson JR, Clouston AD, Ando Y, et al. Angiotensinconverting enzyme inhibition attenuates the progression of rat hepatic fibrosis. Gastroenterology 2001;121:148–155. 394. Tuncer I, Ozbek H, Ugras S, Bayram I. Anti-fibrogenic effects of captopril and candesartan cilexetil on the hepatic fibrosis development in rat. The effect of AT1-R blocker on the hepatic fibrosis. Exp Toxicol Pathol 2003;55:159–166. 395. Schattner A, Kozak N, Friedman J. Captopril-induced jaundice: report of 2 cases and a review of 13 additional reports in the literature. Am J Med Sci 2001;322:236–240. 396. Rahmat J, Gelfand RL, Gelfand MC, et al. Captoprilassociated cholestatic jaundice. Ann Intern Med 1985;102:56–58. 397. Larrey D, Babany G, Bernuau J, et al. Fulminant hepatitis after lisinopril administration. Gastroenterology 1990;99:1832–1833.
398. Nunes AC, Amaro P, Macoas F, et al. Fosinopril-induced prolonged cholestatic jaundice and pruritus: first case report. Eur J Gastroenterol Hepatol 2001;13:279–282. 399. Yeung E, Wong FS, Wanless IR, et al. Ramipril-associated hepatotoxicity. Arch Pathol Lab Med 2003;127:1493–1497. 400. Tabak F, Mert A, Ozaras R, et al. Losartan-induced hepatic injury. J Clin Gastroenterol 2002;34:585–586. 401. Nygaard B, Strandgaard S. Marked hepatotoxicity associated with losartan treatment. Blood Press 1996;5:190–191. 402. Basile G, Villari D, Gangemi S, et al. Candesartan cilexetilinduced severe hepatotoxicity. J Clin Gastroenterol 2003;36:273–275. 403. Andrade RJ, Lucena MI, Fernandez MC, et al. Cholestatic hepatitis related to use of irbesartan: a case report and a literature review of angiotensin II antagonist-associated hepatotoxicity. Eur J Gastroenterol Hepatol 2002;14:887–890. 404. Kiykim A, Altintas E, Sezgin O, et al. Valsartan-induced hepatotoxicity in a HBs-Ag-positive patient. Am J Gastroenterol 2003;98:507. 405. Lewis JH, Ranard RC, Caruso A, et al. Amiodarone hepatotoxicity: prevalence and clinicopathologic correlations among 104 patients. Hepatology 1989;9:679–685. 406. Hilleman D, Miller MA, Parker R, Doering P, Pieper JA. Optimal management of amiodarone therapy: efficacy and side effects. Pharmacotherapy 1998;18:138S–145S. 407. Agozzino F, Picca M, Pelosi G. Acute hepatitis complicating intravenous amiodarone treatment. Ital Heart J 2002;3:686–688. 408. Simon JB, Manley PN, Brien JF, Armstrong PW. Amiodarone hepatotoxicity simulating alcoholic liver disease. N Engl J Med 1984;311:167–172. 409. Singhal A, Ghosh P, Khan SA. Low dose amiodarone causing pseudo-alcoholic cirrhosis. Age Aging 2003;32:224–225. 410. Pollak PT, Shafer SL. Use of population modeling to define rational monitoring of amiodarone hepatic effects. Clin Pharmacol Ther 2004;75:342–351. 411. Polasek TM, Elliot DJ, Lewis BC, Miners JO. Mechanismbased inactivation of human cytochrome P4502C8 by drugs in vitro. J Pharmacol Exp Ther 2004;311:996–1007. 412. Guigui B, Perrot S, Berry JP, et al. Amiodarone-induced hepatic phospholipidosis. A morphological alteration independent of pseudoalcoholic liver disease. Hepatology 1988;8:1063–1068. 413. Hirakawa K, Abe K, Ayabe Y, Nishimura M. Analysis of increased hepatic density during chronic amiodarone therapy. Nippon Igaku Hoshasen Gakkai Zasshi 2003;63:221–224. 414. Berson A, De Beco, V, Letteron P, et al. Steatohepatitisinducing drugs cause mitochondrial dysfunction and lipid peroxidation in rat hepatocytes. Gastroenterology 1998;114:764–774. 415. Kodawara T, Masuda S, Wakasugi H, et al. Organic anion transporter oatp2-mediated interaction between digoxin and amiodarone in the rat liver. Pharm Res 2002;19:738–743. 416. Geltner D, Chajek T, Rubinger D, Levij IS. Quinidine hypersensitivity and liver involvement. A survey of 32 patients. Gastroenterology 1976;70:650–652. 417. Galetin A, Clarke SE, Houston JB. Quinidine and haloperidol as modifiers of CYP3A4 activity: multisite kinetic model approach. Drug Metab Dispos 2002;30:1512–1522. 418. Masubuchi Y, Ose A, Horie T. Diclofenac-induced inactivation of CYP3A4 and its stimulation by quinidine. Drug Metab Dispos 2002;30:1143–1148. 419. Cocozzella D, Curciarello J, Corallini O, et al. Propafenone hepatotoxicity: report of two new cases. Dig Dis Sci 2003;48:354–357. 420. Clark JA, Zimmerman HJ, Tanner LA. Labetalol hepatotoxicity. Ann Intern Med 1990;113:210–213. 421. Chon EM, Middleton RK. Labetalol hepatotoxicity. Ann Pharmacother 1992;26:344–345.
547
Section IV. Toxin Mediated Liver Injury
422. Larrey D, Henrion J, Heller F, et al. Metoprolol-induced hepatitis: rechallenge and drug oxidation phenotyping. Ann Intern Med 1988;108:67–68. 422a. Hagmeyer KO, Stein J. Hepatotoxicity associated with carvedilol. Ann Pharmacother 2001;35:1344–1346. 423. Shaw DR, Misan GM, Johnson RD. Nifedipine hepatitis. Aust NZ J Med 1987;17:447–448. 424. Sawaya GF, Robertson PA. Hepatotoxicity with the administration of nifedipine for treatment of preterm labor. Am J Obstet Gynecol 1992;167:512–513. 425. Traverse JH, Swenson LJ, McBride JW. Acute hepatic injury after treatment with diltiazem. Am Heart J 1994;127:1636–1639. 426. Toft E, Vyberg M, Therkelsen K. Diltiazem-induced granulomatous hepatitis. Histopathology 1991;18:474–475. 427. Burgunder JM, Abernethy DR, Lauterburg BH. Liver injury due to verapamil. Hepatogastroenterology 1988;35:169–170. 428. Farmer JA, Torre-Amione G. Comparative tolerability of the HMG-CoA reductase inhibitors. Drug Saf 2000;23:197–213. 429. Shepherd J, Hunninghake DB, Stein EA, et al. Safety of rosuvastatin. Am J Cardiol 2004;94:882–888. 430. de DS, Spinler SA, Miller K, Peterson AM. Statins and liver toxicity: a meta-analysis. Pharmacotherapy 2004;24:584–591. 431. Wiegman A, Hutten BA, de Groot E, et al. Efficacy and safety of statin therapy in children with familial hypercholesterolemia: a randomized controlled trial. JAMA 2004;292:331–337. 432. Kiortsis DN, Nikas S, Hatzidimou K, Tsianos E, Elisaf MS. Lipid-lowering drugs and serum liver enzymes: the effects of body weight and baseline enzyme levels. Fund Clin Pharmacol 2003;17:491–494. 433. Lipka L, Sager P, Strony J, et al. Efficacy and safety of coadministration of ezetimibe and statins in elderly patients with primary hypercholesterolaemia. Drugs Aging 2004;21:1025–1032. 434. Chalasani N, Aljadhey H, Kesterson J, Murray MD, Hall SD. Patients with elevated liver enzymes are not at higher risk for statin hepatotoxicity. Gastroenterology 2004;126:1287–1292. 435. Pasternak RC, Smith SC Jr, Bairey-Merz CN, et al. ACC/AHA/NHLBI clinical advisory on the use and safety of statins. J Am Coll Cardiol 2002;40:567–572. 436. Roberts WC. Safety of fenofibrate – US and worldwide experience. Cardiology 1989;76:169–179. 437. Etchason JA, Miller TD, Squires RW, et al. Niacin-induced hepatitis: a potential side effect with low-dose time-release niacin. Mayo Clin Proc 1991;66:23–28. 438. Dalton TA, Berry RS. Hepatotoxicity associated with sustainedrelease niacin. Am J Med 1992;93:102–104. 439. Rader JI, Calvert RJ, Hathcock JN. Hepatic toxicity of unmodified and time-release preparations of niacin. Am J Med 1992;92:77–81. 440. Patel SD, Taylor HC. Intrahepatic cholestasis during nicotinic acid therapy. Cleveland Clin J Med 1994;61:70–75. 441. Neuberger J, Williams R. Immune mechanisms in tienilic acid associated hepatotoxicity. Gut 1989;30:515–519. 442. Masubuchi Y, Horie T. Mechanism-based inactivation of cytochrome P450s 1A2 and 3A4 by dihydralazine in human liver microsomes. Chem Res Toxicol 1999;12:1028–1032. 443. Pariente EA, Pessayre D, Bernuau J, Degott C, Benhamou JP. Dihydralazine hepatitis: report of a case and review of the literature. Digestion 1983;27:47–52. 444. Itoh S, Yamaba Y, Ichinoe A, Tsukada Y. Hydralazine-induced liver injury. Dig Dis Sci 1980;25:884–887. 445. Rice D, Burdick CO. Granulomatous hepatitis from hydralazine therapy. Arch Intern Med 1983;143:1077. 446. Clozel M, Breu V, Gray GA, et al. Pharmacological characterization of bosentan, a new potent orally active nonpeptide endothelin receptor antagonist. J Pharmacol Exp Ther 1994;270:228–235.
548
447. Kenyon KW, Nappi JM. Bosentan for the treatment of pulmonary arterial hypertension. Ann Pharmacother 2003;37:1055–1062. 448. Dingemanse J, van Giersbergen PL. Clinical pharmacology of bosentan, a dual endothelin receptor antagonist. Clin Pharmacokinet 2004;43:1089–1115. 449. Fattinger K, Funk C, Pantze M, et al. The endothelin antagonist bosentan inhibits the canalicular bile salt export pump: a potential mechanism for hepatic adverse reactions. Clin Pharmacol Ther 2001;69:223–231. 450. Koeppel TA, Kraus T, Thies JC, et al. Effects of mixed ETA and ETB-receptor antagonist (Ro-47-0203) on hepatic microcirculation after warm ischemia. Dig Dis Sci 1997;42:1316–1321. 451. Sogni P, Moreau R, Gomola A, et al. Beneficial hemodynamic effects of bosentan, a mixed ET(A) and ET(B) receptor antagonist, in portal hypertensive rats. Hepatology 1998;28:655–659. 452. Kuntzen C, Gulberg V, Gerbes AL. Use of a mixed endothelin receptor antagonist in portopulmonary hypertension: A safe and effective therapy? Gastroenterology 2005;128:164–168. 453. Hinterhuber L, Graziadei IW, Kahler CM, Jaschke W, Vogel W. Endothelin-receptor antagonist treatment of portopulmonary hypertension. Clin Gastroenterol Hepatol 2004;2:1039–1042. 454. Molnar C, Alber H, Colleselli D, Vogel W, Kahler CM. Successful switch from inhalative iloprost to oral bosentan in portopulmonary hypertension associated with liver cirrhosis. Wien Klin Wschr 2004;116:627–630. 455. Lewis JH. NSAID induced hepatotoxicity. Clin Liver Dis 1998;2:543. 456. Carson JL, Strom BL, Duff A, Gupta A, Das K. Safety of nonsteroidal anti-inflammatory drugs with respect to acute liver disease. Arch Intern Med 1993;153:1331–1336. 457. Garcia Rodriguez LA, Perez GS, Walker AM, Lueck L. The role of non-steroidal anti-inflammatory drugs in acute liver injury. Br Med J 1992;305:865–868. 458. Bush TM, Shlotzhauer TL, Imai K. Nonsteroidal antiinflammatory drugs. Proposed guidelines for monitoring toxicity. West J Med 1991;155:39–42. 459. Adebajo AO, Eastmond CJ. Hepatotoxicity to several nonsteroidal anti-inflammatory drugs with diclofenac induced histological changes. Clin Rheumatol 1992;11:120–121. 460. Zimmerman HJ. Effects of aspirin and acetaminophen on the liver. Arch Intern Med 1981;141:333–342. 461. Seaman WE, Plotz PH. Effect of aspirin on liver tests in patients with RA or SLE and in normal volunteers. Arthritis Rheum 1976;19:155–160. 462. Martens ME, Chang CH, Lee CP. Reye’s syndrome: mitochondrial swelling and Ca2+ release induced by Reye’s plasma, allantoin, and salicylate. Arch Biochem Biophys 1986;244:773–786. 463. Deschamps D, Fisch C, Fromenty B, et al. Inhibition by salicylic acid of the activation and thus oxidation of long chain fatty acids. Possible role in the development of Reye’s syndrome. J Pharmacol Exp Ther 1991;259:894–904. 464. Glasgow JF, Moore R. Reye’s syndrome 30 years on. Br Med J 1993;307:950–951. 465. Ulshen MH, Grand RJ, Crain JD, Gelfand EW. Hepatoxicity with encephalopathy associated with aspirin therapy in rheumatoid arthritis. J Pediatr 1978;93:1034–1037. 466. Scully RE, Galdabini JJ, Mcneely BU. Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Case 22-1977. N Engl J Med 1977;296:1279–1287. 467. Cook DJ, Achong MR, Murphy FR. Three cases of diflunisal hypersensitivity. Can Med Assoc J 1988;138:1029–1030. 468. Warren JS. Diflunisal-induced cholestatic jaundice. Br Med J 1978;2:736–737.
Chapter 26 DRUG-INDUCED LIVER INJURY
469. Cuthbert MF. Adverse reactions to non-steroidal antirheumatic drugs. Curr Med Res Opin 1974;2:600–610. 470. Kelsey WM, Scharyj M. Fatal hepatitis probably due to indomethacin. JAMA 1967;199:586–587. 471. Jacobs JS. Sudden death in arthritic children receiving large doses of indomethacin. JAMA 1967;199:932–934. 472. Tarazi EM, Harter JG, Zimmerman HJ, Ishak KG, Eaton RA. Sulindac-associated hepatic injury: analysis of 91 cases reported to the Food and Drug Administration. Gastroenterology 1993;104:569–574. 473. McIndoe GA, Menzies KW, Reddy J. Sulindac (Clinoril) and cholestatic jaundice. NZ Med J 1981;94:430–431. 474. Dhand AK, LaBrecque DR, Metzger J. Sulindac (Clinoril) hepatitis. Gastroenterology 1981;80:585–586. 475. Whittaker SJ, Amar JN, Wanless IR, Heathcote J. Sulindac hepatotoxicity. Gut 1982;23:875–877. 476. Klein SM, Khan MA. Hepatitis, toxic epidermal necrolysis and pancreatitis in association with sulindac therapy. J Rheumatol 1983;10:512–513. 477. Banks AT, Zimmerman HJ, Ishak KG, Harter JG. Diclofenacassociated hepatotoxicity: analysis of 180 cases reported to the Food and Drug Administration as adverse reactions. Hepatology 1995;22:820–827. 478. Bhogaraju A, Nazeer S, Al-Baghdadi Y, et al. Diclofenacassociated hepatitis. South Med J 1999;92:711–713. 479. Scully LJ, Clarke D, Barr RJ. Diclofenac induced hepatitis. 3 cases with features of autoimmune chronic active hepatitis. Dig Dis Sci 1993;38:744–751. 480. West SG. Methotrexate hepatotoxicity. Rheum Dis Clin North Am 1997;23:883–915. 481. Bridges SL Jr, Alarcon GS, Koopman WJ. Methotrexateinduced liver abnormalities in rheumatoid arthritis. J Rheumatol 1989;16:1180–1183. 482. Rau R, Karger T, Herborn G, Frenzel H. Liver biopsy findings in patients with rheumatoid arthritis undergoing longterm treatment with methotrexate. J Rheumatol 1989;16:489–493. 483. Zimmerman HJ. Hepatotoxic effects of oncotherapeutic agents. Prog Liver Dis 1986;8:621–642. 484. Perry MC. Chemotherapeutic agents and hepatotoxicity. Semin Oncol 1992;19:551–565. 485. McDonald GB, Tirumali N. Intestinal and liver toxicity of antineoplastic drugs. West J Med 1984;140:250–259. 486. Shorey J, Schenker S, Suki WN, Combes B. Hepatotoxicity of mercaptopurine. Arch Intern Med 1968;122:54–58. 487. Einhorn M, Davidsohn I. Hepatotoxicity of mercaptopurine. JAMA 1964;188:802–806. 488. Topley JM, Benson J, Squier MV, Chessells JM. Hepatotoxicity in the treatment of acute lymphoblastic leukaemia. Med Pediatr Oncol 1979;7:393–399. 489. Greaves MW, Dawber R. Azathioprine in psoriasis. Br Med J 1970;2:237–238. 490. Horsmans Y, Rahier J, Geubel AP. Reversible cholestasis with bile duct injury following azathioprine therapy. A case report. Liver 1991;11:89–93. 491. Knowles SR, Gupta AK, Shear NH, Sauder D. Azathioprine hypersensitivity-like reactions – a case report and a review of the literature. Clin Exp Dermatol 1995;20:353–356. 492. Malekzadeh MH, Grushkin CM, Wright HT Jr, Fine RN. Hepatic dysfunction after renal transplantation in children. J Pediatr 1972;81:279–285. 493. Degott C, Rueff B, Kreis H, et al. Peliosis hepatis in recipients of renal transplants. Gut 1978;19:748–753. 494. Mion F, Napoleon B, Berger F, et al. Azathioprine induced liver disease: nodular regenerative hyperplasia of the liver and perivenous fibrosis in a patient treated for multiple sclerosis. Gut 1991;32:715–717. 495. Watanabe A, Obata T, Nagashima H, Sakagami K, Orita K. Nonicteric liver damage with a gamma-glutamyl transpeptidase
496.
497.
498.
499.
500.
501.
502.
503.
504.
505.
506. 507.
508.
509.
510.
511.
512. 513.
514.
level of 5,609 units/l in a renal-transplant recipient receiving azathioprine. Acta Med Okayama 1984;38: 533–539. Haboubi NY, Ali HH, Whitwell HL, Ackrill P. Role of endothelial cell injury in the spectrum of azathioprine-induced liver disease after renal transplant: light microscopy and ultrastructural observations. Am J Gastroenterol 1988;83:256–261. Marubbio AT, Danielson B. Hepatic veno-occlusive disease in a renal transplant patient receiving azathioprine. Gastroenterology 1975;69:739–743. Liano F, Moreno A, Matesanz R, et al. Veno-occlusive hepatic disease of the liver in renal transplantation: is azathioprine the cause? Nephron 1989;51:509–516. Shepherd PC, Fooks J, Gray R, Allan NC. Thioguanine used in maintenance therapy of chronic myeloid leukaemia causes noncirrhotic portal hypertension. Results from MRC CML. II. Trial comparing busulphan with busulphan and thioguanine. Br J Haematol 1991;79:185–192. Shepherd P, Harrison DJ. Idiopathic portal hypertension associated with cytotoxic drugs. J Clin Pathol 1990;43:206–210. Satti MB, Weinbren K, Gordon-Smith EC. 6-thioguanine as a cause of toxic veno-occlusive disease of the liver. J Clin Pathol 1982;35:1086–1091. Griner PF, Elbadawi A, Packman CH. Veno-occlusive disease of the liver after chemotherapy of acute leukemia. Report of two cases. Ann Intern Med 1976;85:578–582. Ambiru S, Miyazaki M, Ito H, et al. Intraportal infusion of 5FU and lipiodol–aclarubicin after hepatic resection for colorectal liver metastasis. Nippon Geka Gakkai Zasshi 1995;96:145–152. Klotz HP, Weder W, Largiader F. Local and systemic toxicity of intra-hepato–arterial chemotherapy for treatment of unresectable liver metastases of colorectal cancer with 5fluorouracil and high dose leucovorin. Helv Chir Acta 1993;60:283–286. Hohn D, Melnick J, Stagg R, et al. Biliary sclerosis in patients receiving hepatic arterial infusions of floxuridine. J Clin Oncol 1985;3:98–102. Bolton JS, Bowen JC. Biliary sclerosis associated with hepatic artery infusion of floxuridine. Surgery 1986;99: 119–122. Remick SC, Benson AB III, Weese JL, et al. Phase I trial of hepatic artery infusion of 5-iodo-2¢-deoxyuridine and 5fluorouracil in patients with advanced hepatic malignancy: biochemically based combination chemotherapy. Cancer Res 1989;49:6437–6442. Rougier P, Laplanche A, Huguier M, et al. Hepatic arterial infusion of floxuridine in patients with liver metastases from colorectal carcinoma: long-term results of a prospective randomized trial. J Clin Oncol 1992;10:1112–1118. Pettavel J, Gardiol D, Bergier N, Schnyder P. Fatal liver cirrhosis associated with long-term arterial infusion of floxuridine. Lancet 1986;2:1162–1163. Faggioli P, De Paschale M, Tocci A, et al. Acute hepatic toxicity during cyclic chemotherapy in non Hodgkin’s lymphoma. Haematologica 1997;82:38–42. Goodell B, Leventhal B, Henderson E. Cytosine arabinoside in acute granulocytic leukemia. Clin Pharmacol Ther 1971;12:599–606. Traggis DG, Dohlwitz A, Das L, et al. Cytosine arabinoside in acute leukemia of childhood. Cancer 1971;28:815–818. Herzig RH, Wolff SN, Lazarus HM, et al. High-dose cytosine arabinoside therapy for refractory leukemia. Blood 1983;62:361–369. Haskell CM, Canellos GP, Leventhal BG, et al. L-asparaginase: therapeutic and toxic effects in patients with neoplastic disease. N Engl J Med 1969;281:1028–1034.
549
Section IV. Toxin Mediated Liver Injury
515. Shanholtz C. Acute life-threatening toxicity of cancer treatment. Crit Care Clin 2001;17:483–502. 516. Weiss RB, Bruno S. Hypersensitivity reactions to cancer chemotherapeutic agents. Ann Intern Med 1981;94:66–72. 517. Biggs JC, Chesterman CN, Holliday J. L-asparaginase – clinical experience in leukaemia, lymphoma and carcinoma. Aust NZ J Med 1971;1:1–7. 518. Ansfield FJ. Clinical studies with mithramycin. Oncology 1969;23:283–288. 519. Kennedy BJ. Metabolic and toxic effects of mithramycin during tumor therapy. Am J Med 1970;49:494–503. 520. Aviles A, Herrera J, Ramos E, et al. Hepatic injury during doxorubicin therapy. Arch Pathol Lab Med 1984;108:912–913. 521. Kun LE, Camitta BM. Hepatopathy following irradiation and adriamycin. Cancer 1978;42:81–84. 522. Pritchard J, Raine J, Wallendszus K. Hepatotoxicity of actinomycin-D. Lancet 1989;1:168. 523. D’Angio GJ. Hepatotoxicity with actinomycin D. Lancet 1987;2:104. 524. Jayabose S, Shende A, Lanzkowsky P. Hepatotoxicity of chemotherapy following nephrectomy and radiation therapy for right-sided Wilms’ tumor. J Pediatr 1976;88:898. 525. Czauderna P, Katski K, Kowalczyk J, et al. Venoocclusive liver disease (VOD) as a complication of Wilms’ tumour management in the series of consecutive 206 patients. Eur J Pediatr Surg 2000;10:300–303. 526. Hazar V, Kutluk T, Akyuz C, et al. Veno-occlusive disease-like hepatotoxicity in two children receiving chemotherapy for Wilms’ tumor and clear cell sarcoma of kidney. Pediatr Hematol Oncol 1998;15:85–89. 527. Bisogno G, de KJ, Weirich A, Masiero L, et al. Veno-occlusive disease of the liver in children treated for Wilms’ tumor. Med Pediatr Oncol 1997;29:245–251. 528. Flentje M, Weirich A, Potter R, Ludwig R. Hepatotoxicity in irradiated nephroblastoma patients during postoperative treatment according to SIOP9/GPOH. Radiother Oncol 1994;31:222–228. 529. Lorber MI, Van Buren CT, Flechner SM, Williams C, Kahan BD. Hepatobiliary and pancreatic complications of cyclosporine therapy in 466 renal transplant recipients. Transplantation 1987;43:35–40. 530. Kassianides C, Nussenblatt R, Palestine AG, Mellow SD, Hoofnagle JH. Liver injury from cyclosporine A. Dig Dis Sci 1990;35:693–697. 531. el Saghir NS, Hawkins KA. Hepatotoxicity following vincristine therapy. Cancer 1984;54:2006–2008. 532. Tran A, Housset C, Boboc B, et al. Etoposide (VP 16-213) induced hepatitis. Report of three cases following standarddose treatments. J Hepatol 1991;12:36–39. 533. Paschke R, Worst P, Brust J, Queisser W. Hepatotoxicity with etoposide–ifosfamide combination therapy. Onkologie 1988;11:273–275. 534. Menard DB, Gisselbrecht C, Marty M, Reyes F, Dhumeaux D. Antineoplastic agents and the liver. Gastroenterology 1980;78:142–164. 535. Cleland BD, Pokorny CS. Cyclophosphamide related hepatotoxicity. Aust NZ J Med 1993;23:408. 536. Bacon AM, Rosenberg SA. Cyclophosphamide hepatotoxicity in a patient with systemic lupus erythematosus. Ann Intern Med 1982;97:62–63. 537. Umeda K, Lin YW, Watanabe K, et al. Hematopoietic stem cell transplantation with busulfanthiotepa-cyclophosphamide conditioning for pediatric patients with high-risk acute lymphoblastic leukemia. Rinsho Ketsueki 2001;42:685–690. 538. Andersson BS, Gajewski J, Donato M, et al. Allogeneic stem cell transplantation (BMT) for AML and MDS following i.v. busulfan and cyclophosphamide (i.v. BuCy). Bone Marrow Transplant 2000;25 Suppl 2:S35–S38.
550
539. Worth L, Tran H, Petropoulos D, et al. Hematopoietic stem cell transplantation for childhood myeloid malignancies after high-dose thiotepa, busulfan and cyclophosphamide. Bone Marrow Transplant 1999;24:947–952. 540. Lee JL, Gooley T, Bensinger W, Schiffman K, McDonald GB. Veno-occlusive disease of the liver after busulfan, melphalan, and thiotepa conditioning therapy: incidence, risk factors, and outcome. Biol Blood Marrow Transplant 1999;5: 306–315. 541. Modzelewski JR Jr, Daeschner C, Joshi VV, Mullick FG, Ishak KG. Veno-occlusive disease of the liver induced by low-dose cyclophosphamide. Mod Pathol 1994;7:967–972. 542. Pichon N, Bette-Gratien M, Cessot F, et al. Acute cholestatic hepatitis due to chlorambucil. Gastroenterol Clin Biol 2001;25:202–203. 543. Phillips GL, Fay JW, Herzig GP, et al. Intensive 1,3-bis(2chloroethyl)-1-nitrosourea (BCNU), NSC #4366650 and cryopreserved autologous marrow transplantation for refractory cancer. A phase I–II study. Cancer 1983;52:1792–1802. 544. Lokich JJ, Drum DE, Kaplan W. Hepatic toxicity of nitrosourea analogues. Clin Pharmacol Ther 1974;16:363–367. 545. Girgin F, Tuzun S, Demir A, et al. Cytoprotective effects of trimetazidine in carmustine cholestasis. Exp Toxicol Pathol 1999;51:326–329. 546. Quinio P, Bouche O, Lambolais C, et al. Fatal hepatic toxicity of DTIC. A new case. Intens Care Med 1997;23:1099. 547. Voigt H, Caselitz J, Janner M. Veno-occlusive syndrome with acute liver dystrophy following decarbazine therapy of malignant melanoma (author’s transl). Klin Wschr 1981;59:229–236. 548. Asbury RF, Rosenthal SN, Descalzi ME, Ratcliffe RL, Arseneau JC. Hepatic veno-occlusive disease due to DTIC. Cancer 1980;45:2670–2674. 549. Vial T, Descotes J. Clinical toxicity of the interferons. Drug Saf 1994;10:115–150. 550. Manns MP, McHutchison JG, Gordon SC, et al. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 2001;358:958–965. 551. Jones GJ, Itri LM. Safety and tolerance of recombinant interferon alfa-2a (Roferon-A) in cancer patients. Cancer 1986;57:1709–1715. 552. Quesada JR, Talpaz M, Rios A, Kurzrock R, Gutterman JU. Clinical toxicity of interferons in cancer patients: a review. J Clin Oncol 1986;4:234–243. 553. Whiting JF, Green RM, Rosenbluth AB, Gollan JL. Tumor necrosis factor-alpha decreases hepatocyte bile salt uptake and mediates endotoxin-induced cholestasis. Hepatology 1995;22:1273–1278. 554. Kemeny N, Childs B, Larchian W, Rosado K, Kelsen D. A phase II trial of recombinant tumor necrosis factor in patients with advanced colorectal carcinoma. Cancer 1990;66:659–663. 555. Fisher B, Keenan AM, Garra BS, et al. Interleukin-2 induces profound reversible cholestasis: a detailed analysis in treated cancer patients. J Clin Oncol 1989;7:1852–1862. 556. Samlowski WE, Wiebke G, McMurry M, Mori M, Ward JH. Effects of total parental nutrition (TPN) during high-dose interleukin-2 treatment for metastatic cancer. J Immunother 1998;21:65–74. 557. Haga Y, Sakamoto K, Egami H, et al. Changes in production of interleukin-1 and interleukin-2 associated with obstructive jaundice and biliary drainage in patients with gastrointestinal cancer. Surgery 1989;106:842–848. 558. Hoffman M, Mittelman A, Dworkin B, et al. Severe intrahepatic cholestasis in patients treated with recombinant interleukin-2 and lymphokine-activated killer cells. J Cancer Res Clin Oncol 1989;115:175–178.
Section IV: Toxin Mediated Liver Injury
HEPATOTOXICITY OF HERBAL PREPARATIONS
27
Doris B. Strader and Leonard B. Seeff Abbreviations AIH autoimmune hepatitis ALT alanine aminotransferase ATR atractylosides CAM ‘complementary and alternative’ medications
INR NCCAM
international normalized ratio NDGA The National Institute of Health’s National OLT Center for Complementary and Alternative VOD Medicine
INTRODUCTION The use of ‘complementary and alternative’ medications (CAM) to treat a variety of ailments is increasing in the USA.1 The National Institute of Health’s National Center for Complementary and Alternative Medicine (NCCAM) divides CAM into five categories: (1) alternative therapy, which includes homeopathic and naturopathic methods; (2) mind–body interventions, including prayer, meditation, art and dance; (3) biologically based therapy, to which herbals and dietary supplements belong; (4) manipulative and body-based methods, including chiropractic, osteopathic and massage therapy; and (5) energy therapies, including gi-gong, Reiki and electromagnetic field methods. Biologically based therapies, particularly herbal preparations, are by far the most common variety of CAM in the US. It is estimated that since 1997 at least 42% of Americans have used some form of CAM within the previous year.1 In addition, surveys of hepatology clinics across the US reveal that 20–30% of patients with liver disease use herbal remedies.2–4 The rationale for the increased use of CAM in the US is complex and involves a number of factors. First, CAM users report an increased ‘sense of control’ as well as disillusionment with current physician-prescribed medications. Often conventional medicine is perceived as impersonal and technical, with patients expected to play only a passive role in their own care. Recent studies suggest that many CAM users have chronic or incurable diseases, such as diabetes, AIDS, arthritis and cancer, and often feel that conventional medicine has failed them.5,6 CAM therapy reportedly provides patients with the opportunity to actively participate in their own health care and use products with ‘centuries of experience’ behind them. Second, the impression that CAM, particularly herbs and botanicals, is more natural and therefore more healthful than conventional therapies, appeals to the recent desire of Americans to return to a more holistic, nature-oriented lifestyle. Third, many patients have shown an increasing reluctance to have invasive conventional medical/surgical procedures, but rather have embraced the more benign, non-invasive CAM therapies, including massage, acupuncture, biofeedback and imagery. Fourth, the assumption that all medicines packaged in pill form are approved by the Food and Drug Administration, as well as the relative lack of information regarding the adverse effects of alternative medical therapies in the
nordihydroguaiaretic acid orthotopic liver transplantation veno-occlusive disease
lay literature, has created the impression that all CAM is safe.5 Finally, most medical professionals have a limited knowledge of the possible risks and benefits of CAM. Consequently, many patients seek out herbalists, faith-healers, and other alternative health providers to discuss therapies they feel are safer than pharmaceutically derived medications and that may provide them with more options. Herbals and botanicals are the most common form of CAM used in the US. As with other categories of CAM, the primary criticism of herbal use is that claims of efficacy are often based solely on the accumulation of personal experience rather than through scientific study. Accordingly, a number of investigators have attempted to evaluate the proposed benefits of these products by conducting scientifically valid studies. Several animal trials of drug-induced liver injury have demonstrated that some herbal preparations appear to have ‘hepatoprotective’ effects as a consequence of antioxidant, anti-inflammatory, immunomodulatory and antiviral activities.6–16 However, the protective effects noted experimentally have not necessarily translated into reversal of chronic injury or prevention of future injury in humans with liver disease. Further research initiatives to determine the possible benefits of CAM seem warranted, and many such studies are currently under way. On the negative side, however, are the accumulating data demonstrating that some forms of CAM, particularly herbal remedies, can cause toxic adverse events. What follows is a summary of the current literature describing the hepatotoxicity of some herbal medications. It must be emphasized, however, that the attribution of overt liver dysfunction to the use of any medication, be it a conventional drug or a herbal product, can often be very difficult. Even in wellaccepted instances of hepatotoxicity the frequency of such an occurrence is usually rare, and may be the consequence of an idiosyncratic reaction. Therefore, it is often difficult to positively establish causality. Confidence in such an affiliation is increased when there are a large number of reported cases, and when the manifestations of the hepatotoxic response breed true, namely that the resulting liver disease is consistently necroinflammatory in character, simulating hepatitis; is consistently cholestatic in nature, simulating biliary obstruction; or consistently manifests as a mixed pattern. As will be seen in the review that follows, many instances of presumed herbalrelated instances of hepatotoxicity are based on only a single case or on a very small number of cases, not always adequately evaluated,
551
Section IV. Toxin Mediated Liver Injury
so that the validity of the association may be suspect. Nonetheless, there are herbals that have unquestionably caused liver disease, sometimes very serious in nature, and others in which observed liver dysfunction seems associated with usage of a particular agent but less compelling data. Because herbals are not subjected to the more rigorous evaluation that is directed toward evaluating the safety of conventional drugs (and even this approach is sometimes flawed), the true extent of herbal hepatoxicity is still unclear. Information is generally derived from individual case reports. Still, as is the case for conventional drugs, it is just as likely that herbals can cause liverrelated toxicity, and therefore it is imperative to maintain continuing careful surveillance and to undertake sophisticated evaluation and appropriate reporting of probable instances of herbal-related hepatotoxicity in order to fully establish the frequencies and patterns of liver disease provoked by these products.
FACTORS THAT POTENTIALLY PROMOTE HERBAL HEPATOXICITY HERBAL FORMULATION The factor most likely responsible for the hepatotoxicity of herbal preparations is the manner in which herbals are formulated. Unlike conventional medications that generally consist of a standardized formulation of a specific agent in a known concentration, herbal preparations often consist of mixtures of ingredients, often in impure form and varying concentration.2,17 Herbal medications are available in a variety of forms, including roots, seeds, leaves, teas, powders, oils, capsules and tablets. The form a patient receives is dependent on geography, season and culture, as well as the expertise and preference of the practitioner. For example, in rural areas of some developing countries patients are more likely to receive leaves, seeds and teas ‘in season’, whereas in the west, capsules and tablets are the preferred form and are available throughout the year. In developed and developing countries alike, the purity of a particular herbal is uncertain, as standards for harvesting, extracting, and storing herbal agents do not exist. Harper et al. tested 12 samples of red peony root (Paeonia lactiflora) purchased in London and identified variations in the concentration of the active ingredient, Paeoniflorin, ranging from 0.01% to 4.5%.18 In addition, contamination with ingredients such as conventional medications, other substances or pathogenic microorganisms has also been reported. In one trial19 corticosteroids, non-steroidal anti-inflammatory drugs and benzodiazepines were found in ‘ethnic’ remedies of Asian origin. Other studies have identified cadmium and thallium in Ayurvedic remedies, caffeine but no yohimbine in Dutch yohimbine tablets, and mucor fungus in a naturopathic medication ingested by a bone marrow transplant patient.19–21 In addition, botanical misidentification, product substitution or adulteration, and lack of compound stability have also been reported.2,17 It is probable that these factors may increase the likelihood of toxicity, result in diagnostic delay, and lead to difficulty in identifying the causal agent.
interact with their prescribed medications. However, herb–drug interactions have been reported in a number of publications and can be significant.22–28 Most reports focus on herbals used in western countries and describe increased international normalized ratio (INR) or altered platelet function leading to bruising or bleeding, reduction in the concentration of immunosuppressant drugs resulting in decreased immune suppression and possible graft loss, and increase in toxic metabolites producing abnormalities in liver biochemistries.22–24,26,28–30 Table 27-1 provides a list of the most common herb–drug interactions.
TYPES OF LIVER INJURY A wide variety of liver injuries have been associated with herbal use, including abnormal liver tests, acute and chronic hepatitis, cirrhosis, and liver failure.31–36 Although histologic studies of cases of herbal hepatoxicity are relatively sparse, there have been reports of liver biopsy evidence of zonal/bridging necrosis, fibrosis, steatosis and veno-occlusive disease (VOD).37 Some herbals are associated with a specific type of histologic injury, but in most cases the histologic appearance of herbal hepatotoxicity is indistinguishable from that of toxic liver injury of other causes. The form of histologic hepatic injury reported for several herbal remedies is shown in Table 27-2.
HERBAL AND BOTANICAL PREPARATIONS ASSOCIATED WITH HEPATOTOXICITY ATRACTYLIS GUMMIFERA AND CALLILEPSIS LAUREOLA Atractylis gummifera is a thistle indigenous to the Mediterranean region, where 26 species have been identified. The plants secrete a whitish-yellow glue-like substance often used by children as chewing gum.38 Intoxication often occurs by accident, because of confusion
Table 27-1. Hepatotoxic Herb–Drug Interactions Drug
Herb
Interaction
Clinical sign
Anticoagulants
Danshen Devil’s claw Dong quai Feverfew Garlic Gingko Ginseng Papaya St John’s Wort Tamarind St John’s Wort Pyrrolizidines Germander Licorice Sho-saiko-to Licorice
Increased INR Unknown Increased INR Platelet dysfunction Increased INR Platelet dysfunction Decreased INR Increased INR Decreased INR Inc. aspirin conc. CYP3A4 induction CYP3A4 induction CYP3A4 induction Reduced clearance Reduced clearance Mineralocorticoid
Bleeding risk Purpura Bleeding risk Bleeding risk Bleeding risk Bleeding risk Clotting risk Bleeding risk Clotting risk Bleeding risk Rejection risk Hepatotoxicity Hepatotoxicity Hypokalemia Low pred conc Low spir conc
Cyclosporin CYP34A drugs Prednisolone
552
HERB–DRUG INTERACTIONS
Spironolactone
As mentioned above, most consumers of herbal remedies consider them to be safe and rarely consider the possibility that herbals may
Adapted from Stedman et al.
Chapter 27 HEPATOTOXICITY OF HERBAL PREPARATIONS
Table 27-2. Herbs and Associated Liver Injury Herbal remedy
Histologic liver injury
Atractylis gummifera Black cohosh Camphor Cascara Chapparal Chaso (and Onshido) Greater Celandine Germander Impila Ju bu huan Kava Ma huang Mistletoe (skullcap, valerian) Pennyroyal Pyrrolizodine (comfrey, mate, bush tea) Sho-saiko-to (dai-saiko-to, TJ-9)
Diffuse hepatic necrosis Elevated LFTs, liver failure Elevated LFTs, Reye’s syndrome Bridging fibrosis, bile duct prolif. Cholestasis, zone 3 necrosis Elevated LFTs, liver failure Cholestasis Zone 3 necrosis, cirrhosis Hepatic necrosis Periportal fibrosis, steatosis Elevated LFTs, liver failure Elevated LFTs, hepatic necrosis Elevated LFTs, acute hepatitis Acute hepatitis, liver failure VOD (veno-occlusive disease) Bridging fibrosis, steatosis
with wild artichoke, after ingestion of the roots, where the toxins are concentrated.38 Traditionally, Atractylis gummifera has been used as a purgative, emetic, diuretic and antipyretic. The onset of hepatitis is usually acute, beginning a few hours after ingestion.39,40 Headache, vomiting and abdominal pain are characteristic, with neurovegetative symptoms, hepatorenal failure and profound hypoglycemia rapidly ensuing. Fulminant hepatic failure and death have been reported.40–42 Toxicity is felt to be due to atractylosides (ATR) and gummiferin, which have been shown to inhibit mitochondrial function.43,44 ATR competitively inhibits the transport of ADP and ATP, blocking oxidative phosphorylation. It is also thought to induce apoptosis by inducing the mitochondrial membrane permeability transition pore, leading to release of cytochrome c and caspase-activating proteases.43,44 Impila, Callilepsis laureola, is a plant indigenous to the Natal region of South Africa. For years it has been used as a remedy for stomach ailments, tapeworm infestation, impotence, cough, as a fertility enhancer, and to ward off ‘evil spirits’. Several cases of fulminant hepatic failure and renal tubular necrosis have been reported. The clinical features begin abruptly and include abdominal pain, vomiting, diarrhea, convulsions, and acute liver and renal failure and profound hypoglycemia. A mortality rate of >60% is reported within 24 hours, and 91% by 5 days.40 Interestingly, the toxic components of impila have been identified as ATR and carboxyatractyloside, which decomposes to ATR. Antioxidant therapies, including S-adenosyl-L-methionine or betaine have been suggested for the management of toxicity caused by either herb, but are of unproven benefit at present.
BLACK COHOSH Black cohosh, Cimicifuga racemosa, is a leafy, cylindrical black rhizome with white flowers native to Canada and the US, and cultivated in Europe. The medicinal portions are in the fresh and dried roots. Black cohosh has many common names, including black snake root, rattleroot, rattleweed, squaw root, bugbane, bugwort, cimicifuga and richweed. It is used primarily for the relief of menopausal symptoms, but is also used to treat rheumatism, bronchitis, and as
a weight loss aid. Two cases of hepatotoxicity associated with black cohosh have recently been reported. In the first, a 47-year-old woman presented with jaundice, pruritus, elevated serum levels of alanine aminotransferase (ALT), serum bilirubin, and INR 1 week after taking black cohosh for relief of menopausal symptoms.45 Other potential causes of acute liver disease were excluded by serologic, biochemical and radiologic examination. The patient’s condition rapidly deteriorated and an orthotopic liver transplantation was performed. Histologic evaluation of the explanted liver revealed severe hepatitis and multiacinar dropout.45 The patient recovered post transplant without event. In the second reported case, a 52year-old woman presented with deep jaundice, elevated levels of ALT, bilirubin, INR, and alkaline phosphatase 4 weeks after discontinuing use of a herbal preparation she had used for 3 months to treat severe tinnitus.46 The preparation contained ground ivy, golden seal, ginkgo, oat seed and black cohosh. The patient’s condition deteriorated and she developed encephalopathy and hepatorenal failure. She underwent liver transplantation with an uneventful postoperative course. Evaluation of the explant revealed massive hepatic necrosis. It is not certain that black cohosh was the hepatotoxic agent in the above-mentioned cases, as neither preparation was evaluated for the presence of contaminants, and one contained several different herbal agents.47,48 The active ingredients in black cohosh are triterpenes, quinolizidine alkaloids, and phenylpropane derivatives. To date, no studies have implicated these agents in hepatic injury.49 However, the severity of the hepatic injury in these two cases cannot be ignored. Further study of black cohosh is needed to determine what, if any, role it plays in the development of hepatotoxicity.
CAMPHOR OIL Camphor oil is extracted from the camphor tree, Cinnamomum camphora, indigenous to Vietnam and an area extending from southern China to southern Japan. It is used externally as a bronchial secretolytic and hyperemic for cough and bronchitis, rheumatism and arrhythmia. One case of hepatotoxicity has been reported involving a 2-month old girl treated with a camphor-containing cold remedy applied to the skin.50 In this case, the infant, with recent swelling in the right inguinal area, was taken to her local hospital. She was noted to be malnourished due to use of an improperly diluted infant formula and was admitted for nutritional support. During routine laboratory evaluation to monitor for refeeding syndrome, abnormal aminotransferase values were noted. On physical examination, a soft liver edge was palpated 1.5 cm below the costal margin. Viral hepatitis as a cause was ruled out. On questioning, the mother admitted to applying generous amounts of Vicks VapoRub to the baby’s chest and neck three times a day for 5 days. Liver tests returned to normal after application of the rub was discontinued. A second report by Jimenez et al. involved a case of oral ingestion of camphor resulting in toxicity resembling Reye’s syndrome.51 In this report, a 6-month-old child was evaluated for a 2-day history of cough and fever, diagnosed with pneumonia, and treated with ampicillin by his private physician. The following day the infant was lethargic, with worsening pulmonary manifestations and radiographic evidence of bilateral diffuse interstitial infiltrates. Six hours later the infant was unrousable, the liver was palpated 3 cm below the costal margin, and a diagnosis of Reye’s syndrome was suspected.
553
Section IV. Toxin Mediated Liver Injury
On transfer to a tertiary care center the infant was comatose, had an elevated white count with bandemia, aminotransferase levels of 750–1000 U/l, an elevated serum bilirubin level and a prolonged prothrombin time. Liver biopsy results did not show the characteristic pleomorphic, greatly swollen mitochondria with absent dense bodies and stranding of mitochondrial matrix typically seen in fatal Reye’s syndrome; however, given the clinical and neurologic syndrome, a presumptive diagnosis of Reye’s syndrome was made. On questioning the family, it was learned that the infant was regularly treated with a home remedy containing camphor and alcohol. Although the infant’s liver function improved over the next few days, his neurologic status did not and EEG studies revealed an absence of electrical activity. The infant died on the fifth hospital day as a result of cardiac arrest.51 Camphor is a cyclic terpene compound that is a constituent of several medications, including salves, ointments and oral cold remedies. When rubbed on the skin, camphor is a rubefacient which causes local irritation to the skin, thereby blocking pain by ‘counterirration’ (affects the same segmental central nervous system level as that inducing the original pain).52,53 Camphor can be absorbed through the skin, mucous membranes and placenta, leading to significant hepatoneurotoxicity, occasionally culminating in hepatic encephalopathy.54,55 Ingestion of small doses of camphor, characterized by abrupt onset of nausea and vomiting followed by agitation and seizures, can be fatal in young children. The mechanism by which camphor leads to hepatotoxicity is unclear. Ordinarily, camphor is metabolized in the liver and excreted in the urine as an inactive glucuronide compound. Although the exact hepatotoxic metabolite is unknown, it is felt that infants are particularly susceptible to camphor hepatotoxicity because of their immature hepatic detoxification mechanisms.56,57 As a result, it has been recommended that camphor-containing cold remedies (Vicks VapoRub, BenGay, Afrin saline mist) should not be used in children under the age of 2 years.
554
the second a senna alkaloid, have also been associated with the development of chronic hepatitis.59,60 Interestingly, in the first report the authors attributed the hepatotoxicity to the dioctyl sulfosuccinate component rather than the anthraquinone. Cascara is approved by the FDA for use as a laxative and is widely used in the US without hepatic sequelae. Nevertheless, it appears that in extremely rare instances cholestatic hepatitis may occur.
CHAPARRAL Chaparral leaf, Larrea tridentate, is commonly known as creosote bush or greasewood. It is a desert plant indigenous to the southwestern US and Mexico. Native Americans grind the leaf and use it as a tea for a variety of ailments, including the common cold, bone and muscle pain, chicken pox, cancer, tuberculosis, venereal diseases and snake bites.2 Currently, chaparral is packaged as capsules, tablets or balms and used as a ‘liver tonic’ and as treatment for skin lesions. Sheikh et al. reviewed all 18 cases of chaparral-induced injury reported to the FDA since 1990.61 Thirteen of these displayed hepatic injury, ranging from mild hepatitis to cirrhosis and even fulminant hepatic failure. Although the hepatic manifestations were heterogeneous, the predominant pattern of liver injury was cholestatic, with elevations in serum aminotransferases, bilirubin and alkaline phosphatase levels. Only a few patients progressed to cirrhosis, and two required orthotopic liver transplantation (OLT) for hepatic failure. The latter two patients had used chaparral for more than a year, whereas in the remainder it had been used for 1–6 months. The pathophysiology of chapparal toxicity is unknown. It contains a mixture of flavonoids, amino acids, ligans and volatile oils.62 The active ingredient is nordihydroguaiaretic acid (NDGA), which may inhibit cyclooxygenase or cytochrome P450 activity, or act via an immune-mediated mechanism.63,64 In addition, chaparral metabolites exhibit estrogen activity which may contribute to hepatotoxicity.65,66
CASCARA
CHASO AND ONSHIDO
Cascara sagrada is derived from the dried bark of the bush or tree Rhamnus purshiana. The plant is indigenous to the western part of North America and is also cultivated in Canada and eastern Africa. Cascara is used for relief of constipation, hemorrhoids, and as a rectoanal postoperative treatment. The active constituent is an anthracene derivative (O- and C-glycosides). Right upper quadrant pain, jaundice, ascites and portal hypertension were reported in a 48-year-old man 3 days after ingesting cascara sagrada for laxative purposes.58 Liver biopsy revealed moderate portal inflammation with eosinophils and plasma cells, and mild portal–portal bridging fibrosis without cirrhosis. Bile duct proliferation and bile stasis were also noted. The patient recovered fully 3 months after discontinuation of cascara. No other cause for hepatotoxicity was identified and a presumed diagnosis of cascara hepatotoxicity was made. The pathogenesis of cascara hepatotoxicity is unknown, but it is assumed that anthracene glycosides are involved. The temporal association of the ingestion of cascara with the symptoms and liver biopsy evidence of moderate inflammation with lymphocytes, plasma cells and eosinophils suggests an immune-mediated process. Two anthraquinones, the first, Doxidan (a combination of danthron 1,8-hydroxy anthraquinone and dioctyl calcium sulfosuccinate), and
Chaso and Onshido are two widely available Chinese herbal weight loss aids. The manufacturers report that Chaso contains green tea, cassia torae semen, leaves of lotus, Fructus lycii, Fructus crataegi and chrysanthemum flowers. Onshido contains extract of Gynostemma pentaphyllum makino, green tea, aloe, Fructus crataegi and raphani semen. In 2002, 12 Japanese patients (six using Chaso and six using Onshido) presented with symptoms of severe fatigue and anorexia 5–40 days after ingesting the herbs.67 Most mistook their symptoms for those associated with weight loss and did not seek immediate medical attention. On presentation, the levels of the aminotransferases and bilirubin were increased, and the INR was significantly elevated. Two patients developed hepatic encephalopathy and one received a liver transplant 8 days after admission. Another patient died 45 days after admission secondary to intestinal bleeding and infection; the remaining 10 patients improved after discontinuation of the herbals, with the liver biochemical tests all returning to normal.67 On analysis of the ingredients contained in the preparations, N-nitroso-fenfluramine was detected. Fenfluramine was once prescribed for weight loss but was withdrawn from clinical use because of severe cardiac complications. Although fenfluramine was not identified in either Chaso or Onshido, data suggest that N-
Chapter 27 HEPATOTOXICITY OF HERBAL PREPARATIONS
nitroso compounds have been linked with hepatic carcinogenesis.68 The mechanism of injury is felt to be direct hepatotoxicity. Since the publication of this report, 21 cases of Chaso-induced hepatitis and 135 cases of Onshido-induced hepatitis were reported to the Ministry of Health, Labor and Welfare in Japan, most occurring between April and August 2002.69 The herbals have subsequently been removed from the market in the US.
COMFREY AND PYRROLIZIDINE ALKALOIDS (PA) Comfrey, Symphytum officinale, is a plant indigenous to Europe and temperate Asia. The medicinal portions are the leaves and fresh roots, which are used externally for bruises, sprains, rheumatism and pleuritis. The active ingredients, pyrrolizidine alkaloids, are the most important plant toxins associated with liver disease. Heliotroprium, Senecio, Crotalaria and t’u-san-chi’ (Compositae) species are most often responsible for liver injury. Hepatotoxicity related to PAs was first described 70 years ago as Senecio (mate tea) disease in South Africa.70–73 Reports of Jamaican children developing hepatomegaly and evidence of decompensated cirrhosis with ascites after the ingestion of ‘bush tea’ soon followed.74,75 Later, epidemics were reported from India and Afghanistan.76,77 PAs are dose-dependent hepatotoxins that typically cause veno-occlusive disease (VOD). VOD may present as acute, subacute or chronic liver injury, characterized by abdominal distention, hepatomegaly, ascites and edema. Pathophysiologically, non-thrombotic obliteration of terminal centrilobular veins develops, reminiscent of the Budd–Chiari syndrome and leading to disruption of hepatic blood flow.75 High doses of PAs are often responsible for acute liver injury, and the long-term use of relatively low-dose PA has been associated with insidious hepatotoxicity. Recovery from acute disease is the rule, but liver failure with resultant 20–40% mortality has been reported.78 The mechanism of PA-induced hepatotoxicity is unclear. Numerous articles suggest a toxic mechanism which is reproducible in animals and related to biotransformation of alkaloids by cytochrome P450, forming pyrrole derivatives which serve as alkylating agents.79,80 Toxicity can be augmented by the concomitant use of phenobarbital via the induction of cytochrome P450. Discontinuing the herbal can result in resolution of symptoms in some patients, but those with acute or chronic liver failure may require liver transplantation.
Dai-Saiko-to (Sho-Saiko-to, TJ-9, Xiao-ChaiHu-Tang) Dai-Saiko-to is a Japanese kampo formula consisting of seven herbs that has been used since AD100 for the treatment of fatigue, fever, dyspepsia, gallstones, and recently, chronic liver disease.81 DaiSaiko-to differs from Sho-Saiko-to only in the proportion of the herbal constituents, which include bupleurum, pinellia, jujube, ginseng, ginger rhizome, glycyrrhiza and scutellaria.81 Itoh et al. reported four cases of worsening of the aminotransferase levels, which improved with cessation of the drug and recurred with rechallenge.82 In 1997 a case of autoimmune hepatitis was reported that was possibly induced by Dai-Saiko-to.33 The patient was a 55-yearold woman who initially presented with fatigue, fever and abnormal liver tests (ALT 866 IU/l, total bilirubin 13 mg/dl, alkaline phosphatase 317 IU/l). Autoimmune markers and hepatitis serologic
tests were negative; liver histology revealed chronic hepatitis. Within 4 weeks the symptoms and biochemical abnormalities had resolved spontaneously and the ALT values remained normal for 5 years. Subsequently, an increase in the ALT level to 300 IU/l was noted and liver histology revealed chronic hepatitis with severe steatosis and a lipogranuloma. The patient was treated with Dai-Saiko-to and 2 weeks later developed fatigue, fever, an elevated ALT level of 390 IU/l and an antinuclear antibody titer of 1:2560. An International Autoimmune Hepatitis (AIH) diagnostic score of 18 supported the diagnosis of autoimmune disease, and a strongly positive lymphocyte stimulation test for Dai-Saiko-to (340%) suggested this herb was responsible.83 Abnormalities in the ALT values returned to normal after treatment with prednisolone. The mechanism by which the hepatic abnormalities occur is unclear. It has been postulated that scutellaria, which has been implicated in four other cases of hepatotoxicity, might be the culprit.84 Likewise, the saponin in bupleurum, suggested to have toxic effects on cell membranes, may also be involved.2
GERMANDER Germander, Teucrium chamaedrys, is a sub-shrub with a short-lived, main root from which long-reaching thin branched roots grow. It is indigenous to the Mediterranean region as far as the Urals. The active ingredients are diterpenes, iridoide monoterpenes, caffeic acid derivatives and flavonoids. Germander is felt to have cholagogic and antiseptic properties, and is ingested as a capsule or tea to treat dyspepsia, fever, gout and obesity. Following large-scale marketing as a weight loss aid in France in 1992, 30 cases of acute, chronic and fulminant hepatitis were reported.35,85 Most affected patients were women attempting to lose weight, most of whom ingested 600–1600 mg daily. The clinical syndrome, characterized by markedly elevated levels of aminotransferases and bilirubin, and impaired hepatic synthetic function, began approximately 2 months after ingestion. The range of histologic findings included mild chronic hepatitis, fibrosis, cirrhosis, and in some cases acute midzonal hepatocellular necrosis. Those patients without cirrhosis completely recovered after discontinuation of the herb. Analysis of Teucrium chamaedrys revealed the presence of a number of furan-containing neoclerodane diterpenoids which are well-known powerful carcinogens.86–88 In rat hepatocytes, these constituents are oxidized by cytochrome P450 3A4 to reactive metabolites which bind to proteins, deplete cellular glutathione and protein thiols, and cause plasma membrane blebbing and cell disruption.87 Two cell culture studies suggest that germander induces apoptosis after the formation of reactive metabolites.89,90 These reports suggest a reactive metabolite as the mechanism of injury for germander; however, an autoimmune mechanism was proposed after antimicrosomal epoxide hydrolase autoantibodies were found in the sera of some patients.91
GREATER CELANDINE Greater Celandine, Chelidonium majus, is a plant found throughout Europe and the temperate and subarctic regions of Asia. The root is harvested between August and October. Isoquinolone alkaloids and caffeic acid derivatives are felt to be the active ingredients. Celandine is believed to have mild analgesic, central sedative, chol-
555
Section IV. Toxin Mediated Liver Injury
agogic and antimicrobial effects, and is used to treat biliary colic, cholelithiasis, jaundice, gastroenteritis, and diffuse liver and gallbladder complaints. Ten cases of acute hepatitis induced by celandine have been reported.34 All of the patients were women, with the onset of symptoms and moderate elevations of liver chemistries noted approximately 3 months after ingestion of the herb. A cholestatic pattern was noted in five patients; low titer antinuclear antibodies were noted in eight, and portal inflammation with bridging fibrosis and eosinophilic infiltrates was seen on liver biopsy in most of them. Liver chemistries returned to normal 2–6 months after cessation of the herb. The mechanism of action of celandine hepatotoxicity is unknown, but is considered to be idiosyncratic in nature owing to the variable latency period and the lack of dose dependence.31,37 Some have suggested an immune response owing to the presence of serum autoantibodies and eosinophilic infiltrates on liver biopsy, but these findings can be non-specific. More than 20 alkaloids with biologic activity have been identified in celandine; however, the toxic component has yet to be identified.
Ju Bu Huan Ju Bu Huan, Lycopodium serratum, is a plant found worldwide. It is believed to have sedative, analgesic and antispasmodic effects and has been used for more than a millennium as a sleeping aid. Several cases of acute and chronic hepatitis associated with ju bu huan have been published in the literature.92,93 The largest report describes seven patients who developed hepatitis a mean of 20 weeks after ingestion.92 Abdominal pain, constitutional symptoms, jaundice, hepatomegaly and pruritus were characteristic, as were elevations in the aminotransferase and alkaline phosphatase levels. Liver histology generally revealed periportal necrosis and cholestasis. In one patient eosinophilic inflammation was noted, and in another moderate periportal fibrosis, lymphocytic inflammation, focal hepatocellular necrosis, and microvesicular steatosis were present. Symptoms and abnormal laboratory tests resolved within 8 weeks after discontinuation of the herb. Horowitz et al. suggest that the clinical spectrum of ju bu huan toxicity may vary with the mode of ingestion: specifically, the usual dose versus an acute overdose.93 Acute overdose appears to be associated with flaccid weakness, lethargy and respiratory depression (noted in three children), whereas long-term use presents as acute hepatitis without neurologic findings. The mechanism of ju bu huan hepatotoxicity is not fully understood. An L-alkaloid, L-tetrahydropalmatine, which is structurally similar to pyrrolizidine and berberine alkaloids, may be the toxic agent. Long-term berberine use is associated with hyperbilirubinemia in animals, which may be caused by displacement of bilirubin from albumin or disruption of bilirubin conjugation.37,93 Although the exact mechanism of action of ju bu huan is unknown, the presence of fever, skin rash and eosinophilia in some patients suggests an immune-mediated mechanism.
Kava Kava is a rhizome of the pepper plant Piper methysticum. The plant is a 2–3-m tall bush indigenous to the South Sea Islands. The extract of the dried rhizome contains kava lactones, which have central muscle-relaxant, anticonvulsive and antispasmodic effects. The herb
556
also has hypnotic/sedative, analgesic and psychotropic effects, which explain its worldwide use as a therapy for anxiety and tension. Four cases of acute hepatitis and 18 cases of acute liver failure have been reported in connection with its use.94–96 In one report, a 33-year-old woman took Laitan (a sleeping aid containing 210 mg of kava lactones) daily for 3 weeks.94 Two months later she restarted the kava preparation. Three weeks later, 1 day after ingesting 60 g of alcohol, she developed malaise and jaundice. Serum levels of the aminotransferases, bilirubin and alkaline phosphatase were elevated 60-, 15- and 3-fold, respectively. Other forms of acute hepatitis were excluded. Liver biopsy revealed inflammation of the portal tracts, bridging necrosis and canalicular cholestasis. The patient recovered completely within 8 weeks of cessation of kava. A lymphocyte transformation test indicated strong and concentration-dependent T-cell reactivity to Laitan (but not the herb Exsepta, which she was also using over the previous 2 months). Phenotyping of cytochrome P4502D6 activity with debrisoquine showed that the patient was a poor metabolizer. The authors interpreted these data as suggesting that the patient had an immune-mediated reaction to a metabolite of kava. In a second report, a 50-year-old man presented to his physician complaining of jaundice, fatigue, and dark urine of 1 month’s duration.96 He reported taking four capsules of kava extract daily (210–280 mg of kava lactones) for 2 months to treat anxiety. He was on no other medications and did not consume alcohol. Serum aminotransferase levels were elevated >50-fold; alkaline phosphatase, bilirubin and prothrombin time were also elevated. Ultrasound of the liver revealed mild hepatomegaly but no evidence of ascites or portal vein thrombosis. Viral hepatitis was excluded with blood tests. Within 48 hours the patient’s clinical condition deteriorated, he developed stage IV encephalopathy, and received an orthotopic liver transplant 2 days later. Histology of the explant revealed extensive hepatocellular necrosis and extensive lobular and portal infiltration with lymphocytes and eosinophils. The exact mechanism of kava hepatotoxicity is unknown. Some have suggested that a relatively common genetic polymorphism of drug metabolism, CYP2D6 deficiency, may predispose to kava hepatotoxicity.94,97 However, further study is required before this mechanism can be considered definitive.
Ma Huang Ma-huang, Ephedra sinica, is a 30-cm tall, lightly branched shrub which is found mainly in Mongolia and the bordering area of China. The dried young branchlets, which are harvested in the autumn, are considered the medicinal parts. Ma-huang is generally used as a tea for cough and bronchitis, as a weight loss aid, and as an energy enhancer. Instances of ma-huang hepatotoxicity have been reported. In the first case, a 33-year-old woman presented with nausea, vomiting, jaundice and abdominal pain 3 weeks after taking a herbal preparation containing ma-huang.98 She reportedly continued using the preparation despite symptoms, until jaundice developed. Serum aminotransferase and bilirubin levels were elevated without signs of chronic liver disease, and the presumptive diagnosis was acute viral hepatitis. She was admitted to hospital 1 week later with increasing jaundice, worsening aminotransferase levels, and evidence of autoimmune disease with an ANA titer of 1:160 and an antismooth muscle antibody titer of 1:80. Viral hepatitis serologies were nega-
Chapter 27 HEPATOTOXICITY OF HERBAL PREPARATIONS
tive. Liver biopsy demonstrated diffuse hepatic necrosis with occasional eosinophils and plasma cells in the portal tracts. On further questioning, the patient admitted taking another dose of the herbal preparation after her first hospital visit. Symptoms resolved and the liver panel returned to normal 4 months after discontinuation of ma-huang. In a second case, a 58-year-old woman with obesity presented with a history of 4 months of jaundice, fatigue, nausea and abdominal pain of unclear etiology.99 Initial evaluation revealed elevated levels of the aminotransferases, elevated bilirubin, negative viral hepatitis serologies, a negative ANA, an antismooth muscle antibody titer of 1:320, and normal abdominal CT imaging studies. Medication history revealed that she had used ma-huang as a weight loss aid for 4.5 months. Liver biopsy revealed severe infiltration with polymorphonuclear leukocytes, moderate fibrosis and lobular necrosis. She was treated with steroids for presumed autoimmune hepatitis but subsequently developed encephalopathy and was referred for liver transplant evaluation. While awaiting liver transplant, her status improved and she was discharged in a stable condition. The active ingredient in ma-huang is ephedrine, a sympathomimetic used in western medicine to treat asthma and also used as a central nervous system stimulant. Well-known side effects include nervousness, palpitations, headache and insomnia. The exact mechanism by which ephedrine induces hepatotoxicity is unknown, but the presence of autoimmune markers in both cases suggests an immune-mediated process, either as a primary effect or through unmasking of an underlying autoimmune diathesis. In the first case presented above, the concomitant use of other plant extracts in the preparation raises the possibility of herb–herb interaction or other contamination. In both cases the women were obese, and it is possible that non-alcoholic fatty liver disease may have contributed to the liver injury.
MARGOSA OIL (NEEM) Antelaea azadirachta (neem) and Azadirachta indica (margosa oil) are indigenous to the woods of India and Sri Lanka. The bark, leaves and seeds of the deciduous tree are the medicinal portions. Azadirachta indica and Antelaea azadirachta are commonly used in India, Sri Lanka, Burma, Indonesia, Thailand and Malaysia for inflammatory and febrile illnesses, as well as dyspeptic symptoms and worm infestations. There have been several case reports of toxic encephalopathy among infants and young children given small amounts of oral margosa oil.100 Although the oil is generally used externally, some traditional practices include giving small oral amounts to infants and children. Several children presented to physicians with vomiting, drowsiness, tachypnea and recurrent seizures. Laboratory studies revealed a leukocytosis, abnormal liver tests, severe metabolic acidosis, and hepatic lesions consistent with Reye’s syndrome. Supportive care and control of seizure activity led to resolution of symptoms in most children, but a few developed neurologic deficits and some died as a result of hepatic failure. Animal studies of margosa oil ingestion indicate that the injury sequence begins with the rapid development of mitoses and binucleated cells, followed by mitochondrial injury, swelling, and pleomorphism within the nuclei of hepatocytes.101,102 Proliferation
and hypertrophy of the endoplasmic reticulum and subsequent microvesicular steatosis have also been noted. It has been suggested that margosa oil is a mitochondrial uncoupler, increasing mitochondrial respiration and decreasing intramitochondrial ATP.101 These effects may be due to changes in fatty acid metabolism that result in a change in the proportion of acid-soluble and acid-insoluble coenzyme A esters. Even though some suggest that supplementary therapy with L-carnitine and coenzyme A may be useful in the management of margosa oil-induced Reye’s syndrome, avoidance of oral use of this herbal product is clearly prudent.100
MISTLETOE, SKULLCAP, VALERIAN Mistletoe, Viscum album, is a semiparasitic round bush that grows on deciduous trees found primarily in Europe. The medicinal portions include the leaves, stem, and pea-sized berries. Mistletoe has been widely used to treat many illnesses, including degenerative inflammation of the joints, hypertension, asthma, vertigo, diarrhea, epilepsy and nervousness. One case of presumed mistletoe hepatitis was reported in a 49-year-old woman who presented with nausea, general malaise and right upper quadrant pain.103 Aminotransferase levels were elevated, hepatitis B surface antigen was not detected, the cholecystogram was normal, and liver biopsy revealed mild inflammation. Two years later she presented with a similar illness, and questioning revealed that both episodes were preceded by the ingestion of a herbal remedy containing kelp, motherwort, skullcap and mistletoe.103 A challenge test established that the herb was responsible for the symptoms. At the time, mistletoe was the only herb known to contain a potential toxin, lectin, and it was therefore singled out as the causative agent. Later evaluation of the herbal compound suggested that mistletoe was probably not an ingredient, casting doubt on the association between mistletoe and hepatitis. Skullcap, Scutellaria, is a perennial herb 60 cm in height and thickly covered with simple and glandular hairs, which is indigenous to North America and cultivated in Europe. The herb is pulverized and used as a sedative, an antispasmodic and an anti-inflammatory agent, and it is thought to inhibit lipid peroxidation. Valerian, Valeriana officinalis, is a short, cylindrical rhizome with bushy round roots indigenous to Europe and the temperate regions of Asia. It is widely cultivated in England, France, Japan and the US, and is used to treat conditions such as nervousness, insomnia, lack of concentration, headache, and nervous stomach cramps. Often both valerian and skullcap are contained in the same preparation used to relieve stress. Several cases of acute hepatitis have been reported with the use of herbal preparations containing skullcap and valerian. Four cases of jaundice, abdominal pain and dark urine have been reported in women taking two different herbal preparations, Kalms and Neurelax, for relief of stress.84 One woman developed ascites and encephalopathy necessitating intensive medical support. Available liver biopsy tissue revealed a range of abnormalities, from moderate acute hepatitis to bridging fibrosis, to advanced fibrosis and cirrhosis. In all four women the aminotransferase levels returned to normal and symptoms resolved with discontinuation of the herbal compound. Several other cases of jaundice have been reported to the Welsh Drug Information Centre of ingestion of preparations containing skullcap, valerian or both.84 As with mistletoe, the association with hepatitis is presumed; direct experimental evidence of toxicities due to these herbals is currently lacking.
557
Section IV. Toxin Mediated Liver Injury
PENNYROYAL Pennyroyal, Mentha pulegium, is a downy perennial which grows in western, southern and central Europe, Asia, Iran and Ethiopia. It is naturalized in North America. The medicinal portion, the essential oil, is extracted from the fresh plant or dried aerial parts. For centuries it was used as an abortifacient and as a pesticide against fleas, and because it is a source of intoxication it continues to be widely used.104 Current medical uses include digestive disorders, liver and gallbladder disease, amenorrhea, gout, colds, and skin disease. Most cases of hepatotoxicity have been reported in women who use pennyroyal to induce menstruation or abortion.104 In one instance, a 24-year-old woman ingested pennyroyal extract and black cohosh root for 2 weeks in an attempt to induce abortion. When this failed, she ingested additional, unknown amounts of these herbals over a short period. Soon after ingestion she developed abdominal cramping, chills, vomiting and syncope, with difficulty in being roused. Paramedics discovered the patient to be in cardiopulmonary arrest an estimated 7.5 hours after acute ingestion. She was intubated, successfully resuscitated and admitted to the intensive care unit. CT of the abdomen revealed a possible ruptured ectopic pregnancy. Aminotransferase levels, bilirubin and prothrombin time were elevated, and over the next 36 hours the patient developed signs of multiorgan failure and disseminated intravascular coagulation. The patient was comatose and unresponsive to all stimuli, anoxic encephalopathy was confirmed by CT, life support was discontinued and she died within 48 hours of acute ingestion. Two infants developed multiorgan failure after ingestion of mint tea containing pennyroyal, one with confluent hepatic necrosis noted at autopsy.105 Unlike most herbal preparations, the mechanism of pennyroyal hepatotoxicity is well known. The main constituent is R(+)-pulegone, which is oxidized by cytochrome P450 to menthofuran.106 Pulegone depletes hepatic glutathione by the formation of electrophilic metabolites, whereas menthofuran is directly toxic to hepatocytes. As a result of the glutathione loss, replacement of sulfhydryl groups by administering N-acetylcysteine has been advocated as a therapy. As menthofuran toxicity is not greatly affected by glutathione loss, the benefits of this therapy may only be evident in the early phases of pennyroyal poisoning. Despite this, N-acetylcysteine is recommended in cases where more than 10 ml of pennyroyal are ingested.
CONCLUSION The use of herbal preparations to treat various medical conditions is increasing. Among most consumers of herbals there is an implicit assumption of safety, and therefore physician consultation is not sought. Over the past decade, as the popularity of herbal remedies has increased, so have reports of toxicity. The scope of hepatotoxicity ranges from asymptomatic elevations in liver biochemistries to chronic hepatitis, cirrhosis and fulminant hepatic failure. Unfortunately, herbal preparations are neither regulated nor standardized, thereby making precise identification and quantification of ingredients or possible contaminants extremely challenging. This significantly hampers the ability to definitively assign causality to a particular herb when evidence of hepatic injury is observed. In most instances of reported toxicity no attempt at phytochemical analysis is made, but rather, a presumptive association is based on temporal
558
relationships and occasionally, unintentional rechallenge. Despite these drawbacks, the volume of consistent reporting of herb-related hepatotoxicity mandates serious consideration. It is essential that increased public awareness regarding the potential toxicity of herbals be maintained, as well as improved agricultural monitoring, appropriate regulatory systems, and improved scientific evaluation of the potential benefits and hazards of herbal preparations.
REFERENCES 1. Eisenberg DM, Davis RD, Ettner SL, et al. Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. JAMA 1998;280:1569–1575. 2. Stedman C. Herbal hepatotoxicity. Semin Liver Dis 2002;22:195–206. 3. Strader DB. Understudied populations with hepatitis C. Hepatology, 2002;36(Suppl 1): S226–236. 4. Seeff LB, Lindsay KL, Bacon BR, et al. Complementary and alternative medicine in chronic liver disease. Hepatology 2001;34:595–603. 5. Winslow LC, Kroll DJ. Herbs as medicines. Arch Intern Med 1998;158:2192–2199. 6. Brown JS, Marcy SA. The use of botanicals for health purposes by members of a prepaid health plan. Res Nurs Health 1991;14:339–350. 7. Anand KK, Singh B, Saxena AK, et al. 3,4,5-Trihydroxy benzoic acid (gallic acid), the hepatoprotective principle in the fruits of Terminalia belerica bioassay-guided activity. Pharmacol Res 1997;36:315–321. 8. Basaga H, Poli G, Tekkaya C, et al. Free radical scavenging and antioxidative properties of ‘silibin’ complexes on microsomal lipid peroxidation. Cell Biochem Funct 1997;15:27–33. 9. Halim AB, el-Ahmady O, Hassab-Allah S, et al. Biochemical effect of antioxidants on lipids and liver function in experimentally induced liver damage. Ann Clin Biochem 1997;34:656–663. 10. Kayano K, Sakaida I, Uchida K, et al. Inhibitory effects of the herbal medicine Sho-saiko-to (TJ-9) on cell proliferation and procollagen gene expressions in cultured rat hepatic stellate cells. J Hepatol 1998;29:642–649. 11. Lang I, NeKam K, Gonzalez-Cabello R, et al. Hepatoprotective and immunological effects of antioxidant drugs. Tokai J Exp Clin Med 1990;15:123–127. 12. Lin CC, Lin WC, Yang SR, et al. Anti-inflammatory and hepatoprotective effects of Solanum alatum. Am J Chin Med 1995;23:65–69. 13. Miguez MP, Anundi I, Sainz-Pardo LA, et al. Hepatoprotective mechanism of silymarin: no evidence for involvement of cytochrome P450 2E1. Chem Biol Interact 1994;91:51–63. 14. Mira L, Silva M, Manso CF. Scavenging of reactive oxygen species by silibinin dihemisuccinate. Biochem Pharmacol 1994;48:753–759. 15. Pietrangelo A, Borella F, Casalgrandi G, et al. Antioxidant activity of silybin in vivo during long-term iron overload in rats. Gastroenterology 1995;109:1941–1949. 16. Pines M, Knopov V, Genina O, et al. Halofuginone, a specific inhibitor of collagen type I synthesis, prevents dimethylnitrosamine-induced liver cirrhosis. J Hepatol 1997;27:391–398. 17. Bateman J, Chapman RD, Simpson D. Possible toxicity of herbal remedies. Scot Med J 1998;43:7–15. 18. Harper J. Traditional Chinese medicine for eczema. Br Med J 1994;308:489–490. 19. De Smet P. Adverse effects of herbal remedies. Adv Drug React Bull 1997;183:695–698.
Chapter 27 HEPATOTOXICITY OF HERBAL PREPARATIONS
20. Kew J, Leigh PN, Playford ED, et al. Arsenic and mercury intoxication due to Indian ethnic remedies. Br Med J 1993;306:506–507. 21. Keen RW, Deacon AC, Delves HT, et al. Indian herbal remedies for diabetes as a cause of lead poisoning. Postgrad Med J 1994;70:113–114. 22. Karliova M, Treichel U, Malago M, et al. Interaction of Hypericum perforatum (St John’s wort) with cyclosporin A metabolism in a patient after liver transplantation. J Hepatol 2000;33:853–855. 23. Cheng TO. Warfarin–danshen interaction. Ann Thorac Surg 1999;67:894. 24. Fugh-Berman A. Herb–drug interactions. Lancet 2000;355:134–138. 25. Shaw D, Leon C, Kolev S, et al. Traditional remedies and food supplements. A 5-year toxicological study (1991–1995). Drug Safety 1997;17:342–356. 26. Miller LG. Herbal medicinals: selected clinical considerations focusing on known or potential drug–herb interactions. Arch Intern Med 1998;158:2200–2211. 27. Rose KD, Croissant PD, Parliament CF, et al. Spontaneous spinal epidural hematoma with associated platelet dysfunction from excessive garlic ingestion: a case report. Neurosurgery 1990;26:880–882. 28. Janetzky K, Morreale AP. Probable interaction between warfarin and ginseng. Am J Health Syst Pharm 1997;54:692–693. 29. Homma M, Oka K, Ikeshima K, et al. Different effects of traditional Chinese medicines containing similar herbal constituents on prednisolone pharmacokinetics. J Pharm Pharmacol 1995;47:687–692. 30. Chen MF, Shimada F, Kato H, et al. Effect of glycyrrhizin on the pharmacokinetics of prednisolone following low dosage of prednisolone hemisuccinate. Endocrinol Jpn 1990;37:331–341. 31. Farrell GC. Drug-induced liver disease. Edinburgh: Churchill Livingstone, 1994: 513–518. 32. Kao WF, Hung DZ, Tsai WJ, et al. Podophyllotoxin intoxication: toxic effect of Bajiaolian in herbal therapeutics. Hum Exp Toxicol 1992;11:480–487. 33. Kamiyama T, Nouchi T, Kojima S, et al. Autoimmune hepatitis triggered by administration of an herbal medicine. Am J Gastroenterol 1997;92:703–704. 34. Benninger J, Schneider HT, Schuppan D, et al. Acute hepatitis induced by greater celandine (Chelidonium majus). Gastroenterology 1999;117:1234–1237. 35. Larrey D, Vial T, Pauwels A, et al. Hepatitis after germander (Teucrium chamaedrys) administration: another instance of herbal medicine hepatotoxicity. Ann Intern Med 1992;117:129–132. 36. Mattei A, Rucay P, Samuel D, et al. Liver transplantation for severe acute liver failure after herbal medicine (Teucrium polium) administration. J Hepatol 1995;22:597. 37. Chitturi S, Farrell GC. Herbal hepatotoxicity: an expanding but poorly defined problem. J Gastroenterol Hepatol 2000;15:1093–1099. 38. Larrey D, Pageaux GP. Hepatotoxicity of herbal remedies and mushrooms. Semin Liver Dis 1995;15:183–188. 39. Stickel F, Egerer G, Seitz HK. Hepatotoxicity of botanicals. Public Health Nutr 2000;3:113–124. 40. Popat A, Shear NH, Malkiewicz I, et al. The toxicity of Callilepis laureola, a South African traditional herbal medicine. Clin Biochem 2001;34:229–236. 41. Popat A, Shear NH, Malkiewicz I, et al. Mechanism of impila (Callilepis laureola)-induced cytotoxicity in Hep G2 cells. Clin Biochem 2002;35:57–64. 42. Mokhobo KP. Herb use and necrodegenerative hepatitis. S Afr Med J 1976;50:1096–1099. 43. Stewart MJ, Steenkamp V. The biochemistry and toxicity of atractyloside: a review. Ther Drug Monit 2000;22:641–649.
44. Obatomi DK, Brant S, Anthonypillai V, et al. Toxicity of atractyloside in precision-cut rat and porcine renal and hepatic tissue slices. Toxicol Appl Pharmacol 1998;148:35–45. 45. Whiting PW, Clouston A, Kerlin P. Black cohosh and other herbal remedies associated with acute hepatitis. Med J Aust 2002;177:440–443. 46. Lontos S, Jones RM, Angus PW, et al. Acute liver failure associated with the use of herbal preparations containing black cohosh. Med J Aust 2003;179:390–391. 47. Thomsen M, Vitetta L, Sali A, et al. Acute liver failure associated with the use of herbal preparations containing black cohosh. Med J Aust 2004;180:598–9; author reply 599–600. 48. Vitetta L, Thomsen M, Sali A. Black cohosh and other herbal remedies associated with acute hepatitis. Med J Aust 2003;178:411–412. 49. Huntley A, Ernst E. A systematic review of the safety of black cohosh. Menopause 2003;10:58–64. 50. Uc A, Bishop WP, Sanders KD. Camphor hepatotoxicity. South Med J 2000;93:596–598. 51. Jimenez JF, Brown AL, Arnold WC, et al. Chronic camphor ingestion mimicking Reye’s syndrome. Gastroenterology 1983;84:394–398. 52. Antman E, Jacob G, Volpe B, et al. Camphor overdosage. Therapeutic considerations. NY State J Med 1978;78:896–897. 53. Skoglund RR, Ware LL Jr, Schanberger JE. Prolonged seizures due to contact and inhalation exposure to camphor. A case report. Clin Pediatr (Phila) 1977;16:901–902. 54. Weiss J, Catalano P. Camphorated oil intoxication during pregnancy. Pediatrics 1973;52:713–714. 55. Lahoud CA, March JA, Proctor DD. Campho-Phenique ingestion: an intentional overdose. South Med J 1997;90:647–648. 56. Riggs J, Hamilton R, Homel S, et al. Camphorated oil intoxication in pregnancy; report of a case. Obstet Gynecol 1965;25:255–258. 57. Robertson JS, Hussain M. Metabolism of camphors and related compounds. Biochem J 1969;113:57–65. 58. Nadir A, Reddy D, Van Thiel DH. Cascara sagrada-induced intrahepatic cholestasis causing portal hypertension: case report and review of herbal hepatotoxicity. Am J Gastroenterol 2000;95:3634–3637. 59. Beuers U, Spengler U, Pape GR. Hepatitis after chronic abuse of senna. Lancet 1991;337:372–373. 60. Tolman KG, Hammar S, Sannella JJ. Possible hepatotoxicity of Doxidan. Ann Intern Med 1976;84:290–292. 61. Sheikh NM, Philen RM, Love LA. Chaparral-associated hepatotoxicity. Arch Intern Med 1997;157:913–919. 62. Gordon DW, et al. Chaparral ingestion. The broadening spectrum of liver injury caused by herbal medications. JAMA 1995;273:489–490. 63. Katz M, Saibil F. Herbal hepatitis: subacute hepatic necrosis secondary to chaparral leaf. J Clin Gastroenterol 1990;12:203–206. 64. Capdevila J, Gil L, Orellana M, et al. Inhibitors of cytochrome P-450-dependent arachidonic acid metabolism. Arch Biochem Biophys 1988;261:257–263. 65. Smith BC, Desmond PV. Acute hepatitis induced by ingestion of the herbal medication chaparral. Aust NZ J Med 1993;23:526. 66. Obermeyer WR, Musser SM, Betz JM, et al. Chemical studies of phytoestrogens and related compounds in dietary supplements: flax and chaparral. Proc Soc Exp Biol Med 1995;208:6–12. 67. Adachi M, Saito H, Kobayashi H, et al. Hepatic injury in 12 patients taking the herbal weight loss AIDS Chaso or Onshido. Ann Intern Med 2003;139:488–492. 68. Hasegawa R, Futakuchi M, Mizoguchi Y, et al. Studies of initiation and promotion of carcinogenesis by N-nitroso compounds. Cancer Lett 1998;123:185–191.
559
Section IV. Toxin Mediated Liver Injury
69. Ministry of Health, L.a.W., Hepatic injury in cases taking selfimported healthfoods or non-approved drugs. L.a.W. Ministry of Health, Editor. 2002. 70. Ridker PM, McDermott WV. Comfrey herb tea and hepatic veno-occlusive disease. Lancet 1989;1:657–658. 71. Bach N, Thung SN, Schaffner F. Comfrey herb tea-induced hepatic veno-occlusive disease. Am J Med 1989;87:97–99. 72. Weston CF, Cooper BT, Davies JD, et al. Veno-occlusive disease of the liver secondary to ingestion of comfrey. Br Med J (Clin Res Ed) 1987;295:183. 73. Smith LW, Culvenor CC. Plant sources of hepatotoxic pyrrolizidine alkaloids. J Nat Prod 1981;44:129–152. 74. Tyler VE. Herbal medicine in America. Planta Med 1987;53:1–4. 75. Valla D, Benhamou JP. Drug-induced vascular and sinusoidal lesions of the liver. Baillières Clin Gastroenterol 1988;2:481–500. 76. Datta DV, Khuroo MS, Mattocks AR, et al. Herbal medicines and veno-occlusive disease in India. Postgrad Med J 1978;54:511–515. 77. Tandon BN, Tandon HD, Tandon RK, et al. An epidemic of veno-occlusive disease of liver in central India. Lancet 1976;2:271–272. 78. Steenkamp V, Stewart MJ, Zuckerman M. Clinical and analytical aspects of pyrrolizidine poisoning caused by South African traditional medicines. Ther Drug Monit 2000;22:302–306. 79. DeLeve LD, McCuskey RS, Wang X, et al. Characterization of a reproducible rat model of hepatic veno-occlusive disease. Hepatology 1999;29:1779–1791. 80. Wang X, Kanel GC, DeLeve LD. Support of sinusoidal endothelial cell glutathione prevents hepatic veno-occlusive disease in the rat. Hepatology 2000;31:428–434. 81. Shimizu I. Sho-saiko-to: Japanese herbal medicine for protection against hepatic fibrosis and carcinoma. J Gastroenterol Hepatol 2000;15(Suppl):D84–90. 82. Itoh S, Marutani K, Nishijima T, et al. Liver injuries induced by herbal medicine, syo-saiko-to (xiao-chai-hu-tang). Dig Dis Sci 1995;40:1845–1848. 83. Johnson PJ, McFarlane IG. Meeting report: International Autoimmune Hepatitis Group. Hepatology 1993;18:998–1005. 84. MacGregor FB, Abernethy VE, Dahabra S, et al. Hepatotoxicity of herbal remedies. Br Med J 1989;299:1156–1157. 85. Castot A, Larrey D. Hepatitis observed during a treatment with a drug or tea containing wild germander. Evaluation of 26 cases reported to the Regional Centers of Pharmacovigilance. Gastroenterol Clin Biol 1992;16:916–922. 86. Kouzi SA, McMurtry RJ, Nelson SD. Hepatotoxicity of germander (Teucrium chamaedrys L.) and one of its constituent neoclerodane diterpenes teucrin A in the mouse. Chem Res Toxicol 1994;7:850–856. 87. Lekehal M, Pessayre D, Lereau JM, et al. Hepatotoxicity of the herbal medicine germander: metabolic activation of its furano diterpenoids by cytochrome P450 3A depletes cytoskeleton-
560
88.
89. 90.
91.
92.
93. 94. 95.
96. 97. 98.
99.
100. 101.
102.
103. 104.
105.
106.
associated protein thiols and forms plasma membrane blebs in rat hepatocytes. Hepatology 1996;24:212–218. Loeper J, Descatoire V, Letteron P, et al. Hepatotoxicity of germander in mice. Gastroenterology, 1994. 106(2): p. 464–472. Feldmann G. Liver apoptosis. J Hepatol 1997;26(Suppl 2):1–11. Fau D, Lekehal M, Farrell G, et al. Diterpenoids from germander, an herbal medicine, induce apoptosis in isolated rat hepatocytes. Gastroenterology 1997;113:1334–1346. De Berardinis V, Moulis C, Maurice M, et al. Human microsomal epoxide hydrolase is the target of germanderinduced autoantibodies on the surface of human hepatocytes. Mol Pharmacol 2000;58:542–551. Woolf GM, Petrovic LM, Rojter SE, et al. Acute hepatitis associated with the Chinese herbal product jin bu huan. Ann Intern Med 1994;121:729–735. Horowitz RS, Feldhaus K, Dart RC, et al. The clinical spectrum of Jin Bu Huan toxicity. Arch Intern Med 1996;156:899–903. Russmann S, Lauterburg BH, Helbling A. Kava hepatotoxicity. Ann Intern Med 2001;135:68–69. Kraft M, Spahn TW, Menzel J, et al. Fulminant liver failure after administration of the herbal antidepressant Kava-Kava. Dtsch Med Wochenschr 2001;126:970–972. Escher M, Desmeules J, Giostra E, et al. Hepatitis associated with Kava, a herbal remedy for anxiety. Br Med J 2001;322:139. Teschke R, Gaus W, Loew D. Kava extracts: safety and risks including rare hepatotoxicity. Phytomedicine 2003;10:440–446. Nadir A, Agrawal S, King PD, et al. Acute hepatitis associated with the use of a Chinese herbal product, ma-huang. Am J Gastroenterol 1996;91:1436–1438. Borum ML. Fulminant exacerbation of autoimmune hepatitis after the use of ma huang. Am J Gastroenterol 2001;96:1654–1655. Lai SM, Lim KW, Cheng HK. Margosa oil poisoning as a cause of toxic encephalopathy. Singapore Med J 1990;31:463–465. Koga Y, Yoshida I, Kimura A, et al. Inhibition of mitochondrial functions by margosa oil: possible implications in the pathogenesis of Reye’s syndrome. Pediatr Res 1987;22:184–187. Sinniah R, Sinniah D, Chia LS, et al. Animal model of margosa oil ingestion with Reye-like syndrome. Pathogenesis of microvesicular fatty liver. J Pathol 1989;159:255–264. Harvey J, Colin-Jones DG. Mistletoe hepatitis. Br Med J (Clin Res Ed) 1981;282:186–187. Anderson IB, Mullen WH, Meeker JE, et al. Pennyroyal toxicity: measurement of toxic metabolite levels in two cases and review of the literature. Ann Intern Med 1996;124:726–734. Bakerink JA, Gospe SM Jr, Dimand RJ, et al. Multiple organ failure after ingestion of pennyroyal oil from herbal tea in two infants. Pediatrics 1996;98:944–947. Khojasteh-Bakht SC, Chen W, Koenigs LL, et al. Metabolism of (R)-(+)-pulegone and (R)-(+)-menthofuran by human liver cytochrome P-450s: evidence for formation of a furan epoxide. Drug Metab Dispos 1999;27:574–580.
Section IV. Toxin Mediated Liver Injury
28
OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY Thomas D. Schiano and Kristel Hunt Abbreviations ALAD d-aminolevulinic acid dehydratase BEC biliary epithelial cells CAA chloroacetaldehyde CCA chromate copper arsenate CCl4 carbon tetrachloride CEO chloroethylene oxide CSI chemical substance inventory CYP2E1 cytochrome P450 DAPM 4,4¢-diaminodiphenylmethane DDT dichlorodiphenyltrichloroethane DMAC dimethylacetamide
DMF EPA GGTP IARC ICC MDA MDI NICC NIOSH NTP
dimethylformamide environmental protection agency g-glutamyltransferase international agency for research on cancer Indian childhood cirrhosis methylenedianiline 4,4¢-methylenediphenyl di-isocyanate idiopathic copper toxicosis national institute for occupational safety and hazard national toxicology program
INTRODUCTION The pattern of injury of various hepatotoxins and the actual hepatotoxins encountered, both at home and in the workplace, have changed considerably over time. Traditionally recognized occupational hepatotoxins, such as carbon tetrachloride, are now only rarely encountered, largely because of increased awareness among physicians and those workers potentially exposed. It remains difficult to definitively prove the hepatotoxicity of a specific chemical or environmental agent, as the effects are extremely diverse and range from mildly abnormal liver chemistry tests to the development of fulminant liver failure, cirrhosis, and liver cancer. As of December 2004 there were over 25 million organic and inorganic substances registered with the Chemical Abstracts Service, with over 8 million commercially available. Of those, just over 235 000 are regulated through various national and international registries, including the US Toxic Substances Control Act (TSCA) Chemical Substance Inventory (CSI), the Occupational Safety and Hazard Administration (OSHA), the Environmental Protection Agency (EPA), the National Toxicology Program (NTP), and the National Institute for Occupational Safety and Hazard (NIOSH), among others.1 The latest update of the Pocket Guide to Chemical Hazards from February 2004 lists 677 industrial chemicals, with 228 of them listed as capable of causing liver injury.2 Although many of these industrial and occupational toxins are capable of damaging the liver, they rarely do so in the course of typical exposures: the lungs, skin, kidneys, or bone marrow are the more important targets of industrial toxins.1 A list of websites to
OSHA PAs PCBs PVC TCDD TNT TSCA VC VOD
occupational safety and hazard administration pyrrolizidine alkaloids polychlorinated biphenyls polyvinyl chloride tetrachlorodibenzodioxins trinitrotoluene toxic substances control act vinyl chloride veno-occlusive disease
access information on potential occupational and industrial toxins is given in Table 28-1. The first part of this chapter will review known human occupational, chemical, and heavy metal hepatotoxins and their patterns of injury; the remainder will review the pathophysiology and clinical manifestations of specific environmental hepatotoxins: some well known chemical hepatotoxins and their occupational uses are shown in Table 28-2.
TYPES OF EXPOSURE Inhalation Most industrial exposures occur via inhalation, and most of the chemicals are lipid soluble and able to passively cross membrane barriers. Some volatile hepatotoxins may produce severe ocular, mucous membrane and skin irritation that precludes prolonged inhalational hepatotoxicity. Inhalational exposures are fairly well regulated, with well established maximum concentrations that can be tolerated for 40 hours per week without toxic effects.1
Ingestion Uncommon in the workplace, ingestion typically occurs in the domestic setting. Although most ingestions occur accidentally, there are now increasing concerns about intentional contamination of food items by toxic agents that occur in the industrial setting through bioterrorism. At home, toxic ingestion occurs either by accident or intentionally.1 Some household products that contain hepatotoxic chemicals are listed in Table 28-3.
561
Section IV. Toxin Mediated Liver Injury
Table 28-1. Websites to Obtain Information on Potential Occupational and Industrial Toxins American Chemical Society http://www.chemistry.org/portal/a/c/s/1/home.html American College of Occupational and Environmental Medicine Physicians http://www.acoem.org Association for Occupational and Environmental Clinics http://www.aoec.org/ EPA http://www.epa.gov/fedrgstr/EPA-PEST/2005/March/Day11/p4466.htm Haz-Map. Occupational Exposure to Hazardous Agents (National Library of Medicine) http://hazmap.nlm.nih.gov International Agency for Research on Cancer http://www.1arc.fr/ National Center for Toxicological Research of the Food and Drug Administration http://www.fda.gov/nctr National Institute for Occupational Safety and Health of the Centers for Disease Control and Prevention http://www.cdc.gov/niosh/homepage.html National Institute of Environmental Health Sciences of the National Institutes of Health http://www.niehs.nih.gov National Toxicology Program http://ntp-server.niehs.nih.gov University of Kansas Resource and Learning Center. Environmental, Industrial and Occupational Toxicology Information http://library.kumc.edu/omrs/subjects/toxic.html
Table 28-3. Household Products that may be Hepatotoxic Household use
Chemical nature
Carburetor cleaner Christmas tree lights Dry cleaning fluids Furniture polishes and waxes
Chlorobenzene Methylene chloride Chlorinated aliphatic compounds Antimony Nitrobenzene Chlorobenzenes
Moth balls Paint products: Brush cleaners Paints Plasticizers, lacquers, resins Removers, paint, wax Plastic menders, greasers Plasticizers Shoe cleaners Spray repellant Stamping inks Toilet bowl blocks
Cresols Arsenic Varied Chlorinated aliphatic compounds, dimethylformamide Ethylenedichloride Phthalates Aniline Nitrobenzene Vinyl chloride Phenol Paradichlorobenzene
From Zimmerman HJ. Occupational hepatotoxicity. In: Zimmerman HJ. Hepatotoxicity: the adverse effects of drugs and other chemicals on the liver. Philadelphia: Lippincott Williams & Wilkins, 1999
Table 28-2. Some Important Chemicals, Their Uses and Associated Hepatotoxicity Chemical
Uses
Hepatic responses
Arsenic and inorganic salts Beryllium
As pesticides and alloys; in production of dyes, ceramics, drugs, fireworks, paint, petroleum, ink, and semiconductors In alloys, cathode ray tubes, ceramics, electrical equipment, gas mantles, missiles, nuclear reactors, and refractory materials As degreasers, fat processors, fire extinguisher, fumigant, production of solvents: in fluorocarbons, ink, insecticides, lacquer, propellants, refrigerants, rubber and wax As solvent, degreaser, cement component; in production of adhesives, deodorants, detergents, emulsions, fats, glue, lacquer, oil, paint, polish, shoe cream, varnish remover, waxes; in histology laboratories In munitions, pyrotechnics, explosives, smoke bombs, fertilizers, rodenticides, bronze alloys, semiconductors, and luminescent coatings As copper etcher, forensic and biology laboratory chemical; in batteries, colored glass, disinfectants, drugs, dyes, explosives, matches, photography chemicals, and tanneries In cable insulation, dyes, electric equipment, herbicides, lacquers, paper treatment, plasticizers, resins, rubber textiles, flameproofer, transformers, and wood preservation Contaminant of commercial preparations of 2,4,5-trichlorophenoxyacetic acid, polychlorinated biphenyls, and other chlorinated compounds As dry-cleaning agent, fumigant, solvent, degreaser; in production of gaskets, lacquers, paints, phosphorus, resins, varnish, wax As solvent, degreaser, chemical intermediate, fumigant; in production of cellulose esters, gums, rubber, soap, vacuum tubes, wax, wool As explosive As chemical intermediate and solvent; in production of polyvinyl chloride and resins
Acute hepatocellular injury cirrhosis; angiosarcoma Granulomata
Carbon tetrachloride
Dioxane
Phosphorus (yellow)
Picric acid (2,4,6trinitrophenol) Polychlorinated biphenyls 2,3,7,8-Tetrachlorodibenzo-p-dioxin Tetrachloroethane Tetrachloroethylene 2,4,5-Trinitrotoluene Vinyl chloride
Acute hepatocellular injury; cirrhosis
Subacute hepatocellular injury
Acute hepatocellular injury
Acute hepatocellular injury
Subacute hepatocellular injury; cirrhosis
Porphyria cutanea tarda Acute hepatocellular injury Acute hepatocellular injury Acute and subacute hepatocellular injury Fibrosis, non-cirrhotic portal hypertension, cirrhosis, angiosarcoma, carcinoma
Adapted from Gitlin N. Clinical aspects of liver diseases caused by industrial and environmental toxins. In: Zakim D, Boyer TD. Hepatology: A Textbook of Liver Disease, 2nd edn. Philadelphia: WB Saunders Company; 1990.
562
Chapter 28 OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY
Mucosal Absorption The chemical structure and lipid or water solubility of the toxin are the major determinants of absorption.1 Hepatotoxic agents are infrequently absorbed in this manner.
picion needs to be maintained, especially if the temporal relationship is appropriate.
HEPATOTOXIC CHEMICALS
DIAGNOSIS OF CHEMICAL-INDUCED LIVER INJURY
HALOGENATED AROMATIC HYDROCARBONS Polychlorinated Biphenyls
When patients present with liver dysfunction or abnormal liver chemistry tests, it often requires a high index of suspicion to recognize that this may be the result of exposure to an occupational or environmental toxin. Anecdotal reports tend to concentrate on unique, severe, or fatal outcomes, so the overall incidence of environmental and occupational hepatotoxic injury is probably underestimated. Many cases probably go unrecognized because they are never suspected, not properly investigated, or simply not reported.3 A simple severe exposure often leads to an acute clinical presentation, whereas prolonged exposure of a lesser degree may lead to subacute or chronic liver disease.3 The first clinical manifestations of hepatotoxicity can often be non-specific, with constitutional symptoms occurring much earlier than jaundice. The patient may otherwise be healthy but exposed to a known environmental hepatotoxin. When there are no other risk factors for chronic liver disease the diagnosis is usually easy to make in this setting, but unfortunately this scenario is very uncommon. A comprehensive history should be taken in any patient presenting with an acute or chronic cryptogenic liver disease to exclude an occupational or industrial exposure. Occupations that may entail exposure to potentially hepatotoxic chemicals are listed in Table 28-4. Other causes of liver disease must be excluded by careful assessment of clinical, radiologic, biochemical, and serologic testing. Patients with any form of chronic liver disease may develop decompensated liver disease in the presence of a toxic exposure. Some uncommon histological patterns of liver injury, such as angiosarcoma, hepatoportal sclerosis, zonal necrosis, and vascular injury, should raise the suspicion of occupational or environmental hepatotoxicity. The major goals of the clinical evaluation of patients with suspected chemical-induced liver dysfunction are to recognize that this is indeed what is occurring by establishing that there are sufficient criteria to assign a causal relationship to a particular agent. At the same time, other causes of acute and chronic liver diseases must be excluded, as occupationally and environmentally induced liver disease can clinically and histologically mimic almost any known liver disease (Table 28-5). Some drugs and chemicals may themselves be deposited in the liver. Usually this is an incidental finding and has little clinical significance apart from indicating prior exposure. Examples include gold compounds, thorium dioxide, and titanium.
Polychlorinated biphenyls (PCBs), with their significant flame-retardant and insulating properties, have been used since the 1930s in a
Management Early recognition is of utmost importance in the management of chemical-induced hepatotoxicity, as it is in drug hepatotoxicity. As with drug hepatotoxicity, continued exposure to the toxic agent once jaundice appears may lead to acute liver failure and hepatic decompensation. In this regard, anticipation and proactively seeking evidence for hepatotoxicity with the appropriate agencies, such as the Poison Control Center or OSHA, is imperative. Even when a substance is not known to cause hepatotoxicity a high index of sus-
Table 28-4. Occupations that Entail Exposure to Hepatotoxic Chemicals Artificial pearl makers Airplane hanger employees Burnishers Cement (rubber, plastic) makers Chemists Chemical industry workers Chlorinated rubber makers Cobblers Color makers Core makers Degreasers Dry cleaners Dye makers Dyers Electrical transformer and condenser makers Electroplaters Enamel makers Extractors, oil and fats Fire extinguisher makers Galvanizers Garage workers Gardeners (insecticides) Gas (illuminating) workers Glass makers Glue workers Ink makers Insecticide makers Insulators (wire) Lacquer makers and lacquerers Leather workers Linoleum makers Paint removers makers and users Paraffin workers Perfume makers Petroleum refiners Pharmaceutical workers Photographic material workers Polish (metal) makers and users Printers Refrigerator workers Resins (synthetic makers) Rubber workers Shoe factory workers Soap makers Spreaders (rubber works) Straw hat makers Thermometer makers Tobacco dinicotizers Varnish workers Waterproofers Wax makers From: Zimmerman HJ. Occupational hepatotoxicity. In: Zimmerman HJ. Hepatotoxicity: the adverse effects of drugs and other chemicals on the liver. Philadelphia: Lippincott Williams and Wilkins, 1999
563
Section IV. Toxin Mediated Liver Injury
Table 28-5. Liver Injury Patterns for Occupational and Environmental Toxins Category
Examples
Acute hepatocellular injury:
Arsenic, cocaine, copper, halogenated aliphatic hydrocarbons, nitrobenzene, pyrrolizidine alkaloids, tetrachlorethane, TNT Ferrous sulfate, phosphorus Beryllium Arsenic, carbon tetrachloride, mushroom poisoning Arsenic, thorotrast, vinyl chloride Thorotrast Berxllium, copper, cyanobacterial toxins, methylene dianiline, paraquat, toxic rapeseed oil 1,1,2,2-tetrachloroethane, arsenic, cadmium, carbon tetrachloride, pyrrolizidine alkaloids, TNT Vinyl chloride Beryllium, copper Aflatoxin, thorotrast Arsenic, copper, vinyl chloride Copper 1,1,1-trichloroethane, 1,1,2,2tetrachloroethane, alfatoxins, carbon tetrachloride, cocaine, dimethylacetamide, dimethylformamide, margosa oil toxicity, phosphorus Thorotrast, vinyl chloride
Zone 1 Zone 2 Zone 3 Angiosarcoma Cholangiocarcinoma Cholestasis
Cirrhosis
Fibrosis, septal and subcapsular Granulomas Hepatocellular carcinoma Hepatoportal sclerosis Mallory’s hyaline Microvesicular steatosis
Peliosis hepatis Pigment deposition: Anthracite Gold Thorotrast Titanium Veno-occlusive disease
Noted in coal miners Gold compounds used in the treatment of arthritis Occupational exposure Pyrrolizidine alkaloids
variety of industrial applications. Concern about their presence in the environment began in the 1960s, when significant concentrations of PCBs were found in wildlife in Sweden. Subsequent research has shown their long-term persistence and bioaccumulation in the food chain, especially in the fatty tissues of fish and sea mammals. Since the use of these compounds was banned by the US EPA in 1977, the concentrations of PCBs have drastically declined in all environmental reservoirs.3,4
Acute Effects In the 1940s a number of cases of acute PCB toxicity were reported, with symptoms of anorexia, fatigue, nausea, peripheral edema, and rarely jaundice; over 50% of individuals, however, died of acute or subacute hepatic necrosis or of subsequent cirrhosis. Histologically this was described as severe zone 3 necrosis. Since then, there have been two epidemics of acute PCB toxicity, both as a result of ingestion of contaminated rice oil: in 1968 in Japan (Yusho), and in 1979 in Taiwan.5 Both epidemics were associated with liver dysfunction and chloracne, as well as low birthweights as a result of fetal exposure. Although elevated mortality from liver diseases was noted within 3 years of exposure, more recent studies have attributed the hepatotoxicity to contaminating dibenzofurans (PCDFs) instead.4,5
564
Chronic Effects Given the significant carcinogenicity of PCBs in laboratory animals and the initial reports of increased rates of liver cancer in the setting of acute toxicity, in the 1980s PCBs were classified as probable human carcinogens by all major registries. A wealth of literature on occupational exposure, primarily from workers employed in electrical equipment manufacturing with years of PCB exposure and more than 30-year follow-up, however, has not shown any increase in acute or chronic health effects beyond skin and eye irritation and perhaps transient liver enzyme elevation.6 In particular, no increased risks of liver cancer – or any other tumors, for that matter – have been observed.6,7 Based on these data, it is unlikely that PCB exposure from environmental sources can pose a significant risk to humans, and indeed, in 2003, the EPA officially downgraded the perceived cancer risk of PCBs.8 As potent inducers of the enzyme P450 system, however, PCBs can significantly enhance the hepatotoxic effects of other chemical agents or drugs.
Tetrachlorodibenzodioxins (TCDD) Dioxins are a family of chlorinated aromatic compounds formed primarily during combustion of chlorine-containing materials such as PCBs, chlorophenols, and phenoxy herbicides. Like PCBs, they can persist in the environment for years. In experimental animals, dioxins are powerful hepatotoxins with potent P450-inducing abilities and strong carcinogenic potential. Humans, however, are more protected from the hepatotoxic effects than animals because of a lower affinity with dioxin in human cytoplasmic receptors.9 The major identified and chronicled dioxin exposures resulted from the explosion of a herbicide factory in Seveso, Italy, in 1976, and through dioxin-contaminated herbicide use during the Vietnam war (‘Agent Orange’). In the former, progression to porphyria cutanea tarda occurred in some individuals. Despite no early evidence for liver injury, there has been an increased incidence of cholangiocarcinoma and hepatocellular carcinoma, especially in women, in the Seveso cohort after 20 years of follow-up; no direct causal effect of dioxins has been formally established, however.9 No demonstrable hepatic injury or predisposition to liver cancer has been found among the military personnel exposed to Agent Orange, despite high TCDD levels found in blood.10
NITROAROMATIC COMPOUNDS A variety of nitroaromatic compounds have been shown to be hepatotoxic in humans, including nitrobenzene and dinitrobenzene, dinitrophenol, and perhaps one of the best-known hepatotoxins, trinitrotoluene (TNT). Tight regulation of their use in the workplace has dramatically limited the number of any recent occurrences. Nitrobenzene leads to zonal necrosis and is often used as a hepatotoxin in experimental animals. It is primarily used in the production of aniline, and rare instances of human hepatotoxicity have occurred in the industry as a result of inhalation of toxic fumes. A potent hepatic carcinogen in laboratory animals, nitrobenzene is deemed to be a potential human carcinogen as well.2 Dinitrophenol is used in the dye industry and in the manufacturing of photograph developers, as well as a fungicide. The most common human exposure is in the occupational setting by inhalation or dermal contact of workers involved in manufacturing of the
Chapter 28 OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY
compound; there were reports of dinitrophenol toxicity via ingestion in the 1930s, when it was used in the treatment of obesity. Despite its potential for accumulation in the environment, there have been no reports of environmental toxicity in humans. Its mechanism of injury is primarily through its ability to interfere with oxidative phosphorylation, resulting in cholestasis and, rarely, hepatic necrosis.3 Trinitrotoluene (TNT) has been used primarily in the military setting as an explosive in bombs and grenades. The first reports of TNT hepatotoxicity date back to World War I, when severe liver disease with more than a 25% fatality rate was reported among munitions workers in England; in the US, over 400 people were purported to have died as a result of TNT-related liver injuries and aplastic anemia. Another peak of hepatotoxicity occurred during World War II, again through exposure to TNT-containing shells.3,11 Since then, the incidence of toxic effects associated with handling of TNT has decreased sharply, thanks to enhanced protective measures and better ventilation. The manifested hepatic toxicity is typically subacute, with symptoms presenting months after exposure. Initial symptoms of anorexia, nausea, and fatigue that may progress to jaundice are associated with minimal liver enzyme abnormalities. Thereafter, symptoms tend to evolve into one of two classic patterns: some patients develop massive hepatic necrosis within days of the initial symptoms and succumb to liver failure; others develop portal hypertension and ascites, with liver injury ranging from acute or subacute hepatic necrosis to macronodular cirrhosis.3,11 The proposed mechanism of injury is primarily through an increase in free radical levels (superoxide and hydrogen peroxide). The large variability in apparent susceptibility also suggests an idiosyncratic preferential conversion of TNT to toxic metabolites in certain predisposed individuals.
NITROALIPHATIC COMPOUNDS (NITROPARAFFINS) The nitroparaffins (nitromethane and nitroethane, 1-nitropropane and 2-nitropropane) are used primarily as industrial solvents in inks, paints, adhesives, varnishes, polymers, and synthetic materials; 2nitropropane is also used as a component in explosives and propellants, as well as in fuels for internal combustion engines. The primary routes of exposure are inhalation, ingestion, and dermal contact. In the 1970s and 1980s several cases of fulminant liver failure were attributed to 2-nitropropane; since its classification in 1999 by the International Agency for Research on Cancer (IARC) as a possible human carcinogen, no further cases have been reported. In general, nitroparaffins appear less toxic than nitroaromatic compounds.3
CHLORINATED ETHYLENES Vinyl Chloride (VC) Vinyl chloride (VC), historically used as a solvent, propellant, and refrigerant, is now primarily used for the production of polyvinyl chloride (PVC), a material used to manufacture automotive parts and accessories, furniture, packaging materials, pipes, wall coverings, and wire coatings. During the first five decades of its use, no clinically significant toxicity was noted. However, in 1974 a series of cases of angiosarcoma occurring in workers in rubber plants exposed
to VC was reported – a landmark discovery in occupational medicine. Despite its known toxicity, vinyl chloride remains a crucial component of the plastics industry: in 1998, 27 million tons of PVC was produced, accounting for 20% of all plastics production.12 The main route of occupational exposure to VC is via inhalation, and occurs primarily in VC manufacturing plants. Since its recognition in the 1970s as a significant hepatocarcinogen, workplace exposure standards have been tightly regulated, and indeed there have been only rare recent reports of VC-related occupational hepatotoxicity since that time. The exposure of the general population is primarily through accidental spills into the environment, or by chronic exposure to the fumes in the vicinity of VC industry or waste disposal sites. Small amounts of vinyl chloride can also be found in foodstuffs, cosmetic or pharmaceutical products in certain types of PVC packaging, although typically not in sufficient concentrations to lead to significant toxicity.12
Mechanism of Injury After inhalation, VC undergoes oxidation by the cytochrome P450 (CYP2E1) system to form chloroethylene oxide (CEO), which then rapidly forms chloroacetaldehyde (CAA). Both CEO and CAA can alkylate nucleic acid bases and form DNA adducts with strong mutagenic and promutagenic properties. In highly exposed workers with angiosarcoma, point mutations (primarily base-pair substitutions) have been detected both in p53 and K-ras oncogenes.13
Differences in Susceptibility There are significant interspecies and interindividual variations in both CYP2E1 and the glutathione S-transferase isoenzymes (involved in the detoxification process of CEO and CAA), leading to large differences in the dose required to produce VC-related hepatotoxicity. Younger age and female gender, as well as certain HLA alleles, may be associated with higher levels of toxicity. Recent studies have demonstrated clear synergistic effects of both alcohol and viral hepatitis in the setting of VC exposure, leading to a substantially higher risk of hepatocellular carcinoma and cirrhosis in exposed patients who are infected with either hepatitis B or C virus and who consume excessive amounts of alcohol.14 Several other environmental toxins have also been shown to work synergistically with VC: concurrent aflatoxin and PCB exposures have both led to increased rates of liver tumor formation.15
Acute and Chronic Effects The acute effects of vinyl chloride are primarily central nervous system related. With massive exposure, VC leads to cardiac arrhythmias and cardiovascular collapse. More chronic exposure, of the order of months to years, can lead to so-called ‘vinyl chloride disease’, including disorders of skin and connective tissue such as Raynaud’s phenomenon, neurologic symptoms such as headaches, dizziness and blurry vision, as well as hepatosplenomegaly associated with various types of liver pathology.16 The earliest findings of VC exposure are focal hepatocellular hyperplasia and focal mixed hyperplasia. Subcapsular fibrosis, progressive portal fibrosis, and an increase in intralobular connective tissue are precursors to the development of angiosarcoma, which is frequently multicentric.17 Indeed, the association between vinyl
565
Section IV. Toxin Mediated Liver Injury
chloride and angiosarcoma is well described, with over 200 documented cases via either environmental or occupational exposure (Figure 28-1). Other substances associated with angiosarcoma formation include thorium dioxide, long-term exposure to inorganic arsenic, Fowler’s solution in the treatment of psoriasis, and androgens. Angiosarcoma is often peripherally located in the liver and may be associated with a hepatic bruit. Liver fibrosis is over five times more common in individuals exposed to vinyl chloride, and is often associated with significant portal hypertension in the absence of cirrhosis (hepatoportal sclerosis) (Figure 28-2).18,19 Substantially higher risks of hepatocellular carcinoma have also been shown in all the
major cohorts, with a standardized mortality ratio for HCC-related deaths of 1.35 in a pooled analysis.20–24
AROMATIC HYDROCARBONS Aromatic hydrocarbons such as benzene, toluene, xylene, and naphthalene are typically used as solvents and glues and in the manufacture of plastics. In general they are not associated with significant hepatotoxicity, although minor liver enzyme elevations and hepatic steatosis have been described. A recent report has described selflimited hepatitis in association with toluene handling.25
HALOGENATED ALIPHATIC HYDROCARBONS Carbon Tetrachloride (CCl4)
Figure 28-1. Angiosarcoma. Liver needle biopsy showing anastomosing vascular channels that are lined by atypical endothelial cells (arrow). Hematoxylin and eosin, original magnification 20¥. (Courtesy of Dr M.I. Fiel.)
Figure 28-2. Hepatoportal sclerosis. Photomicrograph of wedge biopsy of liver showing a portal tract (arrow) with pronounced herniation of portal veins (asterisks) into adjacent parenchyma. Marked centrilobular sinusoidal dilation is also present (arrowheads). In other areas, portal tracts are severely fibrotic and the portal veins are sclerotic. Hematoxylin and eosin, original magnification 10¥. (Courtesy of Dr M.I. Fiel.)
566
CCl4 is the best-studied occupational hepatotoxin. It provides the prototype for several toxic phenomena: severe liver injury with zone 3 necrosis and steatosis, injury greatly enhanced by alcohol, and injury caused primarily by free radical metabolites causing direct damage to hepatocyte membranes. Until the 1920s, when its potent hepatotoxicity was first noted, it was widely used as a cleaning agent, solvent, fire extinguisher, and grain fumigant, as well as an intermediate for the synthesis of chlorofluorocarbons. Carbon tetrachloride poisoning typically follows inhalation of the vapor in poorly ventilated spaces, but can also occur after ingestion of contaminated foodstuffs or ground water, or via skin absorption.1,3 It is no longer produced on a large scale in the US, and thus occupational and household exposures typically only occur in countries that still allow its use.26–28 Industrial exposure usually occurs by inhalation of the fumes in a poorly ventilated environment. Alcohol ingestion and the use of barbiturates appear to enhance hepatic susceptibility to CC14, possibly via their effects on the P450 system.1,3
Acute Effects Acute carbon tetrachloride poisoning is manifested as a multisystem disorder, with early CNS, renal, and gastrointestinal toxicity. The majority of the deaths attributable to acute CCl4 toxicity are due to acute tubular necrosis and renal failure that frequently lead to pulmonary edema and congestive heart failure. Gastrointestinal symptoms, including nausea, vomiting, diarrhea, and abdominal pain, sometimes also associated with hemorrhagic gastritis, typically occur within the first 24 hours. Features of liver disease manifest within 24–48 hours, with massive hepatomegaly, jaundice, and a bleeding diathesis with spontaneous hemorrhage; ascites and hepatic encephalopathy occur in severe cases.3 The aminotransferases rise to extremely high levels, with AST higher than ALT. The bilirubin levels also soar, reflecting hepatocyte necrosis, hemolysis, disruption of the cytochrome P450 system, and an impaired clearance of bilirubin by the kidneys. In survivors recovery is rapid, with normalization of liver enzymes within 2 weeks. If the hepatic injury is so severe as to result in death, it will typically occur within the first 10 days. The histologic findings in acute toxicity are primarily those of zone 3 necrosis and steatosis, preceded by prominent ballooning and swollen granular cytoplasm with pale nuclei.3 Treatment is mainly supportive.
Mechanism of Injury Great intra- and interspecies variability in hepatotoxic potential has been noted, owing primarily to variability in the mixed-function
Chapter 28 OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY
oxidase system (P450 cytochrome system). In general, males as well as older individuals have been found to be more susceptible to injury. CC14 appears to damage the hepatocellular membrane, with a resultant loss of enzymes and a flux of ions out of the hepatocyte. The CC14 is then metabolized to a toxic free radical by cytochrome P450 in the endoplasmic reticulum. The toxic free radical, via perioxidation of lipids, disrupts membranes further and interferes with the synthesis and transport of lipoproteins, causing fat to accumulate within the hepatocyte.29
1,1,1-Trichloroethane This was widely recognized as a hepatotoxin during the 1940s, when it was used as a solvent for varnish in airplane covers. It is still used as a solvent, and there have been rare reports of its association with fatty liver and cirrhosis.30,31
1,1,2,2-Tetrachloroethane 1,1,2,2-Tetrachloroethane is used as a chemical intermediate, primarily in the manufacture of trichloroethylene and tetrachloroethylene. There have been many reports of human hepatotoxicity causing abnormal liver chemistry tests and acute hepatitis with chronic exposure.1,3
Tetrachloroethylene (Perchloroethylene) This compound is used primarily in dry cleaning and textile processing, as a chemical intermediate in the production of fluorocarbons, and as a metal degreasing agent. It is also used as a solvent, as a pesticide intermediate, and as an antihelminthic.1,3 Transient hepatotoxicity in humans has occurred with single exposures, and abnormal liver enzymes and cirrhosis have been observed with chronic long-standing exposure.
N-SUBSTITUTED AMIDES Dimethyl Acetamide Dimethylacetamide (DMAC) is used as a solvent in the manufacturing of synthetic fibers, some resins and plastics, as well as in film and coating formulations. It is readily absorbed by inhalation as well as dermally. Acute toxicity leads to elevated transaminase levels and hepatomegaly, with focal necrosis seen on liver biopsy.32 Chronic exposures have resulted in microvascular steatosis as well as lipofuscin/hemosiderin accumulation in Kupffer cells, with minimal recovery after withdrawal of the toxin. Biological monitoring of DMAC in exposed workers by measurement of its urinary metabolite (monomethyl acetamide) is recommended.33
Dimethylformamide Dimethylformamide (DMF) has excellent solvent properties and has been used extensively in the production of synthetic leather and resin. Hepatotoxicity occurs in a dose–response fashion after exposure through inhalation, ingestion, and/or skin exposure. Acute exposure typically leads to modest elevations in aminotransferase levels which resolve after withdrawal of the exposure. Chronic exposure leads to microvesicular steatosis, and in severe cases hepatic necrosis can also occur. Both hepatitis B virus and alcohol use have been shown to have a synergistic effect in causing hepatotoxicity with DMF exposure.34,35
METHYLENEDIANILINE Methylenedianiline (MDA) was the agent responsible for a wellknown epidemic of hepatic injury in Epping, UK, in 1965, known as Epping jaundice.36 Bread baked from flour contaminated with MDA caused cholestatic hepatitis in 84 people. Within hours of consuming the bread, two-thirds of the affected individuals developed abdominal pain, fever and rash, and ultimately jaundice. The remaining one-third had minimal or no clinical symptoms. Elevations in serum bilirubin and alkaline phosphatase levels were pathognomonic, but marked transaminase elevations were seen in some cases as well. A similar mixed cholestatic–hepatocellular injury pattern was seen on liver biopsy, with predominant bile stasis, portal inflammatory infiltrates, and varying degrees of hepatocyte necrosis. The majority of patients recovered within 4–6 weeks, but some had prolonged jaundice lasting a few months; several patients had slightly elevated aminotransferase levels persisting for up to 2 years. A follow-up study nearly 40 years later found no increase in mortality rates among the people exposed, and no cases of death specifically from chronic liver disease or from liver cancer.37 Additional reported exposures to MDA have occurred in the occupational setting, with reports of toxicity following ingestion, inhalation, and dermal contact. In each case the hepatitis was self-limited. In rats, glutathione depletion has been shown to drastically enhance and accelerate MDA hepatotoxicity.38 4,4¢-Methylenedianiline (4,4¢-diaminodiphenylmethane; DAPM) is an aromatic diamine used in the production of polyamides, epoxy resins, and 4,4¢-methylenediphenyl di-isocyanate (MDI), the most widely used isocyanate in the production of polyurethanes. These polymers are made into insulation materials, seat cushions, wire coatings, aircraft parts, and medical devices, such as dialysis tubing, orthopedic and odontologic implants, intra-aortic balloons, and vascular grafts.39 In humans, accidental or acute occupational exposure causes jaundice, cholangitis, cholestasis, toxic hepatitis, and skin rash. Studies have demonstrated that acute administration of DAPM to rats produced striking changes in several biliary constituents, followed by necrosis and sloughing of biliary epithelial cells (BEC), cholestasis, and cholangitis. Effects on the biliary system occurred prior to hepatocellular injury, as evidenced by minimal increases in serum indicators of hepatic injury. Furthermore, alterations in biliary constituents were preceded by ultrastructural changes in BEC mitochondria and loss of luminal microvilli, whereas morphologic effects on tight junctions between BEC or between hepatocytes were absent.40 The identity of the DAPM metabolite(s) responsible for injury to BEC remains unknown.
PESTICIDES Dichlorodiphenyltrichloroethane (DDT) DDT, first introduced during World War II, was the first insecticide to be used on a mass scale. As it was highly beneficial for food production, billions of pounds were released into the environment over the next 30 years. Its favorable properties of extreme stability and slow biodegradability also led to its persistence in the environment; indeed, studies have shown that residual compound can be found as long as 10 years after a single application. It is found in the fat stores of animals and is processed up the food chain, leading to significant concentrations of DDT in humans.3
567
Section IV. Toxin Mediated Liver Injury
Acute hepatotoxicity has occurred in rare accidental ingestions of large amounts of DDT, leading to massive hepatic necrosis and hepatic failure. No clear hepatic injury patterns have been described after decades of occupational exposure, and despite its perceived carcinogenicity (it was banned in the US in 1972 for that reason), no clear relationship between tissue DDT levels and cancer risk has been found.41 Similarly to PCBs, DDT is a powerful cytochrome P450 inducer and can enhance the hepatotoxicity or carcinogenicity of other chemicals or drugs.
Paraquat (Dichlorodiphenyltrichloroethane) Paraquat, a bipyridyl, is a widely used herbicide. There are no known hepatic sequelae of chronic paraquat exposure: all known cases of hepatotoxicity have occurred in the setting of substantial paraquat ingestion, usually in a suicide attempt, or rarely through massive skin exposure. The hepatic manifestations of paraquat poisoning usually present within 24–48 hours, and manifest initially as hepatocellular injury leading to zone 3 necrosis. If the individual survives the initial attack, the lesion becomes primarily cholestatic, with destruction of bile ducts. Paraquat poisoning has a high case fatality rate – 50–70% – usually due to initial overwhelming systemic toxicity or to the rapid development of pulmonary fibrosis over the ensuing weeks.3,42
Chlordecone (Kepone) Despite significant hepatic injury and carcinogenicity in laboratory animals exposed to chlordecone, substantial occupational exposures have not resulted in irreversible liver damage. Massive exposure in a manufacturing plant in Virginia in 1970 led to substantial neurologic and testicular damage as well as hepatosplenomegaly – only mild liver dysfunction and minimal steatosis were found on liver biopsy. With cessation of exposure, the liver dysfunction normalized.3,43
METALS Arsenic Arsenic was the first chemical agent with ascribed hepatotoxicity to be identified and was an important early model of experimental liver disease in animals.3,44 Naturally occurring arsenic is ubiquitous in soils, water, and even foods. In industry, arsenic is an essential ingredient in insecticides, herbicides and fungicides, wood preservatives, and dyestuffs, and also is a staple in veterinary medicine for the treatment of parasitic diseases. Chronic arsenic exposure in vineyard workers, miners, and farmers was a frequent cause of chronic liver disease in the early 20th century, but since its identification as a hepatotoxin occupational toxicity has significantly decreased.44
Mechanism of Injury Most of the toxicity of arsenic stems from its ability to inactivate thiol-containing enzymes and substitute arsenate for phosphate groups in a number of biochemical reactions, thus interfering with several key processes such as cellular respiration, gluconeogenesis, glucose uptake, and glutathione metabolism.44 The mechanism for the carcinogenic effects of arsenic is not completely understood, but is probably a combination of induced chromosomal aberrations, oxidative stress, and altered growth factors. Additional mechanisms,
568
including gene amplification, altered DNA repair, enhanced cell proliferation, and suppression of the p53 oncogene, are currently under investigation.45 It is noteworthy that the carcinogenicity of arsenic compounds has not been reproducible in animal models, lending support to individual variations in susceptibility to arsenic toxicity. In additional, other host factors, such as age, sex, and nutritional status (in particular selenium deficiency), appear to be important.46
Acute Effects Acute arsenic poisoning is generally due to accidental or planned ingestion. Although larger doses (1–3 g) are common, severe toxicity can occur with as little as 1 mg. Ingestion of toxic quantities leads to symptoms within 30–60 minutes, with prominent gastrointestinal, neurological, and vascular effects and concomitant hepatic injury; death can ensue in 1–3 days. Histologically there is severe steatosis and varying degrees of necrosis, with preferential distribution in either zones 1 or 3.44
Chronic Effects Historically, occupational arsenic exposure occurred mainly through the widespread use of arsenical pesticides in agriculture and vineyards. Chromate copper arsenate (CCA) – in treated wood in homes and as playground equipment – has recently come under intense media scrutiny, leading to the involvement of the US EPA. Arsenic fumes are encountered in mining, smelting, and metallurgy, as well as in decorative glass-making. More recently, arsenic has become an essential component of semiconductor chips. In industries with known arsenic exposure workers are usually closely monitored to prevent clinical toxicity. The majority of human arsenic exposure is through the use of contaminated water. In certain areas of the world, such as West Bengal, India, and Bangladesh, it is an enormous health hazard, with millions of people exposed to unacceptably high levels of arsenic. Epidemiologic studies from these regions have documented the dose-related effects of arsenic on the endocrine, cardiovascular, and central nervous systems, and especially its tumorigenic effects on skin, bladder, lung, and liver.47–49 Hepatic lesions attributable to chronic arsenic ingestion include hepatoportal sclerosis,50 cirrhosis, and angiosarcoma,51 as well as a possible association with hepatocellular carcinoma.52
Phosphorus Yellow phosphorus has long been known to be a hepatotoxin and was used experimentally over a century ago. Acute poisoning with phosphorus, a primary constituent of matches and firecrackers, was not infrequent in the occupational or household setting until its ban in 1942. Since then, the toxicity has occurred primarily by ingestion of rat poison or firecracker contents, either accidentally or with suicidal intent. Acute phosphorus poisoning is a severely toxic syndrome, with initial severe gastrointestinal manifestations such as nausea, vomiting, gastroduodenal ulceration, and neurologic involvement, followed by a latent relatively symptomatic stage and then by hepatic and renal failure. Diagnosis is suggested by a garlic odor on the breath. Histological findings include a characteristic periportal fatty infiltration and hepatic necrosis.3 Aminotransferases are elevated two- to sixfold. Timely gastric lavage is essential.
Chapter 28 OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY
Beryllium Beryllium is used predominantly in the manufacture of electronics and electrical equipment, and in defense and nuclear weapons technology. Historically, exposures have been by inhalation, with only rare case reports now that industrial hygiene has significantly improved. The two primary lesions are zone 2 necrosis and granulomatous lesions. Beryllium compounds have led to granulomatous lesions in both the liver and the lung.3
Iron Though not directly toxic as an organic molecule, iron in excess concentrations exhibits its hepatotoxicity by the formation of free radicals and lipid peroxidation, resulting in hepatocyte membrane destruction and necrosis. Acute toxicity is seen primarily in the home, occurring either via accidental ingestion (especially among children) or as a suicidal gesture. Increased tissue loads over a prolonged period can lead to the most important chronic toxicity of iron, hemosiderosis. It also contributes to the hepatic injury in alcoholic liver disease, non-alcoholic steatohepatitis, as well as in viral hepatitis.3
Mechanisms of Injury In acute overdose ferrous and ferric ions react with lipid peroxidases, producing free radicals. In contrast to most other toxic agents, the hepatic necrosis in iron-induced hepatotoxicity occurs mostly in zone 1. The periportal region may be primarily affected owing to its high concentrations of both iron and oxygen, substrates for free radical generation. Similarly, in chronic injury iron accumulates primarily in the hepatocytes of zone 1. Initially, cytosolic storage of ferritin and hemosiderin is maximized, but once the tissue stores reach their capacity, free iron is released to produce injury both via the formation of free radicals as well as via catalysis of hydroxyl radical formation.53
Clinical Manifestations Hepatic injury typically develops within 1–3 days after overdose. It is usually characterized by jaundice, elevated aminotransferase levels, and hypoprothrombinemia; in patients with hepatic failure the mortality rate is approximately 50%. However, the majority of patients presenting with iron poisoning will have minimal liver damage. It appears that iron exhibits dose-related toxicity: doses less than 20 mg/kg have been shown to produce no significant injury, whereas doses between 20 and 60 mg/kg can result in mild to moderate injury and doses between 60 and 200 mg/kg in serious injury; doses above 200 mg/kg are likely to lead to death.54 Orthotopic liver transplantation has been performed in the setting of acute iron intoxication.55
Copper Despite its widespread use in electrical wiring and electronics, and in transportation and industrial machinery, household and environmental exposures lead to more episodes of copper toxicity than occupational exposures. Indeed, the only documented occupational hepatotoxicity has been among vineyard workers using a copper solution (Bordeaux mixture) as a fungicide. In the home, exposure can occur from copper-containing supplements and medications, intrauterine conceptive devices, or from drinking water contami-
nated by copper pipes or plumbing components. Copper tubing has led to poisoning in hemodialysis patients. The majority of acute toxicity has occurred with either accidental or intentional (suicidal) ingestion of large amounts of copper.56
Mechanism of Injury The mechanism of copper hepatotoxicity is similar to that of iron toxicity and is caused primarily by free radical production leading to peroxidative injury of hepatocellular membrane lipids. Exposure of DNA to hydrogen peroxide in the presence of a copper– metallothionein complex results in the induction of a variety of types of oxidative damage, including DNA strand breaks and base modifications.
Acute Effects The early clinical manifestations of acute copper toxicity include dysgeusia, nausea, vomiting, and burning abdominal pain; jaundice accompanied by elevated aminotransferase levels, and hepatomegaly occur in approximately a quarter of patients. In severe cases, renal and hepatic failure and shock are the primary causes of death. Histological findings include centrilobular zonal necrosis, cholestasis, and dilated central veins with bile thrombi.57
Chronic Effects Several disease entities are attributable to chronic copper exposure, including Indian childhood cirrhosis (ICC), and idiopathic copper toxicosis (NICC). The role of copper exposure in both ICC and NICC, early childhood diseases that lead to cirrhosis, has been thoroughly investigated. The first reports of ICC were published in 1960, but the disease was not clearly defined until the 1980s, with histologic criteria that included necrosis of hepatocytes with ballooning, Mallory’s hyaline, pericellular intralobular fibrosis, and inflammatory inflammation in the context of a massive copper overload. Minimal fatty changes were observed, and cholestasis was noted only at a very advanced stage. Epidemiologic studies linked the disease to non-breastfed infants in a number of rural Indian villages and related the increase in copper intake to copper utensils used in milk preparation, as well as contamination of drinking of water by copper.58 Since then, similar cases of childhood cirrhosis related to copper exposure have been reported throughout the world, including an epidemiological study of 138 deaths due to infantile cirrhosis in rural Tyrol, Austria, between 1900 and 1974.59 Again, the copper intake was linked both to brass utensils and to high copper levels in the ground water. Current thinking on the pathogenesis of this disease involves a ‘’two-hit’ hypothesis: either excess copper ingestion in the setting of genetic predisposition (no specific mutations have so far been identified, but there is clear familial clustering beyond common copper exposure), or the synergistic effects of environmental toxins. Remarkably, with the copper content of drinking water now tightly regulated in Europe, no further cases of NICC have occurred since the 1970s.60 In vineyard workers exposed to the aerosolized Bordeaux mixture, the primary liver injury occurred in the form of hepatoportal sclerosis and non-caseating granulomas. Despite its carcinogenicity in laboratory animals, copper toxicity has not been linked to liver tumors.
569
Section IV. Toxin Mediated Liver Injury
Lead
Mechanism of Injury
The majority of adult lead exposure is occupational, occurring in workers involved in lead smelting and refinement, and in the manufacture of batteries, pigments, solder, car radiators, and ceramic ware with lead glaze. The lead content of paint was unregulated until 1977, leading to significant exposures both in construction workers and in the home, especially among children. In addition, illegally distilled alcohol (‘moonshine’) is an important source of lead intoxication: data show that habitual users of moonshine are significantly more sensitive to lead exposure, which may lead to considerable blood lead levels and resultant hepatotoxicity.61
Cadmium is a well established hepatotoxin in laboratory animals. Cadmium binds to sulfhydryl groups on critical molecules in mitochondria, leading to oxidative stress, changes in mitochondrial permeability, and mitochondrial dysfunction. Cadmium-dependent activation of Kupffer cells and free radical generation, resulting in the release of a large number of proinflammatory and cytotoxic mediators, leads to further hepatic injury, with subsequent hepatic degeneration, necrosis, and cirrhosis.65 Cadmium’s ability to oxidate nucleic acids and alter DNA repair mechanisms has rendered it a carcinogen, and it is associated with increased rates of lung and renal cell carcinoma in humans.
Mechanism of Injury The mechanism of injury in lead toxicity is complex, but is based primarily on its action as an electropositive metal. As such, it has a high affinity for negatively charged sulfhydryl groups, leading to the inhibition of sulfhydryl-dependent enzymes such as d-aminolevulinic acid dehydratase (ALAD) and ferrochelatase in heme synthesis. The divalent lead acts competitively with calcium in mitochondrial respiration, and interferes with various energy and transport systems. It can also affect nucleic acids by mechanisms that are not yet fully understood. Recently, much research has been concentrated on ALAD polymorphisms and the way they modify lead toxicokinetics, ultimately influencing individual susceptibility to lead poisoning.62
Clinical Manifestations Classic acute cadmium poisoning, known in Japan as itai-itai disease, is characterized by multiple fractures, osteomalacia, osteoporosis, and renal disease. No acute cadmium-related hepatotoxicity has been described. Chronic high-level cadmium exposure has been associated with aminotransferase elevations but non-specific findings on liver histology; its importance lies mainly in its ubiquitous presence in the environment, given its strong synergistic effects with other hepatotoxins and viral hepatitis. Environmental exposure in the absence of underlying liver disease has led to no demonstrable hepatotoxicity so far.65
Toxic Rapeseed Oil Clinical Manifestations Lead poisoning is manifested primarily by damage to the central and peripheral nervous systems, renal toxicity, and anemia; it is also a frequent cause of colicky abdominal pain in children. If diagnosed early, most of the effects are reversible, but chronic high-level exposure can lead to irreversible CNS and kidney damage. No acute lead hepatotoxicity has been reported in the literature, but high blood levels (usually above 75–80 mg/dl) were associated with liver injury in a series of patients from Spain.63 Liver biopsies showed mild centrilobular hepatitis with fatty infiltration and hemosiderin deposition. In that series, markers for lead (blood and urinary lead levels, ALAD inhibition) were more correlated with the degree of liver injury than were liver chemistry tests. Withdrawal of lead exposure or, in severe cases, chelation therapy, led to rapid resolution of the liver dysfunction.
Cadmium The major source of cadmium toxicity is through occupational exposure among individuals engaged in cadmium production and in the industrial use of cadmium in the production of alloys, pigments, and batteries. Over the last century there has been a dramatic increase in environmental contamination with cadmium owing to its presence in household waste and industrial emissions. Soil is commonly contaminated via atmospheric emission and sewage, as well as through the use of cadmium-containing phosphate fertilizers. Cadmium is readily absorbed by plants grown in the contaminated soil, particularly tobacco, grain, rice, and vegetables. Over 50% of the total body cadmium is stored in the liver and kidneys, tightly bound to metallothionein in an intracellular complex.64
570
In 1981 a massive intoxication of tens of thousands of people occurred in Spain following the consumption of adulterated rapeseed oil.66 The syndrome was referred to as the toxic-epidemic syndrome, and more than 350 people died. Initially there was acute multisystem involvement, with fever, arthraglia, rash, myalgia, conjunctivitis, digestive symptoms, and abdominal or chest pain, or both. Eosinophilia, raised levels of transaminases, alkaline phosphatase, g-glutamyl transpepidase, and IgE were commonly found. Three to 18 months after the attack episode a third of the patients who survived the initial toxicity developed scleroderma, sicca syndrome, alopecia, pulmonary hypertension, and respiratory failure. Hepatic involvement occurred in 25% of the victims, especially women in the fourth decade of life. Most of the patients with hepatic injury were asymptomatic.66 Elevation of g-glutamyl transpepidase was universal, and elevated alkaline phosphatase and transaminase levels were common. Histologic features included hepatocyte degeneration, cholestasis, a mixed cellular infiltration with a high cosmophil component, and acidophilic bodies – all similar to a drug-induced cholestatic hepatitis. Long-term histologic reassessment 30 months after the initial poisoning showed that the eventual prognosis could be forecast according to the initial biochemical abnormalities. Those patients with an initial transient elevation of transaminase levels recovered within 2 months. Those with initially raised transaminase and alkaline phosphatase levels mostly recovered or had evidence of a mild ongoing hepatitis. Of the patients who initially had jaundice, half continued to have an abnormal liver biochemistry profile at 30 months, and their histologic picture was suggestive of chronic cholestasis. Many of these patients also had evidence of scleroderma, and there was an increased incidence of HLA-DR3 and HLA-DR4 among them.66
Chapter 28 OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY
ENVIRONMENTAL HEPATOTOXINS MUSHROOM POISONING Of the 5000 known species of mushroom, fewer than 100 are poisonous to humans and fewer than 10 of these are deadly. The gathering and consumption of wild mushrooms, a traditional social practice in western Europe, has become increasingly popular in the United States. Although the number of fatal cases per year reported in this country does not approach the 50–100 deaths occurring annually in western Europe, fatal mushroom poisoning remains a serious public health concern.67 The incidence of mushroom poisoning has been increasing in the United States since the 1970s because of the increasing popularity of natural foods and the popularization of the gourmet qualities of wild mushrooms. Poisonings often occur in amateur mushroom hunters who fail to distinguish between edible and non-edible varieties. In addition, immigrants may come from areas where only edible lookalikes exist. Unsupervised children and persons looking for hallucinogenic substances may be poisoned as well. Many mushroom poisonings involve young children left unattended outdoors who are later found with mushrooms in their mouth. These exposures are rarely serious, because most lawn mushrooms are innocuous, medical evaluation is quickly sought, and fresh specimens of the mushroom are usually available for analysis. Adult exposures tend to be more serious because the mushrooms usually have been eaten in large quantities and were collected in the forest, where poisonous species are more likely to exist. Adults often present much later after ingestion and may eat more than one species of mushroom, which may lead to a confusing clinical presentation.67
Amanita Poisoning Most severe mushroom poisonings are caused by the Amanita species, which contain amatoxin, one of the most potent toxins known. In the United States, most poisonings occur along the cool coastal woodlands of the Pacific Northwest, where Amanita species are most common. Hospitalizations for mushroom poisoning in California have occurred more frequently in the warmer months and reach their peak in August. In general, however, the peak season for wild mushrooms typically coincides with the rainy season, which in California usually is from late October through March. This discordance may in large part be caused by the mistaken ingestion of poisonous Agaricus campestris species, which results in GI symptoms and dehydration. Amanita species also have been identified in the Blue Ridge Mountains of the northeast, as well as in Pennsylvania, New Jersey, suburban New York State, and the Gulf Coast.67 Of the three common Amanita species, A. phalloides, A. verna, and A. virosa, A. phalloides (death cap) has been responsible for more than 90% of fatalities. The mature A. phalloides mushroom can be easily mistaken for similar-appearing non-toxic mushrooms, even by experts. Amanita species tend to grow in oak woodlands or in close proximity to oak trees, and are commonly mistaken for edible species; they have no characteristic smell or taste, and cooking does not destroy their toxin. Color varies with weather, soil, and the age of the mushroom. Amanita can be identified by the presence of white gills (Figure 28-3) underneath the cap, an annulus at
Cap
Annulus Stem
Volva
A
1 cm
B
Figure 28-3. (A) Amanita phalloides. (B) Lepiota helveola. (Adapted from reference 67.)
the top of the stalk, and a pouch at the base of the mushroom. Mushrooms that are cut off at ground level are often misidentified, as this characteristic pouch goes undetected. As Amanita is a mycorrhizal fungus living symbiotically with the roots of the host tree, it cannot be destroyed unless the tree itself is killed.67 A. phalloides exerts its hepatotoxicity through two distinct toxins, phalloidin and amatoxin. Phalloidin is a cyclic heptapeptide that causes irreversible polymerization of G-actin to F-actin, resulting in the ultimate disruption of hepatocyte cell membranes and cell death. Amatoxins are thermostable octapeptides that bind with a subunit of RNA polymerase II and interfere with mRNA synthesis. They are not destroyed by cooking or gastric acidity. The inability to produce vital structural proteins results in cell necrosis, with tissues having high rates of protein synthesis being primary targets for the toxin. The liver and kidney are most commonly involved, although the brain and pancreas also are affected. Amatoxin concentration in mushrooms varies by species, season, region, and local conditions. Although the amount of amatoxin per mushroom is difficult to quantify, a bite-sized piece of one mushroom can contain a lethal dose of toxin. Varying amounts of amatoxin are present in other potentially lethal mushrooms, such as some Galerina and Lepiota species. Fifteen to 20 of these mushrooms can constitute a potentially fatal dose.67 Amatoxins are readily absorbed from the intestinal epithelium and weakly bound by plasma proteins. They penetrate cells rapidly and, in the case of hepatocytes, are transported across the cell membrane via bile transport carriers. In the liver, massive hepatocellular damage results in centrilobular necrosis (Figure 28-4). Sixty per cent of absorbed amatoxin is excreted via the bile and returns to the liver via the enterohepatic circulation, resulting in continued toxin exposure. GI epithelium and the proximal and distal convoluted tubules of the kidney also are severely affected.67
Clinical Presentation Poisoning with non-lethal poisonous mushrooms produces crampy abdominal pain, nausea, emesis, and watery diarrhea soon after ingestion. With Amanita poisoning patients exhibit signs and symptoms that typically occur in stages. The lethal dose is about 50 g,
571
Section IV. Toxin Mediated Liver Injury
tocyte membranes, or via interruption of the enterohepatic recirculation of amatoxin.68 Mortality rates have decreased with improved supportive care, but death still occurs in 20–30% of cases. Once stage 3 hepatic encephalopathy is reached, the patient is unlikely to survive without liver transplantation.
Lepiota Helveola
Figure 28-4. Massive hepatic necrosis secondary to mushroom poisoning. Section taken from an explanted liver, showing confluent lobular necrosis with collapse and extinction of most of the hepatic parenchyma (central area of photomicrograph) and broad areas of bridging necrosis. A few lobules of viable hepatocytes remain (left upper and right lower corners). Hematoxylin and eosin, original magnification 20¥. (Courtesy of Dr M.I. Fiel.)
which corresponds to three medium-sized mushrooms. There is an initial latent asymptomatic period of 6–24 hours; 12–24 hours of severe GI symptoms then ensue, with the patient often being misdiagnosed with viral gastroenteritis. A second latent phase follows, with improvement of GI symptoms but the development of abnormal liver chemistry tests. A hepatic phase occurs 48–96 hours after ingestion, with precipitous elevation of serum aminotransferases into the thousands, coagulopathy, and jaundice. Renal failure secondary to fulminant hepatic failure or to the direct nephrotoxicity of amatoxin also develops. Half of patients with Amanita poisoning have clinical or biochemical evidence of pancreatitis.68
Treatment Early consultation with a poison control center and liver transplant center is essential. The amount of mushroom ingested appears to be the main prognostic factor. Supportive care remains the mainstay of treatment. Induction of emesis can reduce the toxin load, but most patients present 6–8 hours post ingestion, making this of limited use. The administration of activated charcoal with a cathartic and gastric lavage is performed to remove any remaining toxin from the GI tract. As diarrhea can be severe, adequate intravenous fluid and electrolyte maintenance are essential. Although amatoxin and phalloidin are dialyzable, charcoal hemoperfusion and hemodialysis have not been effective in limiting hepatic injury. Although the precise mechanism of action is unknown, high-dose penicillin G (300 000–1 000 000 IU/kg/day) has been used in some patients, with resultant improvement of hepatic dysfunction. The hepatoprotective effects of penicillin G may be secondary to increased renal excretion of amatoxin, or via the inhibition of penetration of the toxin into hepatocytes. Silymarin, an extract of the milk thistle Silybum marianum, has been used widely in Europe in doses of 20–50 mg/kg/day, usually in combination with penicillin G. Its hepatoprotective effects may be related to inhibition of toxin binding to liver cells, prevention of the toxin’s penetration of hepa-
572
The genus Lepiota can be easily mistaken for an edible variety of mushroom. It has a flat pink or brown ochre cap, is covered with fluffy scales, and bears white or cream-colored gills (Figure 28-3). Lepiota helveola contains high concentrations of amatoxin but no phalloidin. It has the same – if not a greater – hepatotoxic potential as Amanita species, and multiple poisonings resulting in FHF and the need for liver transplantation have been reported.67 There are many other types of mushroom that contain toxins with potential GI and neurological toxicities. Symptoms often begin within minutes to several hours of ingestion. Numerous fatalities have resulted from toxin ingestion, though none have been related to hepatic failure. Most toxic ingestions are self-limited, without related hepatotoxicity. As several types of mushroom may grow near each other or be consumed together, clinicians should always be aware of the possibility of the ingestion of both edible and potentially hepatotoxic and lethal varieties, especially if initial symptoms appear to begin after a 6-hour latent period. The presence of symptoms immediately after ingestion should not rule out the possibility of a more serious poisoning.
PYRROLIZIDINE ALKALOIDS Known as hepatotoxins since the 18th century, pyrrolizidine alkaloids (PAs) cause acute and chronic hepatic injury in experimental animals and in humans. Over 300 PAs have been identified in over 6000 plants, including approximately 3% of the world’s flowering plants. Some more common plants containing PA are shown in Table 28-6. The chief genera that produce the toxic alkaloids affecting livestock and humans are the Senecio, Heliotropium, Crotalaria, and Symphytum (comfrey) species. Large outbreaks of PA poisoning have occurred through contamination of wheat crops; significant amounts of PAs in concentrations capable of causing injury in experimental animals have also been found in milk and honey. A number of PAs are used as supplements or in traditional herbal medicines, with comfrey and coltsfoot as popular examples. Systemic bioavailability of PAs after dermal exposure is generally low, but toxicity can occur after use of PA-containing creams and shampoos as well. Liver injury seems to depend on the type of PA and the total dose ingested, along with the susceptibility of the person to the alkaloid.3,67
Mechanism of Injury Once the alkaloids are ingested they pass into the hepatocytes via the sinusoidal blood and are metabolized through the CYP3A to dehydro-PAs, N-oxides and pyrroles. Hydrolysis and N-oxide formation are detoxification reactions and usually do not cause harm to the cell; however, dehydro-PAs are the primary toxic metabolites that react with available nucleophiles within the hepatocyte. Pyrroles are alkylating agents that are highly reactive with proteins and nucleic acids, binding sulfur, nitrogen, and oxygen-containing groups on various macromolecules. They penetrate the nucleus and
Chapter 28 OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY
Table 28-6. Plants Containing Pyrrolizidine Alkaloids Plant
Common names
Alkaloids contained
Geographic distribution
Armsinckia intermedia
Fiddleneck tarweed, firewood, yellow forget-me-nots Rattlebox, rattle pod Wild Lucerne Whiteback Streaked rattle pod Wedge-leaved rattle pod, earring plant Viper’s bugloss, Paterson’s curse, salvation Jane Common heliotrope, caterpillar weed, potato weed Heliotrope
Echiumine, lycopsamine, intermedin
USA
Dicrotaline Dicrotaline Intergerrimine Monocrotaline, retronecine, retusine Echiumine, echimidine
Many countries South Africa South Africa Jamaica Australia, South Africa USA, Australia
Crotolaria Crotolaria dura Crotolaria fulva Crotolaria mucronata Crotolaria reftsa Echium lycopsis Heliotropium europalum Heliotropium lasiocarpum Senecio jacoboae Senecio latifoolius Senecio reetrorus Senecio ridelli Senecio spartioides Senecio vulgaris Symphytum officinale
Common ragweed, stinking Willie Dan’s cabbage, groundsel, ragwort, Rhodesia ragwort Dan’s cabbage, Wooly groundsel Wooly groundsel, Riddels groundsel Brown groundsel Common groundsel
Heliotrine, lasiocarpine, europine, supinine Echimidine, echiumidine, lasiocarpine, heliotune Jacobine, jacodine, jacoline, jacovine Semicriphylline Retrorsine Retrorsine Seneccionine, senociphylline, spartioidine Retrorsine, senecione
Australia, USA Russia, Central Asia US, Jamaica, New Zealand South Africa South Africa USA USA UK, USA Japan
From Zimmerman HJ. Occupational hepatotoxicity. In: Zimmerman HJ. Hepatotoxicity: the adverse effects of drugs and other chemicals on the liver. Philadelphia: Lippincott Williams & Wilkins, 1999
react with DNA, causing cross-linking within DNA and between DNA and nucleoproteins. One of the mutations caused is in codon 249 of the p53 gene, the same target as for aflatoxin; however, this may only be specific for the PAs found in comfrey.68 The predominant feature of PA toxicity, veno-occlusive disease (VOD), is caused by pyrroles passing from the hepatocyte to the perisinusoidal space of Disse into the sinusoidal lumen, where they are able to react with the sinusoidal lining (endothelial cells) and the walls of small hepatic veins (Figure 28-5). They can also become bound on passing red blood cells and thus get carried to the lungs or heart, where they can induce pulmonary hypertension and right ventricular hypertrophy.69 Differences in susceptibility are at least partly explained by isoforms of the CYP3A subfamily. CYP3A catalyzes pyrrole formation as well as N-oxide formation, the latter being primarily responsible for the detoxification process. The differences in enzyme activity can vary by as much as 30-fold, leading to a large difference in the dose required to reach toxicity in different individuals. Older age has also been shown to lead to an increase in susceptibility to hepatotoxicity.3,68
Acute Effects Acute PA toxicity typically occurs through human consumption of contaminated grains with seed containing PA, or through herbal remedies. The syndrome of veno-occlusive disease, characterized by abdominal distention, hepatosplenomegaly, ascites, and peripheral edema, was first described in Jamaican children, and was related to their intake of bush teas made with Senecio and Crotalaria species as therapy for acute illnesses. In other parts of the world, serious outbreaks of PA poisoning have occurred in Afghanistan and central India, both after periods of drought when Heliotropium plants thrived in the region, leading to wheat crop contamination. The
Figure 28-5. Veno-occlusive disease. Liver needle biopsy showing fibroobliterative changes in a terminal hepatic venule consisting of fine fibrous tissue deposition along the wall (arrow) and within adjacent perisinusoidal spaces. The lumen is slightly diminished in caliber. Hematoxylin and eosin, original magnification 20¥. (Courtesy of Dr M.I. Fiel.)
most recent epidemic case occurred in 1992 in Tajikistan, where close to 4000 people developed PA hepatotoxicity. PA toxicity was also endemic in parts of South America in the second half of the 20th century, but with better education and proper identification of plants only sporadic cases are now reported.3,69 In acute PA hepatotoxicity, abdominal pain, jaundice and ascites typically develop within 3–6 weeks of ingestion, with centrilobular hepatic necrosis and sinusoidal dilatation seen on liver biopsy.
573
Section IV. Toxin Mediated Liver Injury
Approximately half of the patients recover completely in a few weeks, but the course is rapidly fatal in approximately 20%. In the remainder of cases, 20% appear to recover clinically but then develop VOD and cirrhosis after several years. Others develop a subacute form of VOD which may eventually resolve or subsequently progress to chronic VOD and cirrhosis.3,69
P450 effect.70 Thus, identifying the concomitant use of ethanol or other medications may be important in predicting the chance of recovery from severe cocaine hepatotoxicity.71 Cocaine intoxication should be considered along with ischemic hepatitis, acetaminophen overdose, and Amanita mushroom poisoning in the differential diagnosis of any patient with acute hepatitis and extremely elevated serum aminotransferases that rise into the several thousand range.
Chronic Effects Chronic hepatotoxicity caused by PA exposure is associated with hepatocellular injury leading to cirrhosis. Owing to the production of persistent electrophilic metabolites that can be released from hepatocytes on an ongoing basis, there is ongoing hepatic injury in addition to VOD. The clinical picture is that of poor hepatic flow akin to the Budd–Chiari syndrome, with hepatosplenomegaly, ascites, and edema. Histologically, there is central fibrosis and bridging between central veins that leads to cirrhosis, similar to that seen in cardiac cirrhosis. Death is usually related either to complications of portal hypertension or to pulmonary hypertension leading to congestive heart failure.3,69
Cocaine Cocaine (benzoylmethylecgonine) is an alkaloid extracted from the leaf of Erythroxylon coa. For many years cocaine was presumed to be directly hepatotoxic to humans because of the frequent findings of jaundice and abnormal liver chemistry tests among cocaine abusers; however, in these persons there often is associated polysubstance abuse, alcohol intoxication, or viral hepatitis. In mice, cocaine is a potent hepatotoxin, causing steatosis and hepatocellular necrosis with high serum transaminase levels. Ten per cent of cocaine is converted in the liver by flavin adenine dinucleotide and cytochrome P450 to the active metabolite, the hepatotoxic free radical norcocaine nitroxide. There is further oxidation of norcocaine nitroxide to the nitrosonium ion, which is highly reactive with glutathione and results in its depletion.70 This ultimately leads to further free radical formation, covalent binding to hepatic proteins, and lipid peroxidation of hepatocyte cell membranes. Although hyperthermia and shock have been associated with cocaine intoxication, and the profound serum aminotransferase elevation found in some cocaine intoxications is similarly seen in patients with ischemic hepatitis, ischemia is not thought to play a major role in cocaine hepatotoxicity.67 Acute hepatotoxic injury from cocaine use is now well described in humans. Severe necroinflammatory hepatitis and liver failure, clinically similar to that seen with acetaminophen ingestion, can occur, with associated renal failure, disseminated intravascular coagulation, and rhabdomyolysis. Histologic changes include pericentral coagulative necrosis and micro- and macrovascular steatosis, with only mild inflammatory infiltration. The most severe histologic changes occur in zone 3 of the liver parenchyma, indistinguishable from changes seen in acetaminophen toxicity. Cocaine and acetaminophen are both metabolized by cytochrome P450. It appears that, as with acetaminophen toxicity, there can be enhancement of necrosis in cocaine hepatotoxicity via P450 induction or glutathione depletion. In mice, cimetidine effectively prevents cocaine-induced hepatic necrosis, whereas ethanol and barbiturates potentiate it, all presumably related to the cytochrome
574
AFLATOXINS Aflatoxins were first isolated in the 1960s after an epidemic of acute hepatic necrosis and death in turkeys that were fed a peanut meal contaminated with Aspergillus flavus (turkey X disease).72 Since that time, aflatoxins have been subject to a great number of studies, confirming their hepatotoxic and hepatocarcinogenic potential in poultry and domestic animals, as well as in humans.73,74 Aflatoxininduced acute hepatotoxicity in humans has followed the ingestion of contaminated maize, soybeans, and cassava.
Epidemiology Aflatoxins are produced as secondary metabolites of Aspergillus flavus and Aspergillus parasiticus fungi in warm, humid temperatures. They are ubiquitous and contaminate a variety of food staples in tropical and subtropical areas, including rice, oats, wheat, corn, ground nuts, and spices. There is a wide (more than 100-fold) intraspecies variation in animal susceptibility to disease, and the predominant patterns of injury vary from species to species as well.74 Within species, older age and male gender appear to be protective factors. Choline-deficient diets and diets low in protein appear protective against acute toxicity, yet enhance the carcinogenic effects of aflatoxins.75,76
Acute Effects Upon ingestion, aflatoxins are rapidly metabolized to the reactive species by the microsomal enzymes of the liver. In binding to guanine residues on DNA these metabolites inhibit the synthesis of nuclear RNA. Acute hepatic injury with steatosis and/or hepatocellular necrosis is thought to result from injury to membranes (plasma, reticular, and mitochondrial) via either injury to ribonuclease or by direct toxic reactions of the metabolites with the membrane proteins. The clinical symptoms of acute aflatoxicosis are ascribed to hemorrhagic necrosis of the liver and bile duct proliferation, leading to hepatic failure.76,77
Chronic Effects The synergistic effects of aflatoxin and hepatitis B and C viruses in the development of hepatocellular carcinoma have been well described. Recent epidemiologic studies from Singapore and Shanghai,78,79 areas that have both seen tremendous economic improvement over the last two decades, have shown that the incidence of HCC has rapidly fallen over the same period. Clearly this is at least partly due to a decrease in aflatoxin contamination of the food supply; aflatoxins have significant late effects in the development of hepatocellular carcinoma. Geographic areas where aflatoxin contamination is prevalent show a strong correlation between the degree of contamination of common foodstuffs and the incidence of hepatocellular carcinoma.
Chapter 28 OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY
High concentrations of aflatoxin B1 have been found in the livers of many patients developing hepatocellular carcinoma in these areas. It appears that aflatoxin contamination of food and coexisting chronic hepatitis B infection are associated with even higher rates of hepatocellular carcinoma in a given population (e.g. in Mozambique, which has both the highest average aflatoxin exposure and the highest incidence of hepatocellular carcinoma in the world). Thus, the effects of these two carcinogenic agents appear synergistic. Chronic feeding of aflatoxin to experimental animals leads reproducibly to the development of hepatocellular carcinoma. Evidence of a causative role for aflatoxin in the pathogenesis of human hepatocellular carcinoma has been further suggested by the description of a high frequency of a unique mutation in codon 249 of the p53 gene.80,81 The p53 gene influences the transcription of important cellular genes involved in regulation of the cell cycle. Mutations and allelic deletions of the p53 gene are the most common genetic alterations found in human tumors, with p53 abnormalities described in several hepatocellular carcinoma cell lines. Mutations at codon 249 of the p53 gene have been found in patients with hepatocellular carcinoma in China and southern Africa, areas of high aflatoxin contamination. The absence of similar mutations has been shown in geographic regions where hepatocellular carcinoma occurs frequently but where aflatoxin contamination is low. These mutations are consistent with those caused by aflatoxin B1 in experimental models of mutagenesis. The clustering of the codon 249 mutation in hepatocellular carcinomas occurring in areas with high aflatoxin contamination may thus be a clue to the carcinogenicity of an environmental toxin.80,81
CYANOBACTERIAL TOXINS Cyanobacteria (also known as blue-green algae) are an ancient group of photosynthetic organisms that grow in water, with habitats that range from hot springs to temporarily frozen ponds in Antarctica and include both fresh and seawater.82 Of the currently known 150 genera with about 2000 species, at least 40 are known to be toxicogenic, affecting primarily either the central nervous system or the liver. The hepatotoxins cylindrospermopsin, microcystins, and nodularin are all small cyclic peptides that are synthesized nonribosomally. The species that are capable of synthesizing these toxins must thus possess the peptide synthesis gene sequence that will be expressed under certain environmental conditions. The presence of cyanobacterial toxins in drinking water supplies poses a serious problem to water treatment facilities, as specific technical procedures that are not widely available are required to remove these toxins effectively.
Acute Effects The majority of reported outbreaks of acute cyanobacterial toxicity are cases of gastroenteritis with mild elevations of g-glutamyltransferase (GGTP). However, in 1996, improperly purified water with high levels of microcystins from the Tabocas reservoir was used at a dialysis center in Caruaru, Brazil, resulting in 101 of 124 exposed subjects developing acute liver injury, with 50 subsequent deaths.83 Clinically, patients developed primarily cholestatic jaundice with high bilirubin and alkaline phosphatase concentrations, as well as increases in hepatic enzymes (aspartate and alanine aminotrans-
ferase). On histopathology, panlobular hepatocyte necrosis with apoptosis was the predominant feature.
Chronic Effects Microcystin and nodularin have been shown to induce the expression of tumor necrosis factor-a and early response genes (c-jun, jun B, jun D, c-fos, fos B, fra-1) in rat liver and hepatocytes.84 In mice, mutations in the K-ras codon 12 as well as DNA fragmentations have been reported after an injection of cyanobacterial extract.85 In China the consumption of microcystin-contaminated drinking water from pond and ditch waters has been found to be associated with a 25-fold increase in incidence of HCC.85,86
REFERENCES 1. Tolman KG, Sirrine RW. Occupational hepatotoxicity. Clin Liver Dis 1998;2:563–589. 2. NIOSH. Pocket Guide to Chemical Hazards. U.S. Government Printing Office; 2004. US Department of Health and Human Services Publication 97-140. Online at http://www.cdc.gov/niosh/npg/npg.html. 3. Zimmerman HJ. Experimental hepatotoxicity. In: Zimmerman HJ. Hepatotoxicity: the adverse effects of drugs and other chemicals on the liver. Philadelphia: Lippincott Williams & Wilkins, 1999: 217. 4. Ross G. The public health implications of polychlorinated biphenyls (PCBs) in the environment. Ecotoxicol Environ Safety 2004;59:275–291. 5. Reggiani G, Bruppacher R. Symptoms, signs and findings in humans exposed to PCBS and their derivatives. Environ Health Perspect 1985;60:225–232. 6. Bosetti C, Negri E, Fattore E, et al. Occupational exposure to polychlorinated biphenyls and cancer risk. Eur J Cancer Prev 2003;12:251–255. 7. Kimbrough RD, Doemland ML, Mandel JS. A mortality update of male and female capacitator workers exposed to polychlorinated biphenyls. J Occup Environ Med 2003;45:271–282. 8. US Environmental Protection Agency, 2003a. Available at: http://www.epa.gov/opptintr/pcb/effects.html. 9. Steenland K, Bertazzi P, Baccarelli A, et al. Dioxin revisited: developments since the 1997 IARC classification of dioxin as a human carcinogen. Environ Health Perspect 2004;112:1265–1268. 10. Frumkin H. Agent Orange and cancer: an overview for clinicians. CA Cancer J Clin 2003:53:245–255. 11. US Department of Health and Human Services, Agency of Toxic Substances and Disease Registry. Toxicological profile for 2,4,6trinitrotoluene. http://www.atsdr.cdc.gov/toxprofiles/tp81.pdf. 12. World Health Organization. United Nations Environment Programme, International Programme on Chemical Safety: Vinyl Chloride, 1999. Accessed at: www.inchem.org/documents/ehc/ehc215htm#SectionNumber:7.10. 13. Weihrauch M, Bader M, Lehnert G, et al. Mutation analysis of Kras-2 in liver angiosarcoma and adjacent nonneoplastic liver tissue from patients occupationally exposed to vinyl chloride. Environ Mol Mutagen 2002;40:36–40. 14. Mastrangelo G, Fedeli U, Fadda E, et al. Increased risk of hepatocellular carcinoma and liver cirrhosis in vinyl chloride workers: synergistic effect of occupational exposure with alcohol intake. Environ Health Perspect 2004;112:1188–1192. 15. Staib F, Hussain SP, Hofseth LJ, et al. TP53 and liver carcinogenesis. Hum Mutat 2003;21:201–216. 16. Pimentel JC, Menezes AP. Liver disease in vineyard sprayers. Gastroenterology 1977;72:275–278.
575
Section IV. Toxin Mediated Liver Injury
17. Maroni M, Mocci F, Visentin S, et al. Periportal fibrosis and other liver ultrasonography findings in vinyl chloride workers. Occup Environ Med 2003;60:60–65. 18. Blendis LM, Smith PM, Lawrie BW et al. Portal hypertension in vinyl chloride monomer workers. Gastroenterology 1978;75:206–211. 19. Hsiao TJ, Wang JD, Yang PM, et al. Liver fibrosis in asymptomatic polyvinyl chloride workers. J Occup Environ Med 2004;46:962–966. 20. Bosetti C, La Vecchia C, Lipworth L, et al. Occupational exposure to vinyl chloride and cancer risk: a review of the epidemiologic literature. Eur J Cancer Prev 2003;12: 427–430. 21. Pirastu R, Baccini M, Biggeri A, et al. Epidemiologic study of workers exposed to vinyl chloride in Porto Maghera: mortality update. Epidemiol Prev 2003;27:161–172. 22. Mundt KA, Dell LD, Austin RP, et al. Historical cohort study of 10,109 men in the North American vinyl chloride industry, 1942–72: update of cancer mortality to 31 December 1995. Occup Environ Med 2000;57:774–781. 23. Ward E, Boffetta P, Andersen A, et al. Update of the follow-up of mortality and cancer incidence among European workers employed in the vinyl chloride industry. Epidemiology 2001;6:710–718. 24. Wong RH, Chen PC, Du CL, et al. An increased standardized mortality ratio for liver cancer among polyvinyl chloride workers in Taiwan. Occup Environ Med 2002;59:405–409. 25. Shimizu H, Kumada T, Nakano S, et al. Liver dysfunction among workers handling 5-nitro-o-toluidine. Gut 2002;50:266–270. 26. Basu S. Carbon tetrachloride-induced lipid peroxidation: eicosanoid formation and their regulation by antioxidant nutrients. Toxicology 2003;189:113–127. 27. Weber LW, Boll M, Stampfl A. Hepatotoxicity and mechanism of action of haloalkanes: carbon tetrachloride as a toxicological model. Crit Rev Toxicol 2003;33:105–136. 28. Brautbar N, Williams J. Industrial solvents and liver toxicity: risk assessment, risk factors and mechanisms. Int J Hyg Environ Health 2002;205:479–491. 29. Williams AT, Burk RF. Carbon tetrachloride hepatotoxicity. An example of free radical-mediated injury. Semin Liver Dis 1990;10:279–284. 30. Hodgson MJ, Heyl AE, Vanthiel DH. Liver disease associated with exposure to 1,1,1-trichloroethanol. Arch Intern Med 1989;149:1793–1798. 31. Thiele DL, Eigenbrodt EH, Ware AJ. Cirrhois after repeated trichloroethylene and 1,1,1-trichloroethane exposure. Gastroenterology 1982;83:996–929. 32. Marino G, Anastopoulos H, Woolf AD. Toxicity associated with severe inhalational and dermal exposure to dimethylacetamide and 1,2-ethanediamine. J Occup Med 1994;36:637–641. 33. Perbellini L, Princivalle A, Caivano M, et al. Biological monitoring of occupational exposure to N,N-dimethylacetamide with identification of a new metabolite. Occup Environ Med 2003;60:746–751. 34. Luo JC, Kuo HW, Cheng TJ, et al. Abnormal liver function associated with occupational exposure to dimethylformamide and hepatitis B virus. J Occup Environ Med 2001;43:474–482. 35. Wrbitzky R. Liver function in workers exposed to N,Ndimethylformamide during the production of synthetic textiles. Int Arch Occup Environ Health 1999;72:19–25. 36. Kopelman H, Robertson MH, Sanders PG, Ash I. The Epping jaundice. Br Med J 1966;5486:514–516. 37. Nichols L. The Epping jaundice outbreak: mortality after 38 years of follow-up. Int Arch Occup Environ Health 2004;77:592–594. 38. Kanz MF, Dugas TR, Liu H, et al. Glutathione depletion exacerbates methylenedianiline toxicity to biliary epithelial cells and hepatocytes in rats. Toxicol Sci 2003;74:447–456.
576
39. Shintani H, Nakamura A. Analysis of a carcinogen, 4,4 methylenedianiline, from thermosetting polyurethane during sterilization. J Anal Toxicol 1989;13:354–357. 40. Kanz MF, Gunasena GH, Kaphalia L, Hammond DK, Syed YA. A minimally toxic dose of methylene dianiline injures biliary epithelial cells in rats. Toxicol Appl Pharmacol 1998;150:414–426. 41. Cocco P, Kazerouni N, Zahm SH. Cancer mortality and environmental exposure to DDE in the United States. Environ Health Perspect 2000;1:1–4. 42. Battaller R, Bragulat E, Nogue S, et al. Prolonged cholestasis after acute paraquat poisoning through skin absorption. Am J Gastroenterol 2000;95:1340–1343. 43. Guzelian P. Comparative toxicology of chlordecone (kepone) in humans and experimental animals. Annu Rev Pharmacol 1982;22:89–113. 44. Tchounwou PB, Centeno JA, Patlolla AK. Arsenic toxicity, mutagenesis, and carcinogenesis – a health risk assessment and management approach. Mol Cell Biochem 2004;255:47–55. 45. Kitchin K. Recent advances in arsenic carcinogenesis: modes of action, animal model systems, and methylated arsenic metabolites. Toxicol Appl Pharmacol 2001;172:249–261. 46. Abernathy CO, Thomas DJ, Calderon RL. Health effects and risk assessment of arsenic. J Nutr 2003;133:1536S–1538S. 47. Tsuda T, Babazono A, Yamamoto E, et al. Ingested arsenic and internal cancer: a historical cohort study followed for 33 years. Am J Epidemiol 1995;141:198–209. 48. Yoshida T, Yamauchi H, Fan Sun G. Chronic health effects in people exposed to arsenic via the drinking water: dose-response relationships in review. Toxicol Appl Pharmacol 2004;198:243–252. 49. Chiu HF, Ho SC, Wang LY, et al. Does arsenic exposure increase the risk for liver cancer? J Toxicol Environ Health A 2004;67:1491–500. 50. Centeno JA, Mullick FG, Martinez L. Pathology related to chronic arsenic exposure. Environ Health Perspect 2002;110:883–886. 51. Ho SY, Tsai CC, Tsai YC, et al. Hepatic angiosarcoma presenting as hepatic rupture in a patient with long-term ingestion of arsenic. J Formos Med Assoc 2004;103:374–379. 52. Guo HR. The lack of a specific association between arsenic in drinking water and hepatocellular carcinoma. J Hepatol 2003;39:382–288. 53. Tenenbeim M. Hepatotoxicity in acute iron poisoning. J Toxicol Clin Toxicol 2001;39:721–726. 54. Henretig FM, Temple AR. Acute iron poisoning in children. Clin Lab Med 1984;3:575. 55. Kozaki K, Egawa H, Garcia-Kennedy R, et al. Hepatic failure due to massive iron ingestion successfully treated with liver transplantation. Clin Transplant 1995;9:85. 56. Pankit AN, Bhave SA. Copper metabolic defects and liver disease: Environmental aspects. J Gastroenterol Hepatol 2002;17:5403–5407. 57. Muller T, Langner C, Fuchsbichler A, et al. Immunohistochemical analysis of Mallory bodies in Wilsonian and non-Wilsonian hepatic copper toxicosis. Hepatology 2004;39:963–969. 58. Tanner MS. Role of copper in Indian childhood cirrhosis. Am J Clin Nutr 1998;67:1074S–1081S. 59. Muller T, Feichtinger H, Berger H, et al. Endemic Tyrolean infantile cirrhosis: an ecogenetic disorder. Lancet 1996;347:877–880. 60. Zietz BP, Dieter HH, Lakomek M, et al. Epidemiological investigation on chronic copper toxicity to children exposed via the public drinking water supply. Sci Total Environ 2003;302:127–144. 61. Lopez CM, Vallejo NE, Pineiro AE, et al. Alteration of biochemical parameters related with exposure to lead in heavy alcohol drinkers. Pharmacol Res 2002;45:47–50.
Chapter 28 OCCUPATIONAL AND ENVIRONMENTAL HEPATOTOXICITY
62. Chia SE, Yap E, Chia KS. Delta-aminolevulinic acid dehydratase (ALAD) polymorphism and susceptibility of workers exposed to inorganic lead and its effects on neurobehavioral functions. Neurotoxicology 2004;25:1041–1047. 63. Sanchez JA, de la Fuente JM, Castrillo JM, et al. Hepatotoxidad por plomo inorganico: resultados de 85 casos de saturnismo agudo. Gastroenterol y Hepatol 1985;8:246–250. 64. Ikeda M, Zhang ZW, Moon CS. Normal liver function in women in the general Japanese population subjected to environmental exposure to cadmium at various levels. Int Arch Occup Environ Health 2000;73:86–90. 65. Rikans LE, Yamano T. Mechanisms of cadmium-mediated acute hepatotoxicity. J Biochem Mol Toxicol 2000;14:110–117. 66. Velicia R, Sanz C, Martinez-Barredo F, et al. Hepatic disease in the Spanish toxic oil syndrome. J Hepatol 1986:3:59–65. 67. Schiano TD. Liver injury from herbs and other botanicals. Clin Liver Dis 1998;2:607–630. 68. Fu PP, Xia Q, Lin G, et al. Pyrrolizidine alkaloids: genotoxicity, metabolism enzymes, metabolic activation, and mechanisms. Drug Metab Rev 2004;36:1–55. 69. Chojkier M. Hepatic sinusoidal-obstruction syndrome: toxicity of pyrrolizidine alkaloids. J Hepatol 2003;39:437–446. 70. Selim K, Kaplowitz N. Hepatotoxicity of psychotropic drugs. Hepatology 1999;29:1347–1351. 71. Ponsoda X, Bort R, Jover R, Gomez-Lechon MJ, Castell JV. Increased toxicity of cocaine on human hepatocytes induced by ethanol: role of GSH. Biochem Pharmacol 1999;58: 1579–1585. 72. Swarbock O. Disease of turkey poults. Vet Rec 1960, 72:652. 73. Ross RK, Yuan JM, Yu MC, et al. Urinary aflatoxin biomarkers and risk of hepatocellular carcinoma. Lancet 1992;339:943– 946. 74. McGlynn KA, Hunter K, LeVoyer T, et al. Susceptibility to aflatoxin B1-related primary hepatocellular carcinoma in mice and humans. Cancer Res 2003;63:4594–4601. 75. Cullen JM, Newberne PM. Acute hepatotoxicity of aflatoxins. In: Eaton DL, Groopman JD, eds. The toxicology of aflatoxins: human health, veterinary, and agricultural significance. London: Academic Press, 1993: 1–26.
76. Williams JH, Phillips TD, Jolly PE, et al. Human aflatoxicosis in developing countries: a review of toxicology, exposure, potential health consequences, and interventions. Am J Clin Nutr 2004;80:1106–1122. 77. Creppy EE. Update on survey, regulation and toxic effects of mycotoxins in Europe. Toxicol Lett 2002;127:19–28. 78. Ming L, Thorgerlsson SS, Gall MH, et al. Dominant role of hepatitis B virus and cofactor role of aflatoxin in hepatocarcinogenesis in Qidong, China. Hepatology 2002;36:1214–1220. 79. Wang LY, Hatch M, Chen CJ, et al. Aflatoxin exposure and risk of hepatocellular carcinoma in Taiwan. Int J Cancer 1996;67:620–625. 80. Yu MC, Yuan JM. Environmental factors and risk for hepatocellular carcinoma. Gastroenterology 2004;127:S72–S78. 81. Kensler TW, Egner PA, Wang JB, et al. Chemoprevention of hepatocellular carcinoma in aflatoxin endemic areas. Gastroenterology 2004;127:S310–S318. 82. Hitzfeld BC, Hoger SJ, Dietrich DR. Cyanobacterial toxins: removal during drinking water treatment, and human risk assessment. Environ Health Perspect 2000;108:113–122. 83. Jochimsen EM, Carmichael WW, An JS, et al. Liver failure and death after exposure to microcystins at a hemodialysis center in Brazil. N Engl J Med 1998;338:873–878. 84. Sueoka E, Sueoka N, Okabe S, et al. Expression of the tumor necrosis factor alpha gene and early response genes by nodularin, a liver tumor promoter, in primary cultured rat hepatocytes. J Cancer Res Clin Oncol 1997;123:413–419. 85. Rao P, Bhattacharya R, Parida MM, et al. Freshwater cyanobacterium Microcystis aeruginosa (UTEX 2385) induced DNA damage in vivo and in vitro. Environ Toxicol Pharmacol 1998;5:1–6. 86. Ueno Y, Nagata S, Tsutsumi T, et al. Detection of microcystins, a blue-green algal hepatotoxin, in drinking water sampled in Haimen and Fusui, endemic areas of primary liver cancer in China, by highly sensitive immunoassay. Carcinogenesis 1996;17:1317–1321.
577
Section IV. Toxin Mediated Liver Injury
29
ALCOHOLIC LIVER DISEASE Stephen F. Stewart and Christopher P. Day Abbreviations ADH alcohol dehydrogenase AMPK adenosine monophosphate-activated protein kinase ALD alcoholic liver disease ALDHs aldehyde dehydrogenases ALT alanine transaminase AP-1 activator protein-1 apoB apolipoprotein B AST aspartate transaminase ATP adenosine triphosphate Bax Bcl-2-associated x protein Bid BH3-interacting domain death agonist COX-2 cyclooxygenase-2 CT computed tomography CTLA-4 cytotoxic T-lymphocyte antigen-4 CYP2E1 cytochrome P450 2E1 ER endoplasmic reticulum DF discriminant function DISC death-inducing signaling complex ECM extracellular matrix ELISA enzyme-linked immunosorbent assay FFA free fatty acids G3P glycerol-3-phosphate
GSH HCC HERs HRS HSC IgA IL-6 iNOS LBP LPS LSP MAA MARS MAT MDA MEOS MMPs mRNA MS MTP NaCNBH3 NO NK
mitochondrial glutathione hepatocellular cancer hydroxyethyl radicals hepatorenal syndrome hepatic stellate cells immunoglobulin A interleukin-6 inducible NO synthase LPS-binding protein lipopolysaccharide liver-specific membrane lipoprotein MDA-acetaldehyde molecular adsorbents recycling system methionine adenosyltransferase malondialdehyde microsomal ethanol-oxidizing system matrix metalloproteinases messenger RNA methionine synthase microsomal triglyceride transfer protein cyanoborohydride nitric oxide natural killer
INTRODUCTION Alcohol is consumed by a large percentage of the world’s population and is an effective anxiolytic and social lubricant. A small proportion of consumers become dependent, and a moderate proportion of these, and many who are not dependent, develop clinically significant liver disease. These problems are not new. Alcohol was recognized to be a cause of liver damage by the ancient Greeks, and is currently the commonest cause of liver disease in the western world. The magnitude and range of the health and socioeconomic problems attributable to alcohol abuse are enormous. Cirrhosis, predominantly alcoholic, is now the fourth commonest cause of death between the ages of 25 and 64 in the USA and alcohol may also make a significant contribution to cardiovascular-related mortality. The overall socioeconomic cost of alcohol abuse in the USA, in terms of health care, crime, and loss of work capacity, has been estimated at over $160 000 million per year. It is, therefore, a significant drain on limited health care resources. In common with all alcohol-related disease, abstinence is the cornerstone of management in patients with alcoholic liver disease (ALD). The development of specific therapies has, however, been hampered by a continued lack of a clear understanding of the mechanisms through which ethanol causes liver injury. Intense research efforts have now
NKT PAP PT PPAR-a
natural killer T phosphatidate phosphohydrolase prothrombin time peroxisome proliferator-activated receptor-a PUFA polyunsaturated fatty acid PTX pentoxifylline RA retinoic acid ROS reactive oxygen species SAH S-adenosylhomocysteine SAME S-adenosylmethionine SOD-2 superoxide dismutase SREBP-1c sterol response element-binding protein-1c TAG triacylglycerol TGF-b transforming growth factor-b TLR4 toll-like receptor 4 TNF-a tumor necrosis factor-a TNFR1 TNF-a receptor 1 TRAIL tumor necrosis factor-related apoptosisinducing ligand UPR unfolded protein response VLDL very-low-density-lipoproteins
highlighted several important metabolic and immunological consequences of excessive alcohol consumption that may contribute to disease pathogenesis and it is hoped that further defining these disease mechanisms may lead to the development of novel treatment strategies. It has also become increasingly clear that individuals are not “all equal” in their susceptibility to ALD. Although cumulative alcohol dose undoubtedly plays a role in determining disease risk, only a small proportion of heavy drinkers go on to develop the more advanced forms of ALD – hepatitis, fibrosis, and cirrhosis. Elucidating the genetic and environmental factors associated with disease progression would be a major step towards disease prevention. This chapter will focus on the pathogenetic mechanisms of ALD and the current treatments available.
EPIDEMIOLOGY Several independent studies have demonstrated a close correlation between deaths from cirrhosis and per capita alcohol consumption. Perhaps the best example of this is the effect of wine rationing in France during the Second World War; this was associated with an 80% reduction in cirrhosis deaths, followed by a return to pre-war levels when restrictions were removed.1 A similar effect was observed during Prohibition in the USA. Figure 29-1 shows the
579
Section IV. Toxin Mediated Liver Injury
30
Second Third Fourth Fifth
Sixth Seventh Eighth
Ninth
Tenth
driven by the question of how alcohol leads to liver injury. With few exceptions they fail to address the fact that most individuals appear to be remarkably resistant to the deleterious effects of ethanol on the liver.
25
Rate per 100 000 population
PATHOGENESIS 20
15
10
5
Males Both sexes Females
0 1910
1920 1930
1940 1950
1960 1970 1980
1990
2000
Year Figure 29-1. Age-adjusted death rates from liver cirrhosis by sex. States with death registration 1910–1932, all United States 1933–2000. Reproduced with permission from National Institute on Alcohol Abuse and Alcoholism (www.niaaa.gov).
decline in cases after the act was passed in 1916 and a gradual increase following the repeal of the act in 1932. The worldwide increase in mortality from cirrhosis observed during the 1950s and 1960s was associated with a similar rise in alcohol consumption, attributed largely to the falling price of alcohol relative to income.2 Conversely, the reduction in per capita alcohol intake that has occurred in several countries since the late 1970s (including in the USA) has recently been reflected in some reduction of deaths due to cirrhosis. More recently, this reduction has stabilized and, once again, there has been a rise in ALD mortality rates in some countries.3 This may be associated with a rise in alcohol consumption, but may also be due, in part, to the increased prevalence of obesity, now recognized to be an important risk factor for the development of ALD.4 In 2001, liver cirrhosis was the 12th leading cause of death in the USA, and 44.4% of the cases of cirrhosis were alcohol-related. These figures translate to around 28 500 deaths from cirrhosis, of which 12 600 are alcohol-related.5 These deaths occur in an estimated total population of 2 million individuals with ALD of varying severities that represent approximately 1 in 7 of the estimated 14 million heavy drinkers in the USA. The potential environmental and genetic explanations for this clear interindividual variation in susceptibility to ALD will be discussed in the section on pathogenesis, below. First it is important to review the putative mechanisms through which this injury occurs in the next section of this chapter on pathogenesis. Most of the studies producing the data presented below were
580
While there is good epidemiological evidence that heavy ethanol intake can result in liver disease in some individuals, there is still much debate about the main pathogenetic mechanisms through which this occurs. Several mechanisms have been proposed, and data from human and animal studies support the fact that more than one is likely to be important. The first and most direct is the effect of ethanol metabolism on liver biochemistry and the resulting steatosis and oxidative stress. The second is the indirect release of cytokines as a result of the increase in gut-derived endotoxin transported to the liver via the portal vein. The third is the liverdirected adaptive immune responses generated as a result of the development of new antigens formed by the reactive intermediates produced by the first two mechanisms. Many of these mechanisms have been elucidated using a variety of animal models. The intragastric ethanol-fed rat model designed by Tsukomoto and French has proven to be the most useful; however, there have also been mouse, guinea pig, hamster, and primate models, each producing its own challenges. As with all animal work, there is often difficulty in interpreting the data with regard to humans. Attempts to minimize this problem led to studies in the baboon, and, using this model, Rubin and Lieber described lesions resembling alcoholic hepatitis and cirrhosis.6 Working with animals so closely related to humans has obvious benefits; however, difficulty replicating the experiments and cost and ethical implications have somewhat limited the usefulness of this approach.7 Later sections of this chapter attempt to describe the putative mechanisms of ethanol-induced liver injury in detail and offer some theories as to how they may interact. First, however, there is a brief description of the absorption, distribution, metabolism, and elimination of alcohol. Alcohol metabolism will be discussed in terms of the fate of a unit of alcohol following ingestion. One unit is equivalent to 10 g or 12.5 ml of absolute alcohol which is present in approximately half a pint (284 ml) of beer and one standard measure of wine (114 ml) or liquor (24 ml).
ABSORPTION, DISTRIBUTION, AND EXCRETION The typical time course of blood alcohol concentration following the ingestion of 1 unit is shown in Figure 29-2. The peak level occurs approximately 20 min after ingestion and reaches between 10 and 15 mg/100 ml. The rate of rise and height of peak is a function of alcohol absorption and tissue distribution. In addition, it has been suggested that the peak value may be influenced by first-pass metabolism of alcohol by alcohol dehydrogenase (ADH) activity within the gastric mucosa; however, the biological importance of this effect is controversial. Alcohol is absorbed from the gastrointestinal tract by simple diffusion.8 Because of slow absorption of ethanol in the stomach, 50–80% of absorption occurs in the duodenum and upper jejunum. The rate of absorption is delayed following a meal and increases in proportion to the alcohol concentration of the drink consumed. Since absorption is more rapid from the intestine than the stomach, any pathological condition, drug, or surgical interven-
Chapter 29 ALCOHOLIC LIVER DISEASE
12 Blood alcohol (mg/100 ml)
On an empty stomach 10
Following a meal
8 6 (Km) of liver alcohol dehydrogenase activity
4 2 0 0
20
40
60
80
100
120
140
Time min Figure 29-2. The typical time course of blood alcohol concentration following the ingestion of 1 unit either following a meal (continuous line) or on an empty stomach (dotted line).
tion that delays or increases gastric emptying will also affect alcohol absorption accordingly. Following absorption, the tissue distribution of alcohol is principally determined by blood flow and water content. Thus, in organs with a rich vasculature such as brain, lungs, and liver, alcohol levels rapidly equilibrate with the blood. Alcohol is poorly soluble in lipids which will take up only 4% of the amount of ethanol that can be dissolved in a corresponding volume of water. As a result, tissues with a high fat/water ratio attain much lower levels than organs such as the kidney, where the high water content results in urinary alcohol levels 1.3 times higher than those in blood. The low lipid solubility of alcohol also explains why, following the ingestion of the same amount of alcohol per unit weight, an obese person attains a higher level of blood alcohol than a thin person. Furthermore, the higher fat content of female body composition compared to male has been invoked as part of the explanation for their higher alcohol levels following the ingestion of similar amounts of alcohol per unit weight.9 Over 90% of circulating alcohol is oxidatively metabolized, primarily in the liver, and excreted as carbon dioxide and water. The remainder is eliminated unchanged in the urine (<1%) and breath (1–5%). In view of the negligible renal and pulmonary excretion, the rate of alcohol elimination is largely determined by the body’s capacity for alcohol oxidation. The rate of alcohol metabolism does not vary widely in the population and above a concentration of 10 mg/ 100 ml occurs at a constant rate of approximately 100 mg/kg body weight per hour: so-called “zero-order” kinetics. A 70-kg man, therefore, eliminates 1 unit of alcohol in about 90 min. An important implication of this type of kinetics is the absence of a rapidfeedback mechanism to increase the rate of alcohol oxidation in response to its concentration. Heavy, repeated alcohol consumption can, however, increase the rate of elimination by up to 100%.
ALCOHOL METABOLISM Site of Alcohol Oxidation Alcohol metabolism is performed almost entirely by the liver, which contains several different high-affinity (low-Km) enzyme systems
capable of oxidizing alcohol. Other organs, including kidney, intestine, and bone marrow, also possess alcohol-oxidizing capacity, but because of the low affinity of the ADH activities present in these tissues, they make an insignificant contribution to overall alcohol oxidation at the concentrations attained following normal “social” drinking. The possible exception is the ADH activity present in the gastric mucosa, where the very high gastric levels of alcohol following ingestion may render the affinity of the enzyme(s) present less critical and the resulting alcohol oxidation significant. This effect has been claimed to contribute to a significant first-pass metabolism of alcohol determining both its bioavailability and its toxic effects.10 This gastric “barrier” may be lower in females and further contribute to their increased susceptibility to alcohol.11 The increased first-pass metabolism due to gastric ADH seems to be primarily a function of men under 40, with older men having similar or lower activity than their female counterparts.12 In addition, certain drugs, such as H2receptor antagonists, may also influence the activity of gastric ADH13 as they have been shown to influence the bioavailability of ethanol by reducing first-pass metabolism.14 A study showing that gastritis results in a reduction in gastric ADH activity without affecting ethanol bioavailability casts doubt on the clinical role of this gastric first-pass effect.15
Oxidation of Alcohol to Acetaldehyde Alcohol oxidation in the liver takes place via three steps (Figure 29-3). First alcohol is oxidized, principally within the cytosol, to acetaldehyde. Then acetaldehyde is further oxidized to acetate, primarily within the mitochondria, and finally, acetate is released into the blood and oxidized to carbon dioxide and water in peripheral tissues. At least three enzyme systems with the capacity to oxidize alcohol to acetaldehyde are present within the liver, although in normal individuals only the ADH enzymes are important. The alcohol dehydrogenase pathway. ADHs catalyze the oxidation of a variety of alcohols to aldehydes and ketones. This includes catalyzing the oxidation of ethanol to acetaldehyde and transferring hydrogen to the cofactor nicotinamide adenine dinucleotide (NAD), which is converted to its reduced form, NADH. The resulting increase in the ratio of NADH/NAD, which is further increased by acetaldehyde oxidation, is partly responsible for the metabolic imbalances that occur following alcohol ingestion and has been considered to play a major role in the initial pathogenesis of alcoholinduced fatty liver. Human ADH exhibits multiple isoenzymes that have been divided into five major classes on the basis of their electrophoretic mobility, kinetic properties, and inhibition by pyrazole.16 They are encoded by at least seven different gene loci, ADH1 to ADH7, encoding the a-, b-, g-, p-, c-, s-, and m-subunits respectively. The class I isoenzymes are formed by random association of the a-, b-, and g-subunits to form the active homo- or heterodimeric isoenzymes, whereas the others are all homodimers. The ADH2 and ADH3 genes are polymorphic, encoding three different b- and two different g-subunits with different kinetic properties, resulting in isoenzymes with different rates of alcohol oxidation in vitro.17 Expression of the ADH genes is tissue-specific. The liver contains the highest levels of class I activity, while class III activity is present equally in all tissues. In humans, the class II isoenzyme, p-ADH, has
581
Section IV. Toxin Mediated Liver Injury
Figure 29-3. The three pathways of alcohol oxidation: ADH, MEOS and catalase. NADPH, nicotinamide-adenine dinucleotide phosphate.
Ethanol O2+NADPH
Alcohol dehydrogenase
MEOS (CYP2E1) H2O+NADP
H2O2
NAD
Peroxisomal catalase
NADH
H2O Acetaldehyde
NAD Aldehyde dehydrogenase NADH
Acetate
Oxidation peripheral tissues
H2O+CO2
been found only in the liver while the class IV enzyme, s-ADH, is present only in the stomach.18 The class I isoenzymes have by far the lowest Km and highest Vmax for alcohol and accordingly are thought to be responsible for the major part of hepatic alcohol oxidation. The overall Km of liver ADH activity is in the order of 1 mmol (4 mg/100 ml), which explains why alcohol follows zeroorder kinetics at anything above very low blood levels. Experiments performed both in vivo and in vitro suggest that the principal regulatory mechanism for the ADH pathway is the capacity of the mitochondria to reoxidize NADH back to NAD.19 The microsomal ethanol-oxidizing system (MEOS) pathway. In addition to ADH, alcohol is metabolized by the MEOS, an accessory pathway that principally involves a specific alcohol-inducible form of cytochrome P450, designated CYP2E1.20 The enzyme is located on the endoplasmic reticulum (ER), is present in greater amounts in perivenular than periportal hepatocytes, and requires oxygen and NADPH. The CYP2E1 protein has been purified and the human gene cloned, sequenced, and localized to chromosome 10.21 The overall contribution of MEOS to alcohol oxidation in vivo is not yet fully clear. Its Km for alcohol is in the order of 50– 80 mg/100 ml, so it appears to play an important role at high blood alcohol levels or following chronic alcohol abuse, in view of its inducibility. Alcohol induction of CYP2E1, and microsomal enzyme systems in general, has also been implicated in the tolerance to various drugs commonly observed in alcoholics and may explain their increased susceptibility to hepatotoxicity by other drugs and xenobiotics which are converted to toxic metabolites by microsomal enzyme systems. An important example of this phenomenon is the increased susceptibility of heavy drinkers to the toxic effects of acetaminophen, where severe liver damage has been reported in alco-
582
holics taking large, but previously considered safe, doses.22,23 More recent data suggest that this effect may, at least in part, be due to an acetaminophen-induced reduction in antioxidants that subsequently renders the hepatocyte sensitive to apoptosis induced by tumor necrosis factor-a (TNF-a), a cytokine found at higher concentrations in the livers of heavy drinkers than of abstainers.24 Nevertheless, it is likely that increased metabolism plays a significant role in the increased sensitivity of drinkers to acetaminophen. Enhanced microsomal enzyme activity may also lead to an increased rate of testosterone breakdown, contributing to low blood levels of hormone already decreased due to inhibition of testosterone production by the direct toxic effects of alcohol on the testes.25 The methods by which ethanol induces CYP2E1 are not entirely clear but are likely to be multifactorial, including increased transcription, increased mRNA stabilization, increased translation, and reduced degradation. The catalase pathway. The third pathway for alcohol oxidation is catalyzed by the enzyme catalase. This enzyme is located in the peroxisomes of most tissues and requires the presence of hydrogen peroxide. The reaction is limited by the availability of hydrogen peroxide which is low in normal circumstances, and suggests that the catalase pathway accounts for less than 2% of overall in vivo alcohol oxidation.26
Oxidation of Acetaldehyde to Acetate Over 90% of the acetaldehyde formed from alcohol oxidation is further oxidized in the liver to acetate by aldehyde dehydrogenases (ALDHs). ALDH, like ADH, uses NAD as a cofactor and further increases the NADH/NAD ratio. Human ALDHs are encoded at four independent loci on four different chromosomes.27 ALDH2 on
Chapter 29 ALCOHOLIC LIVER DISEASE
chromosome 12 encodes the major mitochondrial enzyme, which has a low Km for acetaldehyde and is responsible for the majority of acetaldehyde oxidation. The ALDH2 gene exists in at least two allelic forms, ALDH2*1 and ALDH2*2.28 Isoenzymes present in individuals homozygous for the ALDH2*2 allele have little or no catalytic activity, while those present in heterozygotes have measurable, although reduced, activity compared to the isoenzymes present in ALDH2*1 homozygotes.29 The inactive form of ALDH2 is present in about 50% of orientals but has not been found in caucasian populations. Increased levels of acetaldehyde are thought to be the mechanism through which homozygotes for the ALDH2*2 allele develop the “flushing” reaction after alcohol. Interestingly, heterozygotes for the allele appear to develop advanced liver disease with lower levels of ethanol consumption.30 This, too, is putatively secondary to increased concentrations of acetaldehyde and adds to the extensive evidence that acetaldehyde is centrally involved in the pathogenesis of ALD. The cytosolic form of ALDH, ALDH1 has a higher Km for acetaldehyde than ALDH2 and may play a role following the ingestion of large doses of alcohol. ALDH-inhibitors such as disulfuram (Antabuse) have been used in the treatment of alcoholism to sensitize alcoholics to the unpleasant effects of alcohol intake secondary to high levels of acetaldehyde.
Alterations in Alcohol Metabolism Following Chronic Consumption Many studies have shown that chronic alcohol consumption increases the rate of alcohol elimination except in the presence of severe liver damage. This increase is due both to alcohol induction of the MEOS and to adaptive changes in the ADH pathway. The basis for the increased activity of the ADH pathway is probably increased mitochondrial reoxidation of NADH to NAD, which, as discussed previously, is the important rate-limiting step. It has been suggested that the increased mitochondrial NADH-reoxidation rate is secondary to alcohol-induced stimulation of Na+,K+-ATPase activity, leading to enhanced adenosine triphosphate (ATP) and oxygen consumption.31 This so-called “hypermetabolic state” of the liver has also been implicated in the pathogenesis of alcohol-related liver injury. Alcohol elimination is decreased in jaundiced patients with alcoholic cirrhosis and animals with non-alcohol-related liver disease;32 this probably reflects decreased ADH activity.33 An important consequence of the increased rate of alcohol oxidation in alcoholics is that, following alcohol ingestion, levels of acetaldehyde in both blood and tissues are higher than those seen after similar ingestion in non-alcoholic controls.34,35 This increase is potentiated by a reduction in the capacity of the mitochondria to oxidize acetaldehyde, at least in alcohol-fed rats, and a reduction in total hepatic ALDH activity, observed in chronic alcoholic patients with and without liver disease.36 As discussed, this may have important implications for disease pathogenesis. Having discussed the pathways of ethanol metabolism, the next sections will review the evidence for how this impacts on disease pathogenesis.
ALCOHOL METABOLISM AND THE PATHOGENESIS OF ALD Many studies have focused on the downstream effects of ethanol metabolism in an effort to explain the biochemical and histological
features of ALD. These studies have yielded several mechanisms through which this metabolism may result in the generation of fatty liver (steatosis), oxidative stress/lipid peroxidation, and acetaldehyde, all of which are thought to be important in disease pathogenesis.
The Pathogenesis of Fatty Liver The accumulation of triacylglycerol (TAG), synthesized via the sequential esterification of glycerol-3-phosphate (G3P) within the liver, is an early and reversible effect of alcohol consumption in humans and animal models of ALD. It is the consequence of increased substrate supply (glycerol and free fatty acids (FFA)), increased esterification, and decreased export of TAG from the liver.37 The precise molecular mechanisms that lead to these three main effects have recently been elucidated (Figure 29-4). The role of dietary fat. It is intuitive to suspect that dietary fat will have a role in the development of hepatic steatosis, and, indeed, rat models of ALD have shown that the rate of development of fatty liver is proportional to the fat content of the diet.38 Further studies in these rats in conjunction with human epidemiological data have also highlighted the role that different types of dietary fat may play in influencing the severity of the more advanced forms of ALD such as necroinflammation and fibrosis. These will be discussed in the section on oxidative stress and lipid peroxidation, below. While the quantity of dietary fat can increase the supply of fat to the liver, alcohol intake also increases the lipolysis of adipose tissue, further increasing the concentration of circulating FFA. These are then taken up by the liver and provide the substrate for TAG synthesis. The high concentrations of FFA further promote the synthesis of TAG by increasing the activity of the enzyme phosphatidate phosphohydrolase (PAP), which is the rate-limiting step for TAG synthesis catalyzing the dephosphorylation of phosphatidic acid to diacylglycerol.37,39 Altered redox state. From the three well-characterized metabolic pathways discussed above, two appear to be of clinical and pathogenetic importance: the ADH pathway and the MEOS. The first of these involves the oxidation of ethanol to acetaldehyde by cytosolic ADHs, and subsequent oxidation to acetate by predominantly mitochondrial ALDH. Both of these steps are coupled to the reduction of NAD to NADH. The increased NADH/NAD ratio has profound effects on the metabolism of carbohydrates and lipids. Gluconeogenesis is impaired and substrate flow through the citric acid cycle is diminished, with acetyl coenzyme A diverted towards ketogenesis and fatty acid synthesis. In addition to increased fatty acid synthesis, the altered redox state also directly inhibits fatty acid oxidation. Through these two mechanisms, altered redox state can contribute towards increased substrate supply. The altered NADH/NAD ratio also increases the production of the other key component of TAG, G3P, thereby again promoting TAG synthesis. The role of PPAR-a inhibition. Recent evidence suggests that ethanol also has an effect on fatty acid oxidation through the transcriptional factor peroxisome proliferator-activated receptor-a (PPAR-a). This ligand-activated receptor/transcription factor is a critical component in the regulation of mitochondrial, microsomal
583
Section IV. Toxin Mediated Liver Injury
Substrate supply -oxidation (PPAR-␣-inhibition)
Dietary fat intake
Fatty acid synthesis (SREBP-1c induction)
NADH/NAD
Lipolysis of adipose fat
Figure 29-4. The multiple mechanisms by which ethanol metabolism can result in fatty liver. Ethanol metabolism contributes to fatty liver by increasing substrate supply, increasing fat esterification to triglyceride, and reducing the export of very-lowdensity lipoprotein (VLDL) from the liver. PPAR, peroxisome proliferator-activated receptor; SREBP-1c, sterol regulatory element-binding protein 1c.
Glycerol-3-phosphate
Free fatty acids
Esterification
Phosphatidate phosphohydrolase Triglyceride Export from the liver VLDL Increased by alcohol Inhibited by alcohol
and peroxisomal fatty acid oxidation systems in the liver.40 As FFA are ligands for this receptor, ethanol consumption, which increases hepatic fatty acid levels via the mechanisms outlined, would be expected to result in an induction of enzymes in the oxidation systems and a subsequent increase in fatty acid oxidation. In fact, ethanol-feeding results in a decrease in transcription and activity of many of these enzymes due to an inhibition of the transcriptional and DNA-binding activity of PPAR-a.41 This effect was replicated with acetaldehyde, and inhibited by inhibiting ethanol metabolism, implicating acetaldehyde as the factor leading to the PPAR-a inhibition.41 Treatment with the PPAR-a agonists WY14 643 and clofibrate reversed the effects of ethanol-feeding and the resulting abnormalities in hepatic lipid metabolism in rodent models42,43 while alcohol-fed PPAR-a null mice develop more steatosis, hepatocyte injury, and fibrosis than their wild-type litter mates.44 It appears, therefore, that PPAR-a inhibition plays a critical role in the accumulation of fatty acids in the liver after ethanol-feeding, which will then promote TAG synthesis and potentially necroinflammation and fibrosis. It also seems likely that acetaldehyde is a key component of this inhibition. The precise mechanism of acteldehyde-induced PPAR-a inhibition is still unclear. Mechanisms of altered triglyceride export. In addition to increased substrate supply and esterification resulting in increased levels of TAG in the liver, a decrease in the export of TAG from the liver also appears to contribute to the generation of ethanolinduced steatosis. Usually fat is exported into the circulation in the form of very-low-density-lipoproteins (VLDL). With chronic ethanol ingestion this export mechanism becomes defective for reasons that are not entirely clear but appear to involve the Golgi. Acetaldehyde produced during ethanol metabolism can bind atubulin45 and disrupt microtubule dynamics.46 Possibly as a result of this, fat accumulates in the Golgi and mainly, or at least initially, in the perivenular hepatocytes. This situation is compounded by ethanol-induced down-regulation of microsomal triglyceride trans-
584
fer protein (MTP).47 MTP is the principal enzyme responsible for packaging TAG and apolipoprotein B (apoB) into VLDL particles. In the absence of effective lipidation, apoB is degraded in the proteosome. Role of tumor necrosis factor-a in the pathogenesis of steatosis. The putative role of TNF-a in the pathogenesis of necroinflammation in ALD is described in the section on endotoxin below. Only recently, however, has this cytokine been linked to the development of hepatic steatosis (Figure 29-6). Interest in this area stemmed from a study showing that TNF-a receptor 1 (TNFR1)deficient mice developed considerably less steatosis than their wild-type litter mates when fed ethanol.48 Recently, the mechanisms for this TNF-a-induced steatosis have been better elucidated. They include the down-regulation of MTP (discussed above), the increased expression of sterol response element-binding protein-1c (SREBP-1c),49 a transcription factor critical in controlling de novo hepatic lipogenesis, and inhibition of adiponectin, an antisteatotic adipocytokine.50 Since one of adiponectin’s antisteatotic effects is via PPAR-a, the inhibition of adiponectin by TNF-a provides a potential mechanism for ethanol’s inhibition of PPAR-a activity, discussed above. Inhibition of the methionine cycle and endoplasmic reticular stress. Studies predominantly in micropigs have demonstrated that the inhibition of transmethylation reactions by ethanol is the major mechanism leading to the well-established abnormal methionine metabolism associated with ethanol consumption (Figure 29-5).51 The principal enzyme inhibited by ethanol is methionine synthase (MS),52 and a significant effect of this appears to be the development of hyperhomocysteinemia. Hyperhomocysteinemia has been implicated in the pathogenesis of atherosclerosis and Alzheimer’s disease through a phenomenon known as ER stress. The ER is the main site of protein synthesis from messenger RNA (mRNA), and is also involved in protein trans-
Chapter 29 ALCOHOLIC LIVER DISEASE
ATP
Folate/B12
Methionine 1
2 3 Homocysteine
Betaine
S-adenosylmethionine
Methyl group Glutathione
Figure 29-5. The methionine cycle. (1) Methionine adenosyltransferase catalyzes the synthesis of S-adenosyl-L-methionine (SAME) from methionine and adenosine triphosphate. After donation of a methyl group, SAME becomes S-adenosylhomocysteine, which, through homocysteine, acts as a precursor for glutathione. (2) Methionine synthase regenerates methionine from homocysteine in a reaction that requires normal levels of folate and vitamin B12 and (3) betaine-homocysteine methyltransferase regenerates methionine from homocysteine in a reaction that requires betaine. Ethanol inhibits reactions (1) and (2).
S-adenosylhomocysteine
port and some post-translational modifications.53 When abnormally folded or unfolded proteins build up in the ER this acts as a marker to the cell that the quantity of “client” protein exceeds the ability of the ER to process it and results in a set of responses termed the “unfolded protein response” (UPR) or “ER stress.”54 This response, which can be triggered by homocysteine, has three main arms. The first results in the increased transcription of ER proteins and chaperone proteins that aid processing. The second is to reduce the biosynthesis of other proteins in order to reduce the unfolded client load. The third is to up-regulate proapoptotic protein synthesis so that, in extreme circumstances, the cell will apoptose. In addition, ER stress triggered by homocysteine increases the gene expression of SREBP-1c.55 Previous studies have shown that ethanol induces SREBP-1c in rat hepatoma cell lines and mouse liver with a concomitant increase in the expression of lipogenic genes,56 and that this induction may be related to inhibition of adenosine monophosphate-activated protein kinase (AMPK).57 Whether this inhibition of AMPK by ethanol is related to ER stress is unknown; however, studies with the intragastric ethanol-fed mouse have shown that ethanol induces hyperhomocysteinemia, triggers ER stress, and promotes the features of ALD, including steatosis. Furthermore, feeding betaine, a methyl donor converting homocysteine to methionine, significantly inhibits the development of steatosis, implying that it is the inhibition of methylation and resulting hyperhomocysteinemia that is responsible.58 It has recently been shown that the effect of ethanol on the development of hyperhomocysteinemia is independent of TNF-a, and that these two mechanisms induce steatosis independently and in parallel.59 The role of steatosis in the pathogenesis of advanced ALD. These described mechanisms all appear to promote the rapid and reproducible accumulation of hepatic TAG. This has been shown in humans and several animal models of alcoholic liver injury. The next question is: is this fat accumulation harmful or benign? A growing body of evidence suggests that, rather than being an epiphenomenon of excessive alcohol intake, steatosis may play a direct role in progression to more advanced disease.60 In several prospective studies of heavy drinkers the severity and pattern of steatosis on index biopsy predict the subsequent risk of fibrosis and cirrhosis.61,62 These and other studies have led to steatosis being considered as the “first hit,” increasing the sensitivity of the liver to a variety of “second hits” that result in injury and inflammation. These “second hits” could be gut-derived endotoxin, oxidative stress, or, indeed, immune mechanisms. In support, studies in animal models have shown that steatosis increases endotoxin-mediated necroinflamma-
Table 29-1. Evidence for the role of oxidative stress in the pathogenesis of alcoholic liver disease (ALD) Presence of lipid peroxidation products Site of lipid peroxidation products Multiple potential sources of reactive oxygen species in heavy drinkers Depletion of antioxidant defenses Manipulating oxidative stress in animal models influences disease severity
Found in serum and liver of patients with ALD and correlate with histological severity Found in perivenular region where ALD starts and is most severe Hepatocyte microsomes and mitochondria, Kupffer cells Depletion of glutathione, selenium, and coenzyme Q is found in heavy drinkers Pro-oxidant diets worsen disease while over- or underexpression of antioxidant enzymes affects the degree of liver injury
tion63 and the degree of lipid peroxidation in ethanol- and other drug-induced steatosis.64 Furthermore, genetically obese mice with steatosis have altered proportions of intrahepatic lymphocyte subpopulations. The normal liver contains significant numbers of T cells, B cells, and natural killer (NK) cells and natural killer T (NKT) cells, many of which differ phenotypically and functionally from circulating lymphocytes. Leptin-deficient ob/ob mice have steatotic livers and, interestingly, a selective reduction in the number of NKT cells. While these mice are not a model of alcohol-induced steatosis, this finding does raise questions about whether steatosis itself may alter the intrahepatic immune milieu.65
Oxidative Stress and Lipid Peroxidation Oxidative stress describes a situation where the generation of prooxidant species within or outside the cell overwhelms the endogenous antioxidant systems (Table 29-1). One of the most important consequences of cellular oxidative stress is the peroxidation of the polyunsaturated fatty acid (PUFA) constituents of membrane and lipoprotein lipids which can lead directly to cell death66 and to the release of reactive aldehydes with potent pro-inflammatory, profibrotic, and proimmune properties.67 An accumulating body of evidence now supports a role for oxidative stress and lipid peroxidation in the pathogenesis of ethanol-induced liver injury that can be summarized as follows: 1. Products of lipid peroxidation can be detected in the peripheral blood of heavy drinkers68 and in the livers of patients with ALD,69 and the magnitude of lipid peroxidation correlates with the degree of liver injury.70
585
Section IV. Toxin Mediated Liver Injury
2. In patients and animal models of ALD, lipid peroxidation is most prominent in the perivenular region where the liver injury is typically most severe. 3. A variety of sources of oxidative stress have been identified in patients with ALD and in animal models of disease. 4. Ethanol consumption results in the depletion of endogenous antioxidant capabilities and patients with ALD have evidence of antioxidant deficiencies. 5. In animal models of ALD, dietary and genetic manipulations that increase oxidative stress increase the severity of liver injury and reducing oxidative stress ameliorates injury. Sources of oxidative stress in ALD. The principal pro-oxidant species considered to be important in the pathogenesis of ALD are the reactive oxygen species (ROS), the superoxide anion, hydrogen peroxide, and hydroxyl and hydroxyethyl radicals (HERs). Considerable controversy remains over the most important source of these ROS in ALD but the most likely appear to be microsomal CYP2E1 (the only source of HERs71,72), the mitochondrial electron transport chain,73 inducible nitric oxide (NO) synthase, and Kupffer cells. CYP2E1. In the presence of iron, isolated microsomes can generate sufficient oxidizing species to initiate lipid peroxidation. Most of the ROS in this system come from the ethanol-inducible MEOS. The major component of this system is CYP2E1, which can also oxidize alcohol to acetaldehyde and subsequently to acetate, while concomitantly oxidizing NADPH to NADP. It appears that most ROS in the ethanol-fed rat model appear to be generated by CYP2E1. Spin-trapping studies, experiments which allow the identification of highly reactive radicals produced during reactions for very short periods of time to be identified, have revealed that the HER is the main radical formed when CYP2E1 metabolizes ethanol. The best in vivo evidence for the important role of CYP2E1 metabolites comes from experiments in the ethanol-fed rat discussed above, where enzyme inhibition with diallyl sulfide resulted in decreased lipid peroxidation and amelioration of liver.74,75 Mitochondria. The mitochondrial respiratory electron transport chain generates superoxide radicals during the reoxidation of NADH arising during ethanol metabolism. The superoxide anions are subsequently converted to hydrogen peroxide by mitochondrial superoxide dismutase (SOD-2). Although the hydrogen peroxide is further metabolized to carbon dioxide and water by mitochondrial
Endotoxin CD14 TLR4
586
Inducible nitric oxide synthase. NO has recently emerged as a critical regulator of certain aspects of mitochondrial function.79 In particular, NO can bind to cytochrome-c oxidase and increase the rate of superoxide production.80 After ethanol consumption, liver mitochondria are much more susceptible to NO-dependent inhibition of respiration. Furthermore, mice deficient in the inducible enzyme that synthesizes NO, inducible NO synthase (iNOS), develop much milder liver disease after ethanol consumption.81 The fact that these iNOS–/– mice have dramatically reduced levels of liver lipid peroxidation end-products and reactive nitrogen species provides compelling evidence for the role of this enzyme in oxidative stressrelated injury. As with the mitochondrial production of ROS, TNFa is likely to play a critical role in the generation of oxidative stress via iNOS due to its NFk-B-dependent up-regulation of iNOS gene transcription. Kupffer cells. In addition to playing a central role in the generation of proinflammatory cytokines in response to endotoxin (see below), activated Kupffer cells also act as a rich source of free radicals during ethanol-induced liver injury. Inhibition of Kupffer cells with gadolinium chloride during alcohol exposure has a profound inhibitory effect on the magnitude of lipid peroxidation,82 and mice deficient in Kupffer cell NADPH-oxidase have no increase in free radical production and develop no liver pathology after 4 weeks’ exposure to ethanol.83 Depletion of antioxidant defenses in ALD. Ethanol consumption undoubtedly results in a depletion of endogenous antioxidant capabilities. Consumption of GSH during oxidative stress and inhibition of two enzymes involved in the synthesis of its precursor, Sadenosylmethionine (SAME), MS, and methionine adenosyltransferase (MAT), contribute to the decreased levels of hepatic SAME
Pro-steatotic through SREBP-Ic MTP adiponectin
TNF-␣
Kupffer cell
glutathione (GSH) peroxidase, the increased rate of production after ethanol consumption results in a significant production of ROS.76 As discussed in detail below, TNF-a may play an important role in the mitochondrial production of ROS through inhibition of electron flow in the respiratory chain (Figure 29-6). Perhaps the best evidence that mitochondria are involved in the pathogenesis of ALD are the ultrastructural changes that occur in these organelles in human and experimental ALD.77,78 These changes are almost certainly related to mitochondrial oxidative stress and its downstream effects.
Pro necrotic through ROS if mitochondrial GSH depleted
Pro apoptotic through caspase 8/ mitochondrial death pathway
Figure 29-6. Tumor necrosis factor-a (TNF-a), a central cytokine in alcoholic liver injury. TNF-a induces steatosis through the up-regulation of sterol response element-binding protein-1c (SREBP1-c), the down-regulation of microsomal triglyceride transfer protein (MTP) and the inhibition of adiponectin. Depending on the degree of oxidative stress and the availability of mitochondrial glutathione (GSH), TNF-a can also induce either necrosis or apoptosis through the death-inducing signaling complex, with subsequent activation of caspase 8 and the mitochondrial death pathway.
Chapter 29 ALCOHOLIC LIVER DISEASE
and GSH observed in patients with ALD.84 Depletion of mitochondrial GSH precedes and promotes the progression of alcoholic liver injury in animal models,85 with one mechanism of action being an increased sensitivity of hepatocytes to TNF-a-induced cytotoxicity.86 A recent study has suggested that, rather than reduced GSH levels, a high S-adenosylhomocysteine (SAH)-to-SAME ratio may be responsible for the ethanol-induced enhanced sensitivity of hepatocytes to TNF-a killing, possibly by increasing the activity of caspase-8, a key initiator of the apoptotic cascade.87 Heavy drinkers, including those with ALD, are deficient in the antioxidant trace element selenium,88 which is required for the activity of the antioxidant enzyme GSH peroxidase, the antioxidant vitamins A, C, and E89,90 and coenzyme Q.91 This latter compound is present in plasma and mitochondrial matrix membranes and has emerged as one of the most important, natural free radical scavengers. It is partly derived from the diet, but is also synthesized in the liver. Whether these deficiencies are a cause or an effect of ALD remains unclear, although the fact that antioxidant supplementation appears to be of no benefit in patients with ALD is perhaps more in favor of the latter explanation.92 Manipulating oxidative stress in animal models influences disease severity. Diets that promote oxidative stress increase the severity of ALD in animal models. Perhaps the best example of this comes from studies feeding animals a diet supplemented with different types of dietary fat in addition to alcohol. The degree of liver injury depends on whether the animals had their diet supplemented with beef fat, lard, or corn oil, with beef fat resulting in the least injury and corn oil the most.93 A hypothesis that liver injury correlated with the amount of linoleic acid in the diet was supported by a study in which beef fat was supplemented with linoleic acid. These rats, like the corn oil-fed rats, developed severe liver disease when fed ethanol.94 Severe liver injury was also seen when fish oil was used as the source of dietary fat, implying that polyunsaturated fats, which are much more vulnerable to attack from ROS, have a role in promoting liver injury.95 The mechanisms through which polyunsaturated fats promote ethanol-induced liver injury are not entirely clear, but appear to involve up-regulation of CYP2E1 and an increase in lipid peroxidation.95 In addition, polyunsaturated fats are the precursors of eicosanoids, powerful mediators of inflammation. Interestingly, cyclooxygenase-2 (COX-2), the key enzyme involved in the generation of eicosanoids, is inducible by both endotoxin and TNFa.96 Dietary iron supplementation also leads to an increase in the concentration of the aldehyde end-products of lipid peroxidation within the liver. The addition of this pro-oxidant to the diet in the intragastric ethanol infusion model exacerbates hepatocyte damage and promotes liver fibrogenesis.97 Studies employing genetic manipulation to increase the degree of oxidative stress provide further support for its role as an important disease mechanism in ALD. Cytosolic SOD-1 knockout mice develop more significant liver injury compared to wild-type litter mates following ethanolfeeding,98 while MAT knockout mice have reduced hepatic GSH levels and develop spontaneous steatohepatitis even without alcohol.99 Further evidence supporting the role of oxidative stress in ALD comes from studies aimed at reducing its levels during ethanolfeeding. Reduction of oxidative stress by supplementation with the
GSH precursor and methyl donor SAME reduced cell and mitochondrial injury in the baboon model of ALD100 and led to the pilot study in humans discussed below. Inhibiting the pro-oxidant effect of iron with the oral chelator 1,2-dimethyl-3-hydroxypyrid-4-one reduced hepatic-free iron, lipid peroxidation, and fat accumulation in chronically ethanol-fed rats101 and inhibiting the induction of CYP2E1 with diallyl sulfide and phenylethyl isothiocyanate resulted in the production of fewer free radicals and end-products of lipid peroxidation and ameliorated liver injury in the same model.74,75 Amelioration of liver injury could also be achieved using an adenovirus to deliver the mitochondrial, manganese-dependent SOD-2 to rats fed ethanol,102 and, as discussed above, NADPH-oxidasedeficient mice have also been used to show the importance of this source of oxidants in the development of early alcohol-induced hepatitis.83
Acetaldehyde It is widely considered that acetaldehyde, the first metabolite of ethanol, has a central role in the pathogenesis of ALD, but precisely what role is still unclear. It has been known for some time that acetaldehyde can bind covalently to albumin,103 tubulin,104 hemoglobin,105 plasma proteins,106 collagen,107 and microsomal enzymes108 and results in both stable and unstable adduct formation. These reactions are likely to involve the formation of Schiff bases between the aldehyde and valine, lysine, and tyrosine residues on the carrier protein.105 This binding may affect protein function, and this disruption has been implicated in the pathogenesis of ALD.109 Defects in assembly of microtubules,110 protein excretion,111 and enzymatic activity109 have all been attributed to acetaldehyde; however no firm evidence exists for these mechanisms as disease pathways. With regard to disruption of enzymatic activity, an interesting study identified a 37-kDa protein-acetaldehyde adduct in the liver of alcohol-fed rats to be the enzyme D4-3-ketosteroid 5b-reductase.112 This was felt to be important because children having inborn errors in this enzyme develop intrahepatic cholestasis and in some cases liver failure. While this may be just one of many proteins that adduct to acetaldehyde, it raises questions about whether the binding of acetaldehyde may indeed alter enzyme activity. As well as potentially disrupting protein function, however, the formation of proteinacetaldehyde adducts results in the production of immunodominant antigenic determinants. The consequence of this is discussed in detail under the section on antibodies to acetaldehyde adducts, below. More recently, a further and more complex role for acetaldehyde has been suggested. Hepatocytes are resistant to TNF-a-induced cytotoxicity unless they have been previously exposed to ethanol.113 As discussed above, selective depletion of mitochondrial GSH can induce this sensitization, and is postulated to be the mechanism through which it happens in vivo. When HepG2 cells are exposed to acetaldehyde, there is a selective reduction in mitochondrial GSH and increased sensitization to TNF-a.86 This has been shown to occur even when further metabolism of acetaldehyde is inhibited, and recent evidence suggests it is due to an inhibition of GSH transport into the mitochondria secondary to an increased proportion of cholesterol in the mitochondrial membrane and a resulting increase in viscosity.114 It is postulated that acetaldehyde induces this increase in cholesterol through the unfolded protein and ER stress
587
Section IV. Toxin Mediated Liver Injury
response via SREBP-1c up-regulating the transcription of cholesterol-synthesizing enzymes. This effect on GSH transport will exacerbate the GSH depletion resulting from decreased SAME synthesis and consumption during oxidative stress. In this way, the induction of SREBP-1c by TNF-a and by acetaldehyde and homocysteineinduced ER stress can induce a vicious cycle resulting not only in steatosis, but also in sensitization to the cytotoxic effects of TNFa. This is particularly significant in view of the role of endotoxin in ALD, which acts predominantly as a stimulus to the release of TNFa from hepatic Kupffer cells.
THE INNATE IMMUNE SYSTEM AND THE PATHOGENESIS OF ALD A considerable body of evidence supports a role for the innate immune system activated by gut-derived endotoxin in the pathogenesis of ALD.
toxin; however, with time, this response converts to one of sensitization.126 An early suppressive effect of alcohol on Kupffer cell function and a later induction of CD14, both attributed to increased levels of endotoxin, may explain the time course of this response. Interestingly, when Kupffer cells are inactivated in the ethanol-fed rat model using gadolinium chloride, disease is ameliorated and the steatosis is diminished. This highlights that these cells also have a role in the TAG accumulation, most likely through TNF-a production, as discussed above. Sinusoidal endothelial cells are also actively involved in the response to endotoxin. These cells constitutively express all the surface molecules necessary for antigen presentation, and may induce tolerance or immunity depending on the local microenvironment. This environment may be dictated by Kupffer cell release of immunomodulatory cytokines in response to varying doses of portal endotoxin, with high-dose endotoxin resulting in potentially harmful immunity, and low-dose endotoxin resulting in tolerance through the secretion of IL-10.127
Endotoxin Endotoxin, which refers collectively to the lipopolysaccharide (LPS) components of the cell wall of all Gram-negative bacteria, appears to play a central role in the development of ALD. Ethanol ingestion increases the translocation of endotoxin from the gut lumen115 to the portal circulation where it is normally recognized by cells of the innate immune system, intrahepatic macrophages (Kupffer cells). Endotoxinemia has been found in drinkers with varying degrees of liver disease,116 and in the ethanol-fed rat, the level of plasma endotoxin correlated with the degree of liver injury.117 When Kupffer cells encounter the LPS component of endotoxin in conjunction with LPS-binding protein (LBP)118 they respond by releasing cytokines and ROS. While this can be blocked in vitro by the addition of anti-CD14 monoclonal antibodies,119 implying the importance of this constitutively expressed and inducible receptor, CD14 does not have a transmembrane component, and therefore must rely on alternative receptors for signal transduction. Experiments on mice hyporesponsive to LPS have helped to identify the Toll-like receptor 4 (TLR4) as an important concomitant signal,120 and current data suggest that LPS signaling occurs through activation clusters of these two receptors along with other membrane proteins. The importance of TLRs has been further highlighted recently by the discovery that they control activation of adaptive immune responses as well as innate responses. This occurs through an interleukin-6 (IL-6)-dependent pathway, resulting in suppression of CD4+CD25+ T-regulatory cells that normally specifically inhibit adaptive immune responses.121 This has profound implications for the role of endotoxin in ALD. The role of endotoxin and Kupffer cells in the pathogenesis of ALD has been suggested largely by studies in the rodent model of continuous intragastric feeding developed by Tsukamoto et al.122 As discussed above, these animals develop lesions similar to human alcoholic hepatitis when fed a diet high in ethanol and fat. The eradication of Gram-negative fecal flora with antibiotics reduces endotoxin levels to those of controls not fed alcohol, and prevents liver injury.123 Recent data reveal that this injury is also attenuated in CD14 and TLR4 knockout mice, highlighting the importance of these receptors.124,125 Alcohol can also affect Kupffer cell function directly. Rodents fed alcohol initially become more tolerant of endo-
588
The Role of TNF-a Kupffer cells are the primary intrahepatic source of TNF-a, a cytokine believed to be central to the pathogenesis of ALD. Peripheral blood mononuclear cells produce more basal and LPS-induced TNF-a than controls,128 and plasma levels are higher in patients with more severe disease.129 Furthermore, mice lacking the TNFR1 fail to develop liver injury in the Tsukamoto–French model.48 More recently, it has been suggested that these mice develop an ameliorated form of ALD rather than none at all;59 nevertheless, the findings suggest an important role for the cytokine. How TNF-a contributes to hepatocyte cytotoxicity in ALD, and whether this is via necrosis or apoptosis is an area of dispute; however, recent studies showing a correlation between the degree of apoptosis and clinical indices of severity in patients with alcoholic hepatitis suggest that apoptosis plays an important role.130,131 The interaction of TNFa with its receptor TNFR1 (via the death-inducing signaling complex (DISC)) activates procaspase 8 into caspase 8, which cuts BH3-interacting domain death agonist (Bid).132 Truncated Bid can enter the outer mitochondrial membrane to make this membrane leaky, and it also induces a conformational change in Bcl-2associated x protein (Bax), which translocates to mitochondria and associates with its analogue Bak, to form channels in the outer mitochondria membrane.133 Increased permeability of the outer mitochondrial membrane releases cytochrome-c from the intermembrane space of mitochondria, thus partially blocking the flow of electrons into the respiratory chain and increasing mitochondrial ROS formation.134 The resulting ROS then act on the same or other mitochondria to open an inner-membrane pore, the mitochondrial permeability transition pore. This causes an influx of water into the mitochondrial matrix, mitochondrial swelling, and eventually rupture of the unfolded outer membrane. This leads to the leakage of apoptosis-inducing factors (predominantly cytochrome-c) into the cytosol, where they can activate caspase-9, which initiates the apoptotic cascade. Given the role of ROS in this cascade, it is easy to see how the oxidative stress-related mechanisms of alcohol-induced cell injury can synergize with TNF-a to induce apoptosis and/or necrosis. TNF-
Chapter 29 ALCOHOLIC LIVER DISEASE
a-induced apoptosis is increased by alcohol, an effect that is exaggerated in HepG2 cells with high CYP2E1 activity, presumably related to increased oxidative stress.113 As discussed, the depletion of GSH by alcohol will also increase the sensitivity of hepatocytes to the mitochondrial effects of TNF-a.24,86 It has recently been shown that the mechanism of TNF-a-induced cell death, necrosis, or apoptosis depends on the degree and site of GSH depletion.135 Mitochondrial depletion leads to necrosis due to profound loss of mitochondrial function86 while less severe, predominantly cytosolic, depletion of GSH leads to apoptosis due to inhibition of the NFkB-induced transactivation of survival genes that normally follows TNF-a binding to its receptor. As also discussed above, in addition to TNF-a and oxidative stress, the apoptotic cascade may also be initiated in response to ER stress.
Natural Killer Cells A further, less well-understood arm of the innate immune system that appears to be involved in the pathogenesis of ALD is the NK and NKT cell population. These cells are abundant in the liver136 and both populations have been reported to increase in the peripheral blood of heavy drinkers.137 Recent evidence points to an increase in the number of NKT cells in the livers of mice after alcohol consumption and augmenting their activation using a marine sphingolipid during ethanol-feeding results in fatal hepatotoxicity.138 The relevance of this to clinical liver disease is as yet unclear and is made more complicated by the fact that acute ethanol consumption inhibits innate immunity by suppressing TLR3 signaling,139 and may inhibit NK cell-killing of activated stellate cells, which would have marked profibrogenic effects (see below).
THE ADAPTIVE IMMUNE SYSTEM AND THE PATHOGENESIS OF ALD For some time there has been interest in the role of adaptive immune mechanisms in the pathogenesis of ALD. Several clinical features imply that they may have a role in disease pathogenesis. Abstinent patients who return to drinking have a rapid and aggressive recurrence of their disease, which might imply an immunological anamnestic response. There is a partial response to immunosuppressive steroids in selected groups with severe disease.140 It has been reported that patients receiving interferon therapy for hepatitis C can rapidly develop alcoholic hepatitis with no increase in their daily alcohol intake.141 This is particularly interesting in view of the fact that interferon can act as a trigger for autoimmune thyroid disease in genetically susceptible individuals. Patients with ALD also often display hypergammaglobulinemia and lymphocyte infiltration is a well-recognized histological feature of advanced disease.142 These features have led researchers to look for the specificity of humoral and cellular immune responses in patients with ALD, and at how antigen-specific immune responses may lead to liver injury. Almost all of the work thus far has focused on the humoral immune response.
Humoral Immune Responses in ALD For more than 20 years humoral immune responses have been implicated in the pathogenesis of ALD.143 Although the antigenspecificity of the response was unknown, liver cell membrane
specific antibodies of immunoglobulin A (IgA) and IgG class were detected in patients with alcoholic hepatitis and alcoholic cirrhosis and the percentage of IgG-stained hepatocytes was shown to correlate with the level of transaminase.144 Since these early studies, various groups have attempted to define the main antigenic determinants for these responses. The first efforts focused on determining whether the situation in ALD may mirror that found in immune-mediated drug-induced liver disease. In halothane hepatitis, hepatic proteins are altered by trifluoroacetylation in the course of oxidative biotransformation of halothane,145 and multiple antibodies develop in affected individuals that are specific for these altered proteins.146 Furthermore, autoantibodies to native cytochrome P450 2E1 (CYP2E1), the halothane-metabolizing enzyme, are found in most patients with halothane hepatitis.147 In tielinic acid (a uricosuric drug) hepatotoxicity, a similar situation arises, with autoantibodies to the hepatic microsomal enzyme, CYP2C9, being a specific marker of disease.148 Tielinic acid is known to bind covalently to this enzyme. It is this alteration of the parent enzyme that is then thought to lead to a breakdown in tolerance to the native protein.149 In order to study whether the situation is similar in ALD, humoral immune responses to proteins altered by ethanol metabolism and host proteins were investigated. Initially, work focused on the major reactive metabolite of ethanol metabolism implicated in so many other aspects of liver injury, acetaldehyde.
Antibodies to Acetaldehyde Adducts Studies with acetaldehyde condensates. A study by Israel et al. in 1986 revealed that mice could be immunized with erythrocyte protein-acetaldehyde condensates, yielding antibodies that would cross-react with plasma protein-acetaldehyde condensates, but not with the native plasma proteins.150 This implied that the immune responses developed against the altered component of the protein. In addition to this, the antibodies recognized poly-(-L-lysine) condensates, but not those made with tyrosine or valine, suggesting that it is the adducted lysine residues that are immunogenic in vivo. Importantly, stable adducts can be formed at acetaldehyde concentrations of 200–300 mmol in vitro, concentrations that are found in the livers of ethanol-fed animals. Adducts made at these concentrations, however, were not very immunogenic, and it required heavily conjugated condensates prepared with the reducing agent cyanoborohydride (NaCNBH3) to yield high antibody titers in mice. Nevertheless, the antibodies developed in this way do cross-react with adducts made at concentrations of acetaldehyde found in vivo. A further finding from this study was that mice fed ethanol for 45–50 days developed antibodies that reacted with the condensates. While the need for reduced condensates in ethanol-fed animals has led to suspicion with regard to their direct role in disease pathogenesis, these observations opened up a potential route through which immune mechanisms could lead to liver injury in ALD. Furthermore, and perhaps more importantly, they provide the stimulus for further studies exploring both the development and turnover of these adducts in animals and humans, and the possibility that acetaldehyde is not the only reactive intermediate that may form immunogenic adducts. The discovery that the adduction of acetaldehyde to host proteins could result in immunogenic epitopes was an important step in determining how ethanol metabolism might cause liver injury
589
Section IV. Toxin Mediated Liver Injury
through immune mechanisms. The finding of antibodies to the condensates in the serum of ethanol-fed mice suggested that these adducts were formed in vivo. A further study reporting the presence of these antibodies in the sera of alcoholic patients suggested that they were also formed in humans.151 The first report of the formation of a protein acetaldehyde adduct in vivo was in 1988 when Lin et al. demonstrated an acetaldehyde-modified protein of 37 kDa in the liver of rats fed alcohol for 7 weeks.152 Although the native hepatic protein was not identified, it was shown to be present in non-alcohol-fed rats, implying that it was modified and not induced by the process of alcohol-feeding. In this study the adducts were identified by immunoblotting with serum from rabbits immunized against acetaldehyde-adduct condensates. Confirmation that it was the acetaldehyde component that was recognized was achieved by successfully immunoblotting with antibodies that were obtained by immunizing rabbits with acetaldehyde adducted to either hemocyanin or myoglobin. Interestingly, these antibodies did not recognize liver homogenate incubated with high concentrations of acetaldehyde overnight in vitro, implying that chronic alcoholfeeding of at least 1 week was required to generate the adducts. Since this study, the presence of acetaldehyde-modified proteins has been confirmed in the liver cytosol, membranes, and mitochondria of rats chronically fed ethanol, and their decline over several weeks after ethanol withdrawal has been investigated.153 Acetaldehyde adducts in human ALD. An immunohistochemical study of acetaldehyde adduct distribution in human liver revealed that heavy drinkers showed positive intracellular staining, and that it was particularly hepatocytes in the perivenular region, the first area to develop disease in ALD, that were affected.154 This is of particular interest as this is the main area of distribution of the ethanol-metabolizing enzyme CYP2E1 and the main site of lipid peroxidation. Minimal alcohol consumers had little or no staining. A further, larger human study confirmed that zone 3 of the liver acinus was the first area to become positively stained, but that all hepatocytes in the lobule could stain positively.155 There was also distinct intra- and extracellular staining, with extracellular staining being found more frequently in patients with acute alcoholic hepatitis. The real strength of this second study was that a large number of patients had a second biopsy available for examination that had been taken a mean of 3 years after the first. This longitudinal study allowed fibrosis progression to be correlated with the pattern of staining. Interestingly, extracellular and not intracellular staining correlated with fibrosis progression and this was regardless of whether the patients abstained from alcohol or not. It appears then, that while still a research tool, the detection of acetaldehyde modified proteins in patients’ livers may confirm heavy alcohol intake. Furthermore, the pattern of these proteins may help to determine prognosis. The question still remains, however, as to why some heavy drinkers with acetaldehyde adducts in their liver get ALD, while others do not. While this is still not clear, the answer to this may partly lie in the interindividual variability in the immune response to these adducts. Antiacetaldehyde antibodies and severity of liver disease. As discussed above, the work by Israel et al. in 1986 revealed that mice fed ethanol would develop significant, albeit low, titers of antiacetaldehyde adduct antibodies in vivo. Confirmation of these find-
590
ings, and the later confirmation that adducts were formed in rat livers in vivo, strengthened the argument that the immune response may have a role in liver damage in ALD. These antiacetaldehyde (anti-AcA) adduct antibodies have been found in the sera of heavy drinkers by several groups;155–157 however their relationship to the presence and severity of liver disease is controversial. An initial study showed titers higher in heavy drinkers than controls, with the highest levels being found in patients with acute alcoholic hepatitis.157 A further report looking at serum from 86 biopsy-proven ALD patients found antiadduct IgA titers correlated with serum bilirubin and IL-6 levels, while antiadduct IgG titers correlated with the presence of liver inflammation and necrosis.158 Importantly, this study also had a population (no ALD or NALD group) without liver disease that had been drinking at least 80 g ethanol/day for 4 weeks prior to the study and had a history of repeated inebriations. While this was not the ideal group for selecting patients who will not develop ALD with continual heavy alcohol intake, it was an important addition to the study. Antiacetaldehyde adduct IgA titers were found to be highest in the ALD population, and lower in both the NALD and control groups. Antiacetaldehyde adduct IgG titers were also found to be highest in the ALD and NALD groups, and lower in the control group. The authors concluded that the specific antibodies of an IgA isotype were a marker of liver disease, while those of an IgG isotype better represented ethanol intake.
Antibodies to the Other Products of Ethanol Metabolism While most of the initial studies of humoral immune responses concentrated on the response to acetaldehyde adducts, it is now clear that these are not the only target antigens recognized in ALD. The HER is a reactive intermediate formed by the action of CYP2E1 during ethanol metabolism.159 HERs have been shown to bind to microsomal proteins with high efficiency, and, when these are injected into rabbits, they can induce an immune response in a similar way to acetaldehyde adducts.160 Anti-HER antibodies have also been found in significantly higher titer in patients with alcoholinduced cirrhosis than in non-alcohol-induced cirrhosis or normal controls.161 These do not cross-react with acetaldehyde adducts. Other aldehydes generated as a result of prolonged oxidative stress, such as malondialdehyde (MDA) and 4-hydroxynonenal, which are present in the livers of rats fed ethanol chronically,162 can also adduct to host proteins and induce humoral immune responses in vivo.163 The third and final antigen that has been studied in detail is a combination of acetaldehyde and MDA. As these are both reactive compounds formed at similar concentrations in the liver of ethanol-fed animals, Tuma et al. set out to investigate whether they could react together with the amino groups of proteins and form distinct adducts in vitro, and whether this happened after chronic ethanolfeeding in vivo.164 Interestingly, MDA greatly enhanced the binding of acetaldehyde to albumin and led to the formation of hybrid conjugates they called MDA-acetaldehyde (MAA) adducts. Antibodies could be generated in rabbits, and could detect these adducts in chronically ethanol-fed rats. Antibodies to these hybrid conjugates have now been detected in sera from patients with alcohol-induced liver cirrhosis and alcoholic hepatitis but not in heavy drinkers with no liver disease or healthy controls.165 Importantly, competitive inhi-
Chapter 29 ALCOHOLIC LIVER DISEASE
bition experiments confirmed that the anti-MAA antibodies were unrelated to those against acetaldehyde.
Autoantibodies Non-specific autoantibodies. The concept of ALD being at least partly an autoimmune disease is not a new one. A preparation called liver-specific membrane lipoprotein (LSP) was used in the late 1970s as a source of normal liver membrane proteins for antibody experiments.166 It was found that serum from most patients with active autoimmune hepatitis would react with this preparation.167 Anti-LSP antibodies were also detected in around 30% of patients with ALD.168 These were particularly found in those patients who had interface hepatitis on their liver biopsies, and only rarely in patients with acute alcoholic hepatitis. Further non-specific studies investigating potential autoantigens were performed using liver membrane antigen isolated from rabbit hepatocytes.169 Again, these were found in around a quarter of patients with ALD.143 While a variety of other autoantibodies have been detected in ALD, their significance is unclear. There are, however, two areas of particular interest which may give an insight into disease pathogenesis. The first is with regard to antiphospholipid antibodies, and the second is antibodies to the inducible microsomal enzyme, CYP2E1. Antiphospholipid antibodies. Antiphospholipid antibodies are not exclusive to patients with the classic antiphospholipid syndrome. They are also found in some patients with chronic infections, lymphoproliferative diseases, sickle-cell anemia, and ALD. Most of the binding of these antibodies to antigen appears to be mediated through the interaction between the apoprotein a2-glycoprotein 1 and phospholipids.170 Interestingly, in ALD there appears to be little binding of patients’ serum to phospholipid that has been protected from oxidation. If the phospholipid is allowed to autooxidize, however, 58% of the patients’ sera are reactive.170 This reactivity could be removed by preadsorption on a2-glycoprotein 1-coated enzyme-linked immunosorbent assay (ELISA) plates. This implies that the act of oxidation alters the structure of the interaction between apoprotein and phospholipid, and renders it immunoreactive with patients’ sera. These studies suggest that the oxidative injury that is a well-recognized feature of ALD may alter normal proteins or lipids and result in altered self that can induce an immune response. While these are not true autoantibodies, their presence gives an insight into how tolerance may be broken to host. Other autoantibodies. The other autoantibodies that have been extensively studied are those to the inducible microsomal enzyme CYP2E1. Again, the generation of these antibodies shows not only that tolerance can be broken to host proteins in ALD, but also gives a feasible mechanism through which this may happen. As described above, when ethanol is metabolized by CYP2E1, HER is produced,161 which, when adducted to host proteins, is immunogenic. It has now been demonstrated that HER binds to microsomal proteins in rats and humans, and that these adducts can elicit an immune response when injected into rabbits. Furthermore, a significant proportion of patients with ALD have positive humoral responses to these adducts. HER binds to multiple targets, and, as a result, there are several targets for the anti-HER immune response; however 86% of patients with alcohol-induced cirrhosis had antibodies reactive with HER complexed with CYP2E1.171 This was a
higher percentage than to any of the other identified adducts. As various cytochrome P450s can be targets for autoimmunity in drug hypersensitivity,148 the breakdown of tolerance to native CYP2E1 has also been assessed. Both patients with ALD and rats fed ethanol have been shown to develop these specific autoantibodies.172 Furthermore, their development is prevented by the inhibition of CYP2E1 induction. Interestingly, in humans with ALD, the presence of CYP2E1 autoantibodies is predicted by the presence of antibodies to HERs, suggesting a potential mechanism for tolerance breakdown.173 Although several target antigens have now been identified for the humoral immune response found in advanced ALD, the important question of whether these antibodies are a cause or a consequence of liver disease still remains. The antibodies may be produced in response to liver damage, or directly partake in it. At present this question remains largely unanswered; however, several groups have tried to show a role for the specific antibodies found in ALD in disease progression.
The Role of Antibodies in Disease Pathogenesis In order to try and answer the question of whether these antibody responses play a role in the etiology and pathogenesis of ALD a number of functional studies have been done. The initial studies observing cytotoxicity when lymphocytes from patients with alcoholic hepatitis were co-cultured with rabbit hepatocytes were interpreted as suggesting that antibody-dependent cell cytotoxicity had a role in ALD.174 The cytotoxicity was non-T-cell-dependent and inhibited in these experiments by the addition of normal liver plasma membrane components. In reality these experiments are difficult to interpret. More recently, however, the potential importance of antibody-dependent cell cytotoxicity has been more convincingly suggested by a series of experiments studying HER.175 These studies were performed in the knowledge that CYP2E1 could metabolize ethanol, producing a reactive intermediate that would adduct to the enzyme and induce a humoral immune response. While the humoral immune response to acetaldehyde adducts may be important in disease progression, its potential role in the initial stages of ALD was questioned by the fact that the adducts are not found on the surface of hepatocytes, and therefore, even if an immune response is raised against them, this would not necessarily lead to hepatocyte killing. In contrast, cytochrome P450 isoenzymes, including CYP2E1, are expressed on the outer layer of rat and human hepatocyte plasma membranes.175 Furthermore, HER protein adducts have been found on the surface of isolated rat hepatocytes incubated with ethanol by indirect immunofluorescence. These adducts were not found if the hepatocytes were not ethanoltreated. If sera from patients with ALD were added to the hepatocyte cultures, and antibody binding detected by a fluorescein-labeled antihuman IgG, surface immunofluorescence could be detected that was not present if sera from control subjects were used. These experiments could be repeated using hepatocytes isolated from rats fed ethanol in vivo, with similar results. The fact that the surface immunofluorescence detected is at least partly due to anti-HERCYP2E1 adducts on the hepatocyte surface was confirmed by confocal microscopy using anti-CYP2E1 and anti-HER antibodies. Antibody-dependent cell cytotoxicity was observed when sera from patients with ALD positive for anti-HER antibodies were co-
591
Section IV. Toxin Mediated Liver Injury
Cellular Immune Responses in ALD Both mice and human livers have significant numbers of T cells and some B cells, which recirculate from the peripheral pool; however, in contrast to peripheral blood, there are more CD8+ cells than CD4+ cells.136 In addition there are high frequencies of what appear to be resident liver lymphocytes that are found in very low frequencies in peripheral blood. These are NK cells, recognized by CD56 in humans and NK1.1 in mice, and NKT cells that express these proteins along with the T-cell receptor. In these populations the T-cell receptor is expressed at a lower level than on circulating T cells, and has limited diversity. It is hypothesized that these cells form a link between the innate and adaptive immune responses as they recognize foreign glycolipids associating with non-classical major histocompatibility complex. As such, they will be able to mount specific responses without the need for clonal expansion. In normal liver, these cell types are distributed throughout the parenchyma. In liver disease, this situation changes, and in alcoholic cirrhosis expanded portal tracts contain large numbers of classical CD4+ and CD8+ lymphocytes.142 Again, CD8+ cells outnumber CD4+ cells. No work has been done to determine the antigenspecificity of these intrahepatic lymphocyte populations; however peripheral blood mononuclear cell responses to MDA have been detected in a significant proportion of patients.176 Further work is required to determine whether antigen-specific cellular immune responses are a crucial part of disease progression in ALD or whether they are a relatively insignificant epiphenomenon of liver injury. This work will have obvious therapeutic implications.
produced and an increase in collagen types I, III, and IV.178 In addition, these changes are associated with an up-regulation in numerous integrins, selectins, and soluble growth factors that modulate cell–cell and cell–ECM interactions. Sinusoidal endothelial cells179 and lymphocytes180 may also play important roles in fibrogenesis. More recently, there has been increasing evidence that fibrosis can be reversible, and the discovery of matrix metalloproteinases (MMPs) that degrade collagen has implications for most liver diseases. These MMPs are activated by proteolytic cleavage and are inactivated by tissue inhibitors of metalloproteinases. In ALD fibrosis starts in the perivenular area and ultimately progresses to bridging fibrosis and ultimately cirrhosis (Figure 29-7). As discussed, this early fibrosis occurs at the site of maximal alcoholinduced hepatocyte injury. The principal cell types involved in the activation of HSC in ALD are Kupffer cells and hepatocytes. Kupffer cells are stimulated to release cytokines by endotoxin in ALD, and in turn, these cells can lead to the activation and proliferation of stellate cells through the production of transforming growth factor-b (TGF-b), TNF-a, and ROS. Hepatocytes are rich sources of ROS, lipid peroxidation products, and acetaldehyde
100 90 80
Ferritin (ng/ml serum)
cultured with ethanol-treated rat hepatocytes and peripheral blood mononuclear cells from healthy controls.175 Again, this was not found when sera from healthy individuals were used. Similar cytotoxicity was observed using serum from rabbits immunized with HERs, but none using preimmune serum. Interestingly, cytotoxicity was greatly reduced by preincubating the serum with albumin adducted to the HER. These in vitro studies provide a mechanism whereby antibody responses to HER may be involved in liver damage. They do not, however, confirm that this mechanism is important in disease pathogenesis, and further studies are required to provide evidence for the role of this mechanism of disease in vivo.
70 60 50 40 30 20 10 0 0
2
Fibrosis and ultimately cirrhosis is the final common pathway of most chronic liver disease, and the mechanisms underlying it are discussed elsewhere in this book. It is, however, important to review these briefly and to discuss the factors that are particularly relevant to the pathogenesis of the fibrosis seen in ALD. Hepatic stellate cells (HSC) are found in the space of Disse, between hepatocytes and sinusoidal endothelial cells, and are responsible for producing the majority of the extracellular matrix (ECM).177 In normal liver the space of Disse contains little collagen; however, when these stellate cells become “activated” during liver injury by cytokines and ROS, the composition of the ECM changes as more collagen, glycoproteins, proteoglycans, and glycosaminoglycans are produced. In particular, there is a shift in the type of proteoglycans
592
6
8
10
12
14
16
18
20
22
Years Mean duration of alcohol abuse
MECHANISMS OF ALCOHOL-INDUCED FIBROSIS
4
Mean daily alcohol intake
3.6
8.3
12.9
21.6
(1–5 y)
(6–10 y)
(11–15 y)
(1>15 y)
163g
177g
192g
227g
(144–210)
(160–224)
(197–275)
129
81
51
(average of minimum and maximum) (130–197)
No of cases
73
Cirrhosis of the liver Cirrhosis and potentially precirrhotic lesions (severe steatofibrosis with inflammatory reactions, chronic alcoholic hepatitis) Moderate-to-severe fatty infiltration Figure 29-7. Frequency of cirrhotic and precirrhotic liver lesions according to dose and duration of alcohol consumption in 334 drinkers. (Reproduced from Lelbach WK. Cirrhosis in the alcoholic and its relation to the volume of alcohol abuse. Ann NY Acad Sci 1975; 252:85–105, ©1975 with permission of New York Academy of Science.202)
Chapter 29 ALCOHOLIC LIVER DISEASE Figure 29-8. Alcoholic hepatitis.
during alcohol-induced injury, all of which have been shown to enhance collagen production by HSC.67,181,182 CYP2E1 may be particularly important in this regard given its inducibility by alcohol and a high-fat diet and its perivenular distribution. HSC grown in the presence of hepatocyte cell lines that overexpress CYP2E1 increase their production of collagen, an effect that is prevented by antioxidants or a CYP2E1 inhibitor.183 Hepatocyte apoptosis is a notable feature of alcoholic hepatitis (Figure 29-8),130 and apoptosing hepatocytes express Fas, that can promote stellate cell initiation through the tumor necrosis factor-related apoptosis-inducing ligand (TRAIL).184 Furthermore, apoptosing hepatocytes may also be ingested by Kupffer cells and HSC which subsequently release TGFb, capable of activating HSC.185,186
MECHANISMS OF HEPATOCELLULAR CANCER Epidemiological studies reveal that alcohol plays a major contributory role in the development of hepatocellular cancer (HCC); however the primary mechanisms through which this occurs are not clearly defined (Table 29-2). Cirrhosis itself is a precancerous condition, and alcohol-related HCC without pre-existing cirrhosis is rare. Nevertheless, three features indicate that alcohol may be a cocarcinogen. The first is that heavy alcohol consumption is associated with several extrahepatic cancers (discussed below). The second is that when the incidence of incidental HCCs in liver explants from patients with alcoholic cirrhosis is compared with that from other etiologies, it lies between that of immune-mediated liver disease and viral hepatitis.187 It appears, therefore, that the incidence of HCC
593
Section IV. Toxin Mediated Liver Injury
Table 29-2. The primary mechanisms thought to be involved in ethanolinduced hepatic fibrosis and hepatocellular carcinoma Fibrosis Kupffer cell production of profibrotic cytokines Kupffer cell production of profibrotic reactive oxygen species Hepatocyte production of profibrotic reactive oxygen species Hepatocyte production of acetaldehyde Kupffer cell and hepatic stellate cell production of transforming growth factor-b after ingestion of apoptotic hepatocytes Hepatocellular carcinoma Lipid peroxidation and DNA mutagenesis Activation of carcinogenic xenobiotics Antiapoptotic effect of tumor necrosis factor-a DNA hypomethylation Immunosuppression
is above that of the “baseline” expected due to pure cirrhotic risk. The third is that there are several plausible mechanisms through which alcohol could promote carcinogenesis.
Lipid Peroxidation and DNA Mutagenesis MDA, an end-product of lipid peroxidation, can bind to DNA and form adducts as it does with other endogenous compounds.188 These adducts were found to be highly mutagenic in Escherichia coli189 and are repaired by nucleotide excision repair. They are also found at significant levels in healthy humans, and can induce cell cycle arrest.190 This latter property results in an increase in the number of hepatic progenitor cells (oval cells)191 which are more resistant to oxidative stress than fully differentiated hepatocytes. It has been suggested that this may promote HCC as the oval cells survive through oxidative damage but remain susceptible to mutagenesis.191
Activation of Xenobiotics Another mechanism proposed for the increased rate of HCC seen in alcoholic cirrhosis is the increased production of carcinogenic metabolites from other environmental carcinogens (xenobiotics) that are metabolized through the MEOS and other metabolic pathways up-regulated in heavy drinkers. This mechanism has been suggested to explain the increased cancer risk seen with tobacco smoking,192 aflatoxin,193 and other chemicals.194 In addition, CYP2E1 is responsible for the metabolism of retinoic acid (RA) in the liver.195 Up-regulation of CYP2E1 by ethanol therefore synergizes with its inhibition of RA synthesis and results in reduced RA levels, increased expression of the activator protein-1 (AP-1) (c-Jun and c-Fos) transcriptional complex, and increased hepatocyte proliferation.196 Supplementation with RA reverses this effect.195
TNF-a-induced survival factors As discussed above, TNF-a has both pro- and antiapoptotic properties and the balance of these appears to depend on the local microenvironment and the disease. While apoptosis may reduce the risk of HCC, increased cell survival through TNF-a-induced NFkB activation could have the opposite effect, particularly in combination with the mutagenic effects of lipid peroxidation products.
594
Reduced DNA methylation DNA methylation is an important negative regulator of gene expression and hypomethylation of oncogenes has been shown in human and rat HCC.197 Chronic ethanol consumption results in reduced concentrations of SAME, the main methyl donor (as discussed above), and dietary depletion of SAME increases the risk of HCC in rats.198
Immunosuppression Malnutrition, vitamin deficiencies, and acute ethanol per se can all result in reduced immunosurveillance. Of particular relevance is the effect on NK cells, thought to be central to tumor surveillance. While this is debated, the primary functional effect appears to be one of suppression.137,199 In addition, there are other, more widespread effects on the innate and adaptive immune responses that could all have knock-on effects on tumor surveillance (discussed below). A reduction in immunosurveillance and a subsequent increase in viral replication may also be the mechanism through which alcohol leads to an increased rate of HCC in hepatitis C cirrhosis.200
SUSCEPTIBILITY TO ALCOHOLIC LIVER DISEASE While the majority of heavy drinkers will develop some degree of steatosis (fatty liver), only around a third go on to develop alcoholic hepatitis and only between 1 in 4 and 1 in 12 ever progress to cirrhosis.201 This leads to the obvious question: what factors determine whether or not a heavy drinker develops advanced ALD?
DOSE OF ETHANOL The observation that only a minority of heavy drinkers develop ALD was first reported 30 years ago by Lelbach. This author showed that, although the risk of disease increased in proportion to the duration of intake, only 20% of consumers of more than 200 g of ethanol (around 20 standard “drinks”) per day develop cirrhosis after 13 years and less than 50% after 20 years.202 Further work from around this time showed that women appeared to develop ALD at lower doses of alcohol consumption than men.203 More detailed studies examining the precise dose–response relationship between alcohol intake and risk of ALD, the gender effect, and the risk threshold have been reported in the last 10 years. A large cohort study from Italy involving 6917 subjects between the ages of 12 and 65 reported that the risk of developing ALD begins at 30 g/day of ethanol.201 However, only 5.5% of the individuals drinking this much showed signs of liver disease. The risk increased according to daily dose, reaching 10% at 60 g/day. The study also reported that risk is higher among the over-50s if alcohol is drunk outside mealtimes or consumed in a variety of different beverages rather than one “tipple of choice.” Interestingly, this study showed no gender effect. Further evidence for a dose–response relationship and a risk threshold came from an even larger study from Copenhagen involving 13 285 subjects between the ages of 30 and 79.204 A self-administered questionnaire assessed intake, and incidence of disease was taken from death certificates and hospital
Chapter 29 ALCOHOLIC LIVER DISEASE
medical records. This study revealed a dose-dependent increase in risk, with women having a significant risk above 7–13 units per week, and men 14–27 units per week. This group updated their data analysis recently by looking at type of alcohol consumed.205 Their results suggest that the highest risk is seen in drinkers who do not include wine in their drinking repertoire. Furthermore the relative risk of cirrhosis fell as the proportion of wine increased. This association between wine intake and ALD risk may be confounded by other factors associated with wine drinking such as a lower prevalence of obesity compared to beer and spirit drinkers. While all these studies have their flaws, with data collection being the most obvious, they do allow a number of conclusions to be drawn. No dose of alcohol confers a guarantee of developing cirrhosis regardless of the period it is consumed for, and relatively low doses can cause problems.
DIET The data from ethanol-fed rats linking a diet high in polyunsaturated fats with an increased risk of alcoholic liver injury, discussed above, are supplemented by an epidemiological study linking cirrhosis mortality with pork (high in linoleic acid) consumption and dietary intake of unsaturated fats.206 A further case–control study from France has reported that the risk of cirrhosis is increased by diets high in fat and alcohol and low in carbohydrate.207 A more obvious role for diet in ALD risk has been suggested by two studies showing that obesity and associated hyperglycemia increase the incidence of all stages of ALD in heavy drinkers.4,208 While these studies have provided evidence that dose, pattern, and type of alcohol consumption and dietary (and presumably exercise-related) factors play a role in determining ALD risk, they have also demonstrated that other endogenous factors are likely to be equally, if not more, important.
GENDER AND RISK OF ALD The most obvious endogenous or “genetic” factor determining ALD risk is female gender. It has long been appreciated that women develop ALD at a lower intake of alcohol than men. The traditional explanation has been that women develop higher blood alcohol concentrations per unit of alcohol consumed due to their lower volume of distribution for alcohol. This, in turn, is attributed to their lower body mass index and to fat constituting a higher percentage of their body mass than in men. More recent evidence has, however, suggested an explanation based on disease mechanisms. Enomoto and colleagues have demonstrated in the rat model that estrogen increases gut permeability to endotoxin and accordingly up-regulates endotoxin receptors on Kupffer cells, leading to an increased production of tumor necrosis factor in response to endotoxin.209 These exciting data suggest several new directions for research into human gender-specific susceptibility to ALD.
NON-GENDER-LINKED GENETIC FACTORS AND RISK OF ALD Evidence for non-gender-linked genetic susceptibility to ALD comes principally from a twin study showing that the concordance rate for alcoholic cirrhosis was three times higher in monozygotic than in dizygotic twin pairs.210 This difference in concordance rates was not
entirely explained by the difference in concordance rates for alcoholism per se. Further indirect evidence of a genetic component to disease risk comes from the observation that the death rate from ALD is subject to wide interethnic variation that is not entirely explained by variations in the prevalence of alcohol abuse.211,212 Hispanics appear to be at particularly high risk, for example. Difficulties in performing family linkage studies in ALD have resulted in almost all of the relevant information thus far coming from classical case–control, candidate gene, allele association studies. Accordingly these studies are subject to all the common pitfalls of this type of study design and must be interpreted with caution.213 Many early reports of positive associations are likely to be subject to type I errors (chance findings), while negative reports may be subject to type II errors (false negatives) attributed to small underpowered studies. Given that the most likely mechanisms of hepatocyte injury in excessive drinkers are related to fat accumulation, oxidative stress, endotoxin-mediated release of proinflammatory cytokines, and immunological damage, the majority of studies reported thus far have focused on genes encoding proteins involved in these various pathways.
Genes Influencing the Severity of Steatosis Recognition of the role played by steatosis in the pathogenesis of more advanced liver disease60 suggests that factors determining its severity may play a key role in determining the risk of cirrhosis. Clearly genetic and environmental factors determining the degree of obesity would fall into this category, as would functional polymorphisms of genes encoding enzymes involved in hepatic lipid metabolism. Of interest in this respect is a preliminary report that a “low-activity” promoter polymorphism in the gene encoding MTP, the principal protein responsible for the export of fat from the liver, is associated with an increased risk of advanced ALD.214
Genes Influencing Oxidative Stress and Risk of ALD The principal class of genes that influences the oxidant load in heavy drinkers are those genes encoding enzymes involved in alcohol metabolism. Polymorphisms have been identified in two of the seven genes encoding ADHs (ADH2 and ADH3), in the promoter region of the CYP2E1 gene and in the coding region of the gene encoding the mitochondrial form of ALDH (ALDH2). The genes encoding ADH2 and ALDH2 undoubtedly play a role in determining the risk of alcoholism and, to a lesser extent, ALD in oriental populations.215–217 Previously reported associations with ADH3 probably reflect linkage disequilibrium with ADH2.218 In caucasians, results from studies reported to date support a role for the ADH2 polymorphism in determining the risk of alcoholism, but not ALD.219 Several studies have looked for an association between the c2 promoter (Rsa I) polymorphism of the CYP2E1 gene and ALD with no consistent results emerging in any population, although one study did report that the cumulative lifetime alcohol intake of patients with ALD heterozygous for the c2 (more transcriptionally active) allele was almost half that of patients with ALD homozygous for the c1, wild-type allele.220 The HFE gene is another obvious candidate gene for ALD, since liver iron promotes oxidative stress and iron deposition is common in ALD. Unfortunately, a case–control study
595
Section IV. Toxin Mediated Liver Injury
of over 400 patients and controls found no evidence of an association between ALD and either of the HFE mutations associated with hemochromatosis.221 This lack of association was explained by the observation that hepatic iron content did not differ between patients with and without the mutations. The lack of any striking associations between polymorphisms in genes encoding proteins involved in the generation of ROS and ALD has recently turned attention towards polymorphisms in genes encoding proteins involved in the body’s antioxidant defenses. Manganese-dependent SOD-2 is the most important mitochondrial antioxidant enzyme and a polymorphism altering its mitochondrial targeting sequence has been associated with ALD in a small French study,222 although not confirmed in a larger study from the UK.223 This and other polymorphisms affecting the function of antioxidant defense systems are clearly worthy of further study.
Endotoxin Receptor and Cytokine Genes and Risk of ALD Evidence supporting a role for endotoxin-mediated cytokine release in the pathogenesis of ALD, together with the identification of promoter polymorphisms in genes encoding endotoxin receptors, cytokines, and cytokine receptors, has recently suggested an alternative set of candidates to explain genetic susceptibility to ALD. CD14, an LPS receptor on monocytes, macrophages, and neutrophils, has no intracellular domain but enhances signaling through another LPS receptor, TLR4. A C/T polymorphism is present at position -159 in the CD14 promoter, with the TT genotype associated with increased levels of soluble and membrane CD14.224 A study from Finland has recently reported an association between possession of the TT CD14 genotype and advanced ALD;125 however, this has not been observed in a larger study in north-east England.225 This latter study also showed no association between ALD and possession of the Asp299Gly polymorphism in the TLR4 gene, previously reported to be linked to hyporesponsiveness to LPS.226 With respect to polymorphisms in the cytokine genes, the first such association was reported between alcoholic hepatitis and a polymorphism at position -238 in the TNF-a promoter region.227 The functional significance of this polymorphism is, however, unclear and the association may well be either spurious or due to linkage disequilibrium with another true “disease-associated polymorphism” on chromosome 6. An association with ALD has also been reported for a promoter polymorphism in IL-10. IL-10 is the classical anti-inflammatory cytokine which inhibits: (1) the activation of CD4+ T-helper cells; (2) the function of cytotoxic CD8+T cells and macrophages; (3) class II human leukocyte antigen/B7 expression on antigen-presenting cells; and (4) HSC collagen synthesis. A variant CÆA substitution at position -627 in the IL-10 promoter has been associated with decreased reporter gene transcription, decreased IL-10 secretion by peripheral blood monocytes, and an increased response to a-interferon in patients with chronic hepatitis C – all consistent with the polymorphism being associated with lower IL-10 production. A strong association between possession of the A allele and ALD has been reported from a study of over 500 heavy drinkers with and without advanced liver disease.228 This is consistent with low IL-10 favoring inflammatory and
596
immune-mediated mechanisms of disease as well as HSC collagen production.
Immune Response Genes and Risk of ALD In view of the immunoregulatory functions of IL-10, the association between ALD and a low-activity promoter polymorphism in IL-10 may be considered as further evidence that immune mechanisms are involved in the pathogenesis of ALD. Further evidence supporting a role for immune mechanisms in determining individual susceptibility to ALD has come from a recent study showing that, compared to drinkers with no evidence of ALD, patients with ALD are more likely to have high titers of autoantibodies against CYP2E1173 and to have T-cell responses against oxidative stress-derived adducts.176 Cytotoxic T-lymphocyte antigen-4 (CTLA-4) is a T-cell surface molecule that normally acts to “damp down” the immune response to antigens either directly, by competing with CD28 on the surface of CD4+ Th cells for the antigen-presenting cell co-stimulatory molecule B7, or indirectly, by activating T-regulatory cells which act to inhibit CD4+ Th cell function.229 CTLA-4 knockout mice develop lethal autoreactive lymphoproliferative disease and an AÆG polymorphism in exon 1 leading to a ThrÆAla substitution has recently been associated with autoimmune liver diseases, insulin-dependent diabetes, and autoimmune thyroid disease. These associations strongly suggest that this polymorphism is associated with impaired CTLA-4 function, although recent data suggest that other tightly linked CTLA-4 polymorphisms may be responsible for the functional effect.230 Although the exon 1 polymorphism has been associated with the titer of anti-CYP2E1 antibodies in one study173 and with ALD in another,231 this has yet to be confirmed as an ALDsusceptibility allele in large studies examining the full CTLA-4 gene haplotype. In future, the choice of candidate genes for detailed study as ALD risk factors is likely to be guided by: (1) genome and proteome expression studies in tissue from patients with various stages of ALD; (2) whole-genome single nucleotide polymorphism scans of cases and controls; and (3) mouse mutagenesis studies. Most importantly, however, as with other so-called complex diseases, establishing reliable genetic associations is critically dependent on the collection of large numbers of well-phenotyped cases and controls, which almost certainly require national and multinational collaborations. Only then are we likely to come up with associations that are robust enough to guide targeted treatment and prevention strategies.
CLINICAL FEATURES DIAGNOSIS Chronic alcohol abuse produces a wide range of morphological changes in the liver, the most frequent being fatty liver (steatosis), alcoholic hepatitis, and cirrhosis.232,233 For ease of presentation the three principal lesions will be discussed separately in terms of their pathology, clinical features, and prognosis, but it is important to appreciate that alcohol-related liver damage is a spectrum, with the various lesions occurring more commonly in combination than in isolation. Significantly, the clinical manifestations of each of these histological lesions are extremely variable, ranging from completely
Chapter 29 ALCOHOLIC LIVER DISEASE
asymptomatic forms to a first presentation with severe hepatic failure. Patients with none or minimal symptoms are, however, more likely to have the earlier, more reversible, forms of liver disease and therefore the early recognition of these patients is critical to allow intervention at a stage when it is likely to be of most benefit. Patients most commonly present with symptoms unrelated to the liver, typically non-specific digestive symptoms or vague psychiatric complaints. The patient may seek advice concerning the social effects of alcohol abuse on family life or work performance. Often, physical examination will be normal, other than occasional plethora, suffused conjunctivae, tremulousness, and aggressive behavior. Up to 30% of patients with ALD have no symptoms related to excessive alcohol intake and may present following the chance finding of hepatomegaly or abnormal blood tests at routine medical examination. The key to the early recognition of patients with alcoholrelated disease is a high index of suspicion. Once the diagnosis is suspected it is usually easy to confirm by direct questioning for alcohol history and alcohol-related symptoms, careful clinical examination, and supportive laboratory investigations.
History Features in the history important for both the confirmation of alcohol abuse and to aid in its subsequent management include the amount and duration of alcohol intake, the pattern of intake, precipitating factors of drinking bouts, and evidence of physical dependence such as early-morning tremor, blackouts, and morning drinking. Confirmation of the history should be sought from a family member or close associate. Specific liver-related symptoms, such as jaundice and hematemesis, should be sought but are uncommon, even in patients with established disease. In addition, since all alcoholics with liver disease do not necessarily have disease of alcoholic etiology,234 enquiries should be made concerning other risk factors for liver disease, including a history of foreign travel, blood transfusions, or intravenous drug use.
Clinical Examination Important features to note on examination are the signs of chronic liver disease, including hepatomegaly, and signs indicative of alcoholrelated pathology in other organs, such as hypertension, atrial fibrillation, and a cushingoid appearance. It is important to understand that many of the classical signs of chronic liver disease, including spider nevi, Dupuytren’s contractures, palmar erythema, and parotid swelling, can occur in alcoholics in the absence of cirrhosis. Clinical signs and history cannot be relied upon to distinguish the various histological subtypes of ALD, since patients with cirrhosis can be asymptomatic, while patients with hepatocellular failure may have only severe fatty change.235
Laboratory Investigations Biochemical and hematological tests can confirm the presence of alcohol abuse and indicate the presence of liver damage, but are not useful in determining the severity of the histological lesion. Blood alcohol estimations are an often-underused method of confirming a suspicion of excess drinking, with levels >100 mg/100 ml at a morning clinic or levels >150 mg/100 ml without obvious intoxication strongly suggestive of alcohol abuse. Elevation of g-glutamyl-
transferase has been reported in up to 90% of patients abusing alcohol.236 The rise is mainly due to hepatic microsomal induction and is independent of the presence of liver disease, although hepatocellular necrosis and cholestasis may contribute. It is not specific for alcohol abuse and is raised in other forms of liver injury and in patients taking other enzyme-inducing drugs.237 Its main clinical use is probably in monitoring a period of supposed abstinence, since it falls within a week of cessation of drinking. Other biochemical markers of alcohol abuse rather than liver disease include elevated serum uric acid,238 hypertriglyceridemia, and desialylated transferrin.239 The classical hematological marker of alcohol abuse is a raised mean corpuscular volume, which has been reported to occur in between 80% and 100% of alcoholics with and without liver disease240 and may be more common in alcoholic women. It is due to a direct toxic effect of alcohol on the marrow, although nutritional folate and vitamin B12 deficiencies may contribute in some patients. With regard to biochemical markers of alcohol-related liver damage, a rise in serum aspartate transaminase (AST) activity of up to five times normal is common in patients abusing alcohol and reflects the presence, but not the severity, of liver damage.241 However, unlike non-ALD, alanine transaminase (ALT) activity is raised less often than AST, and the AST/ALT ratio has been suggested as a means of distinguishing liver disease of alcoholic and nonalcoholic etiology.242 Recently, however, it has been appreciated that an AST greater than the ALT can also be a marker of severe nonALD.243 Biochemical markers of the stage of liver disease have so far proved elusive. Possible exceptions include plasma IgA, which is twice normal in less than 30% of alcoholics with early disease and greater than three times normal in 60% of patients with cirrhosis,244 and the procollagen peptides. Levels of procollagen III in particular have been shown to distinguish advanced from early ALD.245
Liver Biopsy Liver biopsy is a mandatory investigation in all patients chronically abusing alcohol who have hepatomegaly and/or abnormal liver blood tests. First, it is used to establish the diagnosis of alcohol-related liver disease. This is important since it has been shown that up to 20% of liver disease in alcoholics with abnormal liver function is due to an alternative etiology.234 Second, it is the only way of accurately staging the disease, which cannot be achieved by any combination of clinical or laboratory data246 Without knowledge of the histological severity, no prognostic information can be given to the patient and no rational treatment plan can be devised.
ALCOHOL-INDUCED FATTY LIVER Pathology Fatty liver is the earliest lesion seen in ALD. The classical appearance is of a single large fat droplet displacing the nucleus occurring predominantly in perivenular hepatocytes (macrovesicular steatosis). Very rarely, the steatosis is panacinar and may be associated with severe cholestasis, cholangiolitis, and clinical presentation with hepatic failure.235 Inflammation is rare in simple fatty liver although occasional lipogranulomata may be seen as a response to the extrusion of cellular lipid. Mild fibrosis may occur in response to lipogranulomata and is usually considered reversible; however, the presence of marked perivenular fibrosis in an otherwise uncompli-
597
Section IV. Toxin Mediated Liver Injury
cated fatty liver may be a marker of high risk of progression to cirrhosis.247 Microvesicular steatosis, in the form of finely dispersed lipid droplets, may also occur in some patients (alcoholic foamy degeneration) and is associated with bilirubinostasis and focal liver necrosis.248 This lesion resolves with abstention. Alcoholic fatty liver is histologically indistinguishable from non-alcoholic fatty liver associated with the metabolic syndrome (hyperlipidemia, hypertension, type II diabetes, and obesity).
Clinical Features Patients with fatty liver are usually asymptomatic or present with non-specific digestive symptoms. Rarely, fatty liver may be associated with hyperlipidemia, hemolytic anemia, and jaundice (Zieve’s syndrome, discussed below) or hepatic failure. Smooth non-tender hepatomegaly is usually the only clinical finding, although signs of portal hypertension may be observed if perivenular fibrosis (central hyaline sclerosis) is present. All, or none, of the laboratory investigations discussed may be abnormal; most commonly, the g-glutamyltransferase, AST, and mean corpuscular volume are mildly raised.
Prognosis It is widely considered that fatty liver is an entirely benign lesion reversible with abstention from alcohol. Fat starts to accumulate in the liver after as little as one weekend of heavy drinking in nonalcoholic human volunteers.249 Fortunately the reverse seems to hold true, and in the majority of patients with fatty liver who stop drinking the laboratory abnormalities quickly return to normal250 and the histological abnormality rapidly regresses.251 Accordingly, no treatment options have been evaluated in patients with fatty liver other than abstention and a well-balanced diet. However, there are reports that alcoholic fatty liver per se is not always benign, with occasional mortality due to hepatic failure, fat emboli, and hypoglycemia. Furthermore, fatty liver may be a precursor of alcoholic cirrhosis. In a study by Sorensen and colleagues,61 it was found that the extent of fatty liver on initial liver biopsy was a better predictor of subsequent progression to cirrhosis 10 years later than alcohol history. Furthermore, a more recent study revealed that even “pure” fatty liver can progress to fibrosis and cirrhosis in a proportion of patients.62 In this study, the presence of mixed macro- and microvesicular fat and giant mitochondria were associated with disease progression. This suggests, as discussed previously, that fatty liver may be causative in the development of cirrhosis rather than simply an epiphenomenon of alcohol abuse.
ALCOHOLIC HEPATITIS Pathology Alcoholic hepatitis consists of a constellation of histological abnormalities. The features obligatory for diagnosis are:232 1. liver cell damage, typified by ballooning degeneration progressing to necrosis ± Mallory bodies. Ballooning degeneration is characterized by hepatocyte swelling, a pale granular cytoplasm, and a small hyperchromatic nucleus. Mallory bodies are intracytoplasmic inclusions staining purplish-red with hematoxylin and eosin and consisting of aggregates of inter-
598
mediate filament proteins, reflecting impaired function of the microtubular system 2. inflammatory cell infiltrate, predominantly neutrophils. These are typically arranged round necrotic hepatocytes that contain Mallory bodies (“satellitosis”) 3. pericellular fibrosis, producing a “chicken-wire” appearance. In addition, there is often fibrous thickening around the hepatic vein radicals and eventual obliteration of the veins, a process referred to as central hyaline sclerosis 4. perivenular distribution, unless cirrhosis is present, when lesions occur at the periphery of nodules. As the severity increases the damage extends to involve the whole lobule Other features, which are often present but are not obligatory for diagnosis, include: fatty change, bridging necrosis, bile duct proliferation, apoptotic bodies, cholestasis, and giant mitochondria. Histological features considered to indicate a high risk of progression to cirrhosis are the extent and degree of fibrosis (central hyaline sclerosis is the worst sign), a panlobular distribution, and widespread Mallory body formation. It is important to highlight that this pattern of lesions can also occur in other conditions, including: diabetes mellitus, obesity, jejunal-ileal bypass, total parenteral nutrition, and following treatment with various drugs, when it is referred to as non-alcoholic steatohepatitis.
Clinical features There is no good correlation between the severity of the histological lesion and the clinical presentation, which can range from asymptomatic to life-threatening hepatic decompensation.252 However, patients with the milder histology are more likely to present with non-specific symptoms, incidental hepatomegaly, or raised transaminases, while patients with severe histology usually present with symptoms specifically related to hepatocellular failure, such as jaundice, ascites, and encephalopathy or variceal bleeding. The episode of decompensation leading to clinical presentation may be precipitated by vomiting, diarrhea, anorexia, increased alcohol intake, or intercurrent infection. The majority of patients have tender, smooth hepatomegaly, with an arterial bruit in severe cases. Signs of chronic liver disease may be present, even without coexisting cirrhosis, and the more advanced cases may also have signs of portal hypertension and encephalopathy. Non-liver signs commonly present include pyrexia, signs of associated vitamin deficiency and malnutrition, a hyperdynamic circulation, and cyanosis due to intrapulmonary arteriovenous shunting. Abnormalities of liver-related blood tests are always present and include decreased albumin and increased gglutamyltransferase, AST, bilirubin, alkaline phosphatase, and prothrombin time (PT). In addition, blood urea and serum sodium and potassium are all low, unless hepatorenal syndrome (HRS) supervenes, and hypoglycemia may be present. Macrocytic anemia, neutrophil leukocytosis, and thrombocytopenia are present in all but the mildest cases. A peculiar clinical feature of patients with severe alcoholic hepatitis is that they often rapidly deteriorate in the days immediately following hospital admission.253 This has been observed in up to 40% of patients and varies from deteriorating blood tests to increasing encephalopathy or variceal bleeding. The pathophysiological basis of this is not clear but suggestions have included the nutritional implications of withdrawing an alcoholic from his/her
Chapter 29 ALCOHOLIC LIVER DISEASE Figure 29-9. Alcohol-induced hepatic fibrosis.
principal source of calorific intake and a reduction in hepatic blood flow consequent upon a reduction in levels of acetaldehyde, which, via conversion to adenosine, has vasodilatory actions.
Prognosis The short-term outcome in patients with alcoholic hepatitis depends largely on the severity of the initial histological lesion. Thus in the Veterans Administration Cooperative study 30-day mortality was 1% in those with mild alcoholic hepatitis, 12% in those with moderately severe histology, and 34% in patients with severe disease.254 One of the management problems is that many patients will have prolongation of the PT that precludes transabdominal liver biopsy. It is these patients who are likely to have severe disease and in whom
aggressive and experimental therapy might be justified. While transjugular liver biopsy is an alternative, it is not always locally available. In view of this difficulty, many clinical and laboratory variables have been suggested as indicators of histological severity and therefore of potential use in predicting short-term mortality in alcoholic hepatitis. Based on these variables there have been three main attempts at creating prognostic indices. First is the modified Child’s criteria which combines the presence of encephalopathy and ascites with serum albumin, bilirubin, and PT.255 Second is a more complex system combining 12 different variables to derive a combined clinical and laboratory index,256 and third is the discriminant function (DF) of Maddrey and colleagues, which is based on PT and bilirubin only.257 This has been confirmed as a useful predictor of mor-
599
Section IV. Toxin Mediated Liver Injury
tality prospectively and, in view of its simplicity, is probably the most clinically useful index at present. Perhaps surprisingly, none of these indices includes the presence of renal failure, which is not uncommon in the most severely ill patients. This may reflect the fact that the occurrence and outcome of the HRS are entirely dependent on the severity of the hepatocellular dysfunction, which is better indicated by other clinical and laboratory variables. Alternatively, renal function may be an important prognostic indicator that has not been utilized in the indices to date. A fourth scoring system, the Glasgow Alcoholic Hepatitis Score, now incorporates this parameter. While this scoring system needs validation in large numbers prospectively, initial work suggests that it is more sensitive and specific than the Maddrey score at determining prognosis.258 If the patient survives to hospital discharge, then the long-term prognosis is determined by the initial histology, the progression to cirrhosis, and the subsequent drinking behavior. Thus, the 5-year survival falls from 70% to 50% in patients with severe compared to mild alcoholic hepatitis,259 and in the Veterans Administration study, patients with mild hepatitis who developed cirrhosis had a 71% 2year survival compared to 81% in those who did not.260 In addition, the 7-year survival has been reported to fall from 80% to 50% in patients who continue to drink compared with abstainers,259 which is presumably due, at least in part, to the influence of intake on the risk of progression to cirrhosis. In men with mild histology, drinking behavior is the major factor determining progression to cirrhosis, while in women and men with severe histology, progression can occur independently of drinking behavior.261
CIRRHOSIS Pathology With progressive injury the features of cirrhosis, namely fibrous septa linking hepatic and portal veins, and regenerative nodules eventually appear. The cirrhosis is usually micronodular, possibly reflecting the inhibition of regenerative activity by alcohol, and frequently reverts to a macronodular cirrhosis with abstention.262 The coexistence of steatosis and hepatitis is common and usually indicates continued consumption. In contrast, alcohol withdrawal at the cirrhotic stage can make the histological determination of etiology almost impossible.
Prognosis The survival of patients with alcoholic cirrhosis is determined by the clinical and histological severity of the disease at presentation and their subsequent drinking behavior. It has also been shown in some studies that gender263 and ethnicity264 may influence survival. Several studies have shown that patients who present with decompensated disease do significantly worse than those presenting with compensated disease.265,266 The influence of drinking behavior on this trend is best illustrated by the seminal study of Powell and Klatskin.267 They showed that in patients with compensated disease, continued drinking reduced the 5-year survival from 89% to 68%. Abstaining patients with ascites or jaundice had lower survival rates than compensated patients, but higher survival rates than patients with ascites or jaundice who continued to drink. The lowest survival was seen in patients with variceal bleeding and alcohol habits had no effect on their mortality. The presence of coexisting alcoholic hepatitis on initial biopsy also adversely affects prognosis.268 HCC develops in 5–15% of patients with alcoholic cirrhosis.269 It is most common in abstaining men and the majority of patients die within a few months of diagnosis.266,270
ASSOCIATED CONDITIONS AND EXTRAHEPATIC MANIFESTATIONS INTRODUCTION The range of health problems associated with excess alcohol consumption extends beyond the liver, with virtually every system in the body affected (Figure 29-10). As with ALD, the pathogenetic pathways are not always clear, but are more complex than the direct effect of ethanol per se. This section will give an overview of these problems.
GASTROINTESTINAL EFFECTS As the first site of exposure after ethanol ingestion, the gastrointestinal system is a prime candidate for toxicity. As well as the liver, alcohol can affect most parts of the gastrointestinal system, as summarized in Table 29-3 and discussed below.
Salivary Glands and Oropharynx Clinical Features As with other forms of cirrhosis the clinical presentation of alcoholic cirrhosis can range from asymptomatic hepatomegaly to hepatic failure and the complications of portal hypertension such as ascites or variceal bleeding. Presentation with severe hepatic decompensation usually implies the presence of continued drinking and superimposed alcoholic hepatitis, but may signal the development of HCC or portal vein thrombosis. The clinical findings will depend on the presence of portal hypertension or encephalopathy and do not differ significantly from those observed in other forms of cirrhosis. Patients with compensated cirrhosis, particularly if abstinent from alcohol, can have completely normal laboratory investigations, while patients with continued intake will have a similar range of abnormal laboratory investigations to those seen in patients with alcoholic hepatitis. In addition, a raised a-fetoprotein suggests the presence of HCC and indicates the need for further investigations.
600
Parotid enlargement is frequently observed in heavy drinkers with and without chronic liver disease. A histological study at necropsy demonstrated an increase in adipose tissue at the expense of acinar tissue in the salivary glands of patients with alcoholic cirrhosis compared to controls.271 It may be this which contributes to the reduction in both basal and stimulated parotid gland salivary flow reported in these patients.272 Whether the reduced secretion and altered gland structure in patients with alcoholic cirrhosis is primarily associated with the liver disease or the effects of prolonged alcohol consumption per se is not entirely clear; however, reports of increased resting salivary flow in alcoholics without liver disease would suggest that the development of liver disease is the important factor leading to reduced secretion. The prevalence of glossitis and stomatitis is higher in alcoholics than controls,273 presumably reflecting their poor nutritional status, which includes deficiencies in B-vitamins and iron. In addition,
Chapter 29 ALCOHOLIC LIVER DISEASE Figure 29-10. Extrahepatic complications of excess alcohol consumption.
Wernike-Korsakoff sydrome Cerebellar disease Brain stem disease Cerebral atrophy Parotid enlargement Oropharyngeal tumours
Arrhythmias, hypertension cardiomyopathy, ischemic heart disease Gastritis Acute/chronic pancreatitis Malabsorption Hematological complications Peripheral neuropathy myopathy
Diarrhoea Sexual dysfunction reduced fertility
Osteoporosis
heavy drinkers have a very significant increase in the incidence of oropharyngeal tumors. Tobacco and alcohol are the principal etiological factors associated with the development of head and neck malignancies and appear to act in synergy. One study has reported a history of alcohol and tobacco use in more than 75% of patients with tumors of the oropharynx.274 As these tumors are more common in Asians carrying the null ALDH2 gene, it has been suggested that acetaldehyde, which has been shown to accumulate in the saliva of these individuals, plays a role in the pathogenesis.275 A history of alcoholism or alcohol-related disease is also associated with a worse prognosis in patients with head and neck malignancy.
Esophagus Acute alcohol ingestion alters esophageal motor function by reducing lower esophageal sphincter pressure, and inhibiting the primary peristaltic movement of the distal esophageal body.276 This reduces esophageal clearance and increases gastroesophageal reflux. Chronic ethanol consumption also results in reduced esophageal clearance, but lower sphincter pressures are increased unless the patient has concomitant autonomic neuropathy. While ethanol is associated with heartburn, there is no good evidence that drinkers are more prone to esophagitis. Nausea and vomiting are frequent in chronic alcoholics and may induce Mallory–Weiss tears.277 Esophageal cancer is the sixth commonest cancer in the world, and alcohol has been identified as a major risk factor since 1962. This association is dose-related, and there is no dose below which there is no increased risk. Smoking is an important cofactor. Alcohol-associated nutritional deficiencies and the enhanced bioactivation of dietary mycotoxins and nitrosamines and tobaccorelated carcinogens may be important cofactors.278
Stomach The effects of ethanol on gastric motility have been inconclusive, but tend towards an inhibitory effect on gastric emptying. Gastric acid secretion is greatly increased by beer and less so by wine. Most spirits do not lead to an increase in gastric acid secretion, leading to the suggestion that it is other products of fermentation that have this effect.279 Acute alcohol consumption causes an acute erosive hemorrhagic form of gastritis, with loss of surface epithelial cells and neutrophil infiltration. This peaks at 60 min and lasts at least 24 h. Both non-steroidal anti-inflammatory drugs and portal hypertensive gastropathy are risk factors. Chronic heavy drinkers are more likely to develop a superficial or atrophic type of gastritis. Whether alcohol abuse per se induces this classical chronic gastritis with a mononuclear cell infiltrate and glandular atrophy is unclear. This may be due to the increased incidence of Helicobacter pylori infection in the gastrointestinal tract of heavy drinkers.280 While there is some debate over whether chronic ethanol consumption increases the risk of duodenal ulcers,281 there is no evidence that the incidence of gastric ulceration is higher than in the general population.
Small Intestine Alcohol is one of the main causes of malnutrition in the western world. It can be severe and is associated with neurological problems, skin abnormalities, and glossitis, and may also contribute to increased susceptibility to infection and malignancy. Malnutrition in an alcoholic can be both primary, due to inadequate nutrient intake, and secondary due to malabsorption or maldigestion resulting from gastrointestinal complications. While pancreatic and hepatic dysfunction can play a role, particularly in fat malabsorption, the most
601
Section IV. Toxin Mediated Liver Injury
Table 29-3. Important extrahepatic gastrointestinal conditions associated with excess alcohol consumption Salivary glands and oropharynx
Esophagus
Small bowel
Colon
Pancreas
Parotid enlargement Glossitis/stomatitis Oropharyngeal malignancy
Gastroesophageal reflux disease Mallory–Weiss tears Esophageal malignancy
High-transit diarrhea Malabsorption
High-transit diarrhea
Acute and chronic pancreatitis Exocrine and endocrine pancreatic insufficiency
important cause of malabsorption is probably altered small-bowel function. Several factors contribute to alcohol-related intestinal dysfunction, including the effects of alcohol on gut motility, cellular structure, and function and blood flow. Significant increases in motility have been reported, the most obvious clinical manifestation of which is reduced transit time and diarrhea.282 Cellular changes observed in the jejunoileal epithelium of alcoholics include abnormal mitochondria, dilated ER, altered membrane fluidity, and focal cytoplasmic degradation.283 These changes are manifest macroscopically by a decrease in villous height,284 biochemically by a decrease in the activity of mucosal disaccharidases,285 and functionally by an increased permeability to water and solutes.286 Intraluminal ethanol also causes regional changes in blood flow within the jejunal mucosa.287 Together, these various effects of alcohol intake impair the absorption of a variety of nutrients and minerals, including glucose, amino acids, trace elements, and vitamins such as thiamine, B12, B6, and folic acid. This intestinal malabsorption can lead to overall weight loss and multiple deficiencies of micronutrients. The role of oxidative stress in the pathogenesis of many alcohol-related diseases highlights the importance of micronutrient deficiencies in antioxidant vitamins and trace elements such as zinc, manganese, and selenium.288
Colon Alcohol has been shown to have direct effects on colonic motility, with alcohol increasing propulsive activity and contributing to alcohol-induced diarrhea. This effect can be observed following alcohol withdrawal when colorectal transit time increases significantly from approximately 25 h to 33 h.289 There is no confirmed impact of alcohol on the incidence of colorectal cancer.
Pancreas Alcohol can cause a chronic, recurrent, calcifying pancreatitis, typically after a period of at least 6 years of heavy consumption. There is also an established association between excessive alcohol intake and acute pancreatitis. In practice, the first clinical episode of acute pancreatitis will occur after the histological changes of chronic pancreatitis have been well established. With time, attacks often become less severe as the features of pancreatic insufficiency set in. The precise mechanisms of alcohol-related pancreatic damage are unclear, though alcohol per se does not seem to be directly toxic.290 As in alcohol-induced liver disease, oxidative stress may play a role.291 Oxidative injury can lead to a block in exocytosis, leading to the shunting of secretions into the interstitium. The resulting inflammatory response leads initially to acute pancreatitis and, if the insult (excess alcohol intake) persists, eventually to chronic pancreatitis as the acini dedifferentiates into tubular structures, losing their secre-
602
tory capacity and fibrosing. The fibrosis is particularly prominent in areas of fat necrosis,292 presumably reflecting the direct fibrogenic effect of lipid peroxidation products. If alcohol abuse continues after the first episode of pancreatitis, the majority of patients will suffer from recurrent attacks of pain, occurring at intervals of weeks or months. Eventually, with progressive loss of acinar tissue, patients develop clinical features of chronic pancreatitis. These include diabetes mellitus, reflecting impaired endocrine function, and malabsorption associated with steatorrhea due to impaired exocrine function. It remains controversial whether or not alcohol abuse is a risk factor for pancreatic cancer.
CARDIOVASCULAR SYSTEM Acute and chronic alcohol ingestion leads to a variety of beneficial as well as deleterious effects on the heart and cardiovascular system. There now seems little doubt that moderate alcohol intake is associated with a decreased risk of ischemic heart disease293 while excessive alcohol intake can lead to hypertension and disordered cardiac rhythm, including sudden cardiac death, cardiomyopathy, and cerebrovascular accidents. This dual effect of alcohol on the cardiovascular system is largely responsible for the well-known U- or J-shaped curve describing the relationship between alcohol intake and total as well as cardiovascular-related mortality.294 This shows that mortality amongst light (1–9 drinks per week) and moderate (10–34 drinks per week) drinkers is lower than in abstainers and heavy drinkers. The left-hand part of the curve is due to an inverse relationship between death from coronary artery disease and alcohol intake, while the right-hand portion is attributable to a greater risk of non-ischemic cardiovascular and non-cardiovascular deaths (accidents, suicide, cancer, liver disease) in heavy drinkers. Importantly, contrary to popular belief, almost 50% of the excess deaths occurring in heavy drinkers are attributable to circulatory diseases rather than to liver disease.295
Hypertension A number of epidemiological studies, controlling for variables such as diet and smoking, have established a dose–response relationship between blood pressure and alcohol consumption.296–298 It has been estimated that 30% of all cases of hypertension may be attributable to alcohol with females apparently less susceptible.299 The threshold for alcohol-associated hypertension appears to be around three standard drinks per day, with some studies showing a dose–response relationship with higher levels of intake.300 Findings from short-term studies have suggested that cessation of alcohol consumption in hypertensive patients results in a decrease in blood pressure.299 Whether alcohol-induced hypertension remains reversible in the
Chapter 29 ALCOHOLIC LIVER DISEASE
long term is unknown. The mechanisms underlying the association between alcohol and hypertension are unclear.
Coronary Artery Disease As discussed, in recent years a number of epidemiological studies have demonstrated a negative correlation between moderate consumption of alcohol and fatal coronary artery disease.301 Case–control studies have also shown a lower incidence of myocardial infarction in moderate drinkers compared to abstainers.302 In these studies “moderate” drinking was no more than two drinks per day in men and one drink per day in women. Supportive evidence for a protective effect of alcohol on ischemic heart disease is provided by its biological plausibility.303 Moderate alcohol consumption increases the plasma levels of the protective high-density lipoprotein cholesterol by as much as 33%.304 The mechanism is likely to be a result of altered hepatic synthesis and secretion of lipoproteins. Alcohol intake is also associated with impaired platelet aggregation305,306 and lower levels of fibrinogen, thereby reducing the risk of thrombo-occlusive events.
Cerebrovascular Disease All types of strokes have been associated with alcohol consumption. This is perhaps not surprising in view of the association between alcohol and most of the established stroke risk factors, including hypertension, cardiomyopathy, arrhythmias, diabetes, and cigarette smoking.307 In view of its negative association with coronary heart disease, it might be expected that moderate consumption would be associated with a reduced risk of ischemic stroke. The consumption of one drink per day has been associated with a reduced risk of ischemic stroke in one study308 but this has not been confirmed in other similar studies, with some reporting a positive association between heavy alcohol intake and cerebral infarction in young men following alcohol “binges.”309 This may be attributed either to dehydration or to the occurrence of alcohol-related supraventricular arrhythmias with resulting embolic events. The expected positive association with hemorrhagic strokes has been reported310 but it remains unclear whether this association is independent of alcohol’s effect on other risk factors, particularly hypertension.
Cardiomyopathy It has been recognized since the early 1960s that long-term, heavy alcohol consumption is the main cause of a non-ischemic, dilated cardiomyopathy. Postmortem and endomyocardial biopsy studies performed in chronic alcoholics both with and without cardiac symptoms have shown dilation of the atria and ventricles, increased myocardial mass, interstitial fibrosis, and small-vessel coronary artery disease.311 While subclinical alcoholic cardiomyopathy, characterized by left ventricular hypertrophy and mild systolic and diastolic dysfunction, appears to be relatively common in heavy drinkers, clinical presentation is relatively uncommon and appears to require at least 5 years of > 90 g ethanol per day. Interestingly, the duration of drinking seems to be more important than the dose over this threshold, with the mean duration in symptomatic patients being 25 years of heavy drinking compared to 16 years in asymptomatic patients.312 The onset is usually insidious, with non-specific fatigue and chest pain associated with palpitations, most commonly
due to atrial fibrillation. As the disease progresses, features of biventricular failure develop. With continued drinking, death from cardiac failure or arrhythmias usually occurs within 4 years of presentation, although in the early stages of disease dramatic recovery can occur with abstention.313
Arrhythmias and Sudden Cardiac Death Heavy drinking increases the risk of cardiac arrhythmias whether or not heart disease is present. This evidence has come from clinical observations, retrospective case–control studies, controlled studies of consecutive admissions for supraventricular tachyarrhythmias, and prospective epidemiological studies.314 The association is best established for atrial fibrillation, although in one study individuals drinking more than six drinks per day had a higher risk of all supraventricular tachyarrhythmias than those drinking less than one drink per day when matched for age, sex, and smoking.315 The tendency of these arrhythmias to present following weekend or holiday “binges” has led to the term “holiday heart syndrome.” Alcohol has also been shown to promote the onset of ventricular tachyarrhythmias316 and this presumably explains the increased incidence of sudden cardiac death observed in heavy compared to occasional or light drinkers.317 The mechanism of alcohol-related arrhythmogenesis is almost certainly multifactorial. Factors that may play a role include subclinical cardiomyopathy producing conduction delays, potassium and magnesium depletion, the hyperadrenergic state accompanying alcohol withdrawal, autonomic neuropathy, and a direct effect of ethanol on cardiac conduction.316 The mechanism of ventricular tachyarrhythmias is most likely early after-depolarizations provoked by catecholamine release and potassium depletion during withdrawal in the presence of a prolonged action potential due to the autonomic neuropathy. In support, patients with a prolonged action potential, manifest on the surface electrocardiogram as QT interval prolongation, have been shown to be at risk of sudden cardiac death.318
EFFECTS ON THE NERVOUS SYSTEM Acute and chronic alcohol intake is associated with a wide range of effects on the nervous system. The depressant effect of alcohol means that acute heavy consumption can lead to blackouts and even coma. After a sudden reduction in alcohol consumption, tremulousness and agitation are common, while the full-blown syndrome of delirium tremens, including hallucinations and seizures, is less frequently seen, and more serious. Alcohol and its metabolite acetaldehyde are almost certainly directly neurotoxic, but associated nutritional deficiencies undoubtedly contribute to the pathogenesis of some, if not all, alcohol-related neurological diseases.319
The Wernicke–Korsakoff Syndrome The Wernicke–Korsakoff syndrome is a nutritional disorder caused by thiamine deficiency, and is predominantly observed in alcoholics. Wernicke’s encephalopathy represents its acute phase, while Korsakoff ’s psychosis represents the chronic continuation of the disease. The major pathologic changes of this syndrome are predominantly in the paraventricular parts of the thalamus and hypothalamus, the mammillary bodies, the periaqueductal gray matter,
603
Section IV. Toxin Mediated Liver Injury
and the floor of the fourth ventricle. An abrupt onset and the triad of oculomotor disturbances, cerebellar ataxia, and mental confusion characterize classical Wernicke’s encephalopathy. The most common ocular abnormality is nystagmus (vertical or horizontal), but bilateral sixth-nerve palsy, palsies of conjugate gaze, and complete ophthalmoplegia are also seen. Ptosis and pupillary abnormalities may also occur. Mental inattention is characterized by disorientation, inattention, and unresponsiveness, which progresses to coma if untreated. Treatment is with parenteral thiamine. The disease can be aggravated by giving intravenous dextrose before thiamine supplementation is administered. Patients either recover within 48– 72 h or progress to Korsakoff ’s psychosis. Korsakoff ’s psychosis is characterized by various degrees of both retrograde and anterograde amnesia, with relative preservation of other intellectual functions. The Korsakoff state is potentially reversible by early intervention with thiamine and prompt treatment of Wernicke’s encephalopathy. Unfortunately, recovery is incomplete in more than 50% of cases and individuals may be left with devastating chronic memory deficits.
Bordeaux, France, revealed that consumption of between 250 and 500 ml of wine per day resulted in a significant reduction in the risk of dementia and Alzheimer’s disease in later life.321 Computed tomography (CT) studies have shown a substantially higher incidence of cerebral cortical atrophy in alcoholics, and brain weights in chronic alcoholics at autopsy are less than half that of agematched controls. However, no correlation has been demonstrated between either the CT or histological changes and the neuropsychological impairment frequently seen in chronic alcoholics. For example, ventricular and sulcal enlargements are often seen on CT in alcoholics with no clinical evidence of cerebral dysfunction. Furthermore, there is little firm evidence of any histological abnormality in the brains of alcoholics other than that related to the complications of alcoholism such as Wernicke’s encephalopathy, post-traumatic changes, and chronic hepatocerebral degeneration. Thus, cerebral atrophy is common in alcoholics, and dementia may occur as a result of the direct toxic effect of ethanol on the brain, but there is no defined clinicopathological entity that constitutes “alcoholic dementia” and the mental disturbances are more likely to be related to other established complications of alcohol abuse.
Cerebellar Disease Alcoholic cerebellar degeneration is characterized clinically by an ataxic gait and truncal ataxia, while typically the upper limbs are unaffected.320 Pathologically there is degeneration of the cerebellar cortex, predominantly of the anterior and superior vermis and anterior lobes. In most cases the syndrome evolves over a period of several weeks or months, after which it remains unchanged for years. Acute cerebellar degeneration may respond to large doses of thiamine and abstinence, but patients usually present long after the onset of their symptoms. At this stage the likelihood of improvement is small, and probably occurs as a result of an improvement in the peripheral neuropathy which is present in around half the patients.
Brainstem Disease Central pontine myelinolysis is a rare demyelinating disease characterized by neuronal dysfunction centered on the pons. It is encountered predominantly in malnourished alcoholics with disordered electrolytes. Cerebral edema associated with either severe hyponatremia or the rapid correction of hyponatraemia during electrolyte replacement may play a role in the pathogenesis. Clinical features include the subacute onset of a progressive quadriparesis, pseudobulbar palsy affecting speech and swallowing, and paralysis of horizontal eye movements. More extensive brainstem dysfunction may result in pupillary abnormalities, decerebrate posturing, altered conscious level, and respiratory paralysis. Not surprisingly, the prognosis of this condition is poor, with the diagnosis often only made at postmortem. Central pontine myelinolysis may be associated with Marchiafava–Bignani syndrome which is a rare demyelinating disease of the corpus callosum also occurring predominantly in alcoholics. This presents with acute bilateral hemispheric dysfunction and has a poor prognosis.
Neuropathies Peripheral neuropathy is another common nutritional complication in alcoholics. The precise mechanism is unclear but histology reveals a non-inflammatory degeneration of myelin sheaths and axon cylinders, which is more intense in distal segments. In advanced cases, degeneration may also be observed in the anterior and posterior roots of the spinal cord. Patients with electrophysiological evidence of peripheral neuropathy can be asymptomatic, or, more typically, present with pain and paresthesia initially affecting the lower limbs. In severe cases weakness and atrophy may be seen. With continued drinking the symptoms progress relentlessly, so that in advanced cases significant distal motor deficits with atrophy may be seen. Treatment consists of abstinence and nutritional supplementation, particularly with B-vitamins. Recovery is slow and often incomplete. An association between alcoholism and autonomic neuropathy was first reported in 1980.322 The subsequent observation that it was more common in alcoholics with liver disease than those without suggested that the liver disease rather than alcohol per se might be the primary cause.323 This hypothesis was supported by a report that the incidence of autonomic neuropathy was similar (45%) in patients with alcohol and non-alcohol-related liver disease.324 More recently, evidence for a reversible metabolic effect of liver disease on autonomic function has been provided by a study demonstrating an improvement in autonomic function 3 months after successful liver transplantation.325 Importantly, autonomic neuropathy is associated with an adverse prognosis in patients with liver disease, attributed either to an impaired response to stresses or to the associated QT interval prolongation and subsequent risk of ventricular arrhythmias.318 As many as 50% of patients with liver disease experience typical symptoms of autonomic neuropathy, including postural dizziness, abnormal sweating, and impotence.324
Alcoholic Dementia While some studies suggest that a high level of alcohol consumption may be a contributing factor in some of cases of dementia, this is an area of great controversy. One study of moderate drinkers in
604
ALCOHOLIC MYOPATHY Alcoholic myopathy can occur in an acute form with variable severity. In the mild form it may represent a mild rise in muscle enzymes,
Chapter 29 ALCOHOLIC LIVER DISEASE
while at its most severe there may be rhabdomyolysis. In the severe form the patient presents, often after a heavy bout of drinking, with muscle pain and weakness. Muscle enzymes are markedly raised and electromyography abnormal. Occasionally the rhabdomyolysis can be severe enough to cause myoglobinuria and acute renal impairment. The condition improves in most cases over days. Alcoholic myopathy also has a more common, chronic form, which presents with progressive, painless weakness and wasting of the proximal muscle groups. The condition is associated with chronic alcohol consumption, and histology reveals a loss of type IIb muscle fibers. While ethanol may partly induce myopathy through the wellcharacterized neuropathy, there appears to be an additional impact of ethanol and its metabolism on muscle per se. These mechanisms remain unclear, but are likely to involve the generation of oxidative stress, mirroring other end-organ damage.
THE FETAL ALCOHOL SYNDROME A conservative estimate for the incidence of fetal alcohol syndrome has been put at 0.33 per 1000 live births, with many more children suffering from various alcohol-related effects not amounting to the full syndrome.326 A similar per capita frequency is likely to occur in other industrialized countries, but few data are available on the magnitude of this problem in the developing or third world. Fetal alcohol syndrome is caused by excessive alcohol consumption during pregnancy which results in a variety of abnormalities in the fetus, thought to be due to a direct effect of alcohol and its metabolite acetaldehyde, rather than to associated nutritional deficiencies or other drugs. The severity of the syndrome depends on both the timing and severity of maternal alcohol consumption during gestation. The diagnostic criteria include features of growth retardation and developmental delay, central nervous system involvement, and characteristic facial dysmorphology in the presence of a maternal alcohol consumption of more than two drinks per day. The central nervous system involvement typically presents as behavioral dysfunction and mental retardation. The characteristic facial features include short palpebral fissures, an elongated mid-face, an indistinct philtrum, a thin vermilion, and a foreshortened maxilla.
ALCOHOL AND CANCER Results from several large epidemiological studies have firmly established that alcohol is associated with a higher cancer incidence and mortality.327 Alcohol consumption is most strongly associated with cancers of the esophagus (as discussed above), oropharynx and larynx, with the increased risk particularly prominent in smokers (Table 29-4). The controversy over the association between alcohol Table 29-4. Cancers confirmed to be associated with excess alcohol consumption Alcohol and cancer Mouth Pharynx Larynx Esophagus Breast Liver
and breast cancer has been resolved by two large meta-analyses. The first was a meta-analysis of six prospective cohort studies.328 This has clearly demonstrated that, for intakes less than 60 g/day, breast cancer risk increases linearly with intake. A daily intake of 30–60 g was associated with a relative risk of 1.41 (1.18–1.69) when compared to non-drinkers and this risk was independent of other known risk factors. The second, more recent study had very similar findings.329 The mechanisms underlying alcohol-related cancers are unclear but several factors have been suggested to play a role. Alcohol may be important in the initiation of cancer, either by increasing the expression of certain oncogenes, or by impairing the cell’s ability to repair DNA, thereby increasing the likelihood that oncogenic mutations will occur. Alcohol may act as a co-carcinogen by enhancing the effect of direct carcinogens such as those found in tobacco and the diet. This effect of alcohol may be, at least in part, via induction of the cytochrome P450 family of enzymes that are found in the liver, lung, and intestine and are capable of metabolizing various tobacco and dietary constituents into cancer-promoting free radicals. Since reduced levels of iron, zinc, and vitamins A, B, and E have been experimentally associated with some cancers, the nutritional deficiencies associated with chronic alcohol intake may also play a role in alcohol-related cancers, possibly by increasing the magnitude of free radical-related oxidative stress. Finally, alcoholism is associated with immunosuppresion, which makes chronic alcoholics more susceptible to infection and theoretically reduces immune surveillance of early tumors.
HEMATOLOGICAL COMPLICATIONS Heavy alcohol consumption, with or without liver disease, can have profound effects on the hematological system. While the earliest and most obvious effects are on erythrocytes, derangements in production, function, and consumption of leukocytes, platelets, and coagulation can also have important consequences.
Erythrocytes It has long been recognized that many alcoholics have increased mean corpuscular volume. This simple macrocytosis can occur in the absence of vitamin deficiency, and is thought to be a direct effect of alcohol, or the products of its metabolism, on the development of the red cell. In line with this, the corpuscular volume returns to normal several weeks after abstinence. In true folate deficiency, there may also be a macrocytic, megaloblastic anemia. This is quite common in alcoholics, who may also suffer from vitamin B12 deficiency, which can cause a similar picture. Alcohol can also promote a sideroblastic anemia in which heme synthesis is impaired. In this condition serum ferritin is raised and the red cells are hypochromic in the peripheral blood, and have ring sideroblasts in the marrow. Heavy drinkers also suffer from the anemias associated with some degree of alcohol-induced liver injury. Zieve’s syndrome is a condition described in 1958,330 typically found in middle-aged male heavy drinkers with alcoholic fatty liver and severe hyperlipidemia. It is rare, and improves with abstinence. This is in contrast to the more concerning spur-cell anemia, which tends to be associated with advanced alcoholic cirrhosis, though it can occasionally occur in
605
Section IV. Toxin Mediated Liver Injury
other types of cirrhosis. Spur cells, or acanthocytes, are caused by the equilibration of the outer lipid layer of the cell membrane with the cholesterol-rich abnormal lipoproteins in the plasma. The problem is compounded by the reduced fluidity of the red-cell membrane seen in cirrhosis due to a reduction in the proportion of polyunsaturated versus saturated fatty acids. The acute development of spur-cell anemia is a poor prognostic indicator in alcoholic cirrhosis, and has been considered an indication for transplantation, which may be curative. Finally, red-cell consumption can occur as a result of congestive hypersplenism resulting from liver disease and portal hypertension.
Leukocytes Chronic alcoholics are more prone to a variety of infections331 and malignancies.332 This sensitivity has often been attributed to defects in innate and acquired immunity due to ALD and ethanol consumption per se. While the effects of ethanol on neutrophil333,334 and macrophage function128,335,336 have been well investigated, revealing in particular a disruption in phagocytosis337 and antigenprocessing and presentation,338 studies into the effects on lymphocyte functioning have been less conclusive. A uniform finding, however, in both chronic alcoholics and chronically ethanol-fed animals is a reduction in circulating T-cell numbers, and, in mice, a reduction in spleen and lymph node size.339,340 Whether this reduction in numbers is primarily due to a failure of proliferation or an increase in apoptotic rates is, however, unclear. It appears therefore that chronic, or even acute, ethanol consumption may alter the host’s ability to mount an appropriate-magnitude immune response.341
Platelets While an acute single dose of ethanol may not affect platelet number or function, chronic heavy drinking does. The changes that occur can do so in the absence of folate deficiency or hypersplenism, although these problems can compound the condition. The thrombocytopenia seen as a consequence of alcohol ingestion appears to be a direct myelosuppressive effect of ethanol on bone marrow megakaryocytes, and is usually mild, and rarely of clinical consequence.342 In addition, ethanol can also affect platelet function, even in the absence of thrombocytopenia. Chronic heavy drinkers have been found to have prolonged bleeding times and platelets that are significantly less responsive to standard platelet aggregation tests and have decreased thromboxane A2 release. When these patients are followed up during an in-hospital period of abstinence, these abnormalities return to normal during 2–3 weeks of alcohol withdrawal.
Coagulation One of the difficult problems in patients with ALD is the derangement in coagulation which can compound acute episodes of gastrointestinal bleeding. These disturbances are common and complex. Liver synthesis of clotting factors can be impaired by hepatocellular dysfunction or inadequate absorption of vitamin K, which is required for the synthesis of factors II, VII, IX, and X. These abnormalities present with an abnormal PT. Treatment requires replacement of the factors plus vitamin K. In many cases of cirrhosis there will also be a reduced fibrinogen level.
606
EFFECTS ON THE ENDOCRINE SYSTEM The pathogenesis of decreased libido and impotence in heavy drinkers is not fully understood. There is evidence from human and rat studies that chronic alcohol consumption reduces testosterone synthesis. While alcoholics may develop the resulting hypoandrogenization, there is not the expected rise in gonadotrophins that normal accompanies this end-organ failure.343 This in turn suggests a problem with the hypothalamic and pituitary feedback mechanisms. These problems are compounded by the hyperestrogenization seen in liver disease and manifested by spider nevi and gynecomastia. Heavy alcohol consumption may also induce changes in peripheral testosterone and estrogen metabolism as well as changes in estrogen receptors. Even in patients without liver disease, alcohol can affect fertility. In men, abnormal spermatogenesis is more frequent, with decreased numbers and motility of sperm. In women, amenorrhea, anovulation, and accelerated onset of menopause have all been associated with alcohol intake. Alcohol-induced pseudo-Cushing’s syndrome has the same characteristics as classical Cushing’s, namely moon face, central obesity, muscle wasting, abdominal striae, fatigue, easy bruising, and hypertension.344 This syndrome can be indistinguishable from true Cushing’s syndrome, except for the fact that it resolves with abstinence and may recur when heavy drinking is resumed. Whether pseudo-Cushing’s is a true identity or simply a syndrome combining several of the clinical features of alcohol abuse is presently in debate. In addition to the endocrinological associations of heavy drinking described above, there are also more subtle effects resulting in a reduction in growth hormone and a rise in prolactin. The first of these has no direct impact apart from enhancing hypoglycemia, described below. The second can exacerbate the effects of hypogonadism and hyperestrogenism.
Hypoglycaemia and ketoacidosis Inhibition of hepatic gluconeogenesis, depleted hepatic glycogen stores, and deranged glucocorticoid secretion may all contribute to the presentation of the alcoholic with severe hypoglycemia. These often malnourished patients are prone to episodes of ketoacidosis which, when compounded with starvation and vomiting, can be lifethreatening.
HYPERURICAEMIA Hyperuricemia is caused by a decrease in the excretion of uric acid, secondary to hyperlactacidemia. Lactate competitively inhibits uric acid clearance by the proximal renal tubule, and consequently reduces its excretion. This situation is exacerbated by the alcoholinduced increase in urate synthesis secondary to accelerated degradation of adenine nucleotides.
OSTEOPOROSIS Even in the absence of liver disease, alcohol can cause osteopenia,345 possibly through a direct toxic effect on osteoblasts and bone remodeling. This loss of bone can result in an increased incidence of fractures in alcoholics. While the pathogenesis of this problem may involve the influence of endocrine factors, such as pseudo-
Chapter 29 ALCOHOLIC LIVER DISEASE
Cushing’s or hypogonadism, nutritional deficiencies associated with alcoholism, low levels of osteocalcin, which rapidly rises on abstinence, point to a more direct effect of alcohol itself on bone formation.346
TREATMENT The age-adjusted death rate from all-cause cirrhosis, the greatest percentage of which is alcohol-related, fell by 45.5% in the USA between 1970 and 2001. This greatly exceeded the age-adjusted allcause mortality, which fell by 28.7%. The total annual per capita ethanol consumption rose steadily from 9 liters (2 gallons) to 12.6 liters (2.8 gallons) between 1955 and 1980 and then dropped back to around 9.9 liters (2.2 gallons) in the following 20 years. In crude terms, it appears, therefore, that the improvement in alcoholic cirrhosis mortality exceeds the reduction in ethanol consumption. This may be due to the fact that there are fewer individuals drinking heavily and more drinking within healthy levels. It may also reflect improvements in the management of patients with ALD. Clearly, any improvement in the treatment of the complications of cirrhosis will have a beneficial effect on the management of patients with end-stage ALD. However, this section focuses on treatment strategies that have been specifically directed at mechanisms involved in the pathogenesis of alcohol-related liver injury and considers their current and future role in the management of patients with the various stages of ALD.
ACHIEVING ABSTINENCE Since cessation or a marked reduction in alcohol intake has been shown to improve the histology and/or survival of patients with all stages of ALD,259,261,267 measures aimed at establishing and maintaining abstinence are critical in the management of patients with ALD. This is best achieved by close liaison between liver physicians and addiction psychiatrists with support from specialist alcohol nurses and trained counselors.347 Available treatments for alcoholdependent patients can be divided into psychological and pharmacological. So-called “brief interventions” are the simplest form of psychological therapy and can be implemented by non-psychiatric staff. This involves educating and informing the patients about the nature of their problem and providing them with advice on how to go about changing their behavior. In spite of the apparent simplicity of this form of management, brief interventions have been shown significantly to increase the chances of heavy drinkers moderating their drinking at 6 and 12 months in an outpatient setting.348 With only minimal training, medical and nursing staff can also deliver a variety of manual-guided psychosocial treatments, including cognitive-behavioral therapy and motivational enhancement therapy, both of which have been shown to reduce drinking in dependent patients in a randomized controlled trial.349 As an alternative, or an addition, to psychological therapies, some patients may derive benefit from pharmacological therapy (Table 295). Both acamprosate and naltrexone have been shown to reduce drinking days and increase abstinence rates in more than one randomized controlled trial and a recent meta-analysis.350–352 Acamprosate is derived from taurine and its beneficial effect is thought to be via binding to the g-aminobutyric acid receptor with a reduc-
Table 29-5. Non-pharmacological and pharmacological therapies to obtain and maintain abstinence Non-pharmacological • Brief intervention • Cognitive therapy • Motivational enhancement therapy • Psychotherapy Pharmacological Acamprosate Naltrexone Disulfiram
• • •
tion in the neuronal excitation that is normally observed during alcohol craving. Importantly for patients with ALD, acamprosate, unlike naltrexone, is well tolerated in all but patients with Child–Pugh C cirrhosis353 and its benefit seems to persist for at least 1 year after treatment withdrawal. Disulfiram, an inhibitor of acetaldehyde dehydrogenase, has been used for many years in the management of alcohol-dependent patients. As discussed previously, it induces an acetaldehyde-mediated adverse reaction to alcohol intake characterized by nausea and flushing. Trials of effectiveness, however, have given conflicting results.354,355 The drug also requires compliance and its potential for hepatotoxicity has limited its use in patients with established ALD.356 Importantly, there have been no formal trials of either psychological or pharmacological therapies in drinkers with ALD. However, previous evidence that the severity of alcohol dependence in ALD patients is less than that observed in an unselected group of alcoholdependent patients357 suggests that these treatments may be even more beneficial in the ALD population. Consistent with the low level of dependency, up to 50% of ALD patients will either abstain completely or achieve a significant reduction in intake after being given simple advice by physicians during their initial presentation, with a significant improvement in survival compared to continued heavy drinkers.358
ALCOHOLIC HEPATITIS Alcoholic hepatitis covers a spectrum of disease from subclinical to a severe, life-threatening disorder. Independent predictors of survival in these patients are serum bilirubin, PT, and the presence of hepatic encephalopathy. As discussed, the two laboratory indices have been combined to derive a DF (bilirubin (mg/dl) + 4.6 (PT prolongation)), and a value of 32 or greater has been shown to predict a high short-term mortality in several prospective studies.140,257,359,360 Accordingly, almost all treatment trials in patients with alcoholic hepatitis have been short-term (usually 1 month) and restricted to patients with a DF > 32 and/or encephalopathy. Patients with less severe disease appear to have a good short-term prognosis even when jaundiced.260 Accordingly, in these patients and the severe patients surviving their initial presentation, treatment is focused on achieving abstinence, which has been convincingly shown to improve long-term outcome.259,261,267 Reports that some patients with alcoholic hepatitis can progress to cirrhosis even with abstention,261 and that patients with coexisting alcoholic hepatitis and cirrhosis have a worse long-term survival than patients with cirrhosis
607
Section IV. Toxin Mediated Liver Injury
only,268 suggest the need for longer-term treatment trials in patients with alcoholic hepatitis. Progress in developing specific treatments for acute alcoholic hepatitis has been hampered by a poor understanding of disease pathogenesis. Reflecting this paucity of information, many treatment modalities have been tried in patients with alcoholic hepatitis; however, none has been consistently shown to have a beneficial effect and, accordingly, none has achieved consensus status among practicing hepatologists.
Despite 13 randomized controlled trials and six meta-analyses, the debate over the use of steroids continues. It appears that they are probably beneficial in patients with severe disease; however, mortality on treatment remains high, particularly when renal impairment is present, and treatment is relatively contraindicated in the large number of patients with concomitant infection and gastrointestinal bleeding. It is because of these limitations that alternative therapeutic strategies have been sought.
Pentoxifylline Corticosteroids Of all the treatments available for patients with severe alcoholic hepatitis, corticosteroids are the most intensively studied, and probably the most effective. Steroids are aimed at suppressing or “switching off ” the hepatic inflammatory response seen in liver biopsies from patients with severe alcoholic hepatitis. The mechanism of this effect is, at least in part, through the inhibition of NFk-B transcriptional activity.361 The transcription of many inflammatory cytokines, chemokines, and adhesion molecules is dependent on the NFk-B signaling cascade.362 Two important side effects of steroids used in medium dose include poor wound-healing and susceptibility to infection, both of which can lead to life-threatening complications in this group of patients. Concern over these adverse effects coupled with a continued uncertainty over efficacy has contributed to the reluctance of many clinicians to prescribe steroids for patients with alcoholic hepatitis. Patients with alcoholic hepatitis form a heterogeneous population, both in severity and probably in disease pathogenesis. Without a liver biopsy it is difficult to differentiate a patient with severe acute inflammatory alcoholic hepatitis from one with alcohol or nonalcohol-induced cirrhosis that has decompensated while drinking. Many initial trials of steroids were poorly designed and included patients with a variety of disease severities and almost certainly patients without alcoholic hepatitis. Most of these trials showed no treatment benefit. However, two randomized controlled trials focused only on patients who had the worst prognosis, defined by a DF of ≥ 32 and/or encephalopathy, and both showed a survival benefit in the steroid-treated patients.257,360 Several meta-analyses have attempted to resolve the controversy, and although most have shown a survival benefit,363–365 this has not been a universal finding.366 Rather than performing a further conventional meta-analysis, the authors of the last three large randomized controlled trials have pooled their individual patient data, only including patients with encephalopathy and/or a DF > 32.367 This study showed that steroids improved survival versus placebo (85 versus 65%), with placebo treatment, increasing age, and creatinine independent predictors of mortality on multivariate analysis. A weakness of this study is that two of the three original trials included gastrointestinal bleeding as a contraindication, while one did not and only one trial required a liver biopsy for diagnosis. Nonetheless, the large numbers (102 on placebo, 113 on steroids) make this the most robust meta-analysis to date. The same group has now published evidence to support the withdrawal of steroids if the bilirubin has not fallen by the seventh day of steroid treatment. This simple clinical observation significantly reduces the length of treatment in nonresponders.368
608
As discussed in the section on pathogenesis, there is good evidence from animal models that TNF-a plays an important role in acute alcohol-mediated liver injury. Pentoxifylline (PTX) is a non-selective phosphodiesterase inhibitor that is approved for use in claudication at a dose of 400 mg three times a day due to its effect on red blood cell deformability. In the late 1980s, PTX was observed to have an anticytokine effect, later attributed to a reduction in TNF-a gene transcription369 and, accordingly, to reduced levels of important downstream TNF-a effectors, including other proinflammatory cytokines, chemokines, and adhesion molecules. The first randomized controlled trial of PTX in 101 patients with alcoholic hepatitis was reported in 2000.370 The effective claudication dose was given for 28 days to patients with a DF > 32 and led to a 40% reduction in mortality compared to placebo. The secondary end-point of HRS was reduced in the treated population by 65%. Importantly, almost all of the improvement in survival was due to a fall in mortality from HRS, suggesting that PTX may have a specific beneficial effect in alcoholic hepatitis patients developing this ominous complication.371 Clearly, further trials are needed to determine whether PTX should become standard treatment for patients with alcoholic hepatitis. In particular, comparisons should be made with steroids and placebo (in patients in whom steroids are contraindicated), and trials of PTX in combination with steroids should be performed.
Nutritional Supplementation Trials investigating the role of nutritional supplementation have been prompted by the degree of protein–calorie malnutrition seen in patients presenting with acute alcoholic hepatitis and the correlation between the severity of malnutrition and mortality.372 Initial trials with parenteral amino acid therapy yielded conflicting results;373,374 however, more consistent and promising results have been reported from two randomized controlled trials of enteral tube-feeding. The first compared enteral tube feeding of an energydense formula supplying 2115 kcal/day with an isocaloric standard oral diet.375 The enteral feed contained whole protein plus branchedchain amino acids, medium- and long-chain triglycerides, and maltodextrin. Thirty-five severely malnourished cirrhotics were randomized and in-hospital mortality was 12% in the tube-fed group compared to 47% in the oral group. This prompted a further study 10 years later comparing enteral feeding to steroids in 71 patients with acute severe alcoholic hepatitis. In this trial, whilst there was no difference in mortality between the groups during the 28-day treatment period, deaths occurred earlier in the steroid-treated patients and the mortality rate was lower in the enterally fed group in the year following treatment. The overall mortality rate at 1 year was 61% and 38%, in steroid- and enteral-treated groups, respec-
Chapter 29 ALCOHOLIC LIVER DISEASE
tively. While this difference did not reach significance (P = 0.26), it must be appreciated that this treatment was being compared with what is currently considered to be the best available treatment. Furthermore, the trial is ongoing,376 with a further 69 patients expected to be recruited. A further pilot study in a small group of patients has shown promising results for enteral nutrition combined with a shortened course of steroids. Further, larger studies are required to confirm this benefit. In summary, nutritional supplementation may have a role in improving medium- to long-term survival in patients with severe alcoholic hepatitis. Which patients benefit most and the mechanisms by which they derive benefit are, as yet, unclear. There is no doubt, however, that this form of treatment deserves further investigation.
Antioxidants Interest in the potential value of antioxidant therapy in the treatment of alcoholic hepatitis has arisen as a result of the growing body of evidence, discussed above, implicating oxidative stress as a key mechanism in alcohol-mediated hepatotoxicity. These considerations have recently led to three trials investigating the effect of antioxidant supplementation in patients with severe alcoholic hepatitis. In the first study, 56 patients were randomized to receive vitamin E, selenium, and zinc supplementation or placebo.377 Whilst treated patients had an in-hospital mortality of 6.5% compared with 40% in the placebo group, the entry criteria and patient details were not clear. The second trial compared steroids with an antioxidant cocktail (vitamins A, C, E, selenium, allopurinol, desferrioxamine, and N-acetylcysteine) and was stopped after an interim assessment found steroid treatment to be associated with a significantly higher survival.378 This trial did not examine whether antioxidants conferred any benefit in patients in whom steroids were contraindicated, or in combination with steroids. The most recent study investigated the role of antioxidants in patients with severe alcoholic hepatitis stratified by gender and steroid treatment. The active group received a loading dose of N-acetylcysteine of 150 mg/kg followed by 100 mg/kg per day for 1 week, and vitamins A–E, biotin, selenium, zinc, manganese, copper, magnesium, folic acid, and coenzyme Q daily for 6 months. The decision to treat with steroids was made by the supervising clinician according to conventional criteria.92 While white blood cell count and bilirubin at trial entry were both associated with increased mortality, antioxidant therapy showed no benefit either alone or in combination with steroids. In summary, on the basis of the data available thus far, high-dose antioxidant therapy confers no survival benefit in patients with severe alcoholic hepatitis.
Hepatic Mitogens The observation that survival in patients with severe alcoholic hepatitis correlates with the intensity of hepatocyte staining for proliferating cell nuclear antigen379 implies that the liver’s capacity for regeneration is an important determinant of outcome and suggests that therapy directed at enhancing proliferation might be beneficial. An infusion of insulin and glucagon has been shown to improve liver regeneration in a rat partial hepatectomy model,380 and to improve survival in a mouse model of fulminant hepatitis.381 These observations were the stimulus to several trials investigating the role of
insulin and glucagon therapy in the treatment of patients with severe alcoholic hepatitis. While the first trial showed a significant reduction in mortality in the treated group,382 two subsequent larger studies showed no benefit, with one reporting a high incidence of hypoglycemia.383,384 At present, therefore, this form of therapy cannot be recommended. Anabolic steroids have also been shown to promote hepatocyte regeneration; however, three large randomized controlled trials with either testosterone or oxandrolone in males with alcoholic hepatitis have reported no treatment benefit.385
Propylthiouracil Centrilobular hypoxia is a feature of animal models of ALD and has been postulated to play a role in the liver injury, which is characteristically most severe in the centrilobular acinar zone 3.386 The hypoxia has been attributed to the hypermetabolic state induced by ethanol which is similar to the hypermetabolic state associated with hyperthyroidism, and can be attenuated in the rodent model of ALD by the antithyroid drug, propylthiouracil.386 Two trials have evaluated the role of this drug in improving short-term mortality in patients with alcoholic hepatitis. Although the first trial reported a more rapid improvement in clinical and laboratory indices, neither trial showed any survival benefit.387,388
Experimental Therapies In addition to the treatments described above, several novel therapies for acute alcoholic hepatitis are currently undergoing investigation. The one that initially showed the most promise was the anti-TNF-a antibody, infliximab. This chimeric human/mouse monoclonal antibody binds to TNF-a and blocks its biological effects. Its potential use in alcoholic hepatitis has been suggested by its reported benefit in several other inflammatory conditions, the putative role of TNF-a in the pathogenesis of ALD, and a report that anti-TNF-a antibodies ameliorate the liver injury in a mouse model of ALD.389 Two initial reports demonstrated an improvement in biochemistry and a satisfactory safety profile when used alone390 or in combination with steroids.391 A further pilot study using a “sister” molecule, etanercept, also showed a satisfactory safety profile, though only 7 of 13 patients had a DF > 32.392 The safety aspect is important since experience with infliximab in other diseases has raised concerns over the risk of infection.393 This could potentially limit the number of patients with AH suitable for treatment and was the rationale for excluding patients with severe disease (DF > 55) from one of the initial trials.391 The beneficial role of TNF-a in promoting liver regeneration is another potential problem for antiTNF-a treatments in patients with alcoholic hepatitis, for the reasons discussed above.394 To date, one clinical trial has examined the role of infliximab in combination with steroids in patients with acute severe alcoholic hepatitis. Infliximab was used at twice the dose given for Crohn’s disease and at 0, 2, and 4 weeks. The study was stopped after 36 patients were randomized because of the high mortality in the treated group. Most of the deaths were infectionrelated, and the Maddrey scores were not found to be different in the treated group after a mean of 2 months.395 While this study may put a halt to further trials of infliximab in this group of patients, its role has not yet been fully assessed. There may be benefit at lower doses, instead of steroids, or for those patients who cannot have steroids because of bleeding. Randomized controlled studies are still
609
Section IV. Toxin Mediated Liver Injury
required to determine which patients might benefit most from treatment, the important adverse effects, and the optimal duration of therapy. A further experimental therapy that may benefit patients with alcoholic hepatitis is the molecular adsorbents recycling system (MARS). The primary aim of this treatment is to support impaired liver function while the liver recovers or the patient undergoes liver transplantation. It may, therefore, have a role in patients with alcoholic hepatitis either alone or in combination with other pharmacological therapies. The principal of the MARS procedure is to dialyze blood against an albumin solution aimed at removing albumin-bound toxins including bilirubin and bile salts. To date, only a small series of patients with alcoholic hepatitis have been treated with this procedure;396 however, clinical improvement has been reported in these cases. In addition, a rapid fall in portal pressure has been observed within hours of starting the dialysis, which may in itself may have therapeutic potential.397 Further randomized trials are awaited with interest.
Treatment of HRS in Alcoholic Hepatitis As alluded to above, in patients with severe AH, the development of renal failure is associated with a survival of less than 10% even with intensive management and renal support.371 Perhaps the most significant advance in the management of patients with advanced liver disease over the past decade has been the introduction of albumin infusions combined with splanchnic vasoconstrictor agents for patients with HRS. This combination appears significantly to improve the survival of patients with cirrhosis who have this lifethreatening complication.398–400 Although no randomized trials have specifically examined this form of therapy in patients with alcoholic hepatitis, the previously reported high mortality in alcoholic hepatitis patients with HRS suggests that it will have a significant and beneficial impact on patient survival.
ALCOHOLIC CIRRHOSIS While the high mortality of severe alcoholic hepatitis, coupled with the young age of many of the patients, makes it an important area for therapeutic trials, the vast majority of patients with ALD in clinical practice have advanced fibrosis or cirrhosis. These patients may be asymptomatic or present with symptoms related to portal hypertension, advanced liver failure, or the development of HCC. As discussed above, the most important therapy is achieving and maintaining abstinence, since this has been shown to improve survival in both well-compensated and decompensated patients.267 Unfortunately, as with alcoholic hepatitis, no adjunctive pharmacotherapies have been consistently shown to improve survival in more than one randomized controlled trial, although some have shown promise and will be reviewed below. Potential reasons for the lack of progress thus far include: (1) a lack of a clear understanding of disease pathogenesis; (2) problems with compliance in long-term treatment trials; and (3) the confounding effect of drinking behavior during the duration of the trial. As a result, at present the management of patients with advanced fibrotic ALD is directed primarily at preventing and treating the complications of portal hypertension, liver failure, and HCC, and deciding if and when to consider patients for orthotopic liver transplantation.
610
Pharmacological Therapy Propylthouracil. In contrast to its lack of effect in patients with acute alcoholic hepatitis, propylthiouracil may improve the longterm survival of patients with alcoholic cirrhosis. There has, however, been only one trial reported thus far.401 In this study, the investigators went to great lengths to assess drinking behavior and compliance by checking daily urine samples for alcohol and a drug biomarker. Treatment for 2 years improved mortality in the patient group as a whole, particularly in patients who continued to drink moderately during the trial. No improvement was seen in abstinent patients, who had an excellent prognosis on drug or placebo or in continued heavy drinkers who had a universally bad prognosis. Although the patient numbers were high,310 a large percentage of patients were either non-compliant or dropped out of the study. For this reason and the lack of any confirmatory studies, propylthiouracil has not been widely adopted by the liver community. In view of the promising results, however, it does seem surprising that no centers have attempted to repeat the study, which remains an excellent model of how to perform a randomized controlled trial in this potentially difficult group of patients. Colchicine. This anti-inflammatory drug has been evaluated in the treatment of patients with alcohol and non-alcohol-related cirrhosis because of its antifibrotic effect in vitro.402,403 To date, clinical results have been conflicting. The most convincing evidence supporting the use of colchicine comes from a study including 100 patients followed up for up to 14 years. Survival was 75% and 34% in treated and placebo groups, respectively. Some patients appeared to have a resolution of their cirrhosis to either minimal fibrosis or normal histology.404 Three further trials, however, with median follow-ups of 1,405 6,406 and 40407 months have all shown no benefit. A recent metaanalysis has also reviewed 14 randomized controlled trials and found no benefit of colchicine treatment on mortality or liver histology.408 This has been confirmed by a further large randomized study.409 Antioxidants. In addition to trials in patients with alcoholic hepatitis the accumulating evidence that oxidant stress is involved in the pathogenesis of ALD has prompted trials of antioxidants in patients with chronic disease. Two trials have evaluated the drug silymarin, which is the active component of the herb milk-thistle and has potent antioxidant properties in vitro410 and in vivo.411,412 The first trial in 170 patients with cirrhosis (92 had ALD) followed up for between 2 and 6 years reported a beneficial effect on survival.413 In contrast, a later, larger study of 200 patients with cirrhotic ALD, followed up for 5 years, showed no benefit.414 SAME, which acts as both an antioxidant by replenishing GSH and a methyl donor maintaining cell membrane fluidity, has also been evaluated in patients with alcoholic cirrhosis. Using death or liver transplantation as a combined end-point, Mato and colleagues reported a significant beneficial effect of SAME treatment in patients with Child’s A and B cirrhosis.415 Clearly further trials with this agent are awaited with interest. Phosphatidylcholine. Phosphatidylcholine is an essential component of all cell membranes and is vulnerable to attack by lipid peroxidation. Through mechanisms that are, as yet, unclear, dietary supplementation with phosphatidylcholine has been shown to attenuate ethanol-induced fibrosis in baboons.416 Potential mechanisms of
Chapter 29 ALCOHOLIC LIVER DISEASE
action include stimulation of collagenase417 and acting as a “sink” for free radicals.418 A long-term trial in patients with alcoholic cirrhosis has just been completed in the USA. While there was a trend to improvement in transaminases and bilirubin in the phosphatidylcholine group in certain patient subgroups (heavy drinkers and those with hepatitis C), overall there was no improvement in liver histology as determined by liver biopsies 24 months apart. The potential benefits of the drug may not have been evaluated appropriately because of the dramatic reduction in drinking seen in the treated and placebo groups of patients that were followed up to completion.419
Liver Transplantation for Advanced ALD Since the initial report of its success in 1988,420 ALD has become an increasingly common indication for orthotopic liver transplantation in both Europe421 and North America.422 However, transplantation for ALD remains controversial, principally due to concerns over the risk of post-transplant recidivism and its effect on outcome and public opinion at a time of increasing donor shortage. This issue, coupled with a perception that these patients are more likely to have contraindications to transplantation due either to extrahepatic complications of excessive alcohol abuse or to an associated lack of selfcare, has contributed to a continued reluctance of many centers to offer transplantation to patients with ALD. An accumulating number of reports of transplantation in patients with ALD have now provided a firm evidence base from which these issues can be addressed. Outcome of liver transplantation for ALD. Several studies have convincingly demonstrated that the survival of patients transplanted for cirrhotic ALD is comparable to patients with cirrhosis of alternative etiologies, with 5- and 10-year survivals lying somewhere between those of patients transplanted for cholestatic and viral hepatitis-related liver disease.423 Furthermore, there is no evidence that patients with ALD have a higher frequency of postoperative complications or resource utilization compared to patients transplanted for other indications, despite being transplanted at a more advanced stage of disease.424 The improvement in quality of life following transplantation also compares favorably with other indications in the short term,425,426 although not after 3 years follow-up.424 The reason for this decline is unclear, but does not seem to be related to a return to problem drinking. As for other indications, the decision to offer transplantation to a patient with ALD is based on their expected survival with and without transplantation. Without transplantation, survival depends on the severity of their liver disease and their subsequent drinking behavior. Patients with Child’s C cirrhosis have a 1-year survival of 50–85% compared to 75–95% in patients with Child’s B.427 This suggests that, in the absence of other predictors of high mortality, such as a history of spontaneous bacterial peritonitis, recurrent variceal hemorrhage, or the development of HCC, transplantation should be restricted to patients with Child’s C cirrhosis. In support of this policy, Poynard and colleagues recently demonstrated that ALD patients whose disease severity approximated to a Child–Pugh score of 11 or higher had a significantly improved 2-year survival compared to matched controls,428 while in the UK, the 1- and 2-year survival rates of 87% and 82% compare well with the 41% and 30%
survival predicted from the Beclere prognostic model.428 The potential effect of abstinence on prognosis of these patients has led most units to adopt a policy of offering transplantation to patients whose Child–Pugh score remains high after a period of abstinence. Post-transplant recidivism. Perhaps the greatest concern when considering transplantation for patients with ALD is the risk of recidivism and its effect on outcome and public opinion. With respect to the frequency of recidivism, this depends critically on its definition. Studies that have considered any alcohol use posttransplantation as a “relapse” have reported recidivism rates as high as 49%,429–431 whilst those that have restricted the definition to heavy or problem drinking have reported lower rates of 10–15%.425,432–436 With respect to the influence of recidivism on outcome, thus far, there are few data on which to base firm conclusions. From the information available, the incidence of graft dysfunction related to recidivism ranges from 0 to 17% and mortality ranges from 0 to 5%.431,432,434,435 In a study from 2001, the recidivism rates were 30%, with many showing evidence of recurrent disease on their protocol biopsies. Neither recidivism nor histology affected 84-month survival rates.437 In spite of this apparently reassuring report there is now evidence that if the follow-up is prolonged to 10 years, mortality in recidivists is significantly higher than in abstainers.438 It is therefore imperative that patients are monitored carefully for relapse following transplantation, with relapsers offered appropriate counseling. It is important that this is done for all ALD patients since, at present, there are few factors have been identified that reliably predict the risk of post-transplant recidivism prior to transplantation. Efforts to minimize the risk of post-transplant recidivism are important not only for the individual patient, but also to avoid the likely adverse effect this has on the organ-donating public. Public opinion. With organ shortage as a significant problem and while the decision to be an organ donor remains voluntary in most of the western world, it is imperative that the public are convinced that donated livers are being given to the most deserving patients. A recent UK study clearly demonstrated that, when compared to other patient groups, the general public, primary care physicians, and gastroenterologists all place patients with ALD well down their list of patients most deserving a liver transplant.439 The perception that patients with ALD have played a significant role in their disease and the widely held belief that “once a drinker, always a drinker” seem likely to be the most important factors contributing to this negative view of ALD patients. It is therefore vital that the public are made aware that patients are only offered transplantation if they fail to recover after a period of abstinence and that the incidence of significant post-transplant recidivism is low. Comorbidity. Excessive alcohol consumption can, and often does, affect many organ systems apart from the liver and this can potentially give rise to contraindications to surgery. An increased risk of pancreatitis, cardiomyopathy, osteoporosis, cerebrovascular disease, dementia, and malnutrition might all be expected to limit the numbers of patients fit for surgery. In practice, however, although most transplant units routinely screen ALD patients for cardiac and cerebral complications of excessive alcohol intake, this results in the
611
Section IV. Toxin Mediated Liver Injury
exclusion of very few patients.436 Similarly, despite the increased risk of psychiatric comorbidity in heavy drinkers, this rarely, if ever, leads to the exclusion of patients at the stage of transplant assessment.436 This seems likely to be attributable to patients with significant physical or psychiatric comorbidities not being referred for formal transplant assessment and/or to the tendency of many alcohol-related morbidities to improve during the period of abstinence required by most units prior to assessment. Preoperative abstinence. In light of the above considerations, it is perhaps not surprising that most centers require patients to have been abstinent for a period of time prior to assessment. This is primarily to give the liver a chance to recover spontaneously; however, it also allows time for other alcohol-related morbidities to recover, thereby improving the patient’s fitness for surgery and, importantly, satisfies public opinion. During this period, the patient can also be put in contact with alcohol treatment services for support both before and after transplantation. Whilst there is broad consensus on the need for a period of pretransplant abstinence, there is far less agreement on the requirement for a minimum duration. Many units have previously insisted on a 6-month period of abstinence, possibly attributable to early reports that this was a positive predictor of post-transplant abstinence.425 While a recent study confirms a 6month period of abstinence to be associated with reduced rates of recidivism at 1 and 2 years post-transplant,440 other studies have shown this period of abstinence to have little if any predictive power for subsequent drinking habits.435 This is obviously a very important question as the price of insisting on a fixed abstinence period may be death in some patients. A recent study demonstrating that the chance of recovery in patients with decompensated ALD can be predicted as early as 3 months441 has led some observers to suggest that, if required at all, the minimum period of abstinence could safely be reduced to 3 months rather than 6 months.442 Currently, it appears that, in practice, most centers do not adhere strictly to a fixed period of abstinence, instead preferring to assess each case on an individual basis and listing the patient when it is considered that recovery is unlikely.423
Transplantation for Alcoholic Hepatitis Clearly, many of these issues are pertinent when it comes to considering the possibility of transplanting patients with severe acute alcoholic hepatitis. A reasonable period of abstinence is not possible to assess the liver’s potential for spontaneous recovery, significant comorbidities are almost universal, and formal psychiatric assessment and pretransplant counseling are often precluded by the severity of the illness. Accordingly, although these patients undoubtedly have a poor prognosis without transplantation, most clinical centers do not consider these patients for liver transplantation. Nonetheless, there have been isolated reports of survival following transplantation of these patients371,443 and a recent report that the presence of histological alcoholic hepatitis in the explanted liver of patients transplanted for apparently chronic stable ALD is not associated with a worse prognosis or an increased risk of recidivism.444 Clearly, some patients with alcoholic hepatitis can benefit from transplantation; however, more data are required before any firm recommendations can be given on which patients (if any) are likely to derive the most benefit.
612
Gut permeability
Alcohol metabolism ADH
CYP2E1/Fe Oxidative stress
Endotoxin
Acetaldehyde
4
1
Immunological damage (adducts)
Lipid peroxidation KC
4
MDA
2 Cytokines
5
Necroinflammation and/or apoptosis 3
Figure 29-11. Putative mechanisms of hepatocyte injury in alcoholic hepatitis with potential targets for therapy: (1) antiendotoxin therapy; antibiotics, probiotics, enteral nutrition; (2) anticytokine therapy; corticosteroids, pentoxifylline, Infliximab; (3) antiapoptotic therapy; caspase inhibitors; (4) antioxidants; (5) immune-based therapy; corticosteroids. CYP2E1/Fe, ADH, MDA, KC.
CONCLUSIONS The short-term social and psychological benefits of alcohol have meant that its use and abuse have been widespread in many societies. End-stage liver disease is the result of prolonged heavy alcohol intake among a small proportion of users. Nevertheless, ALD still accounts for around half the total number of deaths from cirrhosis in the USA, and a great many more patients with fibrosis and alcoholic hepatitis. It therefore makes up a significant proportion of the workload of most liver units. The interaction between the physical and psychological dependence of the drug and the complexity of disease susceptibility makes this patient population a fascinating group. The multiple potential mechanisms of pathogenesis make it an intriguing disease to study, made all the more interesting considering the gray areas around the relative importance of each mechanism (Figure 29-11). Animal work is slowly improving this knowledge. This is filtering down into human trials of novel treatments, but improvements are slow considering the high mortality of acute alcoholic hepatitis, and liver transplantation remains the mainstay of treatment for advanced cirrhosis. Further research to understand the basics of hepatocyte injury are required to fuel further clinical trials and it must be expected that the next 10 years will lead to significant improvements in patient survival.
REFERENCES 1. Lelbach WK. Epidemiology of alcoholic liver disease: continental Europe. In: Hall P, ed. Alcoholic liver disease. London: Edward Arnold; 1985:130–166. 2. Masse L, Juillan JM, Chisloup A. Trends in mortality from cirrhosis of the liver, 1950–1971. World health statistics report. Rapport de Statistiques Sanitaires Mondiales 1976; 29:40–67. 3. Mandayam S, Jamal MM, Morgan TR. Epidemiology of alcoholic liver disease. Semin Liver Dis 2004; 24:217–232. 4. Naveau S, Giraud V, Borotto E, et al. Excess weight risk factor for alcoholic liver disease. Hepatology 1997; 25(1): p. 108–111.
Chapter 29 ALCOHOLIC LIVER DISEASE
5. Yoon Y-H, Yi HY, Smothers BA. Liver cirrhosis mortality in the United States 1970–2001; surveillance report #67. NIAAA: 2004. Available online at www.niaaa.gov. 6. Rubin E, Lieber CS. Experimental alcoholic hepatitis: a new primate model. Science 1973; 182: 712–713. 7. Ainley CC, Senapati A, Brown IM, et al. Is alcohol hepatotoxic in the baboon? J Hepatol 1988; 7:85–92. 8. Berggren S, Goldberg L. The absorption of ethyl alcohol from the gastrointestinal tract as a diffusion process. Acta Physiol Scand 1940; 1:246–270. 9. Marshall AW, Kingstone D, Boss M, Morgan MY. Ethanol elimination in males and females: relationship to menstrual cycle and body composition. Hepatology 1983; 3:701–706. 10. Caballeria J, Baraona E, Lieber CS. The contribution of the stomach to ethanol oxidation in the rat. Life Sci 1987; 41:1021–1027. 11. Frezza M, di Padova C, Pozzato G, et al. High blood alcohol levels in women. The role of decreased gastric alcohol dehydrogenase activity and first-pass metabolism [comment] [erratum appears in N Engl J Med 1990; 323:553]. N Engl J Med 1990; 322:95–99. 12. Parlesak A, Billinger MH, Bode C, Bode JC. Gastric alcohol dehydrogenase activity in man: influence of gender, age, alcohol consumption and smoking in a caucasian population. Alcohol Alcoholism 2002; 37:388–393. 13. Caballeria J, Baraona E, Rodamilans B, Lieber CS. Effects of cimetidine on gastric alcohol dehydrogenase activity and blood ethanol levels [comment]. Gastroenterology 1989; 96:388–392. 14. Brown AS, Fiaterone JR, Day CP, et al. Ranitidine increases the bioavailability of postprandial ethanol by the reduction of first pass metabolism. Gut 1995; 37:413–417. 15. Brown AS, Fiaterone JR, Wood P, et al. The effect of gastritis on human gastric alcohol dehydrogenase activity and ethanol metabolism. Aliment Pharmacol Ther 1995; 9:57–61. 16. Vallee BL, Bazzone TJ. Isozymes of human liver alcohol dehydrogenase. Isozymes: Curr Topics Biol Med Res 1983; 8:219–244. 17. Bosron WF, Li TK. Genetic polymorphism of human liver alcohol and aldehyde dehydrogenases, and their relationship to alcohol metabolism and alcoholism. Hepatology 1986; 6:502–510. 18. Moreno A, Pares X. Purification and characterization of a new alcohol dehydrogenase from human stomach. J Biol Chem 1991; 266:1128–1133. 19. Lester D, Keokosky WZ. Alcohol metabolism in the horse. Life Sci 1967; 6:2313–2319. 20. Lieber CS, DeCarli LM. Ethanol oxidation by hepatic microsomes: adaptive increase after ethanol feeding. Science 1968; 162:917–918. 21. Umeno M, McBride OW, Yang CS, et al. Human ethanolinducible P450IIE1: complete gene sequence, promoter characterization, chromosome mapping, and cDNA-directed expression. Biochemistry 1988; 27:9006–9013. 22. Seeff LB, Cuccherini BA, Zimmerman HJ, et al. Acetaminophen hepatotoxicity in alcoholics. A therapeutic misadventure. Ann Intern Med 1986; 104:399–404. 23. Zimmerman HJ, Maddrey WC. Acetaminophen (paracetamol) hepatotoxicity with regular intake of alcohol: analysis of instances of therapeutic misadventure. Hepatology 1995; 22:767–773. 24. Nagai H, Matsumaru K, Feng G, Kaplowitz N. Reduced glutathione depletion causes necrosis and sensitization to tumor necrosis factor-alpha-induced apoptosis in cultured mouse hepatocytes. Hepatology 2002; 36:55–64. 25. Gordon GG, Altman K, Southren AL, et al. Effect of alcohol (ethanol) administration on sex-hormone metabolism in normal men. N Engl J Med 1976; 295:793–797.
26. Boveris A, Oshino N, Chance B. The cellular production of hydrogen peroxide. Biochem J 1972; 128:617–630. 27. Smith M. Genetics of human alcohol and aldehyde dehydrogenases. Adv Hum Genet 1986; 15:249–290. 28. Goedde HW, Agarwal DP. Polymorphism of aldehyde dehydrogenase and alcohol sensitivity. Enzyme 1987; 37:29–44. 29. Crabb DW, Edenberg HJ, Bosron WF, Li TK. Genotypes for aldehyde dehydrogenase deficiency and alcohol sensitivity. The inactive ALDH2(2) allele is dominant. J Clin Invest 1989; 83:314–316. 30. Enomoto N, Takase S, Takada N, Takada A. Alcoholic liver disease in heterozygotes of mutant and normal aldehyde dehydrogenase-2 genes. Hepatology 1991; 13:1071–1075. 31. Bernstein J, Videla L, Israel Y. Hormonal influences in the development of the hypermetabolic state of the liver produced by chronic administration of ethanol. J Pharmacol Exp Ther 1975; 192(3): 583–591. 32. Lieberman FL. The effect of liver diseases on the rate of ethanol metabolism in man. Gastroenterology 1963. 44: 261–266. 33. Figueroa RB, Klatz AP. Alterations of liver alcohol dehydrogenase and other hepatic enzymes in alcoholic cirrhosis. Gastroenterology 1962. 43(10). 34. Korsten MA, Matsuzaki S, Feinman L, Lieber CS. High blood acetaldehyde levels after ethanol administration. Difference between alcoholic and nonalcoholic subjects. N Engl J Med 1975; 292:386–389. 35. Lindros KO, Stowell A, Pikkarainen P, Salaspuro M. Elevated blood acetaldehyde in alcoholics with accelerated ethanol elimination. Pharmacol Biochem Behav 1980; 13 (suppl 1):119–124. 36. Nuutinen H, Lindros KO, Salaspuro M. Determinants of blood acetaldehyde level during ethanol oxidation in chronic alcoholics. Alcoholism Clin Exp Res 1983; 7:163–168. 37. Day CP, Yeaman SJ. Physical evidence for the presence of two forms of phosphatidate phosphohydrolase in rat liver. Biochim Biophys Acta 1992; 1127:87–94. 38. Nanji AA, Tsukamoto H, French SW. Relationship between fatty liver and subsequent development of necrosis, inflammation and fibrosis in experimental alcoholic liver disease. Exp Mol Pathol 1989; 51:141–148. 39. Elabbadi N, Day CP, Gamouth A, et al. Relationship between the inhibition of phosphatidic acid phosphohydrolase-1 by oleate and oleoyl-CoA ester and its apparent translocation. Biochimie 2005; 87:437–443. 40. Aoyama T, Peters JM, Iritani N, et al. Altered constitutive expression of fatty acid-metabolizing enzymes in mice lacking the peroxisome proliferator-activated receptor alpha (PPARalpha). J Biol Chem 1998; 273:5678–5684. 41. Galli A, Pinaire J, Fischer M, et al. The transcriptional and DNA binding activity of peroxisome proliferator-activated receptor alpha is inhibited by ethanol metabolism. A novel mechanism for the development of ethanol-induced fatty liver. J Biol Chem 2001; 276:68–75. 42. Fischer M, You M, Matsumoto M, Crabb DW. Peroxisome proliferator-activated receptor alpha (PPARalpha) agonist treatment reverses PPARalpha dysfunction and abnormalities in hepatic lipid metabolism in ethanol-fed mice. J Biol Chem 2003; 278:27997–28004. 43. Nanji AA, Dannenberg AJ, Jokelainen K, Bass NM. Alcoholic liver injury in the rat is associated with reduced expression of peroxisome proliferator-alpha (PPARalpha)-regulated genes and is ameliorated by PPARalpha activation. J Pharmacol Exp Ther 2004; 310:417–424. 44. Nakajima T, Kamijo Y, Tanaka N, et al. Peroxisome proliferatoractivated receptor alpha protects against alcohol-induced liver damage [see comment]. Hepatology 2004; 40:972–980.
613
Section IV. Toxin Mediated Liver Injury
45. Tuma DJ, Smith SL, Sorrell MF. Acetaldehyde and microtubules. Ann NY Acad Sci 1991; 625:786–792. 46. Yoon Y, Torok N, Krueger E, et al. Ethanol-induced alterations of the microtubule cytoskeleton in hepatocytes. Am J Physiol 1998; 274:G757–G766. 47. Sugimoto T, Yamashita S, Ishigami M, et al. Decreased microsomal triglyceride transfer protein activity contributes to initiation of alcoholic liver steatosis in rats. J Hepatol 2002; 36:157–162. 48. Yin M, Wheeler MD, Kono H, et al. Essential role of tumor necrosis factor alpha in alcohol-induced liver injury in mice. Gastroenterology 1999; 117:942–952. 49. Ueki K, Kondo T, Tseng YH, Kahn CR. Central role of suppressors of cytokine signaling proteins in hepatic steatosis, insulin resistance, and the metabolic syndrome in the mouse. Proc Natl Acad Sci USA 2004; 101:10422–10427. 50. Xu A, Wang Y, Keshaw H, et al. The fat-derived hormone adiponectin alleviates alcoholic and nonalcoholic fatty liver diseases in mice. J Clin Invest 2003; 112:91–100. 51. Villanueva JA, Halsted CH. Hepatic transmethylation reactions in micropigs with alcoholic liver disease. Hepatology 2004; 39:1303–1310. 52. Kenyon SH, Nicolaou A, Gibbons WA. The effect of ethanol and its metabolites upon methionine synthase activity in vitro. Alcohol 1998; 15:305–309. 53. Mori K. Tripartite management of unfolded proteins in the endoplasmic reticulum. Cell 2000; 101:451–454. 54. Kaufman RJ. Stress signaling from the lumen of the endoplasmic reticulum: coordination of gene transcriptional and translational controls. Genes Dev 1999; 13:1211–1233. 55. Outinen PA, Sood SK, Pfeifer SI, et al. Homocysteine-induced endoplasmic reticulum stress and growth arrest leads to specific changes in gene expression in human vascular endothelial cells. Blood 1999; 94:959–967. 56. You M, Fischer M, Deeg MA, Crabb DW. Ethanol induces fatty acid synthesis pathways by activation of sterol regulatory element-binding protein (SREBP). J Biol Chem 2002; 277:29342–29347. 57. You M, Matsumoto M, Pacold CM, et al. The role of AMPactivated protein kinase in the action of ethanol in the liver. Gastroenterology 2004; 127:1798–1808. 58. Ji C, Kaplowitz N. Betaine decreases hyperhomocysteinemia, endoplasmic reticulum stress, and liver injury in alcohol-fed mice. Gastroenterology 2003; 124:1488–1499. 59. Ji C, Deng Q, Kaplowitz N. Role of TNF-alpha in ethanolinduced hyperhomocysteinemia and murine alcoholic liver injury. Hepatology 2004; 40:442–451. 60. Day CP, James OF. Hepatic steatosis: innocent bystander or guilty party? Hepatology 1998; 27:1463–1466. 61. Sorensen TI, Orholm M, Bentsen KD, et al. Prospective evaluation of alcohol abuse and alcoholic liver injury in men as predictors of development of cirrhosis. Lancet 1984; 2:241–244. 62. Teli MR, Day CP, Burt AD, et al. Determinants of progression to cirrhosis or fibrosis in pure alcoholic fatty liver [comment]. Lancet 1995; 346:987–990. 63. Yang SQ, Lin HZ, Lane MD, et al. Obesity increases sensitivity to endotoxin liver injury: implications for the pathogenesis of steatohepatitis. Proc Natl Acad Sci USA 1997; 94:2557–2562. 64. Letteron P, Fromenty B, Terris B, et al. Acute and chronic hepatic steatosis lead to in vivo lipid peroxidation in mice. J Hepatol 1996; 24:200–208. 65. Guebre-Xabier M, Yang S, Lin R, et al. Altered hepatic lymphocyte subpopulations in obesity-related murine fatty livers: potential mechanism for sensitization to liver damage. Hepatology 2000; 31:633–640. 66. Halliwell B. Free radicals, antioxidants, and human disease: curiosity, cause, or consequence? [see comments] Lancet 1994; 344:721–724.
614
67. Parola M, Robino G. Oxidative stress-related molecules and liver fibrosis. J Hepatol 2001; 35:297–306. 68. Clot P, Tabone M, Arico S, Albano E. Monitoring oxidative damage in patients with liver cirrhosis and different daily alcohol intake. Gut 1994; 35:1637–1643. 69. Paradis V, Kollinger M, Fabre M, et al. In situ detection of lipid peroxidation by-products in chronic liver diseases. Hepatology 1997; 26:135–142. 70. Situnayake RD, Crump BJ, Thurnham DI, et al. Lipid peroxidation and hepatic antioxidants in alcoholic liver disease. Gut 1990; 31:1311–1317. 71. Ekstrom G, Ingelman-Sundberg M. Rat liver microsomal NADPH-supported oxidase activity and lipid peroxidation dependent on ethanol-inducible cytochrome P-450 (P-450IIE1). Biochem Pharmacol 1989; 38:1313–1319. 72. Albano E, Tomasi A, Goria-Gatti L, Dianzini MU. Spin trapping of free radical species produced during the microsomal metabolism of ethanol. Chem-Biol Interact 1988; 65:223–234. 73. Slater TF, Sawyer BC, Strauli UD. Changes in liver nucleotide concentrations in experimental liver injury. 2. Acute ethanol poisoning. Biochem J 1964; 93:267–270. 74. Albano E, Clot P, Morimoto M, et al. Role of cytochrome P4502E1-dependent formation of hydroxyethyl free radical in the development of liver damage in rats intragastrically fed with ethanol. Hepatology 1996; 23:155–163. 75. Morimoto M, Hagbjork AL, Nanji AA, et al. Role of cytochrome P4502E1 in alcoholic liver disease pathogenesis. Alcohol 1993; 10:459–464. 76. Kukielka E, Dicker E, Cederbaum AI. Increased production of reactive oxygen species by rat liver mitochondria after chronic ethanol treatment. Arch Biochem Biophys 1994; 309:377–386. 77. Ma MH. Ultrastructural pathologic findings of the human hepatocyte. I. Alcoholic liver disease. Arch Pathol 1972; 94:554–571. 78. Arai M, Leo MA, Nakano M, et al. Biochemical and morphological alterations of baboon hepatic mitochondria after chronic ethanol consumption. Hepatology 1984; 4:165–174. 79. Brown GC. Nitric oxide regulates mitochondrial respiration and cell functions by inhibiting cytochrome oxidase. FEBS Lett 1995; 369: 136–139. 80. Poderoso JJ, Carreras MC, Lisdero C, et al. Nitric oxide inhibits electron transfer and increases superoxide radical production in rat heart mitochondria and submitochondrial particles. Arch Biochem Biophys 1996; 328:85–92. 81. McKim SE, Gabele E, Isayama F, et al. Inducible nitric oxide synthase is required in alcohol-induced liver injury: studies with knockout mice. Gastroenterology 2003; 125:1834–1844. 82. Niemela O, Parkkila S, Bradford B, et al. Effect of Kupffer cell inactivation on ethanol-induced protein adducts in the liver. Free Radical Biol Med 2002; 33:350–355. 83. Kono H, Rusyn I, Yin M, et al. NADPH oxidase-derived free radicals are key oxidants in alcohol-induced liver disease. J Clin Invest 2000; 106:867–872. 84. Paredes SR, Kozicki PA, Fukuda H, et al. S-adenosyl-Lmethionine: its effect on aminolevulinate dehydratase and glutathione in acute ethanol intoxication. Alcohol 1987; 4:81–85. 85. Hirano T, Kaplowitz H, Tsukamoto H, et al. Hepatic mitochondrial glutathione depletion and progression of experimental alcoholic liver disease in rats. Hepatology 1992; 16:1423–1427. 86. Colell A, Garcia-Ruiz C, Miranda M, et al. Selective glutathione depletion of mitochondria by ethanol sensitizes hepatocytes to tumor necrosis factor. Gastroenterology 1998; 115:1541–1551. 87. Song Z, Zhou Z, Uriarte S, et al. S-adenosylhomocysteine sensitizes to TNF-alpha hepatotoxicity in mice and liver cells: a possible etiological factor in alcoholic liver disease. Hepatology 2004; 40:989–997.
Chapter 29 ALCOHOLIC LIVER DISEASE
88. Dworkin B, Rosenthal WS, Jankowski RH, et al. Low blood selenium levels in alcoholics with and without advanced liver disease. Correlations with clinical and nutritional status. Digest Dis Sci 1985; 30:838–844. 89. Thurnham DI, Davies JA, Crump BJ, et al. The use of different lipids to express serum tocopherol: lipid ratios for the measurement of vitamin E status. Ann Clin Biochem 1986; 23:514–520. 90. Leo MA, Lieber CS. Hepatic vitamin A depletion in alcoholic liver injury. N Engl J Med 1982; 307:597–601. 91. Bianchi GP, Fiorella PL, Bargossi AM, et al. Reduced ubiquinone plasma levels in patients with liver cirrhosis and in chronic alcoholics. Liver 1994; 14:138–140. 92. Stewart SF, Prince M, Bassendine MF, et al. A trial of antioxidant therapy alone or with corticosteroids in acute alcoholic hepatitis. J Hepatol 2002; 36 (suppl 1): 16. 93. Nanji AA, Mendenhall CL, French SW. Beef fat prevents alcoholic liver disease in the rat. Alcohol Clin Exp Res 1989; 13:15–19. 94. Nanji AA, French SW. Dietary linoleic acid is required for development of experimentally induced alcoholic liver injury. Life Sci 1989; 44:223–227. 95. Nanji AA, Zhao S, Sadrzadeh SM, et al. Markedly enhanced cytochrome P450 2E1 induction and lipid peroxidation is associated with severe liver injury in fish oil-ethanol-fed rats. Alcohol Clin Exp Res 1994; 18:1280–1285. 96. Arias-Negrete S, Keller K, Chadee K. Proinflammatory cytokines regulate cyclooxygenase-2 mRNA expression in human macrophages. Biochem Biophys Res Commun 1995; 208:582–589. 97. Tsukamoto H, Horne W, Kamimura S, et al. Experimental liver cirrhosis induced by alcohol and iron. J Clin Invest 1995; 96:620–630. 98. Kessova IG, Ho YS, Thung S, Cederbaum AI. Alcohol-induced liver injury in mice lacking Cu, Zn-superoxide dismutase. Hepatology 2003; 38:1136–1145. 99. Lu SC, Alvarez L, Huang ZZ, et al. Methionine adenosyltransferase 1A knockout mice are predisposed to liver injury and exhibit increased expression of genes involved in proliferation. Proc Natl Acad Sci USA 2001; 98: 5560–5565. 100. Lieber CS. S-Adenosyl-L-methionine and alcoholic liver disease in animal models: implications for early intervention in human beings. Alcohol 2002; 27:173–177. 101. Sadrzadeh SM, Nanji AA, Price PL. The oral iron chelator, 1,2dimethyl-3-hydroxypyrid-4-one reduces hepatic-free iron, lipid peroxidation and fat accumulation in chronically ethanol-fed rats. J Pharmacol Exp Ther 1994; 269:632–636. 102. Wheeler MD, Nakagami M, Bradford BU, et al. Overexpression of manganese superoxide dismutase prevents alcohol-induced liver injury in the rat. J Biol Chem 2001; 276:36664–36672. 103. Donohue TM Jr, Tuma DJ, Sorrell MF. Acetaldehyde adducts with proteins: binding of [14C]acetaldehyde to serum albumin. Arch Biochem Biophys 1983; 220:239–246. 104. Tuma DJ, Jennett RB, Sorrell MF. The interaction of acetaldehyde with tubulin. Ann NY Acad Sci 1987; 492:277–286. 105. Stevens VJ, Fantl WJ, Newman CB, et al. Acetaldehyde adducts with hemoglobin. J Clin Invest 1981; 67(2): 361–369. 106. Tuma DJ, Sorrell MF. Covalent binding of acetaldehyde to hepatic proteins: role in alcoholic liver injury. Progr Clin Biol Res 1985; 183: 3–17. 107. Behrens UJ, Ma XL, Bychenok S, et al. Acetaldehyde-collagen adducts in CCl4-induced liver injury in rats. Biochem Biophys Res Commun 1990; 173:111–119. 108. Behrens UJ, Hoerner M, Lasker JM, Lieber CS. Formation of acetaldehyde adducts with ethanol-inducible P450IIE1 in vivo. Biochem Biophys Res Commun 1988; 154:584–590.
109. Mauch TJ, Donohue TM Jr, Zetterman RK, et al. Covalent binding of acetaldehyde selectively inhibits the catalytic activity of lysine-dependent enzymes. Hepatology 1986; 6:263–269. 110. Matsuda Y, Baraona E, Salaspuro M, Lieber CS. Effects of ethanol on liver microtubules and Golgi apparatus. Possible role in altered hepatic secretion of plasma proteins. Lab Invest 1979; 41:455–463. 111. Baraona E, Pikkarainen P, Salaspuro M, Lieber CS. Acute effects of ethanol on hepatic protein synthesis and secretion in the rat. Gastroenterology 1980; 79:104–111. 112. Zhu Y, Fillenwarth MJ, Crabb D, et al. Identification of the 37kd rat liver protein that forms an acetaldehyde adduct in vivo as delta 4-3-ketosteroid 5 beta-reductase. Hepatology 1996; 23:115–122. 113. Pastorino JG, Hoek JB. Ethanol potentiates tumor necrosis factor-alpha cytotoxicity in hepatoma cells and primary rat hepatocytes by promoting induction of the mitochondrial permeability transition. Hepatology 2000; 31:1141–1152. 114. Lluis JM, Colell A, Garcia-Ruiz C, et al. Acetaldehyde impairs mitochondrial glutathione transport in HepG2 cells through endoplasmic reticulum stress. Gastroenterology 2003; 124:708–724. 115. Nolan JP. Intestinal endotoxins as mediators of hepatic injury – an idea whose time has come again. Hepatology 1989; 10:887–891. 116. Fukui H, Brauner B, Bode JC, Bode C. Plasma endotoxin concentrations in patients with alcoholic and non-alcoholic liver disease: reevaluation with an improved chromogenic assay. J Hepatol 1991; 12:162–169. 117. Nanji AA, Khettry U, Sadrzadeh SM, Yamanaka T. Severity of liver injury in experimental alcoholic liver disease. Correlation with plasma endotoxin, prostaglandin E2, leukotriene B4, and thromboxane B2. Am J Pathol 1993; 142:367–373. 118. Schumann RR, Leong SR, Flaggs GW, et al. Structure and function of lipopolysaccharide-binding protein. Science 1990; 249:1429–1431. 119. Wright SD, Ramos RA, Tobias PS, et al. CD14, a receptor for complexes of lipopolysaccharide (LPS) and LPS binding protein. Science 1990; 249: 1431–1433. 120. Poltorak A, He X, Smirnova I, et al. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 1998; 282:2085–2088. 121. Pasare C, Medzhitov R. Toll pathway-dependent blockade of CD4+CD25+ T cell-mediated suppression by dendritic cells [comment]. Science 2003; 299:1033–1036. 122. Tsukamoto H, Reidelberger RD, French SW, Largman C. Long-term cannulation model for blood sampling and intragastric infusion in the rat. Am J Physiol 1984; 247: R595–R599. 123. Adachi Y, Moore LE, Bradford BU, et al. Antibiotics prevent liver injury in rats following long-term exposure to ethanol. Gastroenterology 1995; 108:218–224. 124. Uesugi T, Froh M, Arteel GE, et al. Toll-like receptor 4 is involved in the mechanism of early alcohol-induced liver injury in mice. Hepatology 2001; 34:101–108. 125. Jarvelainen HA, Orpana A, Perola M, et al. Promoter polymorphism of the CD14 endotoxin receptor gene as a risk factor for alcoholic liver disease [see comment]. Hepatology 2001; 33:1148–1153. 126. Enomoto N, Ikejima K, Bradford B, et al. Alcohol causes both tolerance and sensitization of rat Kupffer cells via mechanisms dependent on endotoxin. Gastroenterology 1998; 115:443–451. 127. Knolle PA, Gerken G. Local control of the immune response in the liver. Immunol Rev 2000; 174:21–34. 128. McClain CJ, Cohen DA. Increased tumor necrosis factor production by monocytes in alcoholic hepatitis. Hepatology 1989; 9:349–351.
615
Section IV. Toxin Mediated Liver Injury
129. Bird G, Sheron N, Goka AK, et al. Increased plasma tumour necrosis factor in severe alcoholic hepatitis. Ann Intern Med 1990; 112:917–920. 130. Natori S, Rust C, Stadheim LM, et al. Hepatocyte apoptosis is a pathologic feature of human alcoholic hepatitis. J Hepatol 2001; 34:248–253. 131. Day CP. Apoptosis in alcoholic hepatitis: a novel therapeutic target? [comment] J Hepatol 2001; 34:330–333. 132. Pessayre D, Fromenty B. NASH: a mitochondrial disease. J Hepatol 2005; 42:928–940. 133. Nechushtan A, Smith CL, Lamensforf I, et al. Bax and Bak coalesce into novel mitochondria-associated clusters during apoptosis. J Cell Biol 2001; 153:1265–1276. 134. Kirkland RA, Windelborn JA, Kasprzak JM, Franklin JL. A Baxinduced pro-oxidant state is critical for cytochrome c release during programmed neuronal death. J Neurosci 2002; 22:6480–6490. 135. Matsumaru K, Ji C, Kaplowitz N. Mechanisms for sensitization to TNF-induced apoptosis by acute glutathione depletion in murine hepatocytes. Hepatology 2003; 37:1425–1434. 136. Crispe IN, Mehal WZ. Strange brew. T cells in the liver. Immunol Today 1996; 17:522–525. 137. Laso FJ, Madruga JI, Giron JA, et al. Decreased natural killer cytotoxic activity in chronic alcoholism is associated with alcohol liver disease but not active ethanol consumption. Hepatology 1997; 25:1096–1100. 138. Minagawa M, Deng Q, Liu ZX, et al. Activated natural killer T cells induce liver injury by Fas and tumor necrosis factor-alpha during alcohol consumption. Gastroenterology 2004; 126:1387–1399. 139. Pruett SB, Schwab C, Zheng Q, Fan R. Suppression of innate immunity by acute ethanol administration: a global perspective and a new mechanism beginning with inhibition of signaling through TLR3. J Immunol 2004; 173:2715–2724. 140. Mathurin P, Duchatelle V, Ramond MJ, et al. Survival and prognostic factors in patients with severe alcoholic hepatitis treated with prednisolone. Gastroenterology 1996; 110:1847–1853. 141. Zylberberg H, Fontaine H, Thepot V, et al. Triggering of acute alcoholic hepatitis by alpha-interferon therapy. J Hepatol 1999; 30:722–725. 142. Chedid A, Mendenhall CL, Moritz TE, et al. Cell-mediated hepatic injury in alcoholic liver disease. Veterans Affairs Cooperative Study Group 275. Gastroenterology 1993; 105:254–266. 143. Burt AD, Anthony RS, Hislop WS, et al. Liver membrane antibodies in alcoholic liver disease: 1. prevalence and immunoglobulin class. Gut 1982; 23:221–225. 144. Trevisan A, Cavigli R, Melocni R, et al. Detection of immunoglobulins G and A on the cell membrane of hepatocytes from patients with alcoholic liver disease. J Clin Pathol 1983; 36:530–534. 145. Van Pelt FN, Kenna JG. Formation of trifluoroacetylated protein antigens in cultured rat hepatocytes exposed to halothane in vitro. Biochem Pharmacol 1994; 48:461–471. 146. Gut J, Christen U, Frey N, et al. Molecular mimicry in halothane hepatitis: biochemical and structural characterization of lipoylated autoantigens. Toxicology 1995; 97:199–224. 147. Eliasson E, Kenna JG. Cytochrome P450 2E1 is a cell surface autoantigen in halothane hepatitis. Mol Pharmacol 1996; 50:573–582. 148. Homberg JC, Andre C, Abuaf N. A new anti-liver-kidney microsome antibody (anti-LKM2) in tienilic acid-induced hepatitis. Clin Exp Immunol 1984; 55:561–570. 149. Beaune P, Dansette PM, Mansuy D, et al. Human antiendoplasmic reticulum autoantibodies appearing in a drug-
616
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
induced hepatitis are directed against a human liver cytochrome P-450 that hydroxylates the drug. Proc Natl Acad Sci USA 1987; 84:551–555. Israel Y, Hurwitz E, Niemela O, Arnon R. Monoclonal and polyclonal antibodies against acetaldehyde-containing epitopes in acetaldehyde-protein adducts. Proc Natl Acad Sci USA 1986; 83:7923–7927. Hoerner M, Behrens UJ, Worner T, Lieber CS. Humoral immune response to acetaldehyde adducts in alcoholic patients. Res Commun Chem Pathol Pharmacol 1986; 54:3–12. Lin RC, Smith RS, Lumeng L. Detection of a proteinacetaldehyde adduct in the liver of rats fed alcohol chronically. J Clin Invest 1988; 81:615–619. Nicholls RM, Fowles LF, Worrall S, et al. Distribution and turnover of acetaldehyde-modified proteins in liver and blood of ethanol-fed rats. Alcohol Alcoholism 1994; 29:149–157. Niemela O, Juvonen T, Parkkila S. Immunohistochemical demonstration of acetaldehyde-modified epitopes in human liver after alcohol consumption. J Clin Invest 1991; 87:1367–1374. Holstege A, Bedossa P, Pynard T, et al. Acetaldehyde-modified epitopes in liver biopsy specimens of alcoholic and nonalcoholic patients: localization and association with progression of liver fibrosis. Hepatology 1994; 19:367–374. Worrall S, De Jersey J, Shanley BC, Wilce PA. Antibodies against acetaldehyde-modified epitopes: presence in alcoholic, non-alcoholic liver disease and control subjects. Alcohol Alcohlism 1990; 25:509–517. Niemela O, Klanjer F, Orrego H, et al. Antibodies against acetaldehyde-modified protein epitopes in human alcoholics. Hepatology 1987; 7:1210–1214. Viitala K, Israel Y, Blake JE, Niemela O. Serum IgA, IgG, and IgM antibodies directed against acetaldehyde-derived epitopes: relationship to liver disease severity and alcohol consumption. Hepatology 1997; 25:1418–1424. Albano E, Tomasi A, Persson JO, et al. Role of ethanol-inducible cytochrome P450 (P450IIE1) in catalysing the free radical activation of aliphatic alcohols. Biochem Pharmacol 1991; 41:1895–1902. Moncada C, Torres V, Varghese G, et al. Ethanol-derived immunoreactive species formed by free radical mechanisms. Mol Pharmacol 1994; 46:786–791. Clot P, Bellomo G, Tabone M, et al. Detection of antibodies against proteins modified by hydroxyethyl free radicals in patients with alcoholic cirrhosis. Gastroenterology 1995; 108:201–207. Niemela O, Parkkila S, Yla-Hertuala S, et al. Covalent protein adducts in the liver as a result of ethanol metabolism and lipid peroxidation. Lab Invest 1994; 70:537–546. Mottaran E, Stewart SF, Rolla R, et al. Lipid peroxidation contributes to immune reactions associated with alcoholic liver disease. Free Radical Biol Med 2002; 32:38–45. Tuma DJ, Thiele GM, Xu D, et al. Acetaldehyde and malondialdehyde react together to generate distinct protein adducts in the liver during long-term ethanol administration. Hepatology 1996; 23:872–880. Rolla R, Vay D, Mottaran E, et al. Detection of circulating antibodies against malondialdehyde-acetaldehyde adducts in patients with alcohol-induced liver disease. Hepatology 2000; 31:878–884. McFarlane IG, Wojcicka BM, Zucker GM, et al. Purification and characterization of human liver-specific membrane lipoprotein (LSP). Clin Exp Immunol 1977; 27:381–390. Kakumu S, Arakawa Y, Goji H, et al. Occurrence and significance of antibody to liver-specific membrane lipoprotein by double-antibody immunoprecipitation method in sera of patients with acute and chronic liver diseases. Gastroenterology 1979; 76:665–672.
Chapter 29 ALCOHOLIC LIVER DISEASE
168. Perperas A, Tsantoulas D, Portmann B, et al. Autoimmunity to a liver membrane lipoprotein and liver damage in alcoholic liver disease. Gut 1981; 22:149–152. 169. Hopf U, Meyer zum Buschenfelde KH, Arnold W. Detection of a liver-membrane autoantibody in HBsAg-negative chronic active hepatitis. N Engl J Med 1976; 294:578–582. 170. Rolla R, Vay D, Mottaran E, et al. Antiphospholipid antibodies associated with alcoholic liver disease specifically recognise oxidised phospholipids. Gut 2001; 49:852–859. 171. Clot P, Albano E, Eliasson E, et al. Cytochrome P4502E1 hydroxyethyl radical adducts as the major antigen in autoantibody formation among alcoholics. Gastroenterology 1996; 111:206–216. 172. Lytton SD, Helander A, Zhang-Gouillon ZQ, et al. Autoantibodies against cytochromes P-4502E1 and P-4503A in alcoholics. Mol Pharmacol 1999; 55:223–233. 173. Vidali M, Stewart SF, Rolla R, et al. Genetic and epigenetic factors in autoimmune reactions toward cytochrome P4502E1 in alcoholic liver disease. Hepatology 2003; 37:410–419. 174. Cochrane AM, Moussouros A, Portmann B, et al. Lymphocyte cytotoxicity for isolated hepatocytes in alcoholic liver disease. Gastroenterology 1977; 72:918–923. 175. Clot P, Parola M, Bellomo G, et al. Plasma membrane hydroxyethyl radical adducts cause antibody-dependent cytotoxicity in rat hepatocytes exposed to alcohol [see comments]. Gastroenterology 1997; 113:265–276. 176. Stewart SF, Vidali M, Day CP, et al. Oxidative stress as a trigger for cellular immune responses in patients with alcoholic liver disease. Hepatology 2004; 39:197–203. 177. Friedman SL. Cellular sources of collagen and regulation of collagen production in liver. Semin Liver Dis 1990; 10:20–29. 178. Gressner AM. The cell biology of liver fibrogenesis – an imbalance of proliferation, growth arrest and apoptosis of myofibroblasts. Cell Tissue Res 1998; 292:447–452. 179. Maher JJ, McGuire RF. Extracellular matrix gene expression increases preferentially in rat lipocytes and sinusoidal endothelial cells during hepatic fibrosis in vivo. J Clin Invest 1990; 86:1641–1648. 180. Safadi R, Ohta M, Alvarez CE, et al. Immune stimulation of hepatic fibrogenesis by CD8 cells and attenuation by transgenic interleukin-10 from hepatocytes. Gastroenterology 2004; 127:870–882. 181. Holt K, Bennett M, Chojkier M. Acetaldehyde stimulates collagen and noncollagen protein production by human fibroblasts. Hepatology 1984; 4:843–848. 182. Ma X, Svegliati-Baroni G, Poniachik J, et al. Collagen synthesis by liver stellate cells is released from its normal feedback regulation by acetaldehyde-induced modification of the carboxylterminal propeptide of procollagen. Alcohol Clin Exp Res 1997; 21:1204–1211. 183. Nieto N, Friedman SL, Cederbaum AI. Cytochrome P450 2E1derived reactive oxygen species mediate paracrine stimulation of collagen I protein synthesis by hepatic stellate cells. J Biol Chem 2002; 277:9853–9864. 184. Canbay A, Higuchi H, Bronk SF, et al. Fas enhances fibrogenesis in the bile duct ligated mouse: a link between apoptosis and fibrosis. Gastroenterology 2002; 123:1323–1330. 185. Canbay A, Taimr P, Torok N, et al. Apoptotic body engulfment by a human stellate cell line is profibrogenic. Lab Invest 2003; 83:655–663. 186. Fadok VA, Bratton DL, Konowal A, et al. Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF. J Clin Invest 1998; 101:890–898. 187. Nair S, Mason A, Eason J, et al. Is obesity an independent risk factor for hepatocellular carcinoma in cirrhosis? [erratum
188. 189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
201.
202.
203.
204.
205.
206. 207.
208.
appears in Hepatology 2002; 36:774]. Hepatology 2002; 36:150–155. Marnett LJ. Oxy radicals, lipid peroxidation and DNA damage. Toxicology 2002; 181–182:219–222. Fink SP, Reddy GR, Marnett LJ. Mutagenicity in Escherichia coli of the major DNA adduct derived from the endogenous mutagen malondialdehyde. Proc Natl Acad Sci USA 1997; 94:8652–8657. Ji C, Rouzer CA, Marnett LJ, Pietenpol JA. Induction of cell cycle arrest by the endogenous product of lipid peroxidation, malondialdehyde. Carcinogenesis 1998; 19:1275–1283. Roskams T, Yang SQ, Koteish A, et al. Oxidative stress and oval cell accumulation in mice and humans with alcoholic and nonalcoholic fatty liver disease. Am J Pathol 2003; 163:1301–1311. Kuper H, Tzonou A, Kaklamani E, et al. Tobacco smoking, alcohol consumption and their interaction in the causation of hepatocellular carcinoma. Int J Cancer 2000; 85:498–502. Bulatao-Jayme J, Almero EM, Castro MC, et al. A case-control dietary study of primary liver cancer risk from aflatoxin exposure. Int J Epidemiol 1982; 11:112–119. Garro AJ, Seitz HK, Lieber CS. Enhancement of dimethylnitrosamine metabolism and activation to a mutagen following chronic ethanol consumption. Cancer Res 1981; 41:120–124. Chung J, Liu C, Smith DE, et al. Restoration of retinoic acid concentration suppresses ethanol-enhanced c-Jun expression and hepatocyte proliferation in rat liver. Carcinogenesis 2001; 22:1213–1219. Wang XD, Liu C, Chung J, et al. Chronic alcohol intake reduces retinoic acid concentration and enhances AP-1 (c-Jun and c-Fos) expression in rat liver. Hepatology 1998; 28:744–750. Shen L, Fang J, Qiu D, et al. Correlation between DNA methylation and pathological changes in human hepatocellular carcinoma. HepatoGastroenterology 1998; 45:1753–1759. Zapisek WF, Cronin GM, Lyn-Cook BD, Poirier LA. The onset of oncogene hypomethylation in the livers of rats fed methyldeficient, amino acid-defined diets. Carcinogenesis 1992; 13:1869–1872. Gallucci RM, Pfister LJ, Meadows GG. Effects of ethanol consumption on enriched natural killer cells from C57BL/6 mice. Alcohol Clin Exp Res 1994; 18:625–631. Miyakawa H, Sato C, Izumi N, et al. Hepatitis C virus infection in alcoholic liver cirrhosis in Japan: its contribution to the development of hepatocellular carcinoma. Alcohol Alcohol Suppl 1993; 1A:85–90. Bellentani S, Saccoccio G, Costa G, et al. Drinking habits as cofactors of risk for alcohol induced liver damage. The Dionysos Study Group [see comments]. Gut 1997; 41:845–850. Lelbach WK. Cirrhosis in the alcoholic and its relation to the volume of alcohol abuse. Ann NY Acad Sci 1975; 252:85–105. Krasner N, Davis M, Portmann B, Williams R. Changing pattern of alcoholic liver disease in Great Britain: relation to sex and signs of autoimmunity. Br Med J 1977; 1:1497–1500. Becker U, Deis A, Sorensen TI, et al. Prediction of risk of liver disease by alcohol intake, sex, and age: a prospective population study. Hepatology 1996; 23:1025–1029. Becker U, Gronbaek M, Johansen D, Sorensen TI. Lower risk for alcohol-induced cirrhosis in wine drinkers [see comment]. Hepatology 2002; 35:868–875. Nanji AA, French SW. Relationship between pork consumption and cirrhosis. Lancet 1985; 1:681–683. Rotily M, Durbec JP, Berthezene P, Sarles H. Diet and alcohol in liver cirrhosis: a case-control study. Eur J Clin Nutr 1990; 44:595–603. Raynard B, Balian B, Fallik D, et al. Risk factors of fibrosis in alcohol-induced liver disease. Hepatology 2002; 35:635–638.
617
Section IV. Toxin Mediated Liver Injury
209. Enomoto N, Takei Y, Kitamura T, et al. Estriol enhances lipopolysaccharide-induced increases in nitric oxide production by Kupffer cells via mechanisms dependent on endotoxin. Alcohol Clin Exp Res 2002; 26 (suppl):66S–69S. 210. Hrubec Z, Omenn GS. Evidence of genetic predisposition to alcoholic cirrhosis and psychosis: twin concordances for alcoholism and its biological end points by zygosity among male veterans. Alcoholism Clin Exp Res 1981; 5:207–215. 211. Stinson FS, Grant BF, Dufour MC. The critical dimension of ethnicity in liver cirrhosis mortality statistics. Alcohol Clin Exp Res 2001; 25:1181–1187. 212. Caetano R, Clark CL. Trends in alcohol-related problems among whites, blacks, and Hispanics: 1984–1995; Alcohol Clin Exp Res 1998; 22:534–538. 213. Daly A, Day CP. Candidate gene case-control association studies: advantages and pitfalls. Br J Clin Pharmacol 2001; 52:489–499. 214. Day CP, Leathart JB, Daly AK. Genetic evidence that fatty liver is involved in the pathogenesis of advanced alcoholic liver disease. Hepatology 2002; 36 (suppl 1): 16. 215. Chao YC, Young TH, Tang HS, Hsu CT. Alcoholism and alcoholic organ damage and genetic polymorphisms of alcohol metabolizing enzymes in Chinese patients. Hepatology 1997; 25:112–117. 216. Chao YC, Wang LS, Hsieh TY, et al. Chinese alcoholic patients with esophageal cancer are genetically different from alcoholics with acute pancreatitis and liver cirrhosis. Am J Gastroenterol 2000; 95:2958–2964. 217. Tanaka F, Shiratori Y, Yokosuka F, et al. High incidence of ADH2*1/ALDH2*1 genes among Japanese alcohol dependents and patients with alcoholic liver disease. Hepatology 1996; 23:234–239. 218. Osier M, Pakstis AJ, Kidd JR, et al. Linkage disequilibrium at the ADH2 and ADH3 loci and risk of alcoholism. Am J Hum Genet 1999; 64:1147–1157. 219. Borras E, Coutelle C, Rosell A, et al. Genetic polymorphism of alcohol dehydrogenase in Europeans: the ADH2*2 allele decreases the risk for alcoholism and is associated with ADH3*1. Hepatology 2000; 31:984–989. 220. Grove J, Brown AS, Daly AK, et al. The RsaI polymorphism of CYP2E1 and susceptibility to alcoholic liver disease in Caucasians: effect on age of presentation and dependence on alcohol dehydrogenase genotype. Pharmacogenetics 1998; 8:335–342. 221. Grove J, Daly AK, Burt AD, et al. Heterozygotes for HFE mutations have no increased risk of advanced alcoholic liver disease. Gut 1998; 43:262–266. 222. Degoul F, Sutton A, Mansouri A, et al. Homozygosity for alanine in the mitochondrial targeting sequence of superoxide dismutase and risk for severe alcoholic liver disease. Gastroenterology 2001; 120:1468–1474. 223. Stewart SF, Leathart JB, Chen Y, et al. Valine-alanine manganese superoxide dismutase polymorphism is not associated with alcohol-induced oxidative stress or liver fibrosis. Hepatology 2002; 36:1355–1360. 224. Baldini M, Lohman IC, Halonen M, et al. A polymorphism* in the 5¢ flanking region of the CD14 gene is associated with circulating soluble CD14 levels and with total serum immunoglobulin E. Am J Respir Cell Mol Biol 1999; 20:976–983. 225. Leathart JB, Day CP, Daly AK. No association between functional SNPs in the endotoxin receptors CD14 and TLR4 and alcoholic liver disease (ALD): is endotoxin important in the pathogenesis of ALD in humans? Hepatology 2001; 34:459A. 226. Arbour NC, Lorenz E, Schutte BC, et al. TLR4 mutations are associated with endotoxin hyporesponsiveness in humans. Nat Genet 2000; 25:187–191. 227. Grove J, Daly AK, Bassendine MF, Day CP. Association of a tumor necrosis factor promoter polymorphism with
618
228.
229. 230.
231.
232. 233. 234.
235.
236.
237.
238.
239.
240.
241.
242. 243.
244.
245.
246. 247.
248.
249. 250.
susceptibility to alcoholic steatohepatitis. Hepatology 1997; 26:143–146. Grove J, Daly AK, Bassendine MF, et al. Interleukin 10 promoter region polymorphisms and susceptibility to advanced alcoholic liver disease [see comment]. Gut 2000; 46:540–545. Waterhouse P, Marengere LE, Mittrucker HW, Mak TW. CTLA4, a negative regulator of T-lymphocyte activation. Immunol Rev 1996; 153:183–207. Ueda H, Howson JM, Epsosito L, et al. Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature 2003; 423:506–511. Valenti L, De Feo T, Fracanzani AL, et al. Cytotoxic Tlymphocyte antigen-4 A49G polymorphism is associated with susceptibility to and severity of alcoholic liver disease in Italian patients. Alcohol Alcohol 2004; 39:276–280. Baptista A. Alcoholic liver disease: morphological manifestations. Review by an international group. Lancet 1981; 1:707–711. MacSween RN, Burt AD. Histologic spectrum of alcoholic liver disease. Semin Liver Dis 1986; 6:221–232. Levin DM, Baker AL, Riddell RH, et al. Nonalcoholic liver disease. Overlooked causes of liver injury in patients with heavy alcohol consumption. Am J Med 1979; 66:429–434. Morgan MY, Sherlock S, Scheuer PJ. Acute cholestasis, hepatic failure, and fatty liver in the alcoholic. Scand J Gastroenterol 1978; 13:299–303. Wu A, Slavin G, Levi AJ. Elevated serum gamma-glutamyltransferase (transpeptidase) and histological liver damage in alcoholism. Am J Gastroenterol 1976; 65:318–323. Rosalki SB, Tarlow D, Rau D. Plasma gamma-glutamyl transpeptidase elevation in patients receiving enzyme-inducing drugs. Lancet 1971; 2:376–377. Drum DE, Goldman PA, Jankowski CB. Elevation of serum uric acid as a clue to alcohol abuse. Arch Intern Med 1981; 141:477–479. Storey EL, Anderson GJ, Mack U, et al. Desialylated transferrin as a serological marker of chronic excessive alcohol ingestion. Lancet 1987; 1:1292–1294. Buffet C, Chaput JC, Albuisson F, et al. [Macrocytosis in chronic alcoholic hepatitis (author’s translation).] Arch Fr Maladies Appareil Digest 1975; 64:309–315. Skude G, Wadstein J. Amylase, hepatic enzymes and bilirubin in serum of chronic alcoholics. Acta Med Scand 1977; 201: 53–58. Clermont RJ, Chalmers TC. The transaminase tests in liver disease. Medicine 1967; 46:197–207. Angulo P, Keach JC, Batts KP, Lindor KD. Independent predictors of liver fibrosis in patients with nonalcoholic steatohepatitis. Hepatology 1999; 30:1356–1362. Bailey RJ, Krasner N, Eddleston AL, et al. Histocompatibility antigens, autoantibodies, and immunoglobulins in alcoholic liver disease. Br Med J 1976; 2:727–729. Bell H, Raknerud N, Orjasaeter H, Haug E. Serum procollagen III peptide in alcoholic and other chronic liver diseases. Scand J Gastroenterol 1989; 24:1217–1222. Bruguera M, Bordas JM, Rodes J. Asymptomatic liver disease in alcoholics. Arch Pathol Lab Med 1977; 101:644–647. Van Waes L, Lieber CS. Early perivenular sclerosis in alcoholic fatty liver: an index of progressive liver injury. Gastroenterology 1977; 73:646–650. Uchida T, Kao H, Quispe-Sjogren M, Peters RL. Alcoholic foamy degeneration – a pattern of acute alcoholic injury of the liver. Gastroenterology 1983; 84:683–692. Rubin E, Lieber CS. Alcohol-induced hepatic injury in nonalcoholic volunteers. N Engl J Med 1968; 278:869–876. Devenyi P, Rutherdale J, Sereny G, Olin JS. Clinical diagnosis of alcoholic fatty liver. Am J Gastroenterol 1970; 54:597– 602.
Chapter 29 ALCOHOLIC LIVER DISEASE
251. Leevy CM. Fatty liver: a study of 270 patients with biopsy proven fatty liver and a review of the literature. Medicine 1962; 41:249–276. 252. Hislop WS, Bouchier IA, Allan JG, et al. Alcoholic liver disease in Scotland and northeastern England: presenting features in 510 patients. Q J Med 1983; 52:232–243. 253. Hardison WG, Lee FI. Prognosis in acute liver disease of the alcoholic patient. N Engl J Med 1966; 275:61–66. 254. Mendenhall CL. Alcohol-related diseases in gastroenterology. Berlin: Springer Verlag; 1985. 255. Gluud C, Henriksen JH, Nielsen G. Prognostic indicators in alcoholic cirrhotic men. Hepatology 1988; 8:222–227. 256. Orrego H, Israel Y, Blake JE, Medline A. Assessment of prognostic factors in alcoholic liver disease: toward a global quantitative expression of severity. Hepatology 1983; 3:896–905. 257. Carithers RL Jr, Herlong HF, Diehl AM, et al. Methylprednisolone therapy in patients with severe alcoholic hepatitis. A randomized multicenter trial. Ann Intern Med 1989; 110:685–690. 258. Forrest EH, Morris JM, Stewart SF, Day CP. The Glasgow Alcoholic Hepatitis Score identifies patients likely to benefit from corticosteroids. Gut 2004; 53 (suppl 3):A13. 259. Alexander JF, Lischner MW, Galambos JT. Natural history of alcoholic hepatitis. II. The long-term prognosis. Am J Gastroenterol 1971; 56:515–525. 260. Goldberg S, Mendenhall C, Anderson S, et al. VA cooperative study on alcoholic hepatitis. IV. The significance of clinically mild alcoholic hepatitis – describing the population with minimal hyperbilirubinemia. Am J Gastroenterol 1986; 81:1029–1034. 261. Pares A, Caballeria J, Bruguera M, et al. Histological course of alcoholic hepatitis. Influence of abstinence, sex and extent of hepatic damage. J Hepatol 1986; 2:33–42. 262. Rubin E, Kus S, Popper H. Pathogenesis of post-necrotic cirrhosis in alcoholics. Arch Pathol 1962; 73:288–299. 263. Gines P, Quintero E, Arroyo V, et al. Compensated cirrhosis: natural history and prognostic factors. Hepatology 1987; 7:122–128. 264. Mendenhall CL, Gartside PS, Roselle GA, et al. Longevity among ethnic groups in alcoholic liver disease. Alcohol Alcoholism 1989; 24:11–19. 265. Saunders JB, Walters JR, Davies AP, Paton A. A 20-year prospective study of cirrhosis. Br Med J Clin Res Ed 1981; 282:263–266. 266. D’Amico G, Morabito A, Pagliaro L, Marubini E. Survival and prognostic indicators in compensated and decompensated cirrhosis. Digest Dis Sci 1986; 31:468–475. 267. Powell WJ Jr, Klatskin G. Duration of survival in patients with Laennec’s cirrhosis. Influence of alcohol withdrawal, and possible effects of recent changes in general management of the disease. Am J Med 1968; 44:406–420. 268. Orrego H, Blake JE, Blendis LM, Medline A. Prognosis of alcoholic cirrhosis in the presence and absence of alcoholic hepatitis. Gastroenterology 1987; 92:208–214. 269. Lee FI. Cirrhosis and hepatoma in alcoholics. Gut 1966; 7:77–85. 270. Pessione F, Ramond MJ, Peters L, et al. Five-year survival predictive factors in patients with excessive alcohol intake and cirrhosis. Effect of alcoholic hepatitis, smoking and abstinence. Liver Int 2003; 23:45–53. 271. Scott J, Burns J, Flower EA. Histological analysis of parotid and submandibular glands in chronic alcohol abuse: a necropsy study. J Clin Pathol 1988; 1:837–840. 272. Dutta SK, Dukehart M, Narang A, Latham PS. Functional and structural changes in parotid glands of alcoholic cirrhotic patients. Gastroenterology 1989; 96:510–518.
273. Larato DC. Oral tissue changes in the chronic alcoholic. J Periodontol 1972; 43:772–773. 274. Erisen L, Basut O, Tezel I, et al. Regional epidemiological features of lip, oral cavity, and oropharyngeal cancer. J Environ Pathol Toxicol Oncol 1996; 15:225–229. 275. Vakevainen S, Tillonen J, Agarwal DP, et al. High salivary acetaldehyde after a moderate dose of alcohol in ALDH2deficient subjects: strong evidence for the local carcinogenic action of acetaldehyde. Clin Exp Res 2000; 24:873–877. 276. Hogan WJ, Viegas de Andrade SR, Winship DH. Ethanolinduced acute esophageal motor dysfunction. J Appl Physiol 1972; 32:755–760. 277. Knauer CM. Mallory–Weiss syndrome. Characterization of 75 Mallory–Weiss lacerations in 528 patients with upper gastrointestinal hemorrhage. Gastroenterology 1976; 71:5–8. 278. Launoy G, Milan CH, Faivre J, et al. Alcohol, tobacco and oesophageal cancer: effects of the duration of consumption, mean intake and current and former consumption. Br J Cancer 1997; 75:1389–1396. 279. Teyssen S, Gonzalez-Calero G, Schimiczek M, Singer MV. Maleic acid and succinic acid in fermented alcoholic beverages are the stimulants of gastric acid secretion. J Clin Invest 1999; 103:707–713. 280. Hauge T, Persson J, Kjerstadius T. Helicobacter pylori, active chronic antral gastritis, and gastrointestinal symptoms in alcoholics. Alcoholism Clin Exp Res 1994; 18:886–888. 281. Piper DW, Nasiry R, McIntosh J, et al. Smoking, alcohol, analgesics, and chronic duodenal ulcer. A controlled study of habits before first symptoms and before diagnosis. Scand J Gastroenterol 1984; 19:1015–1021. 282. Keshavarzian A, Iber FL, Dangleis MD, Cornish R. Intestinaltransit and lactose intolerance in chronic alcoholics. Am J Clin Nutr 1986; 44:70–76. 283. Rubin E, Rybak BJ, Lindenbaum J, et al. Ultrastructural changes in the small intestine induced by ethanol. Gastroenterology 1972; 63:801–814. 284. Hermos JA, Adams WH, Liu YK, et al. Mucosa of the small intestine in folate-deficient alcoholics. Ann Intern Med 1972; 76:957–965. 285. Perlow W, Baraona E, Lieber CS. Symptomatic intestinal disaccharidase deficiency in alcoholics. Gastroenterology 1977; 72:680–684. 286. Robinson GM, Orrego H, Israel Y, et al. Low-molecular-weight polyethylene glycol as a probe of gastrointestinal permeability after alcohol ingestion. Digest Dis Sci 1981; 26:971–977. 287. Buell MG, Beck IT. Effect of ethanol on jejunal regional blood flow in the rabbit. Gastroenterology 1983; 84:81–89. 288. Seal CJ, Ford CL, Day CP. Alcoholism: effects on nutritional status. In: Fallon N, ed. Encyclopedia of human nutrition. London: Academic Press; 1998. 289. Bouchoucha M, Nalpas B, Berger M, et al. Recovery from disturbed colonic transit time after alcohol withdrawal. Dis Colon Rectum 1991; 34:111–114. 290. Wilson JS, Pirola RC. The drinker’s pancreas: molecular mechanisms emerge. Gastroenterology 1997; 113:355–358. 291. Braganza JM. The pathogenesis of chronic pancreatitis. Q J Med 1996; 89:243–250. 292. Kloppel G, Maillet B. Chronic pancreatitis: evolution of the disease. HepatoGastroenterology 1991; 38:408–412. 293. Mukamal KJ, Conigrave KM, Mittleman MA, et al. Roles of drinking pattern and type of alcohol consumed in coronary heart disease in men. N Engl J Med 2003; 348:109–118. 294. Marmot M, Brunner E. Alcohol and cardiovascular disease: the status of the U shaped curve. Br Med J 1991; 303:565–568. 295. Ashley MJ, Rankin JG. Hazardous alcohol consumption and diseases of the circulatory system. J Stud Alcohol 1980; 41:1040–1070.
619
Section IV. Toxin Mediated Liver Injury
296. Saunders JB, Beevers DG, Paton A. Alcohol-induced hypertension. Lancet 1981; 2:653–656. 297. Puddey IB, Vandongen R, Beilin LJ. Pressor effect of alcohol. Lancet 1985; 2:1119–1120. 298. MacMahon SW, Norton RN. Alcohol and hypertension: implications for prevention and treatment. Ann Intern Med 1986; 105:124–126. 299. Klatsky AL, Friedman GD, Armstrong MA. The relationships between alcoholic beverage use and other traits to blood pressure: a new Kaiser Permanente study. Circulation 1986; 73:628–636. 300. Paulin JM, Simpson FO, Waal-Manning HJ. Alcohol consumption and blood pressure in a New Zealand community study. NZ Med J 1985; 98:425–428. 301. Rimm EB, Giovannucci EL, Willett WC, et al. Prospective study of alcohol consumption and risk of coronary disease in men. Lancet 1991; 338:464–468. 302. Lazarus NB, Kaplan GA, Cohen RD, Leu DJ. Change in alcohol consumption and risk of death from all causes and from ischaemic heart disease. Br Med J 1991; 303:553–556. 303. Friedman GD, Klatsky AL. Is alcohol good for your health? N Engl J Med 1993; 329:1882–1883. 304. Gaziano JM, Buring JE, Breslow JL, et al. Moderate alcohol intake, increased levels of high-density lipoprotein and its subfractions, and decreased risk of myocardial infarction. N Engl J Med 1993; 329:1829–1834. 305. Haut MJ, Cowan DH. The effect of ethanol on hemostatic properties of human blood platelets. Am J Med 1974; 56:22–33. 306. Rubin R. Ethanol interferes with collagen-induced platelet activation by inhibition of arachidonic acid mobilization. Arch Biochem Biophys 1989; 270:99–113. 307. Taylor JR. Alcohol and strokes. N Engl J Med 1982; 306:1111. 308. Stampfer MJ, Colditz GA, Willett WC, et al. A prospective study of moderate alcohol consumption and the risk of coronary disease and stroke in women. N Engl J Med 1988; 319:267–273. 309. Wilkins MR, Kendall MJ. Stroke affecting young men after alcoholic binges. Br Med J (Clin Res Ed) 1985; 291:1342. 310. Donahue RP, Abbott RD, Reed DM, Yano K. Alcohol and hemorrhagic stroke. The Honolulu Heart Program. JAMA 1986; 255:2311–2314. 311. Schenk EA, Cohen J. The heart in chronic alcoholism. Clinical and pathologic findings. Pathol Microbiol (Basel) 1970; 35:96–104. 312. Urbano-Marquez A, Estruch R, Navarro-Lopez F, et al. The effects of alcoholism on skeletal and cardiac muscle. N Engl J Med 1989; 320:409–415. 313. Jacob AJ, McLaren KM, Boon NA. Effects of abstinence on alcoholic heart muscle disease. Am J Cardiol 1991; 68:805–807. 314. Koskinen P, Kupari M. Alcohol and cardiac arrhythmias. Br Med J 1992; 304:1394–1395. 315. Cohen EJ, Klatsky AL, Armstrong MA. Alcohol use and supraventricular arrhythmia. Am J Cardiol 1988; 62:971–973. 316. Greenspon AJ, Schaal SF. The “holiday heart”: electrophysiologic studies of alcohol effects in alcoholics. Ann Intern Med 1983; 98:135–139. 317. Wannamethee G, Shaper AG. Alcohol and sudden cardiac death. Br Heart J 1992; 68:443–448. 318. Day CP, James OF, Butler TJ, Campbell RW. QT prolongation and sudden cardiac death in patients with alcoholic liver disease [comment]. Lancet 1993; 341:1423–1428. 319. Butterworth RF. Pathophysiology of alcoholic brain damage: synergistic effects of ethanol, thiamine deficiency and alcoholic liver disease. Metab Brain Dis 1995; 10:1–8. 320. Charness ME. Brain lesions in alcoholics. Alcohol Clin Exp Res 1993; 17:2–11.
620
321. Orgogozo JM, Dartigues JF, Lafont S, et al. Wine consumption and dementia in the elderly: a prospective community study in the Bordeaux area. Rev Neurol (Paris) 1997; 153:185–192. 322. Duncan G, Johnson RH, Lambie DG, Whiteside EA. Evidence of vagal neuropathy in chronic alcoholics. Lancet 1980; 2:1053–1057. 323. Barter F, Tanner AR. Autonomic neuropathy in an alcoholic population. Postgrad Med J 1987; 63:1033–1036. 324. Thuluvath PJ, Triger DR. Autonomic neuropathy and chronic liver disease. Q J Med 1989; 72:737–747. 325. Mohamed R, Forsey PR, Davies MK, Neuberger JM. Effect of liver transplantation on QT interval prolongation and autonomic dysfunction in end-stage liver disease. Hepatology 1996; 23:1128–1134. 326. Abel EL, Sokol RJ. A revised conservative estimate of the incidence of FAS and its economic impact. Alcohol Clin Exp Res 1991; 15:514–524. 327. Longnecker MP, Enger SM. Epidemiologic data on alcoholic beverage consumption and risk of cancer. Clin Chim Acta 1996; 246:121–141. 328. Smith-Warner SA, Spiegelman D, Yaun SS, et al. Alcohol and breast cancer in women: a pooled analysis of cohort studies. JAMA 1998; 279:535–540. 329. Hamajima N, Hirose K, Tajima K, et al. Alcohol, tobacco and breast cancer – collaborative reanalysis of individual data from 53 epidemiological studies, including 58 515 women with breast cancer and 95 067 women without the disease. Br J Cancer 2002; 87:1234–1245. 330. Zieve L. Jaundice, hyperlipemia and hemolytic anemia: a heretofore unrecognized syndrome associated with alcoholic fatty liver and cirrhosis. Ann Intern Med 1958; 48:471–496. 331. Nelson S, Kolls JK. Alcohol, host defense and society. Nature Reviews. Immunology 2002; 2:205–209. 332. Longnecker MP. Alcohol consumption and risk of cancer in humans: an overview. Alcohol 1995; 12:87–96. 333. Patel M, Keshavarzian A, Kottapalli V, et al. Human neutrophil functions are inhibited in vitro by clinically relevant ethanol concentrations. Alcoholism Clin Exp Res 1996; 20:275–283. 334. Stoltz DA, Zhang P, Nelson S, et al. Ethanol suppression of the functional state of polymorphonuclear leukocytes obtained from uninfected and simian immunodeficiency virus infected rhesus macaques. Alcoholism Clin Exp Res 1999; 23:878–884. 335. Hill DB, Devalaraja R, Joshi-Barve S, et al. Antioxidants attenuate nuclear factor-kappa B activation and tumor necrosis factor-alpha production in alcoholic hepatitis patient monocytes and rat Kupffer cells, in vitro. Clin Biochem 1999; 32:563–570. 336. McClain CJ, Hill DB, Song Z, et al. Monocyte activation in alcoholic liver disease. Alcohol 2002; 27:53–61. 337. Silvain C, Patry C, Launay P, et al. Altered expression of monocyte IgA Fc receptors is associated with defective endocytosis in patients with alcoholic cirrhosis. Potential role for IFN-gamma. J Immunol 1995; 155:1606–1618. 338. Szabo G, Verma B, Catalano D. Selective inhibition of antigen-specific T lymphocyte proliferation by acute ethanol exposure: the role of impaired monocyte antigen presentation capacity and mediator production. J Leukocyte Biol 1993; 54:534–544. 339. Ewald SJ, Shao H. Ethanol increases apoptotic cell death of thymocytes in vitro. Alcoholism Clin Exp Res 1993; 17:359–365. 340. Pruett SB, Han YC, Wu WJ. A brief review of immunomodulation caused by acute administration of ethanol: involvement of neuroendocrine pathways. Alcohol Alcoholism 1994; 2 (suppl):431–437. 341. Glassman AB, Bennett CE, Randall CL. Effects of ethyl alcohol on human peripheral lymphocytes. Arch Pathol Lab Med 1985; 109:540–542.
Chapter 29 ALCOHOLIC LIVER DISEASE
342. Lindenbaum J, Hargrove RL. Thrombocytopenia in alcoholics. Ann Intern Med 1968; 68:526–532. 343. Van Thiel DH, Lester R, Vaitukaitis J. Evidence for a defect in pituitary secretion of luteinizing hormone in chronic alcoholic men. J Clin Endocrinol Metab 1978; 47:499–507. 344. Frajria R, Angeli A. Alcohol-induced pseudo-Cushing’s syndrome. Lancet 1977; 1:1050–1051. 345. Peris P, Guanabens N, Pares A, et al. Vertebral fractures and osteopenia in chronic alcoholic patients. Calcif Tissue Int 1995; 57:111–114. 346. Peris P, Pares A, Guanabens N, et al. Bone mass improves in alcoholics after 2 years of abstinence. J Bone Miner Res 1994; 9:1607–1612. 347. Royal College of Physicians. Alcohol – can the NHS afford it? Royal College of Physicians Working Party Report, 2001; 348. Wilk AI, Jensen NM, Havighurst TC. Meta-analysis of randomized control trials addressing brief interventions in heavy alcohol drinkers. J Gen Intern Med 1997; 12: 274–283. 349. Kadden RM. Project MATCH: treatment main effects and matching results. Alcoholism Clin Exp Res 1996; 20 (suppl):196A–197A. 350. O’Malley SS. Kaffe G, Chang RS, et al. Naltrexone and coping skills therapy for alcohol dependence. A controlled study. Arch Gen Psychiatry 1992; 49:881–887. 351. Sass H, Soyka M, Mann K, Zieglgansberger W. Relapse prevention by acamprosate. Results from a placebo-controlled study on alcohol dependence. Arch Gen Psychiatry 1996; 53:673–680. 352. Garbutt JC, West SL, Carey TS, et al. Pharmacological treatment of alcohol dependence: a review of the evidence. JAMA 1999; 281:1318–1325. 353. Delgrange T, Khater J, Capron D, et al. [Effect of acute administration of acamprosate on the risk of encephalopathy and on arterial pressure in patients with alcoholic cirrhosis.] Gastroenterol Clin Biol 1992; 16:687–691. 354. Chick J, Gough K, Falkowski W, et al. Disulfiram treatment of alcoholism. Br J Psychiatry 1992; 161:84–89. 355. Fuller RK, Branchey L, Brightwell DR, et al. Disulfiram treatment of alcoholism. A Veterans Administration cooperative study. JAMA 1986; 256:1449–1455. 356. Peachey JE. A review of the clinical use of disulfiram and calcium carbimide in alcoholism treatment. J Clin Psychopharmacol 1981; 1:368–375. 357. Wodak AD, Saunders JB, Ewusi-Mensah I, et al. Severity of alcohol dependence in patients with alcoholic liver disease. Br Med J Clin Res Ed 1983; 287:1420–1422. 358. Day CP. Moderate alcohol intake is not deleterious in patients with alcoholic liver disease. Hepatology 1996; 24:443A. 359. Maddrey WC, Boitnott JK, Bedine MS, et al. Corticosteroid therapy of alcoholic hepatitis. Gastroenterology 1978; 75:193–199. 360. Ramond MJ, Poynard T, Rueff B, et al. A randomized trial of prednisolone in patients with severe alcoholic hepatitis. N Engl J Med 1992; 326:507–512. 361. Ray A, Prefontaine KE. Physical association and functional antagonism between the p65 subunit of transcription factor NFkappa B and the glucocorticoid receptor. Proc Natl Acad Sci USA 1994; 91:752–756. 362. Ghosh S. Regulation of inducible gene expression by the transcription factor NF-kappaB. Immunol Res 1999; 19:183–189. 363. Reynolds T, Benhamon JP, Blake JE, et al. Treatment of alcoholic hepatitis. Gastroenterol Int 1989; 2:208–216. 364. Imperiale TF, McCullough AJ. Do corticosteroids reduce mortality from alcoholic hepatitis? A meta-analysis of the randomized trials. Ann Intern Med 1990; 113:299– 307.
365. Daures JP, Peray P, Bories P, et al. [Corticoid therapy in the treatment of acute alcoholic hepatitis. Results of a metaanalysis.] Gastroenterol Clin Biol 1991; 15:223–228. 366. Christensen E, Gluud C. Glucocorticosteroids are not effective in alcoholic hepatitis. Am J Gastroenterol 1999; 94:3065–3066. 367. Mathurin P, Mendenhall CL, Carithers RL, et al. Corticosteroids improve short-term survival in patients with severe alcoholic hepatitis (AH): individual data analysis of the last three randomized placebo controlled double blind trials of corticosteroids in severe AH. J Hepatol 2002; 36:480–487. 368. Mathurin P, Abdelnour M, Ramond MJ, et al. Early change in bilirubin levels is an important prognostic factor in severe alcoholic hepatitis treated with prednisolone. Hepatology 2003; 38:1363–1369. 369. Strieter RM, Remick DG, Ward PA, et al. Cellular and molecular regulation of tumor necrosis factor-alpha production by pentoxifylline. Biochem Biophys Res Commun 1988; 155:1230–1236. 370. Akriviadis E, Botla R, Briggs W, et al. Pentoxifylline improves short-term survival in severe acute alcoholic hepatitis: a doubleblind, placebo-controlled trial. Gastroenterology 2000; 119:1637–1648. 371. Mutimer DJ, Burra P, Neuberger JM, et al. Managing severe alcoholic hepatitis complicated by renal failure. Q J Med 1993; 86:649–656. 372. Mendenhall CL, Anderson S, Weesner RE, et al. Protein–calorie malnutrition associated with alcoholic hepatitis. Veterans Administration cooperative study group on alcoholic hepatitis. Am J Med 1984; 76:211–222. 373. Nasrallah SM, Galambos JT. Aminoacid therapy of alcoholic hepatitis. Lancet 1980; 2:1276–1277. 374. Diehl AM, Boitnott JK, Herlong HF, et al. Effect of parenteral amino acid supplementation in alcoholic hepatitis. Hepatology 1985; 5:57–63. 375. Cabre E, Gonzalez-Huix F, Abad-Lacruz A, et al. Effect of total enteral nutrition on the short-term outcome of severely malnourished cirrhotics. A randomized controlled trial. Gastroenterology 1990; 98:715–720. 376. Cabre E, Rodriguez-Iglesias P, Caballeria J, et al. Short- and long-term outcome of severe alcohol-induced hepatitis treated with steroids or enteral nutrition: a multicenter randomized trial. Hepatology 2000; 32:36–42. 377. Wenzel G, Kuklinski B, Ruhlmann C, Ehrhardt D. [Alcoholinduced toxic hepatitis – a “free radical” associated disease. Lowering fatality by adjuvant antioxidant therapy.] Z Gesamte Inn Med Grenzgebiete 1993; 48:490–496. 378. Phillips M, Curtis H, Portmann B, et al. Antioxidants versus corticosteroids in the treatment of severe alcoholic hepatitis: a randomised trial. Hepatology 2001; 34:250A. 379. Fang JW, Bird GL, Nakamura T, et al. Hepatocyte proliferation as an indicator of outcome in acute alcoholic hepatitis. Lancet 1994; 343:820–823. 380. Simek J, Chmelar V, Melka J, et al. Influence of protracted infusion of glucose and insulin on the composition and regeneration activity of liver after partial hepatectomy in rats. Nature 1967; 213:910–911. 381. Farivar M, Wands JR, Isselbacher KJ, Bucher NL. Effect of insulin and glucagon on fulminant murine hepatitis. N Engl J Med 1976; 295:1517–1519. 382. Feher J, Cornides A, Romany A, et al. A prospective multicenter study of insulin and glucagon infusion therapy in acute alcoholic hepatitis. J Hepatol 1987; 5:224–231. 383. Trinchet JC, Balkau B, Poupon RE, et al. Treatment of severe alcoholic hepatitis by infusion of insulin and glucagon: a multicenter sequential trial. Hepatology 1992; 15:76–81. 384. Bird G, Lau JY, Koskinas J, et al. Insulin and glucagon infusion in acute alcoholic hepatitis: a prospective randomized controlled trial. Hepatology 1991; 14:1097–1101.
621
Section IV. Toxin Mediated Liver Injury
385. Rambaldi A, Iaquinto G, Gluud C. Anabolic-androgenic steroids for alcoholic liver disease: a Cochrane review. Am J Gastroenterol 2002; 97:1674–1681. 386. Israel Y, Kalant H, Orrego H, et al. Experimental alcoholinduced hepatic necrosis: suppression by propylthiouracil. Proc Natl Acad Sci USA 1975; 72:1137–1141. 387. Orrego H, Kalant H, Israel Y, et al. Effect of short-term therapy with propylthiouracil in patients with alcoholic liver disease. Gastroenterology 1979; 76:105–115. 388. Halle P, Pare P, Kaptein E, et al. Double-blind, controlled trial of propylthiouracil in patients with severe acute alcoholic hepatitis. Gastroenterology 1982; 82:925–931. 389. Iimuro Y, Gallucci RM, Luster MI, et al. Antibodies to tumor necrosis factor alfa attenuate hepatic necrosis and inflammation caused by chronic exposure to ethanol in the rat. Hepatology 1997; 26:1530–1537. 390. Jalan R, Williams R, Kaser R, et al. Clinical and cytokine response to anti-TNF antibody therapy in severe alcoholic hepatitis. Hepatology 2001; 34:441A. 391. Spahr L, Rubbia-Brandt L, Frossard JL, et al. Combination of steroids with infliximab or placebo in severe alcoholic hepatitis: a randomized controlled pilot study. J Hepatol 2002; 37:448–455. 392. Menon KV, Stadheim L, Kamath PS, et al. A pilot study of the safety and tolerability of etanercept in patients with alcoholic hepatitis. Am J Gastroenterol 2004; 99:255–260. 393. Keane J, Gershon S, Wise RP, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med 2001; 345:1098–1104. 394. Akerman P, Cote P, Yang SQ, et al. Antibodies to tumor necrosis factor-alpha inhibit liver regeneration after partial hepatectomy. Am J Physiol 1992; 263:G579–G585. 395. Naveau S, Chollet-Martin S, Dharancy S, et al. A double-blind randomized controlled trial of infliximab associated with prednisolone in acute alcoholic hepatitis [see comment]. Hepatology 2004; 39:1390–1397. 396. Jalan R, Sen S, Steiner C, et al. Extracorporeal liver support with molecular adsorbents recirculating system in patients with severe acute alcoholic hepatitis [see comment]. J Hepatol 2003; 38:24–31. 397. Sen S, Mookerjee RP, Cheshire LM, et al. Albumin dialysis reduces portal pressure acutely in patients with severe alcoholic hepatitis. J Hepatol 2005; 43:42–48. 398. Ortega R, Gines P, Uriz J, et al. Terlipressin therapy with and without albumin for patients with hepatorenal syndrome: results of a prospective, nonrandomized study. Hepatology 2002; 36:941–948. 399. Uriz J, Gines P, Cardenas A, et al. Terlipressin plus albumin infusion: an effective and safe therapy of hepatorenal syndrome. J Hepatol 2000; 33:43–48. 400. Guevara M, Gines P, Fernandez-Esparrach G, et al. Reversibility of hepatorenal syndrome by prolonged administration of ornipressin and plasma volume expansion. Hepatology 1998; 27:35–41. 401. Orrego H, Blake JE, Blendis LM, et al. Long-term treatment of alcoholic liver disease with propylthiouracil. N Engl J Med 1987; 317:1421–1427. 402. Ehrlich HP, Ross R, Bornstein P. Effects of antimicrotubular agents on the secretion of collagen. A biochemical and morphological study. J Cell Biol 1974; 62:390–405. 403. Diegelmann RF, Peterkofsky B. Inhibition of collagen secretion from bone and cultured fibroblasts by microtubular disruptive drugs. Proc Natl Acad Sci USA 1972; 69:892–896. 404. Kershenobich D, Uribe M, Suarez GI, et al. Treatment of cirrhosis with colchicine. A double-blind randomized trial. Gastroenterology 1979; 77:532–536. 405. Akriviadis EA, Steindel H, Pinto PC, et al. Failure of colchicine to improve short-term survival in patients with alcoholic hepatitis. Gastroenterology 1990; 99:811–818.
622
406. Trinchet JC, Beaugrand M, Callard P, et al. Treatment of alcoholic hepatitis with colchicine. Results of a randomized double blind trial. Gastroenterol Clin Biol 1989; 13:551– 555. 407. Cortez-Pinto H, Alexandrino P, Camilo ME, et al. Lack of effect of colchicine in alcoholic cirrhosis: final results of a double blind randomized trial. Eur J Gastroenterol Hepatol 2002; 14:377–381. 408. Rambaldi A, Gluud C. Colchicine for alcoholic and nonalcoholic liver fibrosis and cirrhosis. Cochrane Database of Systematic Reviews 2001; 3:002148. 409. Morgan TR, Weiss DG, Nemchausky B, et al. Colchicine treatment of alcoholic cirrhosis: a randomized, placebocontrolled clinical trial of patient survival. Gastroenterology 2005; 128:882–890. 410. Carini R, Comoglio A, Albano E, Poli G. Lipid peroxidation and irreversible damage in the rat hepatocyte model. Protection by the silybin–phospholipid complex IdB 1016. Biochem Pharmacol 1992; 43:2111–2115. 411. Pietrangelo A, Borella F, Casalgrandi G, et al. Antioxidant activity of silybin in vivo during long-term iron overload in rats. Gastroenterology 1995; 109:1941–1949. 412. Masini A, Ceccarelli D, Giovannini F, et al. Iron-induced oxidant stress leads to irreversible mitochondrial dysfunctions and fibrosis in the liver of chronic iron-dosed gerbils. The effect of silybin. J Bioenergetics Biomembranes 2000; 32:175–182. 413. Ferenci P, Dragosics B, Dittrich H, et al. Randomized controlled trial of silymarin treatment in patients with cirrhosis of the liver. J Hepatol 1989; 9:105–113. 414. Pares A, Planas R, Torres M, et al. Effects of silymarin in alcoholic patients with cirrhosis of the liver: results of a controlled, double-blind, randomized and multicenter trial. J Hepatol 1998; 28:615–621. 415. Mato JM, Camara J, Fernandez de Paz J, et al. S-adenosylmethionine in alcoholic liver cirrhosis: a randomized, placebo-controlled, double-blind, multicenter clinical trial. J Hepatol 1999; 30:1081–1089. 416. Lieber CS, Robins SJ, Li J, et al. Phosphatidylcholine protects against fibrosis and cirrhosis in the baboon. Gastroenterology 1994; 106:152–159. 417. Li J, Kim CI, Leo MA, et al. Polyunsaturated lecithin prevents acetaldehyde-mediated hepatic collagen accumulation by stimulating collagenase activity in cultured lipocytes. Hepatology 1992; 15:373–381. 418. Lieber CS. Alcoholic liver disease: new insights in pathogenesis lead to new treatments. J Hepatol 2000; 32 (suppl):113–128. 419. Lieber CS, Weiss DG, Groszmann R, et al. II. Veterans Affairs cooperative study of polyenylphosphatidylcholine in alcoholic liver disease. Alcoholism Clin Exp Res 2003; 27:1765–1772. 420. Starzl TE, Van Thiel D, Tzakis AG, et al. Orthotopic liver transplantation for alcoholic cirrhosis. JAMA 1988; 260:2542–2544. 421. European Liver Transplant Registry. www.ELTR.org. 2001. 422. Belle SH, Beringer KC, Detre KM. Liver transplantation for alcoholic liver disease in the United States: 1988 to 1995. Liver Transplant Surg 1997; 3:212–219. 423. Neuberger J, Schulz KH, Day C, et al. Transplantation for alcoholic liver disease. J Hepatol 2002; 36:130–137. 424. Wiesner RH, Lombardero M, Lake JR, et al. Liver transplantation for end-stage alcoholic liver disease: an assessment of outcomes. Liver Transplant Surg 1997; 3:231–239. 425. Bird GL, O’Grady JG, Harvey FA, et al. Liver transplantation in patients with alcoholic cirrhosis: selection criteria and rates of survival and relapse. Br Med J 1990; 301:15–17. 426. Burra P, De Bona M, Canova D, et al. Longitudinal prospective study on quality of life and psychological distress before and one
Chapter 29 ALCOHOLIC LIVER DISEASE
427.
428.
429.
430.
431.
432.
433.
434.
435.
year after liver transplantation. Acta Gastroenterol Belg 2005; 68:19–25. Propst A, Propst T, Zangerl G, et al. Prognosis and life expectancy in chronic liver disease. Digest Dis Sci 1995; 40:1805–1815. Poynard T, Barthelemy P, Fratte S, et al. Evaluation of efficacy of liver transplantation in alcoholic cirrhosis by a case-control study and simulated controls. Lancet 1994; 344:502–507. Lucey MR, Carr K, Beresford TP, et al. Alcohol use after liver transplantation in alcoholics: a clinical cohort follow-up study. Hepatology 1997; 25:1223–1227. Gerhardt TC, Goldstein RM, Urschel HC, et al. Alcohol use following liver transplantation for alcoholic cirrhosis. Transplant Proc 1996; 62:1060–1063. Berlakovich GA, Steininger R, Herst M, et al. Efficacy of liver transplantation for alcoholic cirrhosis with respect to recidivism and compliance. Transplant Proc 1994; 58:560–565. Kumar S, Stauber RE, Gavaler JS, et al. Orthotopic liver transplantation for alcoholic liver disease. Hepatology 1990; 11:159–164. Osorio RW, Ascher NL, Avery M, et al. Predicting recidivism after orthotopic liver transplantation for alcoholic liver disease. Hepatology 1994; 20:105–110. Pageaux GP, Michel J, Coste V, et al. Alcoholic cirrhosis is a good indication for liver transplantation, even for cases of recidivism [see comment]. Gut 1999; 45:421–426. Foster PF, Fabrega F, Karademir S, et al. Prediction of abstinence from ethanol in alcoholic recipients following liver transplantation. Hepatology 1997; 25:1469–1477.
436. Anand AC, Ferraz-Neto BH, Nightingale P, et al. Liver transplantation for alcoholic liver disease: evaluation of a selection protocol. Hepatology 1997; 25:1478–1484. 437. Burra P, Mioni D, Cecchetto A, et al. Histological features after liver transplantation in alcoholic cirrhotics. J Hepatol 2001; 34:716–722. 438. Cuadrado A, Fabrega E, Casafront F, Pons-Romero F. Alcohol recidivism impairs long-term patient survival after orthotopic liver transplantation for alcoholic liver disease. Liver Transplant 2005; 11:420–426. 439. Neuberger J, Adams D, MacMaster P, et al. Assessing priorities for allocation of donor liver grafts: survey of public and clinicians. Br Med J 1998; 317:172–175. 440. Miguet M, Monnet E, Vanlemmens C, et al. Predictive factors of alcohol relapse after orthotopic liver transplantation for alcoholic liver disease. Gastroenterol Clin Biol 2004; 28:845–851. 441. Veldt BJ, Laine F, Guillygomar’h A, et al. Indication of liver transplantation in severe alcoholic liver cirrhosis: quantitative evaluation and optimal timing. J Hepatol 2002; 36:93–98. 442. Lucey MR. Is liver transplantation an appropriate treatment for acute alcoholic hepatitis? J Hepatol 2002; 36:829–831. 443. Shakil AO, Pinna A, Demetris J, et al. Survival and quality of life after liver transplantation for acute alcoholic hepatitis. Liver Transplant Surg 1997; 3:240–244. 444. Tome S, Martinez-Rey C, Gonzalez-Quintela A, et al. Influence of superimposed alcoholic hepatitis on the outcome of liver transplantation for end-stage alcoholic liver disease [see comment]. J Hepatol 2002; 36:793–798.
623
Section V. Liver Diseases Due to Infectious Agents
30
HEPATITIS A Maria H. Sjogren Abbreviations 5¢ NTR 5¢ non-translated anti-HAV antibodies against HAV FHF fulminant hepatic failure
HAV HBV
hepatitis A virus hepatitis B virus
INTRODUCTION Experimental work in humans begun in the 1940s led to the clinical recognition of viruses as etiological agents of hepatitis A (“infectious hepatitis”) and hepatitis B (“serum hepatitis”).1,2 These initial observations were later amplified demonstrating the existence of two hepatitis viruses: hepatitis A virus (HAV) and hepatitis B virus (HBV).3 The hepatitis A viral agent was first characterized in 1973, when scientists discovered the virus in human stools from volunteers who were infected with HAV.4 The subsequent development of sensitive and specific serological techniques for the diagnosis of HAV infection and the isolation of HAV in cell culture5 were important advances which permitted understanding of the epidemiology of HAV infection and ultimately protection against infection by the development of effective vaccines.
VIROLOGY Initially, HAV was classified as an enterovirus type 72 belonging to the picornaviridae family. Subsequent sequencing of HAV nucleotides and amino acid sequences questioned this classification and the new genus Hepatovirus was created for HAV.6 HAV is a non-enveloped virus that measures 27–28 nm in diameter, has a buoyant density of 1.33–1.34 g/cm3 in CsCl, and a sedimentation coefficient of 156–160 S by ultracentrifugation. Hepatitis A survives exposure to ether and an acid environment at pH 3. It also survives heat exposure at 60oC for 60 min, but is inactivated at 85oC for 1 min. Only one serotype of HAV is known which has no antigenic crossreactivity with the hepatitis B, C, D, E, or G agents. The HAV genome consists of a positive-sense RNA, 7.48 kb long, singlestranded, and linear. The HAV RNA has a sedimentation coefficient of 32–33 S and a molecular weight of 2.8 ¥ 104. The HAV RNA has a long open reading frame (6681 nucleotides) and is covalently linked to a 5¢ terminal protein and a 3¢ terminal poly-A tract. Onset of HAV replication in cell culture systems takes a long time, usually months. Primate cells are favored for HAV in-vitro cultivation, among them African green-monkey kidney cells, primary human fibroblasts, human diploid cells (MRC-5), and fetal rhesus kidney cells. The virus is not cytopathic and persistent infection of cell cultures is the rule. Two conditions control the outcome of HAV replication in cell culture.7 First, the genetic make-up of the virus is important, as HAV strains mutate in distinct regions of
IgM PCR
immunoglobulin M polymerase chain reaction
the viral genome, as they become cell-culture-adapted. Another important condition is the metabolic activity of the host cell at the time of infection. Cells in culture, although infected simultaneously, initiate HAV replication in an asynchronous manner, probably due to differences in the metabolic activity of individual cells, but there is no definitive evidence of cell cycle dependence of HAV replication.8 An initial step in the life cycle of a virus is its attachment to a cell surface receptor. Location and function of these receptors determine tissue tropism. Little is known about the mechanism of cell entry of HAV. Scientific work has suggested that HAV could infect cells by a surrogate receptor-binding mechanism (a non-specified serum protein). HAV infectivity in tissue culture was shown to require calcium and to be inhibited by treatment of the cells with trypsin, phospholipases, and b-galactosidase.9 A surface glycoprotein on African green-monkey kidney cells has been identified as a receptor for HAV. The receptor was named HAVcr-1. Blocking of HAVcr-1 with specific monoclonal antibodies prevented infection of otherwise susceptible cells. Whatever the entry mechanism, once HAV enters a cell, the viral RNA is uncoated, cell host ribosomes bind to viral RNA, and polysomes are formed. HAV is translated into a large polyprotein of 2227 amino acids. This polyprotein is organized into three regions: P1, P2, and P3. The P1 region encodes for structural proteins VP1, VP2, VP3, and a putative VP4. The P2 and P3 regions encode for non-structural proteins associated with viral replication. The viral RNA polymerase copies the plus RNA strand (genome) that in turn is used for translation into proteins and for assembly into mature virions. It appears that down-regulation of HAV RNA synthesis occurs as defective HAV particles appear.10 In addition a group of specific RNA-binding proteins have been observed during persistent infection.11 The origin and nature of these proteins are unknown, but they exert activity in the RNA template and are believed to play a regulatory role in the replication of HAV.12 Numerous strains of hepatitis A exist, with considerable nucleotide sequence variability (15–25% difference within the P1 region of the genome). Human HAV strains can be grouped in four different genotypes (I, II, III, and VII), while simian strains belong to genotypes IV, V, and VI.13 Despite the nucleotide sequence heterogeneity, the antigenic structure of human HAV is highly conserved among strains. A recent report suggested that the hepatitis A VP1/2A and 2C genes are responsible for virulence. Investigators studied the ability of recombinant HAV virus to cause acute hepatitis in experimental
627
Section V. Liver Diseases Due to Infectious Agents
Figure 30-1. The reported and estimated cases of acute hepatitis A have declined over the last 10 years, particularly after hepatitis A virus vaccine became commercially available in the USA.16
70000 Vaccine licensed 60000
Cases
50000 40000 30000 20000 10000 0 1966
1970
1974
1978
1982
1986
1990
1994
1998
2002
Year
animals by constructing 14 chimeric virus genomes from two infectious cDNA clones encoding a virulent and an attenuated HAV isolate (HM175 strain). Comparisons of the genotype and phenotype of each virus supported their findings.14 Among the many strains of HAV, the HM175 and CR326 human HAV strains are important because they were used for the production of commercially available vaccines; strain HM175 was isolated in 1978 from human feces from Australian patients who were members of a small hepatitis A outbreak. CR326 was isolated from Costa Rican patients infected with the HAV. The nucleotide and amino acid sequence show 95% identity between these two strains. Vaccines prepared with these strains are expected to provide protection against all relevant human strains. To study a possible viral contribution to the development of fulminant hepatic failure (FHF) during acute infection with HAV, the 5¢ non-translated (5¢ NTR) region of the HAV genome in 84 serum samples, including 12 FHF patients, were sequenced.15 The investigators observed relatively fewer nucleotide substitutions in FHF compared to non-FHF subjects with acute HAV infection (P < 0.001). This finding was most prominent between nucleotides 200 and 500, suggesting that this nucleotide variation in the central portion of the 5¢ NTR of HAV may influence the clinical severity of the infection.
EPIDEMIOLOGY Acute hepatitis A is a reportable infectious disease in the USA, with a rate of infection of 4/100 000.16 In the USA, the number of reported and estimated subjects infected with HAV has been decreasing, as shown in Figure 30-1.16 In the year 2003, 7653 HAV infections were reported; however, taking into consideration the underreporting of cases and asymptomatic infections, the true number of infections was calculated to be 61 000.16 The highest rate of reported disease has been among children 5–14 years of age and 25% of reported cases are among persons aged 20 years or less;17 however, HAV infection can occur in any age group. A recent communication from the Centers for Diseases Control and Prevention reported the current epidemiological risk factors for the US population, as shown in Figure 30-2.18
628
Sexual or household contact 12% International travel 9% Unknown 57%
MSM 8% IVDA 5% Day care 5%
Outbreak 1% Contact with patient 4% Figure 30-2. The epidemiological risks for acute hepatitis A virus infection have not changed significantly over the years. Data depicted in the graph were obtained from reported cases in the USA in 2002.18 MSM, men who have had sex with men; IVDA, intravenous drug abuse.
Hepatitis A infection generally follows one of three epidemiological patterns.19 First, in countries with poor sanitary conditions, most children are infected at an early age. Although earlier seroepidemiological studies routinely showed that 100% of preschool children had detectable anti-HAV, presumably reflecting previous subclinical infection, recent studies show that the average age of infection is rapidly increasing to 5 years and above, when symptomatic infection is more likely. As such, 82% of 1393 Bolivian school children were shown to have detectable anti-HAV in sera; however, when stratified in two groups according to family income, there was a
Chapter 30 HEPATITIS A
significant difference between the groups: 56% of children from high-income families had detectable anti-HAV versus 95% of children from low-income families.20 The second epidemiological pattern is seen in industrialized countries where there is a low prevalence of HAV infection among children and young adults. Thus, in the USA the prevalence of antiHAV is approximately 10% in children, whereas 37% of adults have detectable anti-HAV.21 The third epidemiological pattern is observed in closed or semi-closed communities, in which HAV is capable (through epidemics) of infecting the entire population, which thus becomes immune. Thereafter, newborns remain susceptible but free of anti-HAV until the virus is reintroduced into the community. The primary route of human transmission is fecal–oral by either person-to-person contact or ingestion of contaminated food or water. Although rare, transmission by parenteral route has been documented following blood transfusion22,23 or the use of blood products.24 Cyclic outbreaks among users of injecting and non-injecting drugs and among men who have sex with men (up to 10% in outbreak years) have been reported.25 In the USA the rate of antibody to HAV increases with age:25 9% of adolescents are found to have detectable antibody, while 75% of individuals over 70 years of age have evidence of remote infection with HAV.
nificant problem in adults and older children. Approximately 11–22% of patients25 with acute hepatitis A require hospitalization, with an average cost of $6914 per patient.28 In one outbreak involving 43 persons, the cost was approximately $800 000.25 On average 27 workdays are lost per adult case with a total loss of 829 000 workdays per year.25,28 Combined direct and indirect costs associated with hepatitis A infection in the USA totaled more than $200 million in 1989 and approximately $488.8 millions in 1997.15,29 The clinical characteristics of hepatitis A cases reported in 2002 were similar to previous years with a preponderance of cases among men rather than women in all ages, 72% of cases manifesting jaundice, 25% requiring hospitalization and 0.5% resulting in death.18 Mortality and the need for hospitalization increase with age (Table 30-1), and subjects older than 60 years of age are at increased risk.18 Hepatitis A infection usually presents in one of five different clinical patterns:
PATHOGENESIS
Young children (less than 2 years of age) are usually asymptomatic; only 20% develop jaundice, while most 5-year-old children or older (80%) develop symptoms. This high rate of symptoms is maintained in adulthood. HAV infection with prolonged cholestasis is rare but could lead to invasive procedures, as the diagnosis of acute hepatitis may not be readily accepted in patients with a history of jaundice for several months even in the presence of detectable immunoglobulin M (IgM) anti-HAV.30 Relapsing hepatitis A is observed in approximately 10% of patients with acute hepatitis A; it is a benign syndrome and resolves by itself.31 Cholestasis and relapsing hepatitis do not have an increased mortality and treatment is symptomatic. Shedding of HAV has been documented during the relapse phase.31 Acute hepatitis A, unlike hepatitis E, has not been shown to have an increased mortality in pregnant women.
Once the HAV is ingested and survives the gastric acid, it traverses the small-intestine mucosa and the virus reaches the liver through the portal vein. The putative cellular receptor for the virus has been identified as a surface glycoprotein on an experimental model in African green-monkey kidney cells.26 Once the virus reaches the liver cell, it starts replicating in the hepatocyte cytoplasm where it has been shown as a fine granular pattern, but it is not present in the nucleus. The distribution of HAV is throughout the liver. Although HAV antigen has been detected in other organs (lymph nodes, spleen, kidney), the virus appears to replicate exclusively in liver cells. Once the virus is mature, it reaches the systemic circulation through hepatic sinusoids and is released into the biliary tree through bile canaliculi and the small intestine and during acute hepatitis it is eventually excreted in feces. The pathogenesis of HAV-associated cell injury is not completely defined. The lack of cell injury in cell culture systems does not support the hypothesis of a cytopathic virus. Immunologically mediated cell damage is greatly favored. The hypothesis is that the emergence of antibody to HAV could result in hepatic necrosis during immunologically mediated elimination of the virus.27
1. asymptomatic (not jaundiced) 2. symptomatic (jaundiced): self-limited to approximately 8 weeks 3. cholestatic, in which jaundice lasts 10 weeks or more 4. relapsing, consisting of two or more bouts of acute HAV infection occurring over a 6–10-week period 5. FHF
FULMINANT HEPATIC FAILURE Fulminant hepatic failure (FHF) due to hepatitis A infection is rarely seen in young people. When hepatic failure is due to hepatitis A, it usually becomes manifest in the first week in about 55% of patients and in 90% during the first 4 weeks following the onset of disease; it is rarely seen after this interval.32 There are reports of increased
CLINICAL FEATURES AND DIAGNOSIS
Table 30-1. Age-specific mortality due to hepatitis A virus
Chronic infection with the HAV does not occur, it only manifests as an acute bout of hepatitis, and rarely acute hepatitis A can have prolonged clinical manifestations such as relapsing hepatitis or cholestasis. Commonly, the incubation period is 2–4 weeks, rarely up to 6 weeks. Mortality is low in previously healthy individuals. The impact of the disease is in the morbidity it causes, usually a sig-
<5 5–14 15–29 30–49 > 49 Total
Age group (years)
Case fatality (per 1000) 3.0 1.6 1.6 3.8 17.5 4.1
629
Section V. Liver Diseases Due to Infectious Agents
contribution of HAV to acute liver failure cases in populations classified as hyperendemic for hepatitis A. As such, in a recent communication from India where 276 patients with FHF were seen between 1994 and 1997, 10.6% of the cases among adults were due to HAV infection. HAV had been responsible for only 3.5% of ALF among 206 patients seen in the same community during the years 1978–1981.33 In recent years, it has been appreciated that certain populations are at high risk of developing increased morbidity and acute liver failure due to hepatitis A infection, among them the elderly34 and individuals with chronic liver disease. A 1998 scientific report described the clinical outcome of 256 individuals hospitalized with acute hepatitis A in the state of Tennessee between January 1994 and December 1995.35 All patients had a serological diagnosis of acute hepatitis A based on a detectable IgM antibody to hepatitis A. Figure 30-3 depicts the clinical outcome of these patients. The age of the individuals played a major role when rates of complications were analyzed. Twenty-five percent of patients aged 40 or above and 11% of patients younger than 40 years of age had at least one complication (P = 0.014). Although two recent reports described the further decline of acute viral hepatitis as a cause of fulminant failure,36,37 this decline
100
89%
80
Percentage of patients
60
39%
40
26% 20
5% 0 Died
Serious Prolonged PT complications
Nausea vomiting
Figure 30-3. Hospitalized patients due to acute hepatitis A have significant disease.35 PT, prothrombin time.
630
is principally due to the control of hepatitis B, while the contribution of HAV infection to FHF has remained unchanged over the last three decades, despite the availability of highly efficacious vaccines.
CLINICAL SYMPTOMS OF ACUTE HAV INFECTION Prodromal symptoms include fatigue and weakness, anorexia, nausea and vomiting, and abdominal pain. Less common symptoms include fever, headache, arthralgias, myalgias, and diarrhea. Usually dark urine precedes other symptoms in approximately 90% of infected individuals; this sign occurs within 1–2 weeks of the onset of prodromal symptoms. Symptoms may last a few days, up to 2 weeks, usually decreasing with the onset of clinical jaundice. Tenderness and mild liver enlargement are present in 85% of individuals, splenomegaly in 15%, and cervical lymphadenopathy in 15%. Complete clinical recovery is achieved in 60% of individuals within a period of 2 months and in almost everybody by 6 months. Overall the prognosis of acute hepatitis A infection is excellent in otherwise healthy adults. Some subjects develop complications that can be fatal (FHF).
DIAGNOSIS, BIOCHEMICAL ABNORMALITIES, AND SEROLOGY Acute hepatitis A is clinically indistinguishable from other forms of viral hepatitis. The diagnosis of infection is based on the determination of antibodies against HAV (anti-HAV) in serum. A diagnosis of acute hepatitis A requires demonstration of IgM anti-HAV in serum. This test is positive from the very onset of symptoms29 and usually remains positive for approximately 4 months.38 However, certain patients have been shown to have low levels of detectable IgM anti-HAV for more than a year after the initial infection.38 IgG anti-HAV is also positive at the onset of the disease, remains detectable usually for life, and is interpreted as a marker of previous infection. The diagnosis of acute HAV should be differentiated from any other acute viral hepatitis, autoimmune hepatitis, and others by means of appropriate serological testing. However, some cases may be difficult because individuals may harbor more than one viral infection, such as patients with chronic hepatitis B or chronic hepatitis C with superimposed acute HAV. Testing for HAV RNA is limited to research laboratories. HAV RNA has been detected in serum, stool, and liver tissue. Amplification of viral RNA by polymerase chain reaction (PCR) is available. Using this test, HAV RNA has been documented in human sera for up to 21 days after onset of illness.39 Briefly, the principles of PCR are: HAV RNA is reverse-transcribed (RT-PCR) into complementary cDNA before proceeding to amplify the viral product. Viral amplification is obtained in the presence of specific oligonucleotide primers, which are described in the scientific literature.40 In a recent description of 76 French patients with acute HAV infection, seen between January 1987 and April 2000, 19 were diagnosed as having FHF.41 Ten patients experienced liver death and required orthotopic transplantation and one patient died while awaiting liver transplantation. The article describes the HAV RNA status in 39 of the 50 patients whose sera and clinical data were available, including the 19 with FHF. Of interest, HAV RNA was detected in 36/50 (72%) cases. The likelihood of having an undetectable HAV RNA was more
Chapter 30 HEPATITIS A
pronounced in fulminant hepatitis cases than in non-fulminant cases (P < 0.02). When HAV RNA was detectable, titers were lower in encephalopathic patients than in non-fulminant cases (3.6 log versus 4.4 log, P = 0.02). These data suggest that the finding of a detectable IgM antibody against hepatitis A and undetectable or low-titer HAV RNA in patients with acute liver failure may signal an ominous prognosis and warrant consideration of early referral for liver transplantation. As with other studies, HAV genotypes did not seem to play a role in the clinical manifestations.42
PREVENTION AND TREATMENT Hepatitis A is still an infectious disease of potentially serious consequences due to morbidity and mortality. Recommendations concerning HAV immunoprophylaxis by the Advisory Committee on Immunization Practices were published in December 1999.25 The overall strategy is to protect individuals from disease and to lower the incidence of HAV infection in the USA. The vaccine is not licensed for use in children less than one year of age. At the present time, high-risk populations are targeted for immunization. Table 302 depicts these populations. Since childhood vaccination in high-risk areas was recommended, the overall hepatitis A rate has declined steadily, and in 2002, it was the lowest yet recorded (3.1/100 000). The decline in rates has been greater among children and in states where routine childhood vaccination is recommended, suggesting a positive impact of childhood vaccination. As such, hepatitis rates declined 20-fold during the years 1997–2001 among American Indian and Alaska Native children where routine hepatitis A vaccine was implemented.43 However, a 2003 Centers for Disease Control analysis of hepatitis A vaccination coverage for children aged 24–35 months who reside in the 11 states where the HAV vaccine is routinely recommended showed that immunization ranged from 6.4% to 72.7%,
Table 30-2. High-risk patient groups25a, 25b ∑ All children 12–23 months old ∑ Healthy individuals who travel to endemic areas, work in occupations where likelihood of exposure is high, are family members of infected patients, or adopt infants or children from endemic areas ∑ Persons with liver disease, including chronic hepatitis B or C or those undergoing liver transplantation ∑ HIV-positive patients ∑ Men who have sex with men ∑ Users of injecting and non-injecting illicit drugs ∑ Persons with clotting-factor disorders ∑ Laboratory workers who handle live hepatitis A virus ∑ Persons who live in communities with high or intermediate rates of hepatitis A ∑ Routine vaccination is recommended for children living in 11 states where rates of hepatitis A are at least twice the national average (≥20 cases per 100 000 population). These states are: Alaska, Arizona, California, Idaho, Nevada, New Mexico, Oklahoma, Oregon, South Dakota, Utah, and Washington. In addition, vaccination should be considered for children who live in Arkansas, Colorado, Missouri, Montana, Texas, and Wyoming because the average annual incidence of infection with HAV is 10–20 cases per 100 000 inhabitants HIV, human immunodeficiency virus; HAV, hepatitis A virus.
with an average of 50.9%; while immunization among children of the same age residing in the six states where HAV vaccination should be considered averaged 25% (range 0.6–32.3%). The analysis concluded that HAV immunization rates for children aged 24–35 months are lower than overall rates for other child vaccines.44 It is likely that universal immunization was not recommended in the USA because communities were considered to have high, intermediate, and low rates of hepatitis A and US government surveillance data demonstrated that communities with high and intermediate rates were primarily responsible for an average of 50% of reported HAV cases each year.25 Hence the recommendation was based on the concept that reducing hepatitis A incidence in states with high (Table 30-2) or intermediate average annual incidence of hepatitis A through routine vaccination of children would substantially reduce the national disease incidence. However, recent outbreaks in Georgia, Tennessee, and Pennsylvania, where more than 600 symptomatic cases and 3 deaths were reported, and thousands of exposed individuals needed immediate passive immunization,45 seem to contradict the recommendation for immunization for high or intermediate rates of endemic HAV and it is likely that immunization directed to specific groups would not control infection as efficiently as universal immunization would. There are no specific medications to treat acute hepatitis A; symptomatic treatment is recommended. Sanitary measures and administration of serum Ig were the principal means of prevention against HAV infection in years past. However, the availability of excellent hepatitis A vaccines has rendered the use of Ig for pre-exposure prophylaxis unnecessary. When immune serum globulin is used for postexposure prophylaxis, it should be given within 2 weeks of exposure. In these cases, the recommended dose of Ig is 0.02 ml/kg by intramuscular injection. Although considered safe, it can cause fever, myalgias, and pain at injection sites. Post-exposure prophylaxis with Ig can be safely accompanied by active immunization against hepatitis A.46 Figure 30-4 depicts the magnitude of anti-HAV titer following administration of one dose of Ig or two doses of hepatitis A vaccine. Hepatitis A vaccine was licensed in the USA in 1995; two inactivated hepatitis A vaccines are commercially available in the country. Their extensive use during clinical trials and postmarketing surveillance support the safety and efficacy of these products. Havrix is manufactured by SmithKline Biologicals, Rixensart, Belgium, and Vaqta by Merck Sharp & Dohme, West Point, Pennsylvania. Both vaccines have been prepared with HAV grown in cell culture. The final products are purified and formalin-inactivated and contain alum as an adjuvant. The basic difference between the two commercially available vaccines is the HAV strain used for preparation. Havrix was prepared with the HM175 strain, while Vaqta was prepared with the CR326 strain,47,48 a difference of little or no practical importance since both vaccines are safe and immunogenic. Doses and schedule of immunization are given in Table 30-2. Persistence of anti-HAV serum level is estimated to be detectable for approximately 20 years following immunization with Havrix.49 During clinical trials it was observed that the vaccine was well tolerated and the frequency of adverse events decreased with successive doses. Rare cases of anaphylaxis have been reported and repeated doses should not be given if individuals experience hypersensitivity reactions to the original dose. An incidence of 1–10% has been reported for local reactions at the injection site (induration,
631
Section V. Liver Diseases Due to Infectious Agents
GMT=geometric mean titer 1000 IG 900
Figure 30-4. The magnitude of antibody to hepatitis A virus following immunoglobulin (Ig) administration is usually not detectable by commercial assays and is short-lived. GMT, geometric mean titer.
10000 Vaccine
800 1000 700
500
GMTs
GMTs
600 100
400 300
10
200 100 1
0 Dose of IG
15
30
Primary dose
redness, swelling), fatigue, mild fever, malaise, anorexia, and nausea. An incidence of less than 1% was reported for hematoma at injection site, pruritus, rash, upper respiratory tract infections, abdominal pain, diarrhea, vomiting, arthralgias, myalgias, lymphadenopathy, insomnia, photophobia, and vertigo.50 Post-marketing reports have shown similar events, plus rare side effects such as syncope, jaundice, hepatitis, dyspnea, Guillain–Barré syndrome, multiple sclerosis, and others. Whether these events are related to vaccine administration is unclear. In recent years a combination of hepatitis A and B vaccines has become available (Twinrix), with an excellent record of efficacy and safety.51 The dose schedule is shown in Table 30-3.
IMMUNIZATION AGAINST HEPATITIS A IN SUBJECTS WITH CHRONIC LIVER DISEASE Individuals with chronic liver disease are susceptible to infection with HAV, and increased HAV-related mortality has been observed in these individuals. It is logical to recommend pre-exposure prophylaxis with hepatitis A vaccine for patients with chronic liver disease who are susceptible to infection.52 This recommendation should be extended to pre- or post-liver-transplant individuals, although the efficacy of HAV vaccines is lessened in these individuals.53 Figure 30-5 depicts the seroconversion rate observed in patients with compensated chronic liver disease.52 An acute hepatitis episode in patients with underlying chronic liver disease is fraught with possible exacerbation of symptoms, and increased morbidity and mortality. Although the current guidelines recommend immunization against HAV of all patients with chronic liver disease,25 cost-effective analysis has found controversial results. A report published in 2000 found that it would cost 23 million Canadian dollars to save one life if patients with HCV infection were immunized.54 However, some of the assumptions on this report were challenged and remain unsolved.55 Two other studies on patients with chronic hepatitis C decidedly show an advantage of immunizing these patients against HAV.56,57 The methods used in these studies are dissimilar and some may be insensitive to the inci-
632
180 Days
Seroconversion at mos (%)
100
210 Booster at 6 months
98.2
97.7
94.3
95.2
80 60 40 20 0 Healthy adults Chronic HBV Chronic HCV
Other CLD
Figure 30-5. The seroconversion rate to anti-hepatitis A virus was similar in healthy controls and in subjects with compensated chronic liver disease (CLD); however the anti-hepatitis A virus titer was lower in chronically ill individuals compared to controls.55 HBV, hepatitis B virus; HCV, hepatitis C virus. (Reproduced from Keeffe EB. Vaccination against hepatitis A and B in chronic liver disease. Viral Hepatol 1999; 5:77–88, ©1999, with permission of Blackwell Publishing.)
Table 30-3. Recommended doses of hepatitis A vaccinesa (HAV) Vaccine
Age (years)
Dose
Volume
Dosing schedule
Havrix > 18 Vaqta > 18 Twinrix ≥ 18
2–18
720 EL.U 1440 EL.U 25 U 50 U 720 EL.U HAV 20 mg HBV
0.5 ml 1.0 ml 0.5 ml 1.0 ml 1.0 ml
0, 6–12 months 0, 6–12 months 0, 6–18 months 0, 6–18 months 0, 1, 6 months
a
2–18
Vaccines are injected intramuscularly in the deltoid area. EL.U, ELISA units; HBV, hepatitis B vaccine.
Chapter 30 HEPATITIS A
dence of HAV, underestimation of cases, and economic and societal costs of even one case of FHF. Universal immunization during childhood before the establishment of any chronic liver disease promises excellent prevention of HAV infection.58
REFERENCES 1. MacCallum FO, McFarlan AM, Miles JAR, et al. (eds) Infective hepatitis: studies in East Anglia during the period 1943–1947. Medical Research Council special report no. 273. London: HMSO; 1951:1–144. 2. Havens WP, Ward R, Drill VA et al. Experimental production of hepatitis by feeding icterogenic materials. Soc Exp Biol Med 1944; 57:206. 3. Krugman S, Ward R, Giles JP, et al. Infectious hepatitis: detection of virus during the incubation period and in clinically inapparent infections. N Engl J Med 1959; 261:729–734. 4. Feinstone SM, Kapikian AZ, Purcell RH. Hepatitis A: detection by immune electron microscopy of a viruslike antigen associated with acute illness. Science 1973; 182:1026–1028. 5. Provost PJ, Hilleman MR. An inactivated hepatitis A virus vaccine prepared from infected marmoset liver. Proc Soc Exp Biol Med 1978; 159:201–203. 6. Minor PD. Picornaviridae. Classification and nomenclature of viruses. Fifth report of the International Committee on Taxonomy of Viruses. Arch Virol 1991; 2(suppl):320–326. 7. Siegl G. Replication of hepatitis A virus and processing of proteins. Vaccine 1992; 10:S32–S35. 8. Harmon SA, Summers DF, Ehrenfeld E. Detection of hepatitis A virus RNA and capsid antigen in individual cells. Virus Res 1989; 12;361–369. 9. Seganti L, Superti F, Orsi N, et al. Study of the chemical nature of Frp/3 cell recognition units for hepatitis A virus. Med Microbiol Immunol 1987; 176:21–26. 10. Siegl G, Nüesch JPF, de Chastonay J. DI-particles of hepatitis A virus in cell culture and clinical specimens. In: Brinton MA, Heinz FX, eds. New aspects of positive strand RNA viruses. Washington: American Society for Microbiology; 1990: 102–107. 11. Nüesch JPF, Weitz M, Siegl G. Proteins specifically binding to the 3¢ untranslated region of hepatitis A virus RNA in persistently infected cells. Arch Virol 1993; 128:65–79. 12. Robertson BH, Jansen RW, Khanna B, et al. Genetic relatedness of hepatitis A virus strains recovered from different geographic regions. J Gen Virol 1992; 73:1365–1377. 13. Mathiesen LR, Feinstone SM, Purcell RH, Wagner JA. Detection of hepatitis A antigen by immunofluorescence. Infect Immun 1977; 18:524–530. 14. Emerson SU, Huang YK, Nguyen H, et al. identification of VP1/2A and 2C as a virulence genes of hepatitis A virus and demonstration of genetic instability of 2C. J Virol 2002; 76:8551–8559. 15. Fujiwara K, Yokosuka O, Ehata T, et al. Association between severity of type A hepatitis and nucleotide variations in the 5¢ non-translated region of hepatitis A virus RNA: strains from fulminant hepatitis have fewer nucleotide substitutions. Gut 2002; 51:82–88. 16. Disease burden from viral hepatitis A, B and C in the United States. www.cdc.gov/hepatitis. 17. Centers for Diseases Control and Prevention. Guidelines for viral hepatitis surveillance and case management. Atlanta, GA: CDC; 2005:1–42. 18. Centers for Diseases Control and Prevention. Hepatitis surveillance report no. 59. Atlanta, GA: CDC; 2004:1–60. 19. Gust ID. Epidemiological patterns of hepatitis A in different parts of the world. Vaccine 1992; 10:S56–S58.
20. Gandolfo GM, Ferri GM, Conti L, et al. Prevalence of infections by hepatitis A, B, C and E viruses in two different socioeconomic groups of children from Santa Cruz, Bolivia. Med Clin (Barc) 2003; 120:725–727. 21. Centers for Disease Control and Prevention. Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 1996; 45:1–30. 22. Skidmore SJ, Boxall EH, Ala F. A case report of post-transfusion hepatitis A. J Med Virol 1982; 10:223. 23. Hollinger FB, Khan NC, Oefinger PE, et al. Posttransfusion hepatitis type A. JAMA 1983; 250:2313–2317. 24. Mannucci PM, Gdovin S, Gringeri A, et al. Transmission of hepatitis A to patients with hemophilia by factor VIII concentrates treated with organic solvent and detergent to inactivate viruses. Ann Intern Med 1994; 120:1–7. 25. Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 1999; 48:1–37. 25a Hepatitis A vaccine: what you need to know www.cdc.gov/hepatitis 25b. Immunization Action Coalition www.immunize.org 26. Kaplan G, Totsuka A, Thompson P, et al. Identification of a surface glycoprotein on African green monkey kidney cells as a receptor for hepatitis A virus. EMBO J 1996; 15:4282–4296. 27. Kurane I, Binn LN, Bancroft WH, Ennis FA. Human lymphocyte responses to hepatitis A virus-infected cells: interferon production and lysis of infected cells. J Immunol 1985; 135:2140–2144. 28. Berge JJ, Drennan D, Jacobs J, et al. The cost of hepatitis A infections in American adolescents and adults in 1997. Hepatology 2000; 31:469–473. 29. Liaw YF, Yang CY Chu CM, Huang MJ. Appearance and persistence of hepatitis A IgM antibody in acute clinical hepatitis A observed in an outbreak. Infection 1986; 14:156–158. 30. Gordon SC, Reddy KR, Schiff ER. Prolonged intrahepatic cholestasis secondary to acute hepatitis A. Ann Intern Med 1984; 101:635–637. 31. Sjogren MH, Tanno H, Fay O, et al. Hepatitis A virus in stool during clinical relapse. Ann Intern Med 1987; 106:221–226. 32. William R. Classification, etiology and considerations of outcome in acute liver failure. Semin Liver Dis 1996; 16:343–348. 33. Chadha MS, Walimbe AM, Chobe LP, Arankalle VA. Comparison of etiology of sporadic acute and fulminant viral hepatitis in hospitalized patients in Pune, India during 1978–81 and 1994–97. Ind J Gastroenterol 2003; 22:11–15. 34. Brown GR, Persley K. Hepatitis A epidemic in the elderly. South Med J 2002; 95:826–833. 35. Willner IR, Mark DU, Howard SC, et al. Serious hepatitis A: an analysis of patients hospitalized during an urban epidemic in the United States. Ann Intern Med 1998; 128:111–114. 36. Ostapowicz G, Fontana R, Schiedt FV, et al. Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States. Ann Intern Med 2002; 137:947–954. 37. Schiodt FV, Atillasoy E, Shakill AO, et al. Etiology and outcome for 295 patients with acute liver failure in the United States. Liver Transplant Surg 1999; 5:29–34. 38. Kao HW, Ashcavai M, Redeker AG. The persistence of hepatitis A IgM antibody after acute clinical hepatitis A. Hepatology 1984; 4:933–936. 39. Yotsuyanagi H, Iino S, Koike K, et al. Duration of viremia in human hepatitis A viral infection as determined by polymerase chain reaction. J Med Virol 1993; 40:35–38. 40. Mannucci PM, Gdovin S, Gringeri A, et al. Transmission of hepatitis A to patients with hemophilia by factor VIII concentrates treated with organic solvent and detergent to inactivate viruses. Ann Intern Med 1994; 120:1–7.
633
Section V. Liver Diseases Due to Infectious Agents
41. Rezende G, Roque-Alsonso M, Samuel D, et al. Viral and clinical factors associated with fulminant course of hepatitis A infection. Hepatology 2003; 38:613–618. 42. Fujiwara K, Yokosuka O, Imazeki F, et al. Analysis of the genotype-determining region of hepatitis A viral RNA in relation to disease severities. Hepatol Res 2003; 25:124–134. 43. Bialek SR, Thoroughman D, Hu D, et al. Hepatitis A incidence and hepatitis A vaccination among Anerican Indians and Alaska Natives, 1990–2001. Am J Public Health 2004; 94:996–1001. 44. Hepatitis A vaccination coverage among children aged 24–35 months – United States. MMWR 2005; 54:141–144. 45. Sjogren MH. The clinical profile of acute hepatitis A infection: is it really so severe? Hepatology 2004; 39:572–573. 46. Leentvaar-Kuijpers A, Coutinho RA, Brulein V, Safary A. Simultaneous passive and active immunization against hepatitis A. Vaccine 1992; 10:S138–S141. 47. Andre FE, D’Hondt E, Delem A, Safary A. Clinical assessment of the safety and efficacy of an inactivated hepatitis A vaccine. Vaccine 1992; 10(suppl 1):S160–S168. 48. Provost PJ, Hughes JN, Miller WJ, et al. An inactivated hepatitis A viral vaccine of cell culture origin. J Med Virol 1986; 19:23–31. 49. Van Damme P, Thoelen S, Cramm K, et al. Inactivated hepatitis A vaccine: reactogenicity, immunogenicity, and long-term antibody persistence. J Med Virol 1994; 44:446–451.
634
50. Just M, Berger R. Reactogenicity and immunogenicity of inactivated hepatitis A vaccines. Vaccine 1992; 10(suppl 1):S110–S113. 51. FDA approval for a combined hepatitis A and B vaccine. MMWR 2001; 50:806–807. 52. Reiss G, Keefe EB. Review article: hepatitis vaccination in patients with chronic liver disease. Aliment Pharmacol Ther 2004; 19:715–727. 53. Aeslan M, Wiesner RH, Poterucha JJ, Zein NN. Safety and efficacy of hepatitis A vaccination in liver transplantation recipients. Transplantation 2001; 72:272–276. 54. Myers RP, Gregor JC, Marotta P. The cost-effectiveness of hepatitis A vaccination in patients with chronic hepatitis C. Hepatology 2000; 31:834–839. 55. Jacobs RJ, Koff RS. Cost-effectiveness of hepatitis A vaccination in patients with chronic hepatitis C. Hepatology 2000; 32:873–874. 56. Jacobs RJ, Koff RS, Meyerhoff AS. The cost-effectiveness of vaccinating chronic hepatitis C patients against hepatitis A. Am J Gastroenterol 2002; 97:427–434. 57. Arguedas MR, Heudebert GR, Fallon MB, Stinnett AA. The cost-effectiveness of hepatitis A vaccination in patients with chronic hepatitis C viral infection in the United States. Am J Gastroenterol 2002;97:721–728. 58. Rosenthal P. Cost-effectiveness of hepatitis A vaccination in children, adolescents and adults. Hepatology 2003; 37:44–51.
Section V: Liver Diseases Due to Infectious Agents
31
HEPATITIS B Robert Perrillo, Satheesh Nair Abbreviations AFP a-fetoprotein ALT alanine aminotransferase anti-HBc antibody to HBcAg anti-HBe antibody to HBeAg anti-HBs antibody to HBsAg anti-HDV antibody to hepatitis delta virus AST aspartate aminotransferase BCP basal core promoter cccDNA covalently closed circular DNA CTLs cytotoxic T lymphocytes
HAV HBcAg HBeAg HBIG HBsAg HBV HBV DNA HBX HCC HDAg
hepatitis A virus hepatitis B core antigen hepatitis B e antigen hepatititis B immune globulin hepatitis B surface antigen hepatitis B virus hepatitis B virus deoxyribonucleic acid hepatitis B X antigen hepatocellular carcinoma hepatitis delta antigen
HEPATITIS B: A GLOBAL HEALTH ISSUE There are more than 350 million carriers of hepatitis B virus (HBV) in the world today, of whom 75% reside in Asia and the western Pacific. Effective vaccines have been available for more than 20 years, but perinatal and early life exposure continues to be a major source of infection in high-prevalence areas. Moreover, high-risk behaviors, such as promiscuous heterosexual contact and injecting drug use, account for many new cases in young adults in other parts of the world. Fulminant acute hepatitis B accounts for several hundred deaths per year in the United States, and chronic HBV infection accounts for one million deaths worldwide each year due to complications of end-stage liver disease, including hepatocellular carcinoma. Although significant progress has been made in the area of antiviral therapy, many patients cannot be successfully managed. Thus, universal hepatitis B vaccination is likely to have the greatest impact on liver disease-related mortality in future generations.1
EPIDEMIOLOGY OF HBV INFECTION GEOGRAPHICAL DISTRIBUTION AND SOURCES OF EXPOSURE The prevalence of hepatitis B varies markedly around the world. In highly endemic regions, such as Southeast Asia (excluding Japan), China, and much of Africa, 8% or more of the population are chronic HBV carriers, and the lifetime risk of infection varies between 60% and 80%. In these areas perinatal transmission and horizontal spread among children are the major sources of infection. Nearly half of the HBV carriers in the world reside in these highly endemic areas.2 Areas of intermediate risk include parts of southern and eastern Europe, the Middle East, Japan, the Indian subcontinent, much of the former Soviet Union, and northern Africa. In intermediate-risk
HDV HIV HLA IFN MCSF MHC PCR Si RNA Th1 YMDD
hepatitis delta virus human immunodeficiency virus histocompatability locus antigen interferon macrophage colony-stimulating factor major histocompatibility complex polymerase chain reaction small interfering deoxyribonucleic acid T helper 1 tyrosine-methionine-aspartate-aspartate
areas the lifetime risk of infection is between 20% and 60%. Individuals of all age groups are infected, but as with high-risk areas most infections occur during infancy or early childhood. Regions of low prevalence include North America, western Europe, certain parts of South America, and Australia. Here the lifetime risk of HBV infection is less than 20% and transmission is primarily horizontal (i.e. between young adults). Sexual transmission is the main mode in Europe and North America, with injecting drug use making a major contribution to new cases as well.3 Transmission of infection from an HBV carrier mother to the neonate accounts for the majority of new infections in the world today: 60–90% of hepatitis B surface antigen (HBsAg)-positive mothers who are also hepatitis B e antigen positive will transmit the disease to their offspring, whereas mothers who are positive for antibody to HBeAg (anti-HBe) do so less frequently (15–20%). Other less frequent sources of infection include household contact with an HBV carrier, hemodialysis, exposure to infected healthcare workers, tattooing, body piercing, artificial insemination, and receipt of blood products or organs. Since routine screening of the blood supply was implemented in the early 1970s, transfusion-associated hepatitis B has become rare in the US. Hepatitis B can be transmitted by HBsAg-negative but anti-HBc-positive blood owing to the presence of a small amount of circulating HBV DNA that is only detectable by polymerase chain reaction (PCR) in 10–20% of cases.4 In addition, liver donors who are positive for anti-HBc alone can transmit hepatitis B.5 HBV is efficiently transmitted by percutaneous and mucous membrane exposure to infectious body fluids. The virus is 100 times more infectious than human immunodeficiency virus (HIV) and 10 times more infectious than hepatitis C virus. The presence of HBeAg positivity indicates a higher risk of transmission from mother to child, after needlestick exposure, and in the setting of household contact. HBV DNA has been detected by sensitive testing such as PCR in most body fluids. Although HBV replicates primarily in hepatocytes, the presence of replicative intermediates and virally
635
Section V. Liver Diseases Due to Infectious Agents
Decline among MSM & HCWs
Decline among injecting drug users
16
Cases per 100 000 population
14 12
Vaccine licensed
Universal HBsAg screening of pregnant women Infant immunization
10 8 6
Figure 31-1. Rates of acute hepatitis B infection 1966–2000, and vaccine milestones. Rates of infection have been declining but are most notable in healthcare workers and neonates.
HBsAg screening of high risk pregnant women
OSHA rule Adolescent immunization
4 2
Immunization all children
0 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 Years
encoded proteins in other sites, such as adrenal glands, testis, colon, nerve ganglia, and skin, suggests that there is a vast extrahepatic reservoir for infectious virus.6 Small amounts of HBV DNA have been demonstrated in peripheral mononuclear cells and liver tissue years after apparent resolution of chronic infection.7,8 Extrahepatic localization of low levels of replicating virus explain the relatively high rate of transmission of infection from anti-HBc positive organ donors.9
Other* 5% None identified 16% Heterosexual 42%
RATES OF INFECTION IN THE UNITED STATES Hepatitis B virus has infected an estimated 150 000–450 000 persons in the US each year during the past two decades.10 There has been a 75% reduction in incidence of acute hepatitis B between the years 1987 and 1998 owing to vaccination programs, changes in sexual lifestyle, refinements in blood screening procedures, and the availability of virus-inactivated blood components (Figure 31-1). Most striking has been the decrease among children and healthcare workers, groups with the highest rates of vaccination. Nonetheless, an estimated 78 000 cases of new HBV infections occurred in 2001, with the highest incidence of disease among sexually active young adults (20–29 years old), and higher rates among black and Hispanic people than in white persons.11 Between 1991 and 2001, approximately 40% of cases of acute hepatitis B reported to the Centers for Disease Control were caused by intimate contact among heterosexuals, approximately 18% were due to injecting drug use, and 19% occurred in men who have sex with other men (Figure 31-2). Between 1994 and 1998, more than half of all patients (56%) reported treatment for a sexually transmitted disease or imprisonment prior to their illness, suggesting that these cases might have been prevented by routine immunization in sexually transmitted disease clinics and correctional health care programs.10 According to the third National Health and Nutrition Examination Survey (1988–1994), one or more serologic markers of HBV infection were demonstrated in 4.9% of the US population, and the
636
MSM 19%
Injecting drug use 18% Figure 31-2. Reported risk factors for acute hepatitis B in adults, United States, 1991–2001. Source: Sentinel Counties Study of Viral Hepatitis, CDC, courtesy of Dr Eric Mast. *Other includes household contact, institutionalization, hemodialysis, blood transfusion, and occupational exposure.
prevalence of chronic infection was 0.2%.12 Traditional estimates based on the results of blood donation screening in the late 1970s also indicated a prevalence rate for chronic infection of 0.2–0.4% in the US population. Although it has traditionally been estimated that there are between 1.25 and 1.5 million HBV carriers in the US, this is likely to be a serious underestimate because of changing immigration patterns and under-representation of certain minorities, such as Asians and Pacific Islanders, in field surveys, ethnic groups in which chronic infection rates of 10% are routinely seen.
Chapter 31 HEPATITIS B
CLINICAL OUTCOME OF HBV INFECTION DEFINITIONS In common usage, the term ‘carrier’ has often been used to refer to persistently infected individuals with normal serum aminotransferase levels (sometimes inappropriately referred to as healthy HBV carriers). Because of the potentially confusing nomenclature, it has been proposed that the carrier state be categorized as inactive or active, with the former referring to patients who have evidence for HBV replication by polymerase chain reaction (PCR)-based assay only and normal or only mildly abnormal serum aminotransferase levels.13 Long-term follow-up of inactive carriers suggests that the majority of patients do not have progressive liver disease and do not develop complications. However, some of these patients will ultimately have one or more episodes of reactivated hepatitis in which there an increase in viremia and elevated serum aminotransferase activity. Also, some patients with the inactive carrier state may develop hepatocellular carcinoma. Active carriers, on the other hand, have evidence for HBV replication by non-PCR-based assays for HBV DNA, intermittently or persistently abnormal serum aminotransferase levels, and liver biopsy evidence for chronic hepatitis.
CLINICAL SEQUELAE OF ACUTE HBV INFECTION The age at which an individual becomes infected with HBV determines the clinical outcome. HBV infection in adults with an intact immune system is more likely to cause clinically apparent acute hepatitis B, with only 1–5% of cases becoming chronically infected.3 By contrast, as many as 95% of infected neonates become chronic HBV carriers owing to immunologic tolerance to the virus. In adults, fulminant liver failure due to acute hepatitis B occurs in less than 1% of cases, but this still accounts for 5% of all cases of acute liver failure and approximately 400 deaths annually in the US. Spontaneous survival in acute liver failure due to hepatitis B is approximately 20%. Liver transplantation has resulted in a 50–60% survival rate. Recurrent disease in the allograft is now infrequent owing to the administration of hepatitis B immunoglobulin and antiviral therapy. Rapid viral elimination may result in HBsAg clearance by the time of initial presentation. In these cases, the accurate diagnosis of fulminant liver failure due to hepatitis B may require testing with IgM antibody to hepatitis B core antigen (see Section on Serologic Markers of Infection).
CLINICAL SEQUELAE OF CHRONIC HBV INFECTION Between one-third and one-quarter of people chronically infected with HBV are expected to develop progressive liver disease (including cirrhosis and primary liver cancer). An estimated 15–25% of patients over the age of 40 with chronic HBV infection will die of liver-related causes, including hepatocellular carcinoma. The presence of active viral replication and long-standing necroinflammatory liver disease due to HBV strongly influences the rate
of progression to cirrhosis. The major determinant of survival for patients with chronic hepatitis B is the extent of the liver disease when they first come to medical attention.14 Cirrhosis is associated with decreased survival and a higher frequency of hepatocellular carcinoma. Five- and 20-year survival rates of 55% and 25%, respectively, have been reported in cirrhotic patients, whereas rates of 97% and 63% have been reported for those with mild disease.15,16 The most dramatic difference in survival seems to be between patients with compensated versus decompensated cirrhosis. An 84% 5-year survival was reported for compensated HBV-related cirrhosis. In the same study, a 14% 5-year survival was found for patients with ascites, jaundice, encephalopathy and/or a history of variceal bleeding.17 Multivariate analyses in several large cohort studies have identified that age, ascites, hyperbilirubinemia, and other features of advanced liver disease correlate independently with survival in cirrhotics. Interferon-induced clearance of HBeAg has been associated with longer survival without complications or the need for transplantation.18
MOLECULAR BIOLOGY OF HBV HBV is a small (3.2 kb) DNA virus that belongs to the family Hepadnaviridae. Other members of this family include human HBV-like agents that infect the woodchuck, the ground and tree squirrels, the woolly monkey, crane, heron, Ross goose, and the Pekin duck. HBV has a DNA genome that is in a relaxed, circular, partially doublestranded configuration (Figure 31-3). The genome is composed of four open reading frames in which several genes overlap and use the same DNA to encode viral proteins. The four viral genes are core, surface, X, and polymerase. The core gene encodes the core nucleocapsid protein that is important in viral packaging and HBeAg. The surface gene encodes the pre-S1, pre-S2, and S proteins (large [L], middle [M], and small [S] surface proteins, respectively). The X gene encodes the X protein, which has trans-activating properties and may be important in hepatic carcinogenesis. The polymerase gene has a large open reading frame (approximately 800 amino acids) and overlaps the entire length of the surface open reading frame. It encodes a large protein with functions critical for packaging and DNA replication (including priming, RNA- and DNA-dependent DNA polymerase, and Rnase H activities). Although HBV is a DNA virus, replication is through an RNA intermediate requiring an active viral reverse transcriptase/polymerase enzyme. The reverse transcriptase lacks a proofreading function, and this results in a higher mutation rate than other DNA viruses (estimated to be 1010–11 point mutations per day).19 Complete HBV genomic sequencing has identified a large number of mutations within the HBV genome, many of which are silent or do not alter the amino acid sequence of encoded proteins. Because of genomic overlap, however, some of the silent mutations in one open reading frame (for example the polymerase gene) may result in an amino acid substitution in an overlapping reading frame (HBsAg gene). The clinical implications of this are currently uncertain. Figure 31-4 illustrates the life cycle of HBV. The initial phase of hepadnaviral infection involves the attachment of mature virions to host cell membranes. The human receptor for HBV remains
637
Section V. Liver Diseases Due to Infectious Agents
k 2.4
2.1
NA bR
kb
Figure 31-3. Overlapping HBV genome and major transcripts. The hepatitis B genome is partially double-stranded with four overlapping open reading frames (ORF) or genes. The S gene encodes the viral surface envelope proteins (HBsAg) and is composed of the pre-S1, pre-S2 and S regions. The core gene consists of the precore and core regions, which give rise to the HBeAg and the core protein, respectively. The polymerase gene overlaps the entire S gene and mutations here may give rise to changes in the HBsAg protein. The fourth gene codes for an incompletely understood protein, HBX. Two 11 base-pair direct repeats (DR1 and DR2) are required for strand-specific HBV DNA synthesis during viral replication.
RN A
Pre-S 2 e Pr
1 -S
nd 3.5 kb RNA
+St r an d
-S tr a
-P
-S
RF
F OR
O
OR
F-C
5' DR 1 5'
DR2
Pre -C AA A AA A AA A AA A
ORF X
0.7 kb R
NA
Budding
Excess HBsAg
Assembly ER Uncoating
HBsAg cccDNA
Positive Strand Synthesis
Removal of pregenome
Translation Repair Transcription 2.4 / 2.1 kb RNA 3.5 kb RNA
638
Translation
Negative Strand Synthesis
Figure 31-4. Life cycle of hepatitis B virus. The receptor for viral entry has not been identified. Once inside the cell, the virus undergoes uncoating and nuclear entry of the HBV genome occurs, followed by repair of the single-stranded DNA strand and formation of the covalently closed (ccc) DNA template. Viral transcripts are formed for the HBsAg, DNA polymerase, X protein, and the RNA pregenome; the pregenome and polymerase are incorporated into the maturing nucleocapsid. The surface protein enveloping process occurs in the endoplasmic reticulum. Some of the nonenveloped nucleocapsid recirculates back to the nucleus and the cycle begins again. Excess tubular and spherical forms of HBsAg are secreted in great abundance. (Adapted from Perrillo R, Nair S. Hepatitis B and D. In: Feldman M, Friedman L, eds. Sleisenger and Fordtran’s gastrointestinal and liver disease, 8th edn. Philadelphia: Elsevier, 2005, ©2005, with permission from Elsevier.)
Chapter 31 HEPATITIS B
unknown. Entry of the virus results from fusion of the viral and host membranes as the nucleocapsid is released into the cytoplasm. Mechanisms of intracellular transport of viral genome into the nucleus are poorly understood, but the first step in genomic replication involves conversion of the relaxed circular form of HBV DNA into a double-stranded covalently closed circular form (cccDNA). These cccDNA molecules serve as the template for viral transcription and are the major form of viral DNA in the nucleus of the infected hepatocytes. Subgenomic (0.7–2.4 kb) and pregenomic (3.5 kb) RNA molecules are transcribed from this template. The L protein is translated from the 2.4 kb RNA, the M and S proteins from the 2.1 kb message, and the HBX protein from the 0.7 kb transcript. The pregenomic RNA serves as the template for reverse transcription as well as the mRNA for core and polymerase; the precore RNA codes for the precore gene product. HBV replication begins with encapsidation of the pregenome RNA through complex interactions between host and viral proteins. HBV DNA polymerase reverse transcribes the pregenomic RNA into a negative-strand HBV DNA, which in turn serves as the template for positive-strand synthesis to form a partially doublestranded genome. Concurrent with HBV DNA synthesis, the nucleocapsid undergoes maturation and, through a yet incompletely understood mechanism, interaction occurs with the S protein to initiate viral assembly in the endoplasmic reticulum. S protein is synthesized in the endoplasmic reticulum, where monomer aggregates that exclude host membrane proteins subsequently bud into the lumen as subviral particles. Once formed, the HBsAg undergoes glycosylation in the endoplasmic reticulum and the Golgi apparatus. Non-infectious subviral particles (spherical and filamentous forms of HBsAg) are secreted in great abundance compared to mature virions.
GENOMIC VARIATION HBV Genotypes A genetic classification based on the comparison of complete genomes has demonstrated eight genotypes of HBV (A–H).20 The different genotypes are based on an intergroup divergence of 8% or more in the complete nucleotide sequence. During the last few years there has been a rapid increase in information regarding the epidemiology, molecular characteristics, clinical impact and treatment response of different HBV genotypes (Table 31-1). In the next several years, new extrapolations of the importance of HBV genotypes should emerge because several methods are now available for their detection, including a commercially available line probe assay (INNO-LiPA, Innogenetics, Ghent, Belgium). HBV genotype A is found mainly in Caucasians in northwestern Europe and the USA. Genotypes B and C are highly prevalent in Asia. Recent changes in immigration patterns have resulted in an influx of Asian HBsAg carriers with these genotypes into the US.21 Genotype D has been found in all continents, with the highest prevalence in the Mediterranean region and the Middle East. Genotype E is almost entirely restricted to West Africa and is genetically very similar to genotype D. The most divergent genotype F is found in South and Central America. It is believed that this is the original genotype of the New World.22 Cases of genotype G have been reported in the US and France. A newly defined genotype, H, has been described in Mexico, Nicaragua, and California.
Table 31-1. Hepatitis B Genotypes and Their Possible Clinical Associations Eight well characterized genotypes (A–H) Different geographic distributions A Northwestern Europe, North America, Central Africa B Southeast Asia, including China, Japan and Taiwan (increasing prevalence in North America) C Southeast Asia (increasing prevalence in North America) D Southern Europe, Middle East, India E West Africa F Central and South America, American natives, Polynesia G USA, France H Central and South America: Proposed clinical associations Time to HBeAg seroconversion and probability of HBsAg loss (B shorter than C) Response to treatment with interferon-a (A>B>C>D) Precore/core promoter mutant frequency (precore not selected with genotypes A and F) Liver disease activity and risk of progression (B
Evidence Strong Strong Strong Strong Strong Weak Not known Not known Weak (?higher in genotype A) Not known
Some patients seem to be infected with more than one genotype (for example genotypes B and C and genotypes A and G), which is likely to be due to genetic recombination.22 The mechanism of recombination is unclear; however, this may arise either following simultaneous transmission of several HBV genotypes, or as a result of sequential infections. In addition to the main HBV genotypes, several subtypes have been identified (Aa and Ae, found in Africa and Europe, respectively; Ba and Bj, found in southern China, Taiwan, Vietnam and Japan, respectively). The subtype Ba seems to have been derived as a result of recombination between Bj and the precore/core regions of HBV genotype C (see section on Mutation of HBV genome). There appear to be clinical associations with the various genotypes. At present, the strongest clinical associations appear to be that HBeAg seroconversion occurs earlier in patients with genotype B than in those with genotype C, and that patients with genotype A respond more frequently to interferon therapy than do patients with genotype D.23 Several studies have shown that patients over the age of 50 years with genotype C tend to have a higher rate of cirrhosis and liver cancer.24,25 In a study from Taiwan, patients with liver
639
Section V. Liver Diseases Due to Infectious Agents
cancer below the age of 50 tended to be predominantly genotype B.24 As the genotype B strains in Taiwan and Japan are different – Ba and Bj, respectively – further studies will be needed to clarify whether the clinical differences reported in Japan and Taiwan are related to the presence of different genotype B subtypes.
MUTATION OF THE HBV GENOME The inherent mutability of HBV and its high production rate (1012–13 virions per day) could in theory account for all possible single base changes to be made daily. The vast majority of mutations identified by comparison of nucleotide sequences with wild-type HBV are silent or do not alter the amino acid sequence in a particular open reading frame. Some mutations, however, have been shown to have potentially important disease associations and are defined below.
HBsAg Mutants Mutations in the surface gene can result in changes in the antibody binding domain. Accordingly, both virus neutralization by polyclonal antibody to HBsAg and testing for HBsAg by methods that depend on antibody binding can be affected. Large-scale vaccination programs in endemic regions have revealed a 2–3% incidence of vaccine escape mutants resulting from alterations in the ‘a’ determinant of the HBsAg protein, which is the major neutralizable epitope. Typically, this results in the substitution of glycine for arginine at amino acid position 145, which prevents binding of neutralizing antibodies (i.e. anti-HBs). The clinical significance of these mutants for neonatal vaccination programs remains highly controversial because the frequency of these variants in mothers whose infants respond to vaccination has been found to be similar to that of mothers whose infants failed prophylaxis.26 Another setting where ‘a’ determinant mutants are proposed to have clinical relevance is following liver transplantation for hepatitis B. Patients who develop recurrent HBV infection despite the use of hepatitis B immunoglobulin (HBIG) have been shown to have these ‘escape’ mutants in as many as 50% of cases, and the rate at which they can be detected appears to correlate with the duration of passive immunization.27
Mutations in the Precore, Basal Core Promoter, and Core Genes Mutations in the precore and basal core promoter regions of the HBV genome can influence the production of HBeAg. A precore mutation results in a stop codon at nucleotide 1896 that abolishes the synthesis of HBeAg,28 whereas mutations in the basal core promoter at nucleotides 1762 and 1764 decrease HBeAg synthesis by approximately 70% while maintaining pregenomic RNA levels.29 Both types of mutation have been associated with severe or fulminant hepatitis, which has been attributed to the loss of the immunetolerizing effects of HBeAg antigen. These two types of HBV mutant have been described in the same patients and are particularly common in Asian and European patients with chronic hepatitis B.30 A large serosurvey of HBV carriers residing in the US has found that precore and core promoter mutations are common (27% and 44%, respectively), reflecting the ethnicity and place of birth. Both mutant forms of HBV were observed to occur far more commonly in HBeAg-negative patients (precore 38% versus 9%; core promoter 51% versus 36%).31 In addition to these mutations,
640
upstream mutations in the core gene can influence immunologic responses to HBV. Core gene mutations have been shown to block recognition of HBV by cytotoxic T cells, a key mode of viral clearance, thereby contributing to HBV immune escape and possibly influencing the response to interferon-a.32,33 Core gene mutations within the immunodominant epitopes of the HBV nucleocapsid can also affect CD4+ T-cell reactivity.34 In patients with perinatally acquired chronic hepatitis B, a prolonged immune tolerant phase with minimal to absent hepatic necroinflammatory activity is typically seen for the first 20–30 years of HBV infection. Sequencing studies have shown stable core gene sequences during this phase.35 Precore mutations are also uncommon during this period. Core gene mutations become more common as patients pass from the immune tolerant phase, at which time an increasing number of mutations are observed in the region of the core gene that includes many B- and T-cell epitopes. Both precore stop codon mutants and core gene mutants have been associated with a poor response to interferon (IFN) therapy. Several studies have demonstrated a correlation between HBV genotype and the development of mutations down-regulating the expression of HBe antigen, in particular the core promoter mutation T1762/A1764 and/or the precore stop codon G1896A.36 HBV genotypes B, C, D, and E are prone to develop the precore stop codon owing to a G to A mutation at nucleotide 1896 in the precore region, whereas this very rarely occurs in genotype A. This genotypedependent selection is based on the stability of the stem loop structure of the HBV pregenome encapsidation sequence at the opposite positions 1858 and 1896.37
HBV DNA Polymerase Mutants The polymerase gene product is needed for encapsidation of viral RNA into core particles and conversion of the pregenomic viral RNA into genomic viral DNA. In general, the HBV reverse transcriptase function of the polymerase gene is highly conserved, as major mutations impairing viral replication efficiency lead to selection pressure against such a variant form. After prolonged exposure to lamivudine, a nucleoside analog that inhibits HBV DNA polymerase, nucleotide substitutions have been observed in region B (the template-binding site of the polymerase) and region C (the catalytic site of the polymerase). These mutants are collectively referred to as YMDD, the letters standing for the amino acid residues (Y = tyrosine, M = methionine, D = aspartate) in the catalytic domain. The major site of mutation is the methionine residue in the YMDD amino acid motif. There are two types of mutation at nucleotide position 204 (formerly codon 552) of region C that result in substitution of the amino acid methionine for either isoleucine or valine (designated M204I, M204V, respectively). The M204V tends to occur in conjunction with a mutation in domain B that results in substitution of leucine to methionine (L180M). The M204I mutation or the combined M204V, L180V mutations result in marked resistance to the effect of lamivudine (>10 000-fold reduced susceptibility). The inherent mutability of HBV accounts for the fact that single and double polymerase mutations of this type are likely to pre-exist in patients and are selected during treatment with lamivudine. Thus, mutations in these sites become more frequent as the duration of treatment becomes more prolonged, being found in approximately 15–20% of patients after 1 year of treatment, 30–40% of those
Chapter 31 HEPATITIS B
treated for 2 years, 50% after 3 years, and in more than 65% where treatment has been continued for 4 years.38 These mutant viruses appear to be less replication fit, and although patients with these mutants have a lower chance of HBeAg seroconversion, they may continue to exhibit clinical improvement for a variable period. Persistent infection with the mutant virus, however, is ultimately associated with progression of the disease in many patients and blunting of histological responses to antiviral therapy.39 Severe flares of hepatitis have been reported after the emergence of lamivudineresistant HBV,40 and the acquisition of these mutants may lead to rapidly progressive liver disease after liver transplantation.41
NATURAL HISTORY OF CHRONIC HBV INFECTION Four stages of hepatitis B infection have been described: immune tolerance, immune clearance, low or non-replicative phase (also referred to as the inactive carrier stage) and reactivation phase (Figure 31-5). Patients who are exposed in the perinatal period often have high levels of serum HBV DNA without biochemical evidence of active hepatitis, and are considered to be immunotolerant to HBV. When followed longitudinally, many of these patients ultimately develop abnormalities in their liver chemistries in association with histologic evidence for chronic hepatitis. The trigger mechanisms for this apparent change in tolerance are poorly understood, but are likely to reflect changes in host immunoreactivity. Experiments in transgenic mice suggest that HBeAg induces a state of immunologic tolerance to HBV in neonates. Thus, perinatal transmission of HBe antigen has been considered to be a potential mechanism for the immunologically tolerant state. As patients enter the immune clearance phase there is a decrease in HBV DNA concentrations, increased ALT levels, and histologic activity, reflecting immune-mediated lysis of infected hepatocytes. This second
phase has a variable duration and often lasts many years. The third phase (low-level replication) occurs after seroconversion from HBeAg to antibody to HBeAg (anti-HBe). This is usually preceded by a marked reduction of serum HBV DNA to levels that are only detectable by PCR, followed by normalization of ALT levels and resolution of liver necroinflammation. This is also termed the inactive carrier stage. It may last for a lifetime, but a proportion of patients ultimately undergo spontaneous or immunosuppression-mediated reactivation of HBV replication, with the reappearance of high levels of HBV DNA with or without HBeAg seroreversion and a rise in ALT levels. For unclear reasons, precore or core promoter mutants preventing or down-regulating HBeAg production may be selected during or after HBeAg seroconversion.42 A key event in the natural history of HBeAg-positive chronic hepatitis is HBeAg seroconversion. This is associated with marked reduction in HBV replication and biochemical and histologic remission in the majority of patients. Regression of fibrosis occurs gradually months to year after HBeAg seroconversion. Most studies have found that the mean annual rate of spontaneous HBeAg seroconversion ranges from 8 to 15% in older children or adults with elevated ALT. Longitudinal studies of untreated patients with predominantly HBeAg-positive chronic hepatitis B have indicated that the incidence of developing cirrhosis ranges between 2 and 5 per 100 person-years, and the 5-year cumulative incidence of progression to cirrhosis varies between 8 and 20%.43 It has been suggested that a higher rate of cirrhosis occurs in HBeAg-negative patients than in those with HBeAg. A variety of risk factors for the development of cirrhosis have been identified. Of these, older age, the stage of fibrosis at presentation, and ongoing HBV replication with persistent or intermittent detection of HBV DNA by non-PCR-based assays are perhaps the most clinically apparent. Combined infection with HDV, HCV or HIV, or concomitant alcohol abuse has also been linked to an increased rate of developing cirrhosis.
HBeAg Anti-HBe HBV DNA
Treatment Opportunity 1
Treatment Opportunity 2
Figure 31-5. Natural evolution and stages of chronic hepatitis B virus infection. Treatment opportunities exist whenever there is elevation of ALT and AST indicative of an immune response to HBV. In some patients HBeAg seroconversion is followed by the selection of precore mutant HBV and continuing chronic hepatitis. Cancer may complicate cirrhosis or the inactive carrier state. (Adapted from Perrillo R, Nair S. Hepatitis B and D. In: Feldman M, Friedman L, eds. Sleisenger and Fordtran’s gastrointestinal and liver disease, 8th edn. Philadelphia: Elsevier, 2005, ©2005, with permission from Elsevier.)
ALT
Immune tolerance
Immune clearance
Normal or minimal hepatitis
Chronic hepatitis
Inactive carrier
Reactivation
Normal or Progressive inactive hepatitis fibrosis
Cirrhosis HCC
Liver histology
641
Section V. Liver Diseases Due to Infectious Agents
Once cirrhosis develops, there are two major complications: hepatocellular carcinoma and hepatic decompensation. HBV infection remains the most important risk factor for hepatocellular carcinoma worldwide. In a large European cohort with HBV-related compensated cirrhosis, the 5-year cumulative incidence of hepatic decompensation was 16% and the incidence per 100 person-years was 3.3.44 Similar rates have also been reported in Asians. The cumulative 5-year incidence for hepatocellular carcinoma has been shown to vary between 9 and 14%.44 Several factors have been identified with an increased risk of hepatocellular carcinoma. These include male gender, African or Asian ethnicity, age over 45 years, having a first-degree relative with primary liver cancer, the presence of cirrhosis, HBeAg-positive status, level of HBV DNA in serum, and reversion from anti-HBe to HBeAg status.45Additional risk factors include exposure to aflatoxin, alcohol, and co-infection with hepatitis C virus or hepatitis delta virus. HBsAg-positive persons with none of the identified risk factors can develop hepatocellular carcinoma, although less frequently. In addition, hepatocellular carcinoma has been described in individuals who lose HBsAg. Recommendations concerning the timing of initiation of ultrasound and a-fetoprotein screening for hepatocellular carcinoma are controversial, but screening is generally recommended in cirrhotics and in all HBsAg carriers over the age of 40, when the likely route of transmission has been through perinatal or early childhood exposure. Some authorities recommend screening in those over the age of 30 in this situation.
CLINICAL FEATURES OF ACUTE AND CHRONIC HEPATITIS B
642
months. Delayed clearance of HBsAg has been reported to be preceded by a decline in HBsAg titers. Serum aminotransferase levels of 1000–2000 IU/l are typically seen, with ALT being higher than AST. In patients with icteric hepatitis the increase in bilirubin levels often lags behind that in ALT levels. The peak ALT level has no correlation with prognosis, and prothrombin time is the best indicator of prognosis. Following clinical recovery from acute hepatitis B and HBsAg seroconversion, HBV DNA often remains detectable using PCR assays (see section on Serologic Diagnosis). After resolution of acute hepatitis, the frequency of HBV-specific CD4+ and CD8+ cells in blood and liver decreases rapidly. Nonetheless, T-cell responsiveness remains very high upon re-encounter with HBV antigens, indicating that traces of virus can maintain the CTL response indefinitely following clinical recovery, thereby exerting control over the virus and preventing reactivated infection.46 Fulminant hepatitis occurs in less than 1% of cases. This generally takes place within 4–8 weeks of the onset of symptoms and is associated with encephalopathy, multisystem organ failure, and a high mortality (>80%) if not treated by liver transplantation. Patients over the age of 40 appear to be more susceptible to ‘lateonset liver failure’, in which encephalopathy, renal dysfunction, and other extrahepatic disorders associated with severe liver insufficiency become manifest over the course of several months. The pathogenic mechanisms for fulminant hepatitis are poorly understood, but are presumed to cause massive immune-mediated lysis of infected hepatocytes. This may explain why many patients with fulminant hepatitis B have no evidence of viral replication at presentation and HBV recurrence rates are low after transplantation.47
ACUTE HEPATITIS B
CHRONIC HEPATITIS B General Features
The incubation period varies from a few weeks to 6 months, depending on the amount of replicating virus in the inoculum. Only 30% of patients develop jaundice. The disease may be more severe in patients co-infected with other hepatitis viruses or in those with established underlying liver disease. Alcohol abstention is usually recommended, but the chances for an uneventful recovery do not appear to be affected by the consumption of moderate amounts of alcohol (20–30 g daily) during the convalescent phase. Acute infections are heralded by a serum sickness-like prodrome in 10–20% of patients, with fever, arthralgias/arthritis, and skin rash, most frequently of the maculopapular or urticarial variety. This prodrome results from circulating HBsAg–anti-HBs complexes that activate complement and deposit in the synovium and walls of cutaneous blood vessels. These features generally abate prior to the manifestations of liver disease and peak aminotransferase elevations. Clinical symptoms and jaundice generally disappear after 1–3 months, but some patients have prolonged fatigue even after normalization of ALT levels. In general, elevated ALT and serum HBsAg decline and disappear together, and in approximately 80% of instances HBsAg disappears by 12 weeks after the onset of illness. In 5–10% of cases HBsAg is cleared early and is no longer detectable by the time the patient first presents to the physician. Persistence of HBsAg after 6 months implies the existence of a carrier state, with only a small likelihood of recovery during the next 6–12
A history of acute or symptomatic hepatitis is often lacking in patients with chronic HBV infection. When symptoms are present, fatigue tends to predominate over other constitutional symptoms such as poor appetite and malaise. Right upper quadrant pain may also be found, but is generally low grade in nature. Patients may remain asymptomatic even during periods of reactivated hepatitis. In other instances, particularly when superimposed on cirrhosis, reactivation of infection may be associated with frank jaundice and signs of liver failure (see section on Acute Flares in Chronic Hepatitis B). Physical examination may be normal or there may be hepatosplenomegaly. In decompensated cirrhosis, spider angiomata, jaundice, ascites and peripheral edema are common. Biochemical tests are usually completely normal during the inactive HBV carrier state. In contrast, with the exception of patients in the immunotolerant phase, most patients with the immune clearance phase of chronic HBV infection have mild to moderate elevation in serum AST and ALT. During exacerbations of disease ALT levels may be as high as 1000 IU/l or more, and the patient may have a clinical and laboratory picture that is indistinguishable from that of acute hepatitis B, even including the presence of IgM anti-HBc. Progression to cirrhosis should be suspected whenever there is evidence of hypersplenism, hypoalbuminemia in the absence of nephropathy, or prolongation of the prothrombin time. AST is frequently elevated out of proportion to ALT in patients with advanced cirrhosis.48
Chapter 31 HEPATITIS B
Extrahepatic Manifestations Extrahepatic syndromes are seen in association with acute or chronic hepatitis B (Table 31-2). They are important to recognize because they may occur without clinically apparent liver disease and can be mistaken for independent disease processes in other organ systems. The pathogenesis of these extrahepatic disorders has not been fully elucidated, but this is likely to represent an aberrant immunologic response to extrahepatic viral proteins.49 Many of the extrahepatic manifestations (e.g. arthritis, dermatitis, glomerulonephritis, polyarteritis nodosa, papular acrodermatitis, and polymyalgia rheumatica) are observed in association with circulating immune complexes that activate serum complement. Antiviral therapy may be indicated for persistent disease symptoms.
Cutaneous Disorders In acute HBV infection, skin findings may be seen either alone or as part of the serum sickness-like prodrome. The most common manifestation is urticarial or maculopapular rash, but occasionally purpura, petechiae, or target-like lesions similar to erythema multiforme can be seen. On biopsy, skin lesions of acute HBV typically appear as a leukocytoclastic vasculitis or lymphocytic venulitis with focal necrosis. Patients with chronic HBV infection are more likely to have palpable purpura rather than other cutaneous manifestations, and these have a histologic appearance compatible with neutrophilic necrotizing vasculitis involving small vessels.
Table 31-2. Extrahepatic Manifestations of Hepatitis B Manifestations of acute HBV infection Serum sickness-like syndrome Cutaneous manifestations Maculopapular rash, urticaria > purpuric rash Papular acrodermatitis of childhood (Gianotti–Crosti syndrome) Articular manifestations Polyarthralgia, polyarthritis, tenosynovitis Manifestations of chronic HBV infection Nephritic Membranous and membranoproliferative glomerulonephritis Immunoglobulin A nephropathy Polyarteritis nodosa Cutaneous Purpuric rash > urticaria and maculopapular rash Rare manifestations* Neurologic Guillain–Barré syndrome Peripheral mononeuropathy, mononeuropathy multiplex, seizures Myopathies Polymyositis, dermatomyositis, myocarditis Cutaneous manifestations Lichen planus, pyoderma gangrenosum, erythema nodosum Dermatomyositis-like syndrome, porphyria cutanea tarda Rheumatologic Rheumatoid arthritis, polymyalgia rheumatica, systemic lupus erythematosus Essential mixed cryoglobulinemia (type II, III) Hematologic Aplastic anemia * Data based on isolated case reports and small case series.
The Gianotti–Crosti syndrome, also known as papular acrodermatitis of childhood, is a rare skin disorder associated with acute HBV infection in children aged 2–6 years. Typically, eruptions of flat-topped small papules are observed on the face and extremities in association with generalized lymphadenopathy.
Arthropathy As with skin manifestations, synovial symptoms may present either alone or as part of the serum sickness-like syndrome prior to the onset of hepatic symptoms. The arthropathy can be difficult to distinguish from acute rheumatoid arthritis, as it may afflict the same joints and be associated with morning stiffness. A minority of patients may experience polyarthritis or tenosynovitis. Typically, symptoms begin abruptly 6–20 weeks after exposure to HBV and 2–3 weeks before other systemic symptoms. The joint symptoms almost always resolve before the development of jaundice, should that occur, and do not lead to joint deformity. In some cases, polyarthropathy may persist for several months.
Polyarteritis Nodosa Polyarteritis nodosa (PAN) occurs sporadically or as a serious complication of chronic HBV infection, conferring a mortality rate of 30–50%. The process affects small to medium-sized arteries and arterioles. There is a strong relationship between PAN and HBV. Positive HBV serology has been reported in 36–70% of patients with PAN.50 Conversely, only two in 500 HBV infections result in PAN. PAN typically presents with a non-specific systemic prodrome that includes fever, malaise, myalgias, and arthralgias. Organ-specific complications subsequently develop owing to end-organ damage from fibrinoid necrosis and perivascular inflammation, which cause microaneurysm formation, stenosis, and vessel occlusion. Neurologic manifestations are very common and include polyneuropathy, mononeuropathy, and CNS involvement leading to confusion, memory loss, seizures, or intracranial hemorrhage. Renal failure and hypertension are common. Cutaneous manifestations include tender subcutaneous nodules mostly on the lower extremities, livedo reticularis, urticaria, ulceration and angioedema. Other complications include eosinophilia, polyarthralgia, myalgia, and cardiomyopathy. Whereas PAN is generally associated with the same range of complications for both the sporadic and the HBV-associated forms, some manifestations, such as renal infarction, malignant hypertension, orchiepididymitis and mesenteric ischemia, are seen more commonly with HBV infection. The disease is characterized by circulating immune complexes that contain HBsAg; for this reason plasmapheresis may be indicated. Good therapeutic responses have also been observed with antivirals, either given alone or in combination with plasmapheresis. There is no apparent relationship between the severity of the vasculitis and the severity of the hepatic disease, and often the hepatic disease is relatively mild although high-level viral replication is frequently present. The course of the disease is variable, but the prognosis is gravest for those with significant proteinuria (>1 g/day), renal insufficiency, gastrointestinal involvement, cardiomyopathy, and CNS involvement.
Glomerulonephritis Several types of glomerular lesion have been described in chronic HBV infection, with membranous glomerulonephritis and
643
Section V. Liver Diseases Due to Infectious Agents
membranoproliferative glomerulonephritis being the most common. Renal biopsies have found immune complex deposition and cytoplasmic inclusions in the glomerular basement membrane. These complexes activate complement and cytokine production, with a subsequent inflammatory response. Nephrotic syndrome is the most frequent presentation of HBV-associated glomerulonephritis. During childhood, significant renal failure at presentation is infrequent and a prior history of clinical liver disease is uncommon. Nevertheless, liver biopsies almost always demonstrate varying degrees of chronic viral hepatitis. The diagnosis of HBV-associated glomerulonephropathy is usually established by serologic evidence of HBV antigens or antibodies, the presence of immune complex glomerulonephritis on renal biopsy, and by the demonstration of glomerular deposits of one or more HBV antigens such as HBsAg, HBcAg, or HBeAg by immunohistochemistry. Most patients have detectable HBeAg in serum, and in addition will demonstrate activation of the classic complement cascade, with low serum C3 and occasionally C4 levels. The renal disease often resolves over months to several years, especially in children. Often this resolution occurs with seroconversion of HBeAg. Rarely, however, renal failure may ensue. The natural history of HBV-related glomerulonephritis in adults has not been well defined, but several reports suggest that glomerular disease is often slowly and relentlessly progressive.51 Treatment has been successfully accomplished with interferon-a and has been linked to successful long-term control of HBV replication.52 Nucleoside analogue therapy has resulted in improved renal function and diminished proteinuria.
Pathology of Chronic Hepatitis B Chronic hepatitis B infection is characterized by mononuclear cell infiltration in the portal triads. Periportal inflammation often leads to the disruption of the limiting plate of hepatocytes (referred to as
interface hepatitis), and inflammatory cells can often be seen at the interface between collagenous extensions from the portal tract and liver parenchyma (referred to as active septae). During reactivated hepatitis B there is more intense lobular inflammation that is reminiscent of the features of acute viral hepatitis. Unlike hepatitis C, steatosis is not a feature of chronic hepatitis B. The only histologic feature on routine light microscopy that is specific for chronic hepatitis B is the presence of ground-glass hepatocytes (Figure 31-6). The morphological findings are due to HBsAg particles (20–30 nm in diameter) that accumulate in the dilated endoplasmic reticulum. Owing to the presence of high levels of cystine in HBsAg, ground-glass cells have a high affinity for certain dyes, such as orcein, Victoria blue and aldehyde fuchsin. Groundglass hepatocytes may also be seen in HBsAg carriers, where they may be detected in up to 5% of cells. When present in abundance, they are often indicative of a state of active viral replication. Immunofluorescence and electron microscopic studies have shown HB core antigen inside the hepatocyte nuclei of affected cells. During periods of intense hepatitis activity, cytoplasmic core antigen staining is generally easily observed. With successful treatment with nucleoside analogs the cytoplasmic core antigen staining often disappears, but nuclear core antigen staining may remain, indicating persistence of the cccHBV DNA template.
ACUTE FLARES IN CHRONIC HEPATITIS B Chronic hepatitis B is often punctuated by sudden flares of disease activity that are reflected by an increase in serum aminotransferase levels. Although a uniform definition of a flare is lacking, they have frequently been described as increments that are at least equal to
Figure 31-6. Ground-glass hepatocytes indicated by arrows. The cellular inclusions represent large amounts of HBsAg in the endoplasmic reticulum of infected hepatocytes. This finding is essentially diagnostic of hepatitis B infection in a patient with chronic hepatitis. (Hematoxylin and eosin; ¥630.) Photo courtesy of Dr Gist Farr, Department of Pathology, Ochsner Clinic Foundation.
644
Chapter 31 HEPATITIS B
Table 31-3. Hepatitis Flares in Chronic Hepatitis B Virus Infection Cause
Comment
Spontaneous
Precipitating factors for antecedent viral replication unclear Often observed during withdrawal; requires pre-emptive antiviral therapy
Immunosuppressive therapy
adulthood owing to a breakdown of immunotolerance to HBV. Multiple episodes of reactivation and remission have been shown to accelerate the progression of chronic hepatitis B, and are particularly likely to occur in patients with the precore mutant form.42
IMMUNOSUPPRESSIVE THERAPY
twice the baseline values. These spontaneous flares are an important part of the natural history of hepatitis B, as when repeated they cause histologic progression. Acute flares in chronic hepatitis B occur in association with a number of circumstances and clinical situations (Table 31-3). Most of these flares are due to a change in the balance between immunologic response to HBV and the extent of viral proliferation. Acute flares in chronic hepatitis B that are not explainable by infection with other hepatotropic viruses often occur as a secondary response to increased levels of replicating wild-type or mutant virus, or as a result of therapeutic intervention with immunologic modifiers such as interferon, corticosteroids, or cancer chemotherapy. In most instances, the initiating events for the acute exacerbations in chronic hepatitis B may not be readily identifiable, and these flares are considered to be spontaneous in nature.
Reactivation of HBV replication is a well-recognized complication in patients with chronic HBV infection who receive cytotoxic or immunosuppressive therapy.54 Suppression of the normal immunological responses to HBV leads to enhanced viral replication and is thought to result in widespread infection of hepatocytes. Upon discontinuation of immunosuppressive medications such as cancer chemotherapy, antirejection drugs, or corticosteroids, immune competence is restored and infected hepatocytes are rapidly destroyed. The more potent the immunosuppression, the greater the level of viral replication, and thus the greater the potential for serious clinical consequences of sudden withdrawal and restoration of immunologic competence. Postmortem study of liver tissue in cases of severe liver injury has documented sparse staining of viral antigens, suggesting that patients were in an active state of immune clearance.55 The vast majority of patients have been HBsAg positive before treatment, but some studies have emphasized the reappearance of this marker in patients who were initially positive for anti-HBc alone.56 Reactivated hepatitis in HBsAg-negative patients with antiHBc is explainable by the possible latency of HBV in liver and mononuclear cells and the large extrahepatic reservoir of HBV. Chemotherapy given to cancer patients who are chronic hepatitis B carriers has been shown to be associated with an increased risk of liver-related morbidity and mortality.57 Acute flares of hepatitis B due to cancer chemotherapy and other immunosuppressive drugs are often detected relatively late. The use of antiviral treatment after major biochemical abnormalities have been detected should be anticipated to have relatively little effect on reducing liver injury, because much of the immunologic response to HBV and viral elimination has already occurred. Instead, the key to management of this situation lies in anticipating its occurrence and early antiviral treatment. Current treatment guidelines suggest that nucleoside analog therapy should commence at the onset of cancer chemotherapy or a finite course of immunosuppressive therapy and be maintained for 6 months after completion of the immune-modifiying therapy.
SPONTANEOUS FLARES
ANTIVIRAL THERAPY-INDUCED FLARES
Antiviral therapy Interferon Lamivudine On treatment YMDD mutant Withdrawal*
HIV treatment
Genotypic variation Precore and core promoter mutants Superinfection with other hepatitis viruses
Often during the second to third month; may herald virologic response No more common than placebo Can have severe consequences in patients with advanced disease Caused by rapid re-emergence of wild-type HBV; can have severe consequences in patients with advanced disease As above when using lamivudine; can also occur with immune reconstitution or secondary to antiretroviral drug hepatotoxicity Fluctuations in ALT common with precore mutant May be associated with suppression of HBV replication
*Has also been reported with other nucleoside analogs.
Spontaneous exascerbations are often due to reactivated infection, and it has been shown that an increase in serum HBV DNA often precedes an increase in serum aminotransferase level. Histologic evidence for acute lobular hepatitis superimposed on the changes of chronic viral hepatitis is frequently observed during these flares. Immunoglobulin M (IgM) antibody to hepatitis B core antigen (HBcAg), a marker which is often diagnostic of acute viral hepatitis, also may appear at this time. The reasons for reactivated infection are unknown, but are probably explained by subtle changes in the immunologic control of viral replication. Reactivation seems to occur more commonly in individuals who are infected with HIV-1.53 In individuals who acquire their infection early in life, flares become more common during
Antiviral treatment of chronic hepatitis B can be associated with flares of hepatitis in several ways. Flares may occur during interferon therapy, after withdrawal from nucleoside analogs, following withdrawal of corticosteroids, and in association with lamivudineresistant mutants.
Interferon Interferon-induced flares occur in approximately a third of treated patients and are explainable by the immunostimulatory properties of the drug. Flares generally occur during the second to third month of treatment with the conventional preparations of interferon. It is currently unknown whether or not they occur as commonly with
645
Section V. Liver Diseases Due to Infectious Agents
the newer long-acting (pegylated) forms of interferon. ALT flares have been shown to be a predictor of sustained virologic response, particularly in patients with high-level viremia.58 Flares tend to be particularly common in patients who have decompensated liver disease, having occurred in 50% of patients in one series.59 They are frequently associated with clinical deterioration in these patients.
Nucleoside Analogs ALT flares occur in approximately 20–25% of patients after withdrawal of nucleoside analogs such as lamivudine and adefovir. These are probably due to rapid resurgence of wild-type virus, and although generally well tolerated they have been associated with serious clinical exacerbations in patients with advanced liver disease.60 Reinstitution of therapy is often associated with a decline in HBV DNA and aminotransferase levels. Flares have been seen to follow the emergence of YMDD mutant HBV during lamivudine therapy.61 Initial reports emphasized the temporal occurrence of these flares at the time of or shortly after the detection of lamivudine resistance. However, further follow-up of patients with lamivudine-resistant HBV mutants indicates that the incidence of moderate to severe ALT flares (defined as >5 and >10 times the upper limit of normal) increases with time after detection of lamivudine resistance. In one long-term follow-up study the cumulative incidence of ALT flares was as follows: <1 year (24%), 1–2 years (29%), 2–3 years (30%), 3–4 years (37%) and >4 years (61%).62
Corticosteroid Withdrawal ALT levels increase, often with an inverse decline in HBsAg concentration and serum HBV DNA, following withdrawal of corticosteroids. Clinical trials have been carried out in which a short course of corticosteroids was used prior to conventional antiviral therapy and suggest that this may enhance virologic response rates.63,64 The immune rebound following withdrawal from a 4–8-week course of corticosteroids may be due to increased activation of lymphocytes that promote Th1 cytokine responses at a time when there is increased viral antigen expression. Serious hepatic decompensation has been reported in patients with advanced disease, and this approach is no longer being used.
Antiretroviral Therapy ALT flares occur in HIV–HBV co-infected patients receiving highly active antiretroviral therapy (HAART) owing to a number of potential causes.65 Lamivudine resistance and withdrawal may be associated with ALT flares, and HBV infection clearly increases the risk of toxicity from antiretroviral medication, usually within the first 6 months of starting treatment. Immune reconstitution due to HAART may also be associated with ALT flares. These patients may also be particularly susceptible to flares due to infection with other hepatitis viruses.
Genotypic Variation Chronic infection with precore mutant HBV is often associated with multiple flares of liver cell necrosis interspersed with periods of asymptomatic HBV carriage.42 Approximately 45% of patients have episodic ALT flares with normal levels in between episodes, and 20% have flares superimposed on persistently abnormal ALT.66 These
646
flares have been attributed to increases in the concentration of precore mutants and changes in the proportion of precore to wildtype HBV.67 Mutations at the basal core promoter (BCP) region of the HBV genome are associated with decreased HBeAg synthesis, active liver histology, and increased viral replication. Multiple exacerbations of hepatitis due to reactivated HBV infection have been described in patients with BCP mutation, either alone or in association with precore mutation.68 HBeAg-negative patients who harbor both the precore and the core promoter mutants may be particularly predisposed to severe reactivation episodes following chemotherapy for malignancies.69
Infection with Other Viruses Patients with chronic HBV infection may exhibit severe flares in serum aminotransferases and even frank liver failure when superinfected with other hepatotropic viruses, such as hepatitis A virus (HAV), hepatitis C virus (HCV) and hepatitis delta virus (HDV). Increased mortality has been reported when delta virus superinfection is superimposed on chronic hepatitis B, and chronic hepatitis delta virus infection is often associated with multiple fluctuations in serum aminotransferase levels (see section on Hepatitis Delta Virus). Acute hepatitis C superimposed on chronic hepatitis B has been reported to be as severe as delta superinfection and has been associated with a high rate of liver failure (34%) and death (10%).70 Acute hepatitis C often leads to chronic infection, and the subsequent course also may be characterized by frequent fluctuations in serum aminotransferase levels. Patients with chronic hepatitis B who become infected with other hepatotropic viruses may become HBeAg negative and serum HBV DNA negative by non-PCR-based assays owing to a process of viral interference. This has been described with superimposition of HAV, HCV, and HDV infections and combined infections with HCV and HDV.
SEROLOGIC DIAGNOSIS OF ACUTE AND CHRONIC HEPATITIS B SEROLOGIC MARKERS OF ACTIVE OR PAST INFECTION Hepatitis B surface antigen (HBsAg) appears in serum 2–10 weeks after exposure to the virus and before the onset of symptoms or elevation of serum aminotransferase levels. HBsAg usually becomes undetectable after 4–6 months in self-limited acute hepatitis. Persistence for more than 6 months implies progression to chronic HBV infection. Figure 31-7A indicates the typical time course for acute hepatitis B with complete immunologic recovery. The disappearance of HBsAg is followed several weeks later by the appearance of hepatitis B antibody (anti-HBs). In most patients anti-HBs persists for life and provides long-term immunity. In some cases anti-HBs may not become detectable, but these patients do not appear to be susceptible to recurrent infection. Anti-HBs may not be detectable during a window period of weeks to months after the disappearance of HBsAg. During this period, HBV infection is diagnosed by the detection of IgM antibodies against hepatitis B core antigen.
Chapter 31 HEPATITIS B
HBsAg
HBV DNA
ALT
HBeAg Symptoms 1000
1000 800
800
ALT (IU/L)
700 600
600
500 400
400
300 200
200
100 0
HBV DNA (million copies/ml)
900
Figure 31-7. A Typical acute hepatitis B. There is a brief period of viremia during which HBV DNA and HBeAg is detectable in serum. These events precede the onset of ALT abnormality. Both HBeAg and HBV DNA are generally no longer detectable after 12 weeks from the onset of illness. B Typical HBeAg-positive chronic hepatitis B. During the replicative phase of infection there is variable expression of ALT abnormality and persistence of circulating HBeAg. C Chronic Hepatitis B with transition to inactive carrier state. During the inactive HBV carrier state low levels of HBV DNA are detectable in serum by PCR only. Generally, HBV DNA values are 2 log10 or less. Continued
0 0
1
2
3
4
A
5
6
12
24
36
48
60
Months after exposure
HBV DNA
ALT
HBsAg HBeAg
1000
1000 800
800 ALT (IU/L)
700 600
600
500 400
400
300 200
200
100 0
HBV DNA (million copies/ml)
900
0 0
1
2
3
4
B
5
6
12
24
36
48
60
Months after exposure
HBV DNA
Anti-HBe
ALT
HBsAg HBeAg
1000
1000
800
800
ALT (IU/L)
700 600
600
500 400
400
300 200
200
100 0
0 0
C
HBV DNA (million copies/ml)
900
1
2
3
4
5
6
12
24
36
48
60
72
84
Months after exposure
647
Section V. Liver Diseases Due to Infectious Agents
HBV DNA
Figure 31-7, cont’d. D Chronic hepatitis B: Evolution of HBeAg-negative mutant. In some patients HBeAg-negative mutants of HBV (precore and core promoter mutants) are selected after the loss of HBeAg. Such patients often have fluctuations in serum HBV DNA and ALT, as indicated in the right hand side of the graph.
ALT
HBsAg HBeAg
1000
anti-HBe
1000
800
800
ALT (IU/L)
700 600
600
500 400
400
300 200
200
100 0
0 0
D
3
6
9
12
15
18
21
24
27
30
33
36
Months after exposure
Coexistence of HBsAg and anti-HBs has been reported in approximately 10 to 25% of HBsAg-positive individuals, and this occurs more commonly in those with chronic hepatitis B.71 In most instances the antibodies are low level, non-neutralizing, and heterotypic, i.e. directed against a different subtype of HBsAg than that present in the infected patient. The mechanisms behind this finding are not clear, but it may be due to antibody formed against minor variants of the HBsAg protein. The presence of these heterotypic antibodies is not associated with specific risk factors or change in clinical course, and may occur in the setting of active liver disease and viral replication.72 Hepatitis B core antibody (anti-HBc) is detectable in acute and chronic HBV infection. During acute infection, anti-HBc is predominantly of the IgM class and is usually detectable for 4–6 months after an acute episode of hepatitis, and may rarely persist for up to 2 years. IgM anti-HBc may become detectable during exacerbations of chronic hepatitis B. Anti-HBc persists in those who recover from acute hepatitis B, and also in association with HBsAg in those who progress to chronic infection. In areas with low HBV endemicity the occurrence of isolated presence of anti-HBc has been found in 1–4% of the general population. Isolated reactivity for anti-HBc may occur in a number of situations. It may occur during the window period of acute hepatitis B, when it is predominantly of the IgM class; many years after recovery from acute hepatitis B, when anti-HBs has fallen to undetectable levels; as a false positive serologic test; after many years of chronic infection when the HBsAg titer has decreased below the level of detection; in individuals who are co-infected with hepatitis C, and rarely, as a result of variable sensitivity of HBsAg assays.73 Evidence for co-infection with hepatitis C virus has been demonstrated in up to 60% of individuals with anti-HBc only.74 The results of PCR testing of sera have shown that 0–30% of patients with isolated anti-HBc have HBV DNA in serum. Usually, the HBV DNA is detectable at a low level and not by standard hybridization assays. The presence of low-level viremia in these HBsAg-negative subjects has clinical implications with regard to potential infectivity. It has been shown that anti-HBc testing of
648
HBV DNA (million copies/ml)
900
blood donors prevents some cases of post-transfusion hepatitis B. Also, the risk of transmission of HBV infection from a liver donor with isolated anti-HBc to a liver transplant recipient has been found to be as high as 50–70% in some series, with the risk of transmission depending partly on the HBV serological status of the recipient.75,76 Low-level viral replication also has implications with regard to the possibility of underlying liver disease. HBV DNA in serum and liver tissue has been confirmed by PCR in HBsAg-negative patients with cirrhosis and hepatocellular carcinoma, and PCR has confirmed an association in some cases of fulminant non-A–C hepatitis.77,78 Hepatitis B e antigen (HBeAg) is a soluble viral protein that is found in serum early during acute infection. HBeAg reactivity usually disappears at or soon after the peak in serum enzymes, and persistent detection 3 or more months after the onset of illness indicates a high likelihood of transition to chronic infection. The finding of HBeAg in the serum of an HBsAg carrier indicates greater infectivity and a high level of viral replication that may require antiviral therapy. Using a commercially available PCR assay, nearly 90% of cases with HBeAg positive chronic hepatitis B were found to have serum HBV DNA levels that were persistently above 105 copies/ml, and the mean value was 8.37 log10 copies.79 In contrast, anti-HBepositive patients had much lower serum HBV DNA levels, with the values being higher in those with persistently or intermittently abnormal ALT (mean of 5.1 log10 copies/ml) than patients with persistently normal ALT (3.10 log10 copies/ml). Most HBV carriers who are HBeAg-positive have active liver disease, the exception being HBeAg-positive children and young adults with perinatally acquired HBV infection, who usually have normal ALT levels and minimal inflammation in the liver.12 Figure 31-7B indicates the serologic features of a typical case of HBeAgpositive chronic hepatitis B. In general, seroconversion from HBeAg to anti-HBe is associated with a mean 3 log10 or greater reduction in serum HBV DNA and remission of liver disease. Patients who have normal ALT and detection of small amounts of HBV DNA (103 copies or less) after HBeAg seroconversion are considered to be inactive carriers (Figure 31-7C). Some patients, however,
Chapter 31 HEPATITIS B
Digene corp.
Roche molecular systems
Bayer corp.
1010 109 108
HBV DNA IU/ml
107
Figure 31-8. Commercially available HBV DNA assays. The dynamic ranges of quantification of the available HBV DNA assay vary considerably and none detects the full range of HBV values that can be observed. The expression of results in IU/ml is preferred when trying to equate the results of one assay with another. (Adapted from Standring DN, Bridges EG, Placidi L, et al. Antiviral B-L-nucleosides specific for hepatitis B virus infection. Antivir Chem Chemoter 2001; 12:119–129.)
106 105 104 103 102 10 1 HBV digene HBV digene Ultra- Amplicor Cobas Cobas hybridhybrid- sensitive HBV amplicor taqman capture I capture II digene monitor HBV 48 HBV hybridmonitor capture II
continue to have active liver disease and detectable HBV DNA in serum, due to either low levels of wild-type virus or the selection of precore or core promoter mutations that impairs HBe Ag secretion (Figure 31-7D). HBV DNA can be measured in serum using qualitative or quantitative assays. The clinical utility of testing for serum HBV DNA has been hampered by the absence of a licensed test in the US, as well as an accepted international reference standard. A number of non-PCR-based assays are available, with varying levels of sensitivity from 103 to 105 genomic copies per milliliter of serum (Figure 31-8). Although less sensitive than PCR, the results of the non-PCRbased assays correlate with a clinical response to antiviral therapy, and several of the currently available antiviral therapies were licensed based on clinical trials that used these assays. There are several shortcomings, however, to the use of these less sensitive assays, and as a result most clinical laboratories use one of several commercially available PCR assays with enhanced sensitivity (102 genomic copies/ml or less). The measurement of serum HBV DNA by quantitative PCR is potentially useful in a number of clinical circumstances, perhaps one of the most common of which is in distinguishing HBeAg-negative chronic hepatitis B from the inactive HBV carrier state with another source of liver test abnormalities, such as alcohol or obesity. Quantitative serum HBV DNA testing is also useful in evaluating candidacy for antiviral therapy, as well as to monitor response during treatment. Patients with high serum HBV DNA have been found to respond less frequently to conventional interferon therapy.80 When using solution hybridization testing, a baseline level of 200 pg/ml or
Versant Versant HBV HBV DNA 1.0 DNA 3.0
more (roughly equivalent to 56 million copies by PCR assay) has been found to be associated with a very low rate of virologic response. By contrast, quantitative PCR testing of baseline serum HBV DNA has not been shown to correlate with a response to nucleoside analog therapy owing to more potent inhibition of viral replication. Monitoring the HBV DNA level at key intervals during treatment allows one to assess the likelihood of HBeAg clearance. Several studies have found that the level of serum HBV DNA at 12–24 weeks of nucleoside analog treatment may distinguish those individuals more likely to go on to HBeAg seroconversion.81,82 Other studies have suggested that baseline HBV DNA level during treatment can be used to evaluate the likelihood of relapse after treatment discontinuation, as well as the chance of developing resistance to lamivudine.83,84 Reappearance of HBV DNA, or a 1 log increase in serum HBV DNA during treatment, suggests that drug resistance has occurred.85Finally, high pretreatment levels of serum HBV DNA have been shown to correlate with a higher rate of recurrent infection in liver transplant recipients treated with lamivudine.86,87 Qualitative PCR is an even more sensitive method of detecting HBV DNA than quantitative PCR. Qualitative PCR has altered traditional concepts of clearance of HBV DNA in acute and chronic infection. HBV DNA can be detected in serum and peripheral mononuclear cells years after recovery from acute viral hepatitis.46 The disappearance of HBsAg is followed by the loss of HBV DNA from serum in chronic hepatitis B. Even then, HBV DNA persists in small amounts in liver tissue and peripheral mononuclear cells years after the infection has subsided.8 Detection of HBV DNA before liver transplantation may identify patients who are at a higher
649
Section V. Liver Diseases Due to Infectious Agents
risk of apparent de novo hepatitis after transplantation, and this also may allow the identification of HBV as the cause of liver disease in HBsAg-negative patients.88,89 Finally, detection of minute amounts of HBV DNA may be particularly important in patients with fulminant hepatitis B, who frequently have cleared HBsAg by the time they obtain medical attention.90
ANTIVIRAL THERAPY OF CHRONIC HBV INFECTION VIROLOGIC ENDPOINTS AND DEFINITIONS OF RESPONSE The primary goal of treatment for chronic hepatitis B is durable suppression of HBV DNA to levels below that associated with liver disease. This can be accomplished with either interferon-a or nucleoside analogs. The level at which this occurs is generally considered to be <105 copies/ml for HBeAg-positive chronic hepatitis B and 104 for individuals with HBeAg-negative hepatitis.12 There are varying definitions of response, but the most important clinically is a lasting or durable response off treatment (Table 31-4). HBeAg seroconversion is an additional endpoint that can be used to determine the appropriate length of treatment with nucleoside analogs, and has been used as a virologic endpoint in the nucleoside analog phase III trials. HBsAg clearance occurs so infrequently with current antiviral therapies that it is considered an impractical endpoint. It should be noted, however, that early HBsAg seroconversion occurs significantly more frequently in interferon-treated versus nontreated patients, and this has not been shown with nucleoside analog therapy. A systematic review of clinical trials using interferon-a indicates that HBsAg seroconversion occurs 6% more often in treated patients than in untreated controls.91 Moreover, the rate of HBsAg seroconversion has been shown to increase upon prolonged followup of individuals who demonstrate treatment-induced loss of HBeAg.92
Table 31-4. Treatment of Chronic Hepatitis B: Definitions of Response to Antiviral Therapy Virologic response Serum HBV DNA decrease to <100 000 copies/ml or less than 20 000 IU/ml Loss of HBeAg with or without seroconversion (anti-HBe)* Biochemical response Normalization of ALT On-treatment response Initial response (viral suppression to fewer than 105 copies/ml with or without loss of HBeAg, in addition to normalization of ALT) Maintained response (as above, but requiring continuation of therapy to maintain) Off-treatment response Sustained response As above, observed for 6–12 months after treatment is discontinued Durable response Indefinite virologic and biochemical response after treatment discontinuation * Pertains to HBeAg-positive patients only.
650
FACTORS INVOLVED IN CHOICE OF AGENTS In deciding on the appropriate therapy for patients with chronic hepatitis B, it is important to consider the level of ALT, serum HBV DNA, and liver histology at baseline; the expense of treatment; the potential for and ability to withstand adverse effects; age and other comorbid conditions; and realistic expectations about the need for monitoring. Both interferon and nucleoside analogs have advantages and disadvantages, and no one therapy is suitable for all patients (Table 31-5). An important advantage of interferon therapy is that it can be used for 1 year or less, responses tend to be durable (80–90% during the first post treatment year), and maintenance therapy is not needed. In contrast, prolonged treatment with nucleoside analogs is often necessary to maintain viral suppression.
GUIDELINES FOR THE MANAGEMENT OF HEPATITIS B There are three published sets of guidelines for the treatment of hepatitis B, representing the consensus statements of the American Association for the Study of Liver Diseases, the European Association for the Study of the Liver, and the Asian-Pacific Association for the Study of the Liver.93–95 In general, these guidelines are quite similar (Table 31-6). All guidelines suffer from unavoidable delays in publication that prevent the incorporation of recent data, and thus they are in need of periodic updating. For example, all guidelines emphasize that interferon may be used as an alternative to nucleoside analogs, but only one of them currently incorporates guidelines on the use of the newer pegylated forms of interferon.94 There are many similarities with regard to the recommendations made in the three sets of published guidelines. In general, the guidelines recommend treatment in individuals who have HBV DNA levels in excess of 20 000 IU/ml, or roughly 100 000 (105) copies. Nucleoside analog therapy is specifically recommended in instances of decompensated cirrhosis. Also, emphasis is given to the treatment of patients with ALT levels that are at least twice the upper limit
Table 31-5. Advantages and Disadvantages of Currently Available Antiviral Agents
Interferon
Nucleoside analogs
Advantage
Disadvantage
Finite duration of treatment Durable off treatment response Loss of HBsAg (5–8%) No drug resistance Low response rate with high-level viremia Oral delivery Negligible side effects Potent inhibition of virus replication Less expensive than IFN
Given by injection Frequent side effects Expensive Unpredictable immunologic effects
Drug resistance Long or indefinite treatment duration Low rate of HBsAg disappearance Moderately expensive when given long term
* Average retail price is approximately $200–700US per month, depending on drug.
Chapter 31 HEPATITIS B
Table 31-6. Recommendations for Treatment of Chronic Hepatitis B* HBeAg Status
HBV DNA
ALT
Treatment strategy
+
+**
£2 ¥ ULN
+
+
> 2 ¥ ULN
Low efficacy with current treatments Observe; consider treatment should ALT become elevated IFN-a, lamivudine or adefovir may be used as initial therapy
-
+
> 2 ¥ ULN
+ or -
+
Cirrhosis
+ or -
-
Cirrhosis
Duration of therapy IFN-a, 16 weeks Lamivudine, minimum of 1 year Continue for 3–6 months after HBeAg seroconversion Adefovir, minimum of 1 year IFN-a non-responders/contraindications to IFN-a Æ lamivudine or adefovir Lamivudine resistanceÆadefovir IFN-a, lamivudine, or adefovir may be used as initial therapy, but IFN-a or adefovir preferred owing to need for long-term therapy End-points of treatment—sustained normalization of ALT and undetectable HBV DNA by PCR assay Duration of therapy IFN-a, 1 year Lamivudine, >1 year Adefovir, >1 year IFN-a non-responders/contraindications to IFN-aÆ lamivudine or adefovir Lamivudine resistanceÆadefovir £2 ¥ ULN No treatment required Compensated: lamivudine or adefovir Decompensated: lamivudine or adefovir, refer for liver transplantation; IFN-a contraindicated Compensated: observe Decompensated: refer for liver transplantation
Adapted from reference 136. * Treatment recommendations for compensated hepatitis B are intended primarily for those with moderate to severe hepatitis. ** HBV DNA >105 copies/mL.
of normal (Table 31-6). This recommendation is based on the observation of very low rates of sustained virologic response with either interferon or nucleoside analogs in patients with minimal pretreatment ALT elevation.96All guidelines indicate that decisions should ideally be made in the context of liver histology, and treatment should be preferentially directed to individuals with moderate to severe hepatitis. Some experts feel that the guidelines are too restrictive, however, and that patients with advanced fibrosis may benefit from treatment even when serum HBV DNA is below 105 copies.97 Also, the exclusion from treatment of patients with ALT levels that are normal or less than twice the upper limit of normal is controversial, because such individuals may occasionally have significant fibrosis and necroinflammatory disease.97 Liver biopsy is key in making the most appropriate treatment decision in this situation.
CURRENTLY AVAILABLE ANTIVIRAL AGENTS Interferon-a Interferon is effective after a relatively short course of treatment (6 months to 1 year), and unlike the nucleoside analogs has not been associated with drug resistance. Also, in contrast to nucleoside analogs, interferon has direct immunomodulatory properties. Interferon enhances human leukocyte antigen (HLA) class I antigen expression on the surface of infected hepatocytes and augments CD8+ cytotoxic T-lymphocyte activity. This could be operatively important in reducing the amount of the covalently closed circular (ccc) form of HBV DNA (the genomic template for viral transcrip-
tion), which may explain the loss of HBsAg that occurs in approximately 5–8% of interferon-treated patients. The major disadvantages of interferon are related to its poor acceptance in comparison to nucleoside analogs, its lower level of HBV DNA suppression, and the greater cost during a year of treatment. Flares of ALT have been described during interferon-a therapy, and although these are potentially important in achieving a virologic response, their unpredictability leads to inconsistent antiviral efficacy. The magnitude of an ALT flare has been shown to predict the likelihood of sustained virologic response in patients with high-level viremia, suggesting that vigorous cell-mediated immune responses are often required to overcome high levels of viral replication.58 Pegylated interferon has been found to be more effective than conventional interferon in the treatment of HBV infection.98 Doses of 1.0 mg/kg body weight of pegylated interferon-a2b and 180 mg of pegylated interferon-a2a given once weekly have been studied in clinical trials.99–101 No data are yet available for judging whether the increased effectiveness of pegylated interferon is primarily a function of a more pronounced effect on viral replication or of greater immunomodulatory action.
Impact of Genotype on Response Viral genotype appears to effect the response to interferon. Early studies to recognize the potential importance of genotype in interferon-treated patients were limited by small sample sizes.102 A relationship between virological response and genotype, however, has been recently reaffirmed in a large multicenter study of pegylated
651
Section V. Liver Diseases Due to Infectious Agents
LAM ADV ETV IdT
5V 20 M
23
6T
4V N
20
41 M
20
1 M
G
02 S2
84
V A1
81 A1
80
L
M
FTC
L1
Nucleoside analogs have excellent oral bioavailability, a good safety record, and antiviral efficacy comparable to that observed with interferon-a2b. They are also considerably less expensive than interferon when given for 48–52 weeks, as recommended in the prescribing information. These drugs have proved to be particularly useful in the management of decompensated cirrhosis, a clinical situation in which even small doses of interferon can lead to worsening liver failure and severe infections. Table 31-7 lists the available agents and those in late-stage development. Nucleoside analogs replace natural nucleosides during the synthesis of the first or second strand (or both) of HBV DNA. They thus serve as competitive inhibitors of the viral reverse transcriptase and DNA polymerase. Because nucleoside analogs partially and reversibly suppress viral replication, they have to be given for more than 1 year in most cases to achieve maximal efficacy. Unfortunately, drug resistance occurs with prolonged monotherapy (Table 31-7). Salvage therapy with another nucleoside or nucleotide analog may be possible if the two drugs do not share similar resistance sites (Figure 31-9). Nucleoside analogs have several other limitations as well. With these agents, demonstrating the clearance of the covalently closed circular form of HBV DNA (cccDNA) has been difficult, and unlike with interferon HBsAg clearance rarely occurs after 1 year of treatment. These problems may be partly due to the fact that nucleoside analogs, unlike interferon, do not have a direct enhancing effect on the immunologic response to HBV.56 Also, as indicated above, post-withdrawal ALT flares have been seen in approximately 25% of cases after discontinuation of treatment.
Lamivudine has the distinction of being the first nucleoside analog to be specifically licensed for the treatment of hepatitis B. The drug has been shown to be a relatively potent inhibitor of viral replication, convenient to administer, and free of severe adverse effects. Clinical trials demonstrated that a 1-year course of lamivudine resulted in the suppression of viral replication and histologic improvement.103 In one study, HBeAg seroconversion and HBeAg loss occurred in 17% and 32% of patients, respectively.104 A 2-year course of lamivudine proved to be more effective, resulting in an increase of HBeAg seroconversion from 17% at 1 year to 27% at 2 years.105 Prolongation of treatment beyond 1 year, however, has been associated with incremental changes in viral resistance (approximately 40% at 2 years), and the longer treatment is continued the more frequently resistance is seen (65% at year 5).62 Resistance is even more commonly encountered (90% at 4 years) in patients co-
73
Nucleoside Analogs
Lamivudine
V1
interferon-a2b. In this study, HBeAg-positive patients with genotype A responded more frequently than those with genotypes B, C and D (47%, 44%, 28%, and 25%, respectively).99 These results confirm and extend the finding of earlier studies in HBeAg-positive and HBeAg-negative patients suggesting that patients with genotype A respond more frequently than those with genotype D. The impact that genotype exerts on response to interferon is particularly relevant in the selection of therapy in the growing population of AsianAmericans, who often carry the less responsive genotype C.21
Figure 31-9. HBV drug resistance sites. Reported HBV polymerase mutations by treatment. With the exception of adefovir, dipivoxil and tenofovir (not shown), the mutational patterns of the L-nucleoside analogs overlap with that for lamivudine at nucleotide site 204, suggesting that the selection of lamivudine-resistant mutants affects future treatment options. LAM, lamivudine; ADV, adefovir; ETV, entecavir; LdT, telbivudine; FTC, emtricitabine. Adapted from 85 with permission.
Table 31-7. Available Antiviral Agents and Those Under Late-Stage Development Drug
Antiviral potency at weeks 48–52 (log10 decline HBV DNA)*
Rate of resistance**
Lamivudine Emtricitabine Adefovir
–4 to –4.5 log –3 log –3 to – 3.5 log
Tenofovir Entecavir Telbivudine Pegylated interferon†
–5 to –6 log –5 to –6 log –5 to –6 log –3.5 to –4 log
High rate of resistance (15–20% at year 1, 60–70% at year 4) Similar resistance pattern to lamivudine Effective against lamivudine resistant and wild-type HBV; resistance 3% at year 2; 6% at year 3, 15% at year 4, and 28% at year 5. Not known but presumably similar to adefovir owing to close chemical similarity Uncommon (<10%) in years 1 and 2 and only in lamivudine-resistant patients Four percent at one year; similar resistance site as lamivudine None
* Based on results of PCR testing. ** Observed when used as monotherapy. † Based on data with pegylated interferon-a2a
652
Chapter 31 HEPATITIS B
infected with HIV-1 owing to the early use of lamivudine in HAART regimens.65 Lamivudine resistance for more than 2 years has been associated with a blunted histologic response, and patients with lamivudine resistance experience more hepatitis flares.33,62
Adefovir Dipivoxil Adefovir dipivoxil is the acyclic analog of dAMP. It is approved for HBeAg-positive and HBeAg-negative chronic hepatitis B based on the findings of randomized, controlled trials in the US, Europe, and Asia.106,107 In these pivotal studies, 48 weeks of adefovir treatment resulted in median serum HBV DNA reductions of 3.52 and 3.91 log10 copies in HBeAg-positive and HBeAg-negative patients, respectively. The rate of HBeAg seroconversion and HBeAg loss was slightly lower than that achieved with 52 weeks of lamivudine (12% and 24%, respectively). An increase in the frequency of HBeAg seroconversion and non-detectability of HBV DNA by PCR has been observed during the second and third years of adefovir treatment.108 In contrast to interferon, the same level of HBV DNA suppression has been demonstrated irrespective of genotype.109 Whereas the degree of viral suppression with adefovir is slightly less than that achieved with lamivudine, the two drugs differ greatly in their resistance profiles. Point mutations (A181V, N236T) in the B and D domains, respectively, of the HBV polymerase gene affect HBV susceptibility to adefovir, but this occurs only in 3% of patients at 2 years of treatment, in 6% at year 3, and in approximately 15% at year 4.110,111 HBV isolates with the N236T mutation have remained susceptible to lamivudine and appear to be sensitive to entecavir and telbivudine in vitro (Figure 31-9).112 The low rate of resistance with prolonged treatment makes adefovir particularly suitable as first-line therapy in HBeAg-negative chronic hepatitis B owing to the frequent need for continued viral suppression in this form of the disease. Adefovir is effective both clinically and virologically in patients with lamivudine-resistant HBV. This has been demonstrated in patients with clinically stable disease, decompensated cirrhosis, and recurrent hepatitis B after liver transplantation.113–115 Adefovir has been shown to be effective when used alone or in combination with lamivudine maintenance. In one study, 37% of patients with lamivudine-resistant virus experienced serious ALT flares when switched to adefovir monotherapy.114 These data have led many experts to advise that patients with serious underlying liver disease be maintained on lamivudine in addition to adefovir. Adefovir has the disadvantage of being potentially nephrotoxic, and dose reductions may be necessary in patients likely to develop compromised renal function.115
Emtricitabine Emtricitabine (FTC) is a fluorinated cytosine analog that inhibits HBV DNA polymerase and HIV reverse transcriptase. This drug is very closely structurally related to lamivudine, differing by the attachment of one fluorine atom, and therefore shares similar mutational sites and rates of resistance. Emtricitabine is currently licensed in the US and other parts of the world for HIV-1 infection. Forty-eight weeks of emtricitabine (200 mg daily) reduces serum HBV DNA by a median of 3 log10 copies/ml and significantly improves liver histology.116 Studies in HIV-1 co-infected patients have shown similar degrees of suppression compared to HBV
alone.117 The incidence of YMDD mutations in patients receiving emtricitabine 200 mg daily has been shown to be between 12% at week 48 and 19% at week 96.116,118 In a recently reported doubleblind randomized trial as many patients demonstrated HBeAg seroconversion (12%) as became resistant (12.6%) during a 48-week course of treatment.116 Based on these findings, it is unlikely that this agent will play a significant role in the management of chronic hepatitis B other than in patients with combined HIV-1 infection.
Tenofovir Disoproxil Fumarate Tenofovir is an acyclic nucleotide inhibitor of HBV polymerase and HIV reverse transcriptase with close chemical similarity to adefovir dipivoxil. It has been licensed for the treatment of HIV-1 infection, and its antiviral activity against HBV has been reported to be greater than that of the 10 mg dose of adefovir in lamivudine-resistant patients.119 Reports of therapeutic efficacy have come largely from small clinical trials in patients with lamivudine-resistant HBV, where 3–4 log10 reductions in HBV DNA have been observed after 1 year of treatment.119–122 The N236T mutation that confers resistance to adefovir is not sensitive to tenofovir. Despite its possible greater antiviral potency compared to adefovir, large-scale studies of this drug for chronic hepatitis B are not yet available. Recently, the FDA has approved a combination formulation of tenofovir and emtricitabine for use in HIV infection. This drug could in theory be useful, as HBV that is resistant to emtricitabine remains susceptible to tenofovir, and tenofovir-resistant HBV remains susceptible to emtricitabine.
Entecavir Entecavir is a deoxyguanine nucleoside analog that selectively inhibits HBV replication. The drug has recently been licensed by the US Food and Drug Administration and European health authorities. This drug blocks HBV replication by inhibiting the priming of HBV DNA polymerase as well as the synthesis of the first and second strands of HBV DNA. It has been shown to be effective against both wild-type and lamivudine-resistant HBV. A dose of 0.5 mg of entecavir given daily for 24 weeks reduced HBV DNA by an additional 1.28 log10 copies/ml compared to lamivudine in treatment-naive HBeAg-positive patients.123 In HBeAg-positive patients entecavir caused a 1.5 log10 greater reduction in serum HBV than lamivudine after 48 weeks of treatment, but the rates of HBeAg seroconversion were comparable at 21% and 18%, respectively.124 In HBeAg-negative chronic hepatitis B, entecavir resulted in a greater percentage of patients becoming HBV DNA negative by PCR compared to lamivudine (91% vs 73%, respectively).125 Phase III clinical trials used a dose of 0.5 mg in nucleoside analog-naive HBeAg-positive and HBeAg-negative patients, whereas the 1 mg dose was found to be the most effective in patients with lamivudine resistance.126 Entecavir resistance is rare, and thus far has only been reported in lamivudine-resistant cases.127 Entecavir-resistant HBV remains susceptible to adefovir or tenofovir, but not to lamivudine or telbivudine (see b-L-Nucleosides below).
AGENTS UNDER DEVELOPMENT A number of other treatments for chronic hepatitis B are currently under development, including alternative nucleoside analogs, immunologic modifiers, gene therapy, and viral packaging inhibitors.
653
Section V. Liver Diseases Due to Infectious Agents
b-L-Nucleosides The L-nucleoside analogs L-deoxythymidine (LdT; telbivudine) and valyl-L-deoxycytidine (val-LdC; valtorcitabine) are promising compounds that are potent and selective inhibitors of HBV replication.128 Treatment of hepadnavirus-infected woodchucks resulted in a greater than 8 log10 reduction in viral DNA levels, whereas only a 3 log10 reduction was observed in response to lamivudine.129 There was also evidence for synergism when the animals were treated with a combination of these L-nucleosides. Telbivudine has been shown to preferentially inhibit second-strand HBV DNA synthesis in vitro, whereas valtorcitabine inhibits first- and second-strand synthesis, making them potentially complementary to each other. Valtorcitabine is in early phase II study, but telbivudine has been more extensively studied. Data from a phase IIb study of telbivudine has shown that either a 400 mg or a 600 mg dose of telbivudine given daily for 52 weeks resulted in an approximate 6 log10 reduction in serum HBV DNA.130 HBeAg loss occurred in 33% of patients treated with telbivudine alone, compared to 28% receiving lamivudine monotherapy. HBV DNA became undetectable by PCR in 64% of those treated with telbivudine alone, compared to 32% of those given lamivudine. Resistance to telbivudine has been reported to occur in 4% of patients after 1 year of treatment. Telbivudine resistance is anticipated to be susceptible to adefovir or tenofovir, but not consistently to lamivudine. Extended treatment protocols at the 600 mg dose, alone or in combination with lamivudine, are currently under way.
Clevudine Clevudine is a pyrimidine analog that is a potent inhibitor of HBV replication both in vitro and in vivo. This drug has been studied fairly extensively in woodchucks, but available human data are limited at present. In one study, clevudine given daily for 24 weeks resulted in a greater than 4 log10 reduction in serum HBV DNA.131 Rebound to pretreatment viral levels occurred slowly in this as well as other studies.131,132 The reason for the sustained antiviral activity after drug discontinuation is currently unexplained, but may be due to the long intracellular half-life of the phosphorylated compound and active metabolites. Should this prove to be the case, it may be possible to give this drug less frequently than once daily. Little is known about long-term safety and whether drug accumulation occurs.
OTHER NOVEL THERAPIES Th1 Cytokines Chronic hepatitis B is an immunologically based liver disorder, and treatments that are directed toward augmenting immune responses could be helpful in viral clearance. IL-12 and IL-18 are cytokines secreted by activated macrophages and dendritic cells. These cytokines promote Th1 cellular responses, which are thought to be important for HBV clearance. Whereas the therapeutic administration of these cytokines has shown some efficacy in transgenic mice, human studies are either lacking or inconclusive.133,134 IL-28 and IL29 are cytokines distantly related to interferon-a and IL-10 that have been shown to inhibit HBV replication in a HepG2 cell line These agents induce 2¢5¢ oligoadenylate synthetase, double-stranded RNA activated protein kinase, and MxA gene expression in primary hepatocytes.135 No human studies are available at present.
654
Thymosin a1 is a thymic-derived peptide that enhances T-cell maturation and augments a number of T-cell functions. This drug is available as therapy for hepatitis B in China, India, Mexico, the Philippines, and South America, on the basis of studies indicating that it can improve ALT and other viral replication indices. Published reports on this drug are extremely variable, however, with some studies suggesting no benefit.136 A recent study in patients with HBeAg-negative chronic hepatitis B found significantly higher rates of sustained virologic and biochemical responses when patients were treated with a combination of thymosine and interferon-a2b followed by interferon-a2b, compared to interferon alone or a combination of interferon and lamivudine.137 It is possible that thymosin will have a role in combination regimens, and side effects have been minimal. Further study is warranted, however, before a recommendation can be made as to its usefulness.
Therapeutic Vaccines The current prophylactic vaccines against HBV contain recombinant HBsAg adsorbed to alum, a strongly Th2-biased adjuvant that would not be expected to be effective in inducing the Th1 type responses required for clearance and long-term control of chronic HBV infection. There are a number of Th1-promoting vaccine strategies being developed, but few of these have been tested in HBV chronic carriers. Based on preclinical data, genetic (DNA) vaccines may one day prove to be highly effective; however, many technical hurdles are yet to be overcome. Recently, phase I studies involving an oral vaccine (HBsAg + preS1 + preS2) demonstrated a reduction in the plasma viral load and an increase of cell-mediated immune responses to HBV epitopes in a subset of subjects.138
Monoclonal Antibodies Small clinical trials are under way using a mixture of two monoclonal antibodies to different epitopes of HBsAg in patients with chronic HBV infection.139,140 These early clinical trials have shown a reduction in HBsAg concentration and a significant reduction in serum HBV DNA. This therapy could be potentially useful in combination with drugs that inhibit viral replication. Moreover, it may prove to be a safe, effective, and more economical alternative than HBIG in patients undergoing liver transplantation for hepatitis B.
Small Interfering RNA The use of small interfering RNA molecules (siRNAs) is an innovative molecular approach to silencing or suppressing HBV gene expression. siRNAs are small duplex or hairpin RNA molecules with sequences that match those of target HBV DNA. The activity of these molecules depends upon their proper delivery to the nucleus of HBV-infected cells. The hairpin sequences are exported to the cytoplasm, where they are incorporated into the cytoplasmic RNAinduced silencing complex (RISC) along with HBV mRNA targets, resulting in degradation of the HBV mRNA and subsequent inhibition of viral replication. The feasibility of this approach has been confirmed in vitro and in the mouse model, where reduction in both HBcAg and HBsAg have been observed in serum.141,142 Further studies are necessary to establish the ideal target sequence to inhibit HBV protein expres-
Chapter 31 HEPATITIS B
sion and to determine whether these sequences can be efficiently incorporated into human hepatocytes.
Viral Packaging Inhibitors Several compounds have recently been developed that have a mechanism of action against HBV that is unrelated to the viral polymerase. Phenylpropenamide derivatives have been shown to reduce the production of encapsidated RNA in both wild-type and lamivudine-resistant virus without affecting the production of HBV RNA, core protein, or nucleocapsids.143,144 A second unique series of nonnucleoside inhibitors of HBV capsid formation also have been developed (heteroaryldihydropyrimidines) that directly interfere with viral nucleocapsid formation. Studies have been performed in transgenic mice and in the duck model of hepatitis B.
TREATMENT OF HIV/HBV COINFECTED PATIENTS Therapy for HBV infection is often indicated when the criteria for treatment of isolated hepatitis B are met. There is debate about what is appropriate therapy in patients co-infected with HIV and HBV. It is of the utmost importance, however, that clinicians contemplating anti-HBV therapy take into consideration whether or not the patient is currently receiving, or is anticipated to start, HAART. For those not on HAART who are judged to be immunocompetent, initiation of HBV monotherapy can be considered. Monotherapy with either interferon or adefovir has been proposed as initial treatment.145 Initiation of lamivudine monotherapy should be avoided in co-infected patients not on HAART because of the risk of developing both HBV and HIV resistance. It is recommended that tenofovir monotherapy not be prescribed without a concurrent HAART regimen.145 For patients who will begin HAART, it may be advisable to include multiple agents with activity against HBV. Tenofovir in combination with lamivudine has been advocated.146 Combination therapy of this type should maintain antiviral effectiveness because lamivudine resistance is treatable with tenofovir and tenofovir-resistant HBV is treatable with lamivudine. In co-infected patients already on long-standing HAART that may include lamivudine, it is reasonable to continue this regimen if lamuvidine resistance is not suspected.146 Should the latter occur, then adefovir or tenofovir should be added. Many experts would continue
the lamivudine in this instance to avoid clinical resistance to adefovir or tenofovir.
INNOVATIVE THERAPEUTIC APPROACHES WITH AVAILABLE AGENTS Lamivudine Maintenance in Advanced Fibrosis Nucleoside analog therapy has been associated with improvement in liver histology, including improvement in bridging fibrosis and cirrhosis when treatment is extended beyond 52 weeks.39 Also, a large, double-blind placebo-controlled study of long-term treatment with lamivudine versus placebo has shown that long-term nucleoside analog therapy can prevent disease progression (Table 31-8).147 In this study the initial intention was to treat for 5 years, but it was terminated early (median of 32 months of active treatment) because patients in the placebo arm clearly had reached more clinical indicators of disease progression, including hepatocellular carcinoma. Among lamivudine recipients, those with YMDD mutant HBV had an intermediate response and reached fewer endpoints than did untreated controls (Table 31-8). If confirmed by other studies, this could have a major impact on the way patients with advanced disease are treated.
Combination Nucleoside Analog Treatment In vitro data and studies in the woodchuck model of hepatitis B have shown that combination nucleoside analogs cause greater suppression of HBV replication than does monotherapy. It has also been proposed that combination treatment prevents or delays the emergence of phenotypic drug resistance. Multidrug-resistant HBV has been described following the sequential use of famvir and lamivudine, and rare instances have been described after sequential therapy with famciclovir and lamivudine and lamivudine and entecavir. This has led some authorities to suggest combination therapy as an initial management strategy despite the added cost and potential for increased toxicity. Somewhat surprisingly, however, the results of early clinical trials of combination therapy in treatment-naïve patients with HBeAgpositive chronic hepatitis B have shown that the combination of two nucleoside analogs (telbivudine and lamivudine) or the combination of a nucleoside analog with a nucleotide analog (lamivudine and adefovir) does not result in greater viral inhibition during the first year of treatment.130,148 The reasons for the lack of apparent additive effect in these studies remain unexplained. It is possible that
Table 31-8. Incidence of Clinical Endpoints in Lamivudine Maintenance Study in Chronic Hepatitis B with Advanced Fibrosis Clinical endpoint n (%) Total Increase in Child–Pugh Score ≥ 2 Hepatocellular carcinoma
Lamivudine group Negative for YMDD mutants Positive for YMDD mutants* (n = 221) (n = 209) 11 (5) 1 (< 1) 8 (4)
23 (11) 14 (7) 9 (4)
Placebo group (n = 214)
38 (18) 19 (9) 16 (7)
* Patients in the group with YMDD mutations were more likely to have an increased Child–Pugh score than those without YMDD mutations (p < 0.001) but were less likely to reach an endpoint than patients given placebo (p > 0.05). Taken from 147 with permission.
655
Section V. Liver Diseases Due to Infectious Agents
adefovir may not add to the greater antiviral activity of lamivudine. The possibility exists that nucleoside analogs such as telbivudine and lamivudine taken together sterically inhibit binding to the HBV DNA polymerase or compete for phosphorylation enzymes (kinases) that are required for drug activation. Thus, not only is no benefit seen for a combination of these two drugs, but combination therapy might actually be detrimental compared to either agent alone. Recently, a combination of emtricitabine and adefovir was associated with more rapid and greater HBV DNA suppression, and this was associated with enhanced T-cell responses.149 More studies need to be carried out with combination nucleoside analog therapy before definitive conclusions regarding benefits and costs of combination therapy can be drawn.
Combination Interferon and Nucleoside Analog Therapy From a conceptual standpoint, the combination of interferon with a nucleoside analog might prove to be more effective than either drug alone because these agents have different mechanisms of action. This might also allow for a shorter course of nucleoside analog therapy, thereby reducing the risk of viral resistance. A number of studies in the woodchuck and several clinical trials in humans have provided support for additive or synergistic effects when interferon is used in conjunction with a nucleoside analog. Three large multicenter studies have evaluated the combination of pegylated interferon and lamivudine. In one study, HBeAg-positive patients received pegylated interferon-a2b with either lamivudine or placebo for 1 year.99 Despite a 3 log10 greater decline in serum HBV DNA at the end of treatment in the combination therapy group, the frequency of loss of HBeAg at the end of follow-up was nearly equivalent (35% and 36%, respectively). It is possible that the low dose of pegylated interferon used in this study (100 mg weekly for 8 months, followed by 50 mg weekly until the end of treatment) may have influenced the high relapse rate in the combination therapy group. In a second study, 814 HBeAg-positive patients were treated with either 48 weeks of pegylated interferon-a2a in a dose of 180 mg once weekly or pegylated interferon in the same dose given concurrently with lamivudine; a third arm received lamivudine alone.100 Patients in the combination therapy arm had a greater degree of HBV DNA suppression (-7.2 log10 copies vs -5.8 for lamivudine and -4.5 for interferon monotherapy). At the end of a 6-month follow-up period, however, HBeAg seroconversion had occurred in 32% of the pegylated interferon monotherapy group, 27% in the combination group, and 19% of the lamivudine-only patients. Between 9 and 15% of patients in this study had previously been treated with lamivudine, which may have influenced the findings. In the third study, patients with HBeAg-negative chronic hepatitis B were treated with 180 mg of pegylated interferon-a2a given weekly for 1 year in combination with either placebo or lamivudine, and these two treatment groups were compared against lamivudine monotherapy.101 Both virologic response and ALT normalization were significantly more common at completion of follow-up in patients treated with the interferon-containing regimens, but as with the study in HBeAg-positive patients those receiving combination therapy did not demonstrate a higher rate of sustained virologic response than those receiving pegylated interferon alone. There was, however, a more rapid decline in serum
656
HBV DNA and an approximate 1 log10 greater reduction in serum HBV DNA at the end of treatment in patients who received combination therapy. Thus, at present a combination of lamivudine and pegylated interferon does not appear to offer any durable offtreatment therapeutic advantages over pegylated interferon alone. Unlike the situation with combination nucleoside analog therapy, however, all three studies cited above provide proof of concept that pegylated interferon and lamivudine have additive antiviral effects on therapy, and further studies using pegylated interferon-a and a nucleoside analog in different treatment schedules may be warranted.
FUTURE ISSUES AND AREAS OF NEED Many advances in the treatment of chronic hepatitis B have been made over the past decade, but many unresolved issues remain. Time-limited treatments that induce a durable virologic response while remaining both safe and easily affordable have not been developed. The treatment of patients with normal or near-normal ALT levels, individuals who are immunosuppressed or HIV co-infected, and those with HBeAg-negative chronic hepatitis B remain the greatest clinical challenges. Significant developments with nucleoside analog therapy over the past decade have overshadowed the importance of the host immune response in achieving therapeutic end-points and many authorities have relegated interferon to consideration as second-line therapy. However, data demonstrating the enhanced potency of pegylated interferon now provide reason to re-evaluate this issue. As nucleoside or nucleotide analogs have to be given for extended periods this encourages the selection of drug-resistant HBV mutants. This occurs to some extent with all of these agents when used as monotherapy. Furthermore, recognition that sequential nucleoside analog therapy can result in the selection of mutant HBV that is resistant to more than one drug may ultimately lead to the use of a combination of nucleos(t)ide analogs as first-line therapy in some patients. This strategy is likely to be most useful in patients who have features that correlate with a high probability of drug resistance (for example high serum HBV DNA level at baseline). Immunologic modifiers that are better accepted and more predictably effective than interferon would be a great step forward for this field, and could provide an opportunity for durable viral suppression with much shorter courses of nucleos(t)ide analog therapy. The same could potentially be said of other drugs that reduce serum HBV DNA levels independently of inhibition of viral DNA polymerase.
IMMUNIZATION FOR HEPATITIS B Immunization against HBV can be achieved by vaccination against HBs Ag, which confers active immunity, or by immunoglobulin, which confers passive immunization. Active immunization gives long-term immunity, whereas passive immunization confers only immediate protection (2–4 months).
HEPATITIS B IMMUNOGLOBULINS Hepatitis B immunoglobulin (HBIG) contains high titers of antiHBs antibody. Several clinical trials have established the efficacy of HBIG in preventing HBV in postexposure prophylaxis settings.150–153
Chapter 31 HEPATITIS B
HBIG licensed in the United States has an anti-HBs antibody titer of 1:100 000. In Europe, several preparations of HBIG with different concentrations and pharmacokinetics are available. HBIG is safe, but anaphylactic reactions can rarely occur, particularly when given intravenously. Myalgias, skin rash and arthralgias have been reported, and these are believed to be due to antigen–antibody complexes that can occur in HBsAg positive-patients who have mistakenly been given HBIG.
HEPATITIS B VACCINES At present there are two vaccines that are specifically licensed for hepatitis B prevention in the US: Recombivax HB (Merck, licensed in 1986) and Engerix-B (SmithKline Beecham, licensed in 1989). The two vaccines are similar in efficacy. Aluminum hydroxide is added as an adjuvant and thiomersal as a preservative. Because of concerns about the mercury content in thiomersal, preservative-free vaccines are available. For optimal response, the hepatitis vaccine is administered intramuscularly in the deltoid area of adults and in the anterolateral thigh in infants or neonates. HBV vaccines make use of DNA recombinant technology by introducing HBs antigen gene (S gene) into yeast, Saccharomyces cerevisiae. These vaccines induce HBsAg-specific T-helper cells and T cell-dependent B cells to produce neutralizing HBs antibodies against the epitope a (aa124-148) of HBsAg as early as 2 weeks after the first immunization.154 HBV vaccines are highly efficacious in preventing HBV infection. Recipients of HBV vaccine develop only antibody to HBsAg (antiHBs). The detection of antibody to hepatitis B core antigen implies infection, which is frequently subclinical. Anti-HBs titers greater than 100 MIU/ml confer 100% protection against hepatitis B. Most recipients achieve such a high anti-HBs level. The antibody titer may wane over the years, but even at lower titers there is excellent protection against HBV. A number of factors have been identified (smoking, obesity, chronic liver disease, and age over 50) that affect the antibody response, leading to lower levels of anti-HBs.155,156 These ‘hyporesponders’ may benefit from higher doses of vaccine. The response rate is also lower in immunocompromised patients, such as transplant recipients and patients receiving chemotherapy. Injection of vaccine into the buttocks elicits a lower rate of response than injection in the deltoid or anterolateral thigh. HBV vaccination elicits a lower response in hemodialysis patients, with only 50–60% responding adequately. The factors associated with a poor response are old age, presence of DR3, DR7 DQ2 and absence of A2 alleles.157 Hence patients with chronic renal insufficiency should be vaccinated early before renal disease progresses, so that a better response can be achieved.158 In a recent report, repeated dosing with intradermal vaccination (5 mg every 2 weeks, aiming for a titer of >1000 IU/l) achieved a response rate of 97.6%.159 Approximately 5–10% of vaccine recipients do not achieve detectable antibody levels (non-responders). Recent reports, mostly in animals, indicate that intradermal injection may produce a stronger humoral and cellular immunity than conventional intramuscular doses.160,161 Intradermal injection, by recruiting dendritic cells, stimulates primary MHC class I as well as class II restricted T-cell responses. In one study, all but one of nine non-responders to
conventional intramuscular dosing responded to two to three doses of vaccine given intradermally.161 Despite these interesting observations, recommendations for use of intradermal vaccination are lacking at present owing to limited data with regard to the lasting nature of antibody responses and inherent problems with standardization of intradermal delivery. Because vaccination results in a strong immunologic memory capable of preventing infection even in patients with very low or undetectable antibody titers, there is no role for booster doses in immunocompetent adults and children.162 Current recommendations for a booster dose only include patients undergoing hemodialysis. In these patients anti-HBs titer should be tested annually and a booster dose given if the titer is less than 10 mIU/ml.163 There are no serious side effects to vaccine. An increased incidence of neurologic disease, such as aseptic meningitis or Guillain– Barré syndrome, has not been observed.
High-Risk Groups Table 31-9 indicates the high-risk groups for whom vaccination is recommended. Targeted vaccination has not achieved its objective in certain high-risk groups, such as injecting drug users, but has been moderately successful in healthcare workers. Failure of strategies to vaccinate high-risk adults has led to policies of universal vaccination of newborns and preadolescent children in the US.
Vaccination Schedule The dosage of HBV vaccine and the schedule are listed in Table 3110. The typical schedule is 0, 1, and 6 months. The first two doses are important in the recruitment of responders and the third acts as a booster to achieve the highest anti-HBs titer. In immunocompromised patients and those on hemodialysis four doses are recommended, with the fourth intended to provide the highest possible titer. If a vaccination series is interrupted, the second dose should be administered as soon as possible.163 If the third dose is interrupted it should be given when convenient. The second and third doses should be separated at least by 2 months.163
Table 31-9. High-Risk Groups Requiring Consideration for HBV Vaccine Healthcare workers Public safety workers with likelihood of exposure to blood Staff and clients of institutions for developmentally disabled Hemodialysis patients Patients who are likely to require multiple transfusions with blood or blood products Household contacts and sex partners of HBV carriers or patients with acute hepatitis B International travelers to endemic areas who may have intimate contact with local population or who may take part in medical activities in endemic areas Injecting drug users Sexually active bisexual and homosexual men Sexually active heterosexual men and women if they have more than one partner Inmates of correctional facilities Patients with chronic liver disease Potential organ recipients
657
Section V. Liver Diseases Due to Infectious Agents
Table 31-10. Recommended Dosing for the Currently Available HBV Vaccines*
Infants* and children < 11 yrs Children/adolescents (11–19 yrs) Adults (≥20 yrs) Hemodialysis patients Immunocompromised patients
Recombivax HB (10 mg/ml)
Engerix-B (20 mg/ml)
2.5 mg 5 mg 10 mg 40 mg (1.0 ml)¶ 40 mg (1.0 ml)¶
10 mg 20 mg 20 mg 40 mg (2.0 ml)# 40 mg (2.0 ml)#
* The standard schedule is 0, 1, and 6 months. ** Infants born to HBs-negative mothers. ¶ Special formulation. # Two 1.0 ml doses administered at one site in four-dose schedule (0, 1, 2, 6 months).
Table 31-11. Hepatitis B Prophylaxis for Infants Born to an HbsagPositive Mother Vaccine
First dose Second dose Third dose¶
Recombivax HB 5 mg (0.5 ml)
HBIG or
Engerix B 10 mg (0.5 ml)
Age of the infant 0.5 ml IM*
5 mg (0.5 ml)
10 mg (0.5 ml)
None
Within 12 hours of birth 1 month
5 mg (0.5 ml)
10 mg (0.5 ml)
None
6 months
* HBIG should be administered at a site different from that used for vaccine. ¶ If four doses of vaccine are administered, the third dose is given at 2 months and the fourth at 12–18 months.
HBV vaccine is currently administered to all infants and children as a part of a universal immunization program in the US and in many countries worldwide. In Taiwan universal vaccination of newborns was implemented in 1984, and this has led to a decline in hepatitis B carrier rates among children, from 10% to less than 1%.164 Prevention by vaccination is also the best way to control the complications of chronic infection, such as hepatocellular carcinoma. In Taiwan this has proved to be effective in reducing the incidence of cancer to a quarter to a third of that in children born prior to 1984.165 Unfortunately, as of the year 2000 only 116 of 215 countries had implemented a universal vaccine program, owing to lack of resources and poor education as to its strategic importance.166 Combination vaccine, HBV with DPT and Hib (DTPw-HB/Hib, the current vaccine for immunization of infants) does not affect the immunogenicity of any of the components.167 Adolescents should also be vaccinated if they are at high risk. Table 31-10 lists the highrisk adult groups who are targeted for HBV vaccine.
Postexposure/Perinatal Prophylaxis Table 31-11 outlines the recommendations for prevention of perinatal transmission. Postexposure vaccination should be considered for any percutaneous, ocular or mucous membrane exposure. The type of immunoprophylaxis is determined by the HBsAg status of the source and the vaccination-response status of the exposed person. Table 31-12 outlines the postexposure prophylaxis for exposure to a known HBsAg-positive source.
658
Table 31-12. Postexposure Prophylaxis: if Source is HBsAg positive Vaccination status of exposed person
Immune prophylaxis
Unvaccinated
HBIG (0.06 ml/kg) and initiate HB vaccination series
Previously vaccinated Known responder* Known non-responder Antibody response unknown
No treatment HBIG ¥ 2 doses or HBIG ¥ 1 dose and initiate revaccination Test for anti-HBs If adequate* : no treatment If inadequate**: HBIG ¥ 1 dose and give vaccine booster
* Anti-HBs titer >10 mIU/ml. ** Anti-HBs titer <10 mIU/ml.
RECENT DEVELOPMENTS The immunogenicity of HBV vaccine may be enhanced by using more potent adjuvants. HBVsAg/AS04 vaccine contains 3¢-deacylated monophosphoryl lipid A and MF59-adjuvant vaccine contains surface and pre-S2 antigens. Both have been shown to elicit a better immune response.168–170 Concomitant use of granulocyte colonystimulating factor may also enhance the anti-HBs antibody response.171 Immunization using HBV DNA encoding for hepatitis B surface antigen and nucleoprotein has generated considerable interest as both a prophylactic and a therapeutic tool.172 These DNA-based vaccines, when given to antigen-processing tissues (intradermally or intramuscularly), elicit both humoral and cellular immunity as well as CD4+ and CD8+ T-cell responses. A novel delivery of vaccine to immunize the fetus in utero was successfully attempted in fetal lambs and pigs. HBV DNA vaccine was introduced into the amniotic fluid, and this resulted in high serum antibody and a cell-mediated immune response combined with local immunity in the oral cavity.173 An edible vaccine for hepatitis B is another area of interest. The DNA fragment encoding for HBs antigen was used to obtain transgenic lupin and transgenic lettuce that expressed envelope surface protein. Mice fed transgenic lupin developed hepatitis B-specific antibodies, and human volunteers fed transgenic lettuce developed specific serum IgG responses to plant-derived protein.174
HBV ESCAPE MUTANTS AND IMPLICATIONS FOR IMMUNIZATION Mutations in the HBV genome encoding for HBsAg can result in mutant HBV virus strains that escape neutralization by anti-HBs. The mutation involves the ‘a determinant’ and this has been associated with decreased binding to monoclonal anti-a epitope antibodies.28 Such mutants have been reported worldwide and are more common in areas endemic for HBV infection.175 This mutant virus accounts for some failures of vaccine to prevent perinatal transmission, but the extent to which this occurs varies considerably in the published studies. Accordingly, the area remains controversial. It has been argued that this may diminish the success of vaccine programs in the future, and modification of vaccines to include the mutant antigen have been proposed to deal with this possibility.
Chapter 31 HEPATITIS B
REFERENCES 1. Chang MH, Chen CJ, Lai MS, et al. Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. Taiwan Childhood Hepatoma Study Group. N Engl J Med 1997;336:1855–1859. 2. Alter MJ. Epidemiology of hepatitis B in Europe and worldwide. J Hepatol 2003;39(Suppl 1):S64–69. 3. Lee WM. Hepatitis B virus infection. N Engl J Med 1997;337:1733–1745. 4. Grob P, Jilg W, Bornhak H, et al. Serological pattern ‘anti-HBc alone’: report of a workshop. J Med Virol 2000;62:450–455. 5. Dickson RC, Everhart JE, Lake JR, et al. Transmission of hepatitis B by transplantation of livers from donors positive for antibody to hepatitis B core antigen. The National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database. Gastroenterology 1997;113:1668–1674. 6. Mason A, Wick M, White H, et al. Hepatitis B virus replication in diverse cell types during chronic hepatitis B virus infection. Hepatology 1993;18:78–789. 7. Mason A, Yoffe B, Noonan C, et al. Hepatitis B virus DNA in peripheral-blood mononuclear cells in chronic hepatitis B after HBsAg clearance. Hepatology 1992;16:36–41. 8. Kuhns M, McNamara A, Mason A, et al. Serum and liver hepatitis B virus DNA in chronic hepatitis B after sustained loss of surface antigen. Gastroenterology 1992;103:1649–1656. 9. Fong TL, Bunnapradist S, Jordan SC, et al. Impact of hepatitis B core antibody status on outcomes of cadaveric renal transplantation: analysis of United Network of Organ Sharing database between 1994 and 1999. Transplantation 2002;73:85–89. 10. Goldstein ST, Alter MJ, Williams IT, et al. Incidence and risk factors for acute hepatitis B in the United States, 1982–1998: Implications for vaccination programs. J Infect Dis 2002;185:713–719. 11. Alter MJ. Epidemiology and prevention of hepatitis B. Semin Liver Dis 2003;23:39–46. 12. McQuillan GM, Coleman PJ, Kruszon-Moran D, et al. Prevalence of hepatitis B virus infection in the United States: the National Health and Nutrition Examination Surveys, 1976 through 1994. Am J Public Health 1999;89:14–18. 13. AS Lok, Heathcote EJ, Hoofnagle JH. Management of hepatitis B: 2000 – summary of workshop. Gastroenterology 2001;120:1828–1853. 14. Fattovich G. Natural history of hepatitis B. J Hepatol 2003;39(Suppl 1):S50–58. 15. Weissberg JI, Andres LL, Smith CI, et al. Survival in chronic hepatitis B. An analysis of 379 patients. Ann Intern Med 1984;101:613–616. 16. Cardenas CL, Soetikno R, Robinson WS, et al. Long-term follow-up of patients with chronic hepatitis B: A 25 year prospective study. Hepatology 1999;30:300A [abstract]. 17. de Jongh FE, Janssen HLA, de Man RA, et al. Survival and prognostic indicators in hepatitis B surface antigen-positive cirrhosis of the liver. Gastroenterology 1992;103:1630–1635. 18. Niederau C, Heintgen T, Lange S, et al. Long-term follow-up of HBeAg-positive patients treated with interferon alfa for chronic hepatitis B. N Engl J Med 1996;334:1422–1427. 19. Nowak MA, Bonhoeffer S, Hill AM, et al. Viral dynamics in hepatitis B virus infection. Proc Natl Acad Sci USA 1996;93:4398–4402. 20. Bartholomeusz A, Schaefer S. Hepatitis B virus genotypes: comparison of genotyping methods. Rev Med Virol 2004;14:3–16. 21. Chu CJ, Keeffe EB, Han SH, et al. Hepatitis B virus genotypes in the United States: results of a nationwide study. Gastroenterology 2003;125:444–451.
22. Kidd-Lunjunggre K, Miyakawa Y, Kidd AH. Genetic variability in hepatitis B viruses. J Gen Virol 2002;83:1267–1280. 23. Chu CJ, Lok AS. Clinical significance of hepatitis B virus genotypes. Hepatology 2002;35:1274–1276. 24. Kao JH, Chen PJ, Chen DS. Hepatitis B genotypes correlate with clinical outcomes in patients with chronic hepatitis B. Gastroenterology 2000;118:554–559. 25. Orito E, Ichida T, Sakugawa H, et al. Geographic distribution of hepatitis B virus (HBV) genotype in patients with chronic HBV infection in Japan. Hepatology 2001;34:590–594. 26. Nainan OV, Khristova ML, Byun K-S, et al. Frequency and significance of hepatitis B virus antibody resistant mutants. Antiviral Ther 2000;5(Suppl):29–30 [Abstract]. 27. Ghany MG, Ayola B, Villamil FG, et al. Hepatitis B virus S mutants in liver transplant recipients who were reinfected despite hepatitis B immune globulin prophylaxis. Hepatology 1998;27:213–222. 28. Locarnini S, McMillan J, Bartholomeusz A. The hepatitis B virus and common mutants. Semin Liver Dis 2003;23:5–20. 29. Li J, Buckwold VE, Hon MW, Ou JH. Mechanism of suppression of hepatitis B virus precore RNA transcription by a frequent double mutation. J Virol 1999;73:1239–1244. 30. Lindh M, Andersson AS, Gusdal A. Genotypes, nt 1858 variants, and geographic origin of hepatitis B virus – large scale analysis using a new genotyping method. J Infect Dis 1997;175:1285–1293. 31. Chu CJ, Keeffe EB, Han SH, et al. Prevalence of HBV precore/core promoter variants in the United States. Hepatology 2003;38:619–628. 32. Bertoletti A, Sette A, Chisari FV, et al. Natural variants of cytotoxic epitopes are T-cell receptor antagonist for antiviral cytotoxic T cells. Nature 1994;369:407–410. 33. Naoumov NV, Thomas MG, Mason AL, et al. Genomic variations in the hepatitis B core gene: a possible factor influencing response to interferon alfa treatment. Gastroenterology 1995;108:505–514. 34. Torre F, Cramp M, Owsianka A, et al. Direct evidence that naturally occurring mutations within hepatitis B core epitope alter CD4+ T-cell reactivity. J Med Virol 2004;72:370–376. 35. Bozkaya H, Akarca US, Ayola B, et al. High degree of conservation in the hepatitis B virus core gene during the immune tolerant phase in perinatally acquired chronic hepatitis B infection. J Hepatol 1997;26:508–516. 36. Naoumov N, Hou JL. The emerging role of hepatitis B virus genotypes. Curr Hepatitis Rep 2004;3:145–151. 37. Lok As, Akarca U, Greene S. Mutations in the pre-core region of hepatitis B virus serve to enhance the stability of the secondary structure of the pre-genome encapsidation signal. Proc Natl Acad Sci USA 1994;91:4077–4081. 38. Lai C-L, Dienstag J, Schiff E, et al. Prevalence and clinical correlates of YMDD variants during lamivudine therapy for patients with chronic hepatitis B. Clin Infect Dis 2003;36:687–696. 39. Dienstag JL, Goldin RD, Heathcote EJ, et al. Histological outcome during long-term lamivudine therapy. Gastroenterology 2003;124:105–117. 40. Liaw YF. Impact of YMDD mutations during lamivudine therapy in patients with chronic hepatitis B. Antivir Chem Chemother 2001;12(Suppl 1):67–71. 41. Mutimer D, Pillay D, Shields P, et al. Outcome of lamivudine resistant hepatitis B virus infection in the liver transplant recipient. Gut 2000;46:107–113. 42. Hadziyannis SJ, Vassilopoulos D. Hepatitis B e antigen-negative chronic hepatitis B. Hepatology 2001;34:617–624. 43. Fattovich G. Natural history and prognosis of hepatitis B. Semin Liver Dis 2003;23:47–58. 44. Fattovich G, Pantalena M, Zagni I, et al. Effect of hepatitis B and C virus infection on the natural history of compensated
659
Section V. Liver Diseases Due to Infectious Agents
45. 46.
47.
48.
49. 50. 51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
660
cirrhosis: a cohort study of 297 patients. Am J Gastroenterol 2002;97:2886–2895. McMahon BJ. The natural history of chronic hepatitis B virus infection. Semin Liver Dis 2004;24(Suppl 1):17–21. Rehermann B, Ferrari C, Pasquinelli C, Chisari FV. The hepatitis B virus persists for decades after patients’ recovery from acute viral hepatitis despite active maintenance of a cytotoxic T lymphocyte response. Nature Med 1996;2:1104–1108. Samuel D, Muller R, Alexander G, et al. Liver transplantation in European patients with the hepatitis B surface antigen. N Engl J Med 1993;329:1842–1847. Williams AL, Hoofnagle JH. Ratio of serum aspartate to alanine aminotransferase in chronic hepatitis. Relationship to cirrhosis. Gastroenterology 1988;95:734–739. Prysopoulos NT, Reddy KR. Extrahepatic manifestations of chronic viral hepatitis. Curr Gastroenterol Rep 2001;3:71–78. Theal JJ, Mason A. Pathogenesis of extrahepatic manifestations of hepatitis B. Curr Hepatitis Rep 2004;3:160–166. Lai KN, Li PK, Lui SF, et al. Membranous nephropathy related to hepatitis B virus in adults. N Engl J Med 1991;324:1457–1463. Conjeevaram HS, Hoofnagle JH, Austin HA, et al. Long-term outcome of hepatitis B virus-related glomerulonephritis after therapy with interferon alfa. Gastroenterology 1995;109:540– 546. Di Martino V, Thevenot T, Colin JF, et al. Influence of HIV infection on the response to interferon therapy and the longterm outcome of chronic hepatitis B. Gastroenterology 2002;123:1812–1822. Xunrong L, Yan AW, Liang R, et al. Hepatitis B virus (HBV) reactivation after cytotoxic or immunosuppressive therapy – pathogenesis and management. Rev Med Virol 2001;11:287–299. Lau JYN, Lai CL, Lin HJ, et al. Fatal reactivation of chronic hepatitis B virus infection following withdrawal of chemotherapy in lymphoma patients. Q J Med 1989;73:911–917. Perrillo RP. Acute flares in chronic hepatitis B: the natural and unnatural history of an immunologically mediated liver disease. Gastroenterology 2001;120:1009–1022. Liang R, Lau GKK, Kwong YL. Chemotherapy and bone marrow transplantation for cancer patients who are also chronic hepatitis B carriers: a review of the problem. J Clin Oncol 1999;17:394–398. Nair S, Perrillo R. Serum alanine aminotransferase flares during interferon treatment of chronic hepatitis B: is sustained clearance of HBV DNA dependent on levels of pretreatment viremia? Hepatology 2001;34:1021–1026. Hoofnagle JH, Di Bisceglie AM, Waggoner JG, et al. Interferon alfa for patients with clinically apparent cirrhosis due to chronic hepatitis B. Gastroenterology 1993;104:1116–1121. Honkoop P, de Man RA, Niesters HG, et al. Acute exacerbation of chronic hepatitis B virus infection after withdrawal of lamivudine therapy. Hepatology 2000;32:635–639. Liaw YF, Chien RN, Yeh CT, et al. Acute exacerbation and hepatitis B virus clearance after emergence of YMDD motif mutation during lamivudine therapy. Hepatology 1999;30:567–572. Lok ASF, Lai C-L, Leung N, et al. Long-term safety of lamivudine treatment in patients with chronic hepatitis B. Gastroenterology 2003;125:1714–1722. Krogsgaard K, Marcellin P, Trepo C, et al. Prednisolone withdrawal therapy enhances the effect of human lymphoblastoid interferon in chronic hepatitis B. INTREPED Trial Group. J Hepatol 1996;25:803–813. Liaw YF, Tsai SL, Chien RN, et al. Prednisolone priming enhances Th1 response and efficacy of subsequent lamivudine therapy in patients with chronic hepatitis B. Hepatology 2000;32:604–609.
65. Thio CL. Hepatitis B in the human immunodeficiency virusinfected patient: epidemiology, natural history, and treatment. Semin Liver Dis 2003;23:125–136. 66. Bonino F, Brunetto MR. Chronic hepatitis B e antigen (HBeAg) negative, anti-HBe positive hepatitis B: an overview. J Hepatol 2003;39(Suppl 1):160–163. 67. Oketani M, Oketani K, Xiaohong C, et al. Low level wild-type and pre-core mutant hepatitis B viruses and HBeAg negative reactivation of chronic hepatitis B. J Med Virol 1999;58:332–337. 68. Gerner P, Lausch E, Friedt M, et al. Hepatitis B virus core promoter mutations in children with multiple anti-HBe/HBeAg reactivations result in enhanced promoter activity. J Med Virol 1999;59:415–423. 69. Steinberg JL, Yeo W, Zhong S, et al. Hepatitis B virus reactivation in patients undergoing cytotoxic chemotherapy for solid tumours: precore/core mutations may play an important role. J Med Virol 2000;60:249–255. 70. Liaw YF, Chen YC, Sheen IS, et al. Impact of acute hepatitis C virus superinfection in patients with chronic hepatitis B virus infection. Gastroenterology 2004;126:1024–1029. 71. Shiels MT, Taswell HF, Czaja AJ, et al. Frequency and significance of concurrent hepatitis B surface antigen and antibody in acute and chronic hepatitis B. Gastroenterology 1987;93:675–680. 72. Tsang TK, Blei AT, O’Reilly DJ, et al. Clinical significance of concurrent hepatitis B surface antigen and antibody positivity. Dig Dis Sci 1986;31:620–624. 73. Grob P, Jilg W, Bornhak H, et al. Serological pattern ‘anti-HBc’: report on a workshop. J Med Virol 2000;62:450–455. 74. Weber B, Melchior W, Gehrke R, et al. Hepatitis B virus markers in anti-HBc only positive individuals. J Med Virol 2001;64:312–319. 75. Wachs ME, Amend WJ, Ascher NL, et al. The risk of transmission of hepatitis B from HBsAg(–), HBcAb(+), HBIgM(–) organ donors. Transplantation 1995;59:230–234. 76. Dodson SF, Issa S, Araya V, et al. Infectivity of hepatic allografts with antibodies to hepatitis B virus. Transplantation 1997;64:1582–1584. 77. Paterlini P, Gerken G, Nakajima E, et al. Polymerase chain reaction to detect hepatitis B virus DNA and RNA sequences in primary liver cancers from patients negative for hepatitis B surface antigen. N Engl J Med 1990;323:80–85. 78. Fukai K, Yokosuka O, Fujiwara K, et al. Etiologic considerations of fulminant non-A, non-B viral hepatitis in Japan: analyses by nucleic acid amplification method. J Infect Dis 1998;178: 325–333. 79. Chu C-J, Hussain M, Lok ASF. Quantitative serum HBV DNA levels during different stages of chronic hepatitis B infection. Hepatology 2002;36:1408–1415. 80. Perrillo RP, Schiff ER, Davis GL, et al. A randomized, controlled trial of interferon alfa-2b alone and after prednisone withdrawal for the treatment of chronic hepatitis B. The Hepatitis Interventional Therapy Group. N Engl J Med 1990;323:295–301. 81. Gauthier J, Bourne EJ, Lutz MW, et al. Quantitation of hepatitis B viremia and emergence of YMDD variants in patients with chronic hepatitis B treated with lamivudine. J Infect Dis 1999;180:1757–1762. 82. Werle B, Cinquin K, Marcellin P, et al. Evolution of hepatitis B viral load and viral genome sequence during adefovir dipivoxil therapy. J Viral Hepatol 2004;11:74–83. 83. van Nunen AB, Hansen BE, Suh DJ, et al. Durability of HBeAg seroconversion following antiviral therapy for chronic hepatitis B: relation to type of therapy and pretreatment serum hepatitis B virus DNA and alanine aminotransferase. Gut 2003;52:420–424. 84. Lai CL, Dienstag J. Schiff E, et al. Prevalence and clinical correlates of YMDD variants during lamivudine therapy for
Chapter 31 HEPATITIS B
85.
86.
87. 88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98. 99.
100.
101.
102.
103.
104.
patients with chronic hepatitis B. Clin Infect Dis 2003;36:687–696. Locarnini S, Hatzakis A, Heathcote J, et al. Management of antiviral resistance in patients with chronic hepatitis B. Antiviral Ther 2004;9:679–693. Mutimer D, Pillay D, Dragon E, et al. High pre-treatment serum hepatitis B virus titre predicts failure of lamivudine prophylaxis and graft re-infection after liver transplantation. J Hepatol 1999;30:715–721. Samuel D. Management of hepatitis B in liver transplantation patients. Semin Liver Dis 2004;24:55–62. Roche B, Samuel D, Gigou M, et al. De novo and apparent de novo hepatitis B virus infection after liver transplantation. J Hepatol 1997;26:517–526. Brechot C, Degos F, Lugassy C, et al. Hepatitis B virus DNA in patients with chronic liver disease and negative tests for hepatitis B surface antigen. N Engl J Med 1985;312:270–276. Wright TL, Mamish D, Combs C, et al. Hepatitis B virus and apparent fulminant non-A, non hepatitis. Lancet 1992;339:952–955. Wong DKH, Cheung AM, O’Rourke K, et al. Effect of alpha interferon treatment in patients with hepatitis B e antigenpositive chronic hepatitis B. A meta-analysis. Ann Intern Med 1993;119:312–323. Lau DT, Everhart J, Keiner DE, et al. Long-term follow up of patients with chronic hepatitis B treated with interferon alfa. Gastroenterology 1997;113:1660–1667. Lok ASF, McMahon BJ. Practice Guidelines Committee, American Association for the Study of Liver Diseases (AASLD). Chronic hepatitis B: update of recommendations. Hepatology 2004;39:857–861. Liaw YF, Leung N, Guan R, et al. Asian-Pacific consensus statement on the management of chronic hepatitis B: A 2005 update. Liver International 2005;25:472–489. EASL Jury. EASL International Consensus Conference on Hepatitis B: 13–14 September 2002, Geneva, Switzerland: consensus statement (long version). J Hepatol 2003;39:S3–S25. Perrillo RP, Lai CL, Liaw YF, et al. Predictors of HBeAg loss after lamivudine treatment for chronic hepatitis B. Hepatology 2002;36:186–194. Keeffe EB, Dieterich DT, Han S-HB, et al. A treatment algorithm for the management of chronic hepatitis B virus infection in the United States. Clin Gastroenterol Hepatol 2004;2:87–106. Cooksley WG. Treatment with interferons in patients with hepatitis B. Semin Liver Dis 2004;24(Suppl 1):45–54. Hansen HLA, Senturk H, Zeuzem S, et al. Peginterferon alfa-2b and lamivudine combination therapy compared with peginterferon alfa-2b for chronic HBeAg positive chronic hepatitis B: a randomized, controlled trial in 307 patients. Presented at the 54th Annual Meeting of the American Association for the Study of Liver Diseases; 28 October 2003, Boston, MA. Lau G, Piratvisuth T, Luo KX, et al. PEGinterferon alfa-2a, Iamivudine and the combination for HBeAg-positive chronic hepatitis B. N Engl J Med 2005;352:2682–2695. Marcellin P, Lau GK, Bonino F, et al. Peginterferon alfa-2a alone, lamivudine alone, and the two in combination in patients with HBeAg-negative chronic hepatitis B. N Engl J Med 2004;351:1206–1217. Kao JH, Wu NH, Chen PJ, et al. Hepatitis B genotypes and the response to interferon therapy. J Hepatol 2000;33: 998–1002. Lai CL, Chien RN, Leung NW, et al. A one-year trial of lamivudine for chronic hepatitis. N Engl J Med 1998;339: 61–68. Dienstag JL, Schiff ER, Wright TL, et al. Lamivudine as initial treatment for chronic hepatitis B in the United States. N Engl J Med 1999;341:1256–1263.
105. Liaw YF, Leung NWY, Chang TT, et al. Effects of extended lamivudine therapy in Asian patients with chronic hepatitis B. Gastroenterology 2000;119:172–180. 106. Marcellin P, Chang TT, Lim SG, et al. Adefovir dipivoxil for the treatment of hepatitis antigen-positive chronic hepatitis B. N Engl J Med 2003;348:808–816. 107. Hadziyannis SJ, Tassopoulos NC, Heathcote EJ, et al. Adefovir dipivoxil for the treatment of hepatitis B e antigen-negative chronic hepatitis B. N Engl J Med 2003;348:800–807. 108. Marcellin P, Chang T, Lim S, et al. Adefovir dipivoxil (ADV) 10 mg for the treatment of patients with HBeAg+ chronic hepatitis B: continued efficacy beyond 48 weeks. Hepatology 2002;36:373A [abstract]. 109. Westland C, Delaney W 4th, Yang H, et al. Hepatitis B virus genotypes and virologic response in 694 patients in phase III studies of adefovir dipivoxil. Gastroenterology 2003;125:107–116. 110. Angus P, Vaughan R, Xiong S, et al. Resistance to adefovir dipivoxil therapy associated with the selection of a novel mutation in the HBV polymerase. Gastroenterology 2003;125:292–297. 111. Hadziyannis S, Tassopoulos N, Chang TT, et al. Three year study of adefovir dipivoxil demonstrates sustained efficacy in presumed precore mutant chronic hepatitis B (CHB) patients in a long term safety and efficacy study (LTSES). J Hepatol 2004;40:17 [abstract]. 112. Yang H, Westland CE, Das K, et al. In vitro characterization and molecular modeling analysis of adefovir resistance mutation rtN236T in the HBV polymerase. Programs and abstracts from the Hep Dart Meeting, 14–18 December 2003, Kauai, Hawaii. Abstract 054. 113. Perrillo R, Hann H-W, Mutimer D, et al. Adefovir dipivoxil added to ongoing lamivudine in chronic hepatitis B with YMDD mutant hepatitis B virus. Gastroenterology 2004;126:81–90. 114. Peters MG, Hann HW, Martin P, et al. Adefovir dipivoxil alone or in combination with lamivudine in patients with lamivudineresistant chronic hepatitis B. Gastroenterology 2004;126:91–100. 115. Schiff ER, Lai C-L, Hadziyannis S, et al. Adefovir dipivoxil therapy for lamivudine-resistant hepatitis B in pre- and post-liver transplantation patients. Hepatology 2003;38:1419–1427. 116. Shiffman M, Ng TM, Krastev Z, et al. A double-blind, placebocontrolled trial of emtricitabine (FTC, Emtriva) administered once daily for treatment of chronic hepatitis B virus (HBV) infection. Hepatology 2004;40:172A [abstract]. 117. Raffi F, Snow A, Borroto-Esoda K, et al. Anti-HBV activity of emtricitabine (FTC) in patients co-infected with HIV and hepatitis B virus. Programs and abstracts of the 2nd IAS Conference on Pathogenesis and Treatment; 13–16 July 2003, Paris, France. Abstract 215. 118. Gish R, Leung N, Wang C, et al. Antiviral activity, safety, and incidence of resistance in chronically infected hepatitis B patients (CHB) given once daily emtricitabine for 2 years. Hepatology 2002;36:372A [abstract]. 119. van Bommel F, Wunsche T, Mauss S, et al. Comparison of adefovir and tenofovir in the treatment of lamivudine-resistant hepatitis B virus infection. Hepatology 2004;40:1421–1425. 120. Ristig MB, Crippin J, Aberg JA, et al. Tenofovir disoproxil fumarate therapy for chronic hepatitis B in human immunodeficiency virus/hepatitis B virus-coinfected individuals for whom interferon-alpha and lamivudine therapy have failed. J Infect Dis 2002;186:1844–1847. 121. Nelson M, Portsmouth S, Stebbing J, et al. An open-label study of tenofovir in HIV-1 and hepatitis B virus co-infected individuals. AIDS 2003;17:F7–10. 122. Kuo A, Dienstag JL, Chung RT. Tenofovir disoproxil fumarate for the treatment of lamivudine-resistant hepatitis B. Clin Gastroenterol Hepatol 2004;2:266–272.
661
Section V. Liver Diseases Due to Infectious Agents
123. Lai C-L, Rosmawati M, Lao J, et al. Entecavir is superior to lamivudine in reducing hepatitis B virus DNA in patients with chronic hepatitis B infection. Gastroenterology 2002;123:1831–1838. 124. Chang TT, Gish R, de Man R, et al. Entecavir is superior to lamivudine for the treatment of HBeAg (+) chronic hepatitis B: Results of a phase III study ETV-022 in nucleoside-naïve patients. Hepatology 2004;193A [abstract]. 125. Shouval D, Lai C-L, Cheinquer H, et al. Entecavir demonstrates superior histologic and virologic efficacy over lamivudine in nucleoside-naïve HBeAg (–) chronic hepatitis B. Hepatology 2004;728A [abstract]. 126. Chang T-T, Hadziyannis S, Cianciara J, et al. Sustained viral load and ALT reduction following 48 weeks of entecavir treatment in subjects with chronic hepatitis B who have failed lamivudine. Hepatology 2002;36:300A [abstract]. 127. Tenney DJ, Langley DR, Oliver AJ, et al. Hepatitis B virus resistance to entecavir involves novel changes in the the viral polymerase. Hepatology 2004;40:245A [abstract]. 128. Standring DN, Bridges EG, Placidi L, et al. Antiviral b-Lnucleosides specific for hepatis B virus infection. Antivir Chem Chemother 2001;12(Suppl 1):119–129. 129. Tennant B, Menne S, Baldwin B, et al. Antiviral activities of b-L-thymidine (LdT), Val-b-L-2¢-deoxycytidine (val-LdC), and lamivudine alone and in combination in a 12 week, oral dose study using the woodchuck model of chronic hepatitis B virus infection. Programs and abstracts from the Hep Dart Meeting;16–20 December 2001, Maui, Hawaii. Abstract 027. 130. Lai C-L, Leung NWY, Teo E-K, et al. A 1-year trial of telbivudine lamivudine and the combination in patients with Hepatitis B e antigen-positive chronic hepatitis B. Gastroenterology 2005;129:528–536. 131. Yoo BC, Koh KC, Chung Y-H. Clevudine is highly efficacious in HBeAg(-) chronic hepatitis B patients with a sustained antiviral effect after cessation of therapy. Hepatology 2005;42: 268A[abstract]. 132. Marcellin P, Mommeja-Marin H, Sacks SL, et al. A phase II dose-escalating trial of clevudine in patients with chronic hepatitis B. Hepatology 2004;40:140–148. 133. Cavanaugh VJ, Guidotti LG, Chisari FV. Interleukin-12 inhibits hepatitis B virus replication in the livers of transgenic mice. J Virol 1997;71:3236–3243. 134. Kimura K, Kakimi K, Wieland S, et al. Interleukin-18 inhibits hepatitis B virus replication in the livers of transgenic mice. J Virol 2002;76:10702–10707. 135. Klucher K, Kindsvogel V, Henderson K, et al. Interleukin-28 and interleukin-29 have antiviral activity against hepatitis viruses. Programs and abstracts from the Hep Dart Meeting; 14–18 December 2003, Kauai, Hawaii. Abstract 044. 136. Mutchnick M, Lindsay KL, Schiff ER, et al. Thymosin alpha 1 treatment of chronic hep B: results of a phase III multicentre, randomized, double-blind and placebo controlled study. Viral Hepatol 1999;6:397–403. 137. Saruc M, Ozden N, Turkel N, et al. Long-term outcomes of thymosin-alpha 1 and interferon alpha-2b combination therapy in patients with hepatitis B e antigen (HBeAg) negative chronic hepatitis B. J Pharm Sci 2003;92:1386–1395. 138. Safadi R, Israeli E, Papo O, et al. Treatment of chronic hepatitis B virus infection via oral immune regulation toward hepatitis B virus proteins. Am J Gastroenterol 2003;98: 2505–2515. 139. Galun E, Eren R, Safadi R, et al. Clinical evaluation (phase I) of a combination of two human monoclonal antibodies to HBV: Safety and antiviral properties. Hepatology 2002;35:673–679. 140. Galun E, Safadi R, Milchem A, et al. A phase 2 clinical study evaluating safety and efficacy of HEPEX-B, a mixture of two human monoclonal antibodies to HBsAg in combination with
662
141.
142.
143.
144.
145.
146. 147.
148.
149.
150.
151.
152.
153.
154.
155. 156.
157.
158.
159.
lamivudine in chronic HBV patients. Hepatology 2003;38:707A [abstract]. McCaffrey AP, Nakai H, Pandey K, et al. Inhibition of hepatitis B virus in mice by RNA interference. Nature Biotechnol 2003;21:639–644. Klein C, Bock CT, Wedemeyer H, et al. Inhibition of hepatitis B virus replication in vivo by nucleoside analogs and siRNA. Gastroenterology 2003;125:9–18. Delaney WE IV, Edwards R, Colledge D, et al. The phenylpropenamide derivatives AT-61 and AT-130 inhibit replication of both wild-type and lamivudine resistant strains of hepatitis B virus in vitro. Antimicrob Agents Chemother 2003;46:3057–3060. Feld J, Locarnini S. New targets and possible new therapeutic approaches in the chemotherapy of chronic hepatitis B. Hepatology 2003;38:545–553. Brook M, Gilson R, Wilkins E. BHIVA guidelines for treatment and management of HIV and hepatitis B coinfection. http://www.bhiva.org/guidelines/2003/HBV/index.html. Yachimski P, Chung RT. Hepatitis B virus infection in HIVcoinfected patients. Curr Hepatitis Rep 2004;3:138–144. Liaw YF, Sung JJY, Chow WC, et al. Lamivudine for patients with chronic hepatitis B and advanced liver disease. N Engl J Med 2004;351:1521–1531. Sung JJY, Lai JY, Zeuzem S, et al. A randomized double-blind phase II study of lamivudine compared to lamivudine plus adefovir dipivoxil for treatment naive patients with chronic hepatitis B: week 52 analysis. J Hepatol 2003;38(Suppl 2):25 [abstract]. Lau G, Cooksley H, Ribeiro RM. Randomized, double-blind study comparing adefovir dipivoxil (ADV) plus emtricabine (FTC) combination therapy versus ADV alone in HbeAg (+) chronic hepatitis B: efficacy and mechanisms of treatment response. Hepatology 2004;40: 272A [abstract]. Redeker AG, Mosley JW, Gocke DJ, et al. Hepatitis immune globulin as a prophylactic measure for spouses exposed to acute type B hepatitis. N Engl J Med 1975;293:1055–1059. Wong VC, IP HM, Reesink HW, et al. Prevention of the HBsAg carrier state in newborn infants of mothers who are chronic carriers of HBsAg and HBeAg by administration of hepatitis-B vaccine and hepatitis-B immunoglobulin. Double-blind randomised placebo controlled study. Lancet 1984;1:921–926. Kohler PF, Dubois RS, Merrill DA, et al. Prevention of chronic neonatal hepatitis B virus infection with antibody to the hepatitis B surface antigen. N Engl J Med 1974;291:1378–1380. Reesink HW, Reerink-Brongers F, Lafeber-Schut BJ, et al. Prevention of chronic HBsAg carrier state in infants of HBs Agpositive mothers by hepatitis B immunoglobulin. Lancet 1979;2:436–438. Bocher WO, Herzog-Hauff S, Herr W, et al. Regulation of the neutralizing anti-hepatitis B surface (HBs) antibody response in vitro in HBs vaccine recipients and patients with acute or chronic hepatitis B virus (HBV) infection. Clin Exp Immunol 1996;105:52–58. Keeffe EB, Krause DS. Hepatitis B vaccination of patients with chronic liver disease. Liver Transpl Surg 1998;4:437–439. Wiedmann M, Liebert UG, Oesen U, et al. Decreased immunogenicity of recombinant hepatitis B vaccine in chronic hepatitis B. Hepatology 2000;31:230–234. Peces R, de la Torre M, Alcazar R, et al. Prospective analysis of the factors influencing the antibody response to hepatitis B vaccine in hemodialysis patients. Am J Kidney Dis 1997;29:239–245. Seaworth B, Drucker J, Starling J, et al. Hepatitis B vaccine in patients with chronic renal failure before dialysis. J Infect Dis 1988;157:332–337. Charest AF, McDougall J, Goldstein MB. A randomized comparison of intradermal and intravenous vaccination against hepatitis B virus in incident chronic hemodialysis patients. Am J Kidney Dis 2000;36:976–982.
Chapter 31 HEPATITIS B
160. Wilson CC, Olson WC, Tuting T, et al. HIV-1 specific CTL responses primed in vitro by blood derived dendritic cells and Th1-biasing cytokines. J Immunol 1999;162:3070–3078. 161. Rahman F, Dahmen A, Herzog-Haff S, et al. Cellular and humoral immune response induced by intradermal or intramuscular vaccination with the major hepatitis B surface antigen. Hepatology 2000;31:521–527 162. Banatvala JE, Van Damme P. Hepatitis B vaccine – do we need boosters? J Viral Hepatol 2003;10:1–6. 163. Hepatitis B virus: a comprehensive strategy for eliminating transmission in the United States through universal childhood vaccination. Recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR Morb Mortal Wkly Rep 1991;40 (RR-13): 1–25. 164. Chan CY, Lee SD, Lo KJ. Legend of hepatitis B vaccine: the Taiwan experience. J Gastroenterol Hepatol 2004;19:121–126. 165. Chang MH. Decreasing incidence of hepatocellular carcinoma among children following universal hepatitis B immunization. Liver Int 2003;23:309–314. 166. Alter MJ. Epidemiology and prevention of hepatitis B. Semin Liver Dis 2003;23:39–46. 167. Lopez P, Rubiano L, del Pilar Rubio M, et al. Immunogenicity and reactogenicity of DTPw-HB/Hib vaccine administered to Colombian infants after a birth dose of hepatitis B vaccine. Expert Rev Vaccines 2002;1:277–283. 168. Jacques P, Moens G, Desombere I, et al. The immunogenicity and reactogenicity profile of a candidate hepatitis B vaccine in an
169.
170.
171.
172.
173.
174.
175.
adult vaccine non-responder population. Vaccine 2002;20:3644–3649. Levie K, Gjorup I, Skinhoj P, et al. A 2-dose regimen of a recombinant hepatitis B vaccine with the immune stimulant AS04 compared with the standard 3-dose regimen of Engerix-B in healthy young adults. Scand J Infect Dis 2002;34:610–614. Lewis DJ, Eiden JE, Goilav C, et al. Rapid and frequent induction of protective immunity exceeding UK recommendations for healthcare settings by MF59-adjuvated hepatitis B vaccine. Commun Dis Public Health 2003;6:320–324. Kim MJ, Nafziger AN, Harro CD, et al. Revaccination of healthy nonresponders with hepatitis B vaccine and prediction of seroprotection response. Vaccine 2003;21:1174–1179. Thermet A, Rollier C, Zoulim F, et al. Progress in DNA vaccine for prophylaxis and therapy of hepatitis B. Vaccine 2003;21:659–662 Gerdts V, Babiuk LA, van Drunen Little-van den Hurk, et al. Fetal immunization by a DNA vaccine delivered into the oral cavity. Nature Med 2000;6:929–932. Kapusta J, Modelska A, Figlerowicz M, et al. A plant derived edible vaccine against hepatitis B virus FASEB J 1999;13:1796–1799. Nomura HC, Itoga S, Isobe K, et al. Prevalence of vaccineescape mutants of hepatitis B virus in the adult population in China: a prospective study in 176 restaurant employees. J Gastroenterol Hepatol 2001;16:1373–1377.
663
Section V. Liver Diseases Due to Infectious Agents
32
Hepatitis C Teresa L. Wright and Michael P. Manns Abbreviations ACTG AIDS Clinical Trials Group ALT alanine aminotransferase APRICOT AIDS PEGASYS Ribavirin International Coinfection Trial CDC Centers for Disease Control and Prevention COPILOT Colchicine versus Peg-Intron Long-Term DC dendritic cells EIA enzyme immunoassay EPIC-3 Evaluation of Peg-Intron in Control of Hepatitis C Cirrhosis ETR end of treatment response EVR early virological response FDA Food and Drug Administration HAART highly active antiretroviral therapy
HALT-C HCC HCV HIV IgG IMPDH MC MELD MIU MPGN NHANES
Hepatitis C Antiviral Long-term Treatment Against Cirrhosis hepatocellular carcinoma hepatitis C virus human immunodeficiency virus immunoglobulin G inosine monophosphate dehydrogenase essential mixed cryoglobulinemia model for end-stage liver disease million international units membranoproliferative glomerulonephritis National Health and Nutrition Examination Survey
NHL NK PCR PEG-IFNs RIBA RVR SEER SVR tiw TMA TSH
non-Hodgkin’s lymphoma natural killer polymerase chain reaction pegylated interferons recombinant immunoblot assays rapid virological response Surveillance, Epidemiology, and End Results sustained virological response three times a week transcription-mediated amplification thyroid-stimulating hormone
INTRODUCTION
EPIDEMIOLOGY
Hepatitis C virus (HCV) is a major public health problem, not only in the USA but also around the world.1 In many countries, the prevalence of infection is extrapolated from the prevalence in blood donors and there have been few detailed studies of prevalence in the general population. The virus causing non-A, non-B hepatitis was only identified in 1989, and the first serological assays became available in 1990. Thus, in a relatively short period of time (approximately 15 years) our knowledge of the epidemiology of this infection as well as of the mechanisms of viral replication and pathogenesis has greatly advanced such that now specific therapies to inhibit HCV replication are on the horizon. HCV is an RNA virus with a high degree of genetic variability, that has been classified into six major genotypes and further classified into more than 50 subtypes within genotypes. Further minor genotypic variations of viral sequence with genotypes are known as quasispecies. There is substantial heterogeneity between genotypes that may differ from each other by 31–34% of their nucleotide sequences. This chapter addresses advances in our clinical knowledge of this RNA virus. Information regarding virology and immunopathogenesis is discussed in detail in Chapter 8. The clinical importance of HCV relates largely to the complications of long-standing infection that leads to complications of cirrhosis, liver failure, and hepatocellular carcinoma (HCC). Clinical consequences are increasingly being recognized as occurring outside the liver. These so-called extrahepatic manifestations of infection range from fatigue that can be debilitating to vasculitis such as membranoproliferative glomerulonephritis resulting in renal insufficiency. In this chapter, we review the clinical consequences of HCV infection, and discuss new advances in HCV therapies.
There are estimated to be 170–200 million infections globally, with the highest prevalence of infection being in parts of Eastern Europe (estimate of 10 million infections), South-East Asia (estimate of 30–35 million infections), and Africa (estimate of 30–40 million infections).2 The country with the highest prevalence of infection is Egypt, where 10–28% of blood donors are infected.3 In the USA, HCV-related liver disease accounts for 10 000 deaths and 2000 liver transplantations annually. Estimates from studies conducted by the Centers for Disease Control and Prevention (CDC) suggest that HCV infects more than 2.7 million individuals in the USA,4 with at least 4 million (or 1.8% of the population) who have been HCVexposed. Many of those infected remain undiagnosed. Since 1989, the number of new infections has declined by 80% to fewer than 30 000 new infections per year.5 This decline has resulted from successful screening of blood and blood products for HCV as well as successful public health interventions, such as needle-exchange programs, aimed at reducing transmission of human immunodeficiency virus (HIV) and other bloodborne pathogens. However, the number infected is likely to be higher than those measured in CDC surveillance programs, since US estimates are based on surveys that have not included those who are institutionalized (such as those in correctional facilities) and those who are homeless. There are 1.8 million individuals in state and federal prisons and of these, 30–40% have been found to be HCV-infected, making the current CDC estimate of those infected to be an underestimate by at least 500 000. Revised CDC estimates are that there may be as many as 7 million in the USA infected with HCV and/or HCV-exposed.5 There are global variations in the distribution of different HCV genotypes. In the USA, genotype 1 predominates, accounting for
665
Section V. Liver Diseases Due to Infectious Agents
approximately 70% of infections.1 Approximately 55% of genotype 1 infections are subtype 1a, 40% cent are subtype 1b, and a small proportion is mixed 1a/1b infections. Two-thirds of genotype 1 infection have high viral load (defined as greater than 800 000 IU/ml or 2 000 000 copies/ml)4 and approximately 50% of US patients have genotype 1 high viral load infection, the group that is most resistant to antiviral therapy.6 Most other infections in the USA are genotypes 2 or 3, with a small proportion being genotypes 4–6. In contrast, subtype 1b is most common in western Europe, accounting for 80% of all infections. Genotype 4 infection is prevalent in the Middle East and Egypt, genotype 5 is common in southern Africa, and genotype 6 in parts of South-East Asia, including Vietnam. The geographic distribution of different genotypes likely reflects patterns of human migration and immigration into countries like the USA, and has relevance to treatment outcomes in different populations,7 since interferon-sensitivity is highly linked to infecting genotype. Despite the declining incidence of infection, the number of persons infected for >20 years is predicted to increase substantially before peaking in 2015.8 Those born between 1940 and 1965 appear to have the highest lifetime risk of HCV infection.8 These are precisely the individuals who now, at age 40–65 years, are seeking medical care for HCV and as their liver disease progresses, are being referred for evaluation for liver transplantation. Recent results from the CDC suggest that the prevalence of HCV infections has not greatly changed since the National Health and Nutrition Examination Survey (NHANES) III study conducted a decade ago.4,9 A recent survey conducted during 1999–2002 of 15 079 non-institutionalized persons has shown that the current prevalence of anti-HCV in the USA is 1.6%, corresponding to 3.8 million individuals. Prevalence is higher among blacks (3.0%) than whites (1.5%) or Mexican Americans (1.3%) and higher in males (2.1%) than females (1.1%). The prevalence of anti-HCV peaked in 45–49-year-olds in the ongoing NHANES study, compared with a peak in 35–39-year-olds in NHANES III conducted a decade earlier. These data support an aging cohort of HCV-infected individuals in our society. The prevalence in those aged 45–49 years is 7.1% among men and 2.3% among women. Of all anti-HCVpositive persons, 69.9% are between the ages of 35 and 54 years, an age group that may also be at risk for non-hepatic comorbid conditions.10 Eighty-eight percent of those who are anti-HCV-positive are also HCV RNA-positive. Among black men in this age group, 17.9% tested anti-HCV-positive.9 Consistent with results from NHANES III, the ongoing NHANES study found that the prevalence of anti-HCV depends on a number of demographic and risk factors. Highest prevalence is seen in individuals who have ever injected drugs (57.3%) compared with a prevalence of 3.5% in those who have used non-injection drugs and 0.7% in those who denied any drug use.9 Prevalence increased with increasing numbers of lifetime sexual partners and was higher among those with a history of blood transfusion before 1990 (4.2%) than among those without (1.4%). This recent survey suggests that, while the overall prevalence of HCV infection has not changed in the past 10 years, the peak in age-specific prevalence has shifted to older age groups, resulting in a cohort of people, now aged 40–59 years, with a high prevalence of infection. With prolonged infection comes risk of hepatic
666
complications. The incidence of HCC, a highly lethal complication of long-standing infection, has increased twofold in the USA from 1.3 per 100 000 during 1978–1980 to 3.0 per 100 000 during 1996–1998.11,12 This measured rise in incidence of HCC is consistent with the rise in complications of HCV predicted by the CDC.13 There also appears to be a shift towards the occurrence of HCC in younger individuals.11 The incidence of HCC among those over the age of 65 years continues to rise from 14.2 per 100 000 in 1993 to 18.1 per 100 000 in 1999. While much of this rising incidence is due to HCV, HBV-related HCC also appears to be increasing.14 This change in the incidence of the long-term consequences of HCV infection is of concern. Particularly worrisome is the observation that, once HCC is diagnosed, median survival is only 7–8 months and survival has increased minimally in the past 25 years despite increased screening for HCC and HCC interventions such as radiofrequency ablation and chemoembolization.15 These poor outcomes in patients with advanced HCV-related liver disease underscore the importance of evaluating individuals with hepatitis C for therapy, overcoming barriers to treatment if they exist, and improving outcomes of treatment by maximizing adherence.
MODES OF TRANSMISSION Transmission of HCV infection is highly linked to parenteral exposures such as occurs with injection drug use, from contaminated needles in the health care setting, and through exposure to infected blood products. Risk factors for HCV infection include injection drug use or intranasal cocaine drug use, clotting factors transfused before 1987, blood or blood products transfused before 1992, highrisk sexual activity, mother-to-infant (vertical) transmission, occupational exposure, and tattoos or body-piercing with contaminated needles1 (Table 32-1). Potential exposures that have been evaluated in case-control studies and have been found to be weakly associated, if at all, with HCV infection, include medical, surgical, and dental procedures, tattooing, acupuncture, ear-piercing, incarceration, military service, and foreign travel.16 Prevalence of HCV infection in health care workers is low, although HCV has clearly been transmitted in hospital settings, and outbreaks of HCV have been described in dialysis units and surgical centers.16 Despite the low risk (0.5%) after needlestick exposure from an HCV source patient,
Table 32-1. Sources of infection for hepatitis C
Injection drug use Sexual Transfusion Occupational Unknown Othera a
Previously acquired (before 1990s)
Newly acquired (1995–2000)
60% 15% 10% 4% 10% 1%
68% 18% ~0% 4% 9% 1%
Other includes nosocomial, iatrogenic, perinatal. Data from the Centers for Disease Control and Prevention, Atlanta, GA.
Chapter 32 HEPATITIS C
acquisition of HCV infection in health care workers has been described.16 HCV transmission from infected health care worker to patient has also been described, although this is typically secondary to a breakdown in routine infection control practices.16 HCV is not transmitted through casual household contact, although sharing of toothbrushes and razors with infected household members should be avoided. Serological tests were introduced to screen units of donated blood in 1990 and were improved in 1992. More recently, in the USA, nucleic acid testing that identifies low-level viremia has further improved the safety of the blood supply such that now the risk of acquiring HCV by contaminated blood is less than 1:200 000 000 units transfused. Approximately 6 million people donate blood each year, and 0.6% of these are seropositive for HCV, accounting for 36 000 newly identified infections through this mechanism alone.16 Risk factors in seropositive volunteer blood donors include previous blood transfusion, intranasal cocaine and injection drug use, multiple heterosexual partners, and ear-piercing among men.17 Recommendations for testing for anti-HCV in different at-risk groups are summarized in Table 32-2. While public health measures have led to a marked decline in incident infections, they remain large reservoirs of infection, largely in those with prior or ongoing injection drug use. HCV antibodies are identified in 88.7% of injection drug users and the prevalence of anti-HCV antibodies increased directly with the duration of drug use.18 Injection drug use accounts for two-thirds of new infections in the USA. The prevalence of HCV is increased in injection drug users who are black, infected with HIV, who inject frequently, and who use cocaine. More than 80% of injection drug users acquired HCV within 2 years of initiating drug use, a finding that has been used to estimate the time of initial infection and calculate the duration of infection in many studies.18,19
HCV is transmitted sexually, although the efficiency is low.20 High-risk heterosexual behavior (typically defined as more than 20 or in some studies more than 50 lifetime sexual partners) is independently associated with risk for HCV infection. In sexual partners of HCV-infected persons, prevalence of infection is 2–3%, which is similar to that in the general US population. However, incidence is a little higher (0.4–1.8% per year) in those who have multiple heterosexual partners, sex workers, and men who have sex with men than in those in long-term monogamous partnerships (0–0.6% per year).20 Prevalence of antibody-positive, genotypeconcordant couples varies in different studies from around the world, from 2.8 to 11% in South-East Asia, 0 to 6.3% in northern Europe, to 2.7% in the USA.20 HIV co-infection appears to increase the risk of HCV sexual transmission. In addition, studies from sexually transmitted disease clinics and studies of men who have sex with men suggest that sexual practices associated with trauma are more common in HCV-positive than HCV-negative individuals.20 Whether HCV sexual transmission differs depending on the gender of the infected partner is unclear.20 Current recommendations for preventing HCV transmission differ for those in long-term monogamous versus short-term relationships (Table 32-3).20 Mother-to-infant transmission of HCV is well documented, although the risk to the infant of acquiring infection from a seropositive mother is less than 5%. This risk is higher in mothers with high levels of virus (greater than 106 copies/ml) and those with HCV/HIV co-infection (22% versus 4% in mothers with and without HIV respectively).21 The rate of transmission from viremic mothers to their infants is 3–7%.21 There are no specific recommendations to reduce the risk of perinatal infection. Cesarean section and breast-feeding do not appear to alter the risk.21 Part of the difficulty in evaluating rates of transmission and risk factors for transmission across different studies relates to different definitions of infection and different durations of follow-up from study to study. Infants are frequently anti-HCV-positive from passive trans-
Table 32-2. Recommendations for hepatitis C testing History
Risk for infection and prevalence
Testing recommended
Ever injected drugs Received clotting factors prior to 1987 Persistently abnormal LFTs Chronic hemodialysis Received blood or organs prior to 1992 Born to infected mother Health care workers
High, 60–90% High, 70–90%
Yes Yes
Intermediate, 15% Intermediate, 10% Intermediate, 6%
Yes Yes Yes
Intermediate, 6% Low, 1–2%
Sex with infected partner
Low, 2%
Sex with multiple partners Tattooed Body-piercing Snorted cocaine
Low, 2–5% Low, <1 % Low, <1% Low, <1%
Yes After known exposure On an individual basisa No No No No
a
Counseling and testing of monogamous partners may provide reassurance of low risk LFTs, liver function tests. Reproduced from Centers for Disease Control and Prevention. Morb Mortal Wkly Rep 1998; 47:1–39,5 with permission.
Table 32-3. Recommendations for prevention of sexual transmission of hepatitis C virus (HCV) infection Relationship HCV-positive individuals in long-term monogamous relationships
HCV-infected individuals with multiple or short-term sexual partners Other STDs present, sex during menses, sexual practices that might traumatize the genital mucosa Couples regardless of stability of their relationship
Recommendation No change in sexual practices If couples wish to reduce the low risk of sexual transmission, barrier methods should be used. Partners may be considered for anti-HCV testing Barrier methods or abstinence recommended Routine testing of partners not recommended Barrier precautions
Avoid sharing razors, toothbrushes and nail-grooming equipment
STDs, sexually transmitted diseases. Adapted from Terrault NA. Sexual activity as a risk factor for hepatitis C. Hepatology 2002; 36:S99–S105.
667
Section V. Liver Diseases Due to Infectious Agents
fer of maternal antibody, antibody that is gradually lost in the vast majority of cases with follow-up. To diagnose infection in the infant, tests for HCV RNA should be performed at least 12 months after birth.22 Despite this low rate of transmission of true HCV infection, it is estimated that 10 000–60 000 newborn babies will be infected with HCV each year.21
NATURAL HISTORY An understanding of the natural history of infection is important when counseling patients about their prognosis and the need for treatment. While infection is silent in many and rarely results in liver complications in the short term, disease can clearly be progressive in some, leading to life-threatening complications in approximately 20%. There is a spectrum of liver disease associated with HCV infection, from chronic mild hepatitis to decompensated cirrhosis and liver failure.23 Even in patients with clinically compensated cirrhosis, natural history in the short term is typically benign. In the absence of clinical decompensation, actuarial survival is 83–91% at 5 years and 79% after 10 years.24 Survival drops to 50% at 5 years among those who develop clinical decompensation such as variceal bleeding or ascites. The cumulative probability of developing an episode of decompensation is only 4–5% at 1 year, increasing to 30% at 10 years from diagnosis of cirrhosis.24–27 The risk of developing HCC is 1–4% per year once cirrhosis is established. Much of the controversy regarding the natural history stems from the study design and selection of patients included in these studies. Studies that have focused on outcomes of patients seen in tertiary referral clinics have inevitably portrayed HCV infection as a progressive disease resulting in a high likelihood of developing complications.28 In contrast, population-based cohort studies that have included “unselected” large denominators have tended to portray the natural history of infection as a more benign process.23 Approaches that have been used to evaluate the natural history of chronic HCV infection include: (1) cross-sectional studies in which stage of liver disease is described in patients presenting for care; (2) retrospective/prospective studies in which the stage of established liver disease is related to the estimated duration of infection, calculated from the presumed time of first exposure; (3) prospective studies in which patients are evaluated at the onset of disease and are followed longitudinally for a defined period; and (4) long-term cohort studies of subjects with a defined parenteral exposure, in which the HCV status of the individuals at the time of initial exposure was known and outcome was assessed many years later. Each approach has its limitations. As would be predicted, the most severe outcomes have been described in cross-sectional and retrospective studies. In early cross-sectional studies, anti-HCV was detectable in 50% (8–69%) of patients with cryptogenic cirrhosis and in 6–76% of those with HCC, underscoring the importance of HCV in these diseases. Subsequently, retrospective/prospective studies showed that chronic hepatitis, cirrhosis, and HCC develop 13 ± 11 years, 21 ± 10 years, and 29 ± 13 years after transfusion, respectively.23 Prospective studies that are typically of less than 30 years of follow-up tend to portray HCV infection as benign in many,
668
although have provided evidence of the progressive nature of chronic HCV infection in some individuals. While death from end-stage liver disease is uncommon, liver failure and HCC occur. The classical cohort studies of HCV natural history were performed by Seeff,23 in which 568 patients with non-A, non-B post-transfusional hepatitis were compared with 984 matched controls who received transfusions but who did not develop hepatitis. After 20 years, allcause mortality was no different between the cases and controls (51 and 54%, respectively), but the death rate attributable to liver disease was higher in the former than in the latter (3.3 and 1.5%, respectively). Another important cohort portrayed HCV infection as benign in the majority. In 1977, young Irish women were accidentally infected with HCV genotype 1b from a contaminated lot of anti-D serum. With follow-up of more than 18 years, only 2% had developed cirrhosis and none had developed complications of liver disease.29 The vast majority of women had minimal fibrosis, if any. A similar study was recently reported in Germany involving 152 women also infected through contaminated Rh immune globulin.30 After 15 years of exposure, none of these women had evidence of chronic active hepatitis or cirrhosis. One reason for these apparent discrepancies in natural history may be due to the population under study. Male gender is a strong predictor of disease progression, as is older age at acquisition (see below). The post-transfusion cohort focused on men age 50 at the time of acquisition, whereas the Irish and German cohorts focused on young women. Children in particular appear to have a benign course, again underscoring the importance of age at infection as a predictor of disease progression. In a recent study from Germany, 458 children infected from blood transfusion were followed for a mean of 17 years.31 Almost half of the infected children spontaneously cleared the infection. Of the 17 patients who underwent a liver biopsy, only 2 had signs of portal fibrosis. The study with probably the longest duration of follow-up included 8568 military recruits, of whom 0.2% tested positive for anti-HCV. After 45 years of follow-up, mortality due to liver disease was very low.32 In aggregate, data from these natural history studies suggest that only 15–20% of HCV-infected persons will eventually progress to potentially serious end-stage liver disease, while the remainder will die of causes other than liver disease. Seroreversion (loss of anti-HCV) also seems to occur, although the rate that this occurs is poorly defined. Another concept in considering the natural history of HCV infection is one of competing risks for morbidity and for death. This is particularly true for HCV-infected patients currently considering therapy, who are entering their sixth decade of life. The cumulative risk of death from all causes in those aged 55–64 years is 901 per 100 000 (or 0.9%). In Americans over the age of 65 years, the cumulative risk of death from all causes is 3936 per 100 000 population or 4%. The leading cause of death among adults over 65 years of age is heart disease, followed by malignant neoplasms, cerebrovascular diseases, and chronic obstructive pulmonary disease. These causes are also associated with age, race, and gender. For example, diabetes and kidney diseases are more than twice as likely in older blacks compared with white adults.33 Thus for black patients in whom response to antiviral therapy is low (see below), and the risk of developing potentially life-threatening non-hepatic disease is high, treatment is likely going to be less beneficial than for other races.
Chapter 32 HEPATITIS C
There are additional comorbidities and risks of death associated with substance abuse that may be more common in HCV-infected adults than in the general population. A 33-year follow-up study of narcotic addicts admitted to a compulsory drug treatment program for heroin-dependent criminal offenders found that 49% of study subjects were confirmed to have died.34 The most common cause of death was accidental poisoning or drug overdose (21.6%), followed by death from chronic liver disease (15%), cancer, and cardiovascular diseases. Predictors of death in this study included older age, disability, years since first heroin use, and heavy alcohol use. Homicide, suicide and accidents are also common causes of death in injection drug users, and injection drug users are also at risk for acquiring HIV infection that may further shorten their life expectancy. These health statistics should be taken into account when considering treatment of active injection drug users, although they are likely less relevant to HCV-infected individuals with a remote past history of drug use. Data on natural history, stage of disease, and comborbid medical conditions should all be discussed with the patient when decisions are made regarding initiation of HCV antiviral therapy. While the natural history of HCV infection is highly variable, factors influencing this variability are still inadequately understood. A number of variables have been studied and several, in particular male gender, obesity, and excess alcohol, have been shown to play a role in histological progression of liver disease and development of HCC. The identification of variables influencing disease progression is important in order to focus therapy on patients at greatest risk for progressive liver disease as well as to consider strategies aimed at reducing modifiable risk factors. While viral-related factors have little influence on the course of HCV disease, host and environment factors likely play a major role.35 Age at infection appears to be one of the strongest predictors of outcome. Patients more than 50 years of age at the time of infection have a high likelihood of progressing to cirrhosis within 15 years of infection, whereas the risk of progression to cirrhosis in patients infected in their 40s is lower.35 Studies of the natural history of HCV fibrosis indicate that fibrosis progression is not linear over time, being slower in younger ages of life, increasing with each 10–15-year period, but mostly accelerating after the age of 45–50 years.35 Gender is a strong predictor of progression, with men being at greater risk than women.36 Race has also been implicated in disease progression in a few studies, with African-American patients having a lower risk of progressing to cirrhosis than Caucasians (2.2–22% compared to 7.2–30%).37 Excess alcohol has clearly been shown to increase disease severity (see below) and immunosuppression leads to progressive disease. Studies from patients with humoral (hypogammaglobulinemic patients)38 or cellular immune impairment (liver or kidney transplant recipients39 and HIV-infected patients with low CD4 count40) have shown rates of progression to cirrhosis significantly higher than those observed in immunocompetent patients. Other prognostic factors proposed but less consistently documented include mode of transmission (higher disease progression for those infected through blood transfusion rather than injection drug use) and co-infection with HBV. Metabolic abnormalities and comorbidities, including type 2 diabetes, obesity, increased hepatic iron stores, and liver steatosis, have been associated with accelerated fibrosis progression in several but not all studies.41–43 Importantly, these adverse metabolic conditions
tend to increase with age. Progression also appears to be related to the degree of necroinflammation, with little progression over a period of 4–5 years in patients with none or mild degrees of inflammation on initial liver biopsy, and accelerated progression in those with moderate or marked inflammation. In one of the few published studies that have measured fibrosis progression between two liver biopsies, the extent of elevation of serum alanine aminotransferase (ALT) and necroinflammation on initial liver biopsy appeared to be the strongest predictor of disease progression.44 These data support the recommendation that patients with mild disease activity and scant hepatic fibrosis can delay therapy if they wish to do so until treatments improve and regimens are available that are better tolerated and/or are more effective. In many studies alcohol has been linked to risk of progressive HCV-related liver disease.36,45 Seventy percent of members of western societies drink some alcohol regularly, and the quantity of alcohol that causes harm likely differs from person to person. Patients with chronic HCV are counseled by their physicians to abstain from alcohol. This recommendation is based upon multiple epidemiological studies showing that heavy alcohol intake correlates with worse HCV-associated liver disease. This recommendation is also logical, since alcohol is likely second only to HCV as a cause of liver disease in the western world. However, data supporting an adverse effect of light or moderate alcohol use on HCV-related liver disease are lacking. Moreover, within individuals who drink excessively, there is substantial variation in the severity of liver disease for the same intake of alcohol.45 Alcohol has also been linked to the development of HCC, likely due to the role of alcohol in the development of cirrhosis. Alcohol use results in reduced response to interferon-based treatment of HCV, either through a direct effect of alcohol on viral replication and/or through a suppressive effect of alcohol on the HCV-specific immune response necessary for viral eradication.1 Alcohol might also have an adverse effect on treatment outcomes through diminished adherence to peginterferon plus ribavirin therapy.
CLINICAL MANIFESTATIONS CLINICAL PRESENTATION The majority of patients with chronic hepatitis will have elevated or fluctuating ALT levels, although in one-third, serum ALT values will be persistently normal, despite continued liver injury and detectable virus. The overwhelming complaint of patients with chronic infection is of fatigue. Other frequent manifestations include arthralgias, parasthesias, myalgias, pruritus, and sicca syndrome. Non-specific symptoms include depression, nausea, anorexia, abdominal discomfort, and difficulty with concentration. The severity of these symptoms is not necessarily related to the severity of the underlying liver disease. Once patients develop cirrhosis, they are at risk for complications of portal hypertension (such as ascites, gastrointestinal bleeding, and encephalopathy). Jaundice is rarely seen in chronic infection until significant hepatic decompensation has occurred. Referral for liver transplantation should be considered in patients with complications of portal hypertension or evidence of hepatic
669
Section V. Liver Diseases Due to Infectious Agents
synthetic failure (reduction in serum albumin, rise in serum bilirubin, and/or prolongation of the prothrombin time) (see Chapter 49).
ACUTE HCV INFECTION HCV accounts for approximately 20% of cases of acute hepatitis. Acute infection is however rarely seen in clinical practice since the vast majority of patients are asymptomatic. Jaundice occurs in only 10–20% of patients with acute HCV, with a somewhat greater proportion (20–30%) presenting with non-specific symptoms such as fatigue, nausea, and vomiting. While HCV RNA is detectable within 1–2 weeks of exposure, anti-HCV seroconversion is somewhat delayed (15–90 days). Serum transaminase levels peak at about the first month after exposure, occasionally exceeding 1000 IU/ml, and liver enzymes may fluctuate for the first few months. In those who develop jaundice, peak bilirubin levels are usually less than 12 mg/dl, and jaundice typically resolves within a month. Fulminant liver failure is an extremely rare event following acute infection. A more severe course of acute hepatitis C may develop in patients who drink large amounts of alcohol, or in whom there is co-infection with HBV or HIV. HCV infection is self-limiting in 10–50% of patients, with loss of HCV RNA and normalization of serum ALT. In a person suspected of acute HCV infection who initially tests negative for HCV RNA in serum, repeat testing should be performed for up to 12 months, since late seroconversions can occur. In particular delayed HCV seroconversion has been described in patients with simultaneous acquisition of HIV infection. In patients with communityacquired hepatitis C who resolve their infection spontaneously, loss of virus usually occurs within 3–4 months from clinical onset. The rate of viral persistence varies following acute HCV infection from as low as 40–50% to as high as 90–100% depending on a number of factors; these include patient age and sex (younger and female patients having a lower rate of persistence), source of infection and size of inoculum (injection drug users having a lower rate of persistence than those acquiring HCV infection following blood transfusion), immune status of the host (higher rates of persistence in immunodeficient than in immunocompetent patients), as well as race (higher rates of viral persistence in African-American than in Caucasians and hispanic individuals in the USA). Rates of spontaneous clearance are higher in symptomatic patients who develop jaundice with acute infection than in those who remain asymptomatic.
EXTRAHEPATIC MANIFESTATIONS HCV is increasingly being recognized as a cause of significant extrahepatic disease, including membranoproliferative glomerulonephritis (MPGN), essential mixed cryoglobulinemia (MC), porphyria cutanea tarda, leukocytoclastic vasculitis, focal lymphocytic sialadenitis, Mooren corneal ulcers, lichen planus, rheumatoid arthritis, non-Hodgkin’s lymphoma (NHL), and diabetes mellitus. The strength of these associations with HCV is variable (Table 324). Rheumatologic and skin manifestations are common in patients with chronic HCV. Many HCV-infected patients will have positive serologic tests for non-organ-specific autoantibodies (antinuclear antibodies with a titer >1:40 are observed in 21%, anti-smoothmuscle antibody with a titer >1:40 in 21%, and anti-liver–kidney microsomal antigen in 5% of patients). Cryogolobulins may be positive in as many as 40–50% of patients in some series, and low
670
Table 32-4. Extrahepatic manifestations in hepatitis C Strength of the association Mixed cryoglobulinemia Membranoproliferative glomerulonephritis Arthralgias Autoimmune thyroid disease Porphyria cutanea tarda Lichen planus Diabetes mellitus Non-Hodgkin lymphoma Fatigue
Strong Modest Modest Modest Controversial Modest Strong Controversial Controversial
thyroxine concentrations in 10%. However, disease associated with these abnormalities is much less frequent.
Mixed Cryoglobulinemia The clearest association is for MC. In many parts of the world, in particular in Mediterranean countries, MC type II is caused by HCV infection.46,47 Anti-HCV antibodies are positive in 50–90% of patients with essential MC. However, clinical symptoms develop in only 25–30% of these patients. Symptoms range from fatigue, arthralgias/arthritis, to purpura, Raynaud’s phenomenon, vasculitis, peripheral neuropathies, and glomerulonephritis. MC must be distinguished from the detection of incidental cryoglobulins in hepatitis C patients without symptoms of vasculitis. With sensitive tests, cryoglobulins can be detected in almost all HCV patients, in particular if cirrhosis is present. The immune complexes consist of polyclonal immunoglobulin G (IgG) anti-HCV, monoclonal IgM of unknown specificity, monoclonal IgM rheumatoid factor, and HCVRNA. Antiviral treatment leads to a decline in the cryocrit, a simple test that measures cryogobulins in serum. For practical purposes it is crucial that blood is centrifuged at 37°C and immediately brought to the laboratory, otherwise false-negative and irreproducible results occur. Cryoglobulins decline or disappear if treatment is successful, as do symptoms. Creatinine clearance in those with mild renal insufficiency may show some temporary improvement. However, relapse rates are high when therapy is stopped and most published data are from the era of interferon monotherapy. No controlled studies are available with pegylated interferons (PEG-IFNs) in combination with ribavirin.
Membranoproliferative Glomerulonephritis Several years ago a number of cases with MPGN were described with detection of HCV proteins in the glomerular deposits.48 Glomerular disease generally presents with nephrotic syndrome, non-nephrotic-range proteinuria, and/or renal insufficiency. Hypocomplementemia is frequent and most patients have a positive rheumatoid factor. Interferon therapy in patients with MPGN has been associated with a reduction in proteinuria (65%) but typically no significant changes in renal function. About 10% of patients will progress to end-stage renal failure requiring dialysis. Use of pulse methylprednisolone therapy may provide temporary benefit to some patients with severe progressive disease. While HCV is clearly associated with MPGN, the majority of MPGN is not caused by HCV.
Chapter 32 HEPATITIS C
Thyroid Disease There is clearly an association between HCV, interferon, and thyroid disease. Thyroid autoantibodies are more frequent in HCV than agematched controls with other forms of liver diseases. Autoimmune thyroid disease, in particular Hashimoto’s thyroiditis, occurs before, during, and even after antiviral treatment with interferon-based therapies.49 Thyroid disease can present with hypo- or hyperthyroidism. Thyroid-stimulating hormone (TSH) levels should be tested before treatment is initiated to identify patients at risk, and thyroid abnormalities, if present, should be corrected. Every HCV patient should be informed about the symptoms of hyper- and hypothyroidism and patients should be monitored closely for thyroid disease while on therapy.
Arthralgias Arthralgias are common in acute or chronic viral infections. Immune complexes and their deposition in joint mucosa are regarded as mediators of immunopathogenesis. Complicating patient management further is the well-described joint pain associated with interferons themselves. Arthralgias in hepatitis C may be caused by HCV-associated immune complexes, HCV-induced autoimmune reactions, as well as interferon treatment itself. The correct classification of HCV-associated joint problems is essential for appropriate patient management.
Porphyria Cutanea Tarda Several years ago, Italian investigators described the association of HCV with porphyria cutanea tarda.50 These reports from Mediterranean countries could not be confirmed in northern European countries such as Germany.51 In countries where this association has been described it appears that only a minority of porphyria cutanea tarda is attributable to HCV infection, and in the majority of northern
European cases the etiology of porphyria cutanea tarda remains unknown.
Lichen Planus Lichen planus of the buccal mucosa (Figure 32-1) is frequently seen in association with HCV disease and the etiological association is generally accepted. Lichen planus may worsen on interferon-based therapies, or it may first appear while on treatment. However, lichen planus is not a contraindication to antiviral treatment. In individual cases where disease worsens on therapy, treatment should be stopped.
Diabetes Mellitus The relationship between diabetes and hepatitis C is complex. Type 1 diabetes mellitus certainly can manifest under treatment with cytokines, in particular with interferon-alpha. One hypothesis favors the idea that, in genetically susceptible individuals, cytokines prematurely boost pathogenetic mechanisms that would otherwise have taken years to manifest themselves. HCV antigens have been identified in pancreatic islet tissues. Diabetes is associated epidemiologically with HCV since diabetes is more prevalent in HCV than in HBV patients52 and HCV is more prevalent in diabetic than in nondiabetic persons. The increased prevalence of HCV in diabetics could possibly be due to more medical procedures and injections in diabetics, with consequences of nosocomial acquisition of HCV infection. In addition, HCV infection itself, particularly in those who are obese and/or who have hepatic steatosis, is associated with insulin resistance.
Non-Hodgkin’s Lymphoma HCV infection has also been linked to B-cell NHL, a link that is suggested by a significantly higher prevalence of HCV infection in patients with B-cell NHL (approximately 15%) than that reported
Figure 32-1. Lichen planus of the buccal mucosa in a patient with hepatitis C.
671
Section V. Liver Diseases Due to Infectious Agents
in the general population or in patients with other hematologic malignancies. The mechanisms by which HCV might exert an oncogenic potential are still under investigation. HCV RNA does not integrate into the host genome, and therefore, HCV cannot be considered a typical oncogenic virus. Rather it has been suggested that HCV, a virus that shows lymphotropism, might facilitate the development and selection of abnormal clones through a chronic stimulus to the immune system. The striking geographic differences in the association between HCV and NHL suggest that, while HCV may be the first step in causing a clonal expansion of infected B cells, there remain genetic, environmental, and other viral factors that are probably involved in the pathogenesis of these disorders.
Fatigue Syndrome In recent years chronic fatigue has been described as an HCVassociated syndrome that is independent of the severity of the underlying liver disease. The pathophysiology of fatigue is unknown. Some experts see severe fatigue in itself as an indication for antiviral treatment independent of the severity of the underlying liver disease. In severe cases of fatigue, treatment may be justified and the persistence or disappearance of symptoms may be regarded as indirect proof of the relationship. Data implicating the brain as a site of HCV replication have supported a possible etiological relationship between HCV infection and fatigue.
The Relationship Between Autoimmune Hepatitis and HCV Infection A decade ago, at a time when HCV serological assays lacked specificity, a significant number of patients with autoimmune hepatitis tested anti-HCV-positive. Subsequent studies have demonstrated that these autoantibodies were due to false-positive antibody tests in the presence of hypergammaglobulinemia. This confusion arose in part because these early generations of HCV antibody tests were developed to detect HCV antibodies in healthy blood donors with normal gammaglobulins. Rare cases have been described, however, in individuals who likely have both HCV infection and autoimmune hepatitis. In such cases interferon-based therapies can lead to deterioration of liver disease. On the other hand serological markers of autoimmunity such as antinuclear, smooth-muscle, and liver–kidney microsomal antibodies have been described in a significant number of HCV patients from Mediterranean countries. If HCV RNA is detectable the primary therapeutic goal should be to eradicate HCV infection. While under treatment, patients should be monitored closely for any worsening of liver disease and interferon therapy should be stopped if autoantibody titers rise, and/or if gammaglobuins increase in the absence of cirrhosis. In such cases, immunosuppressive therapy may be necessary as for the management of autoimmune hepatitis in the absence of HCV infection.53
IMMUNOPATHOGENESIS (see Chapter 8) Since HCV is a non-cytopathic virus in most circumstances, it is the immune response rather than the virus itself that is central to the pathogenesis of liver disease. The immune response is also critical to clearance of virus following acute infection. For example, symptomatic patients with acute HCV infection are more likely to
672
recover than asymptomatic patients.54,55 Since symptoms are likely caused by the host’s immune response, a strong cellular immune response appears to be key to viral clearance. Anti-HCV antibodies usually develop between months 2 and 3 of acute HCV infection, a time course that is late compared to other viral infections. The immune response against HCV is complex and generated by various cell types and tissues. Early innate immune responses may play an important role in determining the outcome of infection. The analysis of gene expression profiles in liver biopsies from chimpanzees during early HCV infection shows a very early increase of interferon-response genes, preceding expression of T-lymphocyte surface markers by several weeks. However, HCV has developed several mechanisms to inhibit innate responses, such as direct inhibition of natural killer (NK) cells by HCV envelope proteins via binding to CD8156 or indirect impairment of NK-cell cytotoxicity by upregulation of major histocompatibility complex class I molecules on infected cells.57 Immune mechanisms play a role in the pathogenesis and progression of liver injury, since patients with more severe hepatitis have a higher chance of developing liver cirrhosis and HCC than those with less inflammation. The histological activity of the liver disease is determined by qualitative and quantitative assessment of the cellular infiltrate in the liver. This infiltrate consists mainly of T cells, NK cells, and NKT cells, thus representing an immune response with resulting “hepatitis.” In sharp contrast to hepatitis B, the humoral immune response against HCV does not allow discrimination between different stages of infection (as for example with hepatitis B in which anti-HBcore IgM is indicative of acute HBV infection and anti-HBcore IgG is indicative of chronic or resolved HBV infection). Antibodies against epitopes from all HCV proteins are detectable in acute as well as in chronic infection, and are also present after recovery from HCV. No specific antibody pattern is associated with recovery or with a specific level of replication. An early antibody response against the hypervariable region of the E2 protein (HVR-1) has been associated with a self-limited course of infection. Since there is high variability of the virus in this region, it seems possible that escape from efficient humoral immunity might occur with prolonged periods of viremia. Subsequently, a more heterogeneous humoral immunity against HVR-1 has been associated with viral persistence.58 There seems to be no long-lasting protective humoral immunity against HCV. Anti-HCV antibody titers do decline after recovery from acute HCV infection and may become undetectable after two decades.59 Thus, the prevalence of individuals who have had contact with HCV might be underestimated in the general population by anti-HCV testing alone, since anti-HCV may be negative in those with previous but resolved infection. Adaptive cellular immune responses are induced by dendritic cells (DC) that present antigens to CD4+ and CD8+ T cells. There is some evidence that DC function is altered by HCV,60 although conflicting data have been published in recent years.61 Nevertheless, there is a clear association between a multispecific, strong, and maintained HCV-specific CD4+ and CD8+ T-cell response and viral clearance during acute HCV infection.62 The CD4+ response is maintained for several years after recovery. The CD8+ response also remains detectable, but there are conflicting data as to the extent that the CD8+ response decreases over time following recovery. Not only the frequency but, more importantly, the function of T cells
Chapter 32 HEPATITIS C
determines the outcome of infection. Thus, resolution of HCV has been associated with an early interferon-gamma response by CD8+ T cells63 while functionally impaired CD8+ T cells lead to viral persistence.64 Hyporesponsiveness of T cells may be caused by immunosuppressive functions of HCV proteins. The balance between type 1 (such as interferon-gamma) and type 2 (such as interleukin-4 (IL4) and IL-5) cytokines secreted by CD4+ and CD8+ T cells seems to be altered in chronic hepatitis C, an observation that may have relevance to HCV antiviral therapy. One proposed mode of action of ribavirin has been to shift the cellular immune response to a type 1-dominated immune response.65 Activation of an immune response may also be a novel approach for HCV therapy. Early clinical trials of peptide or protein vaccination have already been performed, although it will likely be some time before therapeutic vaccination becomes part of standard therapeutic regimens for chronic hepatitis C.
DIAGNOSIS AND TESTING DIAGNOSIS Anti-HCV testing is accurate for making the diagnosis of infection in high-risk populations such as injection drug users, but may be negative in immune-compromised populations with HCV infection such as those with HIV, those on hemodialysis, or those following solid organ transplantation, and may be falsely positive in low-risk populations such as blood donors. The presence of anti-HCV indicates exposure to the virus, but does not differentiate between acute, persistent, or resolved infection. Antibodies against HCV persist in patients with spontaneously resolved infection, although titers decrease and may even disappear over time. Virological assays detect HCV RNA sequences, indicative of ongoing infection, and HCV RNA levels may fluctuate half a log even in the absence of therapy. Serologic assays are typically used for screening and initial diagnosis, whereas HCV RNA assays are used for confirming infection and/or for monitoring treatment response.22,66,67
SEROLOGICAL ASSAYS The enzyme immunoassay (EIA) assays detect antibodies against different HCV antigens from the core and non-structural proteins. Serologic assays were first introduced in blood banks to screen donors in 1990, and were improved in 1992. Three generations of EIAs have been developed with increasing sensitivity and progressive decrease in the window period for seroconversion after acute exposure. Since the introduction of serologic assays for screening of donors, the risk of acquiring HCV infection from blood products has declined. The latest third-generation EIAs detect mixed antibodies against HCV core, NS3, NS4, and NS5 antigens, as soon as 7–8 weeks postinfection, with 99% specificity and sensitivity. Recombinant immunoblot assays (RIBA), while frequently used in the past for confirmation of true HCV exposure, have largely been replaced by sensitive virological assays, in which the absence of viral RNA is suggestive of resolved infection.
HCV RNA ASSAYS HCV RNA can be measured by highly sensitive qualitative and quantitative assays.22 Qualitative assays provide information about
the presence or absence of virus and are generally more sensitive than quantitative assays. Qualitative HCV RNA detection may be accomplished by target amplification methods such as polymerase chain reaction (PCR) amplification or transcription-mediated amplification (TMA). Qualitative PCR detects as few as 50 IU/ml, while TMA has a sensitivity of 10 IU/ml. Specificity is 99% with both tests. Qualitative testing is largely used for confirmation of clearance of virus after apparently successful antiviral therapy or for the detection of virus in HCV-seropositive patients with chronic liver disease who lack detectable HCV RNA by quantitative assays. Other clinical situations where either qualitative or quantitative assays may be used include seronegative acute or chronic hepatitis in immunosuppressed patients, and the diagnosis of HCV infection in babies born to HCV-infected mothers. Most anti-HCV-positive patients with infection will have virus detectable by both qualitative and quantitative assays, since HCV RNA levels typically range between 5 ¥ 104 and 5 ¥ 106 IU/ml. US Food and Drug Administration (FDA)-approved tests for qualitative HCV RNA detection include the Amplicor HCV test v2.0 and the Cobas Amplicor HCV test v2.0, both with sensitivities of 50 IU/ml. Qualitative HCV RNA assays (nucleic acid testing) are increasingly being used to test for low-level HCV RNA in blood donors with “serosilent” infection or in acutely infected donors in the “window” period before seroconversion. One in 230 000 donations can be identified to be HCV RNA-positive using nucleic acid testing.68 These donors may transmit infection that may remain “serosilent” in the recipient. For this reason, many blood banks now routinely screen blood with nucleic acid tests, reducing the risk of transfusion-associated HCV infection to as low as 1:2 000 000 units transfused.68 Quantitative assays are useful in monitoring antiviral therapy, particularly 4 and 12 weeks after starting treatment. Patients who lack detectable HCV RNA (by either qualitative or sensitive quantitative assays) at 4 weeks into antiviral therapy are defined as having a rapid virological response (RVR); those who either lack HCV RNA or who have a two-log reduction from baseline values are defined as having an early virological response (EVR). Both these measures are increasingly being used to predict the likelihood of achieving sustained virological response (SVR) with therapy and/or to guide the duration of treatment. Methods to quantify HCV RNA levels in serum include signal and target amplification. The bDNA assay, commercially available through Bayer Diagnostics, is an example of signal amplification, that uses capture and target probes from the conserved 5¢ UTR and core regions of the virus to detect viral RNA. The amount of bound probe is amplified through a series of synthetic branched DNA oligonucleotides. In target amplification techniques, HCV target RNA is reverse-transcribed and amplified using primers to the conserved 5¢ region of the HCV genome and the amount of viral RNA present in the amplified sample is estimated from a standardized dilutional series. HCV RNA levels are typically expressed as international units per milliliter and conversion factors have been derived to calculate IU values from copies for commonly used commercial assays (1 IU/ml corresponds to 0.9 copies/ml in the Amplicor HCV Monitor v2.0, 2.7 copies/ml in the Cobas Amplicor HCV Monitor v2.0, 3.4 copies/ml in the SuperQuant, 3.8 copies/ml in the LCx HCV RNA quantitative assay, and 5.2 copies/ml in the Versant HCV
673
Section V. Liver Diseases Due to Infectious Agents
RNA 3.0 quantitative assay, respectively). Different commercial assays vary in their dynamic range. The lower limit of detection with current assays is approximately 600 IU/ml, while the upper end ranges from >500 000 IU/ml to 1 470 000 IU/ml. The Cobas Amplicor HCV Monitor v2.0 is an automated version of the Amplicor test and has a dynamic range of 600–500 000 IU/ml. Samples above the upper limit should be retested after dilution, particularly in those with high levels of virus prior to therapy, in whom EVR and RVR are going to be measured. There are two commercially available assays for determining HCV genotype, assays based on PCR amplification of the 5¢ non-coding region. With these assays, the six genotypes can be readily identified, although tests are less accurate in measuring HCV subtypes, with errors occurring in 10–25% of cases because of variations in the target 5¢ NC region.
SELECTION FOR ANTIVIRAL THERAPY AND PRETREATMENT ASSESSMENTS All patients with confirmed, chronic HCV infection should be evaluated for antiviral therapy. Because of limitations in efficacy and the potential for toxicity, each patient needs careful assessment of the relative risks and benefits of beginning therapy immediately, delaying therapy until a later time, or deferring treatment indefinitely. Pretreatment evaluations include the following: (1) clinical assessments; (2) laboratory tests; and (3) liver biopsy.
CLINICAL ASSESSMENTS A complete medical evaluation is essential in order to rule out medical comorbid conditions that might contraindicate or might worsen with treatment. Particular attention should be paid to coronary artery disease, since ribavirin-associated anemia may lead to worsening of cardiac ischemia on therapy. Other potentially lifedetermining conditions such as non-skin cancer and chronic obstructive pulmonary disease should be identified since life expectancy may be limited by these comorbid conditions rather than HCVrelated liver disease. In such cases, patients may elect not to be treated. All patients should be evaluated for psychiatric disorders, particularly depression and suicide risk. Uncontrolled depression is an absolute contraindication to interferon-based therapies. Patients with psychiatric disorders in remission or stabilized with antidepressants may receive antiviral therapy, with the involvement of mental health professionals throughout the course of therapy. All patients should be evaluated for current substance use, including alcohol and drug use. Current heavy alcohol use, binge alcohol use (more than four drinks per occasion at least once a month), or active current injection drug use requires referral to an addiction specialist prior to treatment initiation. Since illicit non-injection drug use may affect treatment adherence, each case should be evaluated individually. Establishing abstinence from drugs and alcohol prior to initiating treatment is recommended. Patients should be evaluated for autoimmune disorders since these may worsen on interferon therapy. Controlled autoimmune thyroid disease (on replacement therapy if necessary), and con-
674
trolled diabetes (with normal or near-normal serum glycosylated HbA1C) are not contraindications to HCV antiviral therapy but these patients should be monitored closely for signs of worsening disease. Other medical disorders such as psoriasis, rheumatoid arthritis, and Crohn’s disease might worsen on interferon therapy, so that these diseases must be medically controlled prior to treatment and they should be managed by a specialist familiar with these disorders. Patient adherence to treatment should also be assessed. Evidence of prior non-adherence to medical, psychiatric, or addiction therapies provides indirect evidence of likely non-adherence to HCV therapies and should be considered before a final decision to treat HCV disease is made. A baseline ophthalmic exam should be performed in patients with risk factors for retinal disease (such as hypertension or diabetes) to identify any disease that might be worsened by interferon and to provide a reference point should any symptomatic ophthalmic symptoms occur on therapy.
LABORATORY TESTS Patients should have adequate platelet counts (>75 ¥ 109/l), neutrophils (an absolute neutrophil count of >1.5 ¥ 109/l), and hemoglobin (>13 g/dl for men and >12 g/dl for women) in order to tolerate therapy. Patients with neutrophil and platelet counts below these recommended levels can begin therapy but may require dose reductions. African-Americans with constitutional neutropenia <1.5 ¥ 109/l should not be excluded from treatment since the risk of serious bacterial infection with neutropenia is low. Renal function should be normal or near normal in order to tolerate ribavirin therapy (creatinine <1.5 mg/dl or creatinine clearance of >50 ml/min in those with evidence of mild renal insufficiency), and patients should have evidence of preserved hepatic synthetic function (normal serum bilirubin, albumin, and prothrombin time). Patients with mild to moderate impairment of hepatic function awaiting liver transplantation may be treated cautiously with lower doses of peginterferon plus ribavirin in an attempt to eradicate infection and slow disease progression.69 Pretreatment virological assays include baseline viral load measured using a quantitative HCV RNA assay. Information regarding the viral load may aid in counseling patients as to their likelihood of response. Those with low pretreatment viral loads (<800 000 IU/ml) are more likely to respond to treatment than are those with high baseline viral loads (>800 000 IU/ml). For consistency, the same quantitative assay should be used to evaluate changes in viral load with therapy. HCV genotype should be determined, as genotype is an important predictor of treatment response and will influence treatment duration. A pregnancy test should be obtained from women of childbearing age prior to the initiation of HCV treatment, and if the patient is pregnant or attempting to conceive, HCV treatment is contraindicated and should not be started.
LIVER BIOPSY In general, patients treated for HCV should have evidence of associated liver injury such as persistently abnormal ALT levels or histologic evidence of liver damage. Liver biopsy is the best method for determining the severity of liver injury (degree of fibrosis and stage of disease). Liver biopsy may also be helpful in excluding other causes of liver disease. Since response to therapy is high in patients
Chapter 32 HEPATITIS C
with genotype 2 or 3 HCV infection, antiviral therapy may be initiated in all patients regardless of the severity of liver disease. In such cases, the findings on liver biopsy would not influence the treatment decision and thus biopsy may not need to be performed. Histological scoring systems differ in the numerical range of scores for defining degrees of inflammation and liver fibrosis (Chapter 13). Commonly used scoring systems for hepatitis C in the USA are Ishak (that stages fibrosis from 0 to 6) and BattsLudwig (that stages fibrosis from 0 to 4), whereas the most commonly used system in Europe is the Metavir system (that stages fibrosis from 0 to 4).70–72 Increasingly, liver biopsies are being scored for the amount of hepatic steatosis, as steatosis is linked to risk of fibrosis as well as reduced response to therapy.43,73 Because liver biopsy is an invasive procedure that carries a risk of bleeding (estimated to be 1:1000) and a risk of death (1:10 000), there is increased interest in developing non-invasive serum markers of fibrosis (Chapter 6).74 However, since serum markers are not yet able to distinguish accurately between stages I and II and because distinction between these stages influences decisions regarding treatment, liver biopsy is still highly recommended in patients prior to initiation of therapy.1
CURRENT THERAPY FOR HEPATITIS C TREATMENT-NAIVE PATIENTS Hepatitis C was already treated successfully with interferon-alpha at a time when the disease was still called non-A, non-B hepatitis and when the virus had not yet been discovered.75 Subsequently, interferon-alpha monotherapy led to normalization of transaminases in up to 25% of patients.76 However when more sensitive tests for HCV RNA in serum were applied to evaluate response, it became apparent that clearance of virus was only achieved in about 10% of patients treated for 6 months with three times weekly recombinant human interferon-alpha (Figure 32-2). At that time, a negative test for HCV RNA in serum 6 months after the end of treatment was established as the primary endpoint of treatment, also known as a sustained virological response, or SVR. We now know that achieve-
Sustained virological response (%)
80 PEG-IFN and ribavirin 60
IFN and Ribavirin 48 weeks
40
PEG-IFN 48 weeks
20
IFN 48 weeks IFN 24 weeks
12 kDa PEG-IFN alfa-2b (2001) 40 kDa PEG-IFN alfa-2a (2002)
0 1988 1990 1992 1994 1996 1998 2000 2002 Figure 32-2. Advances in the treatment of hepatitis C. IFN, recombinant interferon-alpha; PEG-IFN, pegylated recombinant interferon-alpha.
ment of SVR is almost synonymous with cure, i.e., long-term eradication of virus. Relapse rates beyond this time point are rare. In 1998 the addition of ribavirin to interferon-alpha significantly improved SVR rates to over 40% (Figure 32-2).77,78 The introduction of two PEG-IFNs, PEG-IFN alpha-2b in 2001 and PEG-IFN alpha-2a in 2002, in combination with ribavirin led to further improvements in response with SVRs for all genotypes of up to 60% (Table 32-5).79–81 The standard of care is now clearly PEG-IFN alpha plus ribavirin.1 The two competing drugs, PEG-IFN alpha-2b and PEG-IFN alpha-2a, differ in their pharmacokinetic profile. They were developed with clinical trials of different designs and combined with different doses of ribavirin. The approval by the two major registration authorities, the FDA in the USA and the European Medicines Evaluation Agency (EMEA) in the European Union led to different labelings of these regimens (Table 32-6). Postapproval commitments and investigator-initiated studies further optimized treatment with PEG-IFN. Despite the different pharmacokinetic profiles of these two molecules and the different dosing regimens, remarkably similar results have been obtained across trials. A headto-head comparison of the two peginterferons has so far only been performed in small pilot studies. Of all the predictive parameters for achieving an SVR, HCV genotype is the most important. Genotype also determines ribavirin dose and treatment duration (Table 32-5). There is general consensus that genotype 1 patients should be treated for 48 weeks, while genotype 2 and 3 patients only need 24 weeks of treatment in most circumstances.81,82 For PEG-IFN alpha-2a the recommended dose is 180 mg interferon for all patients while PEG-IFN alpha-2b should be dosed at 1.5 mg/kg body weight once per week subcutaneously. Table 32-5. Sustained virological response rates (SVR) to pegylated interferon (PEG-IFN)-alpha and ribavirin depending on hepatitis C virus (HCV) genotype Study 1.5 mg/kg PEG-IFN alpha-2b + 800 mg ribavirin 48 weeks HCV genotype 1 HCV genotypes 2 and 3 HCV genotype 4 180 mg PEG-IFN alpha-2a + 1000/1200 mg ribavirin 48 weeks HCV genotype 1 HCV genotypes 2 and 3 180 mg PEG-IFN alpha-2a + 1000/1200 mg ribavirin 48 weeks HCV genotype 1 24 weeks HCV genotype 1 48 weeks HCV genotypes 2 and 3 24 weeks HCV genotypes 2 and 3 180 mg PEG-IFN alpha-2a + 800 mg ribavirin 24 weeks 48 weeks HCV genotype 1 24 weeks HCV genotype 1 48 weeks HCV genotypes 2 and 3 24 weeks HCV genotypes 2 and 3 1.5 mg/kg PEG-IFN alpha-2b + 800–1400 mg ribavirin 24 weeks HCV genotype 2 HCV genotype 3
SVR 54% 42% 82% 50% 56% 46% 76%
51% 41% 77% 78%
Reference Manns et al., 200179
Fried et al., 200280
Hadziyannis et al., 200481
40% 29% 73% 78% Zeuzem et al, 200482 93% 79%
675
Section V. Liver Diseases Due to Infectious Agents
Table 32-6. Recommended doses for hepatitis C therapy
Table 32-7. Common side effects of interferon-alpha and ribavirin
Drug
Dose and duration
Pegylated interferon alpha-2a (Pegasys)
180 mg 1/week SC HCV genotypes 1 (and 4): 48 weeks HCV genotypes 2 and 3: 24 weeks 1.5 mg/kg body weight 1/week SC HCV genotypes 1 (and 4): 48 weeks HCV genotypes 2 and 3: 24 weeks In combination with pegylated interferon alpha-2a per day HCV genotype 1: 1000 mg (< 75 kg), 1200 mg (> 75 kg) HCV genotypes 2/3: 800 mg In combination with pegylated interferon alpha-2b: 800 mg, <65 kg; 1000 mg, 65–85 kg; 1200 mg, 85–105 kg; 1400 mg, 105 kg
Constitutional Fatigue Headache Fever Myalgias Gastrointestinal Nausea Anorexia Diarrhea Insomnia Irritability Depression Dermatological Alopecia Skin rash Laboratory Anemia Neutropenia Thrombocytopenia
Pegylated interferon alpha-2b (Peg-Intron) Ribavirin (Rebetol, Copegus)
HCV, hepatitis C virus.
For genotype 1-infected patients, ribavirin should be given orally in combination with PEG-IFN alpha-2a at doses of 1000 mg/day for patients weighing <75 kg and at 1200 mg/day for all those ≥75 kg. In genotype 2 and 3 patients 800 mg should be given to all patients in combination with PEG-IFN alpha-2a. In contrast, ribavirin doses may be weight-based across a broader range of ribavirin doses for genotype 1 patients receiving PEG-IFN alpha-2b (800 mg < 65 kg, 1000 mg 65–85 kg, 1200 mg > 85 kg, and 1400 mg > 105 kg/day). In order to minimize side effects as well as costs, the “early stopping rule” may be applied to genotype 1 patients who fail to achieve an EVR. This means that quantitative HCV RNA measurement should be performed before and after the initial 12 weeks of treatment. If there is a less than 2 log decline in viral load, treatment should be stopped since further treatment is unlikely to lead to an SVR. Side effects with treatment may be significant and occur in the vast majority of patients (Table 32-7). For interferon these are arthralgias, muscle pain, flu-like symptoms, fever, and others. Concerning laboratory tests, leukopenia, granulocytopenia, and thrombocytopenia are most important and need to be monitored. For ribavirin, anemia is the most prominent adverse event leading to a significant drop in hemoglobin within the first 4 weeks of treatment. In patients with coronary artery disease a careful evaluation of the risk–benefit ratio of this potentially dangerous treatment is essential. In individual patients, interferon-related adverse events may occur that require specific management and, in some cases, termination of therapy. Among the most important side effects are depression, autoimmune thyroid disease, alopecia, and diabetes (Table 32-7). Patients and doctors must be aware of these adverse events and need to react appropriately. Reduction of peginterferon or ribavirin may become necessary in up to 20% of patients due to side effects. Retrospective analysis has shown that it is particularly important for genotype 1 patients to be compliant and take more than 80% of both drugs for more than 80% of the time in order to optimize response rates.83 Treatment of patients with persistently normal ALT is safe and responses are comparable to those with abnormal ALT.84 Thus in most current guide-
676
lines, elevated ALT is not a requirement for therapy.22 In patients with normal ALT a biopsy should be performed and treatment initiated if a fibrosis score of at least 1 (Ishak or Metavir) is present.
NON-RESPONDERS TO PREVIOUS TREATMENT In most countries the present standard treatment with the combination of peginterferon plus ribavirin is only approved and reimbursed for treatment-naive or those previously treated with interferon monotherapy. Unfortunately, a growing number of patients are non-responders to current therapies (more than 50% of genotype 1 patients, 20% of genotype 2 and 3 patients, and 40% of genotype 4 patients). These patients are highly motivated and are often waiting for new approaches. These patients are primary candidates for the new therapies described below. In addition, there have been numerous trials attempting to improve treatment outcomes with the drugs available today. High-dose induction regimens with either peginterferon or ribavirin are under investigation. Several studies have also explored retreatment with low-dose peginterferon for prolonged durations, in order to slow disease progression by exploiting the antifibrogenic and antiproliferative effects of interferons. Long-term trials of this approach are in progress, including Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C), that is exploring low-dose PEG-IFN alpha-2a,85 Evaluation of Peg-Intron in Control of Hepatitis C Cirrhosis (EPIC-3), that is investigating PEG-IFN alpha-2b86 compared to placebo, and Colchicine versus Peg-Intron Long Term (COPILOT), that is evaluating low-dose PEG-IFN alpha-2b compared to colchicine.87 However, significant progress in treatment will only be achieved using innovative new drugs, as described below.
SELECT PATIENT POPULATIONS HCV infection is prevalent in certain populations in the USA. These include 300 000 injection drug users,88 175 000 homeless,89 100 000 children,90 280 000 veterans,91 300 000 with HIV infection,92 310 000 of those who are incarcerated, and 940 000 living below the
Chapter 32 HEPATITIS C
poverty level.4 Many of these patient groups have been excluded from the registration trials of peginterferon plus ribavirin therapy.
AFRICAN-AMERICANS African-Americans have the highest prevalence of HCV infection of any racial group in the USA. African-Americans also have a high prevalence of HCV genotype 1 infection.93 Unfortunately, treatment is particularly problematic in this group. African-Americans have lower rates of spontaneous viral clearance and lower rates of response to peginterferon and ribavirin (SVRs in African-Americans were 18–26%).94,95 In genotype 1 patients, blacks have a slower and quantitatively smaller decline in HCV RNA levels with interferon with or without ribavirin compared with whites.96 In one study of genotype 1 patients, PEG-IFN alpha-2a 180 mg/week plus ribavirin 1000 or 1200 mg/day was administered in a prospective trial for 48 weeks to 78 blacks and 28 whites. End-of-treatment responses were 39 and 52% and SVRs were 26 and 39% in the African-Americans and Caucasians respectively. Histological improvement (defined as an improvement of >2 in histological activity index) was seen in 66% of African-Americans compared with 75% of Caucasians. Treatment was tolerated equally well, except for a higher incidence of severe neutropenia (defined as absolute neutrophil count less than 500/mm3) in the African-American group that was not associated with increased risk of infection. A second trial of PEG-IFN alpha2b (1.5 mg/kg per week) plus ribavirin (1000 mg/day for 12 weeks and 800 mg for an additional 36 weeks) further underscored reduced treatment responses. End-of-treatment responses were 20% and 58% and SVRs were 19% and 52% in African-Americans and Caucasians respectively.95 Reasons for resistance to HCV antiviral therapy are under investigation, but low response rates are not simply explained by the high prevalence of genotype 1 infection seen in this group. As with other patient populations, failure to achieve an early virological response at 12 weeks is strongly predictive of a low likelihood of achieving an SVR with 48 weeks of peginterferon plus ribavirin. Whether prolonging therapy beyond 48 weeks is beneficial is currently unknown.
HEPATITIS C/HIV CO-INFECTION Highly active antiretroviral therapy (HAART) has markedly changed the natural history of HIV infection, such that life expectancy is now comparable to those who are HIV-negative.97 Because of overlapping risk factors for HCV and HIV infection, co-infection with the two viruses is common, particularly in those with HIV who have acquired infection through injection drug use. Of the approximately 1 000 000 HIV-infected individuals in the USA, 30% are estimated to have HCV/HIV co-infection. Similar proportions of HIVinfected patients in Europe seem to be HCV/HIV co-infected, although the proportion is even higher in certain countries (such as Spain, with a prevalence of 50%). The prevalence of HCV depends on the risk factor for HIV infection (85% in injection drug users, 14% heterosexual and 10% with homosexual contact as their risk for HIV98). The CDC and the American Association for the Study of Liver Diseases have recommended that all HIV-infected patients be tested for HCV and that all HIV-positive patients with liver injury of unknown cause should be tested for HCV RNA, even if the anti-HCV test is negative.22 Guidelines have recommended consideration of HCV antiviral therapy in HCV/HIV co-infected
patients, although response to peginterferon plus ribavirin therapy is lower than observed in those with HCV infection alone.22 These guidelines are supported by results of randomized controlled clinical trials.99–102 The need to treat patients with HCV/HIV co-infection is driven in part by concerns that liver disease progression is accelerated in those with co-infection, compared to the natural history in those with HCV monoinfection.44 A meta-analysis of eight separate studies has shown that HCV/HIV-infected patients have a twofold risk of cirrhosis diagnosed on liver biopsy and a sixfold risk of decompensated liver disease with clinical complications when compared to HCV monoinfected patients.103 A 10-year incidence of cirrhosis has been estimated to be 14.9 and 2.6% respectively in HCV-infected patients with and without HIV infection.104 Risk of liver disease progression appears to be particularly true for patients with HIV-related immune compromise.105 Currently only one therapy, PEG-IFN alpha-2a (180 mg/week) plus ribavirin, is FDA-approved for HCV treatment of HCV/HIV co-infection. While the FDA-approved dose of ribavirin is 800 mg/day, and this was the dose of ribavirin included in the phase III trials of peginterferon plus ribavirin,100 several guidelines have advocated the use of higher doses of ribavirin (1000/1200 mg/day based on a 75-kg body weight) in order to give these patients an improved likelihood of treatment response.22,102 Current recommendations are to consider treatment of patients with baseline CD4 counts of >200 cells/ml, or CD4 counts of >100 but <200 cells/ml if HIV RNA is less than 5000 copies/ml. It is important that patients have compensated liver disease since hepatic decompensation with peginterferon in patients with modest impairment of synthetic function (Child–Pugh score > 6) has been described.106 The primary goal of treatment in patients with co-infection is viral clearance (achieving an SVR), although secondary goals include delay of histological and clinical liver disease, as for those with HCV monoinfection. Unfortunately overall response to peginterferon plus ribavirin therapy is reduced in patients with HCV/HIV, in part because of a high prevalence, as variables were shown to be associated with nonresponse in HCV monoinfected patients (genotype 1 infection, high HCV RNA levels, advanced liver disease, and African-American race). Pretreatment HCV RNA levels in an international trial of PEG-IFN alpha-2a plus ribavirin was 14 000 000 copies/ml, more than twofold higher than baseline levels in monoinfected treatment trials.100 There have been three recent published studies of PEG-IFN plus ribavirin in co-infected patients demonstrating superiority to interferon plus ribavirin.99–101 In the AIDS Clinical Trials Group (ACTG) 5071 study, 133 HIV/HCV-co-infected adults were randomized to receive either standard interferon alpha-2a subcutaneously at 3 million international units (MIU) three times a week (tiw) or PEGIFN alpha-2a subcutaneously at 180 mg/week, each combined with increasing doses of ribavirin escalated from 600 to 1000 mg daily. Virologic tests at both 48 weeks end of treatment response (ETR) and 72 weeks (SVR) showed a greater response in the PEG-IFN arm when compared to the standard interferon arm (overall SVR of 27%, and SVR of 14 and 73% in genotype 1 and non-1 in the peginterferon group compared to overall SVR of 12%, and SVR of 6 and 33% in genotype 1 and non-1 in the interferon plus ribavirin arm). Overall tolerability of antiviral therapy was lower than in HCV monoinfected patients with 12% of patients in both arms discon-
677
Section V. Liver Diseases Due to Infectious Agents
tinuing treatment. In the AIDS PEGASYS Ribavirin International Coinfection Trial (APRICOT),100 860 HIV/HCV-co-infected adults were randomized to one of three treatment arms: (1) standard interferon alpha-2b at 3 MIU tiw plus ribavirin at 800 mg/day; (2) PEGIFN alpha-2a at 180 mg/week plus ribavirin 800 mg/day; or (3) PEG-IFN alpha-2a at 180 mg/week plus placebo. Overall SVR in the peginterferon plus ribavirin arm demonstrated superiority to standard interferon plus ribavirin (49 and 14% respectively) with an intermediate response in patients receiving peginterferon monotherapy (SVR of 33%). SVR in genotype 1 and non-1 patients receiving peginterferon plus ribavirin were 29 and 62% respectively, based on an intent-to-treat analysis. Peginterferon also appeared to have an anti-HIV effect, since HIV RNA levels decreased by 0.9 logs. The third multicenter trial (RIBAVIC study) of peginterferon plus ribavirin for co-infection, included 412 patients randomized to either PEG-IFN alpha-2b at 1.5 mg/kg per week compared to standard interferon alpha-2b at 3 MIU tiw, with both arms receiving ribavirin 800 mg/day. As for the PEG-IFN alpha-2a trials, SVR rates in the PEG-IFN arm were superior to the standard interferon arm (27 and 15% respectively). Response to peginterferon plus ribavirin in genotype 1 and non-1 patients were 15 and 46% respectively.99 While all three studies show the superiority of peginterferon plus ribavirin to standard interferon plus ribavirin, the SVR rates reported appear to vary. Several factors may account for the lower overall sustained virologic response rates in the ACTG 5071 trial and the RIBAVIC Study. In the ACTG 5071 trial, the initial starting dose of ribavirin was lower than the other two studies at a total dose of 600 mg/day, and ribavirin dose seems particularly important for response in genotype 1-infected patients. In addition, about 33% of patients in the ACTG 5071 trial were African-American, a population that has consistently been shown to be resistant to HCV therapy. Treatment discontinuations were also high in these trials (42% in RIBAVIC), with severe side effects occurring in 31% of patients. Many patients (80%) had a history of intravenous drug use that may have contributed to non-adherence. Thus, differences in response to PEG-IFN plus ribavirin therapy between trials appear to be multifactorial, in part due to differences in study design, but also due to differences in characteristics of the study populations. Given the lower response rates in HCV/HIV co-infected patients, recent guidelines have recommended ribavirin dosing of 1000/ 1200 mg/day,22,102 although ribavirin-associated hemolytic anemia would be predicted to be significant and may necessitate the use of adjunctive therapy with epoietin, to improve tolerability.107 Recently, the first European Consensus Conference on coinfected patients (1stEEC) published its recommendations.107a
PATIENTS WITH COMPENSATED CIRRHOSIS Patients with cirrhosis have a substantial need for therapy yet have a somewhat lower response to peginterferon plus ribavirin than patients with earlier-stage disease. Goals of treatment include eradication of HCV infection, delay in cirrhosis progression and decompensation and prevention of HCC. Risk of developing HCC is 1.5% per year in a cirrhotic patient, a risk that may be reduced by HCV antiviral therapy.108,109 Data on treating patients with compensated cirrhosis are largely derived from subgroup analyses of clinical trials110; few prospective studies have focused on this population alone. One prospective trial of PEG-IFN alpha-2a in patients with
678
bridging fibrosis and/or cirrhosis demonstrated that viral eradication was achievable in 30% of patients.111 As with all HCV trials, SVR was lower in patients with cirrhosis and genotype 1 infection (SVR in patients with low viral load and high viral load of 16 and 10% respectively). 111 Histological response (defined as >2-point reduction in histological activity index) was seen in 54% of patients. No prospective study has been conducted of peginterferon plus ribavirin in patients with cirrhosis, but subset analysis of patients with stage III and IV fibrosis from the registration trials of peginterferon provides guidance as to expected response. PEG-IFN alpha-2a (180 mg/week) plus ribavirin (1000/1200 mg/day) yielded SVR of 43 and 52% in two separate studies,80,81 while PEG-IFN alpha-2b (1.5 mg/kg per week) plus ribavirin (800 mg/day) yielded a response of 33%.79 Long-term goals of preventing complications of HCV cirrhosis with low-dose peginterferon therapy are under evaluation.
PATIENTS WITH COMPLICATIONS OF CIRRHOSIS HCV-associated end-stage liver disease with or without HCC has become the leading diagnosis in patients undergoing liver transplantation, affecting approximately half of all patients who are potential candidates (Chapter 49). Liver transplantation remains a life-saving option in patients with decompensated HCV cirrhosis, including those with HCV-related HCC, although recurrent HCV infection of the graft is a major cause of morbidity and mortality following liver transplantation (Chapter 52). Thus there is a great unmet medical need to treat HCV infection in patients with complications of cirrhosis. Goals of such treatment include eradication of infection prior to liver transplantation, in order to prevent post-transplantation reinfection, as well as slowing of HCV disease progression with the hope of obviating the need for liver transplantation. Unfortunately, peginterferon plus ribavirin is poorly tolerated in patients with complications of cirrhosis and, as such, these drugs are largely contraindicated. Major concerns about therapy are that leukopenia and thrombocytopenia will be worsened by interferon, making patients at risk for life-threatening infections and/or bleeding complications. Select patients with HCV may be candidates for low doses of interferon and ribavirin therapy provided that therapy is administered under close supervision of providers with experience in the management of end-stage liver disease. Outcomes of a low accelerating-dose regimen of interferon and ribavirin in a case series of 111 patients have recently been reported.69 Sixty-three percent of patients had clinical complications of HCV cirrhosis. While many patients were Child’s A cirrhotics, 45 were Child’s B and 23 were Child’s C. Mean Child–Pugh score of the group was 7 and model for end-stage liver disease (MELD) score was 11. Forty-six percent of patients lost HCV RNA at the end of treatment and 24% were HCV RNA-negative at follow-up. SVR was highly dependent on genotype (13 and 50% in patients with genotype 1 and non-1 infection respectively). Predictors of achieving an SVR included non-1 genotype, Child’s A cirrhosis (for genotype 1 only), treatment duration, and the ability to tolerate full dose therapy. Of the 15 patients who were HCV RNA-negative at the time of liver transplantation, 12 remained virus-free post-transplantation.69 Prevention of posttransplantation reinfection of the graft is a highly desirable goal, given the progressive nature of post-transplantation liver disease.39 These encouraging results support the approach of treating patients with clinical complications of cirrhosis with low doses of interferon
Chapter 32 HEPATITIS C
plus ribavirin. Tolerability of peginterferon plus ribavirin is less clear since cytopenias would be predicted to be worse with peginterferon than with standard interferon preparations, although hematological growth factors may be able to ameliorate some of these side effects.69 While interferon plus ribavirin may provide benefit to select patients with complications of cirrhosis, the medical need for new non-interferon-based options is high.
is common in patients following liver transplantation. Consequently use of ribavirin for treatment of recurrent HCV disease is associated with a high likelihood of treatment-associated hemolysis.119 Difficulties in administering ribavirin in patients following liver transplantation may also contribute to reduced treatment responses in this group (Chapter 52).
PATIENTS WITH RENAL DISEASE
FUTURE THERAPEUTIC APPROACHES AND NOVEL THERAPIES (Figure 32-3, Table 32-8)
Renal disease can be an extrahepatic manifestation of HCV infection with cryoglobulinemia and MPGN112 (see above). Management of HCV-related MPGN is problematic. While improvement in cryoglobulin levels and serum creatinine may occur, benefits appear to be transient, with relapse frequent when treatment is stopped.22 Potential benefits of improvement in renal function and decrease in proteinuria from long-term viral suppression with antiviral treatment have been inadequately studied. There are several other interactions between HCV-associated liver disease and renal dysfunction. The prevalence of HCV antibodies in patients on chronic hemodialysis is 8.6%,113 although prevalence in certain US dialysis centers may be higher, and prevalence from other parts of the world ranges from 5 to 50%.22 HCV infection is an independent risk factor for death in dialysis patients.114 Reasons for the high prevalence of HCV infection include transfusions before the availability of effective screening tests as well as nosocomial transmission of HCV infection in dialysis units. Screening and careful attention to infection control practices are essential to prevent HCV transmission. In patients on hemodialysis, ALT values are typically lower than in non-uremic patients but normal ALT levels do not exclude histologically significant liver disease. The goals of antiviral therapy in patients on dialysis are to slow liver disease progression and to eradicate HCV infection in those who might subsequently undergo kidney transplantation. Eradication of HCV is particularly important since chronic HCV disease adversely affects patient and graft survival after renal transplantation, and HCV antiviral therapy is not possible following kidney transplantation because treatment carries an unacceptable risk of allograft rejection.115 Ribavirin is also contraindicated in patients on dialysis and PEG-IFN alpha-2b should be used with caution in patients with creatinine clearance of less than 50 ml/min. Specific recommendations for dosing of PEG-IFN alpha2a in dialysis patients are available (135 mg/week).22 There is a general sense that treatment responses are greater with interferon monotherapy in dialysis patients than in those with normal renal function, possibly because reduced clearance of interferon effectively increases the administered dose. However, information about treatment outcomes with interferon/peginterferon monotherapy is limited and largely derived from case series. Overall SVRs from these case series are 33% (range 14–71%) with SVRs in genotype 1 patients of 26%.22,116,117 Virological relapse and side effects are common and patients should be monitored closely for toxicity. HCV antiviral therapy is also problematic in HCV-infected patients with lesser degrees of renal dysfunction.118 Since ribavirin is contraindicated in patients with a creatinine clearance of less than 50 ml/min, treatment outcomes are reduced in patients who can only receive peginterferon monotherapy. Because of nephrotoxicities associated with immunosuppressive therapies, renal dysfunction
UNMET NEEDS IN THE TREATMENT OF HEPATITIS C Impressive progress has been made in the treatment of hepatitis C since the discovery of the virus in 1989. Acute hepatitis C has become a treatable disease and chronicity can be prevented with therapy in 90% of cases. Genotype 2 and 3 patients, particularly those with genotype 2 infection, can be cured in around 90%. However effective therapy is still not available for a significant proportion of patients. Fifty percent of genotype 1 patients are nonresponders to standard treatment. These patients are highly motivated and, since many have advanced liver disease, are in substantial need of treatment. In addition, present treatments are
Table 32-8. Future therapeutic approaches and novel hepatitis C therapies Optimization of the use of interferon and ribavirin Optimizing dose and duration Additional agents _ Amantadine _ Histamine hydrochloride _ Thymosin-alpha New interferon delivery systems New generations of interferons, interferon inducers, other cytokines, and growth factors Consensus interferon Omega-interferon Albumin-bound interferon IL-28 A, IL-28 B, IL-29 ANA 245 – oral interferon inducers: TLR 7 and 9 agonists Erythropoietin Thrombopoietin Ribavirin analogs Viramidine Inosine monophosphate dehydrogenase (IMPDH) inhibitors – VX497 Small molecules with direct antiviral modes of action Specific hepatitis C viral enzyme inhibitors • Protease inhibitors – BILN 2061 – VX 950 – SCH 503034 • Polymerase inhibitors – NM 283/valopicitabine • Helicase inhibitors Antisense oligonucleotides Therapeutic vaccines E2 protein vaccine IC41 polypeptide vaccine HCl, hydrochloric acid; IL, interleukin; ANA, antinuclear antibody; IMPDH,
679
Section V. Liver Diseases Due to Infectious Agents
On Market
Ribavirin
IFN & PEG IFN
Phase III
Viramidine Thymosin
Other IFNs Gamma IFN
STAGE
Phase II
antisense
IMPDH inhibitors
Ribozymes
Phase I
Polymerase inhibitors
E2 Vaccine
Histamine HCl
IL 10 & IL 12 others
Figure 32-3. Hepatitis C drug development. Various drugs at various stages of development. Some of these compounds such as levovirin and ribozymes are no longer in clinical development. Polymerase inhibitors are now in both phase I and phase II development. IFN, recombinant interferon alpha; PEG-IFN, pegylated recombinant interferon-alpha; HCl, hydrochloric acid; IMPDH, inosine monophosphate dehydrogenase; IL, interleukin; HCV, hepatitis C virus; siRNA, small interfering RNA. (Modified according to McHutchinson, Falk Liver Week, Freiburg, Germany 2004).
Protease Inhib Levovirin Apoptosis Inhibitors
HCV Immunotherapy
siRNA
Preclinical HCV Vaccines
Research
Many others including Antisense Antifibrotics Immune stimulants Gene therapy
long-term, costly, and associated with significant side effects. Furthermore our knowledge on the treatment of HCV genotypes other than 1, 2, and 3 is very limited. Only recently have we seen reliable data on HCV genotype 4, the most frequent genotype in the Arab world, where the prevalence of HCV in the general population reaches 20%. At present we stratify our treatment according to infecting genotype, although in the future, early viral kinetics may be more important than genotype in guiding treatment regimens. Regardless of evolving treatment strategies, it is likely new innovative drugs will be necessary in order to improve treatment outcomes substantially. The development of innovative drugs has been problematic due to unpredicted adverse events as well as to the early development of resistance. More drugs are failing each year than are successfully entering clinical development. Therefore there is still a strong rationale for optimizing the use of the currently available drugs (peginterferon and ribavirin) and/or for developing next generations of both compounds. Approaches for future therapies are as follows: (1) optimization of the use of interferon and ribavirin; (2) development of new generations of type 1 interferons, interferon inducers, other cytokines, and growth factors; (3) development of ribavirin analogs with an improved safety profile; (4) development of small molecule inhibitors with direct antiviral modes of action; and finally, (5) development of therapeutic vaccines. Each approach will be discussed individually.
OPTIMIZATION OF THE USE OF INTERFERON AND RIBAVIRIN An important question is whether a proportion of patients with genotype 1 infection would benefit from prolonged treatment with peginterferon plus ribavirin. Patients who have cleared the virus
680
from their blood by week 12 have a high chance of becoming a sustained responder with 48 weeks of treatment. Those who clear virus between week 12 and week 24, the so-called “late” or “slow” responders, have a high relapse rate after 48 weeks. The likelihood of achieving an SVR in those with a 2-log drop by week 12, but who still have detectable virus at this time point, is intermediate for achieving an SVR with 48 weeks of treatment. Treatment duration may need to be adjusted according to individual viral kinetics. Initial studies of PEG-IFN alpha-2a plus ribavirin have shown that, while the overall population of genotype 1 patients does not benefit from prolonging treatment to 72 weeks, a subgroup of patients with a slow response may do so.120 In addition, patients known to be difficult to treat, such as those with high body mass index (>25) and those with high viral load (>6.5 log HCV RNA in serum) may also show a better SVR with prolonged therapy.121 However, extending treatment beyond 1 year may also lead to greater treatment discontinuations. Several recent studies have shown that 24 weeks may be more than adequate for certain patients with genotype 2 or 3 infection. The initial study evaluating the combination of PEG-IFN alpha and ribavirin79 included an arm with a lower dose of PEG-IFN alpha-2b (0.5 mg/kg per week) after 1.5 mg/kg per week had been given for the first month. Treatment responses for genotype 2 and 3 patients were equally good or better than the current FDA-approved dose of 1.5 mg/kg per week (80% SVR). Therefore it is logical to evaluate lower doses of peginterferon as well as shorter duration for these easy-to-treat patients. An RVR after 4 weeks of therapy seems to be the crucial factor for the success of a shorter duration of therapy. Patients with this RVR may only be treated for 16 weeks,122 14 weeks,123 or even 12 weeks.124 While in Europe and North America genotype 2 and 3 patients make up to 30% of all HCV patients, in many parts of Asia (India, Thailand, and more), genotype 2 and 3
Chapter 32 HEPATITIS C
patients are the predominant genotype and make up to 80% of all infections. Cost-effective approaches to the treatment of genotype 2/3 infection are particularly important in the developing world.
treat patient populations such as those with HIV co-infection or those prior to or following liver transplantation.
RIBAVIRIN ANALOGS NEW GENERATIONS OF TYPE 1 INTERFERONS, INTERFERON INDUCERS, OTHER CYTOKINES, AND GROWTH FACTORS There are a number of other type 1 interferons on the market or under clinical development. In addition conventional interferon alpha will soon be available as a generic, which will be particularly relevant in parts of the world where patients have to pay for their drugs and/or where social security systems are inadequate for drug reimbursement. Consensus interferon is approved in many countries, although a pegylated formulation is not available. Therefore daily injections of consensus interferon may be necessary to achieve optimum response. Many other cytokines have been tried with limited or no success, among them IL-10 and IL-12. Thymosin alpha is still under evaluation in combination with peginterferon. At present interferon alpha linked to albumin (albuferon) is starting phase III development. This interferon only requires once-monthly dosing.125 A different approach is the application of orally applied interferon inducers, but none has proven successful thus far. Anemia is a major problem in peginterferon and ribavirin combination therapies. Reduction of ribavirin does impair treatment outcomes, particularly for genotype 1 patients. The application of recombinant erythropoietin pre-emptively or when anemia develops has been studied and may offer the advantage of preventing ribavirin dose reductions.126 However, this growth factor adds significantly to the cost of treatment, and epoietin has not been shown to improve SVR. Erythropoietin use is probably most justified in difficult-to-
Adding ribavirin to interferon alpha clearly has been a substantial advance in the treatment of hepatitis C. Non-immune hemolytic anemia is the most important dose-limiting side effect. For treatment of genotype 1 infection in particular, compliance with full dose of ribavirin seems to be an important determinant of treatment response. Therefore several attempts have been made to develop ribavirin analogs with less anemia but comparable or possibly even better antiviral effects as ribavirin. At present viramidine, a “livertargeted” prodrug of ribavirin, is in phase III clinical development. Preliminary analysis of phase II results has suggested that anemia is less severe and less frequent than with ribavirin.127 Whether SVR results will be the same with viramidine as with ribavirin in combination with peginterferon is currently unknown. Results of the phase III trials are awaited. Approaches such as viramidine may be particularly useful for special patient groups such as those with recurrent hepatitis C after liver transplantation and those with renal disease.
SMALL MOLECULES WITH DIRECT ANTIVIRAL MODES OF ACTION (POLYMERASE, HELICASE, AND PROTEASE INHIBITORS) After unraveling the crystal structures of all three major HCV enzymes involved in replication and after the discovery of an HCV replicon system128 that facilitated in vitro testing of novel antiviral agents, there has been a concerted effort in drug discovery for innovative hepatitis drugs (Figure 32-4). Numerous patents have been Figure 32-4. The hepatitis C virus (HCV)-replicon system: milestone for the development of anti-HCV compounds.128 IFN, interferon.
R2884G HCV 5¢
EMCV Neo
IRES
3
4A
4B
5A
5B
3¢
IRES
subgenomic replicon
G418 IFN-a HuH-7 cells with HCV-replicon
IFN-a/G418 resistant cell clones 19 cell clones
expanded cell clones Preparation of total-RNA
2/16
G418-resistant HuH-7 cell clones 16 cell clones
Transfection in naive HuH-7 NUR G418!!!
IFN-a IC50
681
Section V. Liver Diseases Due to Infectious Agents
filed but only a small number of these compounds have so far entered clinical development. The proof concept supporting this approach was established by Boehringer Ingelheim with BILN 2061 that demonstrated a 4-log decrease in viral load after 48 h of oral therapy (Figure 32-5).129,130 These drugs work in treatment-naive as well as in non-responders to prior peginterferon and ribavirin, but show 100-fold less activity against genotype 2 and 3 compared with genotype 1 infection.131 This selectivity for certain genotypes is likely a function of the interaction between small-molecule inhibitors and the HCV protease, since the protease demonstrates sequence differences between genotypes. The in vitro replicon system also suggested that these novel drugs would face problems of drug resistance in vivo. Due to side effects from long-term use in animals, BILN 2061 was withdrawn from further clinical development. Recently, VX 950 became a second protease inhibitor that has also shown a 4.4-log decrease
of HCV viral load after 2 weeks of treatment. Phase II studies are planned to explore this drug in combination with peginterferon.132 Polymerase inhibitors are a different group of drugs with less potent antiviral activity. In monotherapy NM283, or valopicitabine, achieves a 1–2-log decrease in HCV RNA levels133 after 4 weeks. Viral suppression is more potent when this drug is combined with peginterferon, reaching a 3-log reduction in viral load after 24 weeks.134 Addition of interferon seems to increase antiviral efficacy and may prevent resistance development. Apart from gastrointestinal side effects, valopicitabine was well tolerated. Phase II studies are under way. Due to either resistance development and/or to insufficient antiviral activity, all these small molecules acting directly on viral replication will likely need to be administered in combination with drugs such as peginterferon with or without ribavirin. Another antiviral approach has been to cleave the hepatitis C genome in a region that is crucial for the virus. Ribozymes are such
10 000 000
1.GT1b 2.GT1b 3.GT1a 4.GT1a
Virus load: copies RNA/ml
100 000
5.GT1b 6.GT1a 7.GT1b 1000 8.GT1b 9.GT1b 10.GT1b Upper end of linear range
10 000
X Lower end of linearrange X
X
1000 0
20
40
60
80
100
120
140
Administration
160
Hours Figure 32-5. Proof of principle: the BILN 2061 hepatitis C virus (HCV) protease inhibitor. Drop in HCV viral load over time in patients with chronic hepatitis C, genotype 1, naive and non-responders to previous therapy.130
682
Chapter 32 HEPATITIS C
“molecular scissors” that cleave the HCV internal ribosome entry site.135 This concept works in vitro but ribozymes have not been demonstrated to be beneficial in patients. Further concepts are antisense molecules (ISIS-14803)136 and gene silencing using small interfering RNAs (SiRNA).137
THERAPEUTIC VACCINES Patients with chronic HCV infection have a diminished T-lymphocyte reactivity against the various T-cell epitopes compared to patients resolving acute HCV infection. Furthermore T lymphocytes from patients with chronic HCV have less capability to produce cytokines, such that their cellular immune functions are impaired.64 Therefore it is logical to explore whether stimulation of the patients’ immune system with therapeutic vaccines can overcome these deficiencies. Two such therapeutic vaccines are currently in clinical development and data are in the public domain. First, the E2 protein vaccine by Innogenetics has demonstrated reduced liver fibrogenesis despite a lack of viral suppression in a phase I/II study.138 A second vaccine from Intercell in Vienna includes five peptides spanning all the major T-cell epitopes together with an innovative adjuvant. Although CD4 and CD8 T-cell responses could be demonstrated, the associated reduction in viral load has thus far been unimpressive.139 The potential therapeutic value of these vaccines should perhaps be evaluated in easy-to-treat patients in order to prevent relapse, rather than in the difficult-to-treat non-responder population. Only clinical studies will reveal whether HCV therapeutic vaccines will have a place in the management of patients with chronic HCV.
REFERENCES 1. National Institutes of Health Consensus Development Conference Statement. Management of hepatitis C: June 10–12, 2002. Hepatology 2002; 36:S3–S20. 2. World Health Organization. Global distribution of hepatitis A, B and C, 2001. Wkly Epidemol Rec 2002; 77:41–48. 3. Abdel-Aziz F, Habib M, Mohamed MK, et al. Hepatitis C virus (HCV) infection in a community in the Nile delta: population description and HCV prevalence. Hepatology 2000; 32:111–115. 4. Alter MJ, Kruszon-Moran D, Nainan OV, et al. The prevalence of hepatitis C virus infection in the United States, 1988 through 1994. N Engl J Med 1999; 341:556–562. 5. Centers for Disease Control and Prevention. Summary of notifiable diseases, United States, 1998. Morb Mortal Wkly Rep 1998; 47:1–39. 6. Blatt LM, Mutchnick MG, Tong MJ, et al. Assessment of hepatitis C virus RNA and genotype from 6807 patients with chronic hepatitis C in the United States. J Viral Hepatitis 2000; 7:196–202. 7. Simmonds P. The origin and evolution of hepatitis viruses in humans. J Gen Virol 2001; 82:693–712. 8. Kim WR. The burden of hepatitis C in the United States. Hepatology 2002; 36:S30–S34. 9. Armstrong GL, Simard EP, Wasley A, et al. The prevalence of hepatitis C (HCV) infection in the United States, 1999–2002. Hepatology 2004; 40:176A. 10. Kim WR, Poterucha JJ, Benson JT, Therneau TM. The impact of competing risks on the observed rate of chronic hepatitis C progression. Gastroenterology 2004; 127:749–755.
11. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999; 340:745–750. 12. El-Serag HB, Davila JA, Peterson NJ, McGlynn KA. The continued increase in the incidence of hepatocellular carcinoma in the United States: an update. Ann Intern Med 2003; 139:817–823. 13. Armstrong GL, Alter MJ, McQuillan GM, Margolis HS. The past incidence of hepatitis C virus infection: implications for the future burden of chronic liver disease in the United States. Hepatology 2000; 31:777–782. 14. Davila JA, Morgan TO, Shaib Y, et al. Hepatitis C infection and the increasing incidence of hepatocellular carcinoma: a population-based study. Gastroenterology 2004; 127:1372–1380. 15. El-Serag HB, Mason AC, Key C. Trends in survival of patients with hepatocellular carcinoma between 1977 and 1996 in the United States. Hepatology 2001; 33:62–65. 16. Alter MJ. Prevention of spread of hepatitis C. Hepatology 2002; 36:S93–S98. 17. Conry-Cantilena C, Gibble J, Melpolder J, et al. Routes of infection, viremia, and liver disease in blood donors found to have hepatitis C virus infection. N Engl J Med 1996; 334:1691–1696. 18. Thomas D, Vlahov D, Solomon L, et al. Correlates of hepatitis C virus infections among injection drug users. Medicine 1995; 74:212–220. 19. Bell J, Batey RF, Crewe E, et al. Hepatitis C virus in intravenous drug users. Med J Aust 1990; 153:274–276. 20. Terrault NA. Sexual activity as a risk factor for hepatitis C. Hepatology 2002; 36:S99–S105. 21. Roberts EA, Yeung L. Maternal transmission of hepatitis C virus infection. Hepatology 2002; 36:S106–S113. 22. Strader DB, Wright TL, Thomas DL, Seeff LB. Diagnosis, management and treatment of hepatitis C. AASLD practice guideline. Hepatology 2004; 39:1147–1171. 23. Seeff LB. Natural history of chronic hepatitis C. Hepatology 2002; 36:S35–S46. 24. Fattovitch G, Giustina G, Degos F, et al. Morbidity and mortality in compensated cirrhosis type C: a retrospective follow-up study of 384 patients. Gastroenterology 1997; 112:463–472. 25. Serfarty L, Aumaître H, Chazouillères O, et al. Determinants of outcome of compensated hepatitis C virus-related cirrhosis. Hepatology 1998; 27:1435–1440. 26. Hu K-Q, Tong MJ. The long-term outcomes of patients with compensated hepatitis C virus related cirrhosis and history of parenteral exposure in the United States. Hepatology 1999; 29:1311–1316. 27. Degos F, Christidis C, Ganne-Carrie N, et al. Hepatitis C virus related cirrhosis: time to occurrence of hepatocellular carcinoma and death. Gut 2000; 47:131–136. 28. Tong M, El-Farra N, Reikes A, Co R. Clinical outcomes after transfusion-associated hepatitis C infection. N Engl J Med 1995; 332:1463–1466. 29. Kenny-Walsh E for the Irish Hepatology Research Group. Clinical outcomes after hepatitis infection from contaminated antiglobulin. N Engl J Med 1999; 340:1228–1233. 30. Wiese M, Berr F, Lafrenz M, et al. Low frequency of cirrhosis in a hepatitis C (genotype 1b) single-source outbreak in Germany: a 20-year multicenter study. Hepatology 2000; 32:91–96. 31. Vogt M, Lang T, Frosner G, et al. Prevalence and clinical outcomes of hepatitis C infection in children who underwent cardiac surgery before the implementation of blood-donor screening. N Engl J Med 1999; 341:866–870. 32. Seeff LB, Miller RN, Rabkin CS, et al. 45-year follow-up of hepatitis C virus infection in healthy young adults. Ann Intern Med 2000; 132:105–111.
683
Section V. Liver Diseases Due to Infectious Agents
33. Desai MM, Zhuang P, Hennessy CH. Surveillance of morbidity and mortality among older adults – United States, 1995–1996. Morb Mortal Wkly Rep 1999; 48:7–25. 34. Hser Y-I, Hoffman V, Grella CE, Anglin MD. A 33-year follow-up of narcotic addicts. Arch Gen Psychiatry 2001; 58:503–508. 35. Poynard T, Ratziu V, Charlotte F, et al. Rates and risk factors of liver fibrosis progression in patients with chronic hepatitis C. J Hepatol 2001; 34:730–739. 36. Poynard T, Bedossa P, Opolon P, for the OBSVIRC, METAVIR, CLINIVIR and DOSVIRC groups. Natural history of liver fibrosis progression in patients with chronic hepatitis C. Lancet 1997; 349:825–832. 37. Wiley TE, Brown J, Chan J. Hepatitits C virus infection in African Americans: its natural history and histological progression. Am J Gastroenterol 2002; 97:700–706. 38. Bjoro K, Froland SS, Yun Z, et al. Hepatitis C infection in patients with primary hypogammaglobulinemia after treatment with contaminated immune globulin. N Engl J Med 1994; 331:1607–1611. 39. Berenguer M, Ferrell L, Watson J, et al. HCV-related fibrosis progression following liver transplantation: increase in recent years. J Hepatol 2000; 32:673–684. 40. Benhamou Y, Bochet M, Di Martino V, et al. Liver fibrosis progression in human immunodeficiency virus and hepatitis C virus coinfected patients. The Multivirc Group. Hepatology 1999; 30:1054–1058. 41. Hui JM, Sud A, Farrell G, et al. Insulin resistance is associated with chronic hepatitis C virus infection and fibrosis progression. Gastroenterology 2003; 125:1695–1704. 42. Powell EE, Edwards-Smith CJ, Hay JL, et al. Host genetic factors influence disease progression in chronic hepatitis C. Hepatology 2000; 31:828–833. 43. Monto A, Alonzo J, Watson J, et al. Steatosis in chronic hepatitis C: relative contributions of obesity, diabetes mellitus and alcohol consumption. Hepatology 2002; 36:729–736. 44. Zarski JP, McHutchison J, Bronowicki JP, et al. Rate of natural disease progression in patients with chronic hepatitis C. J Hepatol 2003; 38:307–314. 45. Monto A, Bostrom A, Pianko S, et al. Risks of a range of alcohol intake on hepatitis C-related fibrosis. Hepatology 2004; 39:826–834. 46. Agnello V, Chung RT, Kaplan LM. A role for hepatitis C virus infection in type II cryoglobulinemia. N Engl J Med 1992; 327:1490–1495. 47. Ferri C, Greco F, Longombardo G, et al. Association between hepatitis C virus and mixed cryoglobulinemia [see comment]. Clin Exp Rheumatol 1991; 9:621–624. 48. Johnson RJ, Gretch DR, Yamabe H, et al. Membranoproliferative glomerulonephritis associated with hepatitis C virus infection. N Engl J Med 1993; 328:465–470. 49. Tran A, Quaranta JF, Benzaken S, et al. High prevalence of thyroid autoantibodies in a prospective series of patients with chronic hepatitis C before interferon therapy. Hepatology 1993; 18:253–257. 50. Fargion S, Piperno A, Cappellini MD, et al. Hepatitis C virus and porphyria cutanea tarda: evidence of a strong association. Hepatology 1992; 16:1322–1326. 51. Stolzel U, Schuppan D, Tillmann HL, et al. Autoantibodies in porphyria cutanea tarda: a controlled study. J Hepatol 2000; 33:858–859. 52. Mason AL, Lau JY, Hoang N, et al. Association of diabetes mellitus and chronic hepatitis C virus infection. Hepatology 1999; 29:328–333. 53. Dalekos GN, Wedemeyer H, Obermayer-Straub P, et al. Epitope mapping of cytochrome P4502D6 autoantigen in patients with chronic hepatitis C during alpha-interferon treatment. J Hepatol 1999; 30:366–375.
684
54. Wiese M, Berr F, Lafrenz M, et al. Low frequency of cirrhosis in a hepatitis C (genotype 1b) single-source outbreak in Germany: a 20-year multicenter study. Hepatology 2000; 32:91–96. 55. Gerlach JT, Diepolder HM, Zachoval R, et al. Acute hepatitis C: high rate of both spontaneous and treatment-induced viral clearance. Gastroenterology 2003; 125:80–88. 56. Crotta S, Stilla A, Wack A, et al. Inhibition of natural killer cells through engagement of CD81 by the major hepatitis C virus envelope protein. J Exp Med 2002; 195:35–41. 57. Herzer K, Falk CS, Encke J, et al. Upregulation of major histocompatibility complex class I on liver cells by hepatitis C virus core protein via p53 and TAP1 impairs natural killer cell cytotoxicity. J Virol 2003; 77:8299–8309. 58. Farci P, Shimoda A, Coiana A, et al. The outcome of acute hepatitis C predicted by the evolution of the viral quasispecies. Science 2000; 288:339–344. 59. Takaki A, Wiese M, Maertens G, et al. Cellular immune responses persist and humoral responses decrease two decades after recovery from a single-source outbreak of hepatitis C. Nat Med 2000; 6:578–582. 60. Bain C, Fatmi A, Zoulim F, et al. Impaired allostimulatory function of dendritic cells in chronic hepatitis C infection. Gastroenterology 2001; 120:512–524. 61. Longman RS, Talal AH, Jacobson IM, et al. Presence of functional dendritic cells in patients chronically infected with hepatitis C virus. Blood 2004; 103:1026–1029. 62. Wiegend J, Potthoff A, Manns MP, et al. Acute hepatitis C infection: can immunology teach us the right way to treat? Curr Hepatitis Rep 2004; 3:148–156. 63. Thimme R, Oldach D, Chang KM, et al. Determinants of viral clearance and persistence during acute hepatitis C virus infection. J Exp Med 2001; 194:1395–1406. 64. Wedemeyer H, He XS, Nascimbeni M, et al. Impaired effector function of hepatitis C virus-specific CD8+ T cells in chronic hepatitis C virus infection. J Immunol 2002; 169:3447–3458. 65. Cramp ME, Rossol S, Chokshi S, et al. Hepatitis C virusspecific T-cell reactivity during interferon and ribavirin treatment in chronic hepatitis C. Gastroenterology 2000; 118, 346–355. 66. Pawlotsky JM. Use and interpretation of virological tests for hepatitis C. Hepatology 2002; 36:S65–S73. 67. Pawlotsky JM. Molecular diagnosis of viral hepatitis. Gastroenterology 2003; 122:1554–1568. 68. Stramer SL, Glynn SA, Kleinman SH, et al. Detection of HIV1 and HCV infections among antibody-negative blood donors by nucleic acid-amplification testing. N Engl J Med 2004; 351:760–768. 69. Everson GT, Trotter J, Forman L, et al. Treatment of advanced hepatitis C with a low accelerating dosage regimen of antiviral therapy. Hepatology 2005; 42:255–262. 70. Ishak K, Baptista A, Bianchi L, et al. Histological grading and staging of chronic hepatitis. J Hepatol 1995; 22:696–699. 71. Batts KP, Ludwig J. Chronic hepatitis. An update on terminology and reporting. Am J Surg Pathol 1995; 19:1409–1417. 72. Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. The METAVIR Cooperative Study Group. Hepatology 1996; 24:289–293. 73. Brunt EM, Janney CG, Di Bisceglie AM, et al. Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastroenterol 1999; 94:2467–2474. 74. Fontana RJ, Lok AS. Noninvasive monitoring of patients with chronic hepatitis C. Hepatology 2002; 36:S57–S64. 75. Hoofnagle JH, Mullen KD, Jones DB, et al. Treatment of chronic non-A, non-B hepatitis with recombinant human alpha interferon. A preliminary report. N Engl J Med 1986; 315:1575–1578.
Chapter 32 HEPATITIS C
76. Davis GL, Balart LA, Schiff ER, et al. Treatment of chronic hepatitis C with recombinant interferon alfa. A multicenter randomized, controlled trial. Hepatitis Interventional Therapy Group. N Engl J Med 1989; 321:1501–1506. 77. McHutchison JG, Gordon SC, Schiff ER, et al. Interferon alfa2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. Hepatitis Interventional Therapy Group. N Engl J Med 1998; 339:1485–1492. 78. Poynard T, Marcellin P, Lee SS, et al. Randomised trial of interferon alpha2b plus ribavirin for 48 weeks or for 24 weeks versus interferon alpha2b plus placebo for 48 weeks for treatment of chronic infection with hepatitis C virus. International Hepatitis Interventional Therapy Group (IHIT). Lancet 1998; 352:1426–1432. 79. Manns MP, McHutchison JG, Gordon SC, et al. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 2001; 358:958–965. 80. Fried MW, Shiffman ML, Reddy KR, et al. Peginterferon alfa2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 2002; 347:975–982. 81. Hadziyannis SJ, Sette H Jr, Morgan TR, et al. Peginterferonalpha2a and ribavirin combination therapy in chronic hepatitis C: a randomized study of treatment duration and ribavirin dose. Ann Intern Med 2004; 140:346–355. 82. Zeuzem S, Hultcrantz R, Bourliere M, et al. Peginterferon alfa2b plus ribavirin for treatment of chronic hepatitis C in previously untreated patients infected with HCV genotypes 2 or 3. J Hepatol 2004; 40:993–999. 83. McHutchison JG, Manns M, Patel K, et al. Adherence to combination therapy enhances sustained response in genotype1-infected patients with chronic hepatitis C. Gastroenterology 2002; 123:1061–1069. 84. Zeuzem S, Diago M, Gane E, et al. Peginterferon alfa-2a (40 kilodaltons) and ribavirin in patients with chronic hepatitis C and normal aminotransferase levels. Gastroenterology 2004; 127:1724–1732. 85. Shiffman ML, Di Bisceglie AM, Lindsay KL, et al. Peginterferon alfa-2a and ribavirin in patients with chronic hepatitis C who have failed prior treatment. Gastroenterology 2004; 126:1015–1023. 86. Poynard T, Schiff ER, Terg R, et al. Sustained virological response (SVR) in the EPIC3 trial: week 12 virology predicts SVR in previous interferon/ribavirin tretament failures receiving PEG-Intron/Rebetol (PR) weight based dosing (WBD). J Hepatol 2005; 42:S40–S41. 87. Curry M, Cardenas A, Afdhal NH. Effect of maintenance PEG-Intron therapy on portal hypertension and its complications: results from the COPILOT study. J Hepatol 2005; 42:S40. 88. Edlin BR. Prevention and treatment of hepatitis C in injection drug users. Hepatology 2002; 36:S210–S219. 89. Nyamathi AM, Dixon EL, Robbins W, et al. Risk factors for hepatitis C virus infection among homeless adults. J Gen Intern Med 2002; 17:13–143. 90. Jonas MM. Children with hepatitis C. Hepatology 2002; 36:S173–S178. 91. Brau N, Bini EJ, Shahidi A, et al. Prevalence of hepatitis C and coinfection with HIV among United States veterans in the New York City metropolitan area. Am J Gastroenterol 2002; 97:2071–2078. 92. Poles MA, Dieterich DT. Hepatitis C virus/human immunodeficiency virus coinfection: clinical management issues. Clin Infect Dis 2000; 31:154–161. 93. Howell C, Jeffers L, Hoofnagle JH. Hepatitis C in African Americans: summary of a workshop. Gastroenterology 2000; 119:1385–1396.
94. Jeffers LJ, Cassidy W, Howell CD, et al. Peginterferon alfa-2a (40 kd) and ribavirin for black American patients with chronic HCV genotype 1. Hepatology 2004; 39:1702–1708. 95. Muir AJ, Bornstein JD, Killenberg PG. Peginterferon alfa-2b and ribavirin for the treatment of chronic hepatitis C in blacks and non-Hispanic whites. N Engl J Med 2004; 350:2265–2271. 96. McHutchison JG, Ponard T, Pianko S, et al. The impact of interferon plus ribavirin on response to therapy in black patients with chronic hepatitis C. Gastroenterology 2000; 119:1317–1323. 97. Martinez-Sierra C, Arizcorreta A, Diaz F, et al. Progression of chronic hepatitis C to liver fibrosis and cirrhosis in patients coinfected with hepatitis C virus and human immunodeficiency virus. Clin Infect Dis 2003; 36:491–498. 98. Sulkowski MS, Thomas DL. Hepatitis C in the HIV-infected person. Ann Intern Med 2003; 138:197–207. 99. Carrat F, Bani-Sadr F, Pol S, et al. Pegylated interferon alfa-2b vs standard interferon alfa-2b, plus ribavirin, for chronic hepatitis C in HIV-infected patients: a randomized controlled trial. JAMA 2004; 292:2839–2848. 100. Torriani FJ, Rodriguez-Torres M, Rockstroh JK, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. N Engl J Med 2004; 351:438–450. 101. Chung RT, Andersen J, Volberding P, et al. Peginterferon alfa-2a plus ribavirin versus interferon alfa-2a plus ribavirin for chronic hepatitis C in HIV-coinfected persons. N Engl J Med 2004; 351:451–459. 102. Tien PS. Management and treatment of hepatitis C virus infection in HIV-infected adults: recommendations from the Veterans Affairs hepatitis C resource center program and the national hepatitis C program office. Am J Gastroenterol 2005; 100:1–17. 103. Graham CS, Baden LR, Yu E, et al. Influence of human immunodeficiency virus infection on the course of hepatitis C virus infection: a meta-analysis. Clin Infect Dis 2001; 33:562–569. 104. Soto B, Sanchez-Ouijano A, Rodrigo L, et al. Human immunodeficiency virus infection modifies the natural history of chronic parenterally-acquired hepatitis C with an unusually rapid progression to cirrhosis. J Hepatol 1997; 26:1–5. 105. Monto A, Dove LM, Bostrom A, et al. Hepatic steatosis in HIV/hepatitis C coinfection: prevalence and significance compared with hepatitis C monoinfection. Hepatology 2005; 42:310–316. 106. Torriani FJ, Rockstroh J, Rodriguez-Torres M, et al. Final results of APRICOT: a randomized, partially blinded, international trial evaluating peginterferon-alpha-2a + ribavirin vs interferonalpha-2a + ribavirin in the treatment of HCV in HIV/HCV coinfection. Paper presented at: 11th Conference on Retroviruses and Opportunistic Infections, 2004; San Francisco, CA. 107. Sulkowski MS, Dieterich DT, Bini EJ, et al. Epoietin alfa once weekly improves anemia in HIV/hepatitis C virus-coinfected patients treated with interferon/ribavirin: a randomized controlled trial. J AIDS 2005; 39:504–506. 107a. Alberti A, Clumeck N, Collins S, et al. Short statement of the first European Consensus Conference on the treatment of chronic hepatitis B and C in HIV co-infected patients. J Hepatol 2005;42:615–624. 108. Shiratori Y, Ito Y, Yokosuka O, et al. Antiviral therapy for cirrhotic hepatitis C: association with reduced hepatocellular carcinoma development and improved survival. Ann Intern Med 2005; 142:105–114. 109. Cammà C, Giunta M, Andreone P, et al. Interferon and prevention of hepatocellular carcinoma in viral cirrhosis: an evidence-based approach. J Hepatol 2001; 34:593–602. 110. Wright TL. Treatment of patients with hepatitis C and cirrhosis. Hepatology 2002; 36:S185–S194.
685
Section V. Liver Diseases Due to Infectious Agents
111. Heathcote EJ, Feinman SV, Rasenack J, et al. Peginterferon alfa-2a in patients with chronic hepatitis C. N Engl J Med 2000; 343:1673–1680. 112. Meyers CM, Seeff LB, Stehman-Breen CO, et al. Hepatitis C and renal disease. Am J Kidney Dis 2003; 42:631–657. 113. Tokars JI, Miller ER, Alter MJ, et al. National surveillance of dialysis-associated diseases in the United States, 1997. Semin Dial 2000; 13:75–85. 114. Nakayama E, Akiba T, Marumo F, et al. Prognosis of antihepatitis C virus antibody-positive patients on regular hemodialysis therapy. J Am Soc Nephrol 2000; 11: 1896–1902. 115. Fabrizi F, Poordad FF, Martin P. Hepatitis C infection and the patient with end-stage renal disease. Hepatology 2002; 36:3–10. 116. Russo MW, Goldsweig CD, Jacobson IM, et al. Interferon monotherapy for dialysis patients with chronic hepatitis C: an analysis of the literature on efficacy and safety. Am J Gastroenterol 2003; 98:1610–1615. 117. Russo MG, Sigal R, Joshi SH, et al. A multi-center randomized trial of pegylated interferon alfa-2B monotherapy (peg-intron) in patients with chronic hepatitis C and end stage kidney disease on dialysis. Hepatology 2004; 40:399A. 118. Bruchfeld A, Lindahl K, Soderberg M, et al. Interferon and ribavirin treatment in patients with hepatitis C-associated renal disease and renal insufficiency. Nephrol Dial Transplant 2003; 18:1573–1580. 119. Narayan Menon KV, Poterucha JJ, El-Amin OM, et al. Treatment of post-transplantation recurrence of hepatitis C with interferon and ribavirin; lessons on tolerability and efficacy. Liver Transplant 2002; 8:623–629. 120. Sanchez-Tapias JM, Diago M, Escartin P, et al. Longer treatment duration with peginterferon alfa-2A (40kd) (Pegasys®) and ribavirin (Copegus®) in naive patients with chronic hepatitis C and detectable HCV RNA by week 4 of therapy: final results of the randomized, multicenter TeraViC-4 study. Hepatology 2004; 40:218A. 121. Berg T, von Wagner M, Hinrichsen H, et al. (2003). Comparison of 48 or 72 weeks of treatment with peginterferon alfa-2a (40KD) (Pegasys®) plus ribavirin (Copegus®) in treatment-naive patients with chronic hepatitis C infected with HCV genotype 1. Hepatology 2003; 38:317A. 122. von Wagner M, Huber M, Berg T, et al. Peginterferon alfa-2a (40KD) and ribavirin for 16 or 24 weeks in patients with genotype 2 or 3 chronic hepatitis C. Gastroenterology 2005; 129:522–527. 123. Dalgard O, Bjoro K, Hellum KB, et al. Treatment with pegylated interferon and ribavirin in HCV infection with genotype 2 or 3 for 14 weeks: a pilot study. Hepatology 2004; 40:1260–1265. 124. Mangia A, Santoro R, Minerva N, et al. Peginterferon alfa-2b and ribavirin for 12 vs. 24 weeks in HCV genotype 2 or 3. N Engl J Med 2005; 352:2609–2617. 125. Bain V, Kaita K, Yoshida E, et al. A phase 2 study to assess antiviral response, safety, and pharmacokinetics of albuferon in
686
126.
127.
128.
129.
130.
131.
132.
133.
134.
135. 136.
137.
138.
139.
IFN-alfa naive subjects with genotype 1 chronic hepatitis C. J Hepatol 2005; 42:9. Afdhal NH, Dieterich DT, Pockros PJ, et al. Epoietin alfa maintains ribavirin dose in HCV-infected patients: a prospective, double-blind, randomized controlled study. Gastroenterology 2004; 126:1302–1311. Gish RG, Nelson D, Arora S, et al. Virological response and safety outcomes in therapy-naive patients treated for chronic hepatitis C with viramidine in combination with pegylated interferon alfa-2a. J Hepatol 2005; 42:39. Lohmann V, Korner F, Koch J, et al. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 1999; 285:110–113. Lamarre D, Anderson PC, Bailey M, et al. An NS3 protease inhibitor with antiviral effects in humans infected with hepatitis C virus. Nature 2003; 426:186–189. Hinrichsen H, Benhamou Y, Wedemeyer H, et al. Short-term antiviral efficacy of BILN 2061, a hepatitis C virus serine protease inhibitor, in hepatitis C genotype 1 patients. Gastroenterology 2004; 127:1347–1355. Reiser M, Hinrichsen H, Benhamou Y, et al. Antiviral efficacy of NS3-serine protease inhibitor BILN-2061 in patients with chronic genotype 2 and 3 hepatitis C. Hepatology 2005; 41:832–835. Reesink HW, Zeuzem S, van Vliet A, et al. Initial results of a phase 1B, multiple-dose study of VX-950, a hepatitis C virus protease inhibitor. Gastroenterology 2005; 128 (suppl.2):A697. Afdhal N, Rodrriguez-Torres M, Lawitz E, et al. Enhanced antiviral efficacy for valopicitabine (NM283) plus PEGinterferon in hepatitis C patients with HCV genotype-1 infection: results of a phase IIa multicenter trial. J Hepatol 2005; 42:39–40. Afdhal N, Godofsky E, Dienstag J, et al. Final phase I/II trial results for NM283, a new polymerase inhibitor for hepatitis C: antiviral efficacy and tolerance in patients with HCV-1 infection, including previous interferon failures. Hepatology 2004; 40:726A. Krueger M, Beger C, Wong-Staal F. Use of ribozymes to inhibit gene expression. Methods Enzymol 1999; 306:207–225. Gordon SC, Bacon BR, Jacobson IM, et al. Treatment of chronic hepatitis C with ISIS 14803, an antisense inhibitor of HCV, given for 12 weeks. Hepatology 2003; 38:306A–307A. Seo MY, Abrignani S, Houghton M, et al. Small interfering RNA-mediated inhibition of hepatitis C virus replication in the human hepatoma cell line Huh-7. J Virol 2003; 77:810–812. Nevens F, Roskams T, Van Vlierberghe H, et al. A pilot study of therapeutic vaccination with envelope protein E1 in 35 patients with chronic hepatitis C. Hepatology 2003; 38:1289–1296. Manns MP, Berg T, Wedemeyer H, et al. Immunization with the therapeutic hepatitis C virus (HCV) peptide vaccine IC41 in 66 chronic hepatitis C non-responder patients. Hepatology 2004; 40:251A.
Section V: Liver Diseases Due to Infectious Agents
33
HEPATITIS D Mario Rizzetto Abbreviations ALT alanine aminotransferase anti-HD antibody to the HD Ag HBIg hyperimmune serum against HBsAg HBsAg hepatitis B surface antigen HBV hepatitis B virus HDV hepatitis delta virus
IFN IgM IgM antiHD l-HD Ag LKM
interferon immunoglobulin M antibody to the HD Ag large HD Ag liver–kidney microsomal
INTRODUCTION The hepatitis delta virus (HDV) is a unique transmissible human pathogen with an RNA genome which shares similarities with RNA pathogens of higher plants. The HDV is defective and requires the hepatitis B virus (HBV) for transmission; it therefore can only infect individuals with simultaneous HBV infection. It causes both acute and chronic hepatitis D; these are distinct medical entities usually more severe than the disease caused by HBV alone. The pathobiology of hepatitis D is complex, requiring an appreciation of the virologic and pathogenic interactions of two distinct agents within a common host.1
VIROLOGY The HDV is a defective RNA pathogen dependent for infection on obligatory helper functions provided by the HBV; it is the only member of the genus Deltavirus.2 Other hepadnaviruses can support HDV and its infection was experimentally established in the eastern woodchuck carrying the woodchuck HBV. The hepadnavirus provides the hepatitis B surface antigen (HBsAg) coat that is necessary in HDV for binding to the hepatocytes and for virion assembly.3 In vitro, intracellular replication of HDV has been established in mice by the intrahepatic injection of naked viral HDV DNA or RNA sequences4 and by the engraftment of human hepatocytes into heterochimeric severe combined immunodeficiency (SCID) mice.5 The virion is an approximately 36 nm particle containing within an HBsAg coat a 1.7 kilobase single-stranded circular RNA genome and two related structural phosphoproteins sharing a common antigenic reactivity (the hepatitis delta antigen = HD Ag).6 The genome contains only about 1700 nucleotides and is therefore the smallest among animal viruses, resembling viroid RNAs of plants. In analogy with viroids and components of eukaryotic cells, HDV RNA can act as a ribozyme, i.e., it encloses a segment that can self-cleave and self-ligate the genome. The HDV replicates by a rolling-circle mechanism similar to that involved in the replication of viroids; using the circular genomic RNA as a template, linear oligomeric forms of antigenomic polarity
Mu PCR SCID s-HD Ag THF g-2
million units polymerase chain reaction severe combined immunodeficiency small HD Ag thymic humoral factor g-2
are generated, which are then cleaved and ligated to circular monomers by the autocatalytic ribozyme. Replication occurs through RNA-directed RNA synthesis performed by a host RNA polymerase II which is normally DNA-dependent but is redirected by HDV to transcribe its RNA genome. The two forms of HD Ag produced by HDV, the small HD Ag (s-HD Ag) and the large HD Ag (l-HD Ag), differ in their C-terminal 19 amino acids. Their synthesis is regulated through the editing of the antigenomic RNA at position 1012 by cellular double-stranded RNA adenosine deaminase;7 the editing process changes a UAG stop codon to a UGG tryptophan codon that allows transcription to proceed for 19 further amino acid residues, leading to the synthesis of l-HD Ag. The sHDAg is required for viral replication; l-HD Ag inhibits replication of HDV and is required for its assembly.8 Replication of the antigenomic RNA strand requires multiple post-translational modifications, including the phosphorylation and methylation of HDAg.9 In vitro virion assembly is critically dependent on prenyl lipid modification (prenylation) of l-HD Ag.8,10 The HDV has evolved in at least three major genotypes, designated I, II, III, that differ by as much as 40% over the entire genome and have different geographic prevalences; two subgroups have been identified within genotype I and II, named genotype I A and I B and genotype II A and II B.11,12
TRANSMISSION The HDV is transmitted by the same routes as the helper HBV, i.e., by the parenteral route, either overtly or covertly. The highest rates of hepatitis D were reported in parenteral drug addicts; in contrast, vertical transmission from mother to newborn is rare.13 Transmission can occur by sexual contacts, in particular mercenary heterosexual contact, as attested by the prevalence of HDV in prostitutes and sexual partners of HDV-infected persons; however the spread of HDV has not been evident among homosexual man. Household transmission was important in 1970–1980 in endemic areas in southern Europe and cohabitation with an HDV carrier was identified as a major risk factor for the acquisition of the virus;14 molecular studies have confirmed sexual and household spread of HDV.15
687
Section V. Liver Diseases Due to Infectious Agents
The efficiency of HDV transmission is primarily determined by whether the person exposed to HDV is or is not a carrier of HBsAg. In normal (HBsHg-negative) persons, HDV cannot be transmitted unless HBV infection has previously been established; therefore HDV is acquired simultaneously with HBV, i.e., by HBV–HDV coinfection (Figure 33-1A), and transmission efficiency depends on the infectious titer of co-infecting HBV. In the HBsAg carrier the pre-existing HBV state acts as a selective magnet to activate HDV and this infection is therefore rapidly established and amplified – i.e. there is HDV superinfection on prior HBV infection (Figure 33-1B). In the chimpanzee, titers of HDV as low as 10–11 serum dilutions were sufficient to establish infection in animals carrying the HBsAg. As the HBsAg state provides the HDV with indefinite biological support, superinfected HBsAg carriers often also become chronic carriers of HDV. Therefore HBsAg carriers are the selective victims of HDV, and the main epidemiologic reservoir and source of the virus.
EPIDEMIOLOGY Hepatitis D was endemic in the 1980s in many areas of the world but its distribution and ratio relative to the local prevalence of HBV varied widely. Overall prevalence rates were higher in tropical and
ALT
IgM anti-HD
IgM anti-HBc
IgG anti-HD
HDV HBV
HDV HBV Months A
Months
Weeks Course of coinfection with Hepatitis B virus and Hepatitis D virus ALT
IgM anti-HD
IgG anti-HD
HDV HBV
HDV HBV B
Course of Hepatitis D viral superinfection
Figure 33-1. (A) Course of co-infection with hepatitis B virus (HBV) and hepatitis D virus (HDV). (B) Course of hepatitis D viral superinfection. ALT, alanine aminotransferase; IgM, immunoglobulin M.
688
subtropical areas with a higher prevalence of HBV than in north America and northern Europe, where the prevalence of HBV was low. In developing areas hepatitis D occurred as a secondary event in the context of high HBV endemicity; in settings of overcrowding and poor hygienic condition HDV infection mostly occurred in children. In most areas of the developed world, where the prevalence of HBV was low, HDV infection was confined to intravenous drug addicts.16 The pattern of infection was composite in areas with intermediate HBV prevalence such as southern Europe and Taiwan, resulting from an endemic pattern in the general population and an epidemic pattern within drug addicts.17 In Italy virus spread occurred primarily in overcrowded south Italian families and the age of acquisition of infection was adolescence–early adulthood; at-risk groups were also institutionalized patients, prisoners, hemophiliacs, and hemodialysis patients. Genetic analysis of HDV worldwide has shown significant geographic variations.18,19 Genotype I predominates in the USA, Europe,20 Turkey,21 the Middle East, and Mongolia;22 genotype I A predominates in Asia, I B in the USA, and both are common in the Mediterranean basin. Genotype II was found in Japan, Taiwan, and in Yakutia in Russia.23 Genotype III has been identified in northern South America.24 In the late 1970s to early 1980s the prevalence of HDV infection was relatively high in southern Europe but important epidemiologic changes have occurred in this area in the last decade. Changes were best documented in Italy. In a baseline study performed in the years 1978–1981 the prevalence of HDV infection, as determined from the prevalence of the antibody to the HD Ag (anti-HD) in HBsAg carriers with liver disease, was 24%. The prevalence remained stable up to the second half of the 1980s, but by the early 1990s it had diminished to 14% and in a nationwide survey performed in the late 1990s it has further declined to 8%.25 A decline in HDV infection was noted in Spain,26 where the rate of anti-HD in HBsAg carriers with chronic hepatitis dropped from 15% in 1975–1985 to 7.1% in 1986–1992 and a significant decrease in the prevalence of HDV superinfection as the cause of hepatitis flares in HBsAg carriers was also noted in Taiwan.27 The decrease of HDV infection throughout southern Europe is related to the control of HBV achieved in recent years through better public health standards, universal HBV vaccination and the measures introduced to contain the spread of HIV, which is transmitted in the same way as HBV/HDV; in Italy the reduction in family size brought about by social and economic changes had a major additional impact in containing hepatitis D. Sparse information was made available in the last years from other countries.28 A decrease from 47.6 to 15.4% in the prevalence of antiHD in patients with HBsAg-positive cirrhosis was noted in Belarus from 1991 to 1997. No marker of HDV infection was found in Indian children in the reservation of Xinger in central Brazil, a region adjacent to the Amazon jungle where the highest level of HDV endemicity was reported in the 1980s and where outbreaks of hepatitis D have continued to occur (in the Peruvian sector). Hyperendemic foci were reported from west Greenland, Tunisia, and Japan and markers of HDV were found in a consistent proportion
Chapter 33 HEPATITIS D
(18.35%) of HBsAg-positive pregnant women in Moldova as well as in 13.15% of hepatitis B patients in the Shandong province in China.28 In contrast, hepatitis D is rare in Poland29 and no infection marker was detected in rural areas of the high Andean plateau and in Nigeria; the rate of infection was also low, 6.5 and 4% respectively, in chronic HBsAg-positive liver disease in sub-Saharan Africa and in HBsAg carriers in north-west Mexico. In India HDV was implicated as a major risk factor of fulminant hepatitis but was rarely found in a series of hospital patients and mixed clinical HBsAgpositive populations in this country.28 Only a few studies comparing the prevalence of HDV infection over the years in drug addicts are available. In Taiwan, the prevalence of HDV in HBsAg-positive drug addicts has declined from 79% in 1985 to 14% in 2004;30 in this population the average rate of decrease in the prevalence of HDV infection was 4.7% per year.31 No HDV case was found in intravenous drug users in Rio de Janeiro, Brazil, despite a 7.8% prevalence of serum HBsAg in this community; low prevalence rates were found in a community in Gran Canaria, and a 14.7% prevalence rate was found in drug users in Jeddah, Saudi Arabia.28 It is unlikely that the pattern of HDV infection has changed in many endemic areas of the developing world, as the factors determining transmission and circulation of the virus were not modified in the last decade whereas reduction of HBV infection and the consequent shrinkage of the HBsAg pool within drug abusers has diminished the impact of HDV infection in these communities.
DIAGNOSIS Detection of indirect antibody markers is the first step in diagnosing HDV infection.13,19 These are the immunoglobulin M (IgM) antibody to the HD Ag (IgM anti-HD) and total antibody to the HD Ag (anti-HD) which predominantly detects the IgG antibody. The IgM antibody is also composed of monomeric 7S IgM molecules. The antibodies to HDV are not protective; anti-HD may persist at low titer as a serological scar to past HDV infection in both HBsAgpositive and negative individuals. Detection of intrahepatic HD Ag by direct immunohistochemistry was initially the gold standard for the diagnosis of HDV infection, as its finding was diagnostic of viral replication and ongoing infection.13 However, intrahepatic HD Ag is not detectable in as many as 50% of patients with current HDV infection and its expression decreases with evolution to advanced fibrotic disease.32 The detection of HDV RNA in serum by the polymerase chain reaction (PCR) is presently the most specific and sensitive diagnostic method.33 This assay has overcome the limitations of the detection of HD Ag in serum by enzyme or radioimmunoassay, due to antigen sequestration in immune complexes with circulating antibodies; it can detect 10–100 copies of the viral genome. The highest level of diagnostic efficacy is obtained by amplification of the most conserved region, the C-terminal half of the HD Ag gene. The HDV genotype may be determined in serum by restriction fragment length polymorphism analysis of PCR amplification products or by sequencing;20,33 HDV genotypes can also be determined in liver tissue by immunohistochemistry.34
NATURAL HISTORY Hepatitis D comes about through co-infection with the HBV or through superinfection of a HBsAg carrier (Figure 33-1). Individuals protected from HBV by antibodies to the HBsAg (anti-HBs) in serum are also protected from HDV.13 In normal individuals who become co-infected with HBV and HDV, the activation of HDV is dependent on the prior activation of HBV. Expression of HDV may vary from massive to abortive; accordingly, co-infection with hepatitis D presents clinically as an acute hepatitis of variable severity. As the underlying HBV infection is usually self-limiting and the defective HDV cannot survive the clearance of its helper virus, co-infection with hepatitis D usually resolves; fewer than 2% of cases were reported to progress to chronicity.19 In contrast, since the HBV state will indefinitely support HDV, superinfected HBsAg carriers run a high risk of developing chronic hepatitis D; nevertheless, HDV superinfection can occasionally terminate the carriage of HBsAg with clearance of both HBV and HDV, and the rate of HBsAg clearance is increased over the years in chronic HDV patients compared to ordinary HBsAg carriers.35 Triple HBV–HDV–HCV infection can be observed in drug addicts. In studies in Spain36 and France,37 HDV acted as the dominant virus and cirrhosis was more frequent in patients with multiple infections; in a study of triple infections in Taiwan, HCV acted as the dominant virus.38 Superinfection presents clinically as an icteric hepatitis which often runs a severe course, and HDV superinfection was recognized worldwide as an important factor leading to fulminant liver disease.39 Surveys in the Mediterranean area in the 1980s have shown that chronic hepatitis D ran an indolent non-progressive or slowly progressive course in about 10–15% of cases while in the remaining patients the disease was progressive, advancing to cirrhosis within 2–6 years in 40% of cases. Infection with HDV increases the risk for hepatocellular carcinoma threefold and for mortality twofold in patients with HBsAg-positive cirrhosis;40 in cirrhotic patients impressive splenomegalies may develop. Because of the young age at HDV acquisition and the accelerated course of the disease, chronic hepatitis D was recognized as a major cause of juvenile cirrhosis in endemic areas. Evolution to cirrhosis occurs in three phases: (1) an early one characterized by florid disease and infection with significantly elevated alanine aminotransferase (ALT), active HDV replication, and suppressed HBV; (2) a second phase characterized by diminished ALT and HDV synthesis and occasionally by reactivation of HBV; and (3) a late cirrhotic phase characterized by marked reduction of the synthesis of both HBV and HDV.32 In children the course of the disease is similar to that seen in adults. In co-infections the serological pattern includes the markers of acute HBV infection (the IgM antibody to HB core antigen and HBV DNA) with superimposed makers of HDV. Hepatitis D antigenemia occurs early, often is not detectable and, when initially present, becomes rapidly undetectable for sequestration of the serum HD Ag by the homologous antibody. The viremic phase is best detected by HDV-RNA PCR assaying. The IgM antibody to HD rises days to
689
Section V. Liver Diseases Due to Infectious Agents
a few weeks after onset of disease while total (IgG) anti-HD rises a few weeks after. In superinfected carriers, both the IgM and IgG anti-HD rise rapidly to high titers and persist as hepatitis D becomes chronic. The HDV RNA is usually detectable at onset of disease and persists, while the IgM antibody to the HB core antigen is usually negative except if the carrier of HBsAg had chronic liver disease related to active HBV infection. In patients whose acute or chronic disease resolves spontaneously or following interferon (IFN) therapy, the IgM antibody disappears in a few months; in most of these patients IgG anti-HD also disappears but occasionally it may persist at low titers.
CLINICAL COURSE Acute and chronic hepatitis D do not exhibit clinical or histologic features that are specific or distinctive from ordinary hepatitis B.16,18,19 Acute co-infection with hepatitis D may run a biphasic course with two peaks of ALT several weeks apart and expression of HDV in the second peak. In asymptomatic carriers of HBsAg who are unaware of their state, superinfection with hepatitis D may appear as acute hepatitis B; however, in contrast to acute hepatitis B, the IgM antibody to the HB core antigen is usually negative. In carriers with prior chronic hepatitis B, superinfection with hepatitis D may be mistaken for a recrudescence of the underlying HBV infection or it may decompensate a pre-existing stable HBV disease. The concomitant HBV infection is most often inactive (HBV DNA negative by conventional hybridization assays or positive at low titer by PCR; HBe Ag negative and anti-HBe positive in serum). A proportion of patients, usually with early and florid disease, generate a variety of autoantibodies. The most frequent is a liver–kidney microsomal (LKM) antibody which displays a pattern of tissue immunofluorescence similar to that displayed by the LKM1 antibody of autoimmune type 2 hepatitis; it is directed against uridine diphosphate glucuronyltransferase 1 instead of the cytochrome superfamily and is therefore distinguished as LKM3.41 The pathogenicity of HDV appears to be modulated by the degree of replication of the helper HBV; disease is usually most severe in patients whose HDV and HBV infections are both active.42 Variations in pathogenicity may be related to the genotype of HDV; genotype III was associated with outbreaks of fulminant hepatitis in South America24 and Japanese patients with genotype II B showed greater progression of chronic hepatitis to cirrhosis than those with genotype II B.12 The striking changes of the last 15 years in the epidemiology of HBV/HDV in the Mediterranean basin have modified the clinical impact and features of hepatitis D in this area. While most of the HDV patients collected in Italy in the 1980s had a florid chronic active hepatitis, and inactive cirrhosis with residual burnt-out inflammation was seen in fewer than 20% of cases, the proportion of cirrhotic patients has increased in Italy to 70% in the years 1996–2000; these patients represent the survivors to the epidemic of hepatitis D that ravaged HBsAg carriers in the 1970s–1980s.43 Thus the contemporary medical scenario of hepatitis D in areas such as Italy where the circulation of HDV has much diminished in
690
recent years is predominantly made up of advanced fibrotic disease unresponsive to IFN, which is unlikely to achieve benefit from any medical treatment, and for which only liver transplantation offers adequate therapy.
PREVENTION AND THERAPY Successful vaccination against HBV provides protection against HDV. No effective HDV vaccine has been developed to protect the HBsAg carrier. In clinical trials evaluating the long-term administration of IFN-a in chronic hepatitis D, the rate of response (normalization of the serum ALT level and clearance of serum HDV RNA) has varied, occurring at different times from the beginning of treatment.44 It was proportional to the dose of IFN. In a study45, patients given 9 million units (Mu) thrice weekly responded better than patients given lower dosages and, at the end of treatment, the normalization of ALT in responders corresponded to histologic improvement and clearance of HDV RNA from blood; however, most experienced a biochemical and virological relapse shortly after stopping therapy and in the five patients who exhibited normal ALT for up to 4 years, the effect on HDV was not sustained. Nevertheless, after 14 years of followup complete fibrosis regression has been reported in some of the patients who had a persistent biochemical response and lost IgM anti-HD, all of whom had an initial diagnosis of active cirrhosis at the end of therapy.46 A virological and biochemical relapse was also common when IFN was reduced to 3 MU/m2 after a 4-month course with 5 MU/m2.44 Sustained responses are often followed by loss of HBsAg and seroconversion to anti-HBs47 and resolution of chronic hepatitis D has occurred up to 12 years after continuous IFN therapy.48 Responses have occurred in immunodeficiency virus-positive drug abusers with HDV in whom immunologic competence was preserved.49 In HDV patients with concomitant HBV infection, therapy may abrogate HBV replication. Several such patients, both adult and children, cleared HBV DNA and hepatitis B e antigen from serum and seroconverted to anti-HBe during interferon therapy; however the control of HBV infection had no impact on HDV infection or on disease activity. The severity of side effects is proportional to the IFN dose and to the age of the patient. Four patients were reported to experience a severe exacerbation of hepatitis during therapy; one of them had LKM antibodies. Suramin, aciclovir in combination with IFN, ribavirin, and Thymic Humoral Factor g-2 (THF g-2) a synthetic thymus-derived octapeptide, have not produced beneficial therapeutic effects. As HBsAg is required by HDV to perpetuate infection, repression of the helper HBV and of HBsAg production might prevent the spread of virus and thus help to eradicate HDV infection. On this rationale famciclovir and lamivudine, which are inactive on HDV replication but capable of inhibiting HBV, have been tested in hepatitis D. However, results were discouraging. None of 15 patients with chronic hepatitis D treated with famciclovir 500 mg three times a day for 6 months and then followed up for 6 months responded.50
Chapter 33 HEPATITIS D
In a study five patients received lamivudine 100 mg orally daily for 12 months.51 Though HBV DNA became undetectable in four patients and decreased in the other, ALT remained abnormal, HDV RNA detectable, and HBsAg positive in all. Liver biopsy showed no improvement in inflammatory or fibrotic score. A second study52 compared a 12-month with a 24-month course of lamivudine 100 mg/day. Thirty-one patients were randomized to treatment, 11 to placebo, and 20 to lamivudine for 12 months; thereafter all were given lamivudine on an open-label basis for 12 months and followed up for further 16 weeks. At the end of treatment HDV RNA was negative and ALT normal in three patients but only two patients remained virus-free at the end of follow-up. The effects of the combination of IFN with lamivudine were also investigated.53 Eight patients with chronic hepatitis D were treated with lamivudine for at least 24 weeks. Lamivudine was then combined with high-dose IFN followed by 9 MU IFN thrice weekly; the patients were followed up for 12 weeks post-therapy. The HBsAg concentration in serum decreased in two patients. There was no significant reduction of HDV RNA in plasma from baseline during treatment. At the end of treatment ALT normalized in one patient and decreased in three other patients, but three of these four patients showed a biochemical rebound after withdrawal of therapy. Available data indicate that IFN is the only potential – albeit limited – therapy for chronic hepatitis D. To achieve a response, high doses of conventional IFN are required (9–10 MU thrice weekly for 1 year or longer); although there is no information yet on pegylated IFNs, they appear nevertheless to represent a logical therapeutic option for the long-term treatment required for chronic hepatitis D. Parameters predictive of response are still unidentified. As response can take up to 10 months, IFN should be given for at least a year before a patient is regarded as a non-responder. Treated patients can lose HDV markers from serum while on therapy but may relapse when treatment is discontinued if they remain HBsAg-positive; therefore therapy should not be discontinued prematurely on the basis of the clearance of HDV. Loss of HBsAg is a reliable marker of resolution of hepatitis D. Preclinical data in a mouse-based model of HDV infection capable of yelding viremia support a potential role of prenylation inhibitors at clearing viremia through interference within the cxxx box within the 19 amino acids unique to l-HD Ag, which is the substrate for the prenyltransferases required to prenylate the antigen.54
TRANSPLANTATION Liver transplantation provides a valid treatment option for end-stage HDV liver disease. The spontaneous risk of reinfection is lower for HDV than for HBV and the clinical course of recurrent hepatitis is milder than for HBV. The prospect of an uneventful clinical course after transplantation was significantly improved by the long-term administration of hyperimmune serum against HBsAg (HBIg); the 5-year survival rate of 76 transplant recipients in Paris for terminal HDV cirrhosis given HBIg long-term was 88%, with reappearance of HBsAg in only 9%.55 None of 62 HDV patients transplanted in the last years in Turin experienced a viral recurrence; 48 were given HBIg and 14 were given lamivudine before transplantation and lamivudine together with HBIg indefinitely post-transplantation.56
REFERENCES 1. Rizzetto M, Canese MG, Arico S, et al. Immunofluorescence detection of new antigen–antibody system (delta/anti-delta) associated to hepatitis B virus in liver and in serum of HBsAg carriers. Gut 1977; 18:997–1003. 2. Murphy FA. Virus taxonomy. In: Fields BN, Knipe DM, Howley PM, eds. Fields virology. 3rd edn. New York: Raven Press; 1996. 3. Ponzetto A, Cote PJ, Popper H, et al. Transmission of the hepatitis B virus-associated delta agent to the eastern woodchuck. Proc Natl Acad Sci USA 1984; 81:2208–2212. 4. Chang J, Sigal LJ, Lerro A, Taylor J. Replication of the human hepatitis delta virus genome is initiated in mouse hepatocytes following intravenous injection of naked DNA or RNA sequences. J Virol 2001; 75:3469–3473. 5. Ohashi K, Marion PL, Nakai H, et al. Sustained survival of human hepatocytes in mice: a model for in vivo infection with human hepatitis B and hepatitis delta viruses. Nat Med 2000; 6:327–331. 6. Taylor JM. Replication of human hepatitis delta virus recent developments. Trends Microbiol 2003; 11:185–190. 7. Polson AG, Bass BL, Casey JL. RNA editing of hepatitis delta virus antigenome by dsRNA-adenosine deaminase. Nature 1996; 380:454–456. 8. Lai MM. The molecular biology of hepatitis delta virus (review). Annu Rev Biochem 1995; 64:259–286. 9. Li YJ, Stallcup MR, Lai MM. Hepatitis delta virus antigen is methylated at arginine residues, and methylation regulates subcellular localization and RNA replication. J Virol 2004; 78:13325–13334. 10. Yamaguchi Y, Filipovska J, Yano K, et al. Stimulation of RNA polymerase II elongation by hepatitis delta antigen. Science 2001; 293:124–127. 11. Casey JL, Polson AG, Bass BL, Gerin JL. Molecular biology of HDV: analysis of RNA editing and genotype variations. In: Rizzetto M, Purcell RH, Gerin JL, Verme G, eds. Viral hepatitis and liver disease. Turin: Minerva Medica; 1997:290–294. 12. Watanabe H, Nagayama K, Enomoto N, et al. Chronic hepatitis delta virus infection with genotype IIb variant is correlated with progressive liver disease. J Gen Virol 2003; 84:3275–3289. 13. Smedile A, Rizzetto M, Gerin JL. Advances in hepatitis D virus biology and disease. In: Boyer JL, Okner RK, eds. Progress in liver disease, vol. 12. Philadelphia, PA: WB Saunders; 1994:157–175. 14. Sagnelli E, Stroffolini T, Ascione A, et al. The epidemiology of hepatitis delta infection in Italy. Hepatology 1992; 15:211–215. 15. Niro GA, Casey JL, Gravinese E, et al. Intrafamilial transmission of hepatitis delta virus: molecular evidence. J Hepatol 1999; 30:564–569. 16. Rizzetto M, Ponzetto A, Bonino F, Smedile A. Hepatitis delta virus infection: clinical and epidemiological aspects. In: Zuckerman AJ, ed. Viral hepatitis and liver disease. New York: Alan R. Liss; 1988:389–394. 17. Rizzetto M, Hadziyannis S, Hansson BG, et al. Hepatitis delta virus infection in the world, epidemiological patterns and clinical expression. Gastroenterol Int 1992; 5:18–32. 18. Farci P. Delta hepatitis: an update. J Hepatol 2003; 39:S212–S219. 19. Smedile A, Ciancio A, Rizzetto M. Hepatitis D, hepatitis D virus. In: Richman DD, Whitley RJ, Hayden FG, eds. Clinical virology, Washington DC: ASM Press; 2002:1227–1240. 20. Niro GA, Smedile A, Andriulli A, et al. The predominance of hepatitis delta virus genotype I among chronically infected Italian patients. Hepatology 1997; 25:728–734. 21. Bozdayi AM, Aslan N, Bozdayi G, et al. Molecular epidemiology of hepatitis B, C and D viruses in Turkish patients. Arch Virol 2004; 149:2115–2129.
691
Section V. Liver Diseases Due to Infectious Agents
22. Takahashi M, Nishizawa T, Gotanda Y, et al. High prevalence of antibodies to hepatitis A and E viruses and viremia of hepatitis B, C, and D viruses among apparently healthy populations in Mongolia. Clin Diagn Lab Immunol 2004; 11:392–398. 23. Ivaniushina V, Radjef N, Alexeeva M, et al. Hepatitis delta virus genotypes I and II cocirculate in an endemic area of Yakutia, Russia. J Gen Virol 2001; 82:2709–2718. 24. Casey JL, Brown TL, Colan EJ, et al. A genotype of hepatitis D virus that occurs in northern South America. Proc Natl Acad Sci USA 1993; 90:9016–9020. 25. Gaeta GB, Stroffolini T, Chiaramonte M, et al. Chronic hepatitis D: a vanishing disease? An Italian multicenter study. Hepatology 2000; 32:824–827. 26. Navascues CA, Rodriguez M, Sotorrio NG, et al. Epidemiology of hepatitis D virus infection: changes in the last 14 years. Am J Gastroenterol 1995; 90:1981–1984. 27. Huo TI, Wu JC, Lin RY, et al. Decreasing hepatitis D virus infection in Taiwan: an analysis of contributory factors. Gastroenterol Hepatol 1997; 12:745–746. 28. Ciancio A, Rizzetto M. Clinical patterns, epidemiology and disease burden of hepatitis D virus chronic liver disease. Margolis H, Alter M, Liang T, et al, eds. 10th International Symposium on Viral Hepatitis and Liver Disease. London: International Medical Press; 2002:271–275. 29. Chlabicz S, Grzeszczuk A, Lapinski TW, et al. Search for hepatitis delta virus (HDV) infection in hepatitis C patients in north-eastern Poland. Comparison with anti-HDV prevalence in chronic hepatitis B. Eur J Epidemiol 2003; 18:559–561. 30. Huo TI, Wu JC, Wu SI, et al. Changing seroepidemiology of hepatitis B, C, and D virus infections in high-risk populations. J Med Virol 2004; 72:41–45. 31. Kao JH, Chen PJ, Lai MY, Chen DS. Hepatitis D virus genotypes in intravenous drug users in Taiwan: decreasing prevalence and lack of correlation with hepatitis B virus genotypes. J Clin Microbiol 2002; 40:3047–3049. 32. Wu JC, Chen TZ, Huang YS, et al. Natural history of hepatitis D viral superinfection: significance of viremia detected by polymerase chain reaction. Gastroenterology 1995; 108:796–802. 33. Smedile A, Niro MG, Rizzetto M. Detection of serum HDVRNA by RT-PCR. In: Hamatake RK, Lay JYN, eds. Methods in molecular medicine; hepatitis B and D protocols. Totowa, NJ: Humana Press; 2005:85–94. 34. Hsu SC, Syu WJ, Ting LT, Wu JC. Immunohistochemical differentiation of hepatitis D virus genotypes. Hepatology 2000; 32:1111–1116. 35. Niro GA, Gravinese E, Martini E, et al. Clearance of hepatitis B surface antigen in chronic carriers of hepatitis delta antibodies. Liver 2001; 21:254–259. 36. Jardi R, Rodriguez F, Buti M, et al. Role of hepatitis B, C, and D viruses in dual and triple infection: influence of viral genotypes and hepatitis B precore and basal core promoter mutations on viral replicative interference. Hepatology 2001; 34:404–410. 37. Mathurin P, Thibault V, Kadidja K, et al. Replication status and histological features of patients with triple (B, C, D) and dual (B, C) hepatic infections. J Viral Hepatol 2000; 7:15–22.
692
38. Lu SN, Chen TM, Lee CM, et al. Molecular epidemiological and clinical aspects of hepatitis D virus in a unique triple hepatitis viruses (B, C, D) endemic community in Taiwan. J Med Virol 2003; 70:74–80. 39. Rizzetto M. Hepatitis delta virus disease: an overview. Prog Clin Biol Res 1993; 382:425–430. 40. Fattovich G, Giustina G, Christensen E, et al. Influence of hepatitis delta virus infection of morbidity and mortality in compensated cirrhosis type B. The European Concerted Action on Viral Hepatitis (Eurohep). Gut 2000; 46:420–426. 41. Philip T, Durazzo M, Trautwein C, et al. Recognition of uridine disphosphate glucuronosyl transferases by LKM-3 antibodies in chronic hepatitis D. Lancet 1994; 344:578–581. 42. Smedile A, Rosina F, Saracco G, et al. Hepatitis B virus replication modulates pathogenesis of hepatitis D virus in chronic hepatitis D. Hepatology 1991; 13:413–416. 43. Rosina F, Conoscitore P, Cuppone R, et al. Changing pattern of chronic hepatitis D in southern Europe. Gastroenterology 1999; 117:161–166. 44. Niro GA, Rosina F, Rizzetto M. Clinical update: treatment of hepatitis D. J Viral Hepatitis 2005; 12:2–9. 45. Farci P, Mandas A, Coiana A, et al. Treatment of chronic hepatitis D with interferon alfa-2a. N Engl J Med 1994; 330:88–94. 46. Farci P, Roskams T, Chessa L, et al. Long term benefit of interferon alpha therapy of chronic hepatitis D: regression of advanced hepatic fibrosis. Gastroenterology 2004; 126:1740–1745. 47. Lau JY, King R, Tibbs CJ, et al. Loss of HBsAg with interferonalpha therapy in chronic hepatitis D virus infection. J Med Virol 1993; 39:292–296. 48. Lau DT, Kleiner DE, Park Y, et al. Resolution of chronic delta hepatitis after 12 years of interferon alfa therapy. Gastroenterology 1999; 117:1229–1233. 49. Puoti M, Rossi S, Forleo MA, et al. Treatment of chronic hepatitis D with interferon alpha-2b in patients with human immunodeficiency virus infection. J Hepatol 1998; 29:45–52. 50. Yurdaydin C, Bozkaya H, Gurel S, et al. Famciclovir treatment of chronic delta hepatitis. J Hepatol 2002; 37:266–271. 51. Lau DT, Doo E, Park Y, et al. Lamivudine for chronic delta hepatitis. Hepatology 1999; 30:546–549. 52. Niro GA, Ciancio A, Tillman HL, et al. Lamivudine therapy in chronic delta hepatitis: a multicentre randomized-controlled pilot study. Aliment Pharmacol Ther 2005;22:227–232. 53. Wolters LM, van Nunen AB, Honkoop P, et al. Lamivudine-high dose interferon combination therapy for chronic hepatitis B patients co-infected with the hepatitis D virus. J Viral Hepatitis 2000; 7:428–434. 54. Bordier BB, Ohkanda J, Liu P, et al. In vivo antiviral efficacy of prenylation inhibitors against hepatitis delta virus. J Clin Invest 2003; 112:319–321. 55. Samuel D, Zignego AL, Reynes M, et al. Long term clinical and virological outcome after liver transplantation for cirrhosis caused by chronic delta hepatitis. Hepatology 1995; 21:333–339. 56. Marzano A, Gaia S, Ghisetti V, et al. Viral load at the time of liver transplantation and risk of hepatitis B recurrence. Liver Transplant 2005;11:402-409.
Section V: Liver Diseases Due to Infectious Agents
34
HEPATITIS E S.K. Sarin and Manoj Kumar Abbreviations ALT alanine aminotransferase cDNA complementary DNA EIA enzyme immunoassays ET-NANBH enterically transmitted non-A, non-B hepatitis FHF fulminant hepatic failure
HAV HEAg HEV HIV IEM IFM
hepatitis A virus hepatitis E antigen hepatitis E virus immunodeficiency virus immune electron microscopy immune fluorescent microscopy
INTRODUCTION Hepatitis E virus (HEV) is an RNA virus that causes an acute, selflimiting hepatitis. This unclassified virus is enterically transmitted, although other routes of transmission may exist. Infection with HEV may be asymptomatic or may cause hepatitis varying in degree of severity from mild to fulminant disease. Hepatitis E is the commonest form of acute hepatitis in adults in highly endemic regions of Asia, but the disease is rarely diagnosed in industrialized countries. Fulminant hepatitis E has been reported with increased frequency in pregnant women. Evidence for an enterically transmitted virus different from the hepatitis A virus (HAV) came from serologic studies of waterborne epidemics of acute hepatitis in India in the late 1970s. Khuroo1 and Wong et al.2 demonstrated that patients involved in such epidemics of hepatitis in the Kashmir region and in Delhi, India, respectively, lacked serologic evidence of recent HAV infection and only showed evidence of past infection. Therefore, they concluded that another agent must have caused the acute hepatitis. Balayan et al.3 in 1983 provided the first proof of the existence of a newly identified form of acute viral hepatitis by transmitting hepatitis to a volunteer from a patient involved in an outbreak of enterically transmitted non-A, non-B hepatitis in central Asia. The volunteer, who had pre-existing antibody to HAV, developed a severe hepatitis, shed 27–30-nm virus-like particles in his feces detected by immune electron microscopy (IEM), and developed antibodies to the virus-like particles during convalescence. The researchers also inoculated cynomolgus monkeys with the new virus; again, the monkeys developed hepatitis, shed virus-like particles, and developed an immune response to the particles. In 1990, Reyes et al. succeeded in cloning and sequencing a part of the genome of the virus.4 The new form of non-A, non-B hepatitis came to be known as epidemic non-A, non-B hepatitis or enterically transmitted non-A, non-B hepatitis (ET-NANBH), and later, the name of the disease was changed to hepatitis E to conform with the accepted nomenclature for the other types of viral hepatitis.5,6 Ironically, hepatitis occurring in previous centuries and the early 20th century, and known variously as “campaign jaundice” and “infective hepatitis,” was probably not due to HAV but instead was
IG MAbs pORF2 RT-PCR VLPs
immunoglobulin monoclonal antibodies polyprotein open reading frame 2 strand-specific reverse transcriptase polymerase chain reaction virus-like particles
due to HEV because the epidemiological descriptions of such disease resemble those of hepatitis E, not hepatitis A.6
VIROLOGY CLASSIFICATION As per the physicochemical properties, HEV was initially grouped into the Calicivirus family.7 HEV closely resembles the sequence of rubella virus, a member of the virus family Togaviridae, as well as the sequence of beet necrotic yellow vein virus, a plant furovirus.8 The sequences of the putative RNA polymerase and helicase of HEV resemble those of superfamily 3 viruses (rubella, and other alphaviruses) rather than those of superfamily 1 (caliciviruses, picornaviruses, and others).9 The Caliciviridae Study Group recently submitted a proposal to the International Committee on Taxonomy of Viruses to remove HEV from the family Caliciviridae and place it into an “unclassified” status. The decision was based primarily on the lack of phylogenetic relatedness between the HEV and caliciviruses.10,11 as well as profound differences in the types of putative replicative enzymes used by HEV, as described by Koonim and Dolja.12 The final taxonomic status of HEV remains to be determined.13
STRUCTURE Physiochemical Characteristics The buoyant density of HEV is 1.35–1.40 g/cm3 in CsCl.3,14 Its sedimentation coefficient is 183 S.15 The virus is relatively stable to environmental and chemical agents.16 HEV contains an RNA genome enclosed within a capsid that is composed of one or possibly two proteins, but direct analysis of purified virions has not been possible to date.17
Morphology HEV is a spherical, non-enveloped particle that is approximately 27–34 nm in diameter and has an icosahedral symmetry.18,19 It has an indefinite surface structure that is intermediate between that of the Norwalk agent (a member of the Caliciviridae family) and that of HAV (a member of the Picornaviridae family).20 Since HEV is
693
Section V. Liver Diseases Due to Infectious Agents
refractory to growth in continous cell cultures and is not present in large amounts in clinical material, there has been very little characterization of authentic viral particles. However, the expression of truncated polyprotein open reading frame 2 (pORF2: aa 112–660) in the baculovirus system leads to the formation of HEV virus-like particles (VLPs).21,22 At around 27 nm, these VLPs are smaller than the intact virus particle. Xing et al. used cryoelectron microscopy to study the HEV structure.23 Their analysis suggested that HEV VLPs are assembled as a T=1 icosahedral particle containing 30 dimeric subunits of 50-kDa pORF2, with the potential to form an intact virion of the correct size with a T=3 arrangement of 90 dimeric subunits.23 pORF2 dimerization appears to be due to non-covalent interactions in the C-terminal part of the protein and may contribute to the assembly of the immunodominant ORF2.1 epitope. It is not known whether the pORF2 protein is truncated in viral particles (as it is in VLPs), and further studies are needed to enhance the understanding of HEV structure.
GENOME ORGANIZATION Early studies postulated that ET-NANBH was caused by an RNA virus. Therefore Reyes et al. attempted to construct a complementary DNA (cDNA) library. Virus-enriched gallbladder bile from cynomolgus macaques infected with the second passage Burma isolate was used as a source to construct a cDNA library in gglutamyltransferase 10. One clone, designated ET1.1, contained a 1.3-kb cDNA exogenous to non-infected human genomes. Using a sequence-independent single-primer amplification procedure, cDNA from the bile of an infected cynomolgus macaque was amplified and hybridized with the clone ET1.1. Subsequently similar hybridization analyses of human fecal materials collected from outbreaks of ET-NANBH in geographically separated locations indi-
cated that a common pathogen was responsible for the majority of ET-NANBH seen worldwide. A partial nucleotide sequence of the ET1.1 clone was published in 19904 and the entire viral genome was reported in 1991. The genome of HEV is a single-stranded positive-sense RNA molecule approximately 7.2 kilobases in length followed by a polyA tail.9 The HEV has a genome that encodes structural and non-structural proteins through the use of discontinuous and partially overlapping ORFs. The genome consists of a short 5¢ non-translated region, three ORFs, each in a different coding frame, and a short 3¢ nontranslated region that is terminated by a stretch of adenosine residues, organized as 5¢-ORF1–ORF3–ORF2-3¢, with ORF3 and ORF2 largely overlapping (Figure 34-1). The 5¢ and 3¢ untranslated regions are highly conserved and are likely to play roles in RNA replication and encapsidation. Recently it has been shown that the 5¢ end of the genome has a 7-methylguanosine cap.24 ORF1, the largest ORF, begins at the 5¢ end of the viral genome after a 27-bp non-coding sequence and extends 5079 bp to the 3¢ end (in the Burmese prototype strain), and encodes about 1693 amino acids encompassing non-structural, enzymatically active proteins probably involved in viral replication and protein processing. Based on the identification of characteristic amino acid motifs,25 the following genetic elements have been identified, in order, from the 5¢ to the 3¢ end of the ORF: (1) a methyl transferase, presumably involved in capping the 5¢ end of the viral genome; (2) the “Y” domain, a sequence of unknown function that is found in certain other viruses, including rubella virus; (3) a papain-like cysteine protease, a type of protease found predominantly in alphaviruses and rubella virus26 ; (4) a proline-rich “hinge” that may provide flexibility and that contains a region of hypervariable sequence;27,28 (5) an “X” domain of unknown function that has been found adjacent to
Genomic organization of HEV 5' UTR
3' UTR poly (A) tail RNA
Viral
Translation
MT
PORF 1 N
PP
Hel
RdRp C PORF 2 N PORF 3 Poly(cys)
N
C C Poly(pro)
Figure 34-1. Genomic organization of hepatitis E virus. The hepatitis E virus (HEV) genome is a ~7.5kb polyadenylated RNA. At its 5¢ and 3¢ termini, the viral RNA carries two short untranslated regions (UTRs). It is positive-sense and includes three open reading frames (ORFs. The three ORFs are organised as 5¢-ORF1-ORF3ORF2–3¢ and encode the viral proteins. ORF1 encodes a putative nonstructural polyprotein (pORF1) that includes domains found in viral methyltransferases (MT), papain-like cysteine proteases (PP), viral RNA helicases (Hel) and viral RNA-dependent RNA polymerases (RdRp). ORF2 encodes the major viral capsid protein (pORF2). ORF3 encodes a small protein (pORF3) with two hydrophobic domains in its N-terminal half, which includes a polycysteine [Poly(Cys)] region, and prolinerich [Poly(Pro)] sequences in its C-terminal half. pORF3 is phosphorylated at a single (Ser80) amino acid by the cellular mitogen-activated protein kinase (MAPK), associates with the cytoskeleton through its N-terminal end and with proteins carrying the src-homology 3 (SH3) motifs through its C-terminal end. pORF, polyprotein open reading frame.
694
Chapter 34 HEPATITIS E
papain-like protease domains in the polyproteins of other positivestrand RNA viruses;29 (6) a domain containing helicase-like motifs similar to those found in viruses containing type I (superfamily 3) helicases;30 and (7) an RNA-dependent polymerase, with motifs most closely related to those found in viruses containing an RNA polymerase of superfamily 3.31 Expression of pORF1 alone in HepG2 cells or in an in-vitro translation system failed to demonstrate any proteolytic processing into mature products, suggesting that other cofactors may be required for correct processing.32 ORF2, approximately 2000 nucleotides in length, begins approximately 40 nucleotides after the termination of ORF1 and consists of a 5¢ signal sequence, a 300-nucleotide region rich in codons for arginine, probably representing an RNA-binding site, and three potential glycosylation sites.33 ORF2 encodes a 660-amino-acid protein, most likely representing one or more structural or capsid protein(s) of HEV. The pORF2 is an 80-kDa glycoprotein with a potential endoplasmic reticulum directing signal at its N-terminus (a region containing high concentrations of arginine and lysine). It is synthesized as a precursor, then processed through signal sequence cleavage into the mature protein, and probably glycosylated at three potential glycosylation sites, which is common for the surface proteins of the envelope but not for the non-enveloped viruses. In-vitro assays have suggested that the protein is cotranslationally translocated across the endoplasmic reticulum and is expressed intracellularly as well as on the cell surface of the hepatocyte.34 When pORF2 is expressed in mammalian cells, a large proportion of the nascent protein is modified by N-glycosylation.35 However, this glycosylated form of the protein is highly unstable,36 and it remains to be clarified whether the authentic viral particle contains glycosylated capsid proteins. When pORF2 is expressed in insect cells, it is cleaved at a site between amino acids 111 and 112 and at various other sites within the C-terminus of the protein. Some of these truncated forms of the pORF2 have the ability to self-assemble into VLPs or subviral particles.37–39 However, studies on native virus particles are awaited, to confirm the biological relevance of these subviral particles. Recently, by using the yeast two-hybrid system and in-vitro immobilization experiments, it has been shown that full-length and N-terminal deletion fragment of the ORF2 protein can dimerize. This dimerization property may not be amino-acid sequencedependent but instead is a complex formation of a specific tertiary structure that imparts to pORF2 its property to self-associate.40 The ORF2 protein also contains RNA-binding activity and specifically binds to the 5¢ end of the HEV genome. A 76-nucleotode region at the 5¢ end of the HEV genome was responsible for binding the ORF2 protein. This interaction may be responsible for bringing the genomic RNA into the capsid during assembly, thus playing a role in viral encapsidation.41 ORF3, less than 400 nucleotides in length, overlaps ORF1 by one nucleotide at its 5¢ end and overlaps ORF2 by over 300 nucleotides at its 3¢ end. It codes for a 123-amino-acid, 13.5-kDa nonglycosylated protein (pORF3), which is a very basic protein (pI 12.5), and is the most variable protein among the HEV strains. The function of pORF3 is unknown, but it has been reported to be a phosphoprotein. When expressed in animal cells, pORF3 is phosphorylated at a single serine residue (Ser80) in its 123-amino-acid primary sequence, by a mitogen-activated protein kinase, and does not seem to undergo other major post-translational modifications.42
This phosphoprotein, which possesses two very strong hydrophobic regions in the N-terminal half of the molecule (possibly transmembrane a-helices with the potential to be associated with cellular membranes) was found to associate with the hepatocellular cytoskeleton, suggesting a possible role as a cytoskeletal anchor site for the assembly of virus particles.42 Recently a possible role of pORF3 in the modulation of cellular signal transduction pathways has been suggested. By using in-vitro binding assays, pORF3 has been shown to bind to a number of SH3-containing cellular signal transduction pathway proteins, including the protein tyrosine kinases Src, Hck, and Fyn, the p85a regulatory subunit of phosphatidylinositol 3-kinase, phospholipase C, and the adapter protein Grb2.43 Although ORF3 maps in the structural region of the HEV genome, it may have some regulatory functions. Also, as ORF3 interacts with ORF2, the possibility of a non-structural function for the ORF3 protein exists. By using fluorescence-based colocalization in yeast two-hybrid experiments, transiently transfected COS-1 cell (a cell line of saccharomyces cerevisiae) co-immunoprecipitation, and cellfree coupled transcription–translation techniques, it has been shown that ORF3 protein interacts with the ORF2 protein, the interaction being dependent on the phosphorylation at Ser80 of ORF3. Such an interaction suggests a possible well-regulated role for ORF3 in HEV structural assembly.44 The observations that pORF3 can homodimerize in vivo and the overlap of the dimerization domain with the SH3-binding and phosphorylation domains, suggests that pORF3 may have a dimerization-dependent regulatory role to play in the signal transduction pathway.45
GENOTYPES The initial two isolates of HEV from Burma4 and Mexico46 were only about 76% similar in overall genomic organization. In the years immediately following identification of the Burmese and Mexican prototype strains, a number of additional isolates were identified from Pakistan, India, and China. All of these isolates were closely related to the original Burmese isolate, having greater than 93% nucleotide identity across the genome. At that time, the genotypic distribution of HEV consisted of a group of Burmese-like isolates and the lone Mexican isolate.47 Additional genetic diversity was reported in an acute hepatitis patient from the USA.48–50 A number of additional isolates of HEV were subsequently identified from patients from Italy and Greece.51–53 Genetic diversity is seen not only in HEV from different regions, but also from isolates from the same region. Isolates from Argentina and Austria were found to be distinct from each other as well as all previously identified isolates.54–56 Also, the two isolates from patients from Spain were most closely related to one of the Greek isolates.57 Extensive diversity has also been reported between HEV strains from a number of patients from China and Taiwan that are distinct from the original Chinese isolates (genotype IV), and are closely related to the Burmese isolate.58–60 In Japan multiple HEV strains of genotype III or IV have been isolated from patients with acute hepatitis of non-ABC etiology who had never been abroad.61–65 Cocirculating strains within the same country have also been reported from Pakistan and Namibia, South Africa. The 1987 Pakistan-Sar55 isolate clustered with other Asian isolates in subgenotype I-1a, while the 1988 Pakistan-Abb2B isolate clustered with Burmese isolates in subgenotype I-1b.66 Despite the fact that the
695
Section V. Liver Diseases Due to Infectious Agents
1983 and the 1995 Namibian outbreaks occurred in the same area, the 1983 isolates clustered into genotype I67 and the 1995 isolate clustered with a Mexican isolate in genotype II.68 Genetic changes in HEV can occur over time in a given community. Shrestha et al.69 compared the HEV isolates in the Kathmandu valley of Nepal, recovered from 48 patients in 1997, 16 patients in 1999, 14 patients in 2000, and 38 patients in 2002. The annual frequency of cluster 1a-2 isolates declined from 63% in 1997, to 50% in 1999, to 7% in 2000 and none in 2002; cluster 1a-3 isolates were seen in all 4 years; and the annual frequency increased from 5% in 1997 to 95% in 2002. While cluster 1a-1 was only detectable in 1997, clusters 1a-4 and 1a-5 emerged in 2000 and 2002 respectively. HEV subtype 1c was identified in 1997, but not in 1999 and thereafter. To date, HEV 1c has been isolated from India and mainland China, suggesting that the 1c strain was imported from India or China to Nepal in 1997 or before, but was taken over by the cocirculating 1a strains in 1999. Also the observed changing prevalence of the various 1a clusters according to the year of disease onset suggests that the genetic variability of HEV in a community is due to continously occurring genetic changes and that takeover of existing strain(s) possibly occurs by the selected variant having an advantage in transmission in the community or the variant(s) that was imported from other communities. An alternate hypothesis for the variability observed in field isolates is ongoing evolution in alternative host species.69 Based on the phylogenetic relationships of the various isolates, many authors have proposed nomenclature for genotypes of HEV (Table 34-1).47,56,60,70
SEROTYPES AND ANTIGENICITY Despite the presence of genetically different isolates of HEV, there appears to be only one serotype. Antigenic variations have important implications for the serological detection of HEV infection. The type specificity of many epitopes was first recognized by Reyes et al.27 Antibody responses to individual viral antigens are highly variable, due to both strain-specific differences in some epitopes and differences in response to single antigens between individual patients. For example, pORF3 varies greatly between strains, and many experimentally infected animals and some patients fail to develop antibodies to ORF3 protein.71,72 This variable reactivity contributes to the poor sensitivity and concordance of HEV-diagnostic tests based on such antigens.73 Conversely, all isolates of HEV share some important cross-reactive antigens.. Immunization of prehuman primates with recombinant pORF2 proteins conferred immunity to both homologous and heterologous challenge, suggesting that major protection epitopes are common among HEV genotypes (see Active immunoprophylaxis, below).
EPIDEMIOLOGY Hepatitis E is often mentioned among the “novel” diseases,74 which is right only in a sense of “newly described” or “newly identified” and should not be interpreted as a “newly emerged” disease.75 Hepatitis occurring in previous centuries and the early 20th century and known variously as “campaign jaundice” and “infective hepati-
Table 34-1. Proposed Classification of Hepatitis E Virus Genotypes Isolates
Bur1 (Burmese) (2,3,4), Bur2 (Burmese) (2,3,4), I2 (Indian) (2,3,4), I1 (Indian) (3) C1 (Chinese) (2,3,4), C2 (Chinese) (2,3,4), C3 (Chinese) (2,3,4), C4 (Chinese) (2,3,4), C5 (Chinese) (2), C6 (Chinese) (2), P1 (Pakistani) (2,3,4) B1 (Chinese) (2), B2 (Chinese) (2), B6 (Chinese) (2,4), B7 (Chinese) (2,4), S1 (Chinese) (2), S10 (Chinese) (2), S11 (Chinese) (2), S13 (Chinese) (2,4), H8 (Chinese) (2) Tash (Uzbekistan) (2,4), Osh (Kirgizia) (2,4) Mor12 (Morocco) (2,4), Mor23 (Morocco) (2,4), Chad (2,4) Mex (Mexico) (2,3,4) Ni (Nigerian) (4) SwUS1 (swine USA) (3,4) US1 (USA) (2,3,4) US2 (USA) (2,3) S15 (Chinese) (2,3,4), H3 (Chinese) (2,4) B3 (Chinese) (2,4), B4 (Chinese) (2,4), S5 (Chinese) (2,4) Sw NZ1 (swine New Zealand) (4) It1 (Italian) (3,4) Gr1 (Greek) (3,4), Sp1 (Spain) (4), Sp2 (Spain) (4) Gr2 (Greek) (3,4) Ar1 (Argentina) (3,4), Ar2 (Argentina) (3,4) Au1 (Australia) (4) Ct1 (Chinese) (4) Cs9 (Chinese) (4) Ct705 (Chinese) (4), Ct825 (Chinese) (4), Ct845 (Chinese) (4)
696
Arankalle70 (1999) (1)
Wang60 (1999) (2)
Schlauder56 (2000) (3)
Schlauder47 (2001) (4) scheme 1
Schlauder47 (2001) (4) scheme 2
IA
1a
1
I
1
IB
1b
1
I
1
IC
1c
I
1
ID
1d 1e 2
I I II II III III
1 1 2 12 3 3
IV IV III III III III III III IV IV IV
9 10 4 4 5 6 7 7 8 11 10–11
II III
3a 3b 4a 4b
2 3 3 3 4
5 6 7 8
Chapter 34 HEPATITIS E Figure 34-2. Countries with high endemicity for hepatitis E virus (HEV) or sites of reported outbreaks. (Data from Aggarwal R, Krawczynski K. Hepatitis E: an overview and recent advances in clinical and laboratory research. J Gastroenterol Hepatol 2000; 15:9–2000, ©2000, with permission of Blackwell Publishing.368)
Table 34-2. Characteristics of Endemic versus Non-endemic Areas of Hepatitis E Virus (HEV) Features
Endemic
Non-endemic
Incidence of hepatitis E Epidemiological manifestations General sanitation Spread through contaminated water Non-human HEV reservoirs
High Outbreaks
Low Only sporadic cases
Poor Frequent
Adequate Absent
Probable
If exist, do not play essential role Undocumented Moderate Good
Subclinical infection Climatic conditions Hepatitis E reporting system
Frequent Hot Poor
tis” was probably hepatitis E and not hepatitis A, because the epidemiological descriptions of such disease resemble those of hepatitis E, not hepatitis A.6 Therefore it is possible that HEV infection may once have been prevalent in various parts of the world and has only recently become restricted to certain geographic areas, mostly underdeveloped regions with poor sanitation.
INCIDENCE AND PREVALENCE Worldwide, two geographic patterns can be differentiated: (1) areas of high HEV prevalence (endemic regions), in which major outbreaks and a substantial number of sporadic cases occur; and (2) non-endemic regions, in which HEV accounts for a few cases of acute viral hepatitis, mainly among travelers to endemic regions. In connection with epidemiological studies, a concept of endemic versus non-endemic areas is appropriate. Several features have been suggested for characterization of these areas (Table 34-2).75 Endemic disease is geographically distributed around the equatorial belt, including Central America, Africa, and the Middle East, subcontinental India, Asia, and the south-east Pacific (Figure 34-2). In these areas, which are generally characterized by inadequate envi-
ronmental sanitation, major outbreaks of HEV infection involving thousands of cases have been reported. A large epidemic of waterborne viral hepatitis was reported from India in 1955 and 1956, when raw sewage from the flooding Yamuna river resulted in 30 000 cases of jaundice.76 Although initially presumed to be caused by HAV, subsequent analysis of stored serum samples indicated that this epidemic was caused by HEV. It is now recognized that HEV is the commonest cause of epidemic enterically transmitted hepatitis, and HEV is viewed as a significant health problem by the World Health Organization. The largest reported outbreak occurred in the Xinjiang region of China between 1986 and 1988 and involved 120 000 cases.77 Studies of this and other large outbreaks of HEV infection have provided important information on the clinical and epidemiologic features of HEV infection. During the outbreaks, overall attack rates range from 1% to 15%, being much higher among adults 15–40 years (3–30%) than children (0.2–10%).1,78–80 The reason for this pattern of age distribution, which is unusual for an enteric infection, is unknown. It has been suggested that HEV runs a predominantly anicteric course in young age groups followed by gradual loss of immunity.1 It has also been speculated that HEV somehow has a selective tropism for liver cells of adults.81 Nevertheless, young children are susceptible to infection with HEV, because clinical disease has occurred with a similar frequency in all age groups in some epidemics,82 and sporadic clinical hepatitis E in children has been reported.82–91 A male preponderance of cases has been observed in most reports (the male to female ratio varies from 1.5 to 3.5:1).76 It is unclear whether this reflects the greater involvement of men in professional and social activities and, accordingly, their greater exposure to risk factors, or a true difference in susceptibility. Outbreaks of hepatitis E are characterized by high attack rates and mortality in pregnant women.1,76,78 An outbreak may be singlepeaked and short-lived, or multi-peaked and prolonged, lasting for more than a year. In endemic areas, hepatitis E accounts for a substantial proportion of cases of acute sporadic hepatitis in both children and adults.
697
Section V. Liver Diseases Due to Infectious Agents
In India, HEV infection accounts for 50–70% of all cases of sporadic viral hepatitis.1,92,93 The demographic and clinical features of patients with acute sporadic hepatitis E closely resemble those during epidemics of hepatitis E.94 In endemic areas, outbreaks have a periodicity of 5–10 years, which in part reflects the patterns of heavy rainfall. The reservoir for HEV during interepidemic periods is unknown. Sporadic HEV infection in endemic areas may be sufficient to maintain the virus within the community during the interepidemic periods. Another possibility is that a non-human HEV reservoir exists. HEV has been isolated from swine, and antibodies to HEV have been detected in a number of animal species, including swine, sheep, cattle, chickens, rats, and captive monkeys. Moreover, viruses recovered from swine have been identified as variants related to human HEV strains found in the same geographic regions (see Is HEV a zoonosis? below). In non-endemic countries, cases of acute HEV infection are uncommon and most of the cases reported in industrialized countries occur in travelers recently returned from an endemic area.95–106 Secondary transmission has not been reported in these cases. Acute hepatitis E without a history of travel has also been described,107–110 and the source of HEV infection in such cases has not yet been determined. In most non-endemic areas, the disease accounts for fewer than 1% of reported cases of acute viral hepatitis. Although there are still questions about the relative sensitivity and specificity of serologic tests for anti-HEV, a more complete picture of the worldwide distribution and seroprevalence of HEV infection is emerging. Surprisingly, the prevalence of antibody to HEV in developing and documented endemic regions is much lower than expected (3–27%) (Table 34-3).111–143 The paucity of anti-HEV seropositivities in normal human populations of endemic countries has been difficult to explain. Various explanations could be: 1. Anti-HEV immunoglobulin G (IgG) does not persist for a long period of time and disappears early after the acute infection. 2. HEV causes, predominantly, a clinically overt disease: subclinical or silent forms are rare. 3. Currently available serological tests do not capture the whole spectrum of anti-HEV antibodies. Clinical HEV infection in developed countries is frequently associated with recent travel to areas of endemic infection. However, swine HEV strains have a worldwide distribution and infection in pigs from commercial farms is almost ubiquitous. In view of the demonstrated potential of swine HEV to cross species barriers and the close relationship between swine HEV and strains isolated from some humans, HEV infection should be considered a possibility in acute hepatitis patients who do not have a relevant travel history or markers of other hepatitis viruses (see Is HEV a zoonosis? below). Also, the prevalence of anti-HEV in nonendemic regions has been much higher than anticipated (1–28%) (Table 34-4)147–165 and thus, the true rate of HEV infection in developed countries remains controversial. Serological assays based on truncated pORF2 protein expressed using the baculovirus system have yielded seroprevalence rates of over 20% in blood donors from Baltimore, Maryland, USA,144 whereas rates lower than 2% were found in Australian blood donors when using the ORF2.1 protein expressed in Escherichia coli.145 Differences in assay specificity (favoring the ORF2.1 based
698
assay) are a major problem in the detection of HEV prevalence, but differences in assay sensitivity (favoring baculovirus-based assay) may also be important.146 In many studies those individuals who showed the presence of anti-HEV did not differ from the seronegatives with regard to history of recent travel to endemic areas or exposure to local risk factors such as contact with imported hepatitis E cases. Possible explanations for these phenomenon include: 1. inadequacy of the methods employed for detection of antiHEV IgG (false-positivity) 2. circulation of an unrecognized agent capable of eliciting antibodies cross-reacting with HEV 3. a situation where the same virus causes a certain pathology in one climatic zone and a completely different disease in another (e.g., the Epstein–Barr virus causes infectious mononucleosis in moderate climates and Burkitt lymphoma in Africa). In endemic areas, the prevalence of anti-HEV in infants and children has been much lower than expected for a virus transmitted by the fecal–oral route.82,111,113,129,133,135,136,139,143,166 However, in a report from India, anti-HEV was detected in more than 60% of children below 5 years of age.90 Such differences may be related to varying epidemiological conditions or differences in the diagnostic tests used. An increase in the prevalence of anti-HEV has been found among young adults, but not older adults (constant between 10 and 40%).87,111,129,143,166 Although such a pattern of age-specific anti-HEV might suggest a cohort effect representing the disappearance of HEV from endemic regions, as was seen for HAV previously,6 similar age-specific anti-HEV patterns have been reported for sera collected 10 years apart from the same population residing in India, an area that is highly endemic for HEV.143,166 Thus, HEV appears to have epidemiologic characteristics that are quite different from those of most viruses that are transmitted by the fecal–oral route.
TRANSMISSION HEV is transmitted by the fecal–oral route. The most common vehicle of transmission during epidemics has been the ingestion of fecally contaminated water.1,76,78,80 Outbreaks in endemic areas occur most frequently during the rainy season, after floods and monsoons, or following recession of flood waters.76,80,167,168 These climatic conditions in conjunction with inadequate sanitation and poor personal hygiene lead to epidemics of HEV infection, when the sewage waters gain access to open-water reservoirs.76,78,169 In several regions of HEV-endemicity, a pattern of recurrent epidemics has been observed, which is probably related to the permanent existence of conditions in which drinking water is fecally contaminated. In South-East Asian regions, the disposal of human excreta into rivers and the use of river water for drinking, cooking, and personal hygiene have been shown to be significantly associated with a high prevalence of HEV infection: the use of river water over years for various activities can lead to recurrent epidemics.170,171 Both in epidemic and sporadic HEV there is a low rate of clinical illness among household contacts of infected patients, an unexpected finding because the virus is transmitted by the fecal–oral route. Reported secondary attack rates in households of HEVinfected persons range from 0.7% to 2.2%, in contrast to secondary
Chapter 34 HEPATITIS E
Table 34-3. Prevalence of Antibodies to Hepatitis E Virus Among Normal Humans in Endemic Areas Country
Number of samples tested
Taiwan, China
384 adults > 20 years 600 < 20 years 184 urban pregnant woman (Caracas) 204 rural populations (San Camilo, Edo Apure) 223 rural Amerindians (Padamo, Edo Amazonas) 1350 overall 105 second decade 464 third decade 308 fourth decade 157 adults attending sexually transmitted disease clinic 61 129 adults 593 blood donors 174 blood donors 252 blood donors 155 residents of a semiurban village 555 canoeists 227 medical students 1360 blood donors 72 health care workers 241 inmates in state jails 100 Araucanian indians 1850 villagers 503 Inuits 407 urban 360 rural 996 ethnic population 934 134 355 997 165 blood donors with ALT < 2 ¥ ULN 40 blood donors with ALT > 2 ¥ ULN 117 health care workers 53 cleaning service workers Women from a rural area 98 children 100 adults 10 026 rural population 1065 urban children 1005 rural children 487 children 699 1046 185 children 139 adults 96 blood donors 936 884 urban high-socioeconomic group 1497 urban lower-middle-socioeconomic group 1710 rural lower-middle-socioeconomic group 250
Venezuela
Turkey
Central African Republic Brazil Burundi Saudi Arabia Chile Uruguay Egypt South Africa Chile
Egyptian Nile delta West Greenland South Africa French Guiana Tadjikistan Kirghizstan Hong Kong Taiwan Brazil
Tanzania Pakistan Egypt India Amazonian basin, Brazil Rio de Janeiro, Brazil Ankara, Turkey. Tamil Nadu, India Thailand Korea Hong Kong India
India
% seropositive
Reference
10.7 0.3 1.6 3.9 5.4 5.9 0 3.7 9.1 24 4.9 14 16.9 4 1.2 57 1.8 2.6 8 12.5 7.5 17 17.2 3 6.6 15.3 6.4 8.5 4.6 16.1 6.4–8.8 3 7.5 2.6 13.2 7 19.4 16 67.7 28.7 23.8 4.5 2.4 3.8 5.3–16.7 6.5 18 18.8 6.9 10.6 14 4
111 112
113
114 115 116 117 118 119 120 121 122
123 124 125 126 127 128 129 130 131
132 133 134 135 136 137 138 139 140 141 142 143
92
ALT, alanine aminotransferase; ULN, upper limit of normal.
attack rates of 50–75% in households of HAV-infected individuals.85,86,172,173 The reasons for this difference may be related to instability of HEV in the environment, differences in infectious dose needed to produce infection, or a higher frequency of subclinical disease among persons secondarily infected with HEV. Even when multiple cases occur among members of a family, such occurrence
is related to exposure to a common source of contaminated water rather than to person-to-person spread.174 The mode of transmission responsible for sporadic HEV infections is unclear. Contaminated water is probably responsible for most of the cases in this setting. However, foodborne hepatitis E infection after eating uncooked liver of pig or wild boar and meat
699
Section V. Liver Diseases Due to Infectious Agents
Table 34-4. Prevalence of Antibodies to Hepatitis E Virus Among Normal Humans in Non-endemic Areas Country
Number of samples tested
% seropositive
Reference
Northern California (USA) Baltimore, Maryland (USA)
5000 blood donors
1.2–1.4
147
15.9 23 21.3
144
2.5 2.6 0.4 2.81 1.81 2.6 5.4 9.3
148 149 150 151
1 2.1 0.9 2.2 1.1 1.5 1.2 0.5 1.7 6.7 4.6
153 153 153 153 106 154 155 127 156 157
3.8 2.1 0.89 14.8
158 159 160 161
1.7 3 0.4 18 29 25 20 14 9 2.4
162 163
Portugal Italy Australia Israel Italy
UK Germany France Spain Netherlands Belgium Russia Ukraine New Caledonia Japan
Norway Istanbul, Turkey Antalya, Turkey Italy Bolivia Japan USA
Canada
300 homosexual males 300 intravenous drug users 300 blood donors 1473 blood donors 653 279 1139 Jews 277 Arabs 1889 general population 279 intravenous drug users 193 chronic hepatitis disease 1500 972 1007 775 1275 394 168 1721 351 military recruits 478 in a hepatitis C virus-endemic area 775 in a non-hepatitis C virus-endemic area 208 flying personnel 909 children 338 1005 refugee Kurds from Iraq and Turkey 1393 children 200 adults 246 infants 400 blood donors >60 years (55) 50–59 years (65) 40–49 years (104) 30–39 years (95) <30 years (81) 165 adults
152
164
POPULATIONS AT HIGHER RISK FOR HEV ACQUISITION Persons Having Contact With Swine Recently, a high prevalence of antibodies to HEV were found among persons who work with swine.185,186 In Maldova, a country without reported cases of hepatitis E, prevalence of anti-HEV in swine farmers was 51.1% versus 24.5% prevalence among persons without occupational exposure to swine.187 Withers et al. found a 4.5-fold higher anti-HEV prevalence (10.9%) among North Carolina, USA, swine workers, as compared to unexposed subjects (2.4%).188 Meng et al.164 recently reported that, among the 295 swine veterinarians from the eight US states from which normal blood donor data were available, 78 (26%) were positive for anti-HEV when tested with Sar55 antigen and 68 (23%) were positive when tested with swine HEV antigen. In contrast, 73 of 400 normal blood donors (18%) from the same eight US states were positive with Sar55 antigen and 66 (16%) were positive with swine HEV antigen. Thus, human populations with occupational exposure to certain animals have an increased risk of HEV infection. However, whether infection with swine HEV leads to clinical illness is unclear.
Persons Having Contact With Untreated Waste Water 165
of wild deer has been reported.175–177 Recently, it has been suggested that, in countries where HEV is endemic, the transmission of hepatitis E may be associated with blood transfusion. In a recent study from India, 56 transfusion recipients were followed biweekly for 3 months after transfusion. Out of these, 19 were positive and 37 were negative for IgG anti-HEV in the pretransfusion sample. Two of the 37 IgG anti-HEV-negative recepients seroconverted to IgM and IgG anti-HEV 4 and 5 weeks after transfusion, and one developed raised serum aminotransferases. None showed symptoms of hepatitis and attempts to detect HEV RNA in transfused blood were
700
not successful. In contrast, none of the 34 IgG anti-HEV-negative controls seroconverted during the period of follow-up.178 Presumed nosocomial spread of HEV has been reported in South Africa, where acute hepatitis developed in three health care workers 6 weeks after they treated a patient with fulminant hepatic failure (FHF) due to HEV infection.179 Persons working in emergency admission or in surgery in a German study were placed in a highrisk category for acquiring HEV infection.180 In an experimental study, pregnant rhesus monkeys failed to transmit the virus to their offspring.181 However, vertical transmission of HEV infection from mother to infant has been shown to occur. In one study, six of eight babies born to mothers who had either acute uncomplicated hepatitis or FHF due to HEV infection in the third trimester of pregnancy were found to have evidence of HEV infection. Of these, five had HEV RNA in samples of their blood taken at birth, suggesting that infection was transmitted transplacentally.182 More recent studies have shown that mother-to-infant transmission occurred in 50–100% of HEV RNA-positive mothers during pregnancy.183,184
Recently HEV RNA has been identified in a substanial proportion of untreated sewage samples in both non-endemic and endemic areas. One group found that 43.5% (20/46) of urban sewage samples collected in Barcelona, Spain, from 1994 to 2002 tested positive for HEV RNA.189 On the basis of a year-round survey of sewage samples from a sewage treatment plant from Pune, India, 10.97% (9/82) unconcentrated sewage samples before receiving any treatment were found to be HEV RNA-positive.190 In a study from India, antiHEV IgG-positivity was significantly higher among staff members of a sewage treatment plant (56.5%) when compared with controls (18.9%). A sevenfold higher risk of hepatitis E infection was recorded in sewage workers working in close proximity to sewage and a 3.9-fold higher risk in staff members not coming into frequent contact with sewage.191 In a study from Turkey, agricultural workers
Chapter 34 HEPATITIS E
who use untreated waste water for irrigation had an anti-HEV positivity rate of 34.8% as compared to 4.4% in controls.192
Table 34-5. Naturally Occurring Hepatitis E Virus (HEV) in Swine Country
Category of swine
Anti-HEV antibodies: Number positive/number tested (%)
HEV viremia: Number positive/number tested (%)
Australia
Wild Domestic Domestic Domestic Specific Pathogen Free (SPF) Domestic SPF Domestic Domestic Domestic Domestic Domestic Domestic Domestic Domestic
15/59 (17) 12/40 (30) 18/55 102/275 (37) 0/10 (0)
— — 3/47 —
202 202 203 186 186
202/251 (80) 0/10 (0) 12/128 (15) 1448/2500 (58) 29/84 (34.5) 330/419 (78.8) — 122/284 (42.9) 594/998 (59.4) 102/137 (74.4) 53/97 (54.6)
— — 3/128 (2.3) 147/930 (15.8) — 5/236 (1.9) 34/96 (35) 13/284 (4.6) — — —
204 204 141 205 188 206 207 208 165 209 209
HIV-infected Persons An association between anti-HEV seropositivity and human immunodeficiency virus (HIV) infection has been suspected. Over one-third of HIV-infected homosexual men were shown to have IgG anti-HEV.193
Hemodialysis Patients The prevalence rates of IgG anti-HEV are variably reported in hemodialysis patients and asymptomatic blood donors.194,195 While no difference was reported in one study, in the other, a high prevalence (9.7% versus 0.53% in healthy blood donors) was reported. The prevalence of IgM anti-HEV in the patients (4.8%, 4/83) from a dialysis unit in Saudi Arabia was significantly higher than in healthy controls (0.3%, 1/400).196 HEV prevalence was found to be higher in peritoneal dialysis patients as compared to healthy people in a study from Spain.197
Chronic Autoimmune Liver Disease In patients with autoimmune liver disease the rate of anti-HEV was found to be as high as 46%: all anti-HEV positive patients had a more serious disease (compared to anti-HEV-negative patients), rapidly progressing to cirrhosis despite immunosuppressive therapy.198
Other Population Groups A higher prevalence of IgG anti-HEV in various other groups has been found, including hemophiliacs,199,200 patients with sickle-cell anemia and beta-thalassemia major,201 and persons working in emergency rooms or in operating rooms in a German study.180
IS HEV A ZOONOSIS? Animal Strains of HEV: Serological Evidence and Genetic Identification The prevalence of antibody to HEV in a variety of animal and avian species is summarized in Tables 34-5202–209 and 34-6.210–222 Chimpanzees were not found to be exposed to HEV.217,219 In contrast, both wild-caught and domestically raised macaques and rhesus monkeys had serological evidence of prior exposure to HEV,210,218,219 and the prevalence of anti-HEV increased with age.210,217,219,220 That naturally occurring anti-HEV in macaques was indicative of actual infection with an HEV-like agent was proven by demonstrating that the presence of anti-HEV correlated with protection from experimental challenge with virulent HEV.218 Numerous non-primate species also have serological evidence of prior infection with HEV. In particular, a very large proportion of swine, whether raised in developing or industrialized countries, have such antibodies (Table 34-5). The role of rats in the spread of HEV was suspected on the basis of immunoelectron microscope-based demonstration of the virus in rats caught within a distance of 1 km from a hepatitis E epidemic-affected village in the former USSR.214 A large proportion of several species of rats and mice were found to be positive for antiHEV in USA.213,221 As with HEV in swine, HEV naturally infecting rats may represent a unique virus that is well adapted to replicate
Nepal Taiwan
USA Korea Japan USA China USA India Canada India
Reference
in this host. Antibody to HEV has been detected in the sera of a few other species, including cattle, dogs, sheep, goats, and chickens. An agent causing “big liver and spleen disease” of chickens has recently been shown to be genetically related to HEV.222,223,224 In 1997, a novel strain of HEV (designated as swine HEV) was genetically identified and characterized from a pig in USA.204 The genomic organization of swine HEV was very similar to that of the human strain of HEV. Thereafter, many strains of HEV have been isolated from pigs from different countries including USA,207 Taiwan,186,225 Netherlands,225,226 Canada,165,227 New Zealand,228 Korea,141 Japan,205,229,230 and India.208 Most of these studies have found that the swine HEV are genetically heterogenous and are closely related to human strains of HEV of the same geographic area. However, a recent report from India showed that Indian swine and human HEV belong to different genotypes, in contrast to reports from other areas.208 In addition to pigs, variant strains of HEV were reportedly identified from wild-trapped rodents from Kathmandu Valley, Nepal. Sequence analyses indicated that the HEV sequence recovered from rodents is most closely related to the HEV isolates from patients in Nepal.212 The agent causing “big liver and spleen disease” of chickens has recently been shown to be genetically related to human HEV.222,223,224
Cross-Species Infection of HEV It has been shown that swine HEV can cross species barriers. There has been some success in infecting US pigs with a human HEV strain from the USA.231,232 However, in a recent report from India, the investigators were not able to infect Indian pigs with human HEV (AKL-90).209 This could be because Indian swine and human HEV belong to different genotypes, i.e., genotypes 4 and 1 respectively.208 Lambs are reportedly infected with human HEV isolates Osh-225
701
Section V. Liver Diseases Due to Infectious Agents
Table 34-6. Prevalence of Anti-hepatitis E Virus (HEV) in Various Animal Species Country
Species
Anti-HEV Antibodies Number positive/number tested (%)
Japan
Japanese monkeys Cynomolgus monkeys Rhesus monkeys Taiwanese monkeys Norway rats (Rattus norvegicus) Black rats (Rattus rattus) Mice Rattus rattus brunneusculus Bandicota bengalensis Suncus murinus Mus musculus castaneus Unidentified murine rodents Cattle Goat Cattle Dog Goat Rattus rattus rufescens Rattus rattus andamanensis Rattus norvegicus Bandicota bengalensis Mus musculus Rattus norvegicus Rattus rattus Neotoma albiguala Neotoma mexicana Neotoma micropus Oryzomys palustris Sigmodon hispidus Mus musculus Peromyscus boylei Peromyscus eremicus Peromyscus leucopus Peromyscus maniculatus Microtus Clethrionomys gapperi
84/232 (36.2) 2/19 (10.5) 3/83 (3.6) 1/1 (100) 114/362 (31.5)
210 210 210 210 211
12/90 (13.3) 0/55 (0) 45/350 (13) 8/39 (21) 2/108 (2) 0/3 (0) 23/174 (13) 12/190 (6.3) 0/316 (0) 13/188 (6.9) 10/44 (22.7) 0/240 (0) 8/52 (15.8) 2/55 (3.6) 0/32 (0) 12/22 (54.5) 0/12 (0) 135/197 (69) 31/81 (38) 13/22 (59) 48/84 (57) 1/8 (13) 10/41 (24) 37/110 (33) 2/14 (14) 2/24 (8) 3/7 (43) 5/50 (10) 10/91 (11) 0/9 (0) 4/6 (67)
211 211 212 212 212 212 212 206 206 209 209 209 209 209 209 209 209 213 213 213 213 213 213 213 213 213 212 213 213 213 213
Japan
Nepal
China India
USA
Reference
and Osh-228.233 Similarly, Wistar rats,234 and cats235 were reportedly infected with a human stool suspension containing infectious HEV.234 Cases of sporadic acute hepatitis E have been described from Japan from pet cats and domestic swine and deer.177,235,236 The potential for cross-species infection by HEV raises important public health concerns. However, whether a zoonotic reservoir plays a major role in the epidemiology of HEV is not known.
PATHOGENESIS INCUBATION PERIOD The incubation period from exposure to the onset of clinical disease is approximately 28–40 days, based on analysis of waterborne epidemics in which the time of exposure was identified.76 In experimental HEV transmission studies in humans, liver enzyme values peak 42–46 days after ingestion of the virus.3 In experimental infec-
702
Country
Species
Anti-HEV Antibodies Number positive/number tested (%)
Russia
Apodemus sylvaticus Rattus turkestanus Goat Mus musculus Cow Water buffalo Chicken Goat Dog Chicken Owl monkey Squirrel monkey Tamarin Vervet Pig-tailed macaque Captive cynomolgus Captive rhesus Colony-born chimpanzee Wild langur Wild bonnet macaque Colony-born chimpanzee Pig-tailed macaque Captive cynomolgus Captive rhesus SPF rhesus Wild cynomolgus Rattus rattus Rattus norvegicus Rattus exulans Chicken
2/4 (50) 1/4 (25) 0/25 (0) 7/25 (28) 0/52 (0) 0/60 (0) 9/66 (14) 0/75 (0) 15/56 (27) 41/93 (44) 0/2 (0) 0/2 (0) 0/2 (0) 0/2 (0) 0/2 (0) 0/2 (0) 0/2 (0) 0/2 (0) 1/50 (2) 9/47 (19) 0/101 (0) 2/33 (6) 30/30 (100) 152/159 (96) 0/172 (0) 5/31 (16) 102/112 (90) 83/108 (77) 15/18 (83) /1276 (30)
Nepal
Vietnam
USA
India USA
India USA
USA
Reference
214 214 215 215 215 215 215 216 216 216 10 217 217 217 217 217 217 217 218 218 219 219 219 219 219 220 221 221 221 222
tion of pregnant rhesus monkeys, the incubation period varied from 1–2 weeks237 to 4–5 weeks.181
VIRAL REPLICATION (Figure 34-3) The knowledge of HEV replication is poor, due to the lack of practicable cell culture systems for the virus. Several strategies for experimental propogation and production of HEV to study the molecular biology have been reported, but their reproducibility and feasibility need confirmation. Propogation of HEV has been demonstrated in primary cultured macaque hepatocytes,238,239 but the level of replication is very low. This cell culture experiment used in-vivo-infected highly differentiated primate liver cells for the in-vitro studies of the replication of HEV. A strand-specific reverse transcriptase polymerase chain reaction (RT-PCR) technique was used to monitor replication. Both positive-strand and negative-strand HEV RNA were detected in cel-
Chapter 34 HEPATITIS E
1 2
Positive strand RNA
3 NSP
5
Negative strand replicative intermediates
4 Individual functional units 6 Genomic Positive strated RNA’s
7
Figure 34-3. Hepatitis E virus replication. Following attachment to an as-yet-uncharacterised receptor on the surface of hepatocytes, HEV is internalised (1) and then uncoated (2) in the cytoplasm. The genomic positivestrand RNA is translated into NSP (3), the nonstructural polyprotein encoded by ORF1. NSP can be processed into individual functional unites (4) including methyltransferase, protease, helicase and replicase activities. The replicase so generated can use the positive-strand RNA as a template to synthesise the negative-strand replicative intermediates (5). Then genomic (6) and subgenomic (7) positive strand RNAs, are synthesised from the negativestrand RNA intermediates. These subgenomic RNAs are translated to yield pORF1, pORF2 (capsid) and pORF3 (regulatory) proteins (8). The pORF2 and the genomic positivestrand RNA assemble into progeny virions. Release of the progeny virus through the basolateral membrane of the infected cell (10) may result in viremia and infection of adjacent hepatocytes. Transmission to new host is achieved by release of progeny into the bile canaliculi (11) and ultimately into the feces.
Subgenoic positive strand RNA’s 8 9
PORF 1,2 and 3
11 Bile Canalicali 10
lular RNA of the cultured cells and positive-strand HEV RNA was detected in the culture medium (indicating shedding of virus-like particles into the culture medium). No cytopathic effects were observed. Thereafter, using the identical culture system, primary hepatocytes were infected with non-inactivated tissue-culturederived viruses and replication of HEV RNA was demonstrated in this in-vitro model. A neutralizing anti-HEV antibody directed against the ORF2-encoded putative capsid protein blocked the infection of liver cells.238,239 HEV replication has also been reported in continuous cell cultures using cell lines derived from human lung, kidney, or liver (such as 2BS, A549, and Hep-G2).240–243 A Chinese HEV isolate has been grown in A549 cells (human lung carcinoma cells). Cytopathic effects (cell-rounding and monolayer destruction) were visible at day 2 post-inoculation and could be neutralized by specific acutephase antibodies to HEV.241 A major shortcoming of these systems has been the very low level of viral products, precluding any detailed study of viral replication. An infectious cDNA clone of HEV has been constructed in a pSGI vector (a plasmid vector). The three ORFs were amplified separately and then reconstructed to the full-length clone. The invitro-transcribed RNA of the full-length cDNA clone was infective in HepG2 cells grown in tissue culture. Viral replication was detected for six passages (33 days) using a strand-specific PCR.242
Because HEV does not replicate well in cell culture, the mechanisms of HEV pathogenesis and replication are not fully understood. Our understanding of the replication and expression of HEV is based largely on recognized conservative motives of non-structural domains and analogies with other positive-stranded RNA viruses. Studies in rats and swine suggest that extrahepatic tissues such as peripheral blood monocytes, spleen, lymph nodes, and the small intestine are involved in the replication of HEV.244,245 However the main target cells are hepatocytes. After oral ingestion, HEV particles reach the liver, where they attach to an unidentified receptor on the basolateral domain of hepatocytes, leading to virus penetration and uncoating of the genome. After uncoating, the positive-strand, polyadenylated genome of HEV is probably directly translated to yield at least pORF1, via cellular mechanisms that recognize capped RNA. Translated ORF1 is probably cleaved by viral (and perhaps cellular) proteases. The motif of a papain-like protease has been detected in the sequence of ORF1,10,245 but a functional protease has not been demonstrated. The cleavage of pORF1 polyprotein yields the replicative proteins, including a putative RdRp (viral RNA-dependent RNA polymerase that is thought to be encoded by the 3¢ region of ORF1). The RdRp copies the input genome to yield full-length minus-strand RNAs, followed by subgenomic plus-strand RNAs and full-length plus-strand RNAs (new viral genomes). The RdRp is only detectable in the early phase of
703
Section V. Liver Diseases Due to Infectious Agents
COURSE OF HEV INFECTION IN HUMAN STUDIES (Figure 34-4) The incubation period in human volunteers after oral exposure is 4–5 weeks.248,249 HEV can first be detected in stools approximately 1 week before the onset of illness and persists for an initial few weeks,249–252 but in some patients positive RT-PCR results persists for as long as 52 days.253 HEV RNA has regularly been found in serum by RT-PCR in virtually all patients in the first 2 weeks after the onset of illness.254 Except for one study,252 prolonged periods of HEV RNA positivity in serum, ranging from 4 to 16 weeks, have been reported.249,253 HEV has not been looked for in other body fluids.
ALT
IgG anti-HEV
IgM anti-HEV
HEV in blood Antibody titres or ALT levels
replication.244 Recently, the specific binding of purified and refolded RdRp to the 3¢ end of the viral genome has been reported, which requires the structural SL1 and SL2 domains and poly(A) stretch. By forming the specific binding site for viral RdRp and as a template for the synthesis of complementary-strand RNA, the 3¢ end of HEV might assume a possible role as a cis-acting element for the initiation of replication of HEV genome.246 HEV proteins are encoded in three separate ORFs, which suggests that multiple messenger RNA species would be synthesized by RdRp, in addition to full-length genomic and antigenomic strands. Indeed, early studies indicated that structural proteins might be expressed from at least two subgenomic messenger RNAs of 3.7- and 2.0-kilobase length, which are encoded by ORF2 and ORF3.9 However, the existence, origin, and functions of subgenomic RNAs need furthur evaluation. The site of HEV assembly in the cell, and the roles of proteins other than pORF2 in assembly, are poorly understood. The gene product of ORF2 has been tentatively identified as the capsid protein; it contains hydrophobic, signal-like sequences at its amino-terminal end27 as well as an arginine-rich region that probably binds the genomic RNA.9 Encapsulation of the genomic RNA results from its association with the basic capsid protein (pORF2), a process that might involve the cytoskeleton-associated ORF3 phosphorylation.42 The gene product of ORF3 also contains a hydrophobic, signal-like sequence at its amino-terminal end, but it is not known whether this small immunogenic protein is incorporated into the virion.27 Whether release of the virus from infected cells is an active process or the result of virus-mediated cell death is not known. A small amount of HEV is found in plasma during infection, consistent with the release of progeny virus through the basolateral domain of hepatocytes, leading to spread through the liver. However, most of the virus appears to be excreted through the biliary system to complete the replication cycle, consistent with the release of the virus through the apical domain of hepatocytes. Bile appears to be the principal source of HEV in the feces.237,247
HEV in stool Jaundice
Virus in stool
0
2
4
6
8
10
12
14
Weeks after exposure Figure 34-4. Time course of hepatitis E virus infection. Four to eight weeks after exposure to HEV, there is a rise in ALT and the appearance of jaundice whioch persists for 3–12 weeks. Immediately prior to the onset of clinical symptoms, HEV can be detected in the bloodstream for ~1–2 weeks and is shed in the stools for ~3–4 weeks. At the onset of clinical symptoms, HEV is lost from the bloodstream, but continues to be shed in stools. Anti-HEV IgM and IgG titres continue to increase in the asymptomatic phase. The anti-HEV IgM titre peaks during the symptomatic phase and declines thereafter to baseline values within 3–6 months of symptomatic disease. The anti-HEV IgG titre remains detectable for several years. ALT, alanine aminotransferase; HEV, hepatitis E virus; IgG, immunoglobulin G.
Table 34-7. Susceptibility of Animal Species, Based on Hepatitis E Virus (HEV) Transmission Studies Species
Cynomolgus Rhesus Tamarin Chimpanzee Pig-tailed macaque Vervet Owl monkey Squirrel monkey Patas Swine Sheep Rat Chicken
Level of susceptibility
References
Human HEV
Animal HEV and/or animal-adapted human HEV
4+ 4+ 0–3+ 0–4+ 2+
3–4+ 3–4+ ? 3+ ?
217, 220, 247, 254–261 84, 217, 236, 261–270 80, 217, 220, 247, 271 80, 217, 247, 272 217
3–4+ 3+ 1+
? ? ?
217, 220, 254, 273 217, 258 217
1+ 0–3+ 2–3+ 0–2+ ?
? 3–4+ 2–3+ ? 3+
273 204, 230, 231, 274–277 232 214, 233 84, 225
COURSE OF HEV INFECTION IN ANIMAL MODELS Attempts at transmission of HEV have helped to define the host range of this virus (Table 34-7).255–278 Among non-human primates, cynomolgus and rhesus monkeys have been consistently susceptible to infection with HEV strains obtained worldwide. Less consistent has been the response of tamarins and chimpanzees. HEV strains have been serially passaged in tamarins (Saguinus mystax) by some workers,256 but in other lab-
704
oratories, the animals have been either refractory to infection or sparingly susceptible.217,272 The same laboratory that was successful in infecting tamarins was unsuccessful in infecting chimpanzees (at least initially), whereas other laboratories reported typical hepatitis E in experimentally infected chimpanzees.217 The reasons for these discrepant results probably relate to biological differences among different strains of HEV.
Chapter 34 HEPATITIS E
Exposure to HEV is thought to occur usually via the oral route, but the virus can be transmitted parenterally, at least under experimental conditions. Infection of non-human primates via the oral route has been successful in some (but not all) studies. In one study in which quantitative data were available, the infectivity titer of HEV as measured by intravenous inoculation was at least 10 000fold higher than when administered orally.260 Infection transmitted by the intravenous route is more reproducible than the oral route. After the intravenous inoculation of HEV in cynomolgus macaques, the average incubation period for acute hepatitis is about 3 weeks. After exposure (regardless of route), the first evidence of infection with HEV is found in the liver. Shortly thereafter, virus is detected in the blood, bile, and feces.The expression of HEAg in hepatocytes, indicative of viral replication, first appears about day 7 after infection. HEAg can be detected simultaneously in hepatocytes, bile, and feces during the second or third week after inoculation, and before and concurrently with the onset of alanine aminotransferase (ALT) elevation and histopathological changes in the liver.256,257,259 The antigen can be detected in 70–90% of hepatocytes at peak expression and begins to decline after peak ALT activity has been reached. The peak shedding of virus into the blood and bile occurs before onset of clinical disease. The onset of clinical disease usually coincides with first detection of the humoral immune response, diminished replication of the virus, and beginning resolution of the infection. Both IgG and IgM class anti-HEV can usually be detected by the time liver enzymes become elevated and hepatic pathology becomes detectable (Figure 34-3). A site of replication in the intestinal tract has not been identified but is thought to exist. Nevertheless, most virus detected in the intestinal tract is probably there by way of bile.267 HEV replication in the liver is the initial event and a rise in serum ALT level and the presence of mild histologic injury at this time would be consistent with a direct cytopathic effect of the virus or an early immune-mediated effect. Later, hepatic HEVAg becomes undetectable, indicating that viral replication has stopped, and during this time the histologic changes are more pronounced, suggesting that the injury at this time is primarily immune-mediated. In support of this idea is the finding of infiltrating lymphocytes in the liver that have a cytotoxic immunophenotype.258 The delayed appearance of anti-HEV (if true and not the result of insensitivity of antibody testing) suggests that antibody is not essential for initiating hepatocyte injury but may be important in perpetuating it. It is also possible that development of an antibody response occurs independent of hepatocyte injury. In summary, the mechanism of cell death is not known, but early in the course of infection the mechanism may be predominantly direct cytopathicity and later predominantly immune-mediated. In experimental infections of non-human primates, the clinical presentation of hepatitis E is dose-dependent.260 Thus, the severity of infection is directly related to the infectivity titer of the challenge virus, and consistent demonstration of hepatitis in experimentally infected non-human primates has required challenge doses at least 1000 times greater than the minimum dose needed for infection.280,281 It is not known whether such a clinical-to-infectious dose relationship exists for naturally infected humans, but cycles of inapparent infection resulting from exposure to low doses of virus could explain how HEV can be maintained in a population with little or no clinical disease.
IMMUNE RESPONSE SPECIFIC IMMUNE RESPONSES Humoral immune responses Specific IgM and IgG immune responses to HEV occur early in the infection, usually by the onset of clinical illness. In this respect, hepatitis E resembles hepatitis A, and a serologic diagnosis can usually be made at the time of presentation of the patient (Figure 34-3). In patients with hepatitis E, IgM anti-HEV begins to develop just before the peak of ALT activity and reaches a maximal titer around the time of maximal ALT activity. IgM anti-HEV disappears about 4–5 months into the convalescent phase of the disease.282 Of samples of sera collected from patients during various outbreaks of hepatitis E at 1 and 40 days, at 3 and 4 months, and at 6 and 12 months after the onset of jaundice, 100%, 50%, and 40%, respectively were positive for IgM anti-HEV.283 IgG antibodies develop shortly after the IgM antibodies, and the titers increase throughout the acute phase into the convalescent phase, remaining high from 1 to 4.5 years after the resolution of the illness.282,284 IgG anti-HEV appears to diminish in titer at a more rapid rate than does antibody to hepatitis A, raising questions about the duration of immunity following acute hepatitis E. Anti-HEV has been detected as long as 13–14 years after infection;73,285 however the possibility of repeated exposure cannot be ruled out. Antibody responses to individual viral antigens are highly variable, due to both strain-specific differences in some epitopes and differences in response to single antigens between individual patients. For example, pORF3 varies greatly between strains, and many experimentally infected animals and some patients fail to develop antibodies to ORF3 protein.71,72 This variable reactivity contributes to the poor sensitivity and concordance of HEV diagnostic tests based on such antigens.73 In contrast, the 55–63-kDa antigens expressed with a baculovirus system and the ORF 2.1 protein expressed in E. coli detect specific IgG and IgM responses in the great majority of patients. After reaching high levels during the acute phase, HEV pORF2-specific IgG declines rapidly over 6–12 months and might not persist at protective levels for life. Conversely, the responses to pORF3 are highly variable, with a proportion of patients mounting no detectable response to the antigen while others maintain reactivity to pORF3 for many years.286 Anti-HEV of the IgA class (as a correlate of mucosal immunity) has also been detected in the serum of about 50% of naturally infected individuals.287 These antibodies rapidly decline to undetectable levels and the significance of such antibodies is unknown. Since passive immunization with IgG appears to be sufficient for protection, it is likely that IgA is not essential. All isolates of HEV are serologically related, and convalescent antibody produced in response to infection with one strain of HEV probably protects against subsequent exposure to all other strains.15,18,217,248,271
Cell-mediated immune responses Little is known about the cell-mediated immune response to HEV in humans. Recently, Naik et al.288 provided the first evidence of cellular immune responses in patients with acute HEV infection. They examined lymphoproliferative responses to peripheral blood
705
Section V. Liver Diseases Due to Infectious Agents
mononuclear cells in the presence of a small panel of HEV ORF2and ORF3-related 14–23-amino-acid peptides in 22 healthy adult subjects and in 21 patients with acute uncomplicated hepatitis E. They found that responses occurred more often in acute hepatitis E patients than in controls. The results showed that lymphocytes of patients with acute hepatitis E show sensitization to HEV peptides.288 These observations need to be further evaluated using larger number of peptides spanning the entire structural genes of HEV to define the T-cell epitopes.
PATHOLOGY Much of what is known about the pathology of acute HEV infection has been obtained from studies of ET-NANBH epidemics occurring in developing countries.289,290 The morphologic findings are of two main types: (1) a typical acute hepatitis picture; and (2) a cholestatic variant. In the latter, prominent features include bile stasis in canaliculi, gland-like transformation of hepatocytes, and extensive proliferation of small bile ductules. There is also prominent cholestasis in the centroacinar zone. Degenerative changes in hepatocytes and focal areas of necrosis are less frequent than in the non-cholestatic type. Kupffer cells that contain lipofuscin granules are prominent. Portal tracts are expanded; polymorphonuclear leukocytes are conspicuous in the portal tract infiltrates, but lymphocytes predominate. Phlebitis of portal and central veins may be seen. Intralobular infiltrates consist mainly of polymorphonuclear leukocytes and macrophages. With the non-cholestatic type of HEV infection, focal hepatocyte necrosis, ballooned hepatocytes, acidophilic degeneration of hepatocytes, and acidophilic body formation are frequent. An important morphologic feature is focal intralobular areas of hepatocyte necrosis with prominent accumulations of macrophages and activated Kupffer cells in the presence of lymphocytes. The histologic severity of the hepatitis is variable, but in one well-documented epidemic, 78% of biopsy specimens were graded as at least moderately severe.291 In fatal cases, severe acute hepatitis with submassive or massive hepatocyte necrosis is observed. No chronic histologic manifestations have been described. In acute but non-fulminant cases of HEV infection, electron microscopy reveals considerable hepatocyte polymorphism. Some hepatocytes show ballooning degeneration and vesiculation of the perinuclear envelope and rough endoplasmic reticulum, whereas other hepatocytes show shrinking and condensation of cytoplasm
and cell organelles to form a web-like pattern. The bile canaliculi are dilated, and intracanalicular and intracytoplasmic bile stasis is seen. In fulminant cases of HEV infection, hepatocytes show extensive organelle damage.
CLINICAL FEATURES ACUTE INFECTION The incubation period after exposure to HEV has ranged from 2 to 10 weeks during waterborne outbreaks associated with short and well-defined periods of water contamination.76,168
ACUTE ICTERIC HEPATITIS Acute icteric hepatitis is the most common recognizable form of illness associated with HEV infection. Two phases of illness have been described: (1) a prodromal and preicteric phase; and (2) an icteric phase. The prodromal phase, lasting about 1–4 days, is characterized by a variable combination of flu-like symptoms, including fever, chills, abdominal pain, anorexia, nausea, aversion to smoking, vomiting, diarrhoea, arthralgias, asthenia, and urticarial rash.1,291–294 These symptoms are followed within a few days by the appearance of jaundice. The onset of the icteric phase is usually heralded by the appearance of darkening of urine, and may be accompanied by itching or lightening of stool color (Table 34-8). With the onset of jaundice, the fever and other prodromal symptoms tend to diminish and may entirely disappear. The exception is gastrointestinal symptoms which may persist for a longer time. The physical examination reveals jaundice and a mildly enlarged, soft, and slightly tender liver. Splenomegaly is seen in about one-fourth of patients.291 In acute hepatitis, classical biochemical abnormalities are seen. Laboratory test abnormalities include a variable rise in serum bilirubin (predominantly conjugated), markedly elevated aminotransferases, and g-glutamyltransferase levels, and a mild rise in serum alkaline phosphatase activity. An elevated ALT often precedes the onset of symptoms by as much as 10 days and reaches a peak by the end of the first week, coinciding with the onset of the icteric phase. As the illness subsides, the ALT levels decrease significantly, followed by a decrease in serum bilirubin levels which are usually normal by 6 weeks.291 HEV RNA can be detected in the stool for up to 10 days after the onset of the icteric phase, although viral shedding for up to 52 days after the onset of icterus has been described.253 HEV RNA can
Table 34-8. Clinical Findings in Hepatitis E Outbreaks Symptoms and signs Jaundice Malaise Anorexia Abdominal pain Hepatomegaly Nausea, vomiting Fever Pruritus
706
Delhi, India, 1956361 (%) (n = 958)
Accra, Ghana, 1963362 (%) (n = 136)
Kashmir, India, 19781 (%) (n = 275)
Ethiopia, 1989307 (%) (n = 423)
Xinjiang, China 1986–1988306 (%) (n = 85)
Abbottabad, Pakistan, 1988363 (%) (n = 109)
100 — 66 63 62 29 23 14
100 95 95 37 67 48 57 47
100
100 100 100 82 10 100 97 14
91 95 69 55 80 91 53 59
94 — 94 56 — 81 46 16
79 41 85 46 44 20
Chapter 34 HEPATITIS E
be detected in the serum during the preicteric phase and before the detection of virus in stool but becomes undetectable after the peak in serum aminotransferase activity.249 The detection of HEV RNA in the serum during the preicteric phase suggests that sporadic transmission of the virus may occur via the parenteral route. The IgM antibody to HEV becomes detectable just before the peak ALT activity; peak antibody titers occur at approximately the same time as peak ALT levels and decline rapidly thereafter. In the majority of patients, IgM anti-HEV is undetectable 5–6 months after the onset of illness. IgG anti-HEV is detectable shortly after IgM anti-HEV becomes detectable, increases in titer throughout the acute and convalescent phases of infection, and remains detectable in most patients 1 year after acute infection289 (Figure 34-3). The duration of the IgG anti-HEV response is unknown, but high titers have been measured up to 14 years after acute infection.285 The duration of protective immunity is not known; however, in the short term, there appears to be protection from reinfection.290 In non-fatal cases, acute hepatitis is followed by complete recovery without chronic sequelae. No evidence of chronic hepatitis or cirrhosis was detected in patients who were followed clinically or underwent liver biopsy after acute hepatitis E.291,292
PROLONGED CHOLESTATIC HEPATITIS In a few patients, the course is prolonged, with marked cholestasis, persistent jaundice, and itching. In these cases, laboratory tests show a rise in alkaline phosphatase and a persistent rise in bilirubin, even after the transaminases have returned to normal.293 The prognosis is good because the jaundice resolves spontaneously after 2–6 months. Recurrent (bimodal) hepatitis E has not been reported, except in experimentally infected non-human primates.294 In contrast, recurrent hepatitis A is relatively common.295
ANICTERIC HEPATITIS AND ASYMPTOMATIC INFECTION Non-specific symptoms may develop in some infected persons, resembling those of an acute viral febrile illness without jaundice (anicteric hepatitis). Liver involvement in these patients is only recognized after laboratory studies are performed. In its most benign form, HEV infection is entirely inapparent and asymptomatic and passes unnoticed. The exact frequency of anicteric hepatitis and asymptomatic infection is unknown, although it probably far exceeds that of icteric hepatitis. In areas of endemicity, anti-HEV is present in a large proportion of the population; most of these seropositive individuals do not recall having had jaundice and may have had an asymptomatic or anicteric infection. In most of the outbreaks of hepatitis E, the highest rate of clinically evident disease is among persons 15–40 years of age1,78,92,296 – a pattern that contrasts with HAV infection, in which children have the highest attack rates.289 The lower rate of disease among children may be the result of a higher frequency of asymptomatic or anicteric HEV infection in this age group.
LABORATORY DIAGNOSIS Diagnosis of HEV infection in individual patients remains problematic. Assays used in research laboratories are of high sensitivity and
specificity, but routine commercial diagnostic assays are not available in many poor endemic countries and even those available are of questionable sensitivity and specificity. Tests for the diagnosis of HEV infection are based on either the detection of virus or viral components or on serological tests determinig a virus-specific immune response in the host.
VIRUS OR VIRAL COMPONENT DETECTION Virus isolation is not appropriate for HEV because it is largely refractory to routine isolation in cell culture.The first assay for detection of HEV infection used IEM to detect viral particles in feces.3,248,273 This provides a specific diagnostic marker but has very poor sensitivity and is technically demanding. Consensus oligonucleotide primers have now been developed to amplify regions in the HEV genome supposed to encode the helicase, polymerase, or parts of the 3¢ end of ORF2.52,204 These RT-PCR assays are specific and detect HEV RNA in acute-phase sera (or peripheral blood monocytes in blood) and stool as well as in contaminated water and sewage.57 However, a positive result for HEV RNA by RT-PCR is not the equivalent of infectivity. Primers located in highly conserved regions of the genome or degenerate primers must be used to ensure detection of all recognized variants of HEV. Although amplification and detection of HEV RNA from serum, liver, or stool using RT-PCR can be used to diagnose HEV infection, this methodology has been used primarily for research purposes. Quantitative assays for HEV RNA are not available. HEV RNA is also not useful to diagnose acute HEV infection as it is undetectable in the symptomatic phase of the illness. Hepatitis E antigen (HEAg) has been detected in liver tissue with an immunoflourescent probe prepared from convalescent-phase serum.57,297 Because a liver biopsy is needed for this test, it is not useful for clinical diagnosis; but it has been used in experimental studies of HEV infection in primates.
SEROLOGICAL ASSAYS Tests based on detection of anti-HEV include: (1) IEM; (2) fluorescent antibody-blocking assays; and (3) enzyme immunoassays (EIA).
IMMUNE ELECTRON MICROSCOPY AND IMMUNE FLUORESCENT MICROSCOPY (IFM) IEM, the first serological assay to be developed, detects VLPs in clinical specimens.3 HEV particles are precipitated with native antibody to HEV derived from acute- or convalescent-phase sera, which bind to HEV antigens on the surface of viral particles and cause them to aggregate. Anti-HEV concentrations can be determined semiquantitatively by rating the antibody coating. IFM-blocking assays detect semiquantitatively antibodies that react against HEV antigen. This test is based on the ability of anti-HEV in the test serum to block the binding of fluorescein-conjugated anti-HEV IgG to HEV antigen in hepatocytes in frozen liver tissue of experimentally infected cynomolgus macaques.3,297 Although highly specific, this assay had a sensitivity rate of only 50–70% in patients with acute hepatitis during outbreaks of HEV.298 Both IEM and IFM are laborious and expensive and thus not useful for routine diagnosis.
707
Section V. Liver Diseases Due to Infectious Agents
ENZYME IMMUNOASSAYS Antigenic domains have been identified within all three ORFs of HEV. Twelve antigenic domains have been identified throughout ORF1 (particularly in the region of putative RdRp), six antigenic domains within the ORF2 protein, and three within the ORF3 protein.299 Recombinant proteins and synthetic peptides were used to develop solid-phase EIAs to detect IgM and IgG class antibodies to HEV. The antigens are labelled on microtiter plates (EIAs) or stripes (Western blots). Most of these assays use recombinant proteins derived from the C-terminal or large portions of ORF2 and/or the C-terminal or full length of ORF3. Different mechanisms proposed for the antigenic activity of recombinant proteins used in EIAs include: 1. Recombinant proteins or synthetic peptides can represent the linear epitopes. 2. Larger recombinant proteins may have conformational antigenic determinants related to their secondary or tertiary structure. 3. Capsid-like particles originating from a recombinant 111amino acids N-terminal truncated ORF2 protein mimic the three-dimensional structural features of VLPs and probably have an immunogenic activity represented by their quaternary structure.300 Seroprevalence studies have indicated that HEV EIAs based on large antigens expressed from ORF2 or capsid-like particles are superior to those based on short sequences of ORF2 or antigenic epitopes of the ORF3 in detecting convalescent-phase anti-HEV.301 Limited data are available on the comparability of the different methods of antibody detection. The serologic assays utilize different target antigens from different HEV strains and different expression systems for the production of the recombinant proteins, factors that may affect the performance characteristics of the assay. ORF2derived antigens expressed from baculovirus in insect cells and, to a lesser extent, antigens expressed in E. coli have yielded the best serologic tests, whereas synthetic peptides have yielded tests that have been relatively insensitive.284 The assays that use recombinant proteins have greater sensitivity than the fluorescent-blocking assays and detect 80–100% of cases during outbreaks of acute hepatitis E. Assays for detecting IgG anti-HEV and IgM anti-HEV are available for distinguishing acute from past infection. Overall concordance of the different tests for determining the seroprevalence of anti-HEV in a non-diseased population is low. Pairwise comparision of 12 test sets showed a concordance in blood donor sera ranging from 41% to 94% (mean 68%), and a concordance among reactive sera from 0% to 89% (mean 32%).73 EIAs based on synthetic peptides could not detect convalescent-phase anti-HEV reliably. They are used mainly to confirm positive test results from EIAs based on recombinant proteins and to exclude non-specific reactivity. The two EIAs that are available commercially today are the Genelabs EIA and the Abbott EIA. The Genelabs EIA uses four short recombinant proteins derived from the 3¢ terminal end of ORF2 (42 aa) and ORF3 (33 aa) from Burmese and Mexican prototype sequences. The Abbott EIA uses two recombinant proteins derived from complete ORF3 (123 aa) and from a sequence of ORF2 (327 aa) from the Burmese prototype strain. The speci-
708
ficity and sensitivity of these tests for detecting convalescent-phase IgG have not been precisely established, thus limiting the reliability of the results from seroepidemiological studies.
DIAGNOSIS OF HEV INFECTION IN AREAS OF HIGH PREVALENCE Although the titer of HEV-specific IgG tends to decline in the first year after infection, this relationship is not reliable and so cannot form the basis of differential diagnosis.145,301 In endemic countries, a large number of patients have antibodies from past infections, so the detection of IgG anti-HEV is of little use for diagnosis of acute HEV infection. Detection of anti-HEV IgM must be the method of choice for the detection of acute HEV infection in these areas. In endemic areas, the assay sensitivity is of prime concern. Even in research laboratory settings, the current Genelabs IgM assay may fail to detect around 40% of patients with acute HEV infection,302 and may be unsuitable for use in endemic areas. HEV IgM assays based on recombinant antigens expressed in the baculovirus system or the ORF2.1 antigen expressed in E. coli appear to have improved sensitivity, and may be appropriate in endemic settings.38,286,300,302 Since HEV accounts for a substantial proportion of the acute sporadic hepatitis in endemic countries, the specificity of assays is a less important issue. For example, if the false-positive rate of an assay is 2%, only 1 in 35 hepatitis patients in endemic areas might be misdiagnosed as having acute HEV. Higher speceficity would be advantageous and one IgM assay has been reported with a false-positive rate of 0.1%.300
DIAGNOSIS OF HEV INFECTION IN AREAS OF LOW PREVALENCE In areas with a low incidence of clinical HEV infection, assay specificity will have a large impact on the predictive value of HEV serological tests. Assays for HEV-specific IgG have considerable value for the diagnosis of acute hepatitis in travelers returned from endemic areas, among whom the incidence may be much higher than the background rate of reactivity in the healthy population. In a recent study from Taiwan, the sensitivities of IgG and IgM antiHEV assays from Genelabs diagnostics for diagnosis of acute hepatitis E (relative to RT-PCR) were 86.7 and 53.3% respectively. The specificities of these tests were 92.1 and 98.6% respectively, thus concluding that IgG anti-HEV testing has fairly good specificity and sensitivity in detecting acute HEV infection. This assay has good concordance with HEV RNA testing by RT-PCR. While IgG antiHEV test can be used to screen for acute hepatitis E in non-endemic areas,303 the method of choice is IgM HEV. However, the current Genelabs IgM anti-HEV assay has a false-positive rate of 3% in US blood donors.304 It has also been demonstrated that the antigens used in this assay may fail to detect about 40% of acute HEVinfected patients.302 Acute infection diagnosed by the presence of HEV RNA in serum and stool, in the absence of a detectable antibody response, is well described and may be related to the timing of the blood samples in relation to symptomatic infection, as the IgM anti-HEV response may be short-lived or biphasic.253 Alternatively, lack of an antibody response may represent a defective immune response by the host.
Chapter 34 HEPATITIS E
DIFFERENTIAL DIAGNOSIS Acute hepatitis E cannot be clinically distinguished from other forms of acute viral hepatitis. Due to their identical clinical presentation and modes of transmission, hepatitis E may be suspected in the same circumstances as hepatitis A. While hepatitis A is more common than hepatitis E in the developed countries, in endemic areas, HEV is often the most common cause of acute hepatitis. In a patient with symptoms and biochemical evidence of acute hepatitis, serologic tests for excluding acute hepatitis A (IgM anti-HAV), B (HBsAg, IgM anti-HBc), C (anti-HCV), cytomegalovirus, and Epstein–Barr virus are obtained. In non-endemic areas, suspicion of acute HEV should be heightened by a history of travel to areas endemic for HEV. In endemic areas, an outbreak may be associated with a common contaminated water source, and such information should be sought. In endemic areas, infection with HEV may be seen in association with other hepatotropic viruses (A, B, and C), and in the absence of specific anti-HEV testing, the diagnosis of acute HEV coinfection or superinfection may be missed.
NATURAL HISTORY
Table 34-9. Hepatitis E Virus (HEV) as the Cause of Fulminant Hepatic Failure in Eastern Countries Author
Year
Number of cases
% due to HEV
Reference
Acharya et al. Khuroo Jaiswal et al. Arora et al. Khuroo and Kamili Bendre et al.
2000 1997 1996 1996 2003
458 119 95 44 (children) 180
23 48 41 40.9 43.9
309 310 311 312 313
1999
36 (children)
314
Nanda et al.
1994
50 (fulminant non-A, non-B cases)
11.1 (with HAV) 62
315
HAV, hepatitis A virus. (Modified from Aacharya SK, Batra Y, Hhazari S, et al. Etiopathogenesis of acute hepatic failure: eastern versus western countries. J Gastroenterol Hepatol 2002; 17 (suppl 3):S268–S273308.)
Table 34-10. Hepatitis E Virus (HEV) as the Cause of Fulminant Hepatic Failure in Western Countries
In non-fatal cases, acute hepatitis is followed by complete recovery without chronic sequelae. There appears to be protection from reinfection for a time, but the duration of this protection is unknown. Long-term serologic studies will be needed to determine the duration of protective immunity and the nature of the anamnestic response.
Center
Year
No. of cases
% due to HEV
Reference
King’s College Hospital, UK UK France USA Denmark
1990
943
0
316
1994 1990 1999 1990
342 502 295 160
3 2 0 0
317 317 317 317
COMPLICATIONS
(Modified from Aacharya SK, Batra Y, Hhazari S, et al. Etiopathogenesis of acute hepatic failure: eastern versus western countries. J Gastroenterol Hepatol 2002; 17 (suppl 3):S268–S273308.)
MORTALITY AND FULMINANT HEPATIC FAILURE The case-fatality rate in reports based on hospital data has ranged from 0.5% to 4%.296,305 However, the hospital-based data may overestimate mortality. Studies based on data obtained from population surveys during outbreaks have reported lower mortality rates, ranging from 0.07% to 0.6%.78,296,306 In an epidemic among army personnel in Ethiopia, none of the 423 patients with icteric hepatitis developed FHF or died.307 In a small proportion of patients, the disease is more severe and associated with subfulminant or FHF, which can be rapidly fatal. In regions of endemicity, HEV infection is an important cause of FHF (Table 34-9).308–315 In contrast, HEV has not been practically associated with occurrence of FHF in various western reports (Table 34-10).308,316,317
ACUTE HEV SUPERINFECTION IN PATIENTS WITH CIRRHOSIS Acute HEV superinfection in patients with cirrhosis produces decompensation and is associated with a high mortality rate. FHF resulting from HEV and coexistent Wilson’s disease has been reported.318 There is a recent report from Pakistan showing decompensation in 4 chronic liver disease patients due to HEV. Seroprevalence of HAV and HEV patients with stable chronic liver disease was 97.8% and 17.7% respectively. Seroprevalence in controls was 94% and 17.5% respectively for HAV and HEV. There-
fore, the large majority of adult chronic liver disease patients in endemic countries are vulnerable to infection with HEV, but are protected against hepatitis A. This is a group that would be ideal candidates for an HEV vaccine.319 In another recent retrospective study of patients with chronic liver disease and acute icteric hepatitis E, nine patients with chronic liver disease were found to have superinfection with HEV. Out of these, six patients died of liver failure. The seroprevalence of hepatitis A was 99 and 100% and HEV was 56 and 21% in cases and controls, respectively, implying that 44% of patients with chronic liver disease were at risk of developing hepatitis E.320
PREGNANT WOMEN Pregnant women, particularly those in the second and third trimesters, are more frequently affected during hepatitis E outbreaks and have a worse outcome. In fact, the major cause of mortality in epidemics is the high rate of FHF in pregnant women. Mortality rates among pregnant women, especially those infected in the third trimester, range between 15 and 25%.79,80,305,321 In the south Xinjiang epidemic, the maternal mortality rate was 1.5% for infections in the first trimester of pregnancy, 8.5% for those in the second trimester, and 21% for those in the third trimester.77 The frequency of fetal death in utero and immediately after birth is also increased.
709
Section V. Liver Diseases Due to Infectious Agents
In an epidemic in Kashmir, India, attack rates among those in the first, second, and third trimesters were 8.8, 19.4, and 18.6%, respectively, as compared with 2.1% among non-pregnant women and 2.8% among men.321 Further, FHF developed in 22.2% of the affected pregnant women, in comparison with 2.8 and 0% of affected men and non-pregnant women, respectively. Frequency of abortions, stillbirths, and neonatal deaths is also increased among pregnant women with HEV infection above that seen with acute viral hepatitis of other etiologies.79 HEV has not been associated with congenital abnormalities. In a recent report on sporadic acute viral hepatitis in pregnancy from Kashmir, India the proportion of pregnant women in HEV group was 31.7% and 5.3% in non-HEV group. HEV was responsible for 85.5% of acute viral hepatitis cases in pregnant women, as compared to 41.5% in non-pregnant women. The rate of FHF from hepatitis due to HEV was 69.2% in pregnant verses 10% in non-pregnant women. The rate of FHF in HEV cases was 40%, 68.8%, 71%, and 100% during first, second, third trimesters, and puerperium respectively, with mortality rates of 30%, 43.8%, 38.7%, and 37.5% during the same four periods. Higher mortality rates could be due to non-availability of liver transplant facilities in that center.322 In a recently conducted study by the author’s group, 144 pregnant women with median gestational age of 32.2 weeks and presenting with acute viral hepatitis were studied. Sixty-nine percent were in the third trimester and 31% in the second semester. The etiology of viral hepatitis was HEV in 58.2%, HBV in 32.6%, HAV in 0.6%, HCV in 7.6% and co-infection in 1.3%. The maternal complications were coagulation defects (56%), postpartum haemmorhage (20%), acute renal failure (18%) and gastrointestinal bleeding (12.5%). Of the 144 women, 61 (42.3%) had FHF (HEV 77%, HBV 23%). Of 61 women, 41 (67%) died (HEV 90% and HBV 9.8%) and maternal morbidity was two times higher in the third compared to the second trimester (28.2% vs. 13.1%). The perinatal mortality rate was 56.2%. There were 37 (25.6%) live births, 59 (40.9%) still births and 22 (15%) neonatal deaths. The perinatal mortality was 100% in FHF. Perinatal mortality was higher in women with HEV compared to HBV (65.4% vs. 30.8%). Thus, HEV was found to be the most common cause of viral hepatitis and FHF in pregnancy in India and was associated with a very high maternal-perinatal morbidity and mortality, which increases as the gestation advances.323 Pregnant women with FHF have shorter duration of disease, shorter pre-encephalopathy period, lower serum bilirubin levels, and more frequent development of cerebral edema than non-pregnant women with FHF. Therefore, HEV in pregnant women can be an explosive disease with rapid progression of symptoms, development of encephalopathy, and cerebral edema. Also occurrence of disseminated intravascular coagulopathy is one of the most remarkable features of FHF caused by HEV in pregnant women.322 The pathogenesis of disseminated intravascular coagulopathy in pregnant women with FHF is not known; however it may give an important clue to the pathogenesis of the increased severity of HEV in pregnant women.
OTHER COMPLICATIONS Other reported complications include: Guillain–Barré syndrome,324 severe hemolysis and renal failure in glucose-6-phosphate dehydro-
710
genase-deficient patients,325 and acute pancreatitis, which can be severe.326
TREATMENT Like other forms of acute viral hepatitis, the mainstay of therapy is to monitor for the development of complications and provide good nutrition. The HEV infection is usually self-limited, and no specific therapeutic interventions are required. Rarely, if fulminant hepatitis develops, intensive management and options of liver transplantation should be considered (see Chapter 21).
PREVENTION GENERAL MEASURES Improved sanitation is important in controlling diseases that have fecal–oral transmission as a prominent part of their epidemiology. Industrialized countries with a generally high level of public sanitation do not experience epidemics of waterborne hepatitis E or significant endemic or sporadic disease, although hepatitis A continues to be an important cause of clinical disease in some of these countries, for example, the USA. In developing countries, even where the prevalence of antibody to HAV is extremely high, the prevalence of antibody to HEV is much lower.116,119,120,123,126,326 These observations might suggest that HEV is either much less readily spread or much less stable than HAV in the environment or that other factors are involved. It is difficult to assess the importance of animal reservoirs for the maintenance of HEV in the environment, but this is probably involved. Effective prevention relies primarily on maintaining a clean drinking-water supply and paying strict attention to sewage disposal, because immunoprophylaxis is not currently available. Many epidemics in developing countries have occurred due to leakage of sewage pipes into municipal water supply pipes laid in the same or adjacent trenches. A barrier between these two supplies is essential for long-term prevention. During an epidemic, steps to improve water quality can lead to rapid abatement of the occurrence of new cases.327 During an epidemic in India, failure to chlorinate water was followed by a rapid rise in the number of cases, and reinstitution led to rapid abatement of the epidemic.78 Since the virus is transmitted by the fecal–oral route, it must withstand exposure to bile salts during excretion and low pH during ingestion, but it is generally considered to be more labile than hepatitis A. Although limited studies are available, boiling water appears to inactivate HEV effectively.328 Chlorination may be ineffective in the presence of large amounts of organic matter, for example, in water contaminated with feces. Travelers to endemic areas must take precautions against the consumption of contaminated water. They should be advised to avoid drinking water or beverages from sources of unknown purity. Only boiled or bottled water should be used. Although infection via food is much less common for hepatitis E than for hepatitis A, it is important to maintain caution about contaminated food and avoid eating uncooked shellfish, or eating uncooked and unpeeled fruits and vegetables. Women should avoid unnecessary travel to endemic areas during pregnancy. Isolation of affected persons is not indicated as
Chapter 34 HEPATITIS E
person-to-person transmission is uncommon.174 However, infected persons should not be involved in food preparation or handling until their symptoms have fully resolved.
PREVENTION OF PERINATAL AND CONGENITAL INFECTION Efforts should be focused on protection of pregnant women against infection. Vertical transmission of HEV with severe hepatitis can occur in the infant.
PASSIVE IMMUNOPROPHYLAXIS Although four genotypes of HEV have been identified, only one serotype has been described.47 In experimental studies in primates, passive transfer of anti-HEV has been shown significantly to reduce virus-shedding in feces and abrogate disease when given to nonhuman primates challanged with a high dose of homologous HEV.280 Also, a recombinant vaccine consisting of truncated ORF2 protein from the Sar55 strain genotype elicited a humoral immune response in macaques that correlated with protection from hepatitis E regardless of whether the animals were challanged with a high dose of homologous (genotype I) or heterologous (genotype II or III) virus.281,329,330 These data suggest that immunoglobulin (IG) preparations similar to those used for protection against hepatitis A would be efficacious against hepatitis E. However, no reduction in disease rates could be shown in pre- or postexposure prophylaxis studies among recipients of IG preparations manufactured in hepatitis Eendemic areas.331,332 In a recent study, administration of immune serum globulin to pregnant women during an outbreak was shown to reduce the number of total new HEV infections, although the number of clinical cases was unchanged.333 One study failed to demonstrate protection of non-human primates after they were administered with convalescent serum obtained from a volunteer who had been experimentally infected with HEV 4 years earlier, but the authors failed to demonstrate anti-HEV in the infused monkeys at the time of challenge with HEV and they employed a very large but unquantified intravenous dose of challenge virus.334 Such a study may have little relevance to predicting the efficacy of IG with a high titer of anti-HEV administered to individuals with exposure to relatively small doses of HEV acquired by the natural route of infection. In contrast, infusion of non-human primates with convalescent serum or plasma from other primates experimentally infected with HEV has protected them against hepatitis E.280,335 Failure of IG preparations to protect against HEV in humans may reflect the low titers of IgG anti-HEV in IG preparations derived from persons living in endemic areas. The geometric mean titer of anti-HEV in IG from endemic areas such as India is < 1:1000 in the general population. In contrast, the titer of anti-HEV used for passive immunization in studies with cynomolgus monkeys was 1:10 000.280 IG prepared in non-endemic countries would be expected to have even lower levels of anti-HEV and hence would be of little benefit as pre-exposure prophylaxis for travelers to endemic areas. Therefore, pooled normal human IG is unlikely to be useful as an immunoprophylactic agent against HEV and the protective role of anti-HEV antibodies in humans requires further study. The occurrence of large hepatitis E epidemics among adults in disease-endemic areas suggests either that anti-HEV antibody
may not be fully protective or that antibody levels decline with time and gradually reach an unprotective level. As an alternative, an IG preparation consisting of HEV-neutralizing monoclonal antibodies (MAbs) might protect against hepatitis E. Phage display of antibody libraries has provided a powerful tool for the isolation of human MAbs to important viral pathogens. Large repertoires of antibodies can be displayed on the surface of filamentous phage particles, and antibodies with desired specificity can be isolated.336 Recently, Schofield et al.337 characterized 14 MAbs against the ORF2 protein of the Sar55 strain of HEV by phage display from a cDNA library of chimpanzee g1/k antibody genes. Pairwise competition assays with the panel of the 14 chimpanzee MAbs suggested that there were two major antigenic sites on the protein and that each contained distinct non-overlapping epitopes linked by overlapping epitopes. Four of the antibodies have been tested for their ability to neutralize HEV. The two antibodies (HEV #4 and #31) that neutralized HEV recognized overlapping epitopes of antigenic site 2 and mapped to the C-terminal region. The two antibodies (EBL2 and EBL89) that did not neutralize HEV mapped to antigenic site 1 at the amino-terminal end of the protein. Clinical trials of MAbs have not been done and further work is necessary before a definitive role of immune serum globulin can be discerned.
ACTIVE IMMUNOPROPHYLAXIS Rationale for the Development of a Hepatitis E Vaccine HEV infection is a major cause of morbidity and high mortality among infected pregnant women and hence a vaccine would be of immense benefit. Serum antibodies to HEV develop in response to naturally acquired and experimentally induced HEV infections in cynomolgus monkeys.280 IgG anti-HEV has been shown to protect against hepatitis E, though to a limited extent.290,338 Passive immunoprophylaxis studies performed in macaques also suggest that antibodies to hepatitis E could protect against hepatitis E.280 These data, along with evidence that antibody to HEV is broadly reactive against different genotypes of the virus, suggest that a vaccine might be able to prevent hepatitis E.73 The HEV genome contains three ORFs. The ORF1 encodes nonstructural protein(s) and, therefore, ORF1 protein(s) would not be a target for humoral immunity. ORF3 overlaps with ORFs 1 and 2 and encodes a small protein of unknown function but significant antigenicity. However, antibodies to ORF3 did not neutralize virus in an in-vitro assay whereas antibody to ORF2 did.239 ORF 2 encodes the capsid protein and, since the ORF2 protein is the major protein in the virion, it has been the focus of vaccine development. The various options for a vaccine include attenuated or killed virus vaccine, recombinant protein-based vaccine and nucleic acid-based vaccines. The development of an attenuated or killed virus vaccine is not currently feasible, as an efficent cell culture system for HEV does not exist. Therefore, either a recombinant protein-based or nucleic acid-based vaccine is needed. Several antigenic regions of diagnostic relevance have been found within HEV pORF1, pORF2, and pORF3 by using synthetic peptides of different sizes,299,339–342 and recombinant proteins.22,71,72 The HEV neutralization epitopes have not yet been conclusively identified.
711
Section V. Liver Diseases Due to Infectious Agents
There are two approaches to define neutralization epitopes. One well-established approach is by analysis of virus mutants resistant to neutralizing MAbs. However, this is not applicable to HEV because of the absence of characterized HEV mutants, although neutralizing antibodies were recently prepared by phage display.337 A PCR-based in-vitro neutralization assay was developed.343 Antibodies against the HEV recombinant C2 protein comprising the Cterminal two-thirds (225–660 aa) of the HEV Burma pORF2344 were found to neutralize the HEV Burma, Mexico, and Pakistan strains by using the in-vitro neutralization assay.345 This finding suggested that HEV neutralization epitope(s) might be located within the C-terminal part of its pORF2. In a recent study the neutralization epitopes of HEV were studied by an in-vitro neutralization assay using antibodies obtained by immunization of mice with 51 overlapping 30-mer synthetic peptides spanning the region 221–660 aa of the pORF2 and 31 overlapping recombinant proteins of different sizes derived from the entire pORF2 of the HEV Burma strain. Antibodies against synthetic peptides and short recombinant proteins of about 100 amino acids did not neutralize HEV, suggesting the HEV neutralization epitope(s) is conformation-dependent. However, one recombinant protein of about 400 amino acids in length comprising the pORF2 sequence at position 274–660 aa as well as all truncated derivatives of this protein containing region 452–617 aa elicited antibodies, demonstrating HEV-neutralizing activity. The minimal size fragment (designated pB166) that can efficiently model the neutralization epitopes is of 166 amino acids and located at position 452–617 aa of the HEV pORF2.346 In a recent work, the relationship of the dimeric form of recombinant peptide (E2) comprising amino acid 394-606 of the capsid protein of HEV, was analysed using three murine monoclonal antibodies, 8C11, 13D8 and 8H3 (all of which react predominantly against the E2 dimer). Cross-blocking patterns between these antibodies discerned two spatially separate antigenic domains, one identified by 8C11 and 13D8, and the other, by 8H3. Kinetic studies suggested that the epitope to which 8H3 is directed is partially masked, and thus has limited access by the native antibody. However, this was not the case with the smaller Fab. Access to the 8H3 epitope was enhanced by the binding of 8C11, and inhibited by the binding of 13D8 to a distal site on the peptide. Similar to the effects of binding 8H3 to E2, 8C11 was found to enhance immune capture by 8H3, while 13D8 was inhibitory. Also, 8C11 and 8H3 act synergistically to neutralize HEV infectivity. The parallel cross-reaction patterns that these antibodies exhibit against the peptide and the virus, respectively, implicate two interacting conformationally dependent neutralization sites on the HEV particle. These sites might cooperate in the absorption and penetration of the HEV virus.347 Recently, many reports have indicated that the HEV ORF2 protein formed VLPs when expressed in insect cells.39,300 Since virions are the target for neutralizing antibodies, it is logical to assume that VLPs, which model virion macrostructure, would also efficiently model the neutralizing epitope(s) and, therefore, efficiently elicit neutralizing antibodies. In fact, recent experimental evidence has indicated that HEV VLPs are very potent immunogens affording protection against both homologous and heterologous challenge in non-human primates.280,281 The different recombinant
712
vaccines include prokaryote (E. coli) or eukaryote (insect cellderived and plant-derived) vaccines. E. coli-Derived Vaccine. The first candidate vaccine consisted of a recombinant fusion protein comprising tryptophan synthetase and the carboxy-terminal fragment of the ORF2 protein of the Burmese strain (genotype I) expressed in E. coli. Two cynomolgus macaques were vaccinated with the fusion protein, and neither developed hepatitis following experimental challenge. The animal challanged with the heterologous Mexican strain (genotype II) did become infected but did not develop hepatitis.348 In another study, a 23-kDa pE2 protein specified by a 630-bp sequence cloned from 3¢ terminal of ORF2 of a Chinese HEV strain, expressed as a glutathione-S-transferase fusion protein in E. coli, was tested for its efficacy as a vaccine. Three rhesus monkeys were immunized with the purified peptide by 4-weekly 100-μg intramuscular doses and three were given placebo. Both groups of animals were challanged with 105 genome equivalent dose of the homologous strain of HEV. All control animals excreted the virus for 10–12 days beginning 5 days after the infection. The viral genome was also present in the peripheral blood monocytes from two animals, but it was not detected in the plasma samples from any of the animals. The infection in two control animals was accompanied by HEV seroconversion. Immunization was found to abrogate HEV stool excretion in two animals and reduced the viral excretion to 1 day in the third. None of the immunized animals showed detectable HEV in plasma or peripheral blood monocyte samples, nor did any animal show evidence of HEV seroconversion. The study suggested that vaccination with this peptide may prevent experimental infection of primates with the homologous strain of HEV.349 A more recent study evaluated the antigenicity, immunogenicity and efficacy of a candidate recombinant hepatitis E virus (HEV) vaccine (HEV 239 vaccine). The vaccine peptide had a 26 amino acid extension from the N terminal of E2 peptide of the HEV capsid protein. In contrast to E2, the vaccine peptide aggregates to form particles of 13 nm mean radius, and consequently, is more than 240 times more immunogenic than E2. Using alum as adjuvant, immunizing dose determined in mice was 80–250 ng for the vaccine and <60 mg for E2. Rhesus monkeys twice vaccinated with a 10 mg or a 20 mg formulation of this vaccine showed essentially the same antibody response, whereas the response to a 5 mg formulation was delayed but reached similar antibody levels. All the three vaccine formulations gave complete protection against infection with 104 genome equivalent dose of the homologous genotype 1 virus. At higher virus dose of 107, the same vaccine formulation partially protected against the infection and completely protected against hepatitis. The efficacy of the vaccine was essentially the same for the homologous genotype 1 virus and heterologous genotype 4 virus.350 Despite the promising results, vaccines derived from eukaryotic expression systems and especially from insect cells have become the focus of vaccine development strategies. Insect Cell-Derived Vaccine. Truncated baculovirus-expressed ORF2 proteins have been explored most extensively as vaccine candidates. However, the size and extent of modification of the ORF2 proteins in native virions are not known. Therefore it is not clear if any vaccine tested thus far contains all relevant neutralization epitopes, or if the presentation of epitopes mimics their presentation
Chapter 34 HEPATITIS E
Table 34-11. Trials of Recombinant Vaccines Based on Infected Insect cells in Non-human Primates Animal
Vaccine
Vaccine schedule
Challenge
Challenge time
Biochemical hepatitis
Histological heaptitis
Infection with HEV
Reference
Rhesus
55 kDa ORF2 protein of Pakistan strain (Sar55)
2 ¥ 50 mg i.m. (1 month apart): n = 4 2 ¥ 10 mg i.m. (1 month apart): n = 4 2 ¥ 2 mg i.m. (1 month apart): n = 4 2 ¥ 0.4 mg i.m. (1 month apart): n = 4 2 ¥ 50 mg i.m. (1 month apart): n = 4 2 ¥ 385 ng i.m.: n = 3 2 ¥ 385 ng i.m.: n = 3
300 000 MID50 Sar55 300 000 MID50 Sar55 300 000 MID50 Sar55 300 000 MID50 Sar55 100 000 MID50 Mex14 10 000 MID50 Sar55 100 MID50 Sar55 300 000 MID50 Sar55 300 000 MID50 Sar55 300 000 MID50 Sar55 300 000 MID50 Sar55
48 hours before first dose
0/4
0/4
4/4
281
0/4
0/4
4/4
0/4
0/4
4/4
0/4
0/4
4/4
0/4
0/4
3/4
4 week after last dose Month 7
2/3 1/3
0/3 0/3
3/3 0/3
364
1/4
0/4
3/4
329
Month 7
0/2
0/2
1/2
Month 13 Month 13
1/4 0/2
0/4 0/2
4/4 2/2
Rhesus
Rhesus
53 kDa ORF2 protein of Sar 55 56 kDa truncated capsid protein (aa 112–607) from Sar55
2 ¥ 50 mg i.m. (0, 1 month): n = 4 3 ¥ 50 mg i.m. (0, 1, 6 month): n = 2 2 ¥ 50 mg i.m. (0, 1 month): n = 4 3 ¥ 50 mg i.m. (0, 1, 6 months): n = 2
i.m., intramuscularly; MID50 , Monkey Infectious Dose, which causes infection in 50% of inoculated animals.
in infectious virions. Prototype recombinant vaccines based on either full-length or truncated pORF2, expressed in insect cells, have generated high-titer antibody responses.281,329,330,351 Protection against hepatitis but not against HEV infection was reported in most of these studies;281,351 however, full protection in some animals was reported in one study.329 Recently, a preclinical immunogenicity and efficacy trial of a recombinant hepatitis E vaccine in cynomolgus macaques reported full protection in some animals vaccinated with only one or two doses of the vaccine.330 Table 34-11 shows details of some trials of recombinant vaccines based on infected insect cells in non-human primates. Plant-Derived Vaccine. Recently, to explore the feasibility of developing a new type of plant-derived HEV oral vaccine, plant binary expression vector p1301E2, carrying a fragment of HEV ORF2 (named HEV-E2), was constructed by linking the fragment to a constitutive CaMV35s promoter and nos terminator (a regulatory sequence present in several lines of transgenic plants), then directly introduced into Agrobacterium tumefaciens EHA105. Using the leaf disk method, tomato plants medicated by EHA105 were transformed and hygromycin-resistant plantlets were obtained in selective medium containing hygromycin. Seven positive lines of HEV-E2-transgenic tomato plants confirmed by PCR and southern blotting were obtained and the immunoreactivity of recombinant protein could be detected in the extracts of transformants. Transgenic tomatoes may hold promise for producing a new type of lowcost oral vaccine for the HEV.352 A recent study used plant expression cassettes, pHEV101 and pHEV110, for transformation of potato, to express recombinant hepatitis E virus capsid protein (HEV CP). Ten independent transgenic lines of HEV101 and 6 lines of HEV110 were obtained. ELISA for HEV CP was performed on tuber extracts. Although, the
western blot showed that apparently intact HEV CP accumulated, very limited assembly of virus-like particles in potato tubers was observed. Oral immunization of mice with transgenic potatoes failed to elicit detectable anti-CP antibody response in serum, suggesting that VLP assembly is a key factor in orally delivered HEV CP vaccines.353 DNA Vaccines. DNA vaccines induce a broad range of immune responses, due to efficient priming of T lymphocytes.354,355 This novel approach offers several desirable features. First, the DNA is not infectious, it does not replicate, and it encodes only the protein(s) of interest. Second, DNA is stable and it can be made inexpensively in large quantities at a high level of purity and does not need to be stored in a refrigerator. Third, plasmid DNA does not contain any heterologous protein components, as compared with a recombinant virus vaccine, in which host responses to the vector may interfere with the desired response to the recombinant gene product. Fourth, DNA vaccine can induce both cell-mediated and humoral immunity. Finally, antigen expression persists after DNA vaccination, promoting the induction of long-lived memory immune cells.356 Injection of an expression vector pJHEV containing the HEVORF2 gene generated a strong humoral immune response in immunized mice.357 Later it was shown that this antibody can bind to and agglutinate HEV.358 In a study that tested for immunological memory in immunized mice whose current levels of IgG HEV were low or undetectable despite three doses of HEV DNA vaccine 18 months earlier, all mice were administered with a dose of HEV DNA vaccine to simulate an infectious challenge with HEV. Mice previouusly vaccinated with vector alone were controls. By 40 days after the vaccine dose, the level of IgG had increased dramatically in all previously DNA-vaccinated mice, whereas no control mice
713
Section V. Liver Diseases Due to Infectious Agents
became seropositive. This study indicated that HEV DNA vaccine elicits immunological memory in immunized mice, and is capable of generating a response to challenge that could be protective against disease.359 Studies of DNA vaccine in non-human primates have also been done. Intramuscular injection of plasmid DNA encoding HEV ORF2 induced a modest humoral immune response in macaques; 50% of the animals were protected against infection when challenged with a heterologous HEV strain.360 In order to improve the efficacy of that DNA vaccine, it was tested by administering it with a gene gun. Macaques were vaccinated with a plasmid containing a full-length HEV ORF2 sequence (Burmese strain) and subsequently challenged with a heterologous strain of HEV (Mexican strain). The animals administered vaccine by gene gun developed antibodies to HEV, whereas those administered vaccine by intradermal injections and others administered a mock DNA construct did not develop antiHEV. Anti-HEV-positive animals were protected from HEV infection after challenge with an inoculum that produced infection in anti-HEV-negative animals. Thus, DNA vaccine with HEV ORF2 administered by gene gun is protective against a heterologous viral challenge.361 The 56-kDa recombinant vaccine developed at the National Institute of Health, USA, had been licenced to Smith Kline Beecham and was evaluated in humans in South Africa. In a phase I trial, the vaccine was found to be safe and immunogenic in 88 healthy adults aged 18–50. The vaccine was administered in 1, 5, 20, and 40 μg antigen dose escalation in 22 subjects. Three doses were administered intramuscularly to the four groups of volunteers on day 0, month 1, and month 6. At month 7, the vaccine elicited anti-HEV titers of 40 U/ml in at least 88% of vaccinees in the 5-, 20-, and 40μg groups. No serious adverse events caused by the vaccine have been reported.362 A further phase I evaluation was performed in Nepal, where hepatitis E is endemic. Three doses of formulations of 5 and 20 μg respectively were injected into 22 Nepalese volunteers each at 0, and 1 and 6 months. Serious adverse events were not observed. By the second month, 43 of 44 volunteers had seroconverted to anti-HEV. By the seventh month the remaining volunteer also developed antibody to HEV. The study indicated that the HEV candidate vaccine was well tolerared and immunogenic in Nepalese people. A phase II–III efficacy trial has been planned in Kathmandu, Nepal, where as many as 90% of the jaundice cases of hepatitis are caused by HEV.363 In summary, HEV remains an important cause of both sporadic and epidemic viral hepatitis in poor developing and underdeveloped countries, with poor sanitation and other civic amenities. It is also an important cause of maternal mortality in those countries. Why HEV causes severe disease in pregnant women remains to be discovered. Significant research is being done to develop an HEV vaccine, and if available and effective, it would be a boon for select groups of individuals like pregnant women, patients with pre-existing liver disease, and foreign travelers.
REFERENCES 1. Khuroo MS. Study of an epidemic of non-A, non-B hepatitis. Possibility of another human hepatitis virus distinct from posttransfusion non-A, non-B type. J Med 1980; 68:818–824.
714
2. Wong DC, Purcell RH, Sreenivasan MA, et al. Epidemic and endemic hepatitis in India: evidence for non-A/non-B hepatitis virus etiology. Lancet 1980; 2:876–878. 3. Balayan MS, Andjaparidze AG, Savinskaya SS, et al. Evidence for a virus in non-A, non-B hepatitis transmitted via the fecal–oral route. Intervirology 1983; 20:23–31. 4. Reyes GR, Purdy MA, Kim JP, et al. Isolation of a cDNA from the virus responsible for enterically transmitted non-A, non-B hepatitis. Science 1990; 247:1335–1339. 5. Bradley DW. Hepatitis non-A, non-B viruses become identified as hepatitis C and E viruses. In: Melnick JL, ed. Progress in medical virology. Basel: Karger; 1990:101–135. 6. Purcell RH, Ticehurst JR. Enterically transmitted non-A, non-B hepatitis: epidemiology and clinical characteristics. In: Zuckerman A, ed. Viral hepatitis and liver disease. New York: Alan R Liss; 1988:131–137. 7. Cubitt W, Bradley D, Carter M, et al. Caliciviridae. Arch Virol Suppl 1995; 10:359–363. 8. Fry KE, Tam AW, Smith MM, et al. Hepatitis E virus (HEV): strain variation in the nonstructural gene region encoding consensus motifs for an RNA-dependent RNA polymerase and an ATP/GTP binding site. Virus Genes 1992; 6:173–185. 9. Tam AW, Smith MM, Guerra ME, et al. Hepatitis E virus (HEV): molecular cloning and sequencing of the full-length viral genome. Virology 1991; 185:120–131. 10. Koonin E, Gorbalenya A, Purdy M, et al. Computer-assisted assignment of functional domains in the nonstructural polyprotein of hepatitis E virus: delineation of an additional group of positive-strand RNA plant and animal viruses. Proc Natl Acad Sci USA 1992; 89:8259–8263. 11. Berke T, Maston DO. Reclassification of the Caliciviridae into distinct genera and exclusion of hepatitis E from the family on the basis of comparative phylogenetic analysis. Arch Virol 2000; 145:1421–1426. 12. Koonim EV, Dolja VV. Evolution and taxonomy of positivestrand RNA viruses: implications of comparative analysis of aminoacid sequences. Crit Rev Biochem Mol Biol 1993; 28:375–430. 13. Green KY, Ando T, Balayan MS, et al. Taxonomy of the caliciviruses. J Infect Dis 2000; 181 (suppl 2): S322–S330. 14. Favorov MO, Goldberg EZ, Gurov AV, et al. [Some physicochemical properties of non-A, non-B hepatitis virus with the fecal–oral mechanism of transmission and specific diagnosis of the disease (in Russian).] Vopr Virusol 1989; 1:47–50. 15. Bradley DW, Andjaparidze AG, Cook EH Jr, et al. Etiologic agent of enterically-transmitted non-A, non-B hepatitis. J Gen Virol 1988; 69:731–738. 16. Jothikumar N, Aparna K, Kamatchiammal S, et al. Detection of hepatitis E virus in raw and treated wastewater with the polymerase chain reaction. Appl Environ Microbiol 1993; 59:2558–2562. 17. Tam AW, Smith MM, Guerra ME, et al. Hepatitis E virus (HEV): molecular cloning and sequencing of the full-length viral genome. Virology 1991; 185:120–131. 18. Bradley DW. Enterically-transmitted non-A, non-B hepatitis. Br Med Bull 1990; 46:442–461. 19. Ticehurst J. Identification and characterization of hepatitis E virus. In: Hollinger FB, Lemon SM, Margolis H, eds. Hepatitis and liver disease. Baltimore: Williams & Wilkins; 1991:501–513. 20. Kapikian AZ, Chanock RM. Norwalk group of viruses. In: Fields BN, ed. Virology. New York: Raven; 1985:1495–1517. 21. Li TC, Yamakawa I, Suzuki M, et al. Expression and selfassembly of empty virus-like particles of hepatitis E virus. J Virol 1997; 71:7207–7213. 22. Yarbough PO, Tam AW, Fry KE, et al. Hepatitis E virus: identification of type – common epitopes. J Virol 1991; 65:5790–5797.
Chapter 34 HEPATITIS E
23. Xing L. Kato K, Li T, et al. Recombinant hepatitis E capsid protein self-assembles into a dual-domain T=1 particle presenting native virus epitopes. Virology 1999; 265: 35–45. 24. Kabrane-Lazizi I, Meng XJ, Purcell RH, Emerson SU. Evidence that the genomic RNA of hepatitis E virus is capped. J Virol 1999; 73:8848–8850. 25. Koonin EV, Gorbalenya AK, Purdy MA, et al. Computer-assisted assignment of functional domains in the nonstructural polyprotein of hepatitis E virus: delineation of an additional group of positive-strand RNA plant and animal viruses. Proc Natl Acad Sci USA 1992; 89:8259–8263. 26. Gorbalenya AK, Koonin EV, Lai MM-C. Putative papain-related proteases of positive-strand RNA viruses. FEBS Lett 1991; 288:201–205. 27. Reyes GR, Huang C-C, Yarbough PO, et al. Hepatitis E virus. J Hepatol 1991; 13:S155–S161. 28. Tsarev SA, Emerson SU, Reyes GR, et al. Characterization of a prototype strain of hepatitis E virus. Proc Natl Acad Sci USA 1992; 89:559–563. 29. Gorbalenya AK, Koonin EV, Lai MM-C. Putative papain-related proteases of positive-strand RNA viruses. FEBS Lett 1991; 288:201–205. 30. Gorbalenya AK, Koonin EV, Donchenko AP, et al. Two related superfamilies of putative helicases involved in replication, recombination, repair and expression of DNA and RNA genomes. Nucleic Acids Res 1989; 17:4713–4730. 31. Koonin EV. The phylogeny of RNA-dependent RNA polymerases of positive-strand RNA viruses. J Gen Virol 1991; 72:2197–2206. 32. Ansari IH, Nanda SK, Dugapal H, et al. Cloning, sequencing, and expression of hepatitis E virus (HEV) nonstructural open reading frame 1 (ORF1). J Med Virol 2000; 60:275–283. 33. Bradley DW, Purdy MA. Molecular and serological characteristics of hepatitis E virus. In: Nishioka K, Suzuki H, Mishiro S, et al., eds. Viral hepatitis and liver disease. Tokyo: Springer-Verlag; 1993:42–45. 34. Zafrullah M, Ozdener MH, Kumar R, et al. Mutation analysis of glycosylation, membrane translocation, and cell surface expression of the hepatitis E virus ORF2 protein. J Virol 1999; 73:4074–4082. 35. Jameel S, Zafrullah M, Ozdener MH, Panda SK. Expression in animal cells and characterization of hepatitis E virus structural proteins. J Virol 1996; 70:207–216. 36. Torresi J, Meanger P, Lambert P, et al. High level expression of the capsid protein hepatitis E virus in diverse eukaryotic cells using the Semliki Forest virus replicon. J Virol Methods 1997; 69:81–91. 37. Li TC, Yamakawa I, Suzuki M, et al. Expression and selfassembly of empty virus-like particles of hepatitis E virus. J Virol 1997; 71:7207–7213. 38. Robinson RA, Burgess WH, Emerson SU, et al. Structural characterization of recombinant hepatitis E virus ORF2 proteins in baculovirus-infected insect cells. Protein Expression Purif 1998; 12:75–84. 39. Zhang Y, McAtee P, Yarbough PO, et al. Expression, characterization and immunoreactivities of a soluble hepatitis E virus putative capsid protein species expressed in insect cells. Clin Diagn Lab Immunol 1997; 4:423–428. 40. Tyagi S, Jameel S, Lal SK. The full length and N-terminal deletion of ORF2 protein of hepatitis E virus can dimerize. Biochem Biophys Res Commun 2001; 286:214–221. 41. Surjit M, Jameel S, Lal SK. The ORF2 protein of hepatitis E virus binds the 5¢ region of viral RNA. J Virol 2004; 78:320–328. 42. Zafrullah M, Ozdener MH, Panda SK, et al. The ORF3 protein of hepatitis E virus is a phosphoprotein that associates with the cytoskeleton. J Virol 1997; 71:9045–9053.
43. Korkaya H, Jameel S, Gupta D, et al. The ORF3 protein of hepatitis E virus binds to Src homology 3 domains and activates MAPK. J Biol Chem 2001; 276:42389–42400. 44. Tyagi S, Korkaya H, Zafrulla M, et al. The phosphorylated form of the ORF3 protein of hepatitis E interacts with its nonglycosylated form of the major capsid protein ORF2. J Biol Chem 2002; 277:22759–22767. 45. Tyagi S, Jameel S, Lal SK. Self-association and mapping of the interaction domain of hepatitis E virus ORF3 protein. J Virol 2001; 75:2493–2498. 46. Huang CC, Nguyen D, Fernandez J, et al. Molecular cloning and sequencing of the Mexico isolate of hepatitis E virus (HEV). Virology 1992; 191:550–558. 47. Schlauder GG, Mushahwar IK. Genetic heterogeneity of hepatitis E virus. J Med Virol 2001; 65:282–292. 48. Kwo PY, Schlauder GG, Carpenter HA, et al. Acute hepatitis E by a new isolate acquired in the United States. Mayo Clin Proc 1997; 72:1133–1136. 49. Schlauder GG, Dawson GJ, Erker JC, et al. The sequence and phylogenetic analysis of a novel hepatitis E virus isolated from a patient with acute hepatitis reported in the United States. J Gen Virol 1998; 79:447–456. Erratum: J Gen Virol 1998;79:2563. 50. Erker JC, Desai SM, Schlauder GG, et al. A hepatitis E virus variant from the United States: molecular characterization and transmission in cynomolgus macaques. J Gen Virol 1999; 80:681–690. 51. Psichogiou MA, Tassopoulus N, Papatheodoridis GV, et al. Hepatitis E virus (HEV) infection in a cohort of patients with acute non-A non-B hepatitis. J Hepatol 1995; 23: 668–673. 52. Schlauder GG, Desai SM, Zanetti AR, et al. Novel hepatitis E virus (HEV) isolates from Europe: evidence for additional genotypes of HEV. J Med Virol 1999; 57:243–251. 53. Zanetti AR, Schlauder GG, Ramano L, et al. Hepatitis E virus in patients with acute non-A-C hepatitis in Italy: identification of a novel variant. J Med Virol 1999; 57:356–360. 54. Worm HC, Wurzer H, Frosner G. Sporadic hepatitis E in Austria. N Engl J Med 1998; 339:1554–1555. 55. Worm HC, Schlauder GG, Wurzer H, Mushahwar IK. Identification of a novel variant of hepatitis E in Austria. J Gen Virol 2000; 81:2885–2890. 56. Schlauder GG, Frider B, Sookoian S, et al. Identification of two novel isolates of hepatitis E virus in Argentina. J Infect Dis 2000; 182:294–297. 57. Pina S, Buti M, Cotrina M, et al. HEV identified in serum from humans with acute hepatitis and in sewage of animal origin in Spain. J Hepatol 2000; 33:826–833. 58. Huang R, Nakazono N, Ishii K, et al. Existing variations on the gene structure of hepatitis E virus strains from some regions of China. J Med Virol 1995; 47:303–308. 59. Hsieh SY, Yang PY, Ho YP, et al. Identification of a novel strain of hepatitis E virus responsible for sporadic acute hepatitis in Taiwan. J Med Virol 1998; 55:300–304. 60. Wang YR, Ling R, Erker JC, et al. A divergent genotype of hepatitis E virus in Chinese patients with acute hepattis. J Gen Virol 1999; 80:169–177. 61. Takahashi K, Iwata K, Watanabe N, et al. Full-genome nucleotide sequence of a hepatitis E virus strain that may be indigenous to Japan. Virology 2001; 287:9–12. 62. Takahashi K, Kang JH, Ohnishi S, et al. Genetic heterogeneity of hepatitis E virus recovered from Japanese patients with acute sporadic hepatitis. J Infect Dis 2002; 185:1342–1345. 63. Takahashi M, Nishizawa T, Yoshikawa A, et al. Identification of two distinct genotypes of hepatitis E virus in a Japanese patient with acute hepatitis who had not travelled abroad. J Gen Virol 2002; 83:1931–1940.
715
Section V. Liver Diseases Due to Infectious Agents
64. Mizuo H, Suzuki K, Takikawa Y, et al. Polyphyletic strains of hepatitis E virus are responsible for sporadic cases of acute hepatitis in Japan. J Clin Microbiol 2002; 40:3209–3218. 65. Tokita H, Harada H, Gotanda Y, et al. Molecular and serological characterization of sporadic acute hepatitis E in a Japanese patient infected with a genotype III hepatitis E virus in 1993. J Gen Virol 2003; 84:421–427. 66. Van Cuyck-Gandre H, Zhang HY, Tsarev SA, et al. Short report: phylogenetically distinct hepatitis E viruses in Pakistan. Am J Trop Med Hyg 2000; 62:187–189. 67. He J, Binn LN, Tsarev SA, et al. Molecular characterization of a hepatitis E virus isolate from Namibia. J Biomed Sci 2000; 7:334–338. 68. Maila HT, Bowyer SM, Swanepoel R. Identification of a new strain of hepatitis E virus from an outbreak in Namibia in 1995. J Gen Virol 2004; 85:89–95. 69. Shrestha SM, Shrestha S, Tsuda F, et al. Genetic changes in hepatitis E virus of subtype 1a in patients with sporadic acute hepatitis E in Kathmandu, Nepal, from 1997 to 2002. J Gen Virol 2004; 85:97–104. 70. Arankalle VA, Paranjape S, Emerson SU, et al. Phylogenetic analysis of hepatitis E virus isolates from India (1976–1993). J Gen Virol 1999; 80:1691–1700. 71. Li F, Zhuang H, Kolivas S, et al. Persistent and transient antibody responses to hepatitis E virus detected by Western immunoblot using open reading frame 2 and 3 and glutathione S-transferase fusion proteins. J Clin Microbiol 1994; 32:2060–2066. 72. Li F, Torresi J, Locarnini SA, et al. Amino-terminal epitopes are exposed when full-length open reading frame 2 of hepatitis E virus is expressed in Escherichia coli, but carboxy-terminal epitopes are masked. J Med Virol 1997; 52:289–300. 73. Mast EE, Alter MJ, Holland PV, et al. Evaluation of assays for antibody to hepatitis E virus by a serum panel. Hepatology 1998; 27:857–861. 74. Editorial. Meeting reports. J R Soc Med 1992; 85:50–53. 75. Balyan MS. Epidemiology of hepatitis E virus infection. J Viral Hepatitis 1997; 4:155–165. 76. Vishwanathan R: Infectious hepatitis in Delhi (1955–1956): a critical study: Epidemiology. Indian J Med Res 1957; 45 (suppl):1–29. 77. Cao XY, Ma XZ, Liu YZ, et al: Epidemiological and etiological studies on enterically transmitted non-A, non-B hepatitis in the south part of Xinjiang. In: Shikata T, Purcell R, Uchida T, eds. Viral hepatitis C, D and E. Amsterdam: Excerpta Medica; 1991:297–312. 78. Naik SR, Aggarwal R, Salunke PN, et al. A large waterborne viral hepatitis E epidemic in Kanpur, India. Bull WHO 1992; 70:597–604. 79. Tsega E, Hansson BG, Krawczynski K, et al. Acute sporadic viral hepatitis in Ethiopia: causes, risk factors, and effects on pregnancy. Clin Infect Dis 1992; 14:961–965. 80. Kane MA, Bradley DW, Shrestha SM, et al. Epidemic non-A, non-B hepatitis in Nepal: recovery of a possible etiological agent and transmission studies in marmosets. JAMA 1984; 252:3140–3145. 81. Balyan MS. New form of hepatitis with feco–oral mode of spread. In: Zhdanov VM, ed. Soviet medical reviews, section E, virology reviews, vol. 2. Switzerland: Harwood Academic; 1987:235–261. 82. Mushahwar IK, Dawson GJ, Bile KM, et al. Serological studies of an enterically transmitted non-A, non-B hepatitis in Somalia. J Med Virol 1993; 40:218–221. 83. Hyams KC, Purdy MA, Kaur M, et al. Acute sporadic hepatitis E in Sudanese children: analysis based on a new Western blot assay. J Infect Dis 1992; 165:1001–1005. 84. Jameel S, Durgapal H, Habibullah CM, et al. Enteric non-A, non-B hepatitis: epidemics, animal transmission, and hepatitis E
716
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95. 96.
97. 98.
99. 100.
101.
102.
103.
104. 105. 106. 107.
virus detection by the polymerase chain reaction. J Med Virol 1992; 37:263–270. El-Zimaity DMT, Hyams KC, Imam IZE, et al. Acute sporadic hepatitis E in an Egyptian pediatric population. Am J Trop Med Hyg 1993; 48:372–376. Hyams KC, McCarthy MC, Kaur M, et al. Acute sporadic hepatitis E in children living in Cairo, Egypt. J Med Virol 1992; 37:274–277. Clayson E, Shrestha M, Vaughn D, et al. Rates of hepatitis E virus infection and disease among adolescents and adults in Kathmandu, Nepal. J Infect Dis 1997; 176:763–766. Arankalle VA, Chadha MS, Mehendale SM, et al. Outbreak of enterically transmitted non-A, non-B hepatitis among schoolchildren. Lancet 1988; 2:1199–1200. Schlauder G, Dawson G, Mushahwar I, et al. Viraemia in Egyptian children with hepatitis E virus infection. Lancet 1993; 341:378. Aggarwal R, Shahi H, Naik S, et al. Evidence in favour of high infection rate with hepatitis E virus among young children in India. J Hepatol 1997; 26:1425–1430. Ripabelli G, Sammarco ML, Campo T, et al. Prevalence of antibodies against enterically transmitted viral hepatitis (HAV and HEV) among adolescents in an inland territory of central Italy. Eur J Epidemiol 1997; 13:45–47. Khuroo MS, Rustagi VK, Dawson GJ, et al. Spectrum of hepatitis E virus infection in India, J Med Virol 1994; 43:281–286. Arankalle VA, Chobe LP, Jha J, et al. Aetiology of acute sporadic non-A non-B viral hepatitis in India. J Med Virol 1993; 40:121–125 Khoroo MS, Deurmeyer W, Zargar SA, et al. Acute sporadic non-A, non-B, hepatitis in India. Am J Epidemiol 1983; 118:360–364. CDC. Hepatitis E among US travelers, 1989–1992. MMWR Morb Mortal Wkly Rep 1993; 42:1–4. Ghabrah T, Strickland T, Tsarev S, et al. Acute viral hepatitis in Saudi Arabia: seroepidemiological analysis, risk factors, clinical manifestations, and evidence for a sixth hepatitis agent. Clin Infect Dis 1995; 21:621–627. Koshy A, Richards AL, Al-Mufti S, et al. Acute sporadic hepatitis E in Kuwait. J Med Virol 1994; 42:405–408. Smalley D, Brewer S, Dawson G, et al. Hepatitis E virus infection in an immigrant to the United States. South Med J 1996; 89:994–996. Skidmore SJ, Sherratt LM. Hepatitis E infection in the UK. J Viral Hepat 1996; 3:103–105. Wu JC, Sheen IJ, Chiang TY, et al. The impact of travelling to endemic areas on the spread of hepatitis E virus infection: epidemiological and molecular analyses. Hepatology 1998; 27:1415–1420. Smalligan R, Lange W, Frame J, et al. The risk of viral hepatitis A, B, C and E among North American missionaries. Am J Trop Med Hyg 1995; 53:233–236. Drabick J, Gambel J, Gouvea V, et al. A cluster of acute hepatitis E infection in United Nations Bangladeshi peacekeepers in Haiti. Am J Trop Med Hyg 1997; 57:449–454. Clayson E, Innis B, Myint KS, et al. Short report: relative risk of hepatitis A and E among foreigners in Nepal. Am J Trop Med Hyg 1995; 52:506–507. Moaven LD, Fuller AJ, Doultree JC, et al. A case of acute hepatitis E in Victoria. Med J Aust 1993; 159:124–125. Roberts JK, Whitlock RT. Hepatitis E in a traveler to Bangladesh. Ann Intern Med 1992; 117:93. Zaaijer HL, Kok M, Lelie PN, et al. Hepatitis E in the Netherlands: imported and endemic. Lancet 1993; 341:826. Teich N, Tannapfel A, Ammon A, et al. Sporadic acute hepatitis E in Germany: an underdiagnosed phenomenon? Z Gastroenterol 2003; 41:419–423.
Chapter 34 HEPATITIS E
108. Grieco A, Miele L, Gasbarrini G, Grillo R. Sporadic HEV hepatitis in Italy. Gut 2001; 48:580. 109. Levine DF, Bendall RP, Teo CG. Hepatitis E acquired in the UK. Gut 2000; 47:740. 110. Tsang TH, Denison EK, Williams HV, et al. Acute hepatitis E infection acquired in California. Clin Infect Dis 2000; 30:618–619. 111. Lee S-D, Wang Y-J, Lu R-H, et al. Seroprevalence of antibody to hepatitis E virus among Chinese subjects in Taiwan. Hepatology 1994; 19:866–870. 112. Pujol FH, Favorov MO, Marcano T, et al. Prevalence of antibodies against hepatitis E virus among urban and rural populations in Venezuela. J Med Virol 1994;48:234–236. 113. Thomas DL, Mahley RW, Badur S, et al. Epidemiology of hepatitis E virus infection in Turkey. Lancet 1993; 341:1561–1562. 114. Pawlotsky JM, Belec L, Gresenguet G, et al. High prevalence of hepatitis B, C, and E markers in young sexually active adults from the Central African Republic. J Med Virol 1995; 46:269–273. 115. Focaccia R, Sette H Jr, Conceicao OJG. Hepatitis E in Brazil. Lancet 1995; 346:1165. 116. Aubry P, Niel L, Niyongabo T, et al. Seroprevalence of hepatitis E virus in an adult urban population from Burundi. Am J Trop Med Hyg 1997; 57:272–273. 117. Abdelaal M, Zawawi TH, al Sobhi E, et al. Epidemiology of hepatitis E virus in male blood donors in Jeddah, Saudi Arabia. Isr J Med Sci 1998; 167:94–96. 118. Brahm J, Hurtado C, Moraga M, et al. [Hepatitis E virus infection in Chile: preliminary report (in Spanish).] Rev Med Chil 1996; 124:947–949. 119. Cruells MR, Mescia G, Gaibisso R, et al. Epidemiological study of hepatitis A and E viruses in different populations in Uruguay. Gastroenterol Hepatol 1997; 20:295–298. 120. Darwish MA, Faris R, Clemens JD, et al. High seroprevalence of hepatitis A, B, C, and E viruses in residents in an Egyptian village in the Nile delta: a pilot study. Am J Trop Med Hyg 1996; 54:554–558. 121. Grabow WOK, Favorov MO, Khudyakova NS, et al. Hepatitis E seroprevalence in selected individuals in South Africa. J Med Virol 1994; 44:384–388. 122. Ibarra H, Reidemann S, Reinhardt G, et al. [Prevalence of hepatitis E virus antibodies in blood donors and other population groups in southern Chile (in Spanish).] Rev Med Chil 1997; 125:275–278. 123. Kamel M, Troonen H, Kapprell HP, et al. Seroepidemiology of hepatitis E virus in the Egyptian Nile delta. J Med Virol 1995; 47:399–403. 124. Langer BCA, Frosner GG, von Brunn A. Epidemiological study of viral hepatitis types A, B, C, D, and E among Inuits in West Greenland. J Viral Hepat 1997; 4:339–349. 125. Tucker T, Kirsch R, Louw S, et al. Hepatitis E in South Africa: evidence for sporadic spread and increased seroprevalence in rural areas. J Med Virol 1996; 50:117–119. 126. Talarmin A, Kazanji M, Cardoso T, et al. Prevalence of antibodies to hepatitis A, C, and E viruses in different ethnic groups in French Guiana. J Med Virol 1997; 52:430–435. 127. Favorov MO, Khudyakov YE, Fields HA, et al. Enzyme immunoassay for the detection of antibody to hepatitis E virus based on synthetic peptides. J Virol Methods 1994; 46:237–250. 128. Balayan MS, Zamyatina NA, Mikhailov MI, et al. Serological survey on hepatitis E virus infection in an endemic area: diagnostic potential of enzyme immunoassay for detection of anti-IgG antibody. Clin Diag Virol 1994; 2:297–304. 129. Lok ASF, Kwan W-K, Moeckli R, et al. Seroepidemiological survey of hepatitis E in Hong Kong; by recombinant-based enzyme immunoassays. Lancet 1992; 340:1205–1208.
130. Peng CF, Lin MR, Chue PY, et al. Prevalence of antibody to hepatitis E virus among healthy individuals in Southern Taiwan. Microbiol Immunol 1995; 39:733–736. 131. Goncales NS, Pinho JR, Moreira RC, et al. Hepatitis E virus immunoglobulin G antibodies in different populations in Campinas, Brazil. Clin Diagn Lab Immunol 2000; 7: 813–816. 132. Stark K, Poggensee G, Hohne M, et al. Seroepidemiology of TT virus, GBC-C/HGV, and hepatitis viruses B, C, and E among women in a rural area of Tanzania. J Med Virol 2000; 62:524–530. 133. Qureshi H, Shahid A, Mujeeb SA. Exposer rate of hepatitis E (IgG) in a selected population of children and adults in Karachi. J Pak Med Assoc 2000; 50:352–354. 134. Fix AD, Abdel-Hamid M, Purcell RH, et al. Prevalence of antibodies to hepatitis E in two rural Egyptian communities. Am J Trop Med Hyg 2000; 62:519–523. 135. Mathur P, Arora NK, Panda SK, et al. Sero-epidemiology of hepatitis E virus (HEV) in urban and rural children of North India. Indian Pediatr 2001; 38:461–475. 136. Assis SB, Souto FJ, Fontes CJ, Gaspar AM. Prevalence of hepatitis A and E virus infection in school children of an Amazonian municipality in Mato Grosso state. Rev Soc Bras Med Trop 2002; 35:155–158. 137. Santos DC, Souto FJ, Santos DR, et al. Seroepidemiological markers of enterically transmitted viral hepatitis A and E in individuals living in a community located in the north area of Rio de Janeiro, RJ, Brazil. Mem Inst Oswaldo Cruz 2002; 97:637–640. 138. Cesur S, Akin K, Dogaroglu I, et al. Hepatitis A and hepatitis E seroprevalence in adults in the Ankara area. Mikrobiyol Bul 2002; 36:79–83. 139. Mohanavalli B, Dhevahi E, Menon T, et al. Prevalence of antibodies to hepatitis A and hepatitis E virus in urban school children in Chennai. Indian Pediatr 2003; 40:328–331. 140. Louisirirotchanakul S, Myint KS, Srimee B, et al. The prevalence of viral hepatitis among the Hmong people of northern Thailand. Southeast Asian J Trop Med Public Health 2002; 33:837–844. 141. Choi IS, Kwon HJ, Shin NR, Yoo HS. Identification of swine hepatitis E virus (HEV) and prevalence of anti-HEV antibodies in swine and human populations in Korea. J Clin Microbiol 2003; 41:3602–3608. 142. Wong KH, Liu YM, Ng PS, et al. Epidemiology of hepatitis A and hepatitis E infection and their determinants in adult Chinese community in Hong Kong. J Med Virol 2004; 72:538–544. 143. Arankalle VA, Chadha MS, Chitambar SD, et al. Changing epidemiology of hepatitis A and hepatitis E in urban and rural India (1982–98). J Viral Hepat 2001; 8:293–303. 144. Thomas DL, Yarbough PO, Vlahov D, et al. Seroreactivity to hepatitis E virus in areas where the disease is not endemic. J Clin Microbiol 1997; 35:1244–1247. 145. Anderson DA, Li F, Riddell MA, et al. ELISA for IgG class antibody to hepatitis E virus based on a highly conserved, conformational epitope expressed in Escherichia coli. J Virol Methods 1999; 81:131–142. 146. Seow HF, Mahomed NMB, Mak JW, et al. Seroprevalence of antibodies to hepatitis E virus in normal blood donor population and two aboriginal communities in Malaysis. J Med Virol 1999; 59:164–168. 147. Mast E, Kuramoto I, Favorov M, et al. Prevalence of and risk factors for antibody to hepatitis E virus seroreactivity among blood donors in northern California. J Infect Dis 1997; 176:34–40. 148. Queiros L, Condeco J, Tender A, et al. [The seroprevalence for hepatitis E viral antibodies in the northern region of Portugal (in Portuguese).] Acta Med Port 1997; 10:447–453.
717
Section V. Liver Diseases Due to Infectious Agents
149. Stroffolini T, Menchinelli M, Dambruoso V, et al. Prevalence of hepatitis E in a central Italian town at high endemicity for hepatitis C virus. Ital J Gastroenterol 1996; 28:523–525. 150. Moaven L, Asten M, Crofts N, et al. Seroepidemiology of hepatitis E in selected Australian populations. J Med Virol 1995; 45:326–330. 151. Karetnyi YV, Favorov MO, Khudyakova NS, et al. Serological evidence for hepatitis E virus infection in Israel. J Med Virol 1995; 45:316–320. 152. Gessoni G, Manoni F. Hepatitis E virus infection in north-east Italy: serological study in the open population and groups at risk. J Viral Hepat 1996; 3:197–202. 153. Ritter A, Flacke H, Vornwald A, et al. A seroprevalence study of hepatitis E in Europe and Middle East. In: Nishioka K, Suzuki H, Mishiro S, Oda T, eds. Viral hepatitis and liver disease. Tokyo: Springer-Verlag; 1994:432–434. 154. Vranckx R, Van Damme P, Coenjaert A. HEV infection in selected Belgian populations. In: Buisson Y, Coursaget P, Kane M, eds. Enterically-transmitted hepatitis viruses. Joue-les-Tours, France: La Simarre; 1996:225–227. 155. Balayan MS, Fedorova OE, Zamyatina NA, et al. HEV infection: prevalence of IgG class antibody to HEV in endemic verses nonendemic areas. In: Buisson Y, Coursaget P, Kane M, eds. Enterically-transmitted hepatitis viruses. Joue-les-Tours, France: La Simarre; 1996:145–152. 156. Bauduceau O, Berlioz A, Buisson Y. Hepatitis B, C, and E in New Caledonia. Seroepidemiologic study in military recruits. Med Trop 2000; 60:167–170. 157. Tanaka E, Takeda N, Tian-Chen L, et al. Seroepidemiological study of hepatitis E virus infection in Japan using a newly developed antibody assay. J Gastroenterol 2001; 36:317–321. 158. Andenaes S, Lie A, Degre M. Prevalence of hepatitis A, B, C, and E antibody in flying airline personnel. Aviat Space Environ Med 2000; 71:1178–1180. 159. Sidal M, Unuvar E, Oguz F, et al. Age-specific seroepidemiology of hepatitis A, B, and E infections among children in Istanbul, Turkey. Eur J Epidemiol 2001; 17:141–144. 160. Colak D, Ogunc D, Gunseren F, et al. Seroprevalence of antibodies to hepatitis A and E viruses in pediatric age groups in Turkey. Acta Microbiol Immunol Hung 2002; 49:93–97. 161. Chironna M, Germinario C, Lopalco PL, et al. Prevalence rates of viral hepatitis infections in refugee Kurds from Iraq and Turkey. Infection 2003; 31:70–74. 162. Gandolfo GM, Ferri GM, Conti L, et al. Prevalence of infections by hepatitis A, B, C and E viruses in two different socioeconomic groups of children from Santa Cruz, Bolivia. Med Clin (Barc) 2003; 120:725–727. 163. Ding X, Li TC, Hayashi S, et al. Present state of hepatitis E virus epidemiology in Tokyo, Japan. Hepatol Res 2003; 27:169–173. 164. Meng XJ, Wiseman B, Elvinger F, et al. Prevalence of antibodies to hepatitis E virus in veterinarians working with swine and in normal blood donors in the United States and other countries. J Clin Microbiol 2002; 40:117–122. 165. Yoo D, Willson P, Pei Y, et al. Prevalence of hepatitis E virus antibodies in Canadian swine herds and identification of a novel variant of swine hepatitis E virus. Clin Diagn Lab Immunol 2001; 8:1213–1219. 166. Arankalle VA, Tsarev SA, Chadha MS, et al. Age-specific prevalence of antibodies to hepatitis A and E viruses in Pune, India, 1982 and 1992. J Infect Dis 1995; 171:447–450. 167. Sreenivasan MA, Sehgal A, Prasad SR, Dhorje S. A seroepidemiologic study of a water-borne epidemic of viral hepatitis in Kolhapur City, India. J Hyg (Lond) 1984; 93:113–122. 168. Belabbes EH, Bouguermouh A, Benatallah A, Illoul G. Epidemic non-A, non-B viral hepatitis in Algeria: strong evidence for its spreading by water. J Med Virol 1985; 16:257–263. 169. Corwin A, Jarot K, Lubis I, et al. Two years’ investigation of epidemic hepatitis E virus transmission in West Kalimantan
718
170.
171.
172.
173.
174. 175.
176.
177.
178.
179. 180.
181.
182. 183.
184.
185.
186.
187. 188.
189.
190.
(Borneo), Indonesia. Trans R Soc Trop Med Hyg 1995; 89:262–265. Corwin AL, Tien NT, Bounlu K, et al. The unique riverine ecology of hepatitis E virus transmission in South-East Asia. Trans R Soc Trop Med Hyg 1999; 93:255–260. Zhuang H. Hepatitis E and strategies for its control. In: Wen YM, Xu ZY, Melnick JL, eds. Viral hepatitis in China: problems and control strategies. Basel, Switzerland: S. Karger; 1992:126–129. Arankalle VA, Chadha MS, Mehendale SM, Tungatkar SP. Epidemic hepatitis E: serological evidence for lack of intrafamilial spread. Indian J Gastroenterol 2000; 19:24–28. Somani SK, Aggarwal R, Naik SR, et al. A serological study of intrafamilial spread from patients with sporadic hepatitis E virus infection. J Viral Hepat 2003; 10:446–449. Aggarwal R, Naik SR. Hepatitis E. Intrafamilial transmission versus waterborne spread. J Hepatol 1994; 21:718–723. Matsuda H, Okada K, Takahashi K, Mishiro S. Severe hepatitis E virus infection after ingestion of uncooked liver from a wild boar. J Infect Dis 2003; 188:944. Yazaki Y, Mizuo H, Takahashi M, et al. Sporadic acute or fulminant hepatitis E in Hokkaido, Japan, may be food-borne, as suggested by the presence of hepatitis E virus in pig liver as food. J Gen Virol 2003; 84:2351–2357. Tei S, Kitajima N, Takahashi K, Mishiro S. Zoonotic transmission of hepatitis E virus from deer to human beings. Lancet 2003; 362:371–373. Arankalle VA, Chobe LP. Retrospective analysis of blood transfusion recipients: evidence for post-transfusion hepatitis E. Vox Sang 2000; 79:72–74. Robson SC, Adams S, Brink N, et al. Hospital outbreak of hepatitis E. Lancet 1992; 339:1424–1425. Nubling M, Hofmann F, Tiller FW. Occupational risk for hepatitis A and hepatitis E among health care professionals? Infection 2002; 30:94–97. Tsarev SA, Tsarev TS, Emerson SU, et al. Experimental hepatitis E in pregnant rhesus monkeys: failure to transmit hepatitis E virus (HEV) to offspring and evidence of naturally acquired antibodies to HEV. J Infect Dis 1995; 172:31–37. Khuroo MS, Kamili S, Jameel S. Vertical transmission of hepatitis E virus. Lancet 1995; 345:1025–1026. Singh S, Mohanty A, Joshi YK, et al. Mother-to-child transmission of hepatitis E virus infection. Indian J Pediatr 2003; 70:37–39. Kumar RM, Uduman S, Rana S, et al. Sero-prevalence and mother-to-infant transmission of hepatitis E virus among pregnant women in the United Arab Emirates. Eur J Obstet Gynecol Reprod Biol 2001; 100:9–15. Meng XJ, Dea S, Engle RE, et al. Prevalence of antibodies to the hepatitis E virus in pigs from countries where hepatitis E is common or is rare in the human population. J Med Virol 1999; 58:297–302. Hsieh SY, Meng XJ, Wu YH, et al. Identity of a novel swine hepatitis E virus in Taiwan forming a monophyletic group with Taiwan isolates of human hepatitis E virus. J Clin Microbiol 1999; 37:3828–3834. Drobeniuc J, Favorov MO, Shapiro CN, et al. Hepatitis E virus antibody prevalence among persons who work with swine. J Infect Dis 2001; 184:1594–1597. Withers MR, Correa MT, Morrow M, et al. Antibody levels to hepatitis E virus in North Carolina swine workers, non-swine workers, swine, and murids. Am J Trop Med Hyg 2002; 66:384–388. Casares PC, Pina S, Buti M, et al. Hepatitis E virus epidemiology in industrialized countries. Emerg Infect Dis 2003; 9:448–454. Vaidya SR, Chitambar SD, Arankalle VA. Polymerase chain reaction-based prevalence of hepatitis A, hepatitis E and TT
Chapter 34 HEPATITIS E
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
201.
202.
203.
204.
205.
206.
207.
208.
209.
210.
viruses in sewage from an endemic area. J Hepatol 2001; 37:131–136. Vaidya SR, Tilekar BN, Walimbe AM, Arankalle VA. Increased risk of hepatitis E in sewage workers from India. J Occup Environ Med 2003; 45:1167–1170. Ceylan A, Ertem M, Ilcin E, Ozekinci T. A special risk group for hepatitis E infection: Turkish agricultural workers who use untreated waste water for irrigation. Epidemiol Infect 2003; 131:753–756. Montella F, Rezza G, Di Sora F, et al. Association between hepatitis E virus and HIV infection in homosexual men. Lancet 1994; 344:1433. Sylvan SP, Jacobson SH, Christenson B. Prevalence of antibodies to hepatitis E virus among hemodialysis patients in Sweden. J Med Virol 1998; 54:38–43. Dalekos GN, Zervou E, Elisaf M, et al. Antibodies to hepatitis E virus among several populations in Greece: increased prevalence in a hemodialysis unit. Transfusion 1998; 38:589–595. Ayoola EA, Want MA, Gadour MOEH, et al. Hepatitis E virus in hemodialysis patients: a case control study in Saudi Arabia. J Med Virol 2002; 66:329–334. Rivera M, Mateos ML, Teruel JL et al. Hepatitis E virus markers in a peritoneal dialysis population. PDI 2000; 20:576. Sylvan SPE, Hellstrom UB, Hampl H, et al. Hepatitis E in patients with chronic autoimmune liver disease. JAMA 1995; 273:377–378. Barzilai A, Schulman S, Karetnyi YV, et al. Hepatitis E virus infection in hemophiliacs. J Med Virol 1995; 46: 153–156. Buffet C, Laurent-Puig P, Chandot S, et al. A high hepatitis E virus seroprevalence among renal transplantation hemophilia patient populations. J Hepatol 1996; 24:122–123. Al-Fawaz I, Al-Rasheed S, Al-Mugeiren, et al. Hepatitis E virus infection in patients from Saudi Arabia with sickle cell anemia and beta-thalassemia major: possible transmission by blood transfusion. J Viral Hepat 1996; 3:203–205. Chandler JD, Riddell MA, Li F, et al. Serological evidence for swine hepatitis E virus infection in Australian pig herds. Vet Microbiol 1999; 68:95–105. Clayson ET, Innis BL, Myint KS, et al. Detection of hepatitis E virus infections among domestic swine in the Kathmandu Valley of Nepal. Am J Trop Med Hyg 1995; 53:228–232. Meng XJ, Purcell RH, Halbur PG, et al. A novel virus in swine is closely related to the human hepatitis E virus. Proc Natl Acad Sci USA 1997; 94:9860–9865. Takahashi M, Nishizawa T, Miyajima H, et al. Swine hepatitis E virus strains in Japan form four phylogenetic clusters comparable with those of Japanese isolates of human hepatitis E virus. J Gen Virol 2003; 84:851–862. Wang YC, Zhang HY, Xia NS, et al. Prevalence, isolation, and partial sequence analysis of hepatitis E virus from domestic animals in China. J Med Virol 2002; 67:516–521. Huang FF, Haqshenas G, Guenette DK, et al. Detection by reverse transcription-PCR and genetic characterisation of field isolates of swine hepatitis E virus from pigs in different geographic regions of the United States. J Clin Microbiol 2002; 40:1326–1332. Arankalle VA, Chobe LP, Joshi MV, et al. Human and swine hepatitis E viruses from Western India belong to different genotypes. J Hepatol 2002; 36:417–425. Arankalle VA, Joshi MV, Kulkarni AM, et al. Prevalence of antihepatitis E virus antibodies in different Indian animal species. J Viral Hepat 2001; 8:223–227. Hirano M, Ding X, Tran HT, et al. Prevalence of antibody against hepatitis E virus in various species of non-human primates: evidence of widespread infection in Japanese monkeys (Macaca fuscata). Jpn J Infect Dis 2003; 56:8–11.
211. Hirano M, Ding X, Li TC, et al. Evidence for widespread infection of hepatitis E virus among wild rats in Japan. Hepatol Res 2003; 27:1–5. 212. He J, Innis BL, Shrestha MP, et al. Evidence that rodents are a reservoir of hepatitis E virus for humans in Nepal. J Clin Microbiol 2002; 40:4493–4498. 213. Favorov MO, Kosoy MY, Tsarev SA, et al. Prevalence of antibody to hepatitis E virus among rodents in the United States. J Infect Dis 2000; 181:449–455. 214. Karetnyi YV, Dzhumalieva DI, Usmanov RK, et al. Possible involvement of rodents in the spread of viral hepatitis E. Zh Mikrobiol Epidemiol Immunobiol 1993; 4:52–56. 215. Clayson ET, Snitbhan R, Ngarmpochana M, et al. Evidence that the hepatitis E virus (HEV) is a zoonotic virus: detection of natural infections among swine, rats, and chickens in an area endemic for human disease. In: Buisson Y, Coursaget P, Kane M, eds. Enterically-transmitted hepatitis viruses. Joue-les-Tours, France: La Simarre; 1996:329–335. 216. Tien NT, Clayson HT, Khiem HB, et al. Detection of immunoglobulin G to the hepatitis E virus among several animal species in Vietnam. Am J Trop Med Hyg 1998; 59 (suppl):211. 217. Tsarev SA, Tsareva TS, Emerson SU, et al. ELISA for antibody to hepatitis E virus (HEV) based on complete open-reading frame-2 protein expressed in insect cells: identification of HEV infection in primates. J Infect Dis 1993; 168:369–378. 218. Arankalle VA, Goverdhan MK, Banerjee K. Antibodies against hepatitis E virus in Old World monkeys. J Viral Hepat 1994; 1:125–129. 219. Purcell RH, Emersol SU. Animal models of hepatitis A and E. ILAR 2001; 42:161–177. 220. Balayan MS, Zamyatina NA, Ivannikova TA, et al. Laboratory models in the studies of enterically transmitted non-A, non-B hepatitis. In: Shikata T, Purcell RH, Uchida T, eds. Viral hepatitis C, D and E. New York: Elsevier; 1991:213–220. 221. Kabrane-Lazizi Y, Fine JB, Elm J, et al. Evidence for widespread infection of wild rats with hepatitis E virus in the United States. Am J Trop Med Hyg 1999; 61:331–335. 222. Huang FF, Haqshenas G, Shivaprasad HL, et al. Heterogeneity and seroprevalence of a newly identified avian hepatitis E virus from chickens in the United States. J Clin Microbiol 2002; 40:4197–4202. 223. Crerar SK, Cross GM. The experimental production of big liver and spleen disease in broiler breeder hens. Aust Vet J 1994; 71:414–417. 224. Payne CJ, Ellis TM, Plant SL, et al. Sequence data suggests big liver and spleen disease virus (BLSV) is genetically related to hepatitis E virus. Vet Microbiol 1999; 68: 119–125. 225. Wu JC, Chen CM, Chiang TY, et al. Clinical and epidemiological implications of swine hepatitis E virus infection. J Med Virol 2000; 60:166–171. 226. Van der Poel WH, Verschoor F, van der Heide R, et al. Hepatitis E virus sequences in swine related to sequences in humans, the Netherlands. Emerg Infect Dis 2001; 7:970–976. 227. Pei Y, Yoo D. Genetic characterization and sequence heterogeneity of a Canadian isolate of swine hepatitis E virus. J Clin Microbiol 2002; 40:4021–4029. 228. Garkavenko O, Obriadina A, Meng J, et al. Detection and characterisation of swine hepatitis E virus in New Zealand. J Med Virol 2001; 65:525–529. 229. Nishizawa T, Takahashi M, Mizuo H, et al. Characterization of Japanese swine and human hepatitis E virus isolates of genotype IV with 99% identity over the entire genome. J Gen Virol 2003; 84:1245–1251. 230. Takahashi M, Nishizawa T, Okamoto H. Identification of a genotype III swine hepatitis E virus that was isolated from a Japanese pig born in 1990 and that is most closely related to Japanese isolates of human hepatitis E virus. J Clin Microbiol 2003; 41:1342–1343.
719
Section V. Liver Diseases Due to Infectious Agents
231. Meng XJ, Halbur PG, Haynes JS, et al. Experimental infection of pigs with the newly identified swine hepatitis E virus (swine HEV), but not with human strains of HEV. Arch Virol 1998; 143:1405–1415. 232. Meng XJ, Halbur PG, Shapiro MS, et al. Genetic and experimental evidence for cross-species infection by swine hepatitis E virus. J Virol 1998; 72:9714–9721. 233. Usmanov RK, Balayan MS, Dvoinikova OV, et al. An experimental infection in lambs by the hepatitis E virus. Voprosy Virusol 1994; 39:165–168. 234. Maneerat Y, Clayson ET, Myint KSA, et al. Experimental infection of the laboratory rat with the hepatitis E virus. J Med Virol 1996; 48:121–128. 235. Kuno A, Ido K, Isoda N, et al. Sporadic acute hepatitis E of a 47-year-old man whose pet cat was positive for antibody to hepatitis E virus. Hepatol Res 2003; 26:237–242. 236. Yajima Y, Takahashi N, Yamagishi H, et al. An acute hepatitis case domestically infected with a hepatitis E virus whose nucleotide sequence showed a high similarity to that from a domestic swine in Japan. Nippon Shokakibyo Gakkai Zasshi 2003; 100:454–458. 237. Arankalle VA, Chadha MS, Banerjee K, et al. Hepatitis E virus infection in pregnant rhesus monkeys. Indian J Med Res 1993; 97:4–8. 238. Tam A, White WR, Reed E, et al. In vitro propogation and production of hepatitis E virus from in-vivo infected primary macaque hepatocytes. Virology 1996; 215:1–9. 239. Tam A, White WR, Yarbough PO, et al. In-vitro infection and replication of hepatitis E virus in primary cynomolgus macaque hepatocytes. Virology 1997; 238:94–102. 240. Divizia M, Gabrieli R, Degener AM, et al. Evidence of hepatitis E virus replication on cell cultures. New Microbiol 1999; 22:77–83. 241. Huang R, Li D, Wei S, et al. Cell culture of sporadic hepatitis E virus in China. Clin Diagn Lab Immunol 1999; 6:729–733. 242. Huang R, Nakazono N, Ishii K, et al. Hepatitis E virus (87A strain) propogated in A549 cells. J Med Virol 1995; 47:299–302. 243. Huang RT, Li DR, Wei J, et al. Isolation and identification of hepatitis E virus in Xinjiang, China. J Gen Virol 1992; 73:1143–1148. 244. Panda SK, Ansari IH, Durapal H, et al. Synthesized RNA from a cDNA clone of hepatitis E virus is infectious. J Virol 2000; 74:2430–2437. 245. Williams TPE, Kasorndorkbua C, Halbur PG, et al. Meng XJ. Evidence of extrahepatic sites of replication of hepatitis E virus in a swine model. J Clin Microbiol 2001; 39:3040–3046. 246. Agarwal S, Gupta D, Panda SK. The 3¢ end of hepatitis E virus (HEV) genome binds specifically to the viral RNA-dependent RNA polymerase (RdRp). Virology 2001; 282:87–101. 247. Humphrey C, Cook EH, McCaustland K, et al. Entericallytransmitted non-A, non-B hepatitis (ET-NANBH): isolation of infectious virus from gall bladder bile. In Zukermann AJ, ed. The II international symposium on viral hepatitis and hepatocellular carcinoma. Taipei: Alan R. Liss, Inc.; 1988:abstract FP4. 248. Bradley DW, Krawczynski K, Cook EH, et al. Enterically transmitted non-A, non-B hepatitis: etiology of disease and laboratory studies in non human primates. In: Zukermann AJ, ed. Viral hepatitis and liver disease. New York: Alan R Liss; 1988:138–147. 249. Chauhan A, Jameel S, Dilawari JB, et al. Hepatitis E virus transmission to a volunteer. Lancet 1993; 341:149–150. 250. Ticehurst J, Popkin TJ, Bryan JP, et al. Association of hepatitis E virus with an outbreak of hepatitis in Pakistan: serologic responses and pattern of virus excretion. J Med Virol 1992; 36:84–92. 251. Krawczynski K, McCaustand K, Mast E, et al. Elements of pathogenesis of HEV infection in man and experimentally
720
252.
253.
254.
255.
256.
257.
258.
259.
260.
261.
262.
263.
264.
265.
266.
267.
268.
269.
270.
infected primates. In: Buisson Y, Coursage P, Kane M, eds. Enterically transmitted hepatitis viruses. Tours: La Simarre; 1996:317–328. Aggarwal R, Kini D, Sofat S, et al. Duration of viremia and fecal viral excretion in acute hepatitis E. Lancet 2000; 356:1081–1082. Nanda SK, Ansari JH, Acharya SK, et al. Protracted viremia during acute sporadic hepatitis E virus infection. Gastroenterology 1995; 108: 225–230. Claysoii ET, Myint KS, Snitbhan R, et al. Viremia, fecal shedding, and IgM and IgG responses in patients with hepatitis E. J Infect Dis 1995;172:927–933. Andzhaparidze AG, Balayan MS, Savinov AP, et al. Non-A, nonB hepatitis transmitted by the fecal–oral mode experimentally produced in monkeys. Vopr Virusol 1986; 31:73–81. Bradley DW, Krawczynski K, Cook EH Jr, et al. Enterically transmitted non-A, non-B hepatitis: serial passage of disease in cynomolgus macaques and tamarins and recovery of diseaseassociated 27- to 34-nm viruslike particles. Proc Natl Acad Sci USA 1987; 84:6277–6281. Longer CF, Denny SL, Caudill JD, et al. Experimental hepatitis E: Pathogenesis in cynomolgus macaques (Macaca fascicularis). J Infect Dis 1993; 168:602–609. Soe S, Uchida T, Suzuki K, et al. Enterically transmitted non-A, non-B hepatitis in cynomolgus monkeys: morphology and probable mechanism of hepatocellular necrosis. Liver 1989; 9:135–145. Ticehurst J, Rhodes LL Jr, Krawczynski K, et al. Infection of owl monkeys (Aotus trivirgatus) and cynomolgus monkeys (Macaca fascicularis) with hepatitis E virus from Mexico. J Infect Dis 1992; 165:835–845. Tsarev SA, Tsareva TS, Emerson SU, et al. Infectivity titration of a prototype strain of hepatitis E virus in cynomolgus monkeys. J Med Virol 1994; 43:135–142. Uchida T, Suzuki K, Komatsu K, et al. Occurrence and character of a putative causative virus of enterically-transmitted non-A, non-B hepatitis in bile. Jpn J Exp Med 1990; 60:23–29. Uchida T, Win KM, Suzuki K, et al. Serial transmission of a putative causative virus of enterically transmitted non-A, non-B hepatitis to Macaca fascicularis and Macaca mulatta. Jpn J Exp Med 1990; 60:13–21. Arankalle VA, Chadha MS, Chobe LP. Long-term serological follow up and cross-challenge studies in rhesus monkeys experimentally infected with hepatitis E virus. J Hepatol 1999; 30:199–204. Arankalle VA, Favorov MO, Chadha MS, et al. Rhesus monkeys infected with hepatitis E virus (HEV) from the former USSR are immune to subsequent challenge with an Indian strain of HEV. Acta Virol 1993; 37:515–518. Gupta H, Iyenger B, Tandon BN. Localization of a new enteric non-A, non-B (HEV) virus in target organ liver. Gastro Japonica 1993; 28:46–50. Gupta H, Tandon BN, Sriramachari S, et al. Animal transmission of enteric non-A, non-B hepatitis infection to Macaca mulatta by faeco–oral route. Indian J Med Res 1990; 91:87–90. Kawai HF, Koji T, Iida F, et al. Shift of hepatitis E virus RNA from hepatocytes to biliary epithelial cells during acute infection of rhesus monkeys. J Viral Hepat 1999; 6:287–297. Li Z. Experimental infection with hepatitis E virus in rhesus monkeys. Nat Med J (Taipei) 1992; 72:658–660, 701–702. Uchida T, Aye TT, Ma X, et al. An epidemic outbreak of hepatitis E in Yangon of Myanmar: antibody assay and animal transmission of the virus. Acta Pathol Jpn 1993; 43:94–98. van Cuyck-Gandré H, Cockman-Thomas R, Caudill JD, et al. Experimental African HEV infection in cynomolgus macaques (Macaca fascicularis). J Med Virol 1998; 55:197–202.
Chapter 34 HEPATITIS E
271. Yin S, Bao Z, Tian X, et al. Protection of rhesus monkeys from challenge of hepatitis E virus. Bull Acad Mil Med Sci 1992; 16:115–118. 272. Poleshchuk VF, Balayan MS, Andzhaparidze AG, et al. Modeling hepatitis A and enterically transmitted hepatitis non-A, non-B (hepatitis E) in tamarins Sagiunus mystax. Vopr Virusol 1990; 5:379–382. 273. Arankalle VA, Sreenivasan MA, Popper H, et al. Aetiological association of a virus-like particle with enterically transmitted non-A, non-B hepatitis. Lancet 1988; 1:550–554. 274. Sarthou JL, Budkowska A, Sharma MD, et al. Characterization of an antigen–antibody system associated with epidemic non-A, non-B hepatitis in West Africa and experimental transmission of an infectious agent to primates. Ann Virol Inst Pasteur 1986; 137E:225–232. 275. Balayan MS, Usmanov RK, Zamyatina NA, et al. Experimental hepatitis E infection in domestic pigs. J Med Virol 1990; 32:58–59. 276. Platt KB, Yoon K-J, Zimmerman JJ. Susceptibility of swine to hepatitis E virus and its significance to human health. In: Halbur PG, ed. Swine research report. Ames: Iowa State University; 1998:125–126. 277. Usmanov RK, Balayan MS, Dzhumalieva DI, et al. Experimental hepatitis E infection in piglets. Vopr Virusol 1991; 36:212– 216. 278. Usmanov RK, Balayan MS, Kazachkov YA, et al. Further studies of experimental hepatitis E in piglets. Vopr Virusol 1994; 5:208–212. 279. Usmanov RK, Balayan MS, Dvoinikova OV, et al. Experimental infection of lambs with hepatitis E virus. Vopr Virusol 1994; 39:165–168 280. Tsarev SA, Tsareva TS, Emerson SU, et al. Successful passive and active immunization of cynomolgus monkeys against hepatitis E. Proc Natl Acad Sci USA 1994; 91:10198– 10202. 281. Tsarev SA, Tsareva TS, Emerson SU, et al. Recombinant vaccine against hepatitis E: dose response and protection against heterologous challenge. Vaccine 1997; 15:1834–1838. 282. Favorov MO, Fields HA, Purdy MA, et al. Serological identification of hepatitis E virus infections in epidemic and endemic settings. J Med Virol 1992; 36:246–250. 283. Favorov MO, Khudyakov YE, Mast EE, et al. IgM and IgG antibodies to hepatitis E virus (HEV) detected by an enzyme immunoassay based on an HEV-specific artificial recombinant mosaic protein. J Med Virol 1996; 50:50–58. 284. Dawson GJ, Chau KH, Cabal CM, et al. Solid-phase enzymelinked immunosorbent assay for hepatitis E virus IgG and IgM antibodies utilizing recombinant antigens and synthetic peptides. J Virol Methods 1992; 38:175–186. 285. Khuroo MS, Kamili S, Dar MY, et al. Hepatitis E and long-term antibody status. Lancet 1993; 341:1355. 286. Anderson DA. Waterborne hepatitis. In: Spencer SC, Hodinka RL, Young SA, eds. Clinical virology manual. Washington, DC: ASM Press; 2000:295–305. 287. Chau KH, Dawson GJ, Bile KM, et al. Detection of IgA class antibody to hepatitis E virus in serum samples from patients with hepatitis E virus infection. J Med Virol 1993; 40:334–338. 288. Naik S, Aggarwal R, Naik SR, et al. Evidence for activation of cellular immune responses in patients with acute hepatitis E. Indian J Gastroenterol 2002; 21:149–152. 289. Mast E, Krawczynski K: Hepatitis E: an overview. Ann Rev Med 1996; 47:257. 290. Bryan J, Tsarev S, Iqbal M, et al. Epidemic hepatitis E in Pakistan: patterns of serologic response and evidence that antibody to hepatitis E protects against disease. J Infect Dis 1994; 170:517. 291. Gupta DN, Smetana HF. The histopathology of viral hepatitis as seen in the Delhi epidemic (1955–56). Indian J Med Res 1957; 45 (suppl):101–113.
292. Khuroo MS, Saleem M, Teli MR, et al. Failure to detect chronic liver disease after epidemic non-A, non-B hepatitis. Lancet 1980; 2:97–98. 293. Mechnik L, Bergman N, Attali M, et al. Acute hepatitis E virus infection presenting as a prolonged cholestatic jaundice. J Clin Gastroenterol 2001; 33:421–422. 294. Balayan MS. HEV infection: Historical perspectives, global epidemiology, and clinical features. In: Hollinger FB, Lemon SM, Margolis H, eds. Viral hepatitis and liver disease. Baltimore: Williams & Wilkins; 1991:498–501. 295. Glikson M, Galun E, Oren R, et al. Relapsing hepatitis A: review of 14 cases and literature survey. Medicine 1992; 71:14–23. 296. Sanyal MC. Infectious hepatitis in Delhi (1955–56): a critical study. Observations in armed forces personnel. Indian J Med Res 1957; 45 (suppl):91–99. 297. Krawczynski K, Bradley DW. Enterically transmitted non-A, nonB hepatitis: identification of virus-associated antigen in experimentally infected cynomolgus macaques. J Infect Dis 1989; 159:1042–1049. 298. Purdy M, Carson D, McCaustland K, et al. Viral specificity of hepatitis E virus antigens identified by fluorescent antibody assay using recombinant HEV proteins. J Med Virol 1994; 44:212. 299. Khudyakov YE, Lopareva EN, Jue DL, et al. Antigenic domains of the open reading frame 2-encoded protein of hepatitis E virus. J Clin Microbiol 1999; 37:2863–2871. 300. Li TC, Zhang J, Shinzawa H, et al. Empty virus-like particle based enzyme-linked immunosorbent assay for antibodies to hepatitis E virus. J Med Virol 2000; 62:327–333. 301. Ghabrah TM, Tsarev S, Yarbough PO, et al. Comparison of tests for antibody to hepatitis E virus. J Med Virol 1998; 55: 134–137. 302. Yarbough PO, Garza E, Tam AW, et al. Assay development of diagnostic tests for IgM and IgG antibody to hepatitis E virus. In: Buisson Y, Coursaget P, Kane M, eds. Enterically-transmitted hepatitis viruses. Joue-les-Tours, France: La Simarre; 1996:294–296. 303. Lin CC, Wu JC, Chang TT, et al. Diagnostic value of immunoglobulin G (IgG) and IgM anti-hepatitis E virus (HEV) tests based on HEV RNA in an area where hepatitis E is not endemic. J Clin Microbiol 2000; 38:3915–3918. 304. Genelabs. Technical bulletin 1998. 305. Myint H, Soe MM, Khin T, et al. A clinical and epidemiological study of an epidemic of non-A, non-B hepatitis in Rangoon. Am J Trop Med Hyg 1985; 34:1183–1189. 306. Zhuang H, Cao X-Y, Liu C-B, et al. Enterically transmitted nonA, non-B hepatitis in China. In: Shikata T, Purcell RH, Uchida T, eds. Viral hepatitis C, D, and E. New York: Elsevier Science; 1991:277–285. 307. Tsega E, Krawczynski K, Hansson BG, et al. Outbreak of acute hepatitis E virus infection among military personnel in northern Ethiopia. J Med Virol 1991; 34:232–236. 308. Aacharya SK, Batra Y, Hhazari S, et al. Etiopathogenesis of acute hepatic failure: eastern versus western countries. J Gastroenterol Hepatol 2002; 17 (suppl 3):S268–S273. 309. Acharya SK, Panda SK, Saxena A, Gupta SD. Acute hepatic failure in India: a perspective from the east. J Gastroenterol Hepatol 2000;15:473–479. 310. Khuroo MS. Acute liver failure in India. Hepatology 1997; 26:244–246. 311. Jaiswal SB, Chitnis DS, Asolkar MV, et al. Aetiology and prognostic factors in hepatic failure in central India. Trop Gastroenterol 1996; 17:217–20. 312. Arora NK, Nanda SK, Gulati S, et al. Acute viral hepatitis types E, A, and B singly and in combination in acute liver failure in children in north India. J Med Virol 1996; 48:215–221. 313. Khuroo MS, Kamili S. Aetiology and prognostic factors in acute liver failure in India. J Viral Hepat 2003; 10:224–231.
721
Section V. Liver Diseases Due to Infectious Agents
314. Bendre SV, Bavdekar AR, Bhave SA, et al. Fulminant hepatic failure: etiology, viral markers and outcome. Indian Pediatr 1999; 36:1107–1112. 315. Nanda SK, Yalcinkaya K, Panigrahi AK, et al. Etiological role of hepatitis E virus in sporadic fulminant hepatitis. J Med Virol 1994; 42:133–137. 316. Williams R, Wendon J. Clinical syndrome and aetiology of fulminant hepatic failure. In: Williams R, Hughes RD, eds. Acute liver failure: improved understanding and better therapy. Proceedings of the 11th BSG/SK and F international workshop 1990. Herts, UK: SK & F; 1991:1–5. 317. Ostapowicz G, Lee WM. Acute hepatic failure: a western perspective. J Gastroenterol Hepatol 2000; 15:480–488. 318. Sallie R, Chiyende J, Tan KC, et al. Fulminant hepatic failure resulting from coexistent Wilson’s disease and hepatitis E. Gut 1994; 35:849–853. 319. Hamid S, Atiq M, Shezad F, et al. Hepatitis E super infection in patients with chronic liver disease. Hepatology 2002; 36:474–478. 320. Ramachandran J, Eapen C, Gagandeep K, et al. Hepatitis E superinfection produces severe decompensation in patients with chronic liver disease. J Gastroenterol Hepatol 2004; 19:134–138. 321. Khuroo MS, Teli MR, Skidmore S, et al. Incidence and severity of viral hepatitis in pregnancy. Am J Med 1981; 70:252–255. 322. Khuroo MS, Kamili S. Aetiology, clinical course and outcome of sporadic acute viral hepatitis in pregnancy. J Viral Hepat 2003; 10:61–69. 323. Patra S, Trivedi SS, Wadhwan M, et al. Viral hepatitis in pregnancy; etiology, clinical course and outcome: a study of 144 patients. Liver Int 2005, in press. 324. Sood A, Midha V, Sood N. Guillain–Barré syndrome with acute hepatitis E. Am J Gastroenterol 2000; 95:3667–3668. 325. Abid S, Khan AH. Severe hemolysis and renal failure in glucose6-phosphate dehydrogenase deficient patients with hepatitis E. Am J Gastroenterol 2002; 97:1544–1547. 326. Maity SG, Ray G. Severe acute pancreatitis in acute hepatitis E. Indian J Gastroenterol 2002 ;21:37–38. 327. Bile K, Isse A, Mohamud O, et al. Contrasting roles of rivers and wells as sources of drinking water on attack and fatality rates in a hepatitis E epidemic in Somalia. Am J Trop Med Hyg 1994; 51:466–474. 328. Corwin AL, Khiem HB, Clayson ET, et al. A waterborne outbreak of hepatitis E virus transmission in south-western Vietnam. Am J Trop Med Hyg 1996; 54:559–562. 329. Zhang M, Emerson SU, Nguyen H, et al. Recombinant vaccine against hepatitis E: duration of protective immunity in rhesus macaques. Vaccine 2002; 20:3285–3291. 330. Purcell RH, Nguyen H, Shapiro M, et al. Preclinical immunogenicity and efficacy trial of a recombinant hepatitis E vaccine. Vaccine 2003; 21:2607–2615. 331. Khuroo MS, Dar MY, Hepatitis E. Evidence for person-toperson transmission and inability of low dose immune serum globulin from an Indian source to prevent it. Indian J Gastroenterol 1992; 11:113–116. 332. Joshi YK, Babu S, Sarin S, et al. Immunoprophylaxis of epidemic non-A, non-B hepatitis. Indian J Med Res 1985; 81:18–19. 333. Arankalle VA, Chadha MS, Dama BM, et al. Role of immune serum globulins in pregnant women during an epidemic of hepatitis E. J Viral Hepat 1998; 5:199–214. 334. Chauhan A, Dilawari JB, Sharma R, et al. Role of long-persisting human hepatitis E virus antibodies in protection. Vaccine 1998; 16:755–756. 335. Pillot J, Turkoglu S, Dubreuil P, et al. Cross-reactive immunity against different strains of the hepatitis E virus transferable by simian and human sera. CR Acad Sci III 1995; 318:1059–1064.
722
336. Burton DR, Barbas CF. Human antibodies from combinational libraries. Adv Immunol 1994;57:191–280. 337. Schofield DJ, Purcell RH, Nguyen T, Emerson SU. Monoclonal antibodies that neutralise HEV recognise an antigenic site at the carboxyterminus of an ORF2 protein vaccine. Vaccine 2003; 22;257–267. 338. Longer CF, Elliot JE, Caudill JD, et al. Observations on subclinical hepatitis E virus (HEV) infection and protection against reinfection. In: Buisson Y, Coursaget P, Kane M, eds. Enterically-transmitted hepatitis viruses. Joue-les-Tours, France: La Simarre; 1996:362–372. 339. Kaur M, Hyams KC, Purdy MA, et al. Human linear B cell epitopes encoded by hepatitis E virus include determinants in the RNA-dependent RNA polymerase. Proc Natl Acad Sci USA 1992; 89:3855–3858. 340. Coursaget P, Buisson I, Depril N, et al. Mapping of linear B cell epitopes on open reading frame 2 and 3-encoded proteins of hepatitis E virus using synthetic peptides. FEMS Microbiol Lett 1993; 109:252–255. 341. Khudyakov YE, Favorov MO, Jue DL, et al. Immunodominant antigenic regions in a structural protein of the hepatitis E virus. Virology 1994; 198:390–393. 342. Jhudyakov YE, Khudyakova NS, Jue DL, et al. Comparative characterization of antigenic epitopes in the immunodominant region of the protein encoded by open reading frame 3 in Burmese and Mexican strains of hepatitis E virus. J Gen Virol 1994; 75:641–646. 343. Meng JH, Dubreuil P, Pillot J. A new PCR-based seroneutralization assay in cell culture for diagnosis of hepatitis E. J Clin Microbiol 1997; 35:1373–1377. 344. Purdy MA, McCaustland KA, Krawczynski K, et al. Expression of a hepatitis E virus (HEV)-trpE fusion protein containing recognized by antibodies in sera from human cases and experimentally infected primates. Arch Virol 1992; 123:335–349. 345. Meng JH, Pillot J, Dai X, et al. Neutralization of different geographical strains of the hepatitis E virus with anti-hepatitis E virus-positive serum samples obtained from different sources. Virology 1998; 249:316–324. 346. Meng J, Dai X, Chang JC, et al. Identification and characterization of neutralization epitope(s) of hepatitis E virus. Virology 2001; 288:203–211. 347. Zhang J, Gu Y, Ge SX, et al. Analysis of hepatitis E virus neutralization sites using monoclonal antibodies directed against a virus capsid protein. Vaccine 2005; 23:2881–2892. 348. Purdy MA, McCaustland KA, Krawczynski K, et al. Preliminary evidence that a trpE-HEV fusion protein protects cynomolgus macaques against challenge with wild type hepatitis E virus (HEV). J Med Virol 1993; 41:90–94. 349. Im SWK, Zhang JZ, Zhuang H, et al. A bacterially expressed peptide prevents experimental infection of primates by the hepatitis E virus. Vaccine 2001; 19:3726–3732. 350. Li SW, Zhang J, Li YM, et al. A bacterially expressed particulate hepatitis E vaccine: antigenicity, immunogenicity and protectivity on primates. Vaccine 2005; 23:2893–2901. 351. Fuerst TR, Yarbough PO, Zhang Y, et al. Prevention of hepatitis E using a novel Orf2 subunit vaccine. In: Buisson Y, Coursaget P, Kane M, eds. Enterically-transmitted hepatitis viruses. Joue-lesTours, France: La Simarre; 1996:384–392. 352. Man Y, Lin SQ, Gao Y, et al. Expression of ORF2 partial gene of hepatitis E virus in tomatoes and immunoreactivity of expressed products. World J Gastroenterol 2003; 9:2211–2215. 353. Maloney BJ, Takeda N, Suzaki Y, et al. Challenges in creating a vaccine to prevent hepatitis E. Vaccine 2005; 23:1870– 1874. 354. Sarzotti M, Dean TA, Remington MP, et al. Induction of CTL responses in newborn mice by DNA immunization. Vaccine 1997; 15:795–797.
Chapter 34 HEPATITIS E
355. Yokoyama M, Hassett DE, Zhang J, Whitton JL. DNA immunization can stimulate florid local inflammation, and the antiviral immunity induced varies depending on the injection site. Vaccine 1997; 15:553–560. 356. Pardoll DM, Beckery AM. Exposing the immunology of naked DNA vaccine. Immunity 1995; 3:165–169. 357. He J, Hoffamn SL, Hayes CG. DNA inoculation with a plasmid vector carrying the hepatitis E virus structural protein gene induces immune response in mice. Vaccine 1997; 15:357–362. 358. He J, Binn LN, Caudill JD, et al. Antiserum generated by DNA vaccine binds to the native hepatitis E virus (HEV) as determined by PCR and immune electron microscopy (IEM): application for HEV detection by affinity-capture RT-PCR. Vir Res 199; 62:59–65. 359. He J, Hayes CG, Binn LN, et al. Hepatitis E virus DNA vaccine elicits immunologic memory in mice. J Biomed Sci 2001; 8:223–226. 360. Kamili S, Spelbring J, Krawczynski K. DNA vaccination against hepatitis E virus infection in cynomologous macaques. J Gastroenterol Hepatol 2002; 17 (suppl 3):S365–S369. 361. Kamili S, Prelbring J, Carson D, Krawczynski K. Protective efficacy of hepatitis E virus DNA vaccine administered by gene
362. 363. 364. 365.
366.
367.
368.
gun in the cynomologous macaque model of infection. J Infect Dis 2004; 189:258–264. Safary A. Perspectives of vaccination against hepatitis E. Intervirology 2001; 44:162–166. Stevenson P. Nepal calls the shots in hepatitis E vaccine trial. Lancet 355:1623. Vishwanath R, Sidhu AS. Infectious hepatitis: clinical findings. Indian J Med Res 1957; 45 (suppl):49–58. Morrow RH, Smetana HF, Sai FT, et al. Unusual features of viral hepatitis in Accra, Ghana. Ann Intern Med 1968; 68:1250–1264. Bryan JP, Iqbal M, Tsarev S, et al. Epidemic of hepatitis E in a military unit in Abottabad, Pakistan. Am J Trop Med Hyg 2002; 67:662–668. Zhang M, Emerson SU, Nguyen H, et al. Immunogenicity and protective efficacy of a vaccine prepared from 53 kDa truncated hepatitis E virus capsid protein expressed in insect cells. Vaccine 2002; 20:853–857. Aggarwal R, Krawczynski K. Hepatitis E: an overview and recent advances in clinical and laboratory research. J Gastroenterol Hepatol 2000; 15:9–2000
723
Section V: Liver Diseases Due to Infectious Agents
35
OTHER HEPATITIS VIRUSES Mahmoud M. Yousfi, Jorge Rakela Abbreviations anti-S soluble complement-fixing antigens CDC Centers for Disease Control and Prevention CMV cytomegalovirus EBV Epstein-Barr virus EBVNA EBV nuclear antigen FHF fulminant hepatic failure HBsAg hepatitis B surface antigen
HBV HCC HGV HHV-6 HHV-7 HHV-8 HPV-B19
hepatitis B virus hepatocellular carcinoma hepatitis G virus human herpes virus-6 human herpes virus-7 human herpes virus-8 human parvovirus B19
INTRODUCTION There are viral agents apart from hepatitis viruses that may affect the liver as part of systemic involvement. This may include acute hepatitis or, in some instances, acute liver failure. They may lead to fulminant hepatitis. These agents’ ability to cause chronic liver disease has not been proven unequivocally, although it seems unlikely. As the cause of acute liver failure remains unknown in a significant proportion of cases, these viral agents have been evaluated as etiological factors. These agents include cytomegalovirus, Epstein–Barr virus, herpes simplex virus, varicella zoster virus, human herpesviruses 6, 7, and 8, human parvovirus B19, adenoviruses, occult hepatitis B virus, and more recently, SARSassociated coronavirus.
CYTOMEGALOVIRUS (CMV) Human cytomegalovirus (CMV) is the largest member of the b herpesviridae family of viruses. Cytomegaly (giant cell) and prominent intranuclear inclusion bodies characterize the cellular response to CMV infection. CMV infections are quite common, reaching 60–70% in urban populations, and play a significant role as an opportunistic pathogen in immunocompromised hosts. Early recognition of infection, institution of therapy, and prevention of infection are critical in altering the outcome in these patients.1,2 Several factors determine the manifestations and severity of CMV infection. Infection is acquired either in the perinatal period and infancy or in adulthood through sexual contact, blood transfusions, or organ transplantation. Most primary CMV infections in immunocompetent adults are either asymptomatic or associated with a mild mononucleosis-like syndrome. As with other herpes viruses, all primary infections resolve and enter into lifelong latency, in which live virus is sequestered in a non-replicative state. Persons with latent infection and an intact immune system have no symptoms but exhibit antibodies to CMV. Circulating lymphocytes, monocytes, and polymorphonuclear leukocytes may serve as the predominant site of viral latency. The risk for intermittent reactivation is increased with diminished host immune status. In immunocompro-
HSV PCR SARS-CoV SEN-V TTV VCA VZV
herpes simplex virus polymerase chain reaction SARS-associated coronavirus SEN-virus TT virus viral capsid antigens varicella zoster virus
mised patients, CMV disease can result from either a primary infection in a previously uninfected (seronegative) host, or more commonly from reactivation of latent infection. Although adequate anti-CMV antibodies are detected during episodes of reactivation of infection, cell-mediated immunity (characterized by decreased numbers of cytotoxic T lymphocytes and natural killer cells) is defective. The incidence and severity of CMV disease closely parallel the degree of cellular immune dysfunction.3,4 A wide spectrum of clinical syndromes associated with CMV disease ranges from asymptomatic infection, life-threatening congenital CMV syndrome in neonates, infectious mononucleosis syndrome in young adults, to severe pulmonary, retinal, neurological, gastrointestinal, and hepatic diseases in immunocompromised hosts, in whom CMV is a very common opportunistic pathogen.5 Congenital CMV infection is associated with substantial morbidity and mortality and is manifest shortly after birth by jaundice, hepatosplenomegaly, thrombocytopenic purpura, and severe neurological symptoms. Multifocal hepatic necrosis with cytomegalic cells, intranuclear inclusion bodies, inflammatory response, and marked bile stasis may be detected on liver biopsy. If the child survives the jaundice and hepatosplenomegaly may subside, but the neurological sequelae and mental retardation persist.6–8 A different form of neonatal CMV infection occurs as a result of perinatal (from the mother’s cervix during delivery) or postnatal (from breastfeeding) transmission of the virus, resulting in a clinical picture resembling mild infectious mononucleosis syndrome without neurological involvement. A mild self-limiting hepatitis may occur, but usually resolves during the first year of life.9–11 In immunocompetent children and adults CMV infection is usually subclinical, but can sometimes cause a disease that resembles EBV infectious mononucleosis syndrome. Unlike in EBV mononucleosis, pharyngitis and cervical lymphadenopathy are absent and the heterophil response is negative (negative monospot).12,13 The mode of transmission for these patients is through sexual contact, kissing, intrafamilial transmission (sharing objects with contaminated saliva among family members), and blood transfusion. In surgical patients requiring massive blood transfusions CMV infection should be considered as a source of postoperative fever (sometimes called postperfusion syndrome).14
725
Section V. Liver Diseases Due to Infectious Agents
Liver dysfunction is commonly associated with CMV mononucleosis. It is usually mild and rarely symptomatic in the immunocompetent patient. Hepatosplenomegaly and laboratory evidence of mild to moderate hepatic dysfunction are the predominant features, with increased transaminases and alkaline phosphatase values in 88% and 64% of cases, respectively, but lower than commonly encountered in active viral hepatitis.15,16 Rare manifestations of CMV hepatitis include tender hepatomegaly, granulomatous hepatitis (with scattered microscopic granulomas found on liver biopsy), anicteric or icteric cholestatic form of hepatitis, and acute hepatitis with massive hepatic necrosis.17–24 In patients with impaired cell-mediated immunity, disseminated CMV infection results in serious life threatening diseases. CMV is the most common opportunistic viral infection in AIDS patients, causing retinitis, central nervous system infections, esophagitis, and colitis. CMV may also invade the hepatobiliary tract in AIDS patients, causing hepatitis, pancreatitis, and acute acalculous gangrenous cholecystitis.25–27 The presence of cytomegalovirus retinitis, gastrointestinal disease, or viremia in AIDS patients increases the risk for the development of a cholestatic syndrome caused by papillary stenosis and sclerosing cholangitis, which does not usually respond to antiviral therapy.28 Other immunocompromised patients at risk are organ transplant recipients, including liver transplantation. The diagnosis of CMV hepatitis always requires confirmatory laboratory tests, as the clinical presentation alone is not sufficient to establish the diagnosis. Serologic studies of CMV IgM antibodies may be helpful in primary infections.29 Viral culture technique could be greatly enhanced with the use of ‘shell vial’ assays, using CMV early antigens.30,31 Using molecular techniques to detect CMV early antigen or the CMV DNA polymerase increased the sensitivity of detection of CMV infection in the blood or tissue.32,33 Liver biopsy with distinct pathologic findings is important in establishing the diagnosis in CMV hepatitis, especially in the immunocompromised host. Giant multinucleated cell reaction with an inflammatory response, multifocal necrosis, and biliary stasis are commonly found. Large nuclear inclusion-bearing cells, sometimes called ‘owl’s eye’ inclusions, can be found in hepatocytes or bile duct epithelium.34,35 Treatment of CMV with antiviral agents is not always indicated, especially in self-limited disease in immunocompetent adults. For severe and worrisome cases, particularly in patients with impaired cell-mediated immunity, therapy can be life-saving. Aciclovir is ineffective.36 Ganciclovir is a nucleoside analog of guanosine and a homolog of aciclovir that has a longer intracellular half-life. It is considered the antiviral agent of choice against CMV. The duration of therapy should be guided by repeated measurement of CMV in blood samples. Emerging resistant strains to ganciclovir pose a therapeutic challenge, where foscarnet or cidofovir may become alternative antiviral agents.37
EPSTEIN–BARR VIRUS (EBV) Epstein–Barr virus (EBV) shares the characteristic morphologic features of the herpesviridae family. The EBV genome consists of a linear DNA molecule that encodes nearly 100 viral proteins. After infecting B lymphocytes the linear EBV genome becomes circular,
726
forming an episome, which usually remains latent in these B cells. Viral replication is spontaneously activated in only a small percentage of latently infected B cells.38 Transmission of EBV usually occurs via contact with oral secretions (saliva droplets, or possibly cells in saliva). Transmission by blood transfusion has been reported but is very unusual.39 The virus replicates in the nasopharyngeal epithelial cells, and nearly all seropositive persons actively shed virus in the saliva. B cells in the oropharynx may be the primary site of infection.40 Resting memory B cells are thought to be the site of persistence of EBV in the body. Researchers were able to identify various clinical conditions associated with EBV, such as infectious mononucleosis, Burkitt’s lymphoma, nasopharyngeal carcinoma, Hodgkin’s disease, peripheral T-cell lymphoma, and post-transplant lymphoproliferative disease.41 EBV infection is very common, infecting over 90% of humans worldwide and persisting for the lifetime of the person. Hepatosplenomegaly and palatal petechiae may be present in more than 10% of symptomatic patients. Liver involvement is well recognized in EBV infections. Manifestations of liver involvement range from the most commonly encountered mild self-limiting acute hepatitis to occasional reports of fatal acute fulminant hepatitis. Mild elevation of aminotransferase two to three times the upper limit of normal, and elevated lactic dehydrogenase levels are seen in up to 90% of cases of infectious mononucleosis. Typically, the rise in aminotransferases is gradual, reaching a peak that is lower than that commonly encountered in acute viral hepatitis. The rise occurs over a 1–2-week period, then aminotransferases decline gradually over 3–4 weeks.15,42–46 Patients older than 30 years generally have a more severe disease than do children. Mild elevation of alkaline phosphatase levels is also seen in 60% and mild hyperbilirubinemia in about 45%.42 Severe cholestatic jaundice and right upper quadrant abdominal pain may occur in elderly patients. Jaundice may occasionally be the initial clinical presentation, in combination with fever and abdominal pain, and can be mistaken for extrahepatic biliary obstruction. Jaundice occurs predominantly when EBV infection is complicated with autoimmune hemolytic anemia, and occasionally as a direct result of virus-induced cholestasis.47–51 Other occasional clinical settings for EBV liver involvement include post-transfusion hepatitis, granulomatous hepatitis, and fatal fulminant hepatitis. In some cases of granulomatous hepatitis serologic evidence of chronic EBV infection was found.52,53 A detailed clinicopathologic analysis of 30 patients with sporadic fatal infectious mononucleosis was described by Markin et al.54 Cases of fatal fulminant hepatitis with massive hepatic necrosis and disseminated intravascular coagulation were reported in both immunocompromised and immunocompetent hosts.55–59 The diagnosis of infectious mononucleosis is established on the basis of the clinical features, laboratory and serological findings indicative of a recent EBV infection. The most common hematological findings include leukocytosis in 70% of cases, with predominantly lymphocytosis and monocytosis, and mild thrombocytopenia in up to 50%. The ‘monospot’ test that detects heterophile antibodies, although sensitive, is not very specific. EBV-specific IgG and IgM antibodies, directed against the viral capsid antigens (VCA), early antigens (EBV anti-D and anti-R), nuclear antigen (EBVNA),
Chapter 35 OTHER HEPATITIS VIRUSES
and soluble complement-fixing antigens (anti-S), improve sensitivity and specificity in detecting the infection.60–62 With liver involvement, abdominal ultrasound may show a fatty liver appearance or gallbladder wall thickening.63,64 In the vast majority of cases there is no indication for liver biopsy. There may be portal and sinusoidal mononuclear cell infiltration with focal hepatic necrosis or fatty infiltration. Multinucleated giant cells are not seen.39 Of particular utility as diagnostic methods are in situ hybridization, Southern blot analysis, and polymerase chain reaction to identify specific RNA or DNA sequences in the organs involved.65,66 The differential diagnosis of EBV hepatitis includes other viral hepatitis A–E, cytomegalovirus hepatitis, and drug-induced hepatitis. Cervical lymphadenopathy is less common and peripheral monocytosis is encountered as observed with CMV hepatitis.12 There is no specific drug or treatment for EBV infection. Aciclovir inhibits EBV in vitro replication and reduces viral shedding in the oropharynx, but has no effect on the symptoms of infectious mononucleosis (which are primarily due to immune response to the virus) and is therefore not recommended.67 In EBV hepatitis no antiviral agent has proved to be effective. There is one single report of fulminant hepatic failure in an immunocompetent young girl caused by primary EBV infection that was treated by orthotopic liver transplantation.68
HERPES SIMPLEX VIRUS (HSV) Herpes simplex virus (HSV-1 and HSV-2) is a common infection in humans and produces a wide variety of illnesses, including mucocutaneous infection, infections of the central nervous system, and an occasional infection of the visceral organs. The clinical manifestations and course of HSV infections depend mainly on the site involved and the host’s age and immune status. Occasionally, HSV viremia results in visceral involvement, affecting mainly three organs: the esophagus, lungs, and liver. Liver involvement occurs in the following settings: neonatal infections, pregnancy, immunocompromised hosts, and rarely immunocompetent adults. In neonates, hepatitis occurs with multiorgan involvement and usually carries a high mortality rate. Fulminant hepatitis caused by HSV was first described by Hass in 1935 in a neonate with liver and adrenal necrosis associated with distinctive intranuclear inclusions.69 Several subsequent reports have shown that acute fulminant hepatitis and adrenal insufficiency remain the most common causes of death in neonates with disseminated HSV infection.70 The delay in instituting antiviral therapy against HSV, while awaiting confirmation of the diagnosis, results in a catastrophic outcome.71 HSV hepatitis in pregnant women was first reported in 1969 and was seen in the context of disseminated primary infection, usually late in gestation – 65% in the third trimester – and usually manifests as acute fulminant hepatitis.72–74 Mucocutaneous lesions are present in only half of cases; therefore, the clinical suspicion for diagnosis of this condition must be high. Twenty-five percent of cases were not diagnosed until autopsy. Early recognition, with initiation of antiviral therapy, may reverse an otherwise fatal process.75–79 HSV is an uncommon cause of hepatitis in immunocompetent patients. A mild asymptomatic elevation of transaminase levels
can be detected in 14% of healthy adults with acute genital herpes infection.80 Fulminant hepatitis with more than a 100-fold rise in transaminases was reported and associated with a favorable outcome after antiviral therapy.81–83 In immunocompromised hosts HSV hepatitis has occurred during primary and rarely during recurrent infection, with a triad of fever, leukopenia, and markedly elevated aminotransferases being suggestive of the diagnosis. Liver biopsy is essential to establish the diagnosis of HSV hepatitis, especially in pregnancy. It usually shows focal, sometimes extensive, hemorrhagic or coagulative necrosis of the hepatocytes, with limited inflammatory response (usually mononuclear and scattered lymphocytes). Typical intranuclear inclusions (Cowdry A type) are often identified at the margins of the foci of necrosis. The diagnosis is confirmed by the detection of HSV DNA sequences by PCR, which is more sensitive than tissue culture methods.75,81,84,85 HSV hepatitis is one of the infectious disease emergencies associated with a rapid and lethal course and requires early recognition and the institution of antiviral therapy while awaiting confirmation of the diagnosis, in order to improve outcome.86,87 At the Mayo Clinic the incidence of HSV hepatitis was reported to be 6% among all fulminant hepatitis patients reviewed from 1974 to 1982.88 Highdose aciclovir is the antiviral drug of choice (at least 10 mg/kg/day every 8 hours) and has been successfully used.75,89–91 Shanley92 reported a case of a healthy female who developed disseminated HSV-2 infection and fulminant hepatitis during the third trimester of pregnancy requiring high-dose antiviral therapy, which resulted in eradication of HSV mucocutaneous lesions. However, the patient’s condition continued to deteriorate, leading to orthotopic liver transplantation. Recurrence was not observed, suggesting that disseminated HSV infection should not be an absolute contraindication for transplantation in certain clinical settings. A more recently published series demonstrated the utility of liver transplantation in this setting.93
VARICELLA ZOSTER VIRUS (VZV) Varicella zoster virus causes two distinct clinical diseases. Varicella (commonly called chickenpox) is the primary infection, which is characterized as a benign generalized exanthematous rash. Recurrence of infection results in a more localized phenomenon known as herpes zoster (often called shingles).94 Rare non-cutaneous manifestations, such as encephalitis, pneumonitis, myocarditis, and hepatitis, may accompany the skin rash, especially in immunocompromised patients, and may be life-threatening.95 Mild and transient liver enzyme abnormalities are not uncommon in varicella infection in children and can occur in up to 25%.96–98 Primary infection in immunocompetent adults may cause severe acute hepatitis with a more than 10-fold increase in transaminases,8 and sometimes fulminant hepatic failure with evidence of VZV in liver and other organs is only revealed on autopsy.99 In contrast to the rather benign course of zoster (reactivation of infection) in the setting of organ transplantation, primary varicella infection can be quite aggressive.100 Visceral involvement, including the liver, may occur in the immediate postoperative period or may be delayed several months after transplantation. Usually it is associated with rapid onset and fatal fulminant hepatitis.101–105
727
Section V. Liver Diseases Due to Infectious Agents
Serologic testing is of little use, especially in immunocompromised patients. Confirmation of diagnosis is possible through the isolation of VZV from the skin lesions or other affected organ. Liver biopsy often shows foci of coagulative necrosis and intranuclear inclusions with an inflammatory response. PCR and immunoperoxidase techniques may be needed to distinguish VZV from HSV hepatitis. CDC guidelines for the prevention and control of nosocomial infections must be instituted for infection control in hospital personnel.106 Early administration of antiviral therapy is critical in the setting of VZV hepatitis, especially in immunocompromised patients. The drug of choice is intravenous aciclovir 30 mg/kg/day in three divided doses for 7–10 days.107,108
HUMAN HERPES VIRUS-6 (HHV-6) HHV-6 infects nearly all humans by the age of 2 years and usually causes exanthema subitum (roseola infantum; sixth disease), infantile fever without rash, febrile seizures, and occasionally encephalitis.109,110 Liver involvement with HHV-6 infection has been investigated, but attempts to prove an etiologic association have been inconclusive. Elevated aminotransferase levels were not appreciated as a common feature of roseola in a large case series.111 PCR techniques and in situ hybridization led to the isolation of HHV-6 from the liver tissue of infants with chronic hepatitis, suggesting HHV-6 as a causative agent.112,113 Reactivation of infection may occur after solid organ transplantation, with questionable clinical significance.114 Foscarnet has a better in vitro virus sensitivity than aciclovir and ganciclovir against HHV-6.115 A recent study116 reported the involvement of HHV-6 in 15 patients with non-A, non-E hepatitis who underwent liver transplantation for acute liver failure. HHV-6-specific antigens were analyzed in the explanted livers by immunohistochemistry, and the possible presence of the virus in peripheral blood mononuclear cells was demonstrated by the HHV-6 antigenemia test. The involvement of hepatitis viruses and other viral agents, such as CMV and HHV7, was excluded. Of the 15 patients with acute liver failure of unknown cause, 12 (80%) demonstrated HHV-6 antigens in the liver. Most of these patients (10/12) also demonstrated HHV-6 antigenemia. No other viruses were found in the livers of the patients with acute liver failure (ALF). These observations led the authors to speculate that HHV-6 may be a cause of ALF. Although HHV-6 has been reported to cause acute hepatitis and fatal fulminant hepatic failure (FHF),116–119 and demonstrated to be in the blood or liver samples of patients, these reports did not necessarily establish causality.
HUMAN HERPES VIRUSES-7 AND -8 (HHV-7, AND HHV-8) HHV-7 also infects all humans by the age of 5 years, causing febrile syndromes. Hepatitis in association with HHV-7 has been infrequently reported.120 HHV-8 (also called Kaposi’s sarcoma-associated human herpes virus-8) has been detected consistently in Kaposi’s sarcoma,
728
lymphoma, and multicentric Castleman’s disease, in HIV-positive patients, and occasionally in HIV-negative patients. Liver involvement may occur in the visceral type of Kaposi’s sarcoma.
HUMAN PARVOVIRUS B19 (HPV-B19) Human parvovirus B19 is a small DNA virus. It was discovered incidentally in 1974 when parvovirus-like particles were noted in serum specimens from asymptomatic blood donors being tested for hepatitis B surface antigen. Sample number 19 in panel B (hence B19) gave an anomalous ‘false positive’ result that was later recognized as a member of the Parvoviridae family.121,122 Human parvovirus B19 infection produces a spectrum of clinical manifestations including: erythema infectiosum – ‘fifth disease’ – in children; hydrops fetalis and fetal death; an arthritis syndrome associated with acute infections in adults; hematological disorders such as leukopenia, thrombocytopenia, transient aplastic crisis in patients with chronic hemolytic anemia, and chronic anemia in immunocompromised patients including AIDS; other rare organ involvement, including neurologic, cardiac, liver, and vasculitis.122 Hepatic manifestations range from a transient elevation of serum aminotransferases123,124 sometimes seen during the course of erythema infectiosum125,126 to fulminant hepatic failure (FHF).127 HPVB19 DNA has been found in liver samples from 67% of patients with non-A, non-E FHF and aplastic anemia, and in 50% of patients with cryptogenic FHF without aplastic anemia, compared to 15% of control subjects with chronic liver failure. This led some investigators to suggest that HPV-B19 is a possible causative agent of fulminant liver failure.128,129 Definitive diagnosis of acute HPV-B19 infection relies on the detection of HPV-B19 IgM or viral DNA. PCR is much more sensitive for detecting viral DNA in the serum, other body fluids, and fresh and paraffin-embedded tissue. In most cases HPV-B19 infection is benign and self-limited and requires no treatment other than symptomatic relief.122 At the Mayo Clinic we reported two cases of a less severe form of hepatitisassociated aplastic anemia.130
ADENOVIRUSES There are close to 50 serotypes of adenovirus causing acute infections of the respiratory system, conjunctivae, and gastrointestinal tract, and occasionally hemorrhagic cystitis, infantile diarrhea, intussusception, and central nervous system infections.131 Disseminated disease with multiorgan involvement has been reported in immunocompromised patients and associated with an increased mortality.132 The role of adenovirus as an etiologic agent of hepatic damage has been controversial. Fatal cases of adenovirus infection with fulminant hepatitis were reported in these immunosuppressed adult patients. Postmortem liver pathology has revealed widespread hepatic necrosis with intranuclear inclusions within viable hepatocytes. Electron microscopy may show crystalline arrays of virions within hepatocytes.133,134 No specific therapy for adenovirus hepatitis is currently available.
Chapter 35 OTHER HEPATITIS VIRUSES
OCCULT HEPATITIS B VIRUS INFECTION Occult or cryptic hepatitis B virus (HBV) infection is defined as the detection of HBV DNA in the serum or liver tissue of patients who are negative for hepatitis B surface antigen (HBsAg).135 It has been suggested that occult HBV infection maintains pro-oncogenic properties leading to hepatocellular carcinoma (HCC) in HBsAgseronegative patients.135,136 Occult HBV infection was first reported in the mid-1980s when hybridization techniques for the detection of HBV DNA became available. These studies showed that HBV DNA could be detected in HBsAg-negative patients with hepatocellular carcinoma (HCC).137 Recent studies using more sensitive techniques have confirmed these observations.136 Usually, patients with occult HBV infection have low HBV DNA levels: 102–3 copies/ml in the serum and 0.01–0.1 copy per liver cell.138 Therefore, detection of occult HBV infection most often requires ultrasensitive polymerase chain reaction (PCR) assays. Pollicino et al.136 investigated the prevalence and molecular status of occult HBV infection among Italian patients with HCC. They tested tumor tissues from 107 patients with HCC and the corresponding non-tumor liver tissue from 72 of these patients for HBV DNA, utilizing ultrasensitive PCR assays. All cases were hepatitis B surface antigen negative. Viral DNA was detected in 63.5% cases of tumorous tissue with HCC and in 72% of adjacent benign tissue, suggesting that occult HBV is a risk factor for development of HCC. These observations suggest that the same mechanism of oncognesis described for patients with overt HBV infection (HBsAg positive) are operative in patients with occult HBV infection (see Chapter 10). Future studies are needed to determine the exact role of occult HBV infection in the development of HCC.
healthcare facilities to develop operational preparedness and response plans.140 Lymphocytopenia, thrombocytopenia, and elevated levels of Ddimers and activated partial thromboplastin time are common laboratory findings in SARS. The levels of alanine aminotransferase, creatine kinase, and lactate dehydrogenase may be increased. However, these laboratory findings do not allow reliable discrimination between SARS and other causes of community-acquired pneumonia.142 One-third of patients with SARS improve and the other two-thirds develop persistent fever, worsening pulmonary symptoms, and radiographic findings. Some patients develop multiorgan failure and die. Age and coexisting illness, especially diabetes mellitus and heart disease, are consistently found to be independent prognostic factors for the need for intensive care and the risk of death. Liver involvement in SARS is common and has been reported in up to 60% of patients.142–144 The majority of these have been treated with antibiotics, antiviral medications, and steroids, which are potentially hepatotoxic. Hence, whether or not SARS-CoV infection can lead to liver damage per se remains unknown. The most common abnormality is elevated aminotransferases, or the less common ischemic injury in cases of multiorgan failure. However, Chua et al.145 reported the clinical course and liver pathology in three SARS patients with liver impairment. All three had moderate to marked elevation of their liver aminotransferases and common causes of hepatitis were excluded by serologic tests. Histologic examination of the liver specimens revealed prominent mitoses, acidophilic bodies, Kupffer cells, ballooning of hepatocytes, and mild to moderate lobular inflammation as the common histologic features. All of the patients showed positive RT-PCR for SARS-CoV in liver tissue but not in the sera, suggesting that the virus was localized in liver. The investigators concluded that SARS-CoV may infect the liver, leading to mild to moderate lobular inflammation and apoptosis.
SEVERE ACUTE RESPIRATORY SYNDROME (SARS)
ADDITIONAL HEPATITIS AGENTS
SARS is a newly recognized, severe febrile respiratory illness caused by a previously unknown coronavirus, SARS-associated coronavirus (SARS-CoV). It was responsible for the first epidemic of the 21st century, emerging in the southern Chinese province of Guangdong in November 2002.139 The worldwide epidemic was triggered in late February 2003 when an ill physician from Guangdong infected several other guests at a hotel in Hong Kong.140,141 These people subsequently became the index patients for large outbreaks of SARS in many areas of the world. On 12 March 2003, the World Health Organization (WHO) issued a historic global alert for SARS, a deadly new infectious disease with the potential for rapid spread from person to person and via international air travel. WHO and its partners, including the Centers for Disease Control and Prevention (CDC), promptly initiated a rapid, intense and coordinated investigative and control effort that led within 2 weeks to the identification of the etiologic agent, SARS-CoV, and to a series of decisive and effective containment efforts. By the time SARS-CoV transmission was brought to an end in July 2003, more than 8000 cases and 780 deaths had been reported to WHO. The CDC published guidelines in order for localities and
Additional hepatitis agents have been suggested from transfusionassociated hepatitis studies, CDC Sentinel Counties studies, and cases of fulminant hepatitis in whom no agent have been identified in the majority of them. In all these conditions, a viral agent is suspected to exist but no specific virus has been identified. The GB agent and the hepatitis G virus (HGV) are RNA viruses that belong to the Flaviviridae family.146 Extensive investigations have failed to show that these agents play any etiologic role in acute or chronic liver disease.147,148 The TT virus (TTV) has been shown to be a small, non-enveloped single-stranded circular DNA virus in the family of Circoviridae.149,150 It is now clear that TTV is a heterogeneous agent that can be transmitted to humans by both parenteral and non-parenteral routes. The agent is of particularly high prevalence in Japan, where TTV has been detected in healthy persons. Although initially implicated in fulminant hepatitis and cryptogenic chronic liver disease, these associations have not been confirmed and there are currently no proven hepatic diseases associated with this agent.151,152 The newly described viruses in this family have been designated SANBAN and YONBAN.153–156 These agents have similar properties
729
Section V. Liver Diseases Due to Infectious Agents
to TTV but sufficient nucleotide differences to make them distinct members of the Circoviridae family. SEN virus (SEN-V) was discovered independently using amplification strategies with highly degenerate TTV primers.153 Two SENV variants (SEN-D and SEN-C/H) have been studied and have been found as acute infections in 11 of 12 (93%) transfusion-transmitted non-A, non-E hepatitis cases. There is no current evidence that SEN-V is truly a hepatitis virus, and further work is needed.
REFERENCES 1. Kim W, Badley A, Wiesner R, et al. The economic impact of cytomegalovirus infection after liver transplantation. Transplantation 2000;69:357–361. 2. Paya C, Hermans P, Washington JI, et al. Incidence, distribution, and outcome of episodes of infection in 100 orthotopic liver transplantations. Mayo Clin Proc 1989;64:555–564. 3. Goodgame R. Gastrointestinal cytomegalovirus disease. Ann Intern Med 1993;119:924–935. 4. Crumpacker C. Cytomegalovirus. Edinburgh: Churchill Livingstone, 2000. 5. Carey WD, Patel G. Viral hepatitis in the 1990s, Part III: Hepatitis C, hepatitis E, and other viruses. Cleveland Clin J Med 1992;59:595–601. 6. Stagno S, Pass R, Dworsky M, et al. Congenital cytomegalovirus infection: The relative importance of primary and recurrent maternal infection. N Engl J Med 1982;306:945–949. 7. Stagno S, Whitley R. Herpesvirus infections of pregnancy. Part I: Cytomegalovirus and Epstein–Barr virus infections. N Engl J Med 1985;313:1270–1274. 8. Fowler K, Stagno S, Pass R, et al. The outcome of congenital cytomegalovirus infection in relation to maternal antibody status. N Engl J Med 1992;326:663–667. 9. Griffiths P. Cytomegalovirus and the liver. Semin Liver Dis 1984;4:307–313. 10. Levy I, Shohat M, Levy Y, et al. Recurrent ascites in an infant with perinatally acquired cytomegalovirus infection [see comments]. Eur J Pediatr 1989;148:531–532. 11. Chang MH, Huang HH, Huang ES, et al. Polymerase chain reaction to detect human cytomegalovirus in livers of infants with neonatal hepatitis [see comments]. Gastroenterology 1992;103:1022–1025. 12. Watanabe S, Arima K, Nishioka M, et al. Comparison between sporadic cytomegalovirus hepatitis and Epstein–Barr virus hepatitis in previously healthy adults. Liver 1997;17:63–69. 13. Klemola E, Von Essen R, Henle G, et al. Infectiousmononucleosis-like disease with negative heterophil agglutination test. Clinical features in relation to Epstein–Barr virus and cytomegalovirus antibodies. J Infect Dis 1970;121:608–614. 14. Lang D, Scolnick E, Willerson J. Association of cytomegalovirus infection with the postperfusion syndrome. N Engl J Med 1968;278:1147–1149. 15. Horwitz C, Burke M, Grimes P, et al. Hepatic function in mononucleosis induced by Epstein–Barr virus and cytomegalovirus. Clin Chem 1980;26:243–246. 16. Kunno A, Abe M, Yamada M, et al. Clinical and histological features of cytomegalovirus hepatitis in previously healthy adults. Liver 1997;17:129–132. 17. Carter A. Cytomegalovirus disease presenting as hepatitis. Br Med J 1968;3:986. 18. Henson D. Cytomegalovirus hepatitis in an adult. An autopsy report. Arch Pathol Lab Med 1969;88:199–203. 19. Bonkowsky H, Lee R, Klatskin G. Acute granulomatous hepatitis. Occurrence in cytomegalovirus mononucleosis. JAMA 1975;233:1284–1288.
730
20. Reller LB. Granulomatous hepatitis associated with acute cytomegalovirus infection. Lancet 1973;1:20–22. 21. Toghill PJ, Bailey ME, Williams R, et al. Cytomegalovirus hepatitis in the adult. Lancet 1967;1:1351–1354. 22. Mosley JW, Redeker AG, Feinstone SM, et al. Mutliple hepatitis viruses in multiple attacks of acute viral hepatitis. N Engl J Med 1977;296:75–78. 23. Stern H. Cytomegalovirus and EB virus infections of the liver. Br Med Bull 1972;28:180–185. 24. Shusterman N, Frauenhoffer C, Kinsey M. Fatal massive hepatic necrosis in cytomegalovirus mononucleosis. Ann Intern Med 1978;88:810–812. 25. Teixidor H, Honig C, Norsoph E, et al. Cytomegalovirus infection of the alimentary canal: radiologic findings with pathologic correlation. Radiology 1987;163:317–323. 26. Roulot D, Valla D, Brun-Vezinet F, et al. Cholangitis in the acquired immunodeficiency syndrome: report of two cases and review of the literature. Gut 1987;28:1653–1660. 27. Blumberg R, Kelsey P, Perrone T, et al. Cytomegalovirus- and Cryptosporidium-associated acalculous gangrenous cholecystitis. Am J Med 1984;76:1118–1123. 28. Jacobson M, Cello J, Sande M. Cholestasis and disseminated cytomegalovirus disease in patients with the acquired immunodeficiency syndrome. Am J Med 1988;84:218–224. 29. Eeckhout E, Buydens P, Devis G, et al. The importance of a specific IgM antibody assay in the early detection of cytomegalovirus hepatitis [letter]. Am J Gastroenterol 1989;84:79–80. 30. Shuster E, Beneke J, Tegtmeier G, et al. Monoclonal antibody for rapid laboratory detection of cytomegalovirus infections: characterization and diagnostic application. Mayo Clin Proc 1985;60:577–585. 31. Martin WI, Smith T. Rapid detection of cytomegalovirus in bronchoalveolar lavage specimens by a monoclonal antibody method. J Clin Microbiol 1986;23:1006–1008. 32. Mendez J, Espy M, Smith T, et al. Evaluation of PCR primers for early diagnosis of cytomegalovirus infection following liver transplantation. J Clin Microbiol 1999;36:526–530. 33. Persing DH, Rakela J. Polymerase chain reaction for the detection of hepatitis viruses: panacea or purgatory? [editorial; comment]. Gastroenterology 1992;103:1098–1099. 34. Espy M, Paya C, Holley K, et al. Diagnosis of cytomegalovirus hepatitis by histopathology and in situ hybridization in liver transplantation. Diagn Microbiol Infect Dis 1991;14:293– 296. 35. Snover DC, Horwitz CA. Liver disease in cytomegalovirus mononucleosis: a light microscopical and immunoperoxidase study of six cases. Hepatology 1984;4:408–412. 36. Plotkin S, Starr S, Bryan C. In vitro and in vivo responses of cytomegalovirus to acyclovir. Am J Med 1982;73:257–261. 37. Balfour HJ. Antiviral drugs. N Engl J Med 1999;340: 1255–1268. 38. Sixbey J, Nedrud J, Raab-Traub N, et al. Replication of Epstein–Barr virus in human epithelial cells infected in vitro. Nature 1983;306:480–483. 39. White NJ, Juel-Jensen BE. Infectious mononucleosis hepatitis. Semins Liver Dis 1984;4:301–306. 40. Yao Q, Rickinson A, Epstein M. A re-examination of the Epstein–Barr virus carrier state in healthy seropositive individuals. Int J Cancer 1985;35. 41. Cohen J. Epstein–Barr virus infections. N Engl J Med 2000;343:481–492. 42. Hoagland R. The clinical manifestations of infectious mononucleosis. A report of two hundred cases. Am J Med Sci 1960;240:55–63. 43. Baron D, Bell J, Demmett W. Biochemical studies on hepatic involvement in infectious mononucleosis. J Clin Pathol 1960;1965:209–211.
Chapter 35 OTHER HEPATITIS VIRUSES
44. Rosalki S, Jones T, Verney A. Transaminases and liver function studies in infectious mononucleosis. Br Med J 1960;1:929–932. 45. Reichman S, Burke A, Davis WJ. Hepatic involvement of infectious mononucleosis. Am J Dig Dis 1957;2:430–436. 46. Nelson R, Darragh J. Infectious mononucleosis hepatitis. A clinicopathologic study. Am J Med 1956;21:26–33. 47. Madigan N, Newcomer A, Taswell H. Intense jaundice in infectious mononucleosis. Mayo Clin Proc 1973;48:857–862. 48. Edoute Y, Baruch Y, Lachter J, et al. Severe cholestatic jaundice induced by Epstein–Barr virus infection in the elderly. J Gastroenterol Hepatol 1998;13:821–824. 49. Fuhrman S, Gill R, Horwitz C, et al. Marked hyperbilirubinemia in infectious mononucleosis. Analysis of laboratory data in seven patients. Arch Intern Med 1987;147:850–853. 50. Horwitz C, Henle W, Henle G, et al. Infectious mononucleosis in patients aged 40 to 72 years: report of 27 cases, including 3 without heterophil-antibody responses. Medicine 1983;62:256–262. 51. Jacobson IM, Gang DL, Schapiro RH. Epstein–Barr viral hepatitis: an unusual case and review of the literature. Am J Gastroenterol 1984;79:628–632. 52. Biest S, Schubert TT. Chronic Epstein–Barr virus infection: a cause of granulomatous hepatitis? J Clin Gastroenterol 1989;11:343–346. 53. Harrington P, Gutierrez J, Ramirez-Ronda C, et al. Granulomatous hepatitis. Rev Infect Dis 1982;4:638–655. 54. Markin R, Linder J, Zuerlein K, et al. Hepatitis in fatal infectious mononucleosis. Gastroenterology 1987;93: 1210–1217. 55. Donhuijsen-Ant R, Abken H, Westerhausen M, et al. Aggressive hepatitis in a patient with acute myeloid leukaemia during complete remission and detection of Epstein–Barr virus DNA in a liver biopsy. Br J Haematol 1990;76:557–558. 56. Papatheodoridis GV, Delladetsima JK, Kavallierou L, et al. Fulminant hepatitis due to Epstein–Barr virus infection. J Hepatol 1995;23:348–350. 57. Davies M, Morgan-Capner P, Portmann B, et al. A fatal case of Epstein–Barr virus infection with jaundice and renal failure. Postgrad Med J 1980;56:794–795. 58. Pelletier L, Borel D, Romig D, et al. Disseminated intravascular coagulation and hepatic necrosis. Complications of infectious mononucleosis. JAMA 1976;235:1144–1146. 59. Halmos B, Anastopoulos H, Schnipper L, et al. Extreme lymphoplasmacytosis and hepatic failure associated with sulfasalazine hypersensitivity reaction and a concurrent EBV infection – case report and review of the literature. Ann Hematol 2004;83:242–246. 60. Schooly A. Epstein–Barr virus. Edinburgh: Churchill Livingstone, 2000. 61. Edwards JM, Vandervelde EM, Cohen BJ, et al. Laboratory diagnosis of EB virus infection in some cases presenting as hepatitis. J Clin Pathol 1978;31:179–182. 62. Lloyd-Still JD, Scott JP, Crussi F. The spectrum of Epstein–Barr virus hepatitis in children. Pediatr Pathol 1986;5:337–351. 63. Kilpatrick Z. Structural and functional abnormalities of liver in infectious mononucleosis. Arch Intern Med 1966;117:47–53. 64. Sainsbury R, Smith PK, LeQuesne G, et al. Gallbladder wall thickening with infectious mononucleosis hepatitis in an immunosuppressed adolescent. J Pediatr Gastroenterol Nutr 1994;19:123–125. 65. Gan YJ, Sullivan JL, Sixbey JW. Detection of cell-free Epstein–Barr virus DNA in serum during acute infectious mononucleosis. J Infect Dis 1994;170:436–439. 66. Markin R. Manifestations of Epstein–Barr virus-associated disorders in liver. Liver 1994;14:1–13. 67. van der Horst C, Joncas J, Ahronheim G, et al. Lack of effect of peroral acyclovir for the treatment of acute infectious mononucleosis. J Infect Dis 1991;164:788–792.
68. Feranchak AP, Tyson RW, Narkewicz MR, et al. Fulminant Epstein–Barr viral hepatitis: orthotopic liver transplantation and review of the literature. Liver Transpl Surg 1998;4: 469–476. 69. Hass G. Hepato-adrenal necrosis with intranuclear inclusion bodies. Am J Pathol 1935;11:127–142. 70. Hanshaw J. Herpesvirus hominis infections in the fetus and the newborn. Am J Dis Child 1973;126:456–555. 71. Benador N, Mannhardt W, Schranz D, et al. Three cases of neonatal herpes simplex virus infection presenting as fulminant hepatitis. Eur J Pediatr 1990;149:555–559. 72. Flewett T, Parker R, Phillip W. Acute hepatitis due to herpes simplex virus in an adult. J Clin Pathol 1969;22:60–66. 73. Mudido P, Marshall GS, Howell RS, et al. Disseminated herpes simplex virus infection during pregnancy. A case report. J Reprod Med 1993;38:964–968. 74. Fink CG, Read SJ, Hopkin J, et al. Acute herpes hepatitis in pregnancy [see comments]. J Clin Pathol 1993;46:968–971. 75. Klein NA, Mabie WC, Shaver DC, et al. Herpes simplex virus hepatitis in pregnancy. Two patients successfully treated with acyclovir. Gastroenterology 1991;100:239–244. 76. Wertheim RA, Brooks BJ Jr, Rodriguez FH Jr, et al. Fatal herpetic hepatitis in pregnancy. Obstet Gynecol 1983;62:38s–42s. 77. Fairley I, Wilson J. Herpes hepatitis in pregnancy [letter; comment]. J Clin Pathol 1994;47:478. Published erratum appears in J Clin Pathol 1994;47:964. 78. Kang AH, Graves CR. Herpes simplex hepatitis in pregnancy: a case report and review of the literature. Obstet Gynecol Surv 1999;54:463–468. 79. Yaziji H, Hill T, Pitman TC, et al. Gestational herpes simplex virus hepatitis. South Med J 1997;90:347–351. 80. Minuk G, Nicolle L. Genital herpes and hepatitis in healthy young adults. J Med Virol 1986;19:269–275. 81. Chase RA, Pottage JC Jr, Haber MH, et al. Herpes simplex viral hepatitis in adults: two case reports and review of the literature. Rev Infect Dis 1987;9:329–333. 82. Velasco M, Llamas E, Guijarro-Rojas M, et al. Fulminant herpes hepatitis in a healthy adult: a treatable disorder? J Clin Gastroenterol 1999;28:386–389. 83. Farr RW, Short S, Weissman D. Fulminant hepatitis during herpes simplex virus infection in apparently immunocompetent adults: report of two cases and review of the literature. Clin Infect Dis 1997;24:1191–1194. 84. Pellise M, Miquel R. Liver failure due to herpes simplex virus. J Hepatol 2000;32:170. 85. Corey L. Herpes simplex virus. Edinburgh: Churchill Livingstone, 2000. 86. Hofer S, Hunziker S, Ludwig C. Fatal herpes simplex virus hepatitis complicating chemotherapy with weekly docetaxel. Ann Oncol 2003;14:340. 87. Peters D, Greene W, Ruggiero F, et al. Herpes simplex-induced fulminant hepatitis in adults: a call for empiric therapy. Dig Dis Sci 2000;45:2399–2404. 88. Rakela J, Lange SM, Ludwig J, et al. Fulminant hepatitis: Mayo Clinic experience with 34 cases. Mayo Clin Proc 1985;60:289–292. 89. Gabel H, Flamholc L, Ahlfors K. Herpes simplex virus hepatitis in a renal transplant recipient: successful treatment with acyclovir. Scand J Infect Dis 1988;20:435–438. 90. Glorioso DV, Molloy PJ, Van Thiel DH, et al. Successful empiric treatment of HSV hepatitis in pregnancy. Case report and review of the literature. Dig Dis Sci 1996;41:1273–1275. 91. Kaufman B, Gandhi SA, Louie E, et al. Herpes simplex virus hepatitis: case report and review [see comments]. Clinic Infect Dis 1997;24:334–338. 92. Shanley CJ, Braun DK, Brown K, et al. Fulminant hepatic failure secondary to herpes simplex virus hepatitis. Successful
731
Section V. Liver Diseases Due to Infectious Agents
93.
94. 95.
96. 97. 98. 99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109. 110.
111.
112.
113.
114.
115.
732
outcome after orthotopic liver transplantation. Transplantation 1995;59:145–149. Pinna AD, Rakela J, Demetris AJ, et al. Five cases of fulminant hepatitis due to herpes simplex virus in adults. Dig Dis Sci 2002;47:750–754. Whitley R. Varicella-zoster virus. Edinburgh: Churchill Livingstone, 2000. Phuah HK, Chong CY, Lim KW, et al. Complicated varicella zoster infection in 8 paediatric patients and review of literature. Singapore Med J 1998;39:115–1s20. Plotkin SA. Clinical and pathogenetic aspects of varicella-zoster. Postgrad Med J 1985;61:7–14. Ey J, Smith S, Fulginiti V. Varicella hepatitis without neurologic symptoms or findings. Pediatrics 1981;67:285–287. Pitel P, McCormick K, Fitzgerald E, et al. Subclinical hepatic changes in varicella infection. Pediatrics 1980;65:631–633. Anderson D, Schwartz J, Hunter N, et al. Varicella hepatitis: a fatal case in a previously healthy, immunocompetent adult. Report of a case, autopsy, and review of the literature. Arch Intern Med 1993;154:2101–2106. Rubin R, Tolkoff-Rubin N. Viral infection in the renal transplant patient. Proc Eur Dial Transplant Ass 1983;19:513–526. Schiller GJ, Nimer SD, Gajewski JL, et al. Abdominal presentation of varicella-zoster infection in recipients of allogeneic bone marrow transplantation [see comments]. Bone Marrow Transpl 1991;7:489–4s91. Alonso EM, Fox AS, Franklin WA, et al. Postnecrotic cirrhosis following varicella hepatitis in a liver transplant patient. Transplantation 1990;49:650–653. Bensousan TA, Moal MC, Vincent F, et al. Fulminant hepatitis revealing primary varicella in a renal graft recipient. Transplant Proc 1995;27:2512. Morishita K, Kodo H, Asano S, et al. Fulminant varicella hepatitis following bone marrow transplantation [letter]. JAMA 1985;253:511. Patti ME, Selvaggi KJ, Kroboth FJ. Varicella hepatitis in the immunocompromised adult: a case report and review of the literature [see comments]. Am J Med 1990;88:77–80. Williams WW. CDC guidelines for the prevention and control of nosocomial infections. Guideline for infection control in hospital personnel. Am J Infect Control 1984;12:34–63. Alford CA. Acyclovir treatment of herpes simplex virus infections in immunocompromised humans. An overview. Am J Med 1982;73:225–228. Shulman ST. Acyclovir treatment of disseminated varicella in childhood malignant neoplasms. Am J Dis Child 1985;139:137–140. Straus S. Human herpesvirus types 6, 7, and 8. Edinburgh: Churchill Livingstone, 2000. Pruksananonda P, Hall C, Insel R, et al. Primary human herpesvirus 6 infection in young children. N Engl J Med 1992;326:1445–1450. Hall C, Long C, Schnabel K, et al. Human herpesvirus-6 infection in children. A prospective study of complications and reactivation. N Engl J Med 1994;33:432–438. Tajiri H, Tanaka-Taya K, Ozaki Y, et al. Chronic hepatitis in an infant, in association with human herpesvirus-6 infection. J Pediatr 1997;131:473–475. Schmitt K, Deutsch J, Tulzer G, et al. Autoimmune hepatitis and adrenal insufficiency in an infant with human herpesvirus-6 infection. Lancet 1996;348:966. Lunel F, Robert C, Munier P, et al. Hepatitis virus infections in heart transplant recipients. Biomed Pharmacother 1995;49:125–129. Reymen D, Naesens L, Balzarini J, et al. Antiviral activity of selected acyclic nucleoside analogues against human herpesvirus 6. Antiviral Res 1995;28:343–357.
116. Harma M, Hockerstedt K, Lautenschlager I. Human herpesvirus-6 and acute liver failure. Transplantation 2003;76:536–539. 117. Asano Y, Yoshikawa T, Suga S, et al. Fatal fulminant hepatitis in an infant with human herpesvirus-6 infection [letter]. Lancet 1990;335:862–863. 118. Sobue R, Miyazaki H, Okamoto M, et al. Fulminant hepatitis in primary human herpesvirus-6 infection [letter]. N Engl J Med 1991;324:1290. 119. Tajiri H, Nose O, Baba K, et al. Human herpesvirus-6 infection with liver injury in neonatal hepatitis [letter]. Lancet 1990;335:863. 120. Hashida T, Komura E, Yoshida M, et al. Hepatitis in association with human herpesvirus-7 infection. Pediatrics 1995;96:783–785. 121. Cossart Y, Feild A, Cant B, et al. Parvovirus-like particles in human sera. Lancet 1975; 1:72–73. 122. Brown K. Parvoviruses. Edinburgh: Churchill Livingstone, 2000. 123. Tsuda H. Liver dysfunction caused by parvovirus B19 [letter]. Am J Gastroenterol 1993;88:1463. 124. Yoto Y, Kudoh T, Asanuma H, et al. Transient disturbance of consciousness and hepatic dysfunction associated with human parvovirus B19 infection. Lancet 1994;344:624–625. 125. Yoto Y, Kudoh T, Haseyama K, et al. Human parvovirus B19 infection associated with acute hepatitis. Lancet 1996;347:868–869. 126. Drago F, Semino M, Rampini P, et al. Parvovirus B19 infection associated with acute hepatitis and a purpuric exanthem. Br J Dermatol 1999;141:160–161. 127. Karetnyi Y, Beck P, Markin R, et al. Human parvovirus B19 infection in acute fulminant liver failure. Arch Virol 1999;144:1713–1724. 128. Langnas A, Markin R, Cattral M, et al. Parvovirus B19 as a possible causative agent of fulminant liver failure and associated aplastic anemia. Hepatology 1995;22:1661–1665. 129. Sokal E, Melchior M, Cornu C, et al. Acute parvovirus B19 infection associated with fulminant hepatitis of favourable prognosis in young children. Lancet 1998;352:1739–1741. 130. Pardi D, Romero Y, Mertz L, et al. Hepatitis-associated aplastic anemia and acute parvovirus B19 infection: A report of two cases and a review of the literature. Am J Gastroenterol 1998;93:468–470. 131. Baum S. Adenovirus. Edinburgh: Churchill Livingstone, 2000. 132. Zahradnik J, Spencer M, Porter D. Adenovirus infection in the immunocompromised patient. Am J Med 1980;68:725–732. 133. Carmichael GJ, Zahradnik J, Moyer G, et al. Adenovirus hepatitis in an immunosuppressed adult patient. Am J Clin Pathol 1979;71:352–355. 134. Wigger H, Blanc W. Fatal hepatic and bronchial necrosis in adenovirus infection with thymic alymphoplasia. N Engl J Med 1966;275:870–874. 135. Allain J. Occult hepatitis B virus infection. Transfusion Clin Biol 2004;11:18–25. 136. Pollicino T, Squadrito G, Cerenzia G, et al. Hepatitis B virus maintains its pro-oncogenic properties in the case of occult HBV infection. Gastroenterology 2004;126:102–110. 137. Brechot C, Degos F, Lugassy C, et al. Hepatitis B virus DNA in patients with chronic liver disease and negative tests for hepatitis B surface antigen. N Engl J Med 1985;312:270–276. 138. Paterlini P, Poussin K, Kew M, et al. Selective accumulation of the X transcript of hepatitis B virus in patients negative for hepatitis B surface antigen with hepatocellular carcinoma. Hepatology 1995;21:313–331. 139. Navas-Martin S, Weiss S. SARS: lessons learned from other coronaviruses. Viral Immunol 2003;16:461–474. 140. CDC. Public health guidance for community-level preparedness and response to severe acute respiratory syndrome (SARS).
Chapter 35 OTHER HEPATITIS VIRUSES
141.
142. 143.
144.
145.
146.
147.
148.
149.
Volume 2004. Version 2. Department of Health and Human Services, Centers for Disease Control and Prevention, 2004. Lipsitch M, Cohen T, Cooper B, et al. Transmission dynamics and control of severe acute respiratory syndrome. Science 2003;300:1966–1970. Peiris J, Yuen K, Osterhaus A, et al. The severe acute respiratory syndrome. N Engl J Med 2004;349:2431–2441. Lee N, Hui D, Wu A, et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003:1986–1994. Tsang K, Ho P, Ooi G, et al. A cluster of cases of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003;348:1977–1985. Chau T, Lee K, Yao H, et al. SARS-associated viral hepatitis caused by a novel coronavirus: report of three cases. Hepatology 2004;39:302–310. Linnemann CJ, Shea L, Partin J, et al. Reye’s syndrome: epidemiologic and viral studies, 1963–1974. Am J Epidemiol 1975;101:517–526. Laskus T, Wang LF, Radkowski M, et al. Hepatitis G virus infection in American patients with cryptogenic cirrhosis: no evidence for liver replication. J Infect Dis 1997;176:1491–1495. Laskus T, Radkowski M, Wang LF, et al. Lack of evidence for hepatitis G virus replication in the livers of patients coinfected with hepatitis C and G viruses. J Virol 1997;71:7804–7806. Nishizawa T, Okamoto H, Konishi K, et al. A novel DNA virus (TTV) associated with elevated transaminase levels in
150.
151.
152.
153.
154.
155.
156.
posttransfusion hepatitis of unknown etiology. Biochem Biophys Res Commun 1997;241:92–97. Mushahwar I, Erker JC, Muerhoff AS, et al. Molecular and biophysical characterization of TT virus: evidence for a new virus family infecting humans. Proc Natl Acad Sci USA 1999;96: 3177–3182. Naoumov N, Petrova EP, Thomas MG, et al. Presence of a newly described human DNA virus (TTV) in patients with liver disease. Lancet 1998:195–197. Matsumoto A, Yeo AE, Shih JW, et al. Transfusion-associated TT virus infection and its relationship to liver disease. Hepatology 1999;30:283–288. Tanaka Y, Primi D, Wang RY, et al. Genomic and molecular evolutionary analysis of a newly identified infectious agent (SEN virus) and its relationship to the TT virus family. J Infect Dis 2001;183:359–367. Takahashi K, Hijikata M, Samokhvalor EI, et al. Full or near full length nucleotide sequences of TT virus variants (types SANBAN and YONBAN) and the TT virus-like mini virus. Intervirology 2000;43:119–123. Biagini P, Attoui H, Gallian P, et al. Complete sequences of two highly divergent European isolates of TT virus. Biochem Biophys Res Commun 2000;271:837–841. Hijikata M, Takahashi K, Mishiro S. Complete circular DNA genome of a TT virus variant (isolate name SANBAN) and 44 partial ORF2 sequences implicating a great degree of diversity beyond genotypes. Virology 1999;260:17–22.
733
Section V: Liver Diseases Due to Infectious Agents
36
PARASITIC AND HELMINTHIC DISEASE OF THE LIVER John P. Cello and Ruedi F. Thoeni Abbreviations CIET counterimmunoelectrophoresis test CT computed tomography ELISA enzyme-linked immunosorbent assay ERCP endoscopic retrograde pancreatography HBV hepatitis B virus
HCV IEP IFT IgG
hepatitis C virus immunoelectrophoresis indirect fluorescent test immunoglobulin G
INTRODUCTION The primordial burden of humanity is intestinal parasitism. Indeed, prior to the development of modern food preservation together with water and sewage sanitation, among the most common causes of mortality for humans were these infectious diseases. In the industrialized and sanitized environment in which most of us live, intestinal parasites are relatively infrequently encountered. Given, however, the incredible mobility of humans around the world, intestinal parasites and worms are not infrequently acquired by individuals traveling to endemic areas. In addition, clinicians all over the world not infrequently encounter recent residents from areas of the globe where intestinal parasites are common. It is therefore important for clinicians to recognize the clinical manifestations of these illnesses, the diagnostic tests of choice, and the most successful treatments available.
CLONORCHIS SINENSIS Clonorchis sinensis is a biliary fluke found in a high percentage of individuals living along the great rivers and streams in China, North Vietnam, and Korea.1–8 The infection is disproportionately found in individuals employed in the fishing industry, including fishermen, fishmongers, fish processors, and, in general, individuals eating raw or undercooked freshwater fish.1 These fish are contaminated with C. sinensis metacercaria. The intermediary hosts for C. sinensis infection are a wide array of mollusca found in freshwater lakes, rivers, ponds, and streams in South-East Asia. The infection begins with the ingestion of contaminated fish and the exocytosing of larvae in the gastrointestinal tract. The metacercaria migrate via the portal venous system to the lungs, and then pass through the alveolar capillary membrane into the small airways. The immature worms then migrate to the biliary tree where they take up residence. It is important for clinicians to remember that Clonorchis infection of the biliary tree may last up to three decades following the original exposure.
IHA MRCP RPC
indirect hemagglutination assay magnetic resonance cholangiopancreatography recurrent pyogenic cholangitis
CLINICAL PRESENTATION Many, if not most, people with Clonorchis infection in the biliary tree are asymptomatic.1 Rarely, Löffler syndrome may occur in patients; this is manifested by transient pulmonary infiltrates and peripheral eosinophilia. The most common clinical symptoms experienced by patients with infection include epigastric and right upper quadrant discomfort, generalized malaise, inertia, anorexia, dyspepsia, nausea, and intermittent fevers with dizziness and headaches. Later in the course of the disease patients may present with typical bouts of cholangitis.
DIAGNOSIS Diagnostic stool-testing is the most common means of making the diagnosis of C. sinensis infestation. This is usually done by a modified Kato–Katz method.1 This appears to be more sensitive than formalin-ether or direct stick smear stain. In selected areas of China and Korea, a circulating antigen for C. sinensis can be detected by using a dot-enzyme-linked immunosorbent assay (ELISA) serologic test. Since a substantial percentage of these patients present with biliary tract disease, including principally recurring bouts of cholangitis, the majority of symptomatic patients are studied by ultrasound and/or computed tomography (CT) scans.1–8 These recurrent bouts of cholangitis lead to bile duct dilation, strictures, obstruction, and calculi that can be visualized by ultrasound and CT. There is a predilection for ductal involvement in the left lobe of the liver but diffuse intrahepatic disease can be seen. Ultrasound shows dilated intra- and extrahepatic biliary ducts and frequently stones and debris (some of them non-shadowing) in the gallbladder and biliary ducts (Figure 36-1A).2,3 The extrahepatic bile duct can be dilated to as much as 3–4 cm in diameter (Figure 36-B). The dilation of the extrahepatic duct may be secondary to loss of elasticity of the duct wall due to chronic infection or to ampullary narrowing.4 At times, the echogenicity in the periductal areas is increased. On ultrasound, the gallbladder may be distended and its wall thickened.2,3 CT is one of the best imaging modalities in patients with suspected infection with C. sinensis and the findings on CT reflect the
735
Section V. Liver Diseases Due to Infectious Agents
A
B
Figure 36-1. Ultrasound of recurrent pyogenic cholangitis due to Clonorchis. (A) On this right-sided longitudinal scan, the right intrahepatic ducts (long white arrows) are dilated and the common hepatic duct (short black arrows) demonstrate a large filling defect (arrowheads) with posterior shadowing (long black arrows) indicative of a large stone. (B) On this longitudinal scan slightly more to the left than Figure 36-1A, the large filling defect in the common bile duct (white arrows) with posterior shadowing (black arrows) represents a cast of stones that completely fills and markedly distends the bile duct.
pathology.5 CT can demonstrate the extent of disease in the liver more clearly than ultrasound. This pattern of intrahepatic and extrahepatic ductal dilation, especially when the degree of dilatation is pronounced and associated with intraductal stones, suggests the diagnosis of recurrent pyogenic cholangitis (RPC), formerly called oriental cholangiohepatitis (Figure 36-2). RPC is usually caused by Clonorchis or Ascaris infections and its appearance is typical enough to aid in differentiating it from other entities such as ascending cholangitis, primary sclerosing cholangitis, Caroli disease, or choledochal cysts. The calculi are often isodense or slightly hyperdense to bile and typically completely fill the ducts forming ductal casts (Figure 36-3A).2,3,6 With intravenous contrast administration, there may be enhancement in the surrounding liver parenchyma, suggesting the presence of acute hepatic inflammation or pseudotumor (Figure 36-3B).4,7 Gas may be seen within the ducts due to the bacterial infection or surgical intervention. CT is the best imaging modality not only for determining the extent of the disease but also for surgical planning, and for followup. CT can also be used to look for complications such as intrahepatic abscess formation or development of cholangiocarcinoma. The risk for cholangiocarcinoma is increased in patients with RPC and has been reported in up to 5% of patients.4,6 More recently, magnetic resonance cholangiopancreatography (MRCP) has been shown to be superior to direct cholangiography for accurate topographic evaluation of RCP because it is able to depict the entire biliary tree, regardless of obstruction or stenosis.8 However, since MRCP is only a diagnostic tool, endoscopic retrograde pancreatography (ERCP) is still needed in many instances (Figures 36-4 and 36-5). Recurrent bouts of cholangitis due to
736
Clonorchis may be difficult to differentiate from routine stone disease since the majority of patients with Clonorchis do not have large calculi within the gallbladder or the extrahepatic ductal system that would otherwise suggest stone disease. On occasion, clinicians will perform an ERCP and sphincterotomy in patients with Clonorchis and may only make the diagnosis of Clonorchis with the visualization of 2–3 cm long dark black worms in the ductal system. Characteristically the bile of these patients is jet-black in color. Further examination of the bile reveals the dark-black ova sediment on the bottom of a collecting flask or tube. C. sinensis infection of the biliary tree is associated with an increased risk of cholangiocarcinoma. These patients will likely present with the typical presentation of ductal cancer with progressive jaundice, and marked elevation of the serum alkaline phosphatase levels.
TREATMENT The most important medical therapy for Clonorchis infection is praziquantel. Single-dose therapy with 40 mg per kg body weight has only a 25% cure rate.1 Two doses of 30 mg/kg body weight given on a single day has a cure rate of approximately 60%; 25 mg/kg body weight three times a day for 1 day has a cure rate of over 83%. For patients with heavy infestation, a repeat course of 25 mg per/kg b.i.d. for 1 day will achieve a cure rate approaching 100%. There are no clinical studies of ERCP sphincterotomy and/or choledochoenteric anastomosis in patients with persistent C. sinensis infection. It is likely that a simple cholecystectomy will not change the natural history of infection within the intrahepatic and extrahepatic ductal system of patients infected with C. sinensis.
Chapter 36 PARASITIC AND HELMINTHIC DISEASE OF THE LIVER
A
C
HEPATOBILIARY ASCARIASIS IMPORTANCE AND PREVALENCE Over 1 billion individuals, representing more than 25% of the world’s population, are infected with the large roundworm, Ascaris lumbricoides.9–23 Infection is found in individuals living mostly in the tropical and subtropical areas of the world. Risk factors that predispose to the spread of Ascaris include poor sanitation, unplanned development, increasing urbanization of the population, and the use of human excreta as fertilizer. In some regions of the globe nearly 90% of the children and 60% of the adults are infected with A. lumbricoides.11 This is particularly common among the young and female populations. A. lumbricoides is transmitted via the fecal–oral route. When contaminated food or water is ingested, the eggs hatch in the duodenum, releasing the larvae. The immature larvae penetrate the small-bowel mucosa, and travel to the portal venous system
B
Figure 36-2. Computed tomography scan of a patient with recurrent pyogenic cholangitis due to Clonorchis sinensis. Intra- and extrahepatic dilation and cast of stones in extrahepatic biliary system. (A) Markedly dilated intrahepatic ducts (arrows) are evident. (B) Dilated intrahepatic ducts (short white arrows) are seen. An area of increased density representing stones (arrowheads) is seen in the center of the low-density common hepatic duct (long black arrow) that is markedly dilated. A small stone (short black arrow) is present in the right posterior hepatic duct. The portal vein is compressed by the marked dilation of the extrahepatic biliary system and the porta hepatis is filled with collaterals. (C) A scan below the level of (B) demonstrates the common bile duct massively distended by a cast of stones (arrows).
and then to the lungs. Larvae pierce the alveolar capillary membrane and ascend the tracheobronchial tree where they are then swallowed and begin to mature as adults in the small bowel. A. lumbricoides may reach lengths as long as 40 cm and diameters of 3–6 mm. This is in contradistinction to C. sinensis, which is only 8–15 mm in length. Fasciola hepaticum is also quite small compared to Ascaris, being only 15–40 mm in length. The most common complications of Ascaris infestation include weight loss, cramps, and intermittent small-bowel obstruction. Adult worms are highly motile and migrate readily, resulting in worms exiting the nose or patients vomiting up large active worms. Mature adults can migrate easily through the ampulla of Vater and produce cholangitis, biliary colic, pancreatitis, cholecystitis, and hepatic abscess formation. Biliary manifestations of A. lumbricoides are so common that in many areas of the world Ascaris infestation is equivalent to gallstones as a cause of biliary tract disease.10,11 Moreover, ova and/or immature worms serve as a nidus for stone formation. Thus nearly 10% of the biliary stones
737
Section V. Liver Diseases Due to Infectious Agents
A
B
Figure 36-3. Recurrent pyogenic cholangitis due to Clonorchis sinensis that involves the lateral segment of left lobe only. (A) The left intrahepatic duct is dilated (arrows) and filled with stones of intermediate density (arrowheads). (B) The lateral segment of the left hepatic lobe appears shrunken and demonstrates increased enhancement (white arrows) surrounding a low-density mass (black arrowheads). This suggests an inflammatory pseudomass in combination with the dilated and stone-filled biliary duct shown in Figure 36-3A. Note that the right lobe is completely normal.
Figure 36-4. Endoscopic retrograde cholangiopancreatography of patient with recurrent pyogenic cholangitis due to Clonorchis sinensis. Multiple stones are seen in a markedly distended extrahepatic biliary system (black arrows). The common bile duct is filled with a cast of stones (white arrows) corresponding to the mass seen in Figure 36-2C. The intrahepatic ducts are also dilated.
738
Figure 36-5. T tube cholangiogram of patient with recurrent pyogenic cholangitis due to Clonorchis sinensis. Multiple faceted filling defects are seen in the biliary ducts that correspond to stones. The tubular filling defects (arrows) represent the Clonorchis worms.
Chapter 36 PARASITIC AND HELMINTHIC DISEASE OF THE LIVER
found in endemic areas have components of Ascaris incorporated within them.
CLINICAL PRESENTATION A. lumbricoides is most commonly found in the common bile duct (95%), ampulla (3%), and pancreatic duct (2%). In most instances, a single worm is found in the biliary tree, although a larger mass of worms can present as a pseudotumor with bile duct obstruction, mimicking a hepatobiliary tumor. In some studies, biliary ascariasis seems to be more common in patients with previous biliary tract procedures, including cholecystectomy and/or sphincterotomy. Prolonged fasting, particularly during Ramadan in Muslim countries, may also be associated with biliary migration. The typical presentation of biliary ascariasis is right upper quadrant pain, fever, right upper quadrant abdominal tenderness, nausea and vomiting (including the emesis of adult worms), jaundice, and/or right shoulder pain. The biochemical presentation is non-specific and includes elevations of the serum bilirubin levels and modest elevations of the serum alkaline phosphatase and transaminases. Mild eosinophilia is often seen.
DIAGNOSTIC TESTING Given the presentations mentioned above, the most common noninvasive imaging performed in these patients is ultrasonography. Hepatobiliary ascariasis is typically located within the extrahepatic
A
bile duct, resulting in a mildly dilated ductal system as well as a long coiled echogenic structure. Not infrequently, one may find a hyperechoic motile tube with a less echogenic center. Worm bundles, boluses, or a pseudotumor of worms packing the biliary lumen may also be seen. Stones are also associated with biliary ascariasis. Given the relatively small and tortuous cystic duct and the relatively large size of Ascaris, gallbladder ascariasis is rare (< 2%). Gallbladder features of ascariasis include the presence of coiled echogenic structures, echogenic stripes in the lumen, thickening of the wall of the gallbladder, pericholecystic fluid, and even the demonstration of rapidly moving echogenic structures in the lumen of the gallbladder. Only a limited number of CT18–20 and/or MRCP18,21 reports have surfaced concerning identification of ascariasis in the biliary system. On CT, the most common appearance is a relatively high-density tubular structure in the lumen of the extrahepatic biliary system. Occasionally, on both CT and MR scans, one can see a “bull’s-eye,” an “eye-glass,” or a “target” appearance which represents the cross-sectional manifestations of the worm (Figure 36-6).18 These findings may be associated with intraductal stones.23 On ultrasound, tubular structures can be identified which on realtime ultrasound show a writhing motion in the bile ducts and gallbladder. Worms may enter the liver parenchyma and form liver abscesses. On a barium study of the upper gastrointestinal tract, the Ascaris may be demonstrated as a tubular filling defect, occasion-
B
Figure 36-6. Patient with Ascaris lumbricoides in the gallbladder. (A) An axial ultrasound image demonstrates an Ascaris worm (arrows) in the gallbladder. The gastrointestinal tract of the worm is visualized as a hyperechoic line (arrowheads). In real-time ultrasound, the Ascaris was seen writhing around in the gallbladder. (B) The computed tomography scan shows a round filling defect (white arrows) with a central round to oval area of increased density (black arrow). This appearance has been referred to as “bull’s-eye,”“eyeglass,” or “target” lesion and represents the Ascaris worm with the central gastrointestinal tube on cross-section. Additional stones are seen near the neck of the gallbladder (arrowhead).
739
Section V. Liver Diseases Due to Infectious Agents
ally with a central line of barium that represents the gastrointestinal tract of the worm.
TREATMENT Medical therapy for enteric ascariasis includes mebendazole or albendazole. Piperazine citrate 75 mg/kg is also effective at eradicating enteric Ascaris.11 Biliary ascariasis does not reliably respond to oral therapy. While there have been isolated case reports of single biliary worms passing into the feces following the single administration of albendazole, most commonly ERCP has been employed at removing biliary worms within the ductal system. In most instances, a single worm is found within the ductal system and clearance by ERCP is over 90% successful. Worm damage to the papilla often facilitates extraction by baskets without sphincterotomy. Gallbladder ascariasis rarely, if ever, responds to medical therapy. Patients with acute or chronic cholecystitis with biliary ascariasis usually require cholecystectomy. While the response to medical and endoscopic therapy is gratifying in the vast majority of patients, reinfection often occurs because of overall poor sanitation in endemic areas.
OPISTHORCHIASIS Opisthorchis viverrini is an important liver fluke principally in Thailand, Laos, and Cambodia.24–30 In certain areas of north-eastern Thailand, 50% of the population is infected. While children are rarely infected under the age of 5, adult males are more frequently infected than females. The mature worms are usually less than 2 cm in length, living principally in the gallbladder and the intra- and extrahepatic bile ducts. Unlike C. sinensis, which may live up to 30 years, O. viverrini usually lives less than 10 years. The life cycle of O. viverrini is similar to that of C. sinensis in that eggs from contaminated feces are ingested by snails, principally by Bithynia genus. Free-swimming cercaria penetrate a host fish and become encysted as metacercaria. Up to 15 species of freshwater fish, principally of the cyprinoids (carp) family, are infected, with various levels of infection occurring during the winter in South-East Asia. Humans are infected by ingesting uncooked small and medium-sized fish, most commonly as koital in Thailand.
CLINICAL PRESENTATION While 9 million people, principally in Thailand, are infected with O. viverrini, the vast majority are asymptomatic.25,26 As apposed to C. sinensis, there are rarely any acute symptoms associated with early infection. It is estimated that only 5% of infected individuals, principally those heavily infected, have any symptoms. The most common symptoms are mild right upper quadrant pain, flatulence, and fatigue. On physical examination, hepatomegaly may be seen in up to 15% of patients. Liver function test abnormalities are rarely encountered in patients with O. viverrini infection and, when found, usually indicate a complication such as cholangiocarcinoma. Rarely, patients will present with jaundice and ascending cholangitis.
DIAGNOSIS The diagnosis of opisthorchiasis is usually made on the basis of detecting ova in stool.24–27 Many individuals will have between 80
740
and 600 adult worms in the liver and associated with an egg output of 3000–36 000 ova per day. Thus, the stools are usually positive for ova. Ultrasonography shows moderate abnormalities, including a dilated thick-walled gallbladder and usually debris in the dependent portion of the gallbladder. Gallstones are infrequently encountered in patients infected with O. viverrini. Ultrasound and CT commonly demonstrate dilation of the intrahepatic biliary tree of various degrees and a dilated and thick-walled gallbladder. At the time of diagnosis, most patients have coexisting cholangiocarcinoma, some with peritoneal seeding, and/or regional lymph node involvement.27–30 Cholangiocarcinoma is best demonstrated with CT or MR imaging.30 The primary tumor is visualized as an intrahepatic mass or as focal wall thickening of the biliary duct with increased wall enhancement and proximal obstruction. The main clinical problem with this infection is the high incidence of cholangiocarcinoma. The mature flukes attach to the biliary epithelium by oral and ventral suckers, causing desquamation of the biliary epithelium, epithelial hyperplasia, adenomatous changes of the epithelium and periductal fibrosis, and granulomata. An intense inflammatory reaction occurs in the bile ducts and together with ingested mutagens (principally demethylmetrosamines from nitrates) is likely associated with the development of cholangiocarcinoma.
TREATMENT The drug of choice for O. viverrini is praziquantel at a dose of 40–50 mg/kg.24 One-dose therapy eradicates 97% of the flukes. Praziquantel 25 mg/kg given in three doses results in 100% eradication. Mebendazole at 30 mg/day for 3–4 weeks results in up to 95% eradication while albendazole 30 mg twice daily for 7 days results in an eradication rate of only 65%.
PROGNOSIS The principal difficulty in eradicating opisthorchiasis is poor sanitation in endemic areas together with the dietary habits of consuming raw freshwater fish. In addition, the reinfection rate in endemic areas, even after successful treatment, is up to 95% at 1 year. The asymptomatic state in the vast majority of patients also complicates effective eradication.
ECHINOCOCCUS Echinococcus is a naturally occurring cestode parasite, principally of canids, including dogs, foxes, and small mammals.31–44 The tapeworms of Echinococcus occur primarily as adult worms in the small intestine of carnivores. In the usual zoonosis, the hydatid cyst stage occurs in small rodents and ungulates that make up the prey species for the carnivores. Humans are the aberrant intermediate host for the metacestode (larval) state of the disease.33 In the typical human infection with Echinococcus granulosus or E. multilocularis, eggs ingested by humans hatch in the small intestine. Active six-hooked embryos or onchospheres penetrate the small-bowel mucosa and enter the tributaries of the portal venous system. These embryos remain in the liver, grow, and enlarge, producing the hydatid cyst. While E. granulosus is found worldwide, particularly in agraraian societies raising sheep, E. multilocularis is found principally in the
Chapter 36 PARASITIC AND HELMINTHIC DISEASE OF THE LIVER
northern hemisphere. E. multilocularis is the primary tapeworm of red foxes found throughout Alaska, Canada, north central USA, northern Europe, Eurasia, and Japan. E. granulosus usually forms a large single cyst cavity with rim calcification. The more pathogenic E. multilocularis forms an alveolar echinococcosis hydatid cyst in the liver associated with dense granulomatous infiltrates, microcalcification, and necrotic cavitation. Metastatic lesions can occur through the lungs and brain. Ninety-five percent of untreated individuals with E. multilocularis die within 10 years.
CLINICAL PRESENTATION The majority of individuals infected with Echinococcus are asymptomatic.31 Vague right upper quadrant abdominal pain, a palpable epigastric or right upper quadrant mass, and/or hepatomegaly are the most common symptoms in symptomatic patients. Despite the large size of the hydatid cysts, most patients only have modest elevations of serum alkaline phosphatase. Occasionally more impressive elevations of alkaline phosphatase and bilirubin may occur since large cysts may produce bile duct strictures and/or rupture into the bile ducts, releasing biliary sand.
DIAGNOSIS Routine serologic tests of hepatic function are only mildly abnormal, principally with minimal to moderate elevations of serum alkaline phosphatase and bilirubin.31 The diagnosis is strongly suggested by non-invasive imaging, principally ultrasound, CT scans, and, more recently, MRI.33–44 The sensitivity of ultrasonography is probably greater than 60%, while CT and MR have a sensitivity of greater than 90%.39,40,44 Several serologic tests are available, including latex agglutination, immunoelectrophoresis (IEP), immunoglobulin G (IgG) ELISA, and Western blot or serum IgG antibodies. The most sensitive and specific of these appear to be IgG ELISA to the antigen B-rich fraction. This appears to have a sensitivity greater than 95%, as does the sensitivity for the Western blot for serum IgG antibodies. These serologic tests however do not distinguish active from
A
inactive forms of Echinococcus. Percutaneous ultrasound-guided needle aspiration of the cyst is not necessary and is potentially dangerous for the risk of abdominal cystercercosis. Ultrasound needle aspiration can however be extremely helpful when combined with injection of a scolicide. Hepatic cysts due to E. granulosus infection can be single or multiple, uni- or multilocular, and demonstrate wall calcifications in about 25%.38,40 A univesicular cyst in the liver favors the diagnosis of E. granulosus infection, a diagnosis that can be confirmed by immunologic studies. Cysts caused by E. multilocularis tend to be multivesicular. Abnormal areas of parenchyma are nodular or irregular, heterogeneous, and basically echogenic.41 Clustered microcalcifications are encountered within the abnormal parenchyma in 50%.41 In patients with E. granulosus, ultrasound demonstrates separation of the laminated membrane from the pericyst as a “split wall” appearance. Complete collapse of the laminated membrane can result in the ultrasonographic appearance of the “water lily sign.”38 On ultrasound, daughter cysts appear initially solid then become cystic. A fluid level can be due to hydatid sand but layering in a poorly defined cyst indicates an infected or complicated cyst.38,42 Ultrasound can define the relationship of the cyst to the hemidiaphragm better than can CT scan.42 CT can readily diagnose hydatid cysts due to E. granulosus that are often associated with curvilinear or ring calcifications, daughter cysts, and membrane detachment (Figure 36-7).43,44 CT is superior to ultrasound in demonstrating gas within a cyst as well as minute calcifications, and CT excels in anatomical mapping of the parenchymal changes.42 However, ruptured or infected hydatid cysts may be confused with abscesses on CT.45 Lesions caused by E. multilocularis or alveolaris may also be difficult to distinguish radiologically from malignant tumors.40 Generally, in patients infected with E. multilocularis, CT reveals a hypodense mass that often involves more than half of the liver with rim and central calcifications (Figure 36-8). MRI reveals hypointense signal of the infiltrative mass on both T1- and T2-weighted images. On CT and MRI, the portal vein
B
Figure 36-7. Hydatid cysts produced by Echinococcus granulosus. (A) A single cyst with a peripheral rim of calcification (arrows) is seen in the spleen. The patient also had a lesion without calcification in the liver. (B) A multivesicular lesion (arrows) is present in the medial segment of the left lobe of the liver.
741
Section V. Liver Diseases Due to Infectious Agents
the disease. Most importantly, clinicians need to be aware of the potential for echinocccal hydatid cyst disease when encountering patients with complex cystic calcified lesions of the liver.
ENTAMOEBA HISTOLYTICA IMPORTANCE AND PREVALENCE
Figure 36-8. Hydatid cyst produced by Echinococcus multilocularis or E. alveolaris. Almost the entire liver is involved with multiple cystic lesions. Some of the daughter lesions (arrows) appear solid.
branches can be seen coursing through the lesion. MRI may significantly aid in the diagnosis by demonstrating the infiltrative pattern of infestation without a significant effect on vascular structures, or on the signal characteristics.
TREATMENT Radical surgical cyst resection, whether by open surgery or by laparoscopy, is the only potential curative form of treatment.34,35 Ultrasound aspiration and injection of scolicides is usually employed in conjunction with intended subsequent surgery. There have been dramatic advances in the pharmacotherapy of Echinococcus in the past 25 years. Benzimidazole carbamates will kill metacestodes.36 Mebendazole is not generally helpful because of its poor intestinal absorption. Albendazole, a related compound with better absorption (particularly when taken with a fatty meal), is usually administered for many months, occasionally in conjunction with praziquantel. Main side effects of albendazole include alopecia, leucopenia, and occasionally elevated serum transaminase levels. Albendazole, given at 400 mg b.i.d. with meals for 28 days in three successive cycles with a 2-week rest period between cycles, is highly effective, with a cure rate of 82% compared to only 14% “cure” with placebo. In one study, albendazole 10 mg/kg per day plus praziquantel 25 mg/kg per day produced significantly better results than albendazole alone at either 10 or 20 mg/kg per day.
PROGNOSIS As mentioned above, the outcome of patients who go untreated is better with E. granulosus compared to those infected with E. multilocularis. E. granulosus may be indolent for decades and may be found only incidentally on abdominal imaging. E. multilocularis is however nearly universally fatal if untreated. Adequate medical therapy for both these conditions is usually successful at stabilizing
742
Roughly 10% of the world’s population is chronically infected with Entamoeba histolytica.45–55 Only 1% of these individuals develop clinically significant disease, with a small minority developing hepatic abscesses. Amebiasis is the second leading cause of death from parasitic disease worldwide, following only malaria. While morphologically appearing the same, there are two genetically distinct species, E. histolytica, the pathogen, and the other, E. dispar, a commensal organism. The disease is found in high prevalence in India, tropical Africa, Central America, and Amazonia.49,50 Given the extensive nature of worldwide travel, it is now no longer uncommon for clinicians in the northern hemisphere to encounter patients with severe amebic disease. Amebic liver abscesses are the most frequent and most severe of the extraintestinal clinical presentations of amebiasis.48 The trophozoites secrete proteinases that dissolve host tissue, killing host cells on contact, engulfing red blood cells, invading the mucosa of the gut, and traveling via the portal system to the liver. Here in the liver the living trophozoites stimulate a CD8 T cell, CD68 macrophage, and CD15 neutrophil response. The trophozoites produce enzymatic focal necrosis of the hepatocytes and multiple microabscesses coalesce to form single lesions, usually, but not always, in the right lobe of the liver. The center of the abscess contains a thick liquid with a reddish brown-yellow (anchovy-paste) appearance.
CLINICAL PRESENTATION The majority of patients with amebic liver abscesses present with right upper quadrant abdominal pain, fever, and hepatomegaly.44,46 Jaundice is noted in those patients with large and/or multiple abscesses. Based upon clinical parameters it is not possible to differentiate between amebic and pyogenic abscesses of the liver. Routine tests of hepatocellular function usually demonstrate a modest elevation of the bilirubin, elevation of the alkaline phosphatase, and less impressive elevations of serum transaminase levels.
DIAGNOSIS The most common finding on ultrasound and/or CT of the liver in patients with amebiasis is a relatively large (2–15-cm) spaceoccupying mass.45 This semisolid lesion is most frequently located in the posterior aspect of the right lobe of the liver.45,53 Fistulae may develop to the colon, lung, or biliary system. Ultrasound demonstrates a large, hypoechogenic or heterogenic lesion with a light and regular wall.53 CT usually depicts a lowdensity mass with an enhancing wall (Figure 36-9). High-density material and septations may be present within the abscess. CT may be of help in establishing the diagnosis and to demonstrate complications. The most common means of making the diagnosis is serologic testing rather than needle aspiration of the mass. MRI findings seen in hepatic abscesses include high signal intensity on T2-weighted images and perilesional edema. Rim enhance-
Chapter 36 PARASITIC AND HELMINTHIC DISEASE OF THE LIVER
C-reactive protein, and abscess size showed no difference between the groups. Thus routine needle aspiration of abscesses for therapeutic purposes is not necessary in uncomplicated amebic liver abscesses with diameters up to 10 cm. In those patients who fail to respond within the first 48–72 hours there may be a role for needle aspiration. This is particularly helpful if there is any doubt as to the diagnosis. Certainly, abscesses greater than 10 cm should be considered for therapeutic needle aspiration in addition to changing drug therapy to include chloroquine 300 mg b.i.d. for 21 days. In one study however metronidazole alone was effective without additional treatment in over 90% of patients presenting with confirmed amebic liver abscesses. Surgical decompression is rarely necessary, except in patients in whom the diagnosis is in doubt and/or those with extraordinarily large abscesses and/or those in whom imminent intraperitoneal rupture is anticipated.
PROGNOSIS Figure 36-9. Amebic liver abscess. CT demonstrates a large mass with an enhancing rim (white arrowheads) in the posterior segment of the right hepatic lobe. The margins are irregular and some solid areas and air are seen within the low-density mass. A peripheral ring of low density (black arrows) represents edema. A pleural effusion (white arrow) is also present on the right side.
ment and increased conspicuity are present after injection of gadopentetate dimeglumine.55 On T2-weighted images, large amebic abscesses may appear as heterogeneous, low-intensity areas surrounded by a double-layered wall. The inner layer of the abscess is hyperintense and the outer layer hypointense.54,55 The most common means of making the diagnosis is not usually needle aspiration but rather serologic testing. Three tests are available for the immunologic diagnosis of the amebiasis, including the indirect hemagglutination assay (IHA), the indirect fluorescent test (IFT), and the counterimmunoelectrophoresis test (CIET). The IHA test is highly sensitive and specific for amebic liver abscess: more than 90% of the patients are positive on presentation. This test is also a rapid method for discriminating between amebic and pyogenic abscesses. The indirect florescence antibody test may have more enhanced sensitivity. Even among patients coming from endemic areas with space-occupying mass lesions of the liver, an IHA titer greater than 1: 128 is strongly diagnostic for amebic liver abscesses. While rarely necessary, needle aspiration of the abscess for diagnostic purposes will yield typical amebic trophozoites from the periphery of the abscess cavity and/or in the last sample specimens aspirated from the cavity. The initial specimen from the needle aspirate will likely contain only necrotic cellular and amebic debris.
TREATMENT The vast majority of patients with amebic liver abscess respond to medical therapy alone with metronidazole 750 mg t.i.d. for 10 days.46–50 When compared to ultrasonic graphic guided needle aspiration of the abscess plus metronidazole, drug therapy alone was equivalent except for the faster improvement with needle aspiration in right upper quadrant tenderness during the first 3 days. All other parameters in a randomized controlled clinical trial, including erythrocyte sedimentation rate, white blood cell count, hematocrit, and
The vast majority of patients with amebic liver abscesses respond to medical treatment alone, with a minority requiring percutaneous needle drainage of the abscesses. In patients with large amebic abscesses resolution of the sterilized abscesses may take many months and the response to therapy can usually be followed by ultrasonographic techniques. In rare instances, patients with large amebic abscesses may become superinfected with Gram-negative organisms many weeks to months following resolution of the amebic liver abscess. Antibody titers remain elevated for a prolonged period of time following the successful eradication of amebic liver abscesses. There is no increased risk of hepatic neoplasm, development of cirrhosis, or progression to liver failure in patients with amebic liver disease.
SCHISTOSOMIASIS Schistosoma mansoni and S. japonicum are extremely prevalent infections in north-eastern Africa, the Arabian peninsula, and sections of eastern Asia.56–64 The infection is typically acquired by naive individuals swimming or wading in fresh water where snails of certain species are infected with schistosomes. The cercaria are freeliving larval stages of the mature worms which infect a human host by first attaching to and invading the skin passing via the systemic bloodstream to reside as mature males and females in either inferior mesenteric venules of the distal gastrointestinal tract (as is the case with S. mansoni) or in venous tributaries of the superior mesenteric venous system (as is the case with S. japonicum). The ova are covered with cytolytic enzymes that digest their way through the submucosa and mucosa and are ultimately deposited into the gastrointestinal tract. Many hundreds to thousands of ova per day are produced by the females but occasionally some of these ova embolize via the portal system to the liver. They become encapsulated by fibrosis and inflammatory mononuclear cells, producing ultimately a granulomatous and intense fibrotic reaction. In its mature stage, the inflammatory reaction produces periportal or pipestem fibrosis. In some areas of the world, such as the Aswan dam in Africa, millions of individuals are chronically infected with schistosomiasis due to dam developments which have vastly contributed to the spread of schistosomiasis.
743
Section V. Liver Diseases Due to Infectious Agents
CLINICAL PRESENTATION The most common clinical presentation of patients with schistosomiasis is hepatosplenomegaly. Hepatocellular function tests are usually abnormal only in the late stage of the disease, with decreased albumin and elevated serum bilirubin and transaminase levels. A not uncommon mode of presentation of patients chronically infected with schistosomiasis is sudden development of gastrointestinal tract hemorrhage from gastroesophageal varices. Frequently when these patients present they have preserved hepatocellular function tests yet large esophageal varices. More than half the patients with chronic schistosomiasis have esophageal varices and when variceal hemorrhage occurs the mortality from bleeding varices is greater than 10%.
DIAGNOSTIC TESTING The vast majority of patients with schistosomiasis have stools positive for ova. In areas of the world where there may be a high prevalence of coexistent hepatitis C virus (HCV) and/or hepatitis B virus (HBV) infection, the usual means of establishing the diagnosis of schistosomiasis is with supplementary serologic tests, particularly the IHA, with titers greater than 1 to 256.57 In areas of the Nile river delta a very high prevalence of coinfection with HCV has been reported, with a disproportionately higher percentage of these patients with dual infections reporting prior parenteral antischistosomal therapy.57 It appears that those patients with simultaneous schistosomal and HCV infection are more likely to progress to cirrhosis. The risk of variceal hemorrhage and/or the development of hepatoma appears to be greater in those coinfected with HCV and schistomiasis than in those with either infection alone. On ultrasound and CT, the liver often appears nodular.59 Patients with portal hypertension due to hepatosplenic schistosomiasis may develop giant paraesophageal varices that are disproportionate to the degree of portal hypertension.60 In patients with schistosomiasis japonicum, ultrasound may demonstrate a “network” pattern and CT “turtle-back” calcifications.61,62 The combination of findings by ultrasound and CT are characteristic for this type of infection. CT may show both septal and capsular calcifications,64 and enhancement of broad fibrous septa. These features are suggestive of schistomiasis japonicum, especially when no calcification is seen on unenhanced CT scans. MR may also depict these broad fibrous septa in the liver.64
TREATMENT The mainstay of treatment for schistosomiasis is oral praziquantel.57 Although this treatment is very effective in obliterating mature worms, a major problem occurs with a very high reinfection rate in endemic areas. Despite a successful course of therapy leading to eradication of mature worms, the ova emboli represent a persistent problem, leading to relentlessly progressive “pipe stem” fibrosis, and portal hypertension with recurrent variceal hemorrhage. Currently the preferred treatment for variceal hemorrhage in patients with schistosomiasis is band ligation and/or sclerotherapy. Extensive experience in the past was reported in patients treated by surgical portacaval shunting. In patients with severe end-stage liver disease from schistosomiasis there have been reports of successful liver transplants. The mainstay of treatment in endemic areas should
744
however be placed on eradication of the intermediary snail hosts by adequate disposal of feces.
REFERENCES 1. Wang KX, Zhang RB, Cui YB, et al. Clinical and epidemiological features of patients with clonorchiasis. World J Gastroenterol 2004;1:446–448. 2. Kirby CL, Horrow MM, Kotlus-Rosenberg H, et al. Ultrasound case of the day. Radiographics 1995; 15:1503–1506. 3. Lim JH, Ko YT, Lee DH, et al. Oriental cholangiohepatitis: sonographic findings in 48 cases. AJR Am J Roentgenol 1990; 155:511–514. 4. Okuno WT, Whitman GJ, Chew FS. Recurrent pyogenic cholangitis. AJR Am J Roentgenol 1996; 167:484. 5. Chan FL, Man SW, Leong LLY, et al. Evaluation of recurrent pyogenic cholangitis with CT: analysis of 50 patients. Radiology 1989; 170:165–169. 6. Lim JH. Oriental cholangiohepatitis: pathologic, clinical, and radiologic features. AJR Am J Roentgenol 1991; 157:1–8. 7. Yoon KH, Ha HK, Lee JS, et al. Inflammatory pseudotumor of the liver in patients with recurrent pyogenic cholangitis: CT–histopathologic correlation. Radiology 1999; 211:373–379. 8. Park MS, Yu JS, Kim KW, et al. Recurrent pyogenic cholangitis: comparison between MR cholangiography and direct cholangiography. Radiology 2001; 220:677–682. 9. Cha Dy, Song IK, Choi HW, et al. Successful elimination of Ascaris lumbricoides from the gallbladder by conservative medical therapy. J Gastroenterol 2002; 37:758–760. 10. Bude RO, Boweman RA. Case 20. Biliary ascariasis. Radiology 2000; 214:884–887. 11. Misra SP, Dwivedi M. Clinical features and management of biliary ascariasis in a non-endemic area. Postgrad Med J 2000; 76:29–32. 12. Javid G, Wani N, Gulzar GM, et al. Gallbladder ascariasis: presentation and management. Br J Surg 1999; 86:1526–1527. 13. Ng KK, Wong HF, Kong MS, et al. Biliary ascariasis: CT, MR cholangiopancreatography, and navigator endoscopic appearance – report of a case of acute biliary obstruction. Abdom Imaging 1999; 24:470–472. 14. Misra SP, Dwivedi M. Removal of Ascaris lumbricoides from the bile duct using balloon sphincteroplasty. Endoscopy 1998; 30:S6–S7. 15. Schulman A. Ultrasound appearances of intra- and extrahepatic biliary ascariasis. Abdom Imaging 1998; 23:60–66. 16. Sandouk F. Haffar S, Zada MM, et al. Pancreatic-biliary ascariasis: experience of 300 cases. Am J Gastroenterol 1997; 92:2264–2267. 17. Kolt SD, Wirth PD, Speer AG. Biliary ascariasis – a worm in the duct. Med J Aust 1991; 154:629–630. 18. Ng KK, Wong HF, Kong MS, et al. Biliary ascariasis: CT, MR cholangiopancreatography, and navigator endoscopic appearance – report of a case of acute biliary obstruction. Abdom Imaging 1999; 24:470–472. 19. Rocha M de S, Costa NS, Costa JC, et al. CT identification of Ascaris in the biliary tract. Abdom Imaging 1995; 20:317–319. 20. Van Severen M, Lengele B, Dureuil J, et al. Hepatic ascaridiasis. Endoscopy 1987; 19:140–142. 21. Hwang CM, Kim TK, Ha HK, et al. Biliary ascariasis: MR cholangiography findings in two cases. Korean J Radiol 2001; 2:175–178. 22. Schulman A. Intrahepatic biliary stones: imaging features and a possible relationship with Ascaris lumbricoides. Clin Radiol 1993; 47:325–332. 23. Akata D, Ozmen MN, Kaya A, et al. Radiological findings of intraparenchymal liver Ascaris (hepatobiliary ascariasis). Eur Radiol 1999; 9:93–95.
Chapter 36 PARASITIC AND HELMINTHIC DISEASE OF THE LIVER
24. Mairiang E, Mairiang P. Clinical manifestation of opisthorchiasis and treatment. Acta Trop 2003; 88:221–227. 25. Sripa B. Pathobiology of opisthorchiasis: an update. Acta Trop 2003; 88:209–220. 26. Sithithaworn P, Haswell-Elkins M. Epidemiology of Opisthorchis viverrini. Acta Trop 2003; 88:187–194. 27. Okuda K, Nakanuma Y, Miyazaki M. Cholangiocarcinoma: recent progress. Part 1: epidemiology and etiology. J Gastroenterol Hepatol 2002; 17:1049–1055. 28. Watanapa P, Watanapa WB. Liver fluke-associated cholangiocarcinoma. Br J Surg 2002; 89:962–970. 29. Abdel-Rahim AY. Parasitic infections and hepatic neoplasia. Dig Dis 2001; 19:288–291. 30. Damrongsak D, Damrongsak C, Bhothisuwan W, et al. Computed tomography in opisthorchiasis. Comput Radiol 1984; 8:379–385. 31. Craig P. Echinococcus multilocularis. Curr Opin Infect Dis 2003; 16:437–444. 32. Eickhoff A, Schilling D, Benz CA, et al. Endoscopic stenting for postoperative biliary strictures due to hepatic hydatid disease: effectiveness and long-term outcome. J Clin Gastroenterol 2003; 37:74–78. 33. Kenney PJ. Echinococcus multilocularis revisited. AJR Am J Roentgenol 2002; 178:241–242. 34. Khoury G, Abiad F, Geagea T, et al. Laparoscopic treatment of hydatid cysts of the liver and spleen. Surg Endosc 2000; 14:243–245. 35. Strauss M, Schmidt J, Boedeker H, et al. Laparoscopic partial pericystectomy of Echinococcus granulosus cysts in the liver. Hepatogastroenterology 1999; 46:2540–2544. 36. Franchi C, Di Vico B, Teggi A. Long-term evaluation of patients with hydatidosis treated with benzimidazole carbamates. Clin Infect Dis 1999; 29:304–309. 37. Beggs I. The radiological appearances of hydatid disease of the liver. Clin Radiol 1983; 34:555–563. 38. Scherer U, Weinzierl M, Sturm R, et al. Computed tomography in hydatid disease of the liver: a report on 13 cases. J Comput Assist Tomogr 1978; 2:612–617. 39. Didier D, Weiler S, Rohmer P, et al. Hepatic alveolar echinococcosis: correlative US and CT study. Radiology 1985; 154:179–186. 40. Suwan Z. Sonographic findings in hydatid disease of the liver: comparison with other imaging methods. Ann Trop Med Parasitol 1995; 89:261–269. 41. Stoupis C, Taylor HM, Paley MR, et al. The Rocky liver: radiologic-pathologic correlation of calcified hepatic masses. Radiographics 1998; 18:675–685. 42. Pandolfo I, Blandino G, Scribano E, et al. CT findings in hepatic involvement by Echinococcus granulosus. J Comput Assist Tomogr 1984; 8:839–845. 43. Munzer D. New perspectives in the diagnosis of Echinococcus disease. J Clin Gastroenterol 1991; 13:415–423. 44. Katranci N, Elmas N, Yilmaz F, et al. Correlative CT, MRI and histological findings of hepatic Echinococcus alveolaris: a case report. Comput Med Imaging Graph 1999; 23:155–159.
45. Verhaegen F, Poey C, Lebras Y, et al. (X-ray computed tomographic tests in the diagnosis and treatment of amebic liver abscesses.) J Radiol 1996; 77:23–28. 46. Torre A, Kershenobich D. Amebic liver abscess. Ann Hepatol 2002; 1:45–47. 47. Blessmann J, Binh HD, Huang DM, et al. Treatment of amoebic liver abscess with metronidazole alone or in combination with ultrasound-guided needle aspiration: a comparative, prospective and randomized study. Trop Med Int Health 2003; 8:1030–1044. 48. Amarapurkar DN, Patel N, Amarapurkar AD. Amoebic liver abscess. J Hepatol 2003; 39:291–292. 49. Kopterides P. Amebiasis. N Engl J Med 2003; 349:307–308. 50. Sharma MP, Ahuja V. Amebiasis. N Engl J Med 2003; 349:307–308. 51. Abdel-Rahim AY. Parasitic infections and hepatic neoplasia. Dig Dis 2001; 19:288–291. 52. Wynants H, Van den Ende J, Randria J, et al. Diagnosis of amoebic infection of the liver: report of 36 cases. Ann Soc Belg Med Trop 1995; 75:297–303. 53. Abdelouafi A, Ousehal A, Ouzidane L, et al. Ultrasonography in the diagnosis of liver abscesses. Apropos of 32 cases. Ann Radiol (Paris) 1993; 36:286–292. 54. Mendez RJ, Schiebler ML, Outwater EK, et al. Hepatic abscesses: MR imaging findings. Radiology 1994; 190:431–436. 55. Giovagnoni A, Gabrielli O, Coppa GV, et al. MRI appearances in amoebic granulomatous hepatitis: a case report. Pediatr Radiol 1993; 23:536–537. 56. Chaves DM, Sakai P, Mucenic M, et al. Comparative study of portal hypertensive gastropathy in schistosomiasis and hepatic cirrhosis. Endoscopy 2002; 34:199–202. 57. Gad A, Tanaka E, Orii K, et al. Relationship between hepatitis C virus infection and schistosoma liver disease: not simply an additive effect. J Gastroenterol 2001; 36:753–758. 58. Willems M, Van Buuren H, Zondervan P, et al. Schistosomiasis and presinusoidal portal hypertension. Endoscopy 2001; 33:999. 59. Cesmeli E, Vogelaers D, Voet D, et al. Ultrasound and CT changes of liver parenchyma in acute schistosomiasis. Br J Radiol 1997; 70:758–760. 60. Guermazi A, de Kerviler E, Zagdanski AM, et al. Imaging findings in giant pseudotumoral paraesophageal varices. Eur Radiol 1996; 6:451–453. 61. Cheung H, Lai YM, Loke TK, et al. The imaging diagnosis of hepatic schistosomiasis japonicum sequelae. Clin Radiol 1996; 51:51–55. 62. Patel SA, Castillo DF, Hibbeln JF, et al. Magnetic resonance imaging appearance of hepatic schistosomiasis, with ultrasound and computed tomography correlation. Am J Gastroenterol 1993; 88:113–116. 63. Monzawa S, Uchiyama G, Ohtomo K, et al. Schistosomiasis japonica of the liver: contrast-enhanced CT findings in 113 patients. AJR Am J Roentgenol 1993; 161:323–327. 64. Monzawa S, Ohtomo K, Oba H, et al. Septa in the liver of patients with chronic hepatic schistosomiasis japonica: MR appearance. AJR Am J Roentgenol 1994; 162:1347–1351.
745
Section V: Liver Diseases Due to Infectious Agents
37
BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER Birgir Johannsson and Jack T. Stapleton Abbreviations APACHE II Acute Physiology and Chronic Health Evaluation BP bacillary peliosis CT computed tomography ERCP endoscopic retrograde cholangiopancreatography FHC Fitz-Hugh–Curtis syndrome HGA human granulocytic anaplasmosis
HIV HME IL-1 LPS MAC MRI PCR
human immunodeficiency virus human monocytic erhlichiosis interleukin-1 lipopolysaccharides Mycobacterium avium-intracellulare complex magnetic resonance imaging polymerase chain reaction
PDH PID RMSF SIRS TNF-a TPN TSS
progressive disseminated histoplasmosis pelvic inflammatory disease Rocky Mountain spotted fever systemic inflammation response tumor necrosis factor-a total parenteral nutrition toxic shock syndrome
INTRODUCTION
PATHOGENESIS
Pyogenic liver abscesses are purulent collections of infectious material within the liver parenchyma. These can be single or multiple, monomicrobial or polymicrobial. Liver abscesses are not easily recognizable by history, physical examination, and laboratory evaluation, and they represent a final common pathway of many different pathologic processes. The diagnosis of liver abscesses is challenging and requires a high index of suspicion, as the clinical presentation is variable and often non-specific, and standard laboratory studies are not helpful in distinguishing liver abscess from other conditions. Untreated, liver abscesses are uniformly fatal;1 however, early detection, advances in supportive care, and refinement of non-invasive therapeutic techniques have reduced the mortality rate to approximately 15%.2–8
Approximately 50–66% of liver abscesses are solitary and usually confined to the right lobe (75%), or less frequently to the left lobe (20%), and infrequently to the caudate lobe (5%).2–4,6–9,14,15 When multiple, liver abscesses involve the right lobe more commonly than the left, although both lobes are involved in around one-third of cases.3,5,6,14 This distribution pattern is proportional to the mass of each liver lobe, but this may also represent differential patterns of hepatic blood flow. The number of abscesses is not usually reported, but is rarely more than four.8 The size of the abscesses may vary considerably, although the average abscess size in published reports is approximately 6 cm.2,3,5,15–17
EPIDEMIOLOGY Liver abscesses are relatively uncommon, accounting for 13–22 of every 100 000 hospital admissions.7–10 The incidence appears to be increasing, in part due to increased detection. Factors contributing to this increase include liver transplantation, endoscopic procedures, transcatheter chemoembolization, and ethanol injection of malignant liver lesions.7,9,11–13 Patient characteristics have changed, with patients in most early studies being males in their 30s and 40s,1 whereas currently the usual patients are men and women in their 50s and 60s.3,4,7,8,14 This probably relates to changes in predisposing conditions and treatments. In most studies there are no sex, ethnic, or geographic differences in frequency.
ETIOLOGY Liver abscesses are described according to the presumed route of hepatic invasion. First, abscesses without an identifiable underlying disease process are termed cryptogenic. Four other categories of liver abscess result from: (1) a biliary tract source; (2) spread via the portal vein or hepatic artery; (3) direct extension from a contiguous focus of infection; or (4) as a result of trauma (Table 37-1). Diseases of the biliary tree resulting in obstruction and bile stasis are the commonest sources of pyogenic liver abscesses, accounting for 33% to more than 50% of cases described.2,3,5,7–9,14,18,19 Several changes in medical practice, including the use of invasive therapeutic approaches for benign and malignant hepatobiliary and pancreatic disorders, the insertion of foreign bodies into biliary and pancreatic ducts, and surgical resection and reconstruction of the hepatobiliary system, predispose to liver abscess forma-
747
Section V. Liver Diseases Due to Infectious Agents
Table 37-1. Relative Frequency of Possible Underlying Etiologies Leading to Pyogenic Liver Abscess Formation Etiology Biliary origin Cryptogenic Portal orgin Direct extension Hepatic artery Trauma
Frequency (%) 29–62 5–59 11–48 3–29 4–10 1–4
tion.2,6,11,17,20,21 The increase is also partly due to the relative increase in malignant lesions.8,9,22 In the pre-antibiotic era, pyogenic liver abscesses frequently resulted as complications of intra-abdominal infections, particularly appendicitis, and seeding via the portal vein.1 Antibiotics and noninvasive imaging techniques have resulted in early detection and treatment of intra-abdominal processes, and a marked decrease in the incidence of pyelophlebitis and subsequent liver abscess.2,7–9 Currently, diverticulitis and inflammatory bowel disease are the most commonly associated intra-abdominal causes of liver abscesses.23 Any bacteremia may lead to seeding of the liver via the hepatic artery, including endocarditis and intravascular lineassociated sepsis. Bacteremia is responsible for approximately 5% of liver abscesses.2,7–9 Liver abscess formation may occur by direct extension from contiguous infectious foci, most frequently in association with acute cholecystitis or less commonly from paranephric, subphrenic, and paracolic abscesses.2 Penetrating, blunt, or iatrogenic trauma can lead to liver abscess formation. Pathogens introduced with penetrating trauma include skin and/or gastrointestinal flora depending upon the type of injury. The insult may be subtle, for example the ingestion of a foreign object such as toothpick or fish bones.24 Abscesses following blunt trauma may result from bacterial seeding of intrahepatic hematomas or areas of necrosis.25,26 Liver trauma from iatrogenic therapeutic procedures such as transarterial chemoembolization, percutaneous ethanol injection, or radiofrequency ablation may lead to liver abscesses, and patients with biliary tract disease or bilioenteric anastomosis are at particular risk.27–31 Liver abscesses for which no source of infection is identified are termed “cryptogenic.” These are common in recent reports, accounting for one-third to more than 50% of cases.2,7,8,14 Underlying illnesses, including diabetes mellitus, malignancy, cirrhosis, advanced cardiopulmonary disease, and iron overload states, predispose patients to the development of cryptogenic pyogenic liver abscess.6–9,27,32–37 There is not consensus on how extensive an evaluation should be in order to identify an underlying disease process leading to cryptogenic liver abscess. Thorough investigation occasionally identifies underlying conditions (diverticulosis, inflammatory bowel disease, colon or gallbladder carcinoma), and occasionally identifies foreign bodies that may have contributed to the abscess formation.4 However, extensive evaluation is not usually costeffective, and a focused evaluation based upon the individual patient’s clinical picture has been advocated (reviewed by Seet and Rocky8). All patients should be evaluated for evidence of biliary tree disease, gastrointestinal symptoms, bacteremia (before initiating antimicrobial therapy), and gastrointestinal bleeding, and abnormalities should be followed up with appropriate studies.
748
MICROBIOLOGY The microbiology of liver abscesses depends upon the predisposing pathologic process, thus it is difficult to generalize about specific organisms responsible for liver abscess formation. It is important to recognize that antibiotics are frequently administered before definitive cultures have been obtained, and subsequently, bacteria recovered from blood and abscess cultures may not represent the full microbiological content of the abscess.4,8,19,24 In cases in which both blood and abscess cultures are positive, blood cultures tend to identify fewer types of organisms than the abscess cultures8,19,24 because prior antibiotic use more profoundly inhibits the recovery of bacteria from blood than abscess cultures. Traditional Gram stain, blood cultures, and invasive culture of the liver abscess material are complementary, and ideally all should be used.19 Although Gram stain of abscess fluid is effective in detecting Gram-positive cocci, it may miss Gram-negative bacilli due to the lack of contrast between background material and the organisms.19 Nonetheless, Gram stain may provide the only clue that an abscess is polymicrobial.19 Culture of liver abscess material most commonly yields a single organism, although polymicrobial infection occurs in up to one-third of cases.2,4–6,19,20 The rate of positive cultures ranges from 50% to 90% in different studies.2–4,6,8,9,14,16,20 Blood cultures are positive in about 50% of cases, and may identify different organisms than those recovered from abscess material.19 The history of an indwelling biliary stent, prior hepatobiliary surgery, repeated cholangitis, and antibiotic use are all associated with infection with more resistant organisms such as enterococci, Enterobacter and Pseudomonas species.4,9,11,14,20 A weak association has been described between single abscesses having monomicrobial growth, and multiple abscesses demonstrating polymicrobial growth on cultures.8 With the development of better transport media and culture methods, identification of anaerobic organisms has increased.1,2,7,8,38 Anaerobic bacteria are present in ~25% of cases, usually in mixed infections. Gram-negative aerobic bacteria are the most commonly identified pathogens in liver abscess (approximately 50%), followed closely by Gram-positive aerobic bacteria (approximately 33%; Table 37-2).2,7,8,38 Overall Escherichia coli and Klebsiella pneumoniae are the most frequently isolated;2,6–8,14 however, their relative frequency has been changing. K. pneumoniae appears to be occurring more frequently in some series, especially in Taiwan where diabetic patients are particularly afflicted.39,40 Among Grampositive organisms, viridans streptococci, microaerophilic streptococci, enterococcus and staphylococci have all increased in frequency when evaluated for 20-year intervals before and after 1972.9 Bacteroides and Fusobacterium species are the most frequent anaerobes cultured from liver abscesses. Fusobacterium necrophorum is part of the normal flora of the tonsillar crevices, and can cause suppurative pharyngitis that extends to involve the jugular vein with resultant septic thrombosis, bacteremia, and metastatic abscesses, including liver abscess. This constellation of findings is known as Lemierre’s syndrome.41,42
CLINICAL FEATURES Diagnosis of pyogenic liver abscesses is challenging and requires a high degree of suspicion, since presenting symptoms are varied and
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
Table 37-2. Microbiology of Pyogenic Liver Abscesses Common: >10%
Table 37-3. Symptoms Associated with Pyogenic Liver Abscesses
Uncommon: 1–10%
Symptoms
Escherichia coli Klebsiella pneumoniaeb
Pseudomonas spp.a Enterobacter spp.a Proteus spp. Citrobacter spp. Serratia spp. Yersinia spp.
Fever Chills Abdominal pain Right upper quadrant pain Malaise Weight loss Nausea and vomiting Diarrhea Respiratory symptoms Peritoneal irritation
Viridans streptococci
Enterococcus spp.a Staphylococcus aureus Streptococcus pneumoniae
Bacteroides spp.
Fusobacterium spp. Peptostreptococcus spp. Prevotella spp. Clostridium spp. Lactobacillus spp.
Gram-negative aerobic bacteria
Gram-positive aerobic bacteria
Anaerobic bacteria
Microaerophilic bacteria Streptococcus milleri
Frequency (%) 60–100 25–78 36–74 23–69 27–53 5–51 22–41 6–23 9–21 3
Table 37-4. Physical Signs Associated with Pyogenic Liver Abscesses Physical signs Abdominal tenderness Jaundice Hepatomegaly Chest findings Hypotension with altered mental state
Frequency (%) 43–65 7–54 13–48 22 3–20
a
Incidence is increased in patients with a history of biliary tract manipulations and in health care-associated infections. Geographic variations exist. This is the most commonly isolated organism in Asia.
b
Table 37-5. Laboratory Findings Associated with Pyogenic Liver Abscesses
frequently non-specific. The spectrum of presenting symptoms and signs range from long-standing fever of unknown origin to acute abdomen requiring emergent surgical evaluation.4,6,8,38 The diagnosis of liver abscess was made in less than 20% of emergency-room patients prior to admission in one study, illustrating the difficulty of making an early and accurate diagnosis. Among those in whom the correct diagnosis was made, most had a past medical history of orthotopic liver transplantation or biliary tract disease, thus drawing attention to the right upper quadrant.43 Symptoms are present an average of 2 weeks before the diagnosis is made, although one-third of patients may be symptomatic for more than a month.3–5,8,14,16,38,43 The underlying condition predisposing to liver abscess formation may influence the acuity and severity of presenting signs and symptoms. Patients with hematogenous liver abscess have the shortest period of prodromal symptoms, whereas those with pylephlebitis had the longest (3 and 42 days respectively).8 The duration of prodromal syndromes does not appear to be different in patients with single versus multiple abscesses.8,14 Fever, chills, and other constitutional symptoms such as anorexia, malaise, and weight loss are typically present (Table 37-3). However, many patients may be afebrile on presentation and remain so throughout the course of their illness.8 When present, right upper quadrant pain, nausea, and vomiting are still non-specific and the classical triad of fever, right upper quadrant pain, and jaundice are present in only 10% of patients with liver abscess.8 Physical findings may be vague, with diffuse abdominal tenderness, or even misleading, with right chest findings being relatively common.8 Right upper quadrant pain, hepatomegaly, and jaundice are present in about half of patients, and suggest biliary tract disease
Laboratory finding Leukocytosis Elevated alkaline phosphatasea Elevated transaminasesa Anemia Hypoalbuminemia Elevated bilirubina Elevated prothrombin time
Frequency 64–100 55–100 22–90 47–77 59–71 30–63 36–62
a
Elevations more frequent and pronounced in patients with pyogenic liver abscess of biliary origin.
or grave illness.2,3,5,8,9,14,43 A number of patients present with hypotension and altered mental state consistent with septic shock (Table 37-4).2,3,5,6,8,14,40 Laboratory abnormalities may suggest liver disease, but are not specific for liver abscesses (Table 37-5). Leukocytosis is present in most patients and can be quite prominent. Hypoalbuminemia and anemia are often present, and most likely reflect long-standing illness. Although abnormalities in liver-associated enzyme levels are frequent, normal tests do not exclude the presence of pyogenic liver abscess. Bilirubin and alkaline phosphatase elevation are typically more striking than those of the transaminases, especially in patients with underlying biliary tract pathology and those with a history of pancreatobiliary surgery.6,8,20 Patients with multiple liver abscesses appear to have more pronounced leukocytosis and more marked elevations of alkaline phosphatase and bilirubin.3,14 Ultrasound and computed tomography (CT) are essential for the prompt and accurate diagnosis of pyogenic liver abscesses. Previously, plain radiographs and radionuclear imaging were the most
749
Section V. Liver Diseases Due to Infectious Agents
commonly used techniques. Although radionucleotide imaging may identify liver abscesses, it is less sensitive than ultrasound and CT (Table 37-6).2–4,7–9,11,44 Because of the non-specific signs and symptoms, plain radiographs of the chest and abdomen are commonly obtained. These images may be helpful, showing air–fluid levels within an abscess cavity in up to one-third of abdominal radiographs.3,8 Chest radiograph abnormalities are typically confined to the right side and include hemidiaphragm elevation, pleural effusion, and atelectasis or consolidation of the lung base. These findings, though non-specific, occur in one-third to one-half of patients, and when present should place liver abscess in the differential diagnosis.4,7,8 Ultrasound and CT are highly sensitive, and CT is more likely to identify small single abscesses located in the right liver lobe.8,9 Both techniques may provide important information on underlying diseases predisposing to liver abscess, as well as being able to guide therapeutic procedures.3,44 Intravenous contrast media enhancement is essential for CT diagnosis, as it accentuates the unenhanced hypodense lesions. Liver abscesses have a wide range of appearance and rim enhancement is relatively uncommon (Figure 37-1).44 Ultrasound is the method of choice when contrast media and/or radiation needs to be avoided.3,44 On the other hand, CT is preferred in patients with surgical wounds and drains, as well as for guidance during complex drainage procedures (Figure 37-2).44 Features iden-
Table 37-6. Sensitivity of Different Diagnostic Methods in Identifing Pyogenic Liver Abscesses Sensitivity (%) Computed tomography Ultrasound Radionucleotide imaging Abnormal chest radiograph Abnormal abdominal plain film Abscess culture Gram staina Blood culture
93–100 68–95 70–72 23–49 7–29 44–97 79 24–60
a
Sensitivity differs for Gram-positive cocci (90%) and Gram-negative bacilli (52%).
tified on ultrasound can be highly variable and non-specific, ranging from hypo- to hyperechoic lesions with varying degree of internal echoes and debris.44 Nevertheless, in instances in which the clinical suspicion of liver abscess is high, some authors believe that ultrasound should be the primary imaging method used to diagnose liver abscess, and that CT be reserved for instances in which ultrasound is negative or inconclusive.3,44 Magnetic resonance imaging (MRI) is a highly sensitive method capable of detecting small lesions within the liver. However it is limited by the fact that it may not be used to guide therapeutic drainage procedures.3,9,44 Characteristic imaging features on MRI are based on the abscess cavity and wall, perilesional signal, and the dynamic enhancement pattern noted with gadolinium. The abscess cavity is usually hypointense on T1-weighted and hyperintense on T2-weighted images. The most distinctive feature of liver abscesses on MRI is the enhancement of the abscess wall on dynamic postgadolinium images.45 Endoscopic retrograde cholangiopancreatography (ERCP) may be of value in the evaluation of pyogenic liver abscesses, particularly when biliary tract disease is present. ERCP may also help to determine the need for surgical intervention, or if the condition is amenable to endoscopic procedures.3,5,6,8,46 Some authors suggest that a complete evaluation of the biliary tract is indicated in patients with cryptogenic liver abscesses, as these patients frequently have underlying biliary tract disease not identified by laboratory evaluation or diagnostic imaging. 3,5,6,8,14 Although the detection and management of pyogenic liver abscesses have benefited considerably from advances in imaging technologies, two new techniques may improve diagnostic sensitivity in the future. Low mechanical index contrast-enhanced ultrasound is a method that may offer improved detection and characterization of hepatic lesions, utilizing both intermittent and continuous specific microbubble contrast techniques.47 Diffusionweighted MRI is limited by physiologic motion of respiratory and cardiac movement; however, new techniques using ultrafast singleshot echo planar imaging allow diffusion-weighted MRI to be used in the abdominal cavity. This may prove to be a helpful noninvasive modality in differentiating liver abscess from cystic or necrotic tumors in the liver.48
Figure 37-1. (A,B) Computed tomography findings of liver abscess.
A
750
B
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
A
B
C
D
Figure 37-2. (A–D) Percutaneous drainage of liver abscess. The hypodense lesion (A and B) represents an intrahepatic abscess. The percutaneous drain (C) and abscess decompression is shown in panel C. Fluoroscopy demonstrates contrast outlining the abscess carity (D).
Ultimately the definitive diagnosis of pyogenic liver abscess is dependent on obtaining purulent material from the abscess cavity for microbiological culture. The utility of traditional Gram stain, blood cultures, and invasive liver abscess cultures in the diagnosis of liver abscesses is discussed elsewhere and noted in Table 37-6.
DIFFERENTIAL DIAGNOSIS Pyogenic liver abscesses are mistaken for primary and metastatic malignancies and vice versa.6,27,35,49,50 Liver abscesses may also arise within necrotic centers of malignant lesions or proximal to an obstructed bile duct caused by tumor fragment, complicating the diagnosis of either condition.27 Meaningful clues to the presence of hepatocellular carcinoma are history of hepatitis B or C infection, underlying chronic liver disease, unexplained anemia, or marked weight loss.27 The risk of developing a pyogenic abscess within the necrotic center of a hepatocellular carcinoma after iatrogenic manipulations is increased with a large tumor burden, tumor causing biliary tract obstruction, and among patients with diabetes mellitus.35 The converse has been described, and a malignancy has occurred within the cavity of a successfully treated liver abscess.9 As noted previously, diffusion-weighted MRI may help to differentiate liver abscess from cystic or necrotic tumors.48 Pyogenic liver abscesses may be indistinguishable from other infectious lesions based on clinical grounds or by laboratory and imaging, and invasive approaches are the only definitive way to differentiate between these lesions. One exception is that when the history suggests ameba exposure, a positive indirect haemagglutination test for Entamoeba histolytica is highly specific and sensitive for diagnosing a (usually) solitary amebic liver abscess.14,51
ASSOCIATED CONDITIONS Among systemic illnesses and local factors associated with pyogenic liver abscesses, diabetes mellitus is the most frequent comorbid disease, occurring in 15–78% of patients.6–9,14,18,28,34,40 This association occurs worldwide, but is particularly prominent in the eastern hemisphere. As noted, in Taiwan K. pneumoniae liver abscesses are frequently found in patients with diabetes mellitus, and are associated with a higher frequency of serious complications related to septic embolization than in non-diabetics.18,34,38,40,52–54 Various immune defects, including decreased neutrophil chemotaxis, phagocytosis, and killing, as well as monocyte and macrophage dysfunc-
tion, are found in patients with liver abscess. Immunoglobulin levels are usually normal, although cytokine production in response to specific stimuli and complement levels may be decreased.55,56 Other illnesses, including malignancy, cirrhosis, advanced cardiopulmonary disease, and iron overload, also appear to predispose to liver abscess formation.3,6–9,27,32,34–37,40 Inherited disorders resulting in functional or quantitative neutrophil defects, such as Papillon–Lefèvre syndrome, chronic granulomatous disease, and Job’s syndrome, are also predisposing factors for abscess formation both in the liver and elsewhere.10,57,58 Caroli’s disease, an inherited disorder where the bile ducts are enlarged, is associated with stone formation and infection of the biliary tree leading to liver abscesses. In patients with travel to endemic areas, invasion of the biliary tree by Ascaris lumbricoides can lead to the development of liver abscess.59,60 Hepatobiliary and pancreatic carcinomas are local factors commonly encountered in patients with liver abscesses, especially following surgical reconstruction of the biliary tract or vascular supply.6,9,11,20
DISEASE COMPLICATIONS Pyogenic liver abscesses may be complicated by rupture or extension into adjacent structures or septic metastatic embolization (Table 37-7). Rupture or extension may lead to fistula formation between the liver and other intra- or extra-abdominal organs,61–63 or may lead to peritonitis requiring urgent surgery.14,64 Hepatic artery pseudoaneurysm may complicate non-traumatic pyogenic liver abscess; however, this is rare.65 Fewer than 5% of liver abscesses recur.5–8,14,15,17 When this happens, abscesses tend to appear months to years after initial diagnosis.5,6 Late recurrences usually result from persistent local pathology rather than ineffective initial therapy. Bacteremia complicating liver abscesses may lead to sepsis and multiorgan failure.2 Septic metastatic embolization occurs in 3–8% of patients, and may have grave consequences. Septic emboli may lead to bacterial endogenous endophthalmitis.18,52–54 Recognition of endophthalmitis is critical, as prompt diagnosis and treatment may prevent blindness. Symptoms (blurry vision, pain, and redness of the affected eye along with progressive loss of visual acuity and pupillary hypopyon) usually develop within 48–72 hours following the diagnosis of liver abscess.53,54 Endophthalmitis is particularly common in Asian diabetic patients and the pathogen is almost exclu-
751
Section V. Liver Diseases Due to Infectious Agents
Table 37-7. Relative Frequency of Complications Associated with Pyogenic Liver Abscesses Common: >10%
Uncommon: 1–10%
Very rare: <1%
Bacteremia Pleural effusion
Sepsis Endophthalmitisa Wound infection Rupture of abscess Recurrent abscess Acute renal failure Acute respiratory failure Prolonged biliary drainage Postoperative bleeding Subphrenic abscess Pleuropulmonary fistula Fistula to adjacent organs
Postoperative ileus Postoperative pancreatitis Liver failure Hepatobronchial fistula Mesenteric venous thrombosis Hepatic artery pseudoaneurysm
a
Geographic variations exist. More common in Asia in association with Klebsiella pneumoniae in patients with diabetes mellitus.
sively K. pneumoniae (serotypes K1 and K2).18 Treatment includes a combination of intravenous and intravitreal antibiotics that penetrate the aqueous humor of the eye and surgical intervention. Third-generation cephalosporins are generally recommended and ciprofloxacin has been used in the case of penicillin allergy.53,54
TREATMENT Conventional approaches of pyogenic liver abscess therapy include a combination of broad-spectrum antibiotics and abscess drainage.2,8,14,39 Open surgical drainage has largely been replaced with image-guided percutaneous drainage using either intermittent needle aspiration or continuous catheter drainage2,5,8,16 (Figure 372). Laparoscopic surgery has emerged as a less invasive surgical option.15 Endoscopic drainage alone or in combination with percutaneous drainage may also be an effective approach.3,6,17,22 Taking into consideration the many safe and effective therapeutic options available, many of which are complementary to each other, it is clear that the treatment of pyogenic liver abscesses must be highly individualized (Figure 37-3).
therapy.5 Clearly, use of an appropriate oral regimen after the initial stabilization with intravenous antibiotics reduces hospital stay and overall cost, and minimizes the potential risks associated with prolonged use of intravenous catheters.5 When pyogenic liver abscess is suspected by imaging studies, it is often necessary to start empiric antibiotic therapy before abscess cultures have been obtained, although blood cultures should always be obtained prior to antibiotics.66 Initial antibiotic regimens should have activity against aerobic and anaerobic enteric bacteria as well as microaerophilic streptococci. Routine coverage for enterococci is generally not required;66 however, several factors may increase the likelihood of enterococcal infection such as biliary tract disease, history of biliary tract procedures, and health care-associated infections.7,66 A review of the local bacteriological susceptibility profiles is critical, as common pathogens are increasingly resistant to commonly prescribed antibiotics. For example, the Bacteroides fragilis group has increasing rates of resistance to clindamycin, cefotetan, cefoxitin, and quinolones; thus these antibiotics should not be used unless they are combined with metronidazole.66 In addition, E. coli is increasingly resistant to ampicillin/sulbactam.66 An additional consideration in selecting the initial empiric regimen is whether the infection is acquired in the community or in the health care setting, as health care-associated infections are more likely to be caused by Pseudomonas aeruginosa, Enterobacter species, methicillin-resistant Staphylococcus aureus and enterococci.9,11,20,66 Consideration of the underlying process also influences the antibiotic regimen required. For example, pathogens obtained from abscesses resulting from hematogenous spread frequently differ from abscesses resulting from pylephlebitis. The empiric antibiotic regimen needs to be modified based on the susceptibility of the recovered pathogen(s), the patient’s clinical response, and results of repeat imaging studies. When combined with a drainage procedure, a minimum of 1–3 weeks of intravenous antibiotics followed by oral antibiotics to complete a 4–6-week course is usually sufficient. With effective therapy patients usually become afebrile in less than a week.3,5,8,16,39 While the abscess cavity may resolve after successful therapy, in as many as half of cases it does not.10,16 If the remaining cavity is stable on serial imaging studies and the patient is asymptomatic, antibiotic therapy can be discontinued with close follow-up observation.
PHARMACOLOGIC In unusual situations where a drainage procedure is not possible due to extremely small abscess(es) or due to patient instability, use of antibiotics alone may cure liver abscesses.2,8,14,39 This approach should be limited to these situations, as there are no prospective trials comparing antibiotics alone to the more conventional combined approach. Furthermore, it is impossible to determine which patients and what abscess characteristics are most likely to respond to antibiotic therapy alone,39 although retrospective reviews report favorable outcomes with small abscesses and prolonged antibiotic therapy.8,39 There is no clear duration of antibiotic therapy for the treatment of liver abscess with drainage; however, the average duration of successful therapy is around 6 weeks.3–5,8,39 In a retrospective analysis, intravenous antibiotics followed by an appropriate oral antibiotic regimen proved safe and as effective as continuous intravenous
752
PERCUTANEOUS DRAINAGE Using imaging techniques to localize liver lesions accurately in real time, percutaneous drains may be safely inserted into abscess cavities. This is the preferred method for draining liver abscesses,3,4,8,15,16,28,44 and it has markedly improved success rates and decreased complications compared to surgical drainage.2,8,16,28 Aspirated material should be sent for cultures and pathologic examination. A limitation of percutaneous drainage is that underlying conditions may not be addressed, which potentially results in higher abscess recurrence rates.8 Endoscopic evaluation of the biliary tree to identify patients in need of further therapeutic procedures, either surgical or endoscopic, has been recommended by some.8,17,22,67 Although patients with ascites or bleeding diatheses are said to be at higher risk of complications from percutaneous drainage,9,15 one large study found
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
Liver abscess noted on imaging
Obtain blood cultures Consider amebic precipitin test
Initiate broad spectrum antibiotics • Active against: - Aerobic and anaerobic enteric bacteria - microaerophillic streptococci • Consider: - Local bacteriological susceptibility profiles - Community or health-care associated infection - Possible underlying cause
Concomitant: • Intra-abdominal pathology amenable to surgery • Severe ascites • Coagulopathy* • Acceptable surgical risk
No
No
Biliary source suspected
Percutaneous drainage: Intermittent needle aspiration Continuous catheter drainage Yes
Remove catheter when drainage minimal Repeat imaging to evaluate for need to repeat aspiration
Yes
Yes
Surgical drainage
ERCP
No
Clinical improvement in 48–72 hours
Reevaluate diagnosis Reevaluate microbiologic data Ensure coverage for enterococci
Repeat imaging Continue IV antibiotics for 2 weeks Use oral antibiotics for another 4 weeks
Repeat imaging at end of therapy
Persistent abscess
Biliary tract disease
Concomitant surgical disease
Repeat aspiration Reposition catheter
ERCP
Surgical drainage
Clinical improvement Yes
No
Figure 37-3. Algorithm showing an approach to the management of pyogenic liver abscesses. *Permissive coagulative parameters for percutaneous drainage procedures: platelets > 45 000/mm3; international normalized ratio = 1.7, stop anticoagulation medications and aspirin 1 week and 2–3 days prior to procedure, respectively. ERCP, endoscopic retrograde cholangiopancreatography; IV, intravenous.
753
Section V. Liver Diseases Due to Infectious Agents
no major complications or death over a 22-year observation period.28 Percutaneous drainage was carried out if platelets were >45 000 cells/mm3, international normalized ratio was 1.7 or less, and anticoagulation medications and aspirin were discontinued 1 week and 2–3 days before the procedure, respectively. Thus, percutaneous aspiration and drainage appear to be safe in patients with mild to modest coagulation disorders. Transient hyperthermia (up to 40°C) may occur in the first 24 hours after liver abscess drainage.28 Other complications, including bleeding, pneumothorax, septicemia caused by manipulation of abscess cavity, catheter blockage, and dislodgement, are infrequent.16,44 In most instances, the percutaneous catheter is left in place for 1–2 weeks or until drainage has reduced to small volumes and is stable. Percutaneous aspiration without catheter placement may require repeated aspirations; however, approximately 40% of aspirations do not require repeat aspiration, and only two attempts are needed in another 40%.6,16,68 Some studies suggest that continuous catheter drainage is preferred for patients with multiple abscesses,9 although this has been disputed by others.8 For example, there was no difference in treatment outcome among 64 patients randomly assigned to continuous catheter drainage or intermittent needle aspiration,16 including time to defervescence and normalization of liver-associated enzyme levels. Of three outcome measures studied, the length of hospital stay (11 versus 15 days), treatment success rate (93.8 versus 84.4%), and mortality (3.1 versus 12.5%) favored the intermittent needle aspiration group, although the results were not statistically significant.16 Abscess characteristics (number, size, or loculation) did not appear to influence the effectiveness of either intermittent or continuous percutaneous drainage.8,16,68 Intermittent aspiration utilizes a smaller needle that is potentially more acceptable to the patient, and it eliminates problems related to an indwelling catheter. Thus, intermittent needle aspiration appears to be safe and effective, and many consider this the first line of therapy for percutaneous liver abscess drainage.3,16 However, it is important to appreciate that multiple aspirations may be required using this method.
ENDOSCOPIC DRAINAGE Endoscopic placement of a nasobiliary stent has been shown to be effective for liver abscesses of biliary tract origin.22 However, this approach requires concomitant percutaneous abscess drainage. The main disadvantage of primary nasobiliary drainage is that no definitive invasive cultures can be obtained. Furthermore, most abscesses with biliary tract communication respond to percutaneous drainage.22 If there is an obstruction within the biliary tree or the abscess is not responding appropriately to percutaneous drainage alone, endoscopic therapeutic procedures and drainage may eliminate the need for surgical procedures.3,17,22 Although ERCP may have a limited therapeutic role, it is helpful in determining the extent of biliary tract disease and may assist in determining if surgical resection is required.6
SURGICAL DRAINAGE Although surgical drainage of liver abscesses was considered standard therapy in the past, this approach has steadily declined in frequency. In the past 10 years only 2% of people with liver abscess
754
had primary surgical drainage.8 Surgery is now reserved for patients with a complicated abscess, including rupture on presentation, association with hepatolithiasis, or in those who fail to respond to percutaneous therapy.7,15,16,40,44,67 Surgical drainage may be used in patients with concurrent intra-abdominal pathology requiring surgical management or when percutaneous drainage is felt to be contraindicated because of abscess location or factors such as severe ascites or bleeding tendency.7,15,16,44 In rare instances, abscess may lead to necrosis of a segment or a lobe of the liver requiring surgical resection.14 Conventionally, open surgery is used; however, laparoscopic surgery is increasingly being used in patients with liver abscess.15 Intraoperative ultrasonograpy is usually used to localize the abscess, guide the collection of samples for culture and pathologic examination, and to facilitate debridement and irrigation of the abscess cavity as well as to confirm that no residual abscesses are present.15 In general laparoscopic surgery is a safe and effective alternative to open surgery in patients who have failed conventional therapy with percutaneous drainage.15
PROGNOSIS AND NATURAL HISTORY As noted, untreated liver abscesses are invariably fatal. Currently, in spite of effective antimicrobials, enhanced diagnostic imaging capability, and improved drainage methods, mortality remains substantial, with an overall mortality ranging from 3% to 15%.2–8 Increased mortality was associated with multiple lesions; however, this association does not appear to apply with current drainage and antibiotic management.2,3,7–9,14 Nonetheless, complications of liver abscesses may still affect the outcome, with septicemia, underlying illnesses, and general condition of the patient strongly predicting poor outcome or death.2,7,9,40 Of these factors, cirrhosis and underlying malignancy, in particular hepatobiliary or pancreatic carcinomas, are significantly associated with increased mortality.3,4,7–9,11,32,40 Hypoalbuminemia, anemia, and high Acute Physiology and Chronic Health Evaluation (APACHE II) score have also been shown to predict poor outcome.2,9 Less favorable outcomes are observed when biliary tract disorders are the underlying process when compared to cryptogenic abscesses.2,4,7–9
CONCLUSIONS The clinical diagnosis of pyogenic liver abscesses is challenging and requires a high degree of suspicion. Hepatic imaging is essential for the diagnosis of liver abscesses, and ultrasound is usually the preferred method, although contrast-enhanced CT is required in patients for whom the initial evaluation is negative or inconclusive. Treatment options are varied and continue to evolve. Percutaneous intermittent needle aspiration is emerging as the preferred drainage procedure for most patients, and surgical management is not usually required. For patients who require surgical intervention, laparoscopic surgery is becoming a more widely accepted alternative to open surgery. In suspected biliary tract disease, endoscopic evaluation and procedures are complementary to percutaneous drainage.
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
ENDOTOXIN AND INFLAMMATORYRELATED HEPATITIS Six to 65% of neonates, children, and adults with bacterial infections have abnormal levels of liver-associated enzymes and bilirubin.69,70 This is referred to as “parainfectious hepatitis.” These hepatic laboratory abnormalities may accompany the syndrome of generalized systemic inflammation response (SIRS: defined as two or more changes in four factors: (1) body temperature; (2) heart rate; (3) respiratory function; and (4) peripheral leukocyte count). SIRS is frequently, but not always, associated with bacterial infection. Clinically, elevated serum bilirubin levels are frequently out of proportion with the serum alkaline phosphatase and aminotransferase levels, and hyperbilirubinemia is usually more severe among patients with underlying hepatobiliary disease, and may precede positive blood cultures by up to 9 days.69 Mortality is highest in patients with underlying liver disease, and mortality approaches 100% in patients with persistent or increasing hyperbilirubinemia.71 In patients without pre-existing liver disease, mortality is usually lower,70 presumably related to differences in patient populations. The etiology of the underlying infection is most commonly E. coli, although hyperbilirubinemia may occur in association with virtually any bacterial organism. Prognosis appears to be significantly worse when SIRS accompanies S. aureus sepsis.72 Although the primary site of infection is usually from an intra-abdominal source, this syndrome may accompany pneumonia, endocarditis, pyelonephritis, and soft-tissue and pelvic abscesses.73 Sepsis-associated cholestasis should always be considered in the differential diagnosis of jaundice in hospitalized and critically ill patients. If a patient has disproportionate elevations in bilirubin compared to serum alkaline phosphatase and serum aminotransferases, sepsis-associated cholestasis should be considered, even in the absence of fever, leukocytosis, or other signs or symptoms. Early recognition and treatment of the underlying process are critical for reducing morbidity and mortality. Unless prolonged hypotension occurs, liver failure or hepatic necrosis is uncommon. Parainfectious hepatitis appears to be caused by the systemic effects of the release of inflammatory mediators, and active infection is not required for the syndrome to occur. The outer membrane of Gram-negative bacteria contains lipopolysaccharides (LPS or endotoxin), which may be released into the bloodstream even without bacteremia. LPS stimulates macrophages (or, in the case of the liver, Kupffer cells), hepatocytes, and bile duct epithelial cells to release a variety of cytokines, including tumor necrosis factor-a (TNF-a), interleukin-1 (IL-1), 6, and 8, directly affecting hepatocyte function.74 Therapeutic use of TNF-a and IL-2 and administration of LPS to human volunteers results in intrahepatic cholestasis.74 Bacterial products other than LPS can also trigger release of proinflammatory cytokines, resulting in parainfectious hepatitis. Patients with alcoholic liver disease frequently have detectable endotoxin in their portal and systemic venous systems,75 and this may be associated with increased plasma cytokine levels.76 Interventions that block TNF-a (pentoxifylline or anti-TNF-a antibodies) appear to block alcoholic liver injury in experimental animal models, suggesting that these agents may be useful in alcoholic hepatitis. Endotoxin may also play a role in the jaundice associated with
total parenteral nutrition (TPN),77 which may explain why TPNrelated hepatitis can sometimes be prevented or improved with antibiotic therapy.78 In addition to endotoxin-induced parainfectious hepatitis, hepatitis is commonly a complication of bacterial toxin-mediated syndromes, including staphylococcal and streptococcal toxic shock syndrome (TSS). Reduced hepatic function and gastrointestinal symptoms of vomiting and diarrhea are included as criteria for the diagnosis of TSS (reviewed by Hughes and Stapleton79).
PYELOPHLEBITIS Suppurative thrombophlebitis of the portal vein, or pyelophlebitis, is a life-threatening complication of intra-abdominal infections. While pyelophlebitis may not directly involve the liver, bacteremia via the portal system can lead to liver abscesses. Although currently rare, pyelophlebitis was common in the pre-antibiotic era, and was a common cause of liver abscess. Diverticulosis is the most common predisposing cause of portal vein suppurative thrombophlebitis, although it may also develop following appendicitis, necrotizing pancreatitis, inflammatory bowel disease, or any process resulting in bacteremia of the portal vein.80 Signs and symptoms (fever, chills, abdominal pain, diarrhea, jaundice, tender hepatomegaly) are not specific, but suggest an intra-abdominal process or a fever of unknown origin.81 Leukocytosis and blood cultures are positive in 50–80% of cases. A radiographic clue to the diagnosis is gas in the portal area; however, this is relatively insensitive. Contrast CT, ultrasonography, MRI, magnetic resonance angiography and positron emission tomography scanning have all been successfully used to diagnose portal vein thrombosis.82–84 The bacteria isolated reflect bowel flora, with E. coli, Proteus mirabilis, Bacteroides fragilis, and other aerobic Gram-negative bacilli commonly isolated. Therapy consists of prompt initiation of broad-spectrum antibiotics to cover Gram-negative aerobic bacilli, anaerobic bacteria, and streptococcal species. The duration of antibiotics is not clear, although most authors recommend a minimum of 4 weeks, and longer courses in conjunction with surgical or percutaneous drainage may be needed if abscesses complicate the pyelophlebitis.85 The role for anticoagulation therapy is controversial, and no formal studies have been conducted to provide evidence from which to determine their efficacy and safety. In spite of appropriate therapy, pyelophlebitis has a mortality rate of >30%.83
BACTERIAL AND FUNGAL HEPATITIS A variety of bacteria and fungi are capable of directly infecting the liver. Clinical presentation of bacterial and fungal hepatitides may be indistinguishable from viral hepatitis and thus, individuals presenting with elevated liver-associated enzymes or jaundice with fever should be evaluated with blood cultures and appropriate serology. In contrast to viral hepatitis, it is unusual for liver-associated enzymes to be elevated to more than 10 times the upper limits of normal in bacterial or fungal-related hepatitis. Table 37-8 lists specific infections associated with hepatitis. Bacteria are historically classified by size, shape, staining properties, and biochemical properties, and bacterial infections are classified as Gram-positive
755
Section V. Liver Diseases Due to Infectious Agents
Table 37-8. Etiologic Agents of Bacteria and Fungal Hepatitis Bacteria, Chlamydia, and Rickettsia Gram-positive bacilli Clostridium, Listeria, actinomycosis Gram-positive cocci Staphylococcus, Streptococcus Gram-negative bacilli Salmonella, Shigella, Yersinia, brucellosis, legionellosis, Burkhoderia pseudomallei (meliodosis), bartonellosis Gram-negative cocci Meningococcus, Gonococcus Mycobacteria Tuberculosis, non-tuberculous mycobacteria Spirochetes Treponema pallidum, borelliosis, leptospirosis Rickettsia Rocky Mountain spotted fever, rickettsial pox, typus ehrlichiosis, Q fever Fungi Systemic mycoses
Invasive mycoses
Cryptococcosis, coccidioidomycoses, histoplasmosis, paracoccidioidomycosis, Pneumocystis spp. Candida, Fusarium, Aspergillus, Rhizopus, etc.
or Gram-negative bacilli or cocci. However, this approach does not account for many bacterial agents, including mycobacteria, treponemes, other spirochetes, Mycoplasma, Rickettsia, Chlamydia, and Actinomyces. Fungal infections may also be associated with hepatitis, including the systemic mycoses or the invasive fungal infections that frequently result from iatrogenic complications of medical care.
GRAM-POSITIVE BACILLI Clostridium Clostridium species are generally not related to hepatitic disease, and although jaundice frequently occurs in immunocompromised patients with C. perfringens and C. septicum bacteremia, this is due to the release of a toxin, hemolysin, which results in significant hemolysis.86 Clostridium is a contributor to polymicrobial infections of the liver (abscesses) and biliary tract.
Listeria Listeria monocytogenes infection is uncommon except in neonates, pregnant women, elderly individuals, immunosuppressed transplant recipients, and others with impaired cell-mediated immunity. Listeria primarily causes bacteremia and meningoencephalitis, although it may also cause endocarditis, gastrointestinal disease, and pneumonitis. The bacteria is an important zoonosis, and it is increasingly associated with foodborne outbreaks. It is recovered in up to 5% of stool specimens obtained from healthy adults.87 A major source of L. monocytogenes is contaminated foods, including raw vegetables, raw milk, cheese, and meats, including chicken and beef.88 Listeria infection results in disseminated microabscesses and granulomas in neonates, particularly in the liver and spleen. This complication is rare in adults, and when it does occur, it has a predilection for people with underlying liver disease.89
Actinomycosis Actinomycosis is an indolent infection caused by higher bacterium Actinomyces. These organisms are anaerobic or microaerophilic, Gram-positive, filamentous bacilli that most commonly cause
756
oral–cervical disease, although up to 20% of reported cases involve the abdomen.90,91 Hepatic infection is present in approximately 5% of all cases of actinomycosis91 and 16% of cases of abdominal actinomycosis.92 Actinomycosis spreads to the liver either from a contiguous focus or via hematogenous spread, and in one case was associated with a pancreatic stent.93 Single or multiple abscesses are the usual findings, and the indolent nature often leads to mistaken diagnosis of malignancy.94 Actinomycosis typically presents with fever, weight loss, abdominal pain, and anorexia. Rarely, actinomycosis presents with cholangitis, portal vein occlusion, cholestasis, or extension into the thorax.95–97
GRAM-POSITIVE COCCI Staphylococci and streptococci Staphylococci and streptococcal infections involve the liver either as liver abscesses or alternatively, they may induce parainfectious hepatitis via inflammation and/or toxin release, as noted above.
GRAM-NEGATIVE BACILLI Salmonella Hepatitis is common among individuals with acute salmonella infection, particularly Salmonella typhi.98,99 Presenting symptoms, signs, and laboratory studies may mimic acute viral hepatitis, although a history of high fever (>40°C or 104°F) and travel, relative bradycardia, and a left shift on the white blood cell differential all increase the likelihood of typhoid hepatitis. The liver histology is consistent with a parainfectious etiology.100 Undiagnosed and untreated typhoid hepatitis has a mortality rate of 20%, although mortality is much lower with appropriate antibiotic therapy.
Brucellosis Brucellosis is a worldwide zoonosis that is most prevalent in the Mediterranean basin, the Indian subcontinent, and in parts of Mexico and Central and South America.101 Brucellosis presents as a non-specific illness with fever, night sweats, malaise, anorexia, headache, and back pain. Characteristically, the fever has an undulating pattern that begins 2–4 weeks after inoculation. Depression and somatic complaints predominate, and physical abnormalities are less common, leading to consideration of malingering. Symptoms referable to the gastrointestinal tract are common, although liverassociated enzyme levels are normal or mildly elevated. Brucellus abortus is associated with granulomatous hepatitis that can be indistinguishable from sarcoidosis,102 and B. melitensis may have lymphocytic infiltrates resembling mild viral hepatitis.103 Alternatively, epithelioid granuloma may accompany brucellosis.104 Liver abnormalities usually resolve with appropriate antibacterial therapy.
Legionellosis Legionella species are a common cause of pneumonia. Legionella are classified as Gram-negative bacteria, although their staining characteristics are poor and they may be missed on sputum Gram-stain examination. Although no evidence for direct hepatic involvement has been reported, up to 50% of patients with documented L. pneumophila infection will have elevations in liver-associated enzymes, and up to 5% will have jaundice.105,106
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
Burkhoderia pseudomallei Burkhoderia pseudomallei is the causative agent of meloidosis. B. pseudomallei is a water and soil bacterium predominantly found in South-East Asia, Madagascar, Australia, and regions of Central America. Infection may present as an asymptomatic infection or be confined to localized skin ulcers without systemic illness.107,108 At the other end of the spectrum, abscesses may develop throughout the lungs and organs, including the liver. Hepatic involvement may include microabscesses, focal necrosis, and occasionally granulomatous hepatitis.108 B. pseudomallei, like tuberculosis, has the potential for reactivation from a latent focus, and latency periods have been documented to be up to 29 years.109 The infection is diagnosed primarily by culture of involved organs or secretions.
Bartonellosis and peliosis hepatis Bartonella species are closely related to Brucella based on ribosomal RNA sequence.110 Bartonella species have been associated with a number of clinical entities, including Oroya fever, trench fever, catscratch fever, and peliosis hepatis. There are several species within the genus, including B. bacilliformis, the causative agent of Oroya fever, an acute illness with bacteremia occurring 3–12 weeks after inoculation, and a spectrum of disease ranging from mild illness to a profound, systemic disease.111,112 Jaundice due to hemolysis is frequent in severe cases; however, hepatic dysfunction is frequent and appears to result from microvascular thrombosis and ischemia.112 B. quintana is associated with trench fever in human immunodeficiency virus (HIV)-negative, frequently homeless people.113 Trench fever may have many associated symptoms, including fever, hepatosplenomegaly, and elevated liver-associated enzymes. B. henselae causes cat-scratch disease, and may be localized, or lead to persistent bacteremia, more frequently in HIV-positive and other immunosuppressed individuals. Liver lesions may be identified on imaging studies (Figure 37-4) and histological examination may reveal microabscess formation in both liver and lymphoid tissue. Bacillary angiomatosis is pathologically characterized by neovascular proliferation, typically in the skin and regional lymph nodes of HIVinfected or immunosuppressed individuals.114 Bacillary peliosis (BP) has also been identified in the liver and spleen of HIV-infected and other immunosuppressed individuals.115 Numerous blood-filled cystic structures as large as several millimeters may virtually replace the hepatic parenchyma.116 Non-specific symptoms of fever,
A
B
malaise, and associated skin lesions and/or adenopathy are the usual presenting signs and symptoms. It appears that both B. quintana and B. henselae can cause bacillary angiomatosis, although only B. henselae appears to cause bacillary peliosis. Bartonella organisms can be seen on histologic specimens when stained with the Warthin–Starry silver stain.
GRAM-NEGATIVE COCCI Neisseria species Although unusual, Neisseria meningitides sepsis may resemble severe viral hepatitis.117 More commonly, liver enzyme elevations associated with meningococcemia are related to inflammatory mediators released in response to lipo-oligosaccharide. In contrast, N. gonorrhoeae bacteremia is associated with abnormalities in liverassociated enzymes in virtually all patients.118 Jaundice is uncommon, and perihepatitis (Fitz-Hugh–Curtis syndrome or FHC) is the most common complication of disseminated gonococcal infection. FHC syndrome occurs in up to 25% of women with pelvic inflammatory disease (PID). Patients typically present with right upper quadrant abdominal pain and fever, usually in the setting of PID, although some women will lack pelvic signs and symptoms. Fever, hepatic tenderness, right upper quadrant peritoneal inflammatory signs, and occasionally a friction rub over the liver are observed on physical examination.119 FHC syndrome should be considered in the differential diagnosis of right upper quadrant pain in young, sexually active women, and is commonly mistaken for acute cholecystitis or viral hepatitis. Perihepatitis occurs by direct extension of N. gonorrhoeae or Chlamydia trachomatis from the fallopian tubes to the liver capsule and overlying peritoneum, although some lymphatic or hematogenous spread may occur. The latter routes of dissemination explain the rare cases of FHC syndrome that occur in men. Laparoscopic examination may identify violin-string-like adhesions between the liver capsule and peritoneal wall.119
Mycobacteria Mycobacterium tuberculosis. The most common cause of liverassociated enzyme elevation in patients with tuberculosis results from antituberculous therapy-related hepatotoxicity.120 Direct liver involvement is present in up to 50% of patients with miliary tuberculosis.121 Tuberculomas and tubercular abscesses are uncommon, occur in people with disseminated tuberculosis, and may be diffi-
C
D
Figure 37-4. (A–D) Multiple microabscesses in splenectomized patient with disseminated bartonellosis.
757
Section V. Liver Diseases Due to Infectious Agents
cult to diagnose. In the absence of disseminated disease, liver involvement is unusual. When it occurs, hepatic lesions represent localized granulomas, often with central necrosis. Signs and symptoms commonly include right upper quadrant or generalized abdominal pain, hepatomegaly, fever, and weight loss.121–123 Laboratory studies reveal a cholestatic pattern of liver-associated enzyme elevations, and fewer than 25% of patients have jaundice. The diagnosis of hepatic tuberculosis is often not suspected unless active pulmonary disease is present, as it is in approximately 65% of patients. Hepatic lesions may be hypoechoic or hyperechoic on ultrasound.124 Diagnosis relies upon liver biopsy with histologic demonstration or culture of acid-fast bacilli. Up to 33% of patients will not have bacteria identified on histologic specimens, and culture and polymerase chain reaction (PCR) are helpful in these cases. Non-tuberculous mycobacterium. The most common nontuberculous mycobacteria that affects the liver is the Mycobacterium avium-intracellulare complex (MAC).125,126 MAC dissemination with liver involvement is common in HIV-positive individuals, although this is decreasing in countries where effective antiretroviral therapy is common. Disproportionate elevation of serum alkaline phosphatase (up to 40 times the upper limits of normal) may be seen in up to 5% of patients. The other liver-associated enzymes may be remarkably normal, or minimally elevated.126 Patients have non-specific symptoms and liver histology is relatively normal, suggesting that MAC interferes with enzyme metabolism rather than causing cholestasis or hepatic necrosis. Dissemination of mycobacteria, including M. kansasii, M. genavense, and others has been observed,127,128 usually in people with HIV infection or less commonly in organ transplant recipients, patients receiving chronic steroids, or in children with congenital defects in the interferon-g and IL-12 receptor.129
Spirochete diseases Treponema pallidum. Syphilis may affect the liver in any of its various clinical phases.130 Diagnosis of syphilis relies upon serologic testing and maintaining a high index of suspicion. Intrahepatic epithelioid granulomas may complicate congenital syphilis and may on occasion lead to portal or interstitial fibrosis.131,132 In these cases T. pallidum may be visualized in histologic examination using a silver stain. If granuloma formation is extensive, hepatomegaly may occur with development of portal hypertension and ascites. Intrahepatic calcifications may develop, and these should raise the diagnosis of congenital syphilis. Primary syphilis is not associated with hepatic involvement, although treatment may lead to a Jarisch–Herxheimer reaction with systemic inflammation and perihepatitis.133 Although Jarisch–Herxheimer reactions are more frequent with secondary syphilis, they may occur while treating any stage.134 Liver disease is uncommon in secondary syphilis, although mild elevations of serum alkaline phosphatase may occur. Rarely, jaundice and right upper quadrant pain are present, and liver biopsy demonstrates lymphocytic infiltration with organisms identified by silver stain. Late syphilis is a slowly progressive inflammatory disease affecting any organ, including the liver. Clinical illness often occurs years to decades after initial infection.135 Liver disease is related to syphilitic “gummas,” indolent, non-specific, granulomatous-like
758
lesions that vary in size from microscopic defects to large tumor-like masses. The center of large gummas may become necrotic. Gummatous hepatitis may cause fever, epigastric pain, and tenderness, and lead to cirrhosis. Gummas respond rapidly to penicillin therapy, and are quite rare in the antibiotic era. T. pallidum may also cause liver disease through vasculitis or endarteritis. Borreliosis. Borrelia burgdorferi causes Lyme disease, a tick-borne illness consisting of several clinical phases. The first stage usually begins in the summer (stage 1), and presents as an expanding skin lesion called erythema migrans, occurring at the site of a tick bite. Within days to weeks, the spirochete may spread to other sites (stage 2), including skin, heart, joint, and central nervous system.136,137 During stage 2, recurrent erythema migrans may develop, and up to 40% of patients have elevated liver-associated enzymes.138 Stage 2 borrelliosis may include jaundice and tender hepatosplenomegaly.139 Following long periods of latent infection, B. burgdorferi may cause disease in the joints, heart, or central nervous system (stage 3), although liver involvement is unusual. All three stages are usually curable with appropriate antibiotic therapy (summarized by Steere136). Leptospirosis. Leptospirosis is globally distributed and is probably the most common zoonosis.140 Leptospirosis is caused by pathogenic spirochetes of the Leptospira genus, and the disease is thought to be greatly under-reported.140 Human infection results from direct contact with infected animal urine or tissues, or more commonly by indirect exposure to the organisms present in soil and water. The wide spectrum of illness ranges from asymptomatic infection to fatal, multisystem disease. In approximately 90% of people with leptospirosis, the illness is a self-limited febrile illness. In the remaining 10%, any combination of renal failure, liver failure, and pneumonitis may occur.141,142 Weil reported this biphasic illness (consistent with leptospirosis) in 1886, and it is now termed Weil’s disease. This biphasic illness has the initial stage of viral-like illness (associated with bacteremia) after an incubation period of 5–14 days. Five to 7 days later, the fever temporarily declines but is followed in 4–30 days by an immune phase with severe symptoms,141 including the abrupt onset of high fever, headache, chills, rigors, myalgias, conjunctival suffusion without discharge, abdominal pain, anorexia, nausea, and vomiting. Diarrhea, cough, pharyngitis, and a pretibial maculopapular rash may also occur. Conjunctival suffusion and myalgias are the most common findings.143 Forty-six to 85% of patients develop jaundice, frequently with renal failure, headache (with aseptic meningitis), photophobia, and hepatosplenomegaly.143,144 The most significant complication of the immune phase is hepatic and renal failure, and patients developing renal disease have mortality rates as high as 40%.143,144 The rise in liver-associated enzymes is disproportional to the mild pathological findings, with serum bilirubin levels relatively higher than transaminase elevations.143 Liver biopsy reveals cholestasis, and hepatocyte degenerative changes are seen in the absence of hepatocellular necrosis. Erythrophagocytosis and mononuclear cell infiltrates may also be present.145 Leptospira can be recovered from blood and cerebrospinal fluid during the acute phase, and urine sediment abnormalities and organisms can be recovered 5–7 days after the onset of symptoms.143
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
There are insufficient data on which to develop treatment guidelines;146 however, severe disease is usually treated with intravenous penicillin or ceftriaxone, and less severe disease with oral doxycycline. Similar to syphilis and Lyme disease, Jarisch–Herxheimer reactions may occur with therapy. Prophylactic treatment for individuals with unavoidable exposure to leptospirosis in endemic environments consists of using doxycycline.147
RICKETTSIA SPOTTED FEVERS Spotted fevers and other rickettsial infections are caused by a group of 12 Rickettsia species transmitted by ticks, fleas, and mites.148 Rickettsia rickettsii is the etiologic agent of Rocky Mountain spotted fever (RMSF) and is transmitted by ticks. RMSF is classically recognized as the triad of fever, headache, and rash; however, only about half of patients with RMSF have the rash develop in the first 3 days of illness. Gastrointestinal symptoms (nausea, vomiting, abdominal pain, diarrhea, jaundice, hepatosplenomegaly) may occur before the rash, and up to 60% of patients have elevated serum aspartic aminotransferase levels.149,150 Diagnosis is usually based on clinical suspicion, as serologic approaches to diagnosis are negative early in infection and may not be reliable. Detection of antigen in skin biopsies by immunofluorescence is a sensitive approach to diagnosis, although this is not widely available.151 PCR for R. rickettsii is insensitive early in disease.152
TYPHUS R. prowazekii (epidemic typhus) is transmitted by lice, and causes devastating epidemics associated with conditions of poor personal hygiene such as war, poverty, natural disasters including floods and earthquakes, and homelessness.153 In epidemic areas the disease is readily diagnosed, and consists of fever, headache, and myalgia, usually with a rash. In non-endemic areas, the diagnosis may be confused with typhoid fever, hemorrhagic viruses, syphilis, measles, and meningococcemia.154 R. typhi is found worldwide, and is most prevalent in tropical and subtropical regions where the vectors (fleas) and reservoirs (rats) are most common.155,156 Scrub typhus (caused by Orientia tsutsugamuchi) is transmitted by chigger bites and is found in a triangular region of the world bordered by Japan, Australia, and India.157,158 Epidemic, murine, and scrub typhus may present with severe multisystem illness, commonly with neurologic signs, pneumonia, fever, and rash. Jaundice and elevated liver-associated enzymes are present in up to 24% of patients with murine typhus and may suggest viral hepatitis.157 Multiple organ dysfunction, hemorrhage, and elevated liver-associated enzymes occur in severe cases.159,160 Diagnosis of typhus relies on a fourfold rise in convalescent antibody titers, and PCR may be useful for epidemic and murine typhus, although the test is not usually available in regions where it is needed.161,162 The diagnostic sensitivity and specificity of PCR have not been determined for murine typhus.162 Treatment with doxycycline is usually effective, although relapse may occur.163 Due to the retrospective nature of diagnostic approaches, therapy should be initiated based on clinical suspicion.
ERLICHIOSIS Members of the genera Erhlichia, Anaplasma, and Neorickettsia within the family Anaplasmataceae cause human illness referred to as ehrlichiosis, although the clinical manifestations and etiologic agents are distinct.148 Ehrlichiosis are zoonoses transmitted by ticks, and are generally classified as “human monocytic erhlichiosis” (HME) caused by Ehrlichia chaffeensis164 or “human granulocytic anaplasmosis” (HGA; previously human granulocytic ehrlichiosis), caused by Anaplasma phagocytophilum.165 Disease in immunocompetent people ranges from mild systemic illness to severe multisystem illness;165,166 however, in immunocompromised patients, E. chaffeensis commonly causes fatal, multisystem disease.166,167 Both illnesses typically start with fever, headache, myalgia, and malaise, and subsequently progress to include gastrointestinal symptoms including nausea, anorexia, vomiting, and diarrhea. HME is more frequently complicated by abdominal pain, and elevations in liverassociated enzymes occur in up to 86%.168,169 Hepatosplenomegaly is common, and a macular rash occurs in fewer than 50% of HME and rarely in HGA. Neutropenia is common in HGA and HME, while leukopenia and lymphopenia are also common in HME. Diagnosis is based on epidemiology and clinical and laboratory findings. In patients with fever, leukopenia, thrombocytopenia, elevated liver-associated enzymes, and a history of a tick bite in endemic regions during the late spring to mid-summer, HME and HGA should be suspected. HME cannot be distinguished from RMSF clinically, although a rash is less common and leukopenia more common in HME. E. chaffeensis morulae in circulating mononuclear cells visualized by Wright-stained blood smears are detected in fewer than 7% of patients with HME,170 and diagnosis is made by acute and convalescent serologic testing.171 PCR testing for E. chaffeensis (HME) appears promising.172 Morulae in polymorphonuclear cells occur in up to 80% of patients with HGA, and this is diagnostic.173 While culture may be successful, PCR detection of A. phagocytophilum may also be useful (55% and 86% sensitive, >95% specific).174,175 Serology provides a retrospective diagnosis. Doxycycline or tetracycline therapy should be initiated based on clinical suspicion, and delay may result in progression to multiorgan failure and significant mortality. Rifampin is active in vitro, and has been used in pregnant women, although there are insufficient data on therapies other than tetracyclines. Up to 21% of patients with HGA have serologic evidence of Borrelia burgdorferi or Babesia microti infection, as these pathogens and A. phagocytophilum are transmitted by Ixodes spp. ticks.176 It is unclear if concurrent infection alters the severity, incubation period, duration of illness, or likelihood of sequelae.177
COXIELLA BURNETII (Q FEVER) Q fever is a worldwide zoonosis causing a systemic febrile illness. The etiologic agent is Coxiella burnetii,178,179 which is found in cattle, sheep, and goats. Infected animals shed organisms into urine, feces, milk, and placenta178,179 and humans become infected by inhaling the rickettsia. The most common illness is a self-limited febrile illness lasting 2–14 days, although pneumonia, endocarditis, hepatitis, osteomyelitis, and neurological syndromes may occur. Hepatitis is the most common manifestation of Q fever in France and approximately 60% of US cases manifest as hepatitis.179–182
759
Section V. Liver Diseases Due to Infectious Agents
Jaundice occurs in up to 30% of infected patients, and alkaline phosphatase is disproportionately elevated compared to other liverassociated enzymes. Histologically, a characteristic “ring granuloma” consisting of fat vacuoles surrounded by a ring of fibrinoid necrosis, histiocytes, and lymphocytes may be seen. This lesion may also be seen in several other diseases, including cytomegalovirus and lymphoma.183,184
FUNGAL HEPATITIS Fungal hepatitis may occur with any of the disseminated mycoses in immunocompetent hosts, although the frequency of liver involvement and diversity of organisms increase in immunocompromised patients. Cell-mediated immune defects, particularly diseases and/or drugs affecting CD4+ T-lymphocyte function, are the major immune factor predisposing to disseminated mycoses. In contrast, invasive mycoses rarely involve the liver in the absence of severe immunocompromise, and typically require defects in neutrophil function or number, or defects in the natural barriers to infection (skin and gastrointestinal mucosa).
HISTOPLASMOSIS Histoplasmosis is acquired by inhalation of Histoplasma capsulatum, one of the commonest causes of infection in the midwest and southeast USA. Although histoplasmosis is distributed in various regions of the world, the most endemic regions are the Ohio and Mississippi river valleys. H. capsulatum is a soil-based fungus that prefers growth in moist soil containing guano. Bats carry and shed the fungus, whereas birds do not.185 The predilection for H. capsulatum growth in soils containing high nitrogen content explains the high concentration of organisms found in environmental sources associated with chickens, and other birds. H. capsulatum resides in macrophages and dendritic cells186 and is distributed throughout the reticuloendothelial system. Most individuals who acquire histoplasmosis do not develop clinical disease, and retrospective diagnosis is made serologically.187 Hepatic calcifications are common in endemic areas, and these occasionally contain H. capsulatum. Immunecompetent people with acute histoplasmosis may develop a mild, influenza-like illness, although occasionally severe pulmonary disease with systemic illness occurs with transient elevations in liverassociated enzymes, particularly alkaline phosphatase. Severe disease may result when the inhaled inoculum is large. Progressive disseminated histoplasmosis (PDH) occurs in immune-competent individuals; however, it is more common in people with cellmediated immune defects, especially HIV infection. PDH may cause hepatosplenomegaly, elevated liver-associated enzymes, and rarely isolated hepatic lesions.188 Liver histology may reveal portal lymphocytic and histiocytic infiltrates or granulomas. Diagnosis may be made by the detection of yeast forms in liver tissue; however, non-invasive diagnosis is possible using blood cultures and serum and urine histoplasma antigen detection enzyme-linked immunosorbent assay methods.187 Antibodies are common in people from endemic regions, limiting the diagnostic utility of serology. Treatment usually consists of a combination of amphotericin B and oral azoles, and practice guidelines for histoplasmosis therapy have been published.189
760
CRYPTOCOCCOSIS Cryptococcus neoformans is acquired by inhalation, and frequently causes subclinical infection in healthy people. The extent of disease is primarily determined by the immune status of the host, although virulence factors have been identified190 and the inoculum size may influence the outcome. Disease usually presents as pneumonia or meningitis,190,191 although C. neoformans infects any organ. Hepatitis, though rarely severe, is described in both immune-competent and compromised hosts and histology reveals a lymphocytic infiltrate or granulomatous hepatitis.191 Diagnosis can be made by identification of yeast in liver biopsy materials and cultures; however, non-invasive diagnosis can be made by culture or detection of serum or cerebrospinal fluid cryptococcal antigen using latex agglutination.190,191 Treatment varies based on immune status and must include attempts to restore host immunity. Treatment strategies may utilize a combination of amphotericin (or liposomal amphotericin) frequently combined with flucytosine, frequently followed by oral azole therapy.192,193 Specific recommendations for patients with HIV infection have been published.194 Relapse of cryptococcal disease is common in HIV-positive people in whom HIV is not well controlled with antiretroviral therapy. Drug resistance testing should be considered in individuals not responding to therapy or with severe disease.195
COCCIDIOIDO- AND PARACOCCIDIOIDOMYCOSES Coccidiodes immitis infection occurs in the south-western USA and the San Joaquin valley of California. Paracoccidiodes brasiliensis is found in South and Central America. Like H. capsulatum, these two organisms are dimorphic fungi that commonly cause subclinical infection. Most disease occurs in immunocompromised hosts, and consists of pneumonia, fever, and dissemination via the reticuloendothelial system. Elevated liver-associated enzymes are commonly found in disseminated infection, and hepatic granuloma formation may occur.196 While hepatomegaly is common, jaundice is infrequent. Diagnosis of coccidioidomycosis requires a variety of serologic tests to detect antibodies197 or C. immitis antigen in blood,198 and PCR-based methods appear promising.199 Paracoccidioidomycosis diagnosis is made by culture and direct examination of histologic specimens, although PCR is under development.200
PNEUMOCYSTIS SPECIES Pneumocystis species are unicellular fungi acquired by most humans during early childhood.201 Infection is controlled by host immunity, and disease is unusual except in cases of impaired cellular immunity. HIV infection is the commonest cause of reactivation of Pneumocystis, although malnutrition and drug-related immunosuppression also result in disease. Although interstitial pneumonia is the commonest disease manifestation, extrapulmonary disease frequently occurs in HIV-infected people. Cholestatic jaundice is common in patients with HIV202–204 and less frequently in transplant recipients.107,205 Diagnosis is difficult, and relies upon identification of organisms in frothy, eosinophilic honeycombed materials found in affected tissues.206 Treatment involves reversing immune defects and trimethoprim-sulfamethoxazole, although alternative drugs are available when intolerance to trimethoprim-sulfamethoxazole occurs.204
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
INVASIVE FUNGAL INFECTIONS Fungemia with Candida, Aspergillus, Fusarium, Zygomycetales and others is increasingly common. These organisms are normal flora of the skin, respiratory, gastrointestinal mucosa, or environment. Candida species gain access to the bloodstream by contamination of vascular catheters and devices, or by the loss of integrity of gastrointestinal or upper respiratory mucosa (frequently the result of cytotoxic therapy), or via abdominal perforation and surgery. Candidemia frequently results in liver infection; however, clinical disease is rare among patients with normal numbers of functional neutrophils. In contrast, patients with impaired neutrophil function or severe neutropenia develop microabscesses in the liver that may progress to frank hepatic candidiasis207,208 requiring prolonged therapy. Aspergillus, Fusarium, and Zygomycetales and other opportunistic fungi and molds may also reach the bloodstream by contaminated vascular devices or disrupted mucosal surfaces. These infections have a propensity to invade terminal vessels and tissues, causing either localized or widespread tissue and organ dysfunction.209–211
REFERENCES 1. Ochsner A, Debakey M, Murray S. Pyogenic abscess of the liver. Am J Surg 1938; XL:292–319. 2. Perez JAA, Gonzalez JJ, Baldonedo RF, et al. Clinical course, treatment, and multivariate analysis of risks factors for pyogenic abscess. Am J Surg 2001; 181:177–186. 3. Wong W, Wong BC-Y, Hui CK, et al. Pyogenic liver abscess: retrospective analysis of 80 cases over a 10-year period. J Gastroenterol Hepatol 2002; 17:1001–1007. 4. Mohsen AH, Green ST, Read RC, Mckendrick MW. Liver abscess in adults: ten years experience in a UK centre. Q J Med 2001; 95:797–802. 5. NG FH, Wong WM, Wong BCY, et al. Sequential intravenous/oral antibiotic vs. continuous intravenous antibiotic in the treatment of pyogenic liver abscess. Aliment Pharmacol Ther 2003; 16:1083–1090. 6. Lam Y-H, Wong SK-H, Lee DW-H, et al. ERCP and pyogenic abscess. Gastrointest Endosc 1999; 50:1–9. 7. Barrio J, Cosme A, Ojeda E, et al. Pyogenic liver abscesses of bacterial origin. A study of 45 cases. Rev Esp Enferm Dig 2000; 92:236–239. 8. Seet RK, Rocky DC. Pyogenic liver abscess: changes in etiology, management, and outcome. Medicine 1996; 75:99–113S. 9. Huang C-J, Pitt HA, Lipsett PA, et al. Pyogenic hepatic abscess: changing trends over 42 years. Ann Surg 1996; 5:600–609. 10. Johannsen EC, Sifri CD, Madoff LC. Pyogenic liver abscesses. Infect Dis Clin North Am 2000; 14:547–563. 11. Kubo S, Kinoshita H, Hirohashi K, et al. Risks factors for and clinical findings of liver abscess after biliary-intestinal anastomosis. Hepato-Gastroenterol 1999; 46:116–120. 12. Stange BJ, Glanemann M, Nuessler N, et al. Hepatic artery thrombosis after adult liver transplantation. Liver Transplant 2003; 9:612–620. 13. Gunsar F, Rolando N, Pastacaldi S, et al. Late hepatic artery thrombosis after orthotopic liver transplantation. Liver Transplant 2003; 9:605–611. 14. Chou F-F, Shen-Chen S-M, Chen M-C. Single and multiple pyogenic liver abscesses: clinical course, etiology, and results of treatment. World J Surg 1997; 21:384–389. 15. Weng W, Lee W-J, Wei P-L, et al. Laparoscopic drainage of pyogenic liver abscesses. Surg Today 2004; 34:323–325.
16. Yu SCH, Ho SSM, Lau WY, et al. Treatment of pyogenic liver abscess: prospective randomized comparison of catheter drainage and needle aspiration. Hepatology 2004; 39:932–938. 17. Sugiyama M, Atomi Y. Pyogenic hepatic abscess with biliary communication. Am J Surg 2001; 183:205–208. 18. Cheng HP, Chang FY, Fung CP, Sui LK. Klebsiella pneumoniae liver abscess in Taiwan is not caused by a clonal spread strain. Microbiol Immunol Infect 2002; 35:85–88. 19. Chemaly RF, Hall GS, Keys TF, Procop GW. Microbiology of liver abscesses and the predictive value of abscess Gram stain and associated blood cultures. Diagn Microbiol Infect Dis 2003; 46:245–248. 20. Furugaki K, Chijiwa K, Ogawa Y, Tanaka M. Pyogenic liver abscess after hepatobiliary and pancreatic surgery. Int Surg 2001; 86:67–71. 21. De Palma GD, Pezzullo A, Rega M, et al. Unilateral placement of metallic stents for malignant hilar obstruction: a prospective study. Gastrointest Endosc 2003; 58:50–53. 22. Dull JS, Topa L, Balgha V, Pap A. Non-surgical treatment of biliary liver abscesses: efficacy of endoscopic drainage and local antibiotic lavage with nasobiliary catheter. Gastrointest Endosc 2000; 51:1–7. 23. Usdan LS, Massinople C. Multiple pyogenic liver abscesses associated with occult appendicitis and possible Crohn’s disease. Tennessee Med 2002; 95:463–464. 24. Drnovsek V, Fontanez-Garcia D, Waskabayasbi MN. Gastrointestinal case of the day. RadioGraphics 1999; 19:820–822. 25. Hsieh C-H, Hsu Y-P. Early-onset liver abscess after blunt liver trauma: report of a case. Surg Today 2002; 33:392–394. 26. Hsieh CH. Comparison of hepatic abscess after operative and nonoperative management of isolated blunt liver trauma. Int Surg 2002; 87:178–184. 27. Yeh T-S, Jan Y-Y, Jeng L-B, et al. Hepatocelluar carcinoma presenting as pyogenic liver abscess: characteristics, diagnosis, and management. Hepatocell Carcinoma Liver Abscess 1998; 26:1224–1226. 28. Giorgio A, Tarantino L, De Stefano G, et al. Complications after interventional sonography of focal liver lesions: a 22-year singlecenter experience. Am Inst Ultrasound Med 2003; 22:193–205. 29. de Baere T, Risse O, Kuoch V, et al. Adverse events during radiofrequency treatment of 582 hepatic tumors. AJR Am J Roentgenol 2003; 181:695–700. 30. Huang SF, Ko CW, Chang CS, Chen GH. Liver abscess formation after transarterial chemoembolization for malignant hepatic tumor. Hepatogastroenterology 2003; 50:1115–1118. 31. Shibata T, Yamamoto Y, Yamamoto N, et al. Cholangitis and liver abscess after percutaneous ablation therapy for liver tumors: incidence and risk factors. J Vasc Interv Radiol 2003; 14:1535–1542. 32. Molle I, Thulstrup AM, Vilstrup H, Sorensen HT. Increased risk and case fatality rate of pyogenic liver abscess in patients with liver cirrhosis: a nationwide study in Denmark. Gut 2001; 48:260–263. 33. CDC. First Report of AIDS. MMWR 2001; 50:429–439. 34. Fung C-P, Chang F-Y, Lee S-C, et al. A global emerging disease of Klebsiella pneumoniae liver abscess: is stereotype K1 an important factor for complicated endophthalmitis? Gut 50, 420–425. 2002. 35. Lee C-C, Poon S-K, Chen G-H. Spontaneous gas-forming liver abscess caused by salmonella within hepatocellular carcinoma. Digest Dis Sci 2002; 47:586–589. 36. Hopfner M, Nitsche R, Rohr D, et al. Yersinia enterocolitica infection with multiple liver abscesses uncovering a primary hemochromatosis. Scand J Gastroenterol 2001; 2:220–224. 37. Wang S-C, Hsin K-H, Chern JPS, et al. Severe bacterial infection in transfusion-dependent patients with thalassemia major. Clin Infect Dis 2003; 37:984–988.
761
Section V. Liver Diseases Due to Infectious Agents
38. Bamberger D. Outcome of medical treatment of bacterial abscess without therapeutic drainage: review of cases reported in the literature. Clin Infect Dis 1996; 23:592–603. 39. Chang S-C, Fang C-T, Hsueh P-R, et al. Klebsiella pneumoniae isolates causing liver abscess in Taiwan. Diagn Microbiol Infect Dis 2000; 37:279–284. 40. Lee K-T, Wong S-R, Sheen P-C. Pyogenic liver abscess: an audit of 10 years’ experience and analysis of risk factors. Digest Surg 2001; 18:459–466. 41. Clarke MG, Kennedy NJ, Kennedy K. Serious consequences of a sore throat. Ann R Coll Surg Engl 2003; 85:242–244. 42. Athavale NY, Leitch DG, Cowling P. Liver abscesses due to Fusobacterium spp that mimick malignant metastatic liver disease. Eur J Clin Microbiol Infect Dis 2002; 21:884–886. 43. Hernandez JL, Ramos C. Pyogenic hepatic abscess: clues for diagnosis in the emergency room. Clin Microbiol Infect 2001; 7:567. 44. Ralls PW. Inflammatory disease of the liver. Clin Liver Dis 2002; 6:1–18. 45. Balci NC, Sirvanci M. MR imaging of infective liver lesions. MRI Clin North Am 2002; 1:121–135. 46. Krige JEJ, Beckinghman IJ. ABC of diseases of liver, pancreas, and biliary system liver abscesses and hydatid disease. Br Med J 2001; 322:537–540. 47. Catalano O, Sandomenico F, Raco MM, Siani A. Low mechanical index contrast-enhanced sonographic findings of pyogenic hepatic abscess. AJR 2004; 182:447–450. 48. Chan JHM, Tsui EYK, Luk SH, et al. Diffusion-weighted MR imaging of the liver: distinguishing hepatic abscess from cystic or necrotic tumor. Abdom Imaging 2000; 26:161–165. 49. Ryan RS, Hashimi HA, Lee MJ. Hepatic abscesses in elderly patients mimicking metastatic disease. Irish J Med Sci 2004; 170:251–253. 50. Athavale NV, Leitch DG, Cowling P. Liver abscesses due to Fusobacterium spp that mimick malignant metastatic liver disease. Eur J Gastroenterol Hepatol 2002; 21:884–886. 51. Ahsan T, Jehangir MU, Mahmood T, et al. Amoebic versus pyogenic liver abscess. J Pak Med Assoc 2002; 52:497–501. 52. Cahill M, Chang B, Murray A. Bilateral endogenous bacterial endophthalmitis associated with pyogenic hepatic abscess. Br J Ophthalmol 2000; 84:1432. 53. Tan Y-M, Chee S-P, Soo K-C, Chow P. Ocular manifestations and complications of pyogenic liver abscess. World J Surg 2004; 28:28–42. 54. Tan Y-M, Chong C-K, Chow PKH. Pyogenic liver abscess complicated by endogenous endophthalmitis. J Surg 2001; 71:744–746. 55. Geerling S, Hoepelman AIM. Immune dysfunction in patients with diabetes mellitus (DM). Immunol Med Microbiol 1999; 26:259–265. 56. Moutschen MP, Scheen AJ, Lefebvre PJ. Impaired immune responses in diabetes mellitus: analysis of the factors and mechanisms involved. Relevance to the increased susceptibility of diabetic patients to specific infections. Diabetes Metab 1992; 18:187–201. 57. Almuneef M, Khenaizan SA, Ajaji SA, Al-Anazi A. Pyogenic liver abscess and Papillon–Lefevre syndrome: not a rare association. Pediatrics 2003; 111:85–88. 58. Anuradha S, Agarwal S, Singh N, et al. An unusual presentation of the Papillon–Lefevre syndrome as recurrent liver abscesses. J Assoc Phys India 2002; 50:974–976. 59. Ogata H, Tsuji H, Hizawa K, et al. Multilocular pyogenic hepatic abscess complicating Ascaris lumbricoides infestation. Int Med 2000; 39:228–230. 60. Khuroo MS, Zargar SA, Mahajan R. Hepatobiliary and pancreatic ascariasis in India. Lancet 1990; 335:1503–1506. 61. Warshauer DM. Image of the month. Am Gastroenterol Assoc 2002; 123:666.
762
62. Ala A, Safar-Aly H, Millar A. Metallic cough and pyogenic liver abscess. Eur J Gastroenterol Hepatol 2001; 13:967–969. 63. Yamada T, Murakami K, Tsuchida K, et al. Ascending cholangitis as a cause of pyogenic liver abscesses complicated by a gastric submucosal abscess and fistula. J Clin Gastroenterol 2000; 30:317–320. 64. Ukikusa M, Inomoto T, Kitai T, et al. Pneumoperitoneum following the spontaneous rupture of a gas-containing pyogenic liver abscess: report of a case. Surg Today 2001; 31:76–79. 65. Kim MD, Kim H, Kang SW, Jeong BG. Nontraumatic hepatic artery pseudoaneurysm associated with acute leukemia: a possible complication of pyogenic liver abscess. Abdom Imaging 2002; 27:458–460. 66. Solomkin JS, Mazuski JE, Baron EJ, et al. Guidelines for the selection of anti-infective agents for complicated intra-abdominal infections. Clin Infect Dis 2003; 37:997–1005. 67. Sharara AI, Rockey DC. Pyogenic liver abscess. Curr Treat Options Gastroenterol 2002; 5:437–442. 68. Martinek A, Hrabovsky V, Klvana P. Puncture aspiration using ultrasound guidance in the treatment of abscesses and other purulent foci in the liver, pancreas and kidneys. Vnitr Lek 2003; 49:630–636. 69. Franson TR, Hierhozer WJ Jr, LaBrecque DR. Frequency and characteristics of hyperbilirubinemia associated with bacteremia. Rev Infect Dis 1985; 7:1–9. 70. Sikuler E, Guetta V, Keynan A, et al. Abnormalities in bilirubin and liver enzymes levels in adult patients with bacteremia. A prospective study. Arch Intern Med 1989; 149:2246–2248. 71. Franson TR, LaBrecque DR, Buggy BP, et al. Serial bilirubin determinations as a prognostic marker in clinical infections. Am J Med Sci 1989; 297:149–152. 72. Quale JM, Mandel LJ, Bergasa NV, Straus EW. Clinical significance and pathogenesis of hyperbilirubinemia associated with Staphylococcus aureus septicemia. Am J Med 1988; 85:618. 73. Trauner M, Fickert P, Strauber RE. Inflammation-induced cholestasis. J Gastro Hepatol 1999; 14:946–959. 74. Mosely RH. Sepsis and cholestasis. Clin Liver Dis 2004; 8:83–94. 75. Nolan JP. Endotoxin, reticuloendothelial function, and liver injury. Hepatology 1981; 1:458–465. 76. Bird GLA, Sheron N, Goka AKL, Alexander RS. Increased plasma tumor necrosis factor in severe alcoholic hepatitis. Ann Intern Med 1990; 112:917–920. 77. Latham PS, Menkes E, Phillips MJ, Jeejeebhoy KN. Hyperalimentation-associated jaundice: an example of a serum factor inducing cholestasis in rats. Am J Clin Nutr 1985; 41:61–65. 78. Freund HR, Muggia-Sullam M, LaFrance R, et al. A possible beneficial effect of metronidazole in reducing TPN-associated liver function derangements. J Surg Res 1985; 38:356–363. 79. Hughes D, Stapleton JT. Postoperative toxic shock syndrome. Iowa Med 1991; 81:55–58. 80. Ayatunde AA, Debrah SA. Portal vein thrombosis complicating appendicitis. W Africa J Med 2004; 23:332–335. 81. Wireko M, Berry PA, Brennan J, Aga R. Unrecognized pylephlebitis causing life-threatening septic shock: a case report. World J Gastroenterol 2005; 11:614–615. 82. Pitcher R, McKenzie C. Simultaneous ultrasound identification of acute appendicitis, septic thrombophlebitis of the portal vein and pyogenic liver abscess. S Afr Med J 2003; 93: 426–428. 83. Singh P, Yadav N, Visvalingeam V, et al. Pylephlebitis – diagnosis and management. Am J Gastroenterol 2001; 96:1312–1313. 84. Balthazar EJ, Gollapudi P. Septic thrombophlebitis of the mesenteric and portal veins: CT imaging. J Comput Assist Tomogr 2000; 24:755–760. 85. Pelsang RE, Johlin F, Dhadha R, et al. Management of suppurative pylephlebitis by percutaneous drainage: placing a
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
86.
87.
88. 89.
90. 91.
92. 93.
94.
95.
96.
97.
98.
99. 100.
101.
102. 103. 104.
105. 106.
107.
108. 109.
110.
drainage catheter into the portal vein. Am J Gastroenterol 2001; 96:3192–3194. Larson CM, Bubrick MP, Jacobs DM, et al. Malignancy, mortality and medicosurgical management of Clostridium septicum infection. Surgery 1995; 118:592–598. Schuchat A, Deaver K, Hayes PS, et al. Gastrointestinal carriage of Listeria monocytogenes in household contacts of patients with listeriosis. J Infect Dis 1993; 167:1262. Farber JM, Peterkin PI. Listeria monocytogenes, a food-borne pathogen. Microbiol Rev 1991; 55:476–511. Yu VL, Miller WP, Wing EJ, et al. Disseminated listeriosis presenting as acute hepatitis. Case reports and review of hepatic involvement in listeriosis. Am J Med 1982; 73:773–777. Brown J. Human actinomycosis: a study of 181 subjects. Hum Pathol 1973; 4:319–330. Bennhoff D. Actinomycosis: diagnostic and therapeutic considerations and a review of 32 cases. Laryngoscope 1984; 94:1198–1217. Putman H, Dockerty M, Waugh JM. Abdominal actinomycosis: an analysis of 122 cases. Surgery 1950; 28:781–800. Harsch IA, Benninger J, Niedobitek G, et al. Abdominal actinomycosis: complications of endoscopic stent in chronic pancreatitis? Endoscopy 2001; 33:1065–1069. Sharma M, Briski LE, Khatib R. Hepatic actinomycosis: an overview of salient features and outcome of therapy. Scand J Infect Dis 2002; 34:386–391. Kasano Y, Tanimura H, Yamaue H, et al. Hepatic actinomycosis infiltrating the diaphragm and right lung. Am J Gastroenterol 1996; 91:2418–2420. Ruutu P, Pentikainen P, Larinkari U, Lempinen M. Hepatic actinomycosis presenting as repeated cholestatic reactions. Scand J Infect Dis 1982; 14:235–238. Ubeda B, Vilana R, Bianchi L, Pujol T. Primary hepatic actinomycosis: association with portal vein thrombosis. AJR Am J Roentgenol 1995; 164:231–232. El-Newihi HM, Alamy ME, Reynolds TB. Salmonella hepatitis: analysis of 27 cases and comparison with acute viral hepatitis. Hepatology 1996; 24:516. Osler W. Hepatic complications of typhoid fever. Johns Hopkins Hosp Rev 1899; 8:373–387. Calva JJ, Ruiz-Palacios GM. Salmonella hepatitis: detection of salmonella antigens in the liver of patients with typhoid fever. J Infect Dis 1986; 154:373–374. Corbel MJ. Brucella: epidemiology and prevalence worldwide. In: Young EJ, Corbel MJ, eds. Brucellosis: clinical and laboratory aspects. Boca Raton, FL: CRC Press; 1989:76–80. Spink WW, Hoffbauer FW, Walker WW. Histopathology of the liver in human brucellosis. J Lab Clin Med 1949; 34:40–58. Young EJ. Brucella melitensis hepatitis: the absence of granulomas. Ann Intern Med 1979; 91:414–415. Young EJ. Brucellosis. In: Connor DH, Chandler FW, Manz HJ, eds. Pathology of infectious diseases. Stamford, CT: Appleton & Lange; 1997:447–451. Cunha BA. Clinical features of legionnaires’ disease. Semin Respir Infect 1998; 13:116–127. Verneau A, Beaugrand M, Dournon E, et al. Severe jaundice in legionnaires’ disease: a case with early hepatic biopsy. Gastroenterol Clin Biol 1987; 11:254–257. Eickhoff TC, Olin DB, Anderson RJ, Schafer LA. Current problems and approaches to diagnosis of infection in renal transplant recipients. Transplant Proc 1972; 4:693–697. Howe C, Sampath A, Sportnitz M. The pseudomallei group: a review. J Infect Dis 1971; 124:598–606. Chodimella U, Hoppes WL, Whalen S, et al. Septicemia and suppuration in a Vietnam veteran. Hosp Practice 1997; 21:219–221. Relman DA, Lepp PW, Sadler KN, Schmidt TM. Phylogenetic relationships among the agent of bacillary angiomatosis,
111.
112.
113.
114.
115.
116.
117. 118. 119.
120. 121. 122.
123. 124.
125.
126. 127.
128.
129.
130. 131. 132. 133.
134.
Bartonella bacciliformis, and other alpha-proteobacteria. Mol Microbiol 1992; 6:1801–1807. Kosek M, Lavarello R, Gilman RH, et al. Natural history of infection with Bartonella bacilliformis in a nonendemic population. J Infect Dis 2000; 182:865–872. Maguina C, Garcia PJ, Gotuzzo E, et al. Bartonellosis (Carrion’s disease) in the modern era. Clin Infect Dis 2001; 33:772–779. Jackson LA, Spach DH. Emergence of Bartonella quintana infection among homeless persons. Emerg Infect Dis 2005; 2:141–144. Cockerell CJ, Webster GF, Whitlow MA, Friedman-Kien AE. Epithelioid angiomatosis: a distinct vascular disorder in patients with the acquired immunodeficiency syndrome or AIDS-related complex. Lancet 1987; 2:6544–6546. Leong SS, Cazen RA, Yu GSM, et al. Abdominal visceral peliosis associated with bacillary angiomatosis: ultrastructural evidence of endothelial cell destruction by bacilli. Arch Pathol Lab Med 1992; 116:866–871. Perkocha LA, Geaghan SM, Yen TS, et al. Clinical and pathological features of bacillary peliosis hepatitis in association with human immunodeficiency virus infection. N Engl J Med 1990; 323:1581–1586. Cowan GO. Meningococcal hepatitis: a case report. J Trop Med Hyg 1987; 90:95–96. Litt IF, Cohen MI. Perihepatitis associated with salpingitis in adolescents. JAMA 1978; 240:1253. Peter NG, Clark LR, Jaeger JR. Fitz–Hugh–Curtis syndrome: a diagnosis to consider in women with right upper quadrant pain. Cleveland Clin J Med 2004; 71:233–239. Stuart RL, Grayson ML. Health care workers and tuberculosis. Aust NZ J Med 2000; 30:367–372. Morris E. Tuberculosis of the liver. Am Rev Tuberc 1930; 22:585–592. Kok KyY, Yapp SK. Isolated heaptic tuberculosis: report of five cases and review of the literature. J Hepatobiliary Pancreat Surg 1999; 6:195–198. Alvarez SZ. Hepatobiliary tuberculosis. J Gastroenterol Hepatol 1998; 13:833–839. Tan TC, Cheung AY, Wan WY, Chen TC. Tuberculoma of the liver presenting as a hyperechoic mass on ultrasound. Br J Radiol 1997; 70:1293. Pitchenik AE, Fertel D, Bloch AB. Mycobacterial disease: epidemiology, diagnosis, treatment, and prevention. Clin Chest Med 1988; 9:425–441. Young LS. Mycobacterium avium complex infection. J Infect Dis 1988; 157:863–867. Lillo M, Orengo S, Cernoch P, Harris RL. Pulmonary and disseminated infection due to Mycobacterium kansasii: a decade of experience. Rev Infect Dis 1990; 2:760–767. Akiyama H, Maruyama T, Uetake T, et al. Systemic infection due to atypical mycobacteria in patients with chronic myelogenous leukemia. Rev Infect Dis 1991; 13:815–818. Patel R, Roberts GD, Ketaing MR, Paya CV. Infections due to nontuberculous mycobacteria in kidney, heart, and liver transplant recipients. Clin Infect Dis 1994; 19:263–273. Hook EW3. Acquired syphilis in adults. N Engl J Med 1992; 326:1060–1069. Shah MC, Barton LL. Congenital syphilitic hepatitis. Pediatr Infect Dis J 1989; 8:891–892. Herman TE. Extensive hepatic calcification secondary to fulminant neonatal syphilitic hepatitis. Pediatr Radiol 1995; 25:120–122. Taniguchi Y, Nakae Y, Ikoma K, et al. Subclinical syphilitic hepatitis, which was markedly worsened by a Jarisch– Herxheimer reaction. Am J Gastroenterol 1999; 94:1694–1699. Silberstein P, Lawrence R, Pryor D. A case of neurosyphilis with florid Jarisch–Herxheimer reaction. J Clin Neurosci 2002; 9:689–690.
763
Section V. Liver Diseases Due to Infectious Agents
135. Maincent G, Labadie H, Fabre M. Tertiary hepatic syphilis. A treatable cause of multinodular liver. Dig Dis Sci 1997; 42:447–450. 136. Steere AC. Lyme disease. N Engl J Med 2001; 345:115–125. 137. Steere AC, Malawista SE, Syndman DR, et al. Lyme arthritis: an epidemic of oligoarticular arthritis in children and adults in three Connecticut communities. Arthritis Rheum 1977; 20:7–17. 138. Horowitz HW, Dworkin B, Forseter G. Liver functions in early Lyme disease. Hepatology 1996; 23:1412–1417. 139. Steere AC, Bartenhagen NH, Craft JE. The early clinical manifestations of Lyme disease. Ann Intern Med 1983; 99:76–82. 140. World Health Organization. Leptospirosis worldwide. Wkly Epidemiol Rec 1999; 74:237–242. 141. Faine S, Adler B, Bolin C, Percolat P. Leptospira and leptospirosis, 2nd edn. Melbourne: MedSci; 1999. 142. Edwards GA, Domm BM. Human leptospirosis. Medicine 1960; 39:117–156. 143. Levett PN. Leptospirosis. Clin Microbiol Rev 2001; 14:296–326. 144. Edwards CN, Nicholson GD, Hassell TA, et al. Leptospirosis in Barbados: a clinical study. W Indian Med J 1990; 39:27–34. 145. Zaki SR, Spiegel RA. Leptospirosis. In: Nelson AM, Horsburgh CR, eds. Pathology of emerging infections. Washington, DC: American Society for Microbiology Press; 1998:73–92. 146. Guidugli F, Castro AA, Atallah AN. Antibiotics for treating leptospirosis (Cochrane review), vol. 3. Cochrane Library. Oxford: Update Software; 2003. 147. Takafuji ET, Kirkpatrick JW, Miller RN, et al. An efficacy trial of doxycycline chemoprophylaxis against leptospirosis. N Engl J Med 1984; 310:497–500. 148. Dumler JS, Barbet AF, Bekker CPJ. Reorganisation of genera in the families Rickettsiacease and Anaplasmatacease in the order Rickettsiales: unification of some species of Ehrlichia with Anaplasma, Cowdria with Ehrlichia and Ehrlichia with Neorickettsia; descriptions of six new species combinations and designation of Ehrlichia equi and ‘HGE agent’ as subjective synonyms of Ehrlichia phagocytophila. Int J Syst Evol Microbiol 2001; 51:2145–2165. 149. Helmick CG, Gernard KW, D’Angelo LJ. Rocky Mountain spotted fever: clinical, laboratory, and epidemiological features of 262 cases. J Infect Dis 1984; 150:480–486. 150. Kaplowitz LG, Fischer JJ, Sparling PF. Rocky Mountain spotted fever: a clinical dilemma. In: Remingon JB, Schwartz MN, eds. Current clinical topics in infectious diseases, vol. 2. New York: McGraw-Hill; 1981:89–108. 151. LaScola B, Raoult D. Laboratory diagnosis of rickettsioses: current approaches to diagnosis of old and new rickettsial diseases. J Clin Microbiol 1997; 35:2715–2727. 152. Sexton DJ, Kanj SS, Wilson K, et al. The use of a polymerase chain reaction as a diagnostic test for Rocky Mountain spotted fever. Am J Trop Med Hyg 1994; 50:59–63. 153. Raoult D, Ndihokubwayo JB, Tissot-Dupont H, et al. Outbreak of epidemic typhus associated with trench fever in Burundi. Lancet 1998; 352:353–358. 154. Niang M, Brouqui P. Epidemic typhus imported from Algeria. Emerg Infect Dis 1999; 8:1515–1518. 155. Azad AF, Radulovic S, Higgins JA, et al. Flea-borne rickettsioses: ecologic considerations. Emerg Infect Dis 1997; 3:319–327. 156. Comer JA, Paddock CD, Childs JE. Urban zoonoses caused by Bartonella, Coxiella, Ehrlichia, and Rickettsia species. Vector Borne Zoonotic Dis 2001; 1:91–118. 157. Silpapojakul K, Chayakul P, Krisanapan S. Murine typhus in Thailand: clinical features, diagnosis and treatment. Q J Med 1993; 86:43–47. 158. Tamura A, Ohashi N, Urakami H, Miyamura S. Classification of Rickettsia tsutsugamushi in a new genus, Orientia gen nov, as Orientia tsutsugamushi comb. nov. Int J Syst Bacteriol 1995; 45:589–591.
764
159. Tattersall RN. Tsutsugamushi fever on the India–Burma border. Lancet 1945; 2:392–394. 160. Berman SJ, Kundin WD. Scrub typhus in South Vietnam. A study of 87 cases. Ann Intern Med 1973; 79:26–30. 161. Eremeeva ME, Balayeva NM, Raoult D. Serological response of patients suffering from primary and recrudescent typhus: comparison of complement fixation reaction, Weil–Felix test, microimmunofluorescence, and immunoblotting. Clin Diagn Lab Immunol 1995; 1:318–324. 162. Carl M, Tibbs CW, Dobson ME. Diagnosis of acute typhus infection using the polymerase chain reaction. J Infect Dis 1990; 161:791–793. 163. Perine PL, Krause DW, Awoke A, McDade JE. Single-dose doxycycline treatment of louse-borne relapsing fever and epidemic typhus. Lancet 1974; 2:742–744. 164. Bakken JS, Dumler JS, Chen SM, et al. Human granulocytic ehrlichiosis in the upper Midwest United States. A new species emerging? JAMA 1994; 272:212–218. 165. Ijdo JW, Meek JI, Cartter ML, et al. The emergence of another tickborne infection in the 12-town area around Lyme Connecticut: human granulocytic ehrlichiosis. J Infect Dis 2000; 181:1388–1393. 166. Olano JP, Walker DH. Human ehrlichiosis. Med Clin North Am 2002; 86:375–392. 167. Paddock CD, Sumner JW, Shore GM. Isolation and characterization of Ehrlichia chaffeensis strains from patients with fatal ehrlichiosis. J Clin Microbiol 1997; 35:2496–2502. 168. Moskovitz M, Fadden R, Min T. Human erhlichiosis: a rickettsial disease associated with severe cholestasis and multisystemic disease. J Clin Gastroenterol 1991; 13:86–90. 169. Paddock CD, Suchard DP, Grumbach KL, et al. Brief report: fatal seronegative ehrlichiosis in a patient with HIV infection. N Engl J Med 1993; 329:1164–1167. 170. Paddock CD, Folk SM, Shore GM, et al. Infections with Ehrlichia chaffeensis and Ehrlichia ewingii in persons coinfected with human immunodeficiency virus. Clin Infect Dis 2001; 33:1586–1594. 171. Dawson JE, Fishbein DB, Eng TR. Diagnosis of human ehrlichiosis with the indirect fluorescent antibody test: kinetics and specificity. J Infect Dis 1990; 162:91–95. 172. Everett ED, Evans KA, Henry RB. Human ehrlichiosis in adults after tick exposure. Diagnosis using polymerase chain reaction. Ann Intern Med 1994; 120:730–735. 173. Bakken JS, Krueth J, Wilson-Nordskog C, et al. Clinical and laboratory characteristics of human granulocytic ehrlichiosis. JAMA 1996; 275:199–205. 174. Edelman DC, Dumler JS. Evaluation of an improved PCR diagnostic assay for human granulocytic ehrlichiosis. Mol Diagn 1996; 1:41–49. 175. Horowitz HW, Aguero-Rosenfeld ME, McKenna DF, Holmgren D. The clinical and laboratory spectrum of culture proven human granulocytic ehrlichiosis: comparison with culture negative cases. Clin Infect Dis 1998; 27:1314–1317. 176. Magnarelli LA, Dumler JS, Anderson JF, et al. Coexistence of antibodies to tick-borne pathogens of babesiosis, ehrlichiosis, and Lyme borreliosis in human sera. J Clin Microbiol 1995; 33:2054–2057. 177. Belongia EA. Epidemiology and impact of coinfections acquired from Ixodes ticks. Zoonotic Dis 2002; 2:265–273. 178. Baca OG, Paretsky D. Q fever and Coxiella burnetii: a model for host–parasite interaction. Microbiol Rev 1983; 47:127–149. 179. Raoult D. Rickettsial diseases. Medicine (Baltimore) 1996; 24:71–75. 180. Hofmann CER, Heaton JW Jr. Q fever hepatitis. Clinical manifestations and pathological findings. Gastroenterology 1982; 83:474–479. 181. Dupont HL, Hornick EV, Levin HA, et al. Q fever hepatitis. Ann Intern Med 1971; 74:198–206.
Chapter 37 BACTERIAL AND MISCELLANEOUS INFECTIONS OF THE LIVER
182. Qizilbash AH. The pathology of Q fever as seen on liver biopsy. Arch Pathol Lab Med 1983; 107:364–367. 183. Westlake P, Price LM, Russell M, Kelly JK. The pathology of Q fever hepatitis. A case diagnosed by liver biopsy. J Clin Gastroenterol 1987; 9:357–363. 184. Greiner TC, Mitros FA, Stapleton JT, Van Rybroeck J. Fine needle aspiration findings of the liver in a case of Q fever. Diagn Cytopathol 1992; 8:181–184. 185. DiSalvo AF, Ajello L, Palmer JW, Winkler WG. Isolation of Histoplasma capsulatum from Arizona bats. Am J Epidemiol 1969; 89:606–614. 186. Newman SL. Macrophages in host defense against Histoplasma capsulatum. Trends Microbiol 1999; 7:67–71. 187. Wheat LJ. Laboratory diagnosis of histoplasmosis: update 2000. Semin Respir Infect 2001; 16:131–140. 188. Martin RC, Edwards MJ, McMasters KM. Histoplasmosis as an isolated liver lesion: review and surgical therapy. Am Surg 2001; 67:430–431. 189. Wheat J, Sarosi G, McKinsey D, et al. Practice guidelines for the management of patients with histoplasmosis. Clin Infect Dis 2000; 30:688–695. 190. Perfect JR, Casadevall A. Cryptococcosis. Infect Dis Clin North Am 2002; 16:837–874. 191. Perfect JR. Cryptococcosis. Infect Dis Clin North Am 1989; 3:77–102. 192. Leenders AC, Reiss P, Portegies P, et al. Liposomal amphotericin B (Ambisome) compared with amphotericin B followed by oral fluconazole in the treatment of AIDS-associated cryptococcal meningitis. AIDS 1997; 11:1463–1471. 193. Stern JJ, Hartman BJ, Sharkey P, et al. Oral fluconazole therapy for patients with the acquired immunodeficiency syndrome and cryptococcosis: experience with 22 patients. Am J Med 1988; 85:477–480. 194. van der Horst C, Saag MS, Cloud GA, et al. Treatment of cryptococcal meningitis associated with the acquired immunodeficiency syndrome. N Engl J Med 1997; 337:15–21. 195. Ghannoun MA, Ibrahim AS, Fu Y. Susceptibility testing of Cryptococcus neoformans: a microdilution technique. J Clin Microbiol 1992; 30:2881–2886. 196. Zangerl B, Edel G, von Manitius J, Schmidt-Wilcke HA. Coccidioidomycosis as the cause of granulomatous hepatitis. Med Klin 1998; 93:170–171. 197. Kaufman L, Sekhon AS, Moledina N, et al. Comparative evaluation of commercial Premier EIA and
198.
199. 200.
201.
202.
203. 204.
205. 206.
207.
208.
209.
210. 211.
microimmunodiffusion and complement fixation tests for Coccidioides immitis antibodies. J Clin Microbiol 1995; 33:618–629. Galgiani JN, Grace GM, Lundgren LL. New serologic tests for early detection of coccidioidomycosis. J Infect Dis 1991; 163:645–649. Bowman BH. Designing a PCR/probe detection system for pathogenic fungi. Clin Immunol 1992; 12:66–69. Gomes GM, Cisalpino PS, Taborda CP, Camargo ZP. PCR for diagnosis of paracoccidioidomycosis. J Clin Microbiol 2000; 38:3478–3480. Vargas SL, Hughes WT, Santolyay ME. Search for primary infection by Pneumocystis carinii in a cohort of normal, healthy infants. Clin Infect Dis 2001; 32:855–861. Persaud D, Bangaru B, Greco MA, et al. Cholestatic hepatitis in children infected with the human immunodeficiency virus. Pediatr Infect Dis J 1993; 12:492–498. Singh N, Yu VL. Abdominal pain and cholestatic hepatitis in an AIDS patient. Hosp Practice 1993; 28:104–108. Barry SM, Johnson MA. Pneumocystis carinii pneumonia: a review of current issues in diagnosis and management. HIV Med 2001; 2:123–132. Eickhoff TC. Infectious complications in renal transplant recipients. Transplant Proc 2003; 5:1233–1238. Kroe DM, Kirsch CM, Jensen WA. Diagnostic strategies for Pneumocystis carinii pneumonia. Semin Respir Infect 1997; 12:70–78. Tashjian LS, Abramson JS, Peacock JE Jr. Focal hepatic candidiasis: a distinct clinical variant of candidiasis in immunocompromised patients. Rev Infect Dis 1984; 6:689–703. Johnson TL, Barnett JL, Appelman HD, Nostrant T. Candida hepatitis. Histopathologic diagnosis. Am J Surg Pathol 1988; 12:716–720. Vairani G, Rebeschini R, Barbazza R. Hepatic and subcutaneous abscesses due to aspergillosis. Initial diagnosis of a case by intraoperative fine needle aspiration cytology. Acta Cytol 1990; 34:891–894. Singh N, Paterson DL. Aspergillus infections in transplant recipients. Clin Microbiol Rev 2005; 18:44–69. Pfaller MA, Diekema DJ. Rare and emerging opportunistic fungal pathogens: concern for resistance beyond Candida albicans and Aspergillus fumigatus. J Clin Microbiol 2004; 42:4419–4431.
765
Section V. Liver Diseases Due to Infectious Agents
38
GRANULOMATOUS DISEASES OF THE LIVER Shobha Sharma Abbreviations BCG bacille Calmette-Guérin FUO fever of unknown origin HCV hepatitis C virus HIV human immunodeficiency virus
IFN-g R1 IL MAI MTb
interferon-g receptor interleukin Mycobacterium avium intracellulare Mycobacterium tuberculosis
INTRODUCTION The interpretation of hepatic granulomata is more difficult than is generally appreciated, first because the etiology of such lesions can seldom be established on histologic grounds alone, and second because the presence of granulomata in the liver does not necessarily imply an underlying systemic granulomatous process. Indeed, unless these two points are borne in mind, the discovery of granulomata in the liver may prove to be misleading rather than diagnostic.1
DEFINITION Granulomas (Figure 38-1) are composed of discrete aggregates of epithelioid cells. Epithelioid cells are transformed macrophages that have abundant cytoplasm, rich in endoplasmic reticulum and with fewer phagolysosomes than macrophages. These characteristics are consistent with a secretory function for epithelioid cells. In contrast, macrophages are primarily phagocytic. Based on these differences, it is possible to distinguish an epithelioid cell granuloma from an aggregate of macrophages using the PAS (periodic acid–Schiff) stain. Epithelioid cell granulomas are PAS negative, in contrast to an aggregate of macrophages which may contain PAS-positive debris. Macrophages within an aggregate are always separate from one another, whereas epithelioid cells fuse to form a syncytium as well as multinucleated giant cells. In addition to epithelioid cells, granulomas also contain a variable number of multinucleated giant cells and other inflammatory cells, such as lymphocytes and macrophages.2
IMMUNOLOGY The purpose of a granuloma is to destroy or contain an injurious agent that cannot be disposed of either directly or indirectly by the humoral limb of the immune system. The injurious agent may be intracellular, as in Mycobacterium, extracellular as in schistosomiasis, or unknown as in sarcoidosis. It is well recognized that intracellular pathogens elicit a cytokine response that is distinct and different from that induced by extracellular pathogens. In a type 1
PAS PBC TNF
periodic acid-Schiff primary biliary cirrhosis tumor necrosis factor
or Th1 response, the cytokines secreted are interferon-g, interleukin-2 and interleukin-12 (IL-2, IL-12). The type 1 response develops against intracellular pathogens and is typically seen in mycobacterial and sarcoidal granulomas.3–5 Granulomas that develop under the influence of a Th1 response are larger, poorly formed, and more destructive than those that develop in response to Th2 cytokines. In contrast, the type 2 or Th2 response is characterized by secretion of interleukins-4, -5, -6 and -10 (IL-4, IL-5, IL-6, IL-10)3,6,7 and is primarily directed against extracellular antigens. Schistosomal ova induce a Th2 response against soluble egg antigens, contain prominent numbers of eosinophils, and may be associated with fibrosis, features that are attributed to IL-5 and IL4, respectively. Antigen-presenting cells such as macrophages and dendritic cells process antigens and present them to MHC class II restricted helper T cells. These macrophages also secrete interleukin-12, which stimulates the differentiation of CD4 lymphocytes, which in turn secrete INF-g and IL-2 (Th1 response).8 In a positive feedback loop, INF-g and IL-2 amplify the immune response by stimulating proliferation of T cells and the further production of INF-g and IL-2. In turn, these cytokines recruit monocytes and stimulate them to differentiate into macrophages. The activated macrophages secrete tumor necrosis factor-a (TNF-a), which up-regulates the expression of intercellular adhesion molecules (ICAM), allowing inflammatory cells to adhere to endothelial cells and localize to the antigenic stimulus. TNF-a also stimulates T-cell proliferation and the secretion of INF-g.9,10 In schistosomiasis, an initial type 1 response usually evolves into a type 2 response.9,11 This shift in cytokine profiles appears to be mediated by IL-10. Both IL-12 (Th1) and IL-10 (Th2) are cross-regulatory cytokines. For example, interferon-g and IL-12 secreted in a Th1 response can suppress the production of IL-4 and IL-10 associated with the Th2 response, and similarly IL-10 can suppress a Th1 response.8,12 Which response will predominate in a reaction depends on the antigen and the timing of the cytokines secreted in relation to the chronologic evolution of the granuloma. Thus, though the type 1 or type 2 responses may be present in an individual they are not mutually exclusive,3,4 and one response can evolve into the other. This flexibility may provide protection to the host. This is illustrated in the case of schistosomiasis, where an unabated Th1 response results in hepatotoxicity and death in a nude
767
Section V. Liver Diseases Due to Infectious Agents
Figure 38-1. High-power photomicrograph of margin of epithelioid granuloma. Epithelioid cells (E) have abundant cytoplasm and indistinct cytoplasmic borders. Other inflammatory cells, such as lymphocytes (L) and macrophages (M), are present at the junction of the granuloma and the hepatocytes (H). (Hematoxylin and eosin, ¥480)
mouse model, whereas in other animals with schistosomiasis the evolution of a Th1 response into a Th2 response reduces the risk of hepatotoxicity.13 Type 1 and 2 cytokine responses are not restricted to T-helper cells, and similar but not identical responses have been demonstrated in cytotoxic T cells, B cells, natural killer cells and dendritic cells.3,4 In addition to immune induction, a non-immune granulomatous reaction can be induced around indigestible foreign material such as talc. However, talc more frequently accumulates in macrophages within the portal tract and Kupffer cells, rather than forming distinct granulomas.
morphologic grounds. An attempt should be made to examine granulomas for foreign material such as talc, which is birefringent under polarized light.14,15 The value of acid-fast and fungal stains is questionable in non-necrotizing granulomas and must be correlated with the indication for the biopsy, as well as the immune status and geographic background of the patient. The likelihood of finding acid-fast bacilli or fungal elements is low, particularly when the granulomas are found incidentally during the work-up for chronic hepatitis. Organisms are more likely to be found when the liver biopsy is performed in the investigation of pyrexia of unknown origin.16–18 Granulomas other than those associated with primary biliary cirrhosis, sarcoidosis and schistosomiasis are rarely destructive and not associated with consistent derangements of liver tests. Damage to bile ducts is seen in primary biliary cirrhosis and less frequently in sarcoidosis. Destruction of hepatic and portal veins with subsequent obliteration and scarring is implicated as a mechanism of portal hypertension that can develop in patients with granulomatous hepatitis.
LIPOGRANULOMAS Lipogranulomas are distinctive but inconsequential granulomas in the liver (Figure 38-2). These granulomas are composed of lipid-laden histiocytes, lipid vacuoles, and a variable number of chronic inflammatory cells. They are well circumscribed and are usually located around the central vein, though they may be present in the portal areas. The surrounding hepatic parenchyma may be normal or steatotic.2,22 In non-fatty livers, these granulomas appear to develop in response to exogenous mineral oils that are widely used in food processing. Using thin layer and gas–liquid chromatography studies, lipid extracted from liver tissue, candies and polished skins of apples and cucumbers show similar characteristics to mineral oil. Moreover, the incidence of lipogranulomas has increased over the years, which is consistent with the widespread use of mineral oils in the food industry.22
HISTOPATHOLOGY There is a discrepancy between the detection of granulomas in needle biopsies and that in autopsy samples,19 and this reflects the amount of tissue sampled as well as the ability to select grossly abnormal areas for sampling in autopsy material. Also, the likelihood of finding granulomas increases with the number of biopsies obtained. For example, the probability of finding granulomas increases from 50% when one biopsy core is obtained to 100% when three cores are obtained.20,21 The morphologic appearance of granulomas is very variable. They may be distinct and well formed, as in sarcoidosis, or they may be ill defined as in some drug reactions. They may be necrotizing in infections such as tuberculosis, or non-necrotizing as in sarcoidosis. The location of granulomas within the liver, i.e. portal or acinar, is unlikely to be of diagnostic use unless primary biliary cirrhosis (PBC) is the consideration. In general, granulomas in PBC are portal, though they can be found in the acini as well.1,2 In the absence of acid-fast organisms, fungi, parasites or foreign material, it is not possible to identify the etiology of a granuloma on
768
INCIDENCE AND CAUSES The incidence of granulomas detected in liver biopsies varies from 0.8% to 15%.1,19,23–30 Hepatic granulomas may reflect a systemic granulomatous process, and the two major causes of granulomas in liver biopsies are tuberculosis and sarcoidosis (Table 38-1). Sarcoidosis is more frequent in developed countries, whereas tuberculosis is the more common cause in underdeveloped nations. Fungal and schistosomal granulomas are seen in areas endemic for these infectious agents.19,28 After excluding a pot pourri of diagnoses that includes drugs, malignancies, and a variety of infections listed in Table 38-2, there remains a category of granulomas of unknown cause and significance. These may be incidental findings in patients being staged for chronic viral hepatitis,1,24,31–33 though there is a small group of patients in whom idiopathic granulomatous hepatitis is responsible for the clinical symptoms and disordered liver tests. Thus, the determining the cause and the significance of hepatic granulomas is largely dependent on the clinical indication for the biopsy.
Chapter 38 GRANULOMATOUS DISEASES OF THE LIVER Figure 38-2. Lipogranuloma. Note small lipid droplets within the granuloma as well as within adjacent hepatocytes. (Hematoxylin and eosin, ¥480)
Table 38-1. Frequency of Granulomas in the US and Other Countries Ref
Year
City/Country
23 19 24 25 1 26 27 28 29
1953 1966 1970 1974 1977 1979 1988 1990 1994
Cincinatti, USA Texas, USA Scandinavia Cleveland, USA New York, USA Washington, DC Australia Saudi Arabia Ireland
No of bx
Bx with gran
Sarcoidosis (%)
Tb (%)
Unknown (%)
1100 1505 2813 2086 6075 N/A N/A 404 4124
54 35 21 50 565 73 59 59 163
11 23 29 22 38 55 12 0 18
24 20 48 10 12 12 4 32 18
18.50 37 36 21 3 17 0 11
Misc*(%)
54 schistosomiasis 55 PBC
*Miscellaneous causes of granulomas included syphilis,23 lymphogranuloma venereum,23 lymphoma,19,23,25–27 brucellosis,1,19,26,28 mycoses,19,26 drugs,26,27,29 Crohn’s disease,26,29 cytomegalovirus,26,27 berylliosis,1 temporal arteritis,1 Q fever,27 renal and hepatocellular carcinoma27 and typhoid.28
DISEASES ASSOCIATED WITH HEPATIC GRANULOMAS Granulomas may be found in liver biopsies performed for the following indications: ∑ ∑ ∑ ∑ ∑ ∑
Grading and staging in a patient with chronic viral hepatitis Investigation and staging of primary biliary cirrhosis Investigation of fever of unknown origin (FUO) Confirmation of diagnosis of sarcoidosis Investigation of portal hypertension Investigation of liver test abnormalities of undetermined etiology.
GRANULOMAS AND HEPATITIS C INFECTION Incidental hepatic granulomas have been documented in patients who undergo liver biopsy for the grading and staging of their hepa-
titis C infection (HCV), and the reported incidence varies from 0.73% to 10%,31–33 though it is probably closer to the latter. In one large study of 435 patients, non-necrotizing granulomas of unknown etiology were observed in 4.5% of biopsies from 155 patients with HCV infection.34 This was significantly higher than in the control groups of hepatitis B infection (0.66% of 151 patients) and alcoholic liver disease (none of 129 patients). The etiology of these granulomas is unclear, and patients do not appear to have symptoms attributable to this finding. There are anecdotal reports describing the development of non-necrotizing granulomas in the liver subsequent to treatment with interferon-a. In two patients the appearance of the granulomas appeared to correlate with a flare in transaminase activity after an initial 3–4-month biochemical response to interferon-a.35 In a third patient granulomas were no longer identified in sequential biopsies performed 6 months after withdrawal of treatment.36 The role of interferon-a in inducing hepatic granulomas is suggested by the absence of these granulomas prior to initiation of therapy, and their resolution following withdrawal of the drug.
769
Section V. Liver Diseases Due to Infectious Agents
Table 38-2. Causes of Hepatic Granulomas Found incidentally in the evaluation of primary liver disease Hepatitis C – incidental Hepatitis B – incidental27 Associated with primary liver disease Primary biliary cirrhosis Sarcoidosis – destruction of bile ducts and portal veins Associated with portal hypertension Schistosomiasis Sarcoidosis Associated with pyrexia of unknown origin Usual Sarcoidosis Tuberculosis – miliary and pulmonary Atypical mycobacteria Q fever27 Brucellosis 1,23,28 Cat-scratch disease Mycoses, e.g. histoplasmosis, Coccidiodes immitis, South American blastomycosis, candidiasis Drugs Idiopathic granulomatous hepatitis Rare Cytomegalovirus26,27 Temporal arteritis1 Listeriosis Leprosy – reactional states Miscellaneous Lymphoma23,26,27 Hepatic adenomas Hepatocellular carcinoma27 Renal carcinoma27 Leprosy Syphilis23 Lymphogranuloma venereum23 Crohn’s disease26,29 Berylliosis1
Finally, there are reports of patients with hepatitis C who developed a sarcoidosis-like illness with dry cough, dyspnea, pulmonary interstitial infiltrates and nodules or hilar lymphadenopathy. In two patients the sarcoid-like illness became apparent approximately 6 years after the diagnosis of HCV infection was made, and responded to steroid therapy.37 Neither of the patients had been treated with interferon-a. In three other patients symptoms appeared 3–5 months after initiating treatment with interferon-a. Symptoms and radiologic abnormalities subsided 3–8 months after cessation of medication and without steroids.38
PRIMARY BILIARY CIRRHOSIS Granulomas are a common finding in the liver of patients with primary biliary cirrhosis. This disease predominantly affects middleaged Caucasian women who may be diagnosed during the investigation of non-specific symptoms such as fatigue, or in whom the diagnosis is suspected because of cholestatic symptoms and signs such as pruritus and jaundice. The laboratory investigations show a two- to fivefold increase in alkaline phosphatase and the presence of the diagnostic antimitochondrial antibodies. The granulomas
770
are portal in location and may be intimately associated with interlobular bile duct damage.39,40 PBC is discussed more extensively in Chapter 42.
FEVER OF UNKNOWN ORIGIN Among the many causes of fever of unknown origin (FUO) in immunocompetent patients, intra-abdominal infections and neoplasms are diagnosed using increasingly sophisticated imaging modalities. Similarly, increasingly sensitive and specific serologic tests identify patients with a variety of collagen vascular disorders and infections. A thorough clinical history pertaining to medications and travel may provide clues to a drug-induced hepatitis and suggest the need to perform a serologic search for unusual infections. For these reasons, the diagnostic utility of a blind liver needle biopsy in the evaluation of FUO is arguable. In patients with FUO, hepatomegaly and abnormal liver tests, the diagnostic yield on liver biopsy for informative granulomas such as mycobacterial or fungal infection varies from 0 to 17%.16–18
HUMAN IMMUNODEFICIENCY VIRUS In contrast, hepatic granulomas are found in 16–75% of human immunodeficiency virus (HIV)-infected patients being investigated for the cause of an FUO and elevated liver tests or hepatomegaly.41–46 The most likely cause for these granulomas is infection with either Mycobacterium tuberculosis (MTb) or M. avium intracellulare (MAI). If the organisms are identified on AFB stains, a definitive diagnosis is quickly established and treatment instituted.42 Comparable to observations in non-HIV infected patients, acid-fast organisms are easier to find when the agent is MAI rather than MTb. The role of blood and tissue culture in HIVinfected patients is controversial but is useful to speciate the organism and to identify additional cases in which the liver or bone biopsy does not show diagnostic pathology.43 In both immunocompetent and immunocompromised patients the diagnostic yield for fungi and mycobacterial infections is greater in liver biopsy than in bone marrow biopsy.42,47
CAUSES OF HEPATITIC GRANULOMAS SARCOIDOSIS Sarcoidosis is a systemic granulomatous disease characterized by a Th1 response against an unknown antigen. Clustering of patients suggests a temporal and spatial relationship that would be expected if the cause was either a transmissible agent or a shared environmental factor. The fact that the clustering has been shown to be greater between relatives of patients than among their spouses suggests a genetic predisposition, as the latter would be considered most susceptible if contact or a shared environment were the only factors involved in the clustering of cases.48–51 This disease afflicts those in the 20–40-year age groups and affects both Caucasians and African-Americans, but Swedes, Danes and African-Americans have the highest prevalence rates. In most patients the disease is indolent, though in a small proportion patients may be quite symptomatic and the disease can result in death. The
Chapter 38 GRANULOMATOUS DISEASES OF THE LIVER
course tends to be more severe in African-Americans and Swedes, and poor prognostic features include African-American origin, age over 40 years, and involvement of three organ systems.52,53 Patients may present with fever of unknown origin, nonproductive cough and chest pain, malaise, and weight loss. The organs most commonly affected are the lungs (100%) and mediastinal lymph nodes (90%). The diagnosis is one of exclusion and the clinical history must exclude occupational or environmental factors that could cause granulomas. Other tests include chest Xray, pulmonary function testing, electrocardiogram, liver tests, ophthalmic examination and tuberculin skin test. These tests identify involvement of the lung, heart, liver and eye, and screen for tuberculosis. Confirmation of the diagnosis is by biopsy identification of the granulomas, and the preferred sites of biopsy are the tracheobronchial tree and the lung.52,53 Hepatic involvement is usually silent and liver dysfunction is an unusual manifestation of sarcoidosis. Presentation of the disease as either hepatitis or carditis is unusual and is seen in 4–7% of patients.53 Symptoms and signs suggestive of hepatic involvement include abdominal pain, hepatomegaly, jaundice and portal hypertension.54–56 Patients may present with FUO, and in a third of cases there is a disproportionate elevation in alkaline phosphatase in comparison to the amino transaminases.54,57–59 Non-necrotizing granulomas are identified in the liver in 24–75% of patients with sarcoidosis,57,60,61 and sarcoidosis is one of the commonest causes of hepatic granulomas (see Table 38-1). The incidence is slightly higher in biopsy series than in autopsies, and this is probably because a liver biopsy is more likely to be performed when there are symptoms and signs of liver involvement in a patient with established sarcoidosis.57,60,62 The granulomas are characteristically tight, well formed and nonnecrotizing. They are uniformly distributed throughout the liver, and although they tend to be periportal they are also seen within lobules. A component of lobular hepatitis and portal triaditis may be present.62 Most sarcoidal granulomas resolve spontaneously, though some heal by fibrosis and scarring. Cirrhosis may result because of scarring of granulomas or a coexistent disease such as viral hepatitis. Portal hypertension can develop in the absence of cirrhosis, and explanations for its development include presinusoidal portal hypertension secondary to scarring and obliteration of small portal and hepatic veins by portal granulomas, and arteriovenous shunts within the granulomas.54,58,63–65 Destruction of interlobular bile ducts produces a histologic picture that is indistinguishable from that of primary biliary cirrhosis.55,56,59,62 Bile duct damage resembling sclerosing cholangitis has also been described.59,66 Although patients with hepatic involvement may improve symptomatically and biochemically when placed on steroids, structural damage, particularly ductopenia, is irreversible.67,68 Ursodeoxycholic acid has also been used to alleviate cholestatic symptoms.69,70 Recurrence of the disease following liver transplantation has been documented, though this is unusual.71–73
Africa to the South Pacific and Southern China. Also, as a result of travel and migration 400 000 people with schistosomiasis live in the United States.74–77 Schistosomes are digenetic (sexual and asexual reproduction in alternating generations) trematodes, the most common species of which are S. hematobium (Africa and Middle East), S. mansoni (South America, Caribbean, Africa and Middle East) and S. japonicum (Far East). The adult worms reside in the mesenteric (S. mansoni and japonicum) and perivesical venous plexuses (S. hematobium). The adult worm is not immunogenic, but the schistosoma ova are highly antigenic.78–80 The ova elicit a Th2 cytokine response resulting in the development of eosinophil-rich granulomas in the portal tracts (Figure 38-3). In mice with severe combined immunodeficiency, the inability to elicit a granulomatous response leads to severe hepatotoxicity and death due to the egg antigens.13 Interestingly, though, this granulomatous response kills at least a third of eggs and protects the host from the toxicity of the egg antigens. It is also utilized by the parasite to protect the egg from further host damage and permit egg migration.13,78 Complications within the liver include portal fibrosis due to the release of fibrogenic cytokines such as IL-4. Granulomas and fibrosis cause obliteration and/or compression of the portal veins, resulting in presinusoidal portal hypertension. In a small proportion of patients extensive fibrosis develops along the portal venous system and is also known as the pipe-stem fibrosis of Symmers. The fibrosis follows the distribution of the portal veins but does not transect the hepatic parenchyma, and therefore is distinct from cirrhosis.81–84 Hepatic fibrosis does not develop in all patients. This complication is most often seen in young adults (5–15 years) who have had prolonged intense infection (15–20 years). Using segregation and linkage analysis, a genetic locus controlling for infection intensity was identified in a Brazilian cohort. This locus mapped to chromosome 5q31-q33. The genes for IL-4 and IL-5 are also located in this region.85 Similar studies identified a locus controlling for fibrosis on chromosome 6q22-q23. The gene on chromosome 6 is closely linked
SCHISTOSOMIASIS Schistosomiasis (bilharziasis) is another example of a granulomatous hepatitis that can be associated with portal hypertension. It affects 200 million persons worldwide and the infestation is seen in a wide geographic belt that extends from South America, across the Caribbean islands, sub-Saharan Africa and the Middle East, South
Figure 38-3. Portal granulomatous response and fibrosis around schistosome ova. (Hematoxylin and eosin, ¥55)
771
Section V. Liver Diseases Due to Infectious Agents
to the IFN-g R1 (interferon-g receptor) gene. The exact genes and their protein products have yet to be identified.86 Certain polymorphisms within the interferon-g gene itself, such as IFN-g +2109, seem to be associated with severe fibrosis, whereas polymorphisms on IFN-g +3810 appear to protect against the development of fibrosis.87 Patients with hepatosplenic schistosomiasis have high levels of INF-g, TNF-a and soluble TNF receptors and low levels of IL-5 in their serum. This Th1 cytokine profile suggests that patients who do not develop a Th2 response are more likely to develop hepatic fibrosis.88 Understanding how cytokines modulate the development of granulomas may allow therapeutic immunomodulation to achieve the balance between isolation of the toxic egg antigens in a granulomatous response and limiting the development of fibrosis.80 A good clinical history that elicits origin or travel to endemic regions is perhaps the first clue to suspecting the clinical diagnosis. Though most patients are asymptomatic in the early phase of the infection, there may be a local skin rash at the site of entry of schistosomal cercaria (cercarial dermatitis). Persons living in endemic regions are usually asymptomatic. In contrast, visitors to endemic areas develop fever, chills, cough, diarrhea, malaise and arthralgias. These symptoms begin 4–10 weeks after infection. The physical examination of acute schistosomiasis is characterized by hepatosplenomegaly, and the blood work-up shows peripheral blood eosinophilia. Liver tests demonstrate a mild elevation of aminotransferases.76,77 In endemic areas, the diagnosis is made on examination of the stool and urine for ova. However, when the infestation is light, this may be an insensitive tool. In this situation detection of antibodies using sensitive and specific ELISA assays developed against microsomal antigens of the adult worm can be performed. Antibody testing can only be performed 6–8 weeks after exposure.89 The preferred treatment is with praziquantel, or oxamniquine is an alternative in patients infected with Schistosoma mansoni.76,77
TUBERCULOSIS Tuberculosis may manifest primarily as liver disease because of either hepatomegaly or abnormal liver tests. The majority of these patients have miliary tuberculosis, and caseating granulomas are seen in over 80% of cases.90 In one study of 36 patients with miliary tuberculosis, granulomas were present in 91% of the liver biopsies, 52% of which were caseating. In contrast, granulomas were only noted in 53% of the bone marrow biopsies performed on the same patients. Moreover, the hepatic granulomas were present in 78% of those whose bone marrow biopsies did not show granulomas. The conclusion of this study was that although bone marrow biopsies are safer to perform, the diagnostic yield of the liver biopsy in miliary tuberculosis is higher.47 Non-necrotizing hepatic granulomas are seen in 25% of patients with pulmonary tuberculosis, and it is unusual to find acid-fast organisms in these biopsies.91 Similarly, acid-fast bacilli are only identified in 9% of hepatic granulomas in miliary tuberculosis.90 Therefore, although the acid-fast stain should be performed in all cases of granulomatous inflammation associated with PUO, it should be recognized that the stain is insensitive in detecting acid-fast organisms and results vary from 0 to 35%. Similarly, culture of biopsied material yields organisms in less than 10% of cases. PCR for Mycobacterium tuberculosis has been performed on formalin-fixed
772
paraffin embedded sections. Although it has a specificity of 96%, the sensitivity is only 53%. However, this is a relatively rapid method of detection, with a 90% positive predictive value and a 76% negative predictive value.92
ATYPICAL MYCOBACTERIA The characteristic appearance of Mycobacterium avium intracellulare (MAI) in immunocompromised patients and children is the presence of collections of foamy macrophages filled with acid-fast organisms. Infection with Mycobacterium gevanese has also been described in patients immunocompromised by HIV infection, and the morphology is similar.93 The morphology of MAI is different in immunocompetent persons in whom well-formed granulomas have been described in the liver and spleen. In these patients, it is unusual to demonstrate acid-fast organisms in the tissue sections and confirmation is by culture.94 A hepatitic presentation has been described in a patient infected with Mycobacterium scrofulaceum. Noncaseating granulomas were identified on liver biopsy and the diagnosis confirmed on tissue culture.95
BACILLE CALMETTE–GUÉRIN (BCG) BCG is an attenuated strain of Mycobacterium bovis that is used in the immunotherapy of superficial bladder carcinoma. Hepatic dysfunction and granulomatous hepatitis is a very rare complication of intravesical therapy.96 In a large series of 2602 patients tuberculous hepatitis/pneumonia developed in 18 (0.7%), although the frequency of hepatic granulomas in the absence of liver dysfunction is probably higher.96,97 It is rare to culture or to find acid-fast organism in the granulomas,98,99 and this has made the distinction between a hypersensitivity reaction and systemic mycobacteremia difficult. Therefore, the goal of treatment in patients who develop symptoms and signs of hepatic dysfunction is to cover both possibilities, and includes a 6-month course of rifampin and isoniazid, accompanied by steroids if indicated, and possibly cessation of immunotherapy.96
Q FEVER ‘Q’ or query fever was described in 1937 as an occupational disease among slaughterhouse workers and dairy farmers,100 and in 1999 it became a notifiable disease. It is a zoonotic infection caused by an intracellular Gram-negative rickettsial organism, Coxiella burnetii. The primary but not exclusive reservoirs are cattle, goats and sheep, and the infection is maintained in ticks and other arthropods. The organisms are excreted into the milk, urine and feces of the animals, and there is a high concentration in the placenta and amniotic fluid. For this reason, in sheep-farming communities the incidence rises during the lambing season. The organisms are resistant to heat and drying, and the most common method of infection is inhalation of aerosolized bacteria. Ingestion of contaminated milk and tick bites are other sources of infection. The incubation period is between 2 and 3 weeks.101 Most infected patients are asymptomatic. In those who do develop symptoms the illness is self-limited and characterized by a high spiking fever (38.5–40∞C), malaise, bifrontal headache, and pneumonia.102 In most patients there is resolution in 2–3 weeks, but in up to 16% chronic Q fever develops that is characterized by endocarditis.103
Chapter 38 GRANULOMATOUS DISEASES OF THE LIVER
Liver abnormalities are found in 11–65% of cases, and in a study of 72 patients 85% had abnormal liver tests and 65% had hepatomegaly.104 Liver biopsies may be performed in patients who present with fever of unknown origin. Although non-specific granulomas may be seen against a background of non-specific reactive hepatitis and steatosis, a characteristic fibrin ring granuloma (Figure 38-4) has been described. The granuloma surrounds a clear space felt to represent a lipid vacuole. A fibrin ring is present that either surrounds the lipid vacuole within the granuloma or is at the periphery of the entire granuloma.105 Variations on this histology include a granuloma around a lipid vacuole without a fibrin ring, or a granulomatous response around fragmented fibrin material.106,107 In fact, the recognition of this characteristic histology has led to serologic testing and confirmation of Coxiella burnetti infection in some patients with fever of unknown origin.106,108 In some patients who respond to treatment and are followed up with a repeat biopsy, there is resolution of the histologic findings in the liver. The granulomas do not show a preferential lobular or periportal distribution in the liver. Although there are individual reports of fibrosis developing in these patients, based on sequential biopsies, these reports are prior to the availability of hepatitis C testing and are so rare that an association between Q fever and the development of hepatic fibrosis cannot be made with any certainty.109 Although fibrin ring granulomas are characteristically described in acute Q fever and have led to serologic testing and confirmation of the diagnosis, they are not seen in all such patients108 and are not specific for this disease. Isolated case reports have described similar granulomas in liver biopsies from patients with viral hepatitis A, temporal arteritis, Epstein–Barr virus infection, cytomegalovirus infection, systemic lupus erythematosus, leishmaniasis and allopuri-
nol-induced hepatitis. Although Q fever was considered in all these cases, it was excluded on serologic testing.110–116 Confirmation of diagnosis is by identification of specific antibodies. The organism exists in two antigenic phases and antibodies to phase II antigens are seen early in the disease. Chronicity should be suspected if there are rising antibodies to phase I antigens with either constant or falling levels of phase II antibodies.117,118 The treatment of choice is doxycycline 100 mg twice daily for 15–21 days.100
BRUCELLOSIS Brucellosis is a zoonotic infection found in a variety of farm animals, including goats, pigs, cattle and dogs. Human infection is caused by Brucella abortus (cattle), B. suis (pigs) and B. melitensis (goats). Although B. abortus is most widely prevalent in the USA, worldwide the most clinically important of these species is B. melitensis.119 Human infection occurs through contact with animals or animal products, such as cheese made from unpasteurized milk. Similar to Q fever, the mode of entry is either through aerosols of the organisms, ingestion of foods or contamination of wounds. Hence those who consume these products and workers in abattoirs, animal inspectors and handlers, and veterinarians are at greatest risk. Based on information from the CDC, brucellosis is not a common infection in the US (<0.5 cases per 100000) because of strict animal control regulations and vaccination programs. The majority of cases are reported from California, Florida, Texas and Virginia. However, people in countries bordering the Mediterranean Basin, i.e. Portugal, Spain, Southern France, Italy, Greece, Turkey, North Africa, and South and Central America and the Middle East are at high risk, particularly from unpasteurized cheeses.
Figure 38-4. Fibrin ring granuloma in Q fever. Note fibrinoid ring (arrows) around the central vacuole. (Hematoxylin and eosin, ¥460)
773
Section V. Liver Diseases Due to Infectious Agents
In the acute phase, patients present with recurrent episodes of high fever, drenching sweats, frontal and occipital headaches, body, chest and abdominal pains and anorexia. Physical examination reveals splenomegaly and variable hepatomegaly. Resolution is slow and may take weeks or months.120,121 Although hepatic involvement is not a common presentation, symptoms of liver disease were noted in 40 of 82 patients in one series.122 The exact nature of the symptoms was not detailed, but these 40 patients had mild abnormalities in their serum transaminases and alkaline phosphatase. In addition, 65% of them had hepatomegaly. The main finding on liver biopsy was a non-specific reactive hepatitis. Small non-necrotizing granulomas (Figure 38-5) were noted in 28 patients. The development of granulomas appears to be an early event, based on the observation that of the 28 patients with granulomas only three had had symptoms for longer than 100 days. As these granulomas are smaller than those usually associated with sarcoidosis or tuberculosis, their appearance may suggest this diagnosis in a patient being investigated for a FUO.122 Confirmation of the diagnosis is by serologic testing and, exceptionally, through blood and bone marrow culture of the organism. Antibody tests performed 2 weeks apart show an increase in the antibody titers.121 Treatment is difficult. The WHO recommends a combination of doxycycline 200 mg and rifampin 600–800 mg daily for 6 weeks, and the efficacy of this treatment has been substantiated in more recent studies.123,124
CAT-SCRATCH DISEASE Cat scratch disease is caused by Bartonella henselae.125 The classic presentation is with lymphadenopathy associated with mild general symptoms such as malaise, aches and pains, and abdominal pain. Directed questioning will elicit a history of contact with a kitten (a kitten is defined as an animal younger than 1 year old). The organism is introduced into the skin through a scratch or wound and produces a papule at the site of inoculation within 3–5 days. Lymphadenopathy develops in the region draining that location and
is noted 2 weeks later.126 On histopathologic examination the lymph nodes show stellate necrotizing granulomas. The organisms can be identified on a Warthin–Starry stain. The disease appears to be seasonal, preferentially occurring in the second half of the year. Occasionally the disease presents with hepatic involvement in the absence of peripheral lymphadenopathy.127–131 This presentation has mainly been described in children who develop fever, malaise and abdominal pain. The reason why some children develop visceral manifestations is unclear. In one case report, the child had a persistent deficiency of T lymphocytes, suggesting an underlying immunologic abnormality, but this was not further elucidated.132 Blood tests show mild but non-specific abnormalities in liver tests. A CT scan may show filling defects suspicious for a malignant neoplasm; however, on biopsy the necrotizing granulomas and organism are identified. Retrospectively a history of contact with a kitten and a positive cat-scratch skin test confirm the diagnosis. The symptoms and findings resolve spontaneously, though when severe they can be treated with rifampicin, ciprofloxacillin or gentamicin.133
DRUG-INDUCED HEPATITIS For a drug to be considered responsible for granulomas within the liver, a number of criteria need to be fulfilled. First, other causes such as those listed above should be excluded. Additionally, a time course demonstrating that the symptoms leading to the liver biopsy developed after the start of therapy must be established. Lastly, both symptoms/liver test abnormalities and liver biopsy findings should resolve after stopping the drug, and should return on rechallenge. Obviously it is not always possible to establish cause and effect in this fashion, although it has been done for drugs such as allopurinol,134–136 carbamazepine,137–140 chlorpropamide,141 phenylbutazone,142 methyldopa147 and quinidine.143–146 Other drugs that have been implicated in the development of granulomas in the liver include hydralazine,148 diltiazem,149,150 phenytoin,151 mebendazole,152 gold,153 quinine,154,155 halothane,156 amoxicillin–clavulinic acid,157 aluminum,158 pyrazinamide,159 paracetamol,160 gliburide,161 162 163 mesalamine and rosiglitazone (Table 38-3). Both gold and aluTable 38-3. Some Drugs Associated with Granulomas in the Liver Drug
Use 135,136
Allopurinol Carbamazepine137,138 Gliburide161 Chlorpropamide141 Phenylbutazone142 Quinidine143–146 Methyldopa147 Hydralazine148 Diltiazem149,150 Phenytoin151 Mebendazole152 Quinine154,155 Halothane156 Amoxycillin–clavulinic acid157 Pyrazinamide159 Paracetamol160 Mesalamine162 Figure 38-5. Small poorly formed granuloma in patient with brucellosis. (Hematoxylin and eosin, ¥460)
774
Rosiglitazone
Antihyperuricemic Anticonvulsant Oral hypoglycemic Oral hypoglycemic Anti-inflammatory Antiarrhythmic Antihypertensive Antihypertensive Calcium channel blocker Anticonvulsant Antihelminthic Skeletal muscle relaxant Anesthetic gas Antimicrobial Antituberculous agent Analgesic/antipyretic Anti-inflammatory (sulfasalazine derivative) Oral hypoglycemic163
Chapter 38 GRANULOMATOUS DISEASES OF THE LIVER
minum have been demonstrated in granulomas using spectroscopic methods. The granulomas are usually seen on a background of lobular hepatitis and or portal triaditis. Cholangitis is a more variable finding, and eosinophils may or may not be a prominent component of the inflammatory infiltrate. Patients usually present with symptoms such as fever, arthralgias, nausea and vomiting within a month of starting the new medication, although symptoms may develop a few months later. Physical examination may reveal jaundice and hepatomegaly, and liver tests show abnormalities in transaminases and alkaline phosphatase or isolated elevations of alkaline phosphatase. Hyperbilirubinemia is an inconstant finding and peripheral blood eosinophilia is not seen in all patients, but when present is suggestive of drug-induced hepatitis.
GRANULOMATOUS HEPATITIS, NOS After all the more common causes of fever or abnormalities in liver tests resulting in granulomatous inflammation in the liver have been excluded, there remain a group of patients who have granulomatous hepatitis of unknown etiology. Although the etiology is unclear, most of the literature does not specifically address the question of an autoimmune cause.164–166, 168–170 Many of these patients present with a high fever that is often associated with chills, rigors and significant weight loss.164–166,169 The duration of symptoms varies from months to years. Liver biopsy is performed in the work-up for fever of unknown origin and abnormal liver tests. These patients do not respond to empiric antituberculous therapy, which supports the belief this is not due to tuberculosis. In some patients there is spontaneous resolution of symptoms, whereas others require treatment with steroids, to which they show a dramatic response. Although some are cured with short-term steroids, others must be maintained on medication for long periods otherwise they will relapse when it is withdrawn. The duration of steroid treatment depends on the response to withdrawal and can vary from months to years. In those patients who have undergone sequential liver biopsies, the granulomas disappear on therapy and reappear when the patient again becomes symptomatic.165,166 Treatment, in patients who do not recover spontaneously, consists of prednisone 0.75–1.0 mg/kg/day.169 The mean duration of therapy in one study was approximately 1 year. A prophylactic course of antituberculous medication is advised, particularly if the Mantoux test is positive or there is an unexplained anergic response.165,169 In one child who presented with high fever and a palpable liver the ultrasound showed abnormal findings, leading to a CT scan of the liver. This showed multiple focal lesions suspicious for malignancy. The biopsy showed necrotizing granulomatous inflammation which responded to indomethacin after cultures and acid-fast stains had ruled out other causes.168 Methotrexate has been used in patients who are steroid resistant.171 The GLUS syndrome, or granulomatous lesions of unknown significance, describes patients who present with fever, abdominal pain and weight loss, and who are found to have multiorgan granulomatous inflammation that appears to be primarily subdiaphragmatic, involving the liver, spleen and lymph nodes. These granulomas differ from sarcoidal granulomas in that they are rich in B rather than T lymphocytes. Because none of the other studies discussed under idiopathic granulomatous hepatitis have examined the phenotype of the inflammatory cells within the granulomas it is possible that
all these entities form a spectrum of granulomatous hepatitis of unknown etiology.172
INCIDENTAL GRANULOMAS LYMPHOMA Hepatic granulomas have been identified in patients with both Hodgkin’s and non-Hodgkin’s lymphoma. They are usually discovered as a part of the staging procedure and not because of liver dysfunction.173–175 The presence of granulomas is not synonymous with involvement by lymphoma, though it warrants a careful examination of the lymphoid cells intermixed in the granuloma and within the adjacent parenchyma. In one study more than half of the liver biopsies examined as a part of the staging protocol for Hodgkin’s disease showed granulomas in the absence of hepatic involvement by Hodgkin’s disease.173 In another report, a hypodense lesion was noted on a CT scan of the liver in a patient with Hodgkin’s disease. The liver biopsy showed the lesion to be composed of a necrotizing granulomatous reaction. An abnormal lymphoid infiltrate was also noted in the adjacent portal tracts within the biopsy. No microorganisms were identified and the lesion resolved with treatment of the lymphoma.174 Examples of both B- and T-cell lymphomas have been described in which involvement of the liver by the non-Hodgkin’s lymphoma has been associated with a granulomatous reaction, either separate from175 or intimately admixed with the abnormal lymphoid infiltrate.176 It is hypothesized that this reaction may represent a host response to the lymphoma.
LEPROSY Leprosy is rare in the US, but may be seen in immigrants from neighboring countries.177 After nerves, skin and lymph nodes, the liver is the most commonly involved organ.178,179 Granulomatous involvement of the liver (Figure 38-6) occurs in all forms of leprosy
Figure 38-6. Granulomata composed of foam cells in a patient with lepromatous leprosy. These foam cells would be filled with acid-fast bacilli on an acidfast bacillus stain. (Hematoxylin and eosin, ¥92)
775
Section V. Liver Diseases Due to Infectious Agents
Figure 38-7. Coccidiodes (arrow) in hepatic granuloma. (Hematoxylin and eosin, ¥92)
(excluding neuritic) but is not associated with significant abnormalities of liver tests or clinical symptoms.178 Epithelioid granulomas are seen in tuberculoid leprosy and the lesions contain occasional acidfast organisms, in contrast to lepromatous leprosy in which granulomas are composed of foamy histiocytes laden with acid-fast bacilli.177,179 This latter pattern is unusual in M. tuberculosis, but may be seen in M. avium intracellulare infection. Presentation with PUO is uncommon except in reactional states. In this situation, the diagnosis is likely to be misdiagnosed as tuberculosis unless clues such as the macular skin lesion and hair loss are noted, leading to skin testing for acid-fast organisms.177
whether the patient is immunocompetent or immunocompromised, and whether they come from a developed or an underdeveloped country. A non-necrotizing granuloma found in a biopsy from an immunocompetent individual that is being graded and staged for HCV infection is likely to be incidental, non-infectious, and possibly related to interferon-a therapy. On the other hand, a nonnecrotizing granuloma found in the work-up for hepatitis of unknown etiology may point to drug-induced hepatitis. The value of performing acid-fast and fungal stains in these biopsies is minimal. Necrotizing granulomas are most often due to an infectious etiology, irrespective of whether organisms are found or not.
MISCELLANEOUS Case reports and studies of miscellaneous causes and associations of granulomas in the liver include listeriosis,180,181 candidiasis in leukemic patients,182 histoplasmosis,183 coccidioidomycosis (Figure 38-7),184 cryptococcosis,185 blastomycosis,186 toxoplasmosis,187 cytomegalovirus infection,188 giant cell arteritis,189 and hepatocellular neoplasms.190 Listerial and candidal granulomas are necrotizing180–182 and associated with a neutrophilic inflammatory infiltrate. The different fungi and toxoplasmosis were identified within the lesions, in contrast to listeriosis and cytomegalovirus infection in which the organisms were not noted either within the granulomas or in the surrounding hepatic parenchyma. Mycotic granulomas are usually a part of disseminated infection and develop mainly in patients from endemic regions. Mycotic granulomas may be seen in both immunocompetent and immunocompromised patients. Hepatic histoplasmosis is only associated with granulomas in 19% of cases.183 In conclusion, the clues to determining the significance of a hepatic granuloma are found in the indication for the biopsy,
776
REFERENCES 1. Klatskin G. Hepatic granulomata. Problems in interpretation. Mt Sinai J Med 1977;44:798–812. 2. Denk H, Scheuer PJ, Baptista A, et al. Guidelines for the diagnosis and interpretation of hepatic granulomas. Histopathology 1994;25:209–218. 3. Lucey DR, Clerici M, Shearer GM. Type 1 and type 2 cytokine disregulation in human infectious, neoplastic, and inflammatory diseases. Clin Microbiol Rev 1996;9:532–562. 4. Baumer I, Zissel G, Schlaak M, et al. Th1/Th2 cell distribution in pulmonary sarcoidosis. Am J Respir Cell Mol Biol 1997;16:171–177. 5. Agostini C, Basso U, Semenzato G. Cells and molecules involved in the development of sarcoid granuloma. J Clin Immunol 1998;18:184–192. 6. Infante-Duarte C, Kamradt T. Th1/Th2 balance in infection. Springer Semin Immunopathol 1999;21:317–338. 7. Mitchison NA, Schuhbauer D, Muller B. Natural and induced regulation of Th1/Th2 balance. Springer Semin Immunopathol 1999;21:199–210.
Chapter 38 GRANULOMATOUS DISEASES OF THE LIVER
8. Chensue SW, Ruth JH, Warmington K, et al. In vivo regulation of macrophage IL-12 production during type 1 and type 2 cytokine-mediated granuloma formation. J Immunol 1995;155:3546–3551. 9. Chensue SW, Warmington K, Ruth J, et al. Cytokine responses during mycobacterial and schistosomal antigen-induced pulmonary granuloma formation. Production of Th1 and Th2 cytokines and relative contribution of tumor necrosis factor. Am J Pathol 1994;145:1105–1113. 10. Wynn TA, Cheever AW. Cytokine regulation of granuloma formation in schistosomiasis. Curr Opin Immunol 1995;7:505–511. 11. Chensue SW, Warmington KS, Hershey SD, et al. Evolving T cell responses in murine schistosomiasis. Th2 cells mediate secondary granulomatous hypersensitivity and are regulated by CD8+ T cells in vivo. J Immunol 1993;151:1391–1400. 12. Chensue SW, Warmington K, Ruth JH, et al. Effect of slow release IL-12 and IL-10 on inflammation, local macrophage function and the regional lymphoid response during mycobacterial (Th1) and schistosomal (Th2) antigen-elicited pulmonary granuloma formation. Inflamm Res 1997;46:86–92. 13. Amiri P, Locksley RM, Parslow TG, et al. Tumor necrosis factor a restores granulomas and induces parasite egg-laying in schistosome-infected SCID mice. Nature 1992;356:604–607. 14. Molos MA, Litton N, Schubert TT. Talc liver. J Clin Gastroenterol 1987;9:198–203. 15. Allaire GS, Goodman ZD, Ishak KG, et al. Talc in liver tissue of intravenous drug abusers with chronic hepatitis. A comparative study. Am J Clin Pathol 1989;92:583–588. 16. Mitchell DP, Hanes TE, Hoyumpa AM, et al. Fever of unknown origin. Assessment of the value of percutaneous liver biopsy. Arch Intern Med 1977;137:1001–1004. 17. Masana L, Guardia J, Clotet B, et al. Liver biopsy in fever of unknown origin. Gastroenterol Clin Biol 1980;4:215–218. 18. Holtz T, Moseley RH, Scheiman JM. Liver biopsy in fever of unknown origin. A reappraisal. J Clin Gastroenterol 1993;17:29–32. 19. Guckian JC, Perry JE. Granulomatous hepatitis. An analysis of 63 cases and review of the literature. Ann Intern Med 1966;65:1081–1100. 20. Abdi WA, Millan JC, Mezey E. Sampling variability on percutaneous liver biopsy. Arch Intern Med 1979;139: 667–669. 21. Maharaj B, Maharaj RJ, Leary WP, et al. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancet 1986;8480:523–525. 22. Dincsoy HP, Weesner RE, MacGee J. Lipogranulomas in nonfatty human livers. A mineral oil induced environmental disease. Am J Clin Pathol 1982;78:35–41. 23. Wagoner GP, Anton AT, Gall EA, et al. Needle biopsy of the liver. VIII. Experiences with hepatic granulomas. Gastroenterology 1953;25:487–494. 24. Iversen K, Christoffersen P, Poulsen H. Epithelioid cell granulomas in liver biopsies. Scand J Gastroenterol 1970;7(Suppl):61–67. 25. Mir-Madjilessi SH, Farmer RG, Hawk WA. Spectrum of hepatic manifestations of granulomatous hepatitis of unknown etiology. Am J Gastroenterol 1974;62:221–229. 26. Irani SK, Dobbins III WO. Hepatic granulomas: A review of 73 patients from one hospital and a survey of the literature. J Clin Gastroenterol 1979;1:131–143. 27. Anderson CS, Nicholls N, Rowland R, et al. Hepatic granulomas: a 15-year experience in the Royal Adelaide Hospital. Med J Aust 1988;148:71–74. 28. Satti MB, Al-Freihi H, Ibrahim EM, et al. Hepatic granuloma in Saudia Arabia: A clinicopathological study of 59 cases. Am J Gastroenterol 1990;85:669–674.
29. McCluggage WG, Sloan JM. Hepatic granulomas in Northern Ireland: a thirteen year review. Histopathology 1994;25: 219–228. 30. Sabharwal BD, Malhotra N, Garg R, et al. Granulomatous hepatitis: a retrospective study. Indian J Pathol Microbiol 1995;38:413–416. 31. Barcena R, Sanroman AL, Del Campo S, et al. Posttransplant liver granulomatosis associated with hepatitis C? Transplantation 1998;65:494–1495. 32. Yamamoto S, Iguchi Y, Ohomoto K, et al. Epithelioid granuloma formation in type C chronic hepatitis: report of two cases. Hepatogastroenterology 1995;42:291–293. 33. Emile JF, Sebagh M, Feray C, et al. The presence of epithelioid granulomas in hepatitis C virus related cirrhosis. Hum Pathol 1993: 24:1095–1097. 34. Goldin RD, Levine TS, Foster GR, Thomas HC. Granulomas and hepatitis C. Histopathology 1996;28:265–267. 35. Propst A, Propst T, Dietz O, et al. Development of granulomatous hepatitis during treatment with interferon-a2b. [Letter] Dig Dis Sci 1995;40:2117–2118. 36. Veerabagu MP, Finkelstein SD, Rabinovitz M. Granulomatous hepatitis in a patient with chronic hepatitis C treated with interferon-a. Dig Dis Sci 1997;42:1445–1448. 37. Bonnet F, Morlat P, Dubuc S, et al. Sarcoidosis-associated hepatitis C virus infection. Dig Dis Sci 2002;47:794–796. 38. Hoffman RM, Jung M-C, Motz R, et al. Sarcoidosis associated with interferon-a therapy for chronic hepatitis C. J Hepatol 1998;28:1058–1063. 39. Kaplan MM. Primary biliary cirrhosis. N Engl J Med 1996;335: 1570–1580. 40. Portmann BC, MacSween RNM. Diseases of the intrahepatic bile ducts. In: MacSween RNM, Antony PP, Scheuer PJ, et al., eds. Pathology of the liver, 3rd edn. Edinburgh: Churchill Livingstone, 1994: 477. 41. Comer GM, Mukherjee S, Scholes JV, et al. Liver biopsies in the acquired immune deficiency syndrome: influence of endemic disease and drug abuse. Am J Gastroenterol 1989;84: 1525–1531. 42. Prego V, Glatt AE, Roy V, et al. Comparative yield of blood culture for fungi and mycobacteria, liver biopsy, and bone marrow biopsy in the diagnosis of fevers of undetermined origin in human immunodeficiency virus-infected patients. Arch Intern Med 1990;150:333–336. 43. Bach N, Theise ND, Schaffner F. Hepatic histopathology in the acquired immunodeficiency syndrome. Semin Liver Dis 1992;12:205–212. 44. Wilkins MJ, Lindley R, Dourakis SP, et al. Surgical pathology of the liver in HIV infection. Histopathology 1991;18:459–464. 45. Orenstein MS, Tavitian A, Yonk B, et al. Granulomatous involvement of the liver in patients with AIDS. Gut 1985;26:1220–1225. 46. Lebovics E, Thung SN, Schaffner F, et al. The liver in the acquired immunodeficiency syndrome: a clinical and histologic study. Hepatology 1985;5:293–298. 47. Cucin RL, Coleman M, Eckardt JJ, et al. The diagnosis of miliary tuberculosis: utility of peripheral blood abnormalities, bone marrow and liver needle biopsy. J Chronic Dis 1973;26: 355–361. 48. Parkes SA, Baker SB de C, Bourdillon RE, et al. Incidence of sarcoidosis in the Isle of Man. Thorax 1985;40:284–287. 49. Parkes SA, Baker SB de C, Bourdillon RE, et al. Epidemiology of sarcoidosis in the Isle of Man – 1: A case-controlled study. Thorax 1987;42:420–426. 50. Hills SE, Parkes SA, Baker SB de C. Epidemiology of sarcoidosis in the Isle of Man – 2: Evidence for space time clustering. Thorax 1987;42:427–430. 51. Mandel J, Weinberger SE. Clinical insights and basic science correlates in sarcoidosis. Am J Med Sci 2001;1:99–107.
777
Section V. Liver Diseases Due to Infectious Agents
52. Newman LS, Rose CS, Maier LA. Sarcoidosis. N Engl J Med 1997;336:1224–1234. 53. Statement on Sarcoidosis. The joint statement of the American Thoracic Society (ATS), the European Respiratory Society (ERS) and the World Association of Sarcoidosis and other granulomatous disorders (WASOG). Am J Respir Crit Care Med 1999;160:736–755. 54. Maddrey WC, Johns CJ, Boitnott JK, et al. Sarcoidosis and chronic hepatic disease: a clinical and pathologic study of 20 patients. Medicine 1970;49:375–395. 55. Rudzki C, Ishak KG, Zimmerman HJ. Chronic intrahepatic cholestasis of sarcoidosis. Am J Med 1975;59:373–387. 56. Nakanuma Y, Ohta G, Yamazaki Y, et al. Intrahepatic bile duct destruction in a patient with sarcoidosis and chronic intrahepatic cholestasis. Acta Pathol Jpn 1979;29:211–219. 57. Lehmuskallio E, Hannuksela M, Halme H. The liver in sarcoidosis. Acta Med Scand 1977;202:289–293. 58. Valla D, Pessegueiro-Miranda H, Degott C, et al. Hepatic sarcoidosis with portal hypertension. A report of seven cases with a review of the literature. Q J Med 1987;63:531–544. 59. Ishak KG. Sarcoidosis of the liver and bile ducts. Mayo Clin Proc 1998;73:467–472. 60. Hercules HC, Bethlem NM. Value of liver biopsy in sarcoidosis. Arch Pathol Lab Med 1984;108:831–834. 61. Rasmussen SM, Neukirch F. Sarcoidosis. A clinical study with special reference to the choice of biopsy procedure. Acta Med Scand 1976;199:209–216. 62. Devaney K, Goodman ZC, Epstein MS, et al. Hepatic sarcoidosis. Clinicopathologic features in 100 patients. Am J Surg Pathol 1993;17:1272–1280. 63. Tekeste H, Latour F, Levitt RE. Portal hypertension complicating sarcoid liver disease: case report and review of the literature. Am J Gastroenterol 1984;79:389–396. 64. Moreno-Merlo F, Wanless IR, Shimamatsu K, et al. The role of granulomatous phlebitis and thrombosis in the pathogenesis of cirrhosis and portal hypertension in sarcoidosis. Hepatology 1997;26:554–560. 65. Nakanuma Y, Kouda W, Harada K, et al. Hepatic sarcoidosis with vanishing bile duct syndrome, cirrhosis, and portal phlebosclerosis. Report of an autopsy case. J Clin Gastroenterol 2001;32:181–184. 66. Alam I, Levenson SD, Ferrell LD, et al. Diffuse intrahepatic biliary strictures in sarcoidosis resembling sclerosing cholangitis. Case report and review of the literature. Dig Dis Sci 1997;42: 1295–1301. 67. Israel HL, Goldstein RA. Hepatic granulomatosis and sarcoidosis. Ann Intern Med 1973;79:669–678. 68. Murphy JR, Sjogren MH, Kikendall JW, et al. Small bile duct abnormalities in sarcoidosis. J Clin Gastroenterol 1990;12: 555–561. 69. Becheur H, Dall’Osto H, Chatellier G, et al. Effect of ursodeoxhycholic acid on chronic intrahepatic cholestasis due to sarcoidosis. Dig Dis Sci 1997;42:789–791. 70. Baratta L, Cascino A, Delfino M, et al. Ursodeoxycholic acid treatment in abdominal sarcoidosis. Dig Dis Sci 2000;45: 1559–1562. 71. Casavilla FA, Gordon R, Wright HI, et al. Clinical course after liver transplantation in patients with sarcoidosis. Ann Intern Med 1993;118:865–866. 72. Fidler HM, Hadziyannis SJ, Dhillon AP, et al. Recurrent hepatic sarcoidosis following liver transplantation. Transplant Proc 1997;29:2509–2510. 73. Hunt J, Gordon FD, Jenkins RL, et al. Sarcoidosis with selective involvement of a second liver allograft: report of a case and review of the literature. Mod Pathol 1999;12:325–328. 74. Wiest PM, Olds GR. Clinical schistosomiasis. R I Med J 1992;75:179–185.
778
75. Second Report of the WHO Expert Committee. The control of schistosomiasis. WHO Technical Report Series 1993;830:1–87. 76. Elliot DE. Schistosomiasis. Pathophysiology, diagnosis and treatment. (Review). Gastroenterol Clin North Am 1996;25:599–625. 77. Bica I, Hamer DH, Stadecker MJ. Hepatic schistosomiasis. Infect Dis Clin North Am 2000;3:583–604. 78. McKerrow JH. Cytokine induction and exploitation in Schistosome infections. Parasitology 1997;115:S107–S112. 79. Boros DL. T helper cell populations, cytokine dynamics, and pathology of the schistosome egg granuloma. Microbes Infect 1999;1:511–516. 80. Stadecker MJ. The development of granulomas in schistosomiasis: genetic backgrounds, regulatory pathways, and specific egg antigen responses that influence the magnitude of disease. Microbes Infect 1999;1:505–510. 81. Cheever AW, Andrade ZA. Pathological lesions associated with Schistosoma mansoni infection in man. Trans Roy Soc Trop Med Hyg 1967;61:626–639. 82. Edington GM, Lichtenberg F von, Nwabuebo I, et al. Pathologic effects of schistosomiasis in Ibadan, Western State of Nigeria. I. Incidence and intensity of infection; distribution and severity of lesions. Am J Trop Med Hyg 1970;19:982–995. 83. Cheever AW. Pipe-stem fibrosis of the liver. Trans Roy Soc Trop Med Hyg 1972;66:947–948. 84. Bhagwandeen SB. Bilharzial pipe-stem portal fibrosis in Zambia. J Pathol 1973;111:23–30. 85. Marquet S, Abel L, Hillaire D, et al. Genetic localization of a locus controlling the intensity of infection by Schistosoma mansoni on chromosome 5q31–q33. Nature Genet 1996;14: 181–184. 86. Dessein AJ, Hillaire D, Eldin N, et al. Severe hepatic fibrosis in Schistosoma mansoni infection is controlled by a major locus that is closely linked to the interferon-g receptor gene. Am J Hum Genet 1999 ;65:709–721. 87. Chevillard C, Moukoko CE, Elwali N-E MA, et al. INF-g polymorphisms (INF-g +2109 and INF-g +3810) are associated with severe hepatic fibrosis in human hepatic schistosomiasis (Schistosoma mansoni). J Immunol 2003;171:5596–5601. 88. Mwatha JK, Kimani G, Kamau T, et al. High levels of TNF, soluble TNF receptors, soluble ICAM-1 and IFN-g, but low levels of IL-5, are associated with hepatosplenic disease in human schistosomiasis mansoni. J Immunol 1998;160: 1992–1999. 89. Tsang VC, Wilkins PP. Immunodiagnosis of schistosomiasis. Immunol Invest 1997;26:175–188. 90. Essop AR, Posen JA, Hodkinson JH, et al. Tuberculosis hepatitis: a clinical review of 96 cases. Q J Med 1984;53:465–477. 91. Bowry S, Chan CH, Weiss H, et al. Hepatic involvement in pulmonary tuberculosis. Histologic and functional characteristics. Am Rev Respir Dis 1970;101:941–948. 92. Diaz ML, Herrera T, Lopez-Vidal Y, et al. Polymerase chain reaction for the detection of Mycobacterium tuberculosis DNA in tissue and assessment of its utility in the diagnosis of hepatic granulomas. J Lab Clin Med 1996;127:359–363. 93. Maschek H, Georgii A, Schmidt RE, et al. Mycobacterium genavense. Autopsy findings in three patients. Am J Clin Pathol 1994;101:95–99. 94. Farhi DC, Mason UG 3d, Horsburgh CR Jr. Pathologic findings in disseminated Mycobacterium avium-intracellulare infection. A report of 11 cases. Am J Clin Pathol 1986;85:67–72. 95. Patel KM. Granulomatous hepatitis due to Mycobacterium scrofulaceum: report of a case. Gastroenterology 1981;81: 156–158. 96. Lamm DL, Mander Meijden ADPM, Morales A, et al. Incidence and treatment of complications of Bacillus Calmette–Guerin intravesical therapy in superficial bladder cancer. J Urol 1992;147:596–600.
Chapter 38 GRANULOMATOUS DISEASES OF THE LIVER
97. Bodurtha A, Kim YH, Laucius JF, et al. Hepatic granulomas and other hepatic lesions associated with BCG immunotherapy for cancer. Am J Clin Pathol 1974;61:747–752. 98. Proctor DD, Chopra S, Rubenstein SC, et al. Mycobacteremia and granulomatous hepatitis following initial intravesical Bacillus Calmette–Guerin instillation for bladder carcinoma. Am J Gastroenterol 1993;88:1112–1115. 99. McParland C, Cotton DJ, Gowda KS, et al. Miliary Mycobacterium bovis induced by the intravesical Bacillus Calmette–Guerin immunotherapy. Am Rev Respir Dis 1992;146:1330–1333. 100. Maurin M, Raoult D. Q fever. Clin Microbiol Rev 1999;12: 518–553. 101. Anon. Q fever among slaughterhouse workers – California. MMWR Morb Mortal Wkly Rep 1986;35:223–226. 102. Hofmann CE, Heaton JW Jr. Q fever hepatitis: clinical manifestations and pathological findings. Gastroenterology 1982;83:474–479. 103. Turck WPG, Howitt G, Turnberg LA, et al. Chronic Q fever. Q J Med 1976;45:193–217. 104. Powell OW. Liver involvement in ‘Q’ fever. Australas Ann Med 1961;10:52–58. 105. Pellegrin M, Delsol G, Auvergnat JC, et al. Granulomatous hepatitis in Q fever. Hum Pathol 1980;11:51–57. 106. Qizilbash AH. The pathology of Q fever as seen on liver biopsy. Arch Pathol Lab Med 1983;107:369–367. 107. Travis LB, Travis WD, Li CY, et al. Q fever. A clinicopathologic study of five cases. Arch Pathol Lab Med 1986;110:1017– 1020. 108. Westlake P, Price LM, Russell M, et al. The pathology of Q fever hepatitis. A case diagnosed by liver biopsy. J Clin Gastroenterol 1987;9:357–363. 109. Atienza P, Ramond MJ, Degott C, et al. Chronic Q fever hepatitis complicated by extensive fibrosis. Gastroenterology 1988;95:478–481. 110. Ruel M, Sevestre H, Henry-Biabaud E, et al. Fibrin ring granulomas in hepatitis A. Dig Dis Sci 1992;37:1915– 1917. 111. De Bayser L, Roblot P, Ramassamy A, et al. Hepatic fibrin-ring granulomas in giant cell arteritis. Gastroenterology 1993;105:272–273. 112. Nenert M, Mavier P, Dubuc N, et al. Epstein–Barr virus infection and hepatic fibrin-ring granulomas. Hum Pathol 1988;19:608–610. 113. Lobdell DH. ‘Ring’ granulomas in cytomegalovirus hepatitis. Arch Pathol Lab Med 1987;9:357–363. 114. Murphy E, Griffiths MR, Hunter JA, et al. Fibrin-ring granulomas: a non-specific reaction to liver injury? Histopathology 1991;19:91–93. 115. Moreno A, Marazuela M, Yebra M, et al. Hepatic fibrin-ring granulomas in visceral leishmaniasis. Gastroenterology 1988;95:1123–1126. 116. Vanderstigel M, Zafrani ES, Lejonc JL, et al. Allopurinol hypersensitivity syndrome as a cause of hepatic fibrin-ring granulomas. Gastroenterology 1986;90:188–190. 117. Hunt JG, Field PR, Murphy AM. Immunoglobulin responses to Coxiella burnetii (Q fever): single-serum diagnosis of acute infection using an immunofluorescence technique. Infect Immun 1983;39:977–981. 118. Peter O, Dupuis G, Burgdorfer W, et al. Evaluation of the complement fixation and indirect immunofluorescence tests in the early diagnosis of primary Q fever. Eur J Clin Microbiol 1985;4:394–396. 119. Corbel MJ. Brucellosis: an overview. Emerg Infect Dis 1997;3:213–221. 120. Williams E. Brucellosis. Br Med J 1973;2:791–793. 121. Williams E. Brucellosis in humans: its diagnosis and treatment. APMIS 1988;3(Suppl):21–25.
122. Cervantes F, Bruguera M, Carbonell J, et al. Liver disease in brucellosis. A clinical and pathological study of 40 cases. Postgrad Med J 1982;58:346–350. 123. Joint FAO/WHO expert committee on brucellosis. 6th Report. WHO Technical Report Series 1986;740:56–58. 124. Ariza J, Gudiol F, Pallares R, et al. Treatment of human brucellosis with doxycycline plus rifampin or doxycycline plus streptomycin. Ann Intern Med 1992;117:25–30. 125. Bass JW, Vincent JM, Person DA. The expanding spectrum of Bartonella infections: II. Cat-scratch disease. Pediatr Infect Dis J 1997;16:163–79. 126. Carithers HA. Cat scratch disease. An overview based on a study of 1200 patients. Am J Dis Child 1985;139: 1124–1133. 127. Malatack JJ, Jaffe R. Granulomatous hepatitis in three children due to cat-scratch disease without peripheral adenopathy. An unrecognized cause of fever of unknown origin. Am J Dis Child 1993;147:949–953. 128. Dangman BC, Albanese BA, Kacica MA, et al. Cat scratch disease in two children presenting with fever of unknown origin: imaging features and association with a new causative agent, Rochalimaea henselae. Pediatrics 1995;95: 767–771. 129. Destuynder O, Vanlemmans P, Mboyo A, et al. Systemic cat scratch disease: hepatic and splenic involvement about 3 pediatric cases. Eur J Pediatr Surg 1995;5:365–368. 130. Lamp LW, Gray GF, Scott MA. The histologic spectrum of hepatic cat scratch disease. A series of six cases with confirmed Bartonella henselae infection. Am J Surg Pathol 1996;20: 1253–1259. 131. Rivera-Penera T, Nielsen K, Hall TR. Radiological case of the month. Granulomatous hepatitis in cat-scratch disease. Arch Pediatr Adolesc Med 1998;152:87–88. 132. Kahr A, Kerbl R, Gschwandtner K, et al. Visceral manifestation of cat scratch disease in children. A consequence of altered immunological state? Infection 2000;28:778–784. 133. Margileth AM. Antibiotic therapy for cat-scratch disease: clinical study of therapeutic outcome in 268 patients and a review of the literature. Pediatr Infect Dis J 1992;11:474–478. 134. Simmons F, Feldman B, Gerety D. Granulomatous hepatitis in a patient receiving allopurinol. Gastroenterology 1972;62: 101–104. 135. Espiritu CR, Alalu J, Glueckauf LG, et al. Allopurinol-induced granulomatous hepatitis. Am J Dig Dis 1976;21:804–806. 136. Swank LA, Chejfec G, Nemchausky BA. Allopurinol-induced granulomatous hepatitis with cholangitis and a sarcoid-like reaction. Arch Intern Med 1978;138:997–998. 137. Levander HG. Granulomatous hepatitis in a patient receiving carbamazepine. Acta Med Scand 1980;208:333–335. 138. Levy M, Goodman MW, Van Dyne BJ, et al. Granulomatous hepatitis secondary to carbamazepine. Ann Intern Med 1981;95:64–65. 139. Mitchell MC, Boitnott JK, Arregui A, et al. Granulomatous hepatitis associated with carbamazepine therapy. Am J Med 1981;71:733–735. 140. Soffer EE, Taylor RJ, Bertram PD, et al. Carbamazepine-induced liver injury. South Med J 1983;76:681–683. 141. Rigberg LA, Robinson MJ, Espiritu CR. Chlorpropamide-induced granulomas. A probable hypersensitivity reaction in liver and bone marrow. JAMA 1976;235:409–410. 142. Ishak KG, Kirchner JP, Dhar JK. Granulomas and cholestatichepatocellular injury associated with phenylbutazone. Report of two cases. Am J Dig Dis 1977;22:611–617. 143. Chajek T, Lehrer B, Geltner D, et al. Quinidine-induced granulomatous hepatitis. Ann Intern Med 1974;81:774–776. 144. Geltner D, Chajek T, Rubinger D, et al. Quinidine hypersensitivity and liver involvement. A survey of 32 patients. Gastroenterology 1976;70:650–652.
779
Section V. Liver Diseases Due to Infectious Agents
145. Bramlet DA, Posalaky Z, Olson R. Granulomatous hepatitis as a manifestation of quinidine hypersensitivity. Arch Intern Med 1980;140:395–397. 146. Knobler H, Levij IS, Gavish D, et al. Quinidine-induced hepatitis. A common and reversible hypersensitivity reaction. Arch Intern Med 1986;146:526–528. 147. Miller AC Jr., Reid WM. Methyldopa-induced granulomatous hepatitis. JAMA 1976;235:2001–2002. 148. Jori GP, Peschile C. Hydralazine disease associated with transient granulomas in the liver. A case report. Gastroenterology 1973;64:1163–1167. 149. Sarachek NS, London RL, Matulewicz TJ. Diltiazem and granulomatous hepatitis. Gastroenterology 1985;88:1260–1262. 150. Toft E, Vyberg M, Therkelsen K. Diltiazem-induced granulomatous hepatitis. Histopathology 1991;18:474–475. 151. Cook IF, Shilkin KB, Reed WD. Phenytoin induced granulomatous hepatitis. Aust NZ J Med 1981;11:539–541. 152. Colle I, Naegels S, Hoorens A, et al. Granulomatous hepatitis due to mebendazole. J Clin Gastroenterol 1999;28:44–45. 153. Landas SK, Mitros FA, Furst DE, et al. Lipogranulomas and gold in the liver in rheumatoid arthritis. Am J Surg Pathol 1992;16: 171–174. 154. Katz B, Weetch M, Chopra S. Quinine-induced granulomatous hepatitis. Br Med J 1983;286:264–265. 155. Mathur S, Dooley J, Scheuer PJ. Quinine induced granulomatous hepatitis and vasculitis. Br Med J 1990;300:613. 156. Shah IA, Brandt H. Halothane-associated granulomatous hepatitis. Digestion 1983;28:245–249. 157. Silvain C, Fort E, Levillain P, et al. Granulomatous hepatitis due to a combination of amoxicillin and clavulanic acid. Dig Dis Sci 1992;37:150–152. 158. Kurumaya H, Kono N, Nakanuma Y, et al. Hepatic granulomata in long-term hemodialysis patients with hyperaluminumemia. Arch Pathol Lab Med 1989;113:1132–1134. 159. Knobel B, Buyanowsky G, Dan M, et al. Pyrazinamide-induced granulomatous hepatitis. J. Clin Gastroenterol 1997;24: 264–266. 160. Lindgren A, Aldenborg F, Norkrans G, et al. Paracetamolinduced cholestatic and granulomatous liver injuries. J Intern Med 1997;241:435–439. 161. Saw D, Pitman E, Maung M, et al. Granulomatous hepatitis associated with gliburide. Dig Dis Sci 1996;41:322–325. 162. Braun M, Fraser GM, Kunin M, et al. Mesalamine-induced granulomatous hepatitis. Am J Gastroenterol 1999;94: 1973–1974. 163. Dhawan M, Agrawal R, Ravi J, et al. Rosiglitazone-induced granulomatous hepatitis. J Clin Gastroenterol 2002;34: 582–584. 164. Eliakim M, Eisenberg S, Levij IS, et al. Granulomatous hepatitis accompanying a self-limited febrile disease. Lancet 1968;1: 1348–1352. 165. Simon HB, Wolff SM. Granulomatous hepatitis and prolonged fever of unknown origin: a study of 13 cases. Medicine 1973;52:1–21. 166. Penchas S, Ligumski M, Eliakim M. Idiopathic granulomatous hepatitis with a prolonged course: effect of corticosteroid therapy. Digestion 1978;17:46–55. 167. Shee CD, Creamer B. Idiopathic granulomatous hepatitis and abdominal pain. Postgrad Med J 1980;56:342–343. 168. Berlin CM Jr, Boal DK, Zaino RJ, et al. Hepatic granulomata presenting with prolonged fever. Resolution with antiinflammatory treatment. Clin Pediatr 1990;29:339–342.
780
169. Zoutman DE, Ralph Ed, Frei JV. Granulomatous hepatitis and fever of unknown origin. An 11-year experience of 23 cases with three years’ follow-up. J Clin Gastroenterol 1991;13:69–75. 170. Sartin JS, Walker RC. Granulomatous hepatitis: a retrospective review of 88 cases at the Mayo Clinic. Mayo Clin Proc 1991;66:914–918. 171. Knox TA, Kaplan MM, Gelfand JA, et al. Methotrexate treatment of idiopathic granulomatous hepatitis. Ann Intern Med 1995;122:592–595. 172. Brincker H. Granulomatous lesions of unknown significance: the GLUS syndrome. In: James DG, ed. Sarcoidosis and other granulomatous disorders. New York: Marcel Dekker, 1994:69–86. 173. Kadin ME, Donaldson SS, Dorfman RF. Granulomas in Hodgkin’s disease. N Engl J Med 1970;283:859–861. 174. Johnson LN, Iseri O, Knodell RG. Caseating hepatic granulomas in Hodgkin’s lymphoma. Gastroenterology 1990;99:1837–1840. 175. Braylan RC, Long JC, Jaffe ES, et al. Malignant lymphoma obscured by concomitant extensive epithelioid granulomas: report of three cases with similar clinicopathologic features. Cancer 1977;39:1146–1155. 176. Saito K, Nakanuma Y, Ogawa S, et al. Extensive hepatic granulomas associated with peripheral T cell lymphoma. Am J Gastroenterol 1991;86:1243–1246. 177. Weissman JB, Neu HC. Lepromatous leprosy masquerading as disseminated tuberculosis. Am J Med 1979;67:113–116. 178. Karat ABA, Job CK, Rao PSS. Liver in leprosy. Histologic and biochemical findings. Br Med J 1971;1:307–310. 179. Chen TS, Drutz DJ, Whelan GE. Hepatic granulomas in leprosy. Their relation to bacteremia. Arch Pathol Lab Med 1976;100:182–185. 180. De Vega T, Echevarria S, Crespo J, et al. Acute hepatitis by Listeria monocytogenes in an HIV patient with chronic HBV hepatitis. J Clin Gastroenterol 1992;15:251–255. 181. Henderson JR, Ramsey CA. Miliary granulomata in a fatal adult case of listerial meningitis. Postgrad Med J 1967;43:794–796. 182. Jones JM. Granulomatous hepatitis due to Candida albicans in patients with acute leukemia. Ann Intern Med 1981;94: 475–477. 183. Lamps LW, Molina CP, West AB, et al. The pathologic spectrum of gastrointestinal and hepatic histoplasmosis. Am J Clin Pathol 2000;113:64–72. 184. Howard PF, Swith JW. Diagnosis of disseminated coccidiomycosis by liver biopsy. Arch Intern Med 1983;143: 1335–1338. 185. Lefton HB, Farmer RG, Buchwald R, et al. Cryptococcal hepatitis mimicking primary sclerosing cholangitis. A case report. Gastroenterology 1974;67:511–515. 186. Teixeira F, Gayotto LC, De Brito T. Morphological patterns of the liver in South American blastomycosis. Histopathology 1978;2:231–237. 187. Weitberg AB, Alper JC, Diamond I, et al. Acute granulomatous hepatitis in the course of acquired toxoplasmosis. N Engl J Med 1979;300:1093–1096. 188. Clarke J, Craig RM, Saffro R, et al. Cytomegalovirus granulomatous hepatitis. Am J Med 1979;66:264–269. 189. Litwack KD, Bohan A, Silverman L. Granulomatous liver disease and giant cell arteritis. Case report and literature review. J Rheumatol 1977;4:307–312. 190. Malatjalian DA, Graham CH. Liver adenoma with granulomas. The appearance of granulomas in oral contraceptive-related hepatocellular adenoma and in the surrounding nontumorous liver. Arch Pathol Lab Med 1982;106:244–246.
Section V: Liver Diseases Due to Infectious Agents
39
HIV-ASSOCIATED HEPATOBILIARY DISEASE C. Mel Wilcox, Miguel R. Arguedas Abbreviations AIDS acquired immunodeficiency virus Anti-HBs antihepatitis antibody CMV cytomegalovirus CT computed tomography DNA deoxyribonucleic acid HAART highly active antiretroviral therapy HAV hepatitis A virus
HBsag HBV HCV HEV HIV IVDU
hepatitis B surface antigen hepatitis B virus hepatitis C virus hepatitis E virus human immunodeficiency virus intravenous drug use
INTRODUCTION A major breakthrough in the treatment of human immunodeficiency virus (HIV) infection occurred in 1996 when the protease inhibitors became widely available. These drugs, when combined with other antiretroviral agents termed highly active antiretroviral therapy (HAART), profoundly suppressed HIV replication and consequently led to immune reconstitution. Because of HAART, there has been a marked reduction in the incidence of opportunistic infections and neoplasms and improved survival for HIV-infected patients, including those with AIDS.1,2 With this fall in opportunistic processes and increased life expectancy, there has also been a corresponding shift to the management of chronic diseases as well as drug-induced complications.3 One such chronic disease is hepatitisC virus (HCV) infection, which is prevalent in HIV-infected patients because of a common exposure. Indeed, with the remarkable changes observed with HAART, liver disease has assumed increasing importance, as evidenced by reports demonstrating that end-stage liver disease, most often due to HCV, is the leading cause of death, and liver-related complications are the most common reason for hospitalization in HIV-infected patients.4,5 A thorough understanding of the causes and management of hepatobiliary disease in HIV infection and AIDS remains pertinent because of the number of infected patients throughout the world, and because of the rising morbidity and mortality from liver disease. To review hepatobiliary disease associated with HIV infection, this chapter will address the causes of liver disease, the appropriate evaluation and treatment for these disorders, as well as outline a diagnostic strategy to evaluate suspected hepatobiliary disease. The impact of HAART on these topics will also be highlighted.
CAUSES OF LIVER DISEASE The spectrum of hepatobiliary diseases associated with HIV infection has been well characterized (Table 39-1). Several general observations can be made. First, opportunistic infections are common
KS MAC NHL PCR RNA TB
Kaposi’s sarcoma Mycobacterium avium complex Non-Hodgkin’s lymphoma polymerase chain reaction ribonucleic acid tuberculosis
causes of parenchymal liver disease in AIDS. In aggregate Mycobacteria avium complex (MAC) is the most frequently identified opportunistic cause of liver disease, followed by cryptococcus and CMV; autopsy studies often have higher frequencies of identification of CMV. Although uncommon, non-Hodgkin’s lymphoma (NHL) is the most frequent symptomatic hepatic neoplasm, whereas Kaposi’s sarcoma (KS) is the most common hepatic neoplasm found in autopsy studies. Second, the identified cause(s) of disease vary between studies because of differences in patient populations, including geographic setting, risk factors for HIV infection, severity of immunodeficiency, and indications for and method of acquiring liver tissue. For example, studies of intravenous drug users (IVDU) show a high prevalence of HCV, whereas reports from underdeveloped countries uniformly report higher prevalence rates of tuberculosis (TB).6 Studies that include many patients with only modest immunodeficiency (CD4 lymphocyte count 200–500/mm3) would not be expected to identify MAC or cytomegalovirus (CMV) disease, given the relationship between opportunistic infections and absolute CD4 count.1 The method by which liver tissue is acquired will also dictate the spectrum of identified causes. For example, patients undergoing percutaneous liver biopsy are more likely to have an acute presentation and a specific diagnosis established, whereas autopsy studies identify disorders that were often clinically silent.6
PATHOLOGIC FINDINGS Autopsy and liver biopsy studies have shown that liver histopathology is rarely normal in HIV-infected patients. The most common finding is steatosis, which is usually macrovesicular and clinically unimportant except as a cause of hepatomegaly and abnormal liver tests. In most cases steatosis is related to malnutrition, with other potential causes including alcohol abuse, total parenteral nutrition, and HCV. A drug-induced syndrome of severe microvesicular steatosis associated with lactic acidosis, hepatic failure and potential death caused by specific antiretroviral agents is now well recognized (see below).
781
Section V. Liver Diseases Due to Infectious Agents
Table 39-1. Reported Causes of Hepatic Disease in Patients with HIV Infection Viral
Protozoal
Neoplastic
Hepatitis AÆE, G Cytomegalovirus Epstein–Barr Herpes simplex type 2 Varicella
Cryptosporidia Microsporidia Isospora Toxoplasma Sporothrix Malaria
Lymphoma Kaposi’s sarcoma Hepatoma Metastatic tumors
Mycobacterial
Bacterial
Drug-induced
M. avium complex M. tuberculosis Other mycobacteria
Bartonellla henselae Pneumocystis carinii
Sulfa Antiretrovirals Antifungals Others
A
Fungal Cryptococcus Histoplasma Coccidioides Penicillium marneffie Candida
Table 39-2. Causes of Hepatic Granulomas in Patients with HIV Common*
Uncommon
Mycobacterium avium complex Idiopathic Other mycobacteria Mycobacterium tuberculosis Cryptococcus Histoplasma Medications (trimethoprim-sulfa)
Cytomegalovirus Candida Microsporidia Schistosoma
*In approximate order of prevalence.
Granulomas are the next most common histopathologic abnormality. These granulomas rarely have the typical appearance of epithelioid granulomas with histiocytes and giant cells; rather, they are often poorly formed (Figure 39-1A). Foamy blue histiocytes are common in granulomas associated with MAC (Figure 39-1B), whereas caseating granulomas are typical for TB. Most granulomas in HIV-infected patients have an infectious etiology, usually MAC, TB, or fungi, although they can be observed in a number of other conditions (Table 39-2), and in some patients remain idiopathic. Histopathologic changes of viral hepatitis are very common in HIV-infected patients, and their frequency is highly dependent on the risk group being studied and the stage of immunodeficiency. Cirrhosis is not infrequent and is usually caused by chronic viral hepatitis or alcohol.
SPECIFIC CAUSES OF LIVER DISEASE DRUG-INDUCED Drugs are the most common cause of acute liver injury and should be considered in every patient with suspected liver disease. However, the true prevalence of drug-induced liver disease is
782
B
Figure 39-1. Hepatic granuloma caused by Mycobacterium avium complex. (A) Well formed granuloma with normal surrounding hepatic architecture. (B) Acid-fast staining demonstrates multiple organisms. Cultures of the biopsy grew Mycobacterium avium complex.
Chapter 39 HIV-ASSOCIATED HEPATOBILIARY DISEASE
difficult to determine because (1) potentially hepatotoxic medications are often discontinued when elevated liver enzymes are detected, without further investigation; (2) liver biopsy is rarely performed to evaluate for drug-induced hepatic injury; and (3) the histopathologic findings of drug-induced liver disease are often nonspecific. Some medications, such as trimethoprim–sulfamethoxazole, have a well recognized increased risk of toxicity in these patients.7 The mechanisms of drug-induced injury are multifactorial, including idiosyncratic, allergic, drug interactions, the concomitant use of alcohol, or the presence of underlying chronic liver disease. Improvement in immune function with HAART can exacerbate viral hepatitis, particularly that due to HBV. Although a variety of medications have been linked to hepatotoxicity in HIVinfected patients, we will focus on the clinical features and toxicity associated with some of the most common offenders.
Antiretroviral Agents The armamentarium of antiretroviral agents continues to grow. Cohort studies identify significant drug-induced hepatotoxicity in approximately 4–10% of those starting HAART.8 The major risk factors associated with the development of liver injury are coexistent viral hepatitis, elevated baseline ALT levels and older age.3,9,11 Other suggested risk factors are patients who are naive to HAART, discontinuation of lamivudine, lower baseline CD4 count, and rise in CD4 count >50 cells after HAART. The antiretroviral drugs most often associated with drug toxicity include nevirapine and ritonavir.10,12 Indinavir can cause mild indirect hyperbilirubinemia and rarely acute hepatitis. Drug discontinuation is indicated when an allergic reaction occurs, whereas continuing these drugs may be reasonable in other settings if the elevations are only mild with close follow-up. The most feared liver injury associated with nucleoside reverse transcriptase inhibitors is mitochondrial damage and the lactic acidosis syndrome.12 The mechanism of toxicity is believed to be mitochondrial DNA polymerase depletion;13 stavudine has been implicated most frequently. The syndrome is characteristically a multisystem disease manifesting as liver injury (jaundice), lactic acidosis, and frequently death. Pancreatitis and myopathy are also common. Patients typically present with dyspnea, nausea, vomiting and abdominal pain, but may be asymptomatic; the liver tests are usually only mildly increased and hepatomegaly is often pronounced. Elevated serum lactate levels and a metabolic anion gap acidosis are characteristic. Prompt recognition with drug discontinuation is critical, as mortality rates up to 100% are common. No defined treatments are available except liver transplantation.
Trimethoprim–Sulfamethoxazole The high frequency of side effects, including hepatotoxicity, of trimethoprim–sulfamethoxazole in patients with AIDS is well recognized.7 Hepatitis occurs relatively soon after drug initiation, with features of an allergic reaction including rash, fever and eosinophilia. The most common liver test findings are a raised bilirubin associated with mild to marked increases in serum aminotransferases. Histologically, granulomas are commonly observed along with bile stasis and hepatocyte necrosis.
Antimycobacterial Therapy Antituberculous agents such as isoniazid and rifampin are common causes of liver injury. Drug-induced hepatotoxicity may be even more frequent in HIV-infected patients because of the frequent coexistence of alcoholism and the concomitant use of other drugs.13 Isoniazid and rifampin-induced liver injury usually occurs within the first several months of drug ingestion, but can occur at any time. Clinical features of hepatitis and jaundice are common. Prompt discontinuation of these medications usually results in complete resolution of symptoms, although failure to recognize isoniazid hepatotoxicity with continued drug administration may be fatal.
VIRAL DISEASES Among HIV-infected patients, viruses constitute the most important hepatic pathogens. The high prevalence of infection with hepatotropic viruses is not surprising, as the routes of transmission and the risk factors for acquiring these viruses and HIV are similar. Compared to immunocompetent hosts, a number of important differences in the natural history of viral hepatitis, as well as unique diagnostic and management challenges, are observed in HIVinfected patients.
HEPATITIS A VIRUS The epidemiology, natural history and outcome of hepatitis A virus (HAV) infection in HIV-infected patients have been understudied. Risk factors for the development of HAV infection in homosexual men include the number of sexual partners, as well as oral–anal and digital–rectal intercourse.14 Among an HIV-positive population of predominantly intravenous drug users (IVDU) and heterosexual patients, a higher prevalence of anti-HAV antibodies has not been observed compared to routine blood donors.15 Case reports and small case series have reported higher serum titers of HAV RNA, as well as more prolonged viremia and higher transaminase levels, among HIV-infected patients.16,17 Despite these observations, there are no data to suggest that HIV-infected patients develop more severe hepatic disease or worse outcomes than non-HIV-infected patients. Hepatitis A vaccine is safe in HIV-positive patients, although less immunogenic.18
HEPATITIS B VIRUS Epidemiology Prevalence Studies have shown that 65–96% of HIV-infected patients have had prior exposure to the hepatitis B virus (HBV), as defined by surface antibody (anti-HBs) or core antibody (anti-HBc) positivity.19 However, HBV seropositivity depends on the risk group studied, with the highest prevalence seen in homosexual men and IVDU.
Incidence Incidence studies of HBV infection in HIV-infected patients are limited. In one study of 57 patients, 6 (10%) acquired HBV over a median follow-up of 18 months.20 Decreasing rates of HBV infection have been observed among high-risk individuals (IVDU) with or without HIV, suggesting the effectiveness of public health efforts,
783
Section V. Liver Diseases Due to Infectious Agents
including risk-behavior modification and vaccination, and perhaps HAART.21,22
Natural History Influence of HIV Infection on HBV Disease The outcome of HBV infection is strongly influenced by coexistent HIV infection. HIV-infected patients who develop acute HBV infection have a higher likelihood of chronic infection than do HIV-negative individuals, although no differences in the clinical presentation or severity of hepatitis have been observed.20 Studies have shown that HIV-infected patients are also more likely to lose antibodies and/or to have lower HBV antibody titers than are seronegative controls.20 Some patients who clear HBeAg and become anti-HBe positive and later become HIV-infected may have a return of HBV viremia and reactivation of disease. In addition, an increased frequency of positive isolated anti-HBc (with undetectable anti-HBs) has been observed in HIV-infected individuals, probably resulting from a failure to develop anti-HBs in the context of immunodeficiency.23 Studies examining the degree of HBV-related liver injury, as assessed by transaminase levels and hepatic histology, have yielded conflicting results. Most studies have found lower ALT levels and less severe histopathologic injury in HIV-infected patients than in controls. In addition, when stratified based on absolute CD4 lymphocyte count, there appears to be a positive correlation between hepatic injury and CD4 count. The impact of HIV co-infection on the outcome of chronic HBV infection is unclear, although some studies have suggested an accelerated progression towards decompensated cirrhosis in co-infected patients.24
Influence of HBV Infection on HIV Disease Studies have failed to demonstrate a major influence of HBV infection on progression of HIV infection to AIDS.25
Therapy a-Interferon therapy in HIV-infected patients has been associated with a poor response, which may be related to the overall immunosuppression and/or abnormalities in the cytokine profile related to HIV.26 Lamivudine promptly inhibits HBV replication, but the emergence of resistance to lamivudine is of concern.26 Adefovir has been shown to be effective in the treatment of lamivudine-resistant HBV in HIV/HBV-co-infected patients. Tenofovir has recently been shown to have significant activity against both HIV and HBV.26 Immune restoration with antiretroviral therapy has been associated with an acute elevation of serum transaminases (flare) followed by HBV seroconversion, but this phenomenon appears to be relatively infrequent.27 Nevertheless, when it occurs it may be severe, resulting in fulminant hepatic failure. As noted earlier, HBsAg seropositivity has been identified as an independent predictor of HAART-related hepatotoxicity.9
Prevention Given the increased prevalence of serological markers of prior HBV exposure and the high incidence of HBV infection, particularly among homosexual men and IVDU, vaccination has been recommended. However, as with other immunosuppressed hosts, seroconversion rates and, among responders, anti-HBs titers are lower,
784
especially for patients with low CD4 counts. Therefore, it is important to implement vaccination in HIV-positive patients early in their disease to maximize the effectiveness of vaccination.21,28
HEPATITIS DELTA VIRUS Hepatitis delta virus infection is uncommon in the United States compared to other areas of the world. The primary risk factor for HDV infection is IVDU, although having an increased number of sexual partners has also been identified.29 HIV infection abrogates the suppressive effects of HDV on HBV replication, and higher ALT levels suggest more severe hepatic injury.30 The role of interferon in HIV-infected patients is not established, although promising results have been observed in HIV-negative patients.31 To date, there is no evidence that HDV infection alters the natural history of HIV infection.
HEPATITIS C VIRUS Epidemiology The prevalence of HCV infection among HIV-infected patients is highly dependent on the group studied: among homosexual or bisexual men prevalence rates vary from 1.6% to 11.7%, and among IVDU from 13% to 40%, and HCV positivity is very common in hemophiliacs, approaching 90–100%.32 Some patients with HCV/HIV co-infection may lose HCV antibody over time, thus potentially underestimating the true prevalence of HCV infection. HCV infection has important implications for both HCV and HIV transmission. HIV infection, probably related to higher levels of HCV viremia, increases the risk of sexual and parenteral transmission of HCV to HIV-seronegative female sex partners, and to newborns of HIV-infected mothers.32
Natural History Influence of HIV Infection on HCV Disease The outcome and clinical presentation of acute HCV infection in HIV-infected patients are not well defined. However, HIV infection is now recognized as an important cofactor in an accelerated progression of HCV disease.33–37 HCV RNA is more commonly detected and levels are higher in HIV-infected patients than in HIVnegative controls, which suggests an increased rate of viral replication and/or decreased clearance by the host, perhaps owing to underlying immunosuppression. No difference in ALT levels between HCV+/HIV+ compared to HCV+/HIV- controls can be shown,38 although liver biopsy specimens have demonstrated a higher degree of necroinflammatory activity and fibrosis in patients with HIV infection, especially in those with lower CD4 cell counts and HCV genotype 1b.34,37,39 Factors that predict fibrosis and progression to cirrhosis in co-infected patients include older age at time of infection, higher alanine aminotransferase levels, higher inflammatory activity, alcohol consumption of >50 g/day, and CD4+ T-cell count of <500/mm3.34,39,40 Compelling data suggest that fibrosis occurs more rapidly and in a greater proportion of HIV/HCV co-infected patients. A cohort study of 310 hemophiliacs from England showed that death rates 25 years after exposure to HCV were 57% for all-causes of mortality and 21% for liver disease-related mortality in HIV/HCV coinfected patients, compared to 8% and 3%, respectively, in
Chapter 39 HIV-ASSOCIATED HEPATOBILIARY DISEASE
HIV-negative patients.35 In another cohort of patients with hemophilia36 the cumulative risk of death from liver disease and hepatocellular carcinoma was 6.5% in co-infected patients at all ages 25 years after the first exposure to potentially infected blood products, compared to 1.4% in HCV+/HIV- patients. In summary, the preponderance of data shows that co-infected patients, especially hemophiliacs, have a greater incidence of progression to cirrhosis and liver-related complications. The mechanism(s) by which HIV potentiates liver injury from HCV in these latter patients is unknown, but may result from a longer period of HCV infection, HCV genotype, or possibly greater levels of viremia. As with HIV, in the liver transplant patient they also have higher levels of virus and develop cirrhosis in as many as 8% of patients 5 years following transplant, suggesting that any type of immunosuppression has an adverse effect on the liver disease of HCV-infected patients.41
Influence of HCV Infection on HIV Disease The influence of chronic HCV infection on the natural history of HIV and AIDS is less certain, but most data from both prospective and retrospective studies have found no relationship between HCV infection and the progression of HIV disease or AIDS-related death.33
Influence of HAART The incidence of antiretroviral drug-associated liver enzyme elevations and hepatic dysfunction is increased in co-infected patients, but overall, clinically important toxicity remains infrequent.8,9,42 Hepatotoxicity ranges from mild hyperbilirubinemia and transaminase elevations to overt liver failure. The mechanism(s) of increased hepatotoxicity are unknown, but possible explanations include viral interference (HCV replication facilitated by HIV inhibition), decreased endogenous interferon levels and immune reconstitution. Immune restoration after HAART has been associated with an increase in HCV RNA replication, worsening liver inflammation and histology, and the development of hepatic decompensation.43,44 Other studies, however, have failed to show an increase in transaminase levels, changes in HCV RNA associated with HAART, or significant differences in biochemical or histological parameters between HIV-infected patients administered or not administered HAART.38 In addition, antiretroviral therapy significantly reduced long-term liver-related mortality in these patients, and biopsy data suggest a beneficial effect as well.45 In summary, antiretroviral therapy can be safely administered to co-infected patients, with special attention to closely monitoring liver tests. Further research into the mechanisms of drug hepatotoxicity and the effects of immune reconstitution is needed.
and in HIV-infected patients, the severity of immunosuppression (i.e. CD4 count). Response to a-interferon monotherapy in the setting of co-infection is associated with dismal sustained virological response (SVR) rates. a-Interferon in combination with ribavirin has led to improved end-of-treatment response and improved SVR rates (15–40%) with no significant changes in HIV viral load or CD4 count.46 Large clinical trials are currently ongoing to evaluate the efficacy of pegylated a-interferon and ribavirin in HIV/HCV coinfected individuals. Preliminary data and those derived from smaller studies suggest SVR of 31–35% (naive patients) and 16% (prior a-interferon relapsers or non-responders).47,48 Concerns exist regarding an increased incidence of HCV-related treatment side effects in co-infected patients, although the most recent studies suggest an incidence of side effects and discontinuation rates similar to those observed in HIV-negative patients. One particular concern, though, is the development of anemia in patients receiving ribavirin, but the use of erythropoietin can significantly increase hemoglobin levels and allow for the use of higher doses of ribavirin, which may be associated with a higher SVR rate. The interaction of ribavirin with zidovudine, zalcitabine, stavudine and didanosine may result in decreased anti-HIV activity or increased concentrations of antiretrovirals and the potential for toxicity. Studies have documented stable HIV viral loads and a 10–25% decline in the absolute number of CD4 cells in co-infected patients receiving anti-HCV combination therapy on stable doses of antiretrovirals. Thus although ribavirin may inhibit the phosphorylation of reverse transcriptase inhibitors in vitro, in vivo, ribavirin does not appear to detrimentally affect its antiretroviral efficacy. To date, no important drug interactions between ribavirin and nucleoside analogues have been noted.
OTHER HEPATOTROPIC VIRUSES The natural history of other hepatotropic viruses has not been well studied in the setting of HIV infection. Some studies suggest a higher seroprevalence of hepatitis E virus (HEV) infection, but the clinical epidemiology, presentation, natural history and outcome of HEV infection in patients with HIV and AIDS remains poorly defined.49 Hepatitis G virus (HGV), a flavivirus, is prevalent but is not known to be associated with any chronic disease in healthy individuals. HGV RNA was found in 17% of HIV-infected patients, and, interestingly these patients had a better survival, slower disease progression to AIDS, and lower HIV viral loads.50 The mechanism(s) of these effects remain undefined.
OTHER VIRUSES THERAPY Since the advent of HAART therapy, survival for HIV-infected patients has improved substantially. Given the apparent aggressive course of HCV infection and the potential for increased end-stage liver disease-related mortality in these patients, consideration for anti-HCV therapy is warranted. Indications and contraindications for treating HCV in this population are similar to those for HIVnegative patients. Treatment response is highly dependent on the particular antiHCV drug and regimen used, liver histology and HCV genotype,
Immunohistochemical studies of liver tissue have demonstrated HIV within Kupffer cells, but rarely within hepatocytes.51 However, using PCR, HIV RNA has not been consistently demonstrated in all patients, and no correlation between hepatic histology and concentrations of HIV-1 RNA have been shown.52 Given these findings, it is unclear what role, if any, HIV infection of the liver plays in causing either non-specific histopathologic abnormalities, dysfunction of Kupffer cells, or as a reservoir of viral production. Despite being the most common opportunistic infection in AIDS, clinically apparent liver disease caused by CMV is rare. Most studies
785
Section V. Liver Diseases Due to Infectious Agents
report prevalence rates of less than 5%; autopsy studies tend to have higher rates of identification.53 Histologically, CMV is usually observed in sinusoidal endothelial cells, often with a minimal inflammatory response, suggesting that CMV may be an incidental finding. Rarely, CMV may present as acute hepatitis; in these circumstances, antiviral therapy for CMV is indicated. Epstein–Barr virus and varicella have also been reported to involve the liver in AIDS.
MYCOBACTERIAL INFECTIONS MYCOBACTERIUM AVIUM COMPLEX Organisms of the Mycobacterium avium complex, M. avium and M. intracellulare, are among the most common opportunistic infections in patients with AIDS. The organism is acquired from the environment, and disseminated infection results from recent infection rather than reactivation of previous infection.54 The gastrointestinal tract and lung are important portals of entry from which the organism may disseminate. The major risk factor for infection is the level of immune dysfunction; the mean CD4 count in infected patients is 60/mm3, with infection rare above 100/mm. Clinical features which are highly suggestive of disseminated MAC include high fever, anemia, night sweats, diarrhea, and weight loss greater than 10%.54 Abdominal pain, hepatomegaly and increased alkaline phosphatase are frequently present. Although the spleen is the most common site for dissemination, liver involvement is frequent but generally asymptomatic53 (see Figure 39-1). Lymph nodes and bone marrow are also common sites of involvement. In most series, MAC is the most common opportunistic infection involving the liver.53 Histologically, granulomas are often present but are usually poorly formed (see Figure 39-1). Foamy blue histiocytes are the histologic hallmark of MAC infection. Mycobacterial staining of biopsy tissue often demonstrates striking numbers of mycobacteria, and may be positive in the absence of granulomas. Culture of liver tissue is more sensitive than histology and may be positive in the absence of organisms by special stains. Because blood culture positivity may take several weeks, thereby delaying diagnosis, more rapid diagnostic methods have been evaluated. In one study, liver biopsy was more sensitive than bone marrow examination; the diagnosis was established histologically in 75% of liver biopsies compared to 25% from bone marrow, although both techniques yielded similar results by culture.55 Abdominal imaging studies may be helpful in suggesting disseminated MAC. Liver and splenic involvement is usually diffuse, causing hepatosplenomegaly; focal abnormalities are rarely identified. Lymphadenopathy is prominent and, in contrast to TB, the involved nodes infrequently have central necrosis and are much smaller.56 Before HAART, long-term survival following the diagnosis of disseminated MAC was poor, with a median of 3–6 months. With HAART, survival has been substantially improved and long-term antimycobacterial therapy may not be required.57
MYCOBACTERIUM TUBERCULOSIS The AIDS epidemic has been associated with a dramatic resurgence in the prevalence of TB. In Africa, TB is the most common cause of death, is identified in approximately 50% of autopsies, and is disseminated in over 80% of patients.58 Unlike MAC, TB can occur in earlier stages of immunodeficiency, often with CD4 counts >200/mm3. In
786
most patients, disease results from reactivation rather than primary infection. Regardless of the stage of immunodeficiency, pulmonary disease is the most common presentation, and dissemination to almost any organ, including the liver, has been described. Hepatic involvement may be suspected by liver test abnormalities (raised alkaline phosphatase) in the setting of disseminated tuberculosis. Similar to MAC, clinical expression of liver disease is rare. Focal liver disease may present with right upper quadrant abdominal pain and fever. CT findings of disseminated TB include lymphadenopathy with evidence of central necrosis.56 In contrast to MAC, focal lesions often occur in other organs and characteristically are necrotic. Pulmonary disease may be absent. The diagnosis may be established by fine needle aspiration of liver lesions, lymph nodes or other focal lesions, with appropriate staining. Blood cultures are often positive. Unlike the therapy for MAC, multidrug regimens for tuberculosis yield a clinical and microbiologic cure, with relapse of infection being infrequent. Monitoring of the liver tests in selected patients is important to detect hepatotoxicity early. In the HIV-infected patient with a positive PPD, prophylactic therapy with isoniazid has been recommended,59 and although it decreases the incidence of both pulmonary and disseminated tuberculosis, liver toxicity has been reported in ª10% of patients.59
FUNGAL DISEASES Despite the prevalence of fungal diseases in other organ systems in AIDS patients, involvement of the liver is uncommon clinically. Fungal diseases have important differences in prevalence, based on the epidemiologic and geographic setting. For example, disseminated histoplasmosis is much more common in the central USA, whereas coccidiodomycosis is a common opportunistic pathogen in the southwest. Cryptococcus is a frequent pathogen in AIDS, and the most common fungus reported to involve the liver.6 Disseminated disease is rare in those with meningitis but it can involve lung, liver, or bone marrow. Amphotericin B is effective for acute disease, whereas fluconazole is administered chronically to maintain a remission. Disseminated histoplasmosis is an AIDS-defining illness and most often involves the lung and hematopoietic organs at the time of presentation. In immunocompetent hosts, pulmonary involvement is the most common manifestation; pulmonary involvement is observed in most AIDS patients with disseminated histoplasmosis.60 The most common clinical presentation is recurrent high fevers, and in some patients a clinical picture consistent with disseminated intravascular coagulation. High LDH concentrations >600–1000 IU/l suggest the diagnosis.61 The diagnosis can be established by staining of peripheral blood smears, where the organism may be observed in polymorphonuclear leukocytes; bone marrow biopsy with culture and occasionally liver biopsy may be diagnostic. Amphotericin B and itraconazole are effective for acute treatment and chronic prophylaxis, respectively. Immune reconstitution with HAART will negate the need for long-term maintenance antifungal therapy.57 Other fungi reported to involve the liver in AIDS include blastomycosis, coccidiodomycosis, and rarely candida. Penicillium marneffie is a fungus endemic to Southeast Asia and has been increasingly
Chapter 39 HIV-ASSOCIATED HEPATOBILIARY DISEASE
recognized as an opportunistic infection that can involve the liver in patients with AIDS.6
BACTERIAL INFECTIONS Bacterial infections of the liver are rare. Peliosis hepatica is now recognized as being caused by the bacteria Bartonella henselae. Patients with marked immunodeficiency are at greatest risk and present with disseminated disease involving the skin, bones and liver, similar to cat-scratch disease.62 Fever, abdominal pain and hepatosplenomegaly are common with hepatic involvement. The serum alkaline phosphatase is usually markedly elevated. Abdominal CT scan demonstrates hepatomegaly with low-density lesions representing vascular lakes. Biopsy of these lesions reveals vascular channels, and specific stains (Warthin–Starry) aid in identifying these bacteria.63 Culture of liver tissue may also be positive. Antibiotic therapy with doxycycline or erythromycin is usually effective; relapse may occur.
Figure 39-2. Hepatic non-Hodgkin’s lymphoma: multiple filling defects of variable size. Ultrasound-guided biopsy demonstrated high-grade lymphoma.
PROTOZOAL INFECTIONS Protozoa are distinctly uncommon hepatic pathogens. Although rare cases of microsporidial or cryptosporidial hepatitis have been noted, cryptosporidia and microsporidia typically involve the biliary tree rather than hepatic parenchyma (see below). Disseminated Pneumocystis carinii (PCP) may involve the liver, usually in patients using inhaled prophylactic pentamidine to prevent PCP. Liver involvement is diffuse and thus hepatic mass lesions seen on CT are uncommon. Other reported causes of liver infection in AIDS include Isospora belli, Leishmania64 and Toxoplasma. Amebic liver abscess has been observed in high-prevalence areas.
NEOPLASMS Neoplasms may be the initial manifestation of AIDS and can present with isolated liver involvement. In early reports of the AIDS epidemic non-Hodgkin’s lymphoma (NHL) was present in up to 2.3% of patients,65,66 but the rate of most lymphomas has decreased in the HAART era.67 In contrast to immunocompetent hosts, AIDSrelated lymphomas typically present in extranodal sites, are more advanced, and have a poor prognosis.66 HIV-associated lymphoma may have a primary presentation in the liver, and autopsy studies show hepatic lymphoma which was unsuspected antemortem.68 Hepatic lymphoma is usually manifested by abnormal liver tests, often with striking elevations of the alkaline phosphatase or LDH; jaundice is common. Hepatomegaly with or without abdominal pain is common, whereas fever is inconsistent. Peripheral lymphadenopathy is uncommon. Although abdominal ultrasound is frequently diagnostic, abdominal CT is the diagnostic modality of choice.68 Liver involvement appears as one or more focal lesions of variable size (see Figure 392). Abdominal lymphadenopathy and splenic involvement are common, whereas mediastinal adenopathy is rare. Percutaneous radiographically directed biopsy of identified mass lesions will safely and reliably establish the diagnosis. With HAART, the response rate and long-term disease-free survival following chemotherapy have dramatically improved.69
Kaposi’s sarcoma (KS) was recognized early on as a common initial manifestation of AIDS. This neoplasm, caused by human herpes virus-8 (HHV-8), becomes manifest under conditions of immunodeficiency.70 This virus can be detected in blood, and sexual contact is an important route of transmission. The incidence of KS has fallen dramatically, coincidental with the use of effective antiretroviral therapies.71 Cutaneous involvement is the most common presentation; visceral involvement is common and usually asymptomatic. Hepatic KS is generally clinically silent and was a frequent finding at autopsy before HAART. The lesions are generally multiple and hyperechoic. Chemotherapy is relatively effective in controlling disease but, remarkably, tumor regression occurs with HAART therapy alone, making it the treatment of choice.72 Several other tumors involving the liver have been reported in HIV-infected patients, including metastatic adenocarcinoma, cholangiocarcinoma, melanoma and hepatoma. Whether these neoplasms are related to HIV infection or are a complication of other diseases (excluding HCV) is unknown.
CAUSES OF ASCITES As in any patient, ascites may be caused by hepatic (portal hypertension), extrahepatic or peritoneal disease(s). The most common cause of ascites in these patients is cirrhosis and portal hypertension secondary to chronic viral hepatitis. However, in about 25% of patients opportunistic infections and neoplasms are causative, usually from disseminated disease with peritoneal involvement. A syndrome of non-specific peritonitis has been described.73 These reported patients presented with abdominal pain and overt ascites. Ascitic fluid analysis demonstrated high protein concentrations and leukocytosis, but no specific identifiable pathogens. Laparoscopy and laparotomy also failed to disclose a specific etiology; in some patients adhesions and peritonitis were found. The cause of peritoneal disease in these patients is unknown. The evaluation of ascites should generally parallel that in other patients. Patients with high-protein ascites should also have a sample
787
Section V. Liver Diseases Due to Infectious Agents
submitted for cytological analysis to exclude NHL. CT may be useful to identify peritoneal mass lesions (e.g. NHL) or other intraabdominal processes; mass lesions may then be safely biopsied percutaneously under CT guidance. The presence of chylous ascites suggests underlying disruption of the lymphatic system caused by KS or tuberculosis. Laparoscopy with inspection of the liver and peritoneum may play a role when other tests are non-diagnostic.
BILIARY TRACT ABNORMALITIES ETIOLOGY Disorders of the biliary tree in AIDS patients, termed AIDS cholangiopathy, were recognized early in the AIDS epidemic. However, like other opportunistic processes in AIDS, the frequency of this entity has fallen coincidentally with the introduction of HAART. The cause of ductal disease in most patients is infection. Liver biopsies, which included biliary ductal epithelium, biopsies of the common bile duct and/or papilla after endoscopic sphincterotomy, biliary brushings and cytology, as well as periampullary small bowel biopsies, have all identified an infectious cause in selected cases. However, in a substantial number no cause(s) can be found. The most frequent identifiable pathogen is Cryptosporidium; other infectious causes include microsporidia (E. bienusi, Septata intestinalis),74 Cyclospora, CMV, MAC and Giardia. Based on autopsy studies and bile duct biopsies at ERCP, these infections cause severe inflammatory changes that result secondarily in the observed cholangiographic abnormalities. Non-infectious causes of biliary disease, both benign and malignant, include stones and strictures.
CHOLANGIOGRAPHIC PATTERNS
A
The most frequent cholangiographic finding is papillary stenosis in association with intrahepatic sclerosing cholangitis, occurring in approximately 50% of patients with the AIDS cholangiopathy syndrome (Figure 39-3). The next most frequent pattern is intra- and extrahepatic sclerosing cholangitis without papillary stenosis, followed by papillary stenosis alone, or intrahepatic sclerosing cholangitis alone. Isolated strictures of the common bile duct may result from primary common bile duct lymphoma, or pancreatic disease caused by chronic pancreatitis, infections, or neoplasms. Pancreatic duct lesions have also been described, perhaps related to ampullary obstruction.75
CLINICAL PRESENTATIONS The typical presentation of AIDS cholangiography is right upper quadrant pain; papillary stenosis is usually present in patients with severe pain and clinical cholangitis, whereas intrahepatic cholangitis results in milder pain or may be asymptomatic. Diarrhea is common and related to coexistent small bowel involvement with cryptosporidial or microsporidial infection. Asymptomatic elevation of liver tests may be the first clue to the diagnosis. Serum alkaline phosphatase is usually elevated, with mean values in most series of 700–800 IU/l. Mild increases in ALT are common, but jaundice is rare. Rarely liver tests may be normal.
DIAGNOSIS When using ERCP as the gold standard, ultrasound has a sensitivity of approximately 75–87%76 and may demonstrate striking ductal
788
B
Figure 39-3. AIDS cholangiopathy. (A) The common bile duct and cystic duct are markedly dilated to the level of the ampulla. There is also nodularity of the duct. (B) The intrahepatic ducts have areas of stricture and dilation typical for sclerosing cholangitis. A biliary sphincterotomy was performed.
thickening in some patients. HIDA scanning may suggest biliary obstruction when there is delayed excretion of the tracer. In the symptomatic patient without jaundice ultrasound should be the initial study, reserving CT for the patient with jaundice where intrahepatic mass lesions, abdominal adenopathy, and biliary dilation can be identified. ERCP should be reserved for those in whom endoscopic therapy is anticipated.
Chapter 39 HIV-ASSOCIATED HEPATOBILIARY DISEASE
TREATMENT Treatment of AIDS cholangiopathy is primarily endoscopic. For those patients with abdominal pain or cholangitis associated with papillary stenosis, endoscopic sphincterotomy generally provides benefit, with long-term symptom relief in most patients following sphincterotomy for papillary stenosis. It is possible that coexistent pancreatic disease may be responsible for the incomplete pain relief in some patients.75 On long-term follow up the serum alkaline phosphatase may continue to rise owing to the progression of associated intrahepatic sclerosing cholangitis. Sphincterotomy is not indicated for sclerosing cholangitis in the absence of papillary stenosis and common bile duct dilation, and may be associated with a higher complication rate. For those with a dominant common bile duct stricture, endoscopic stenting and/or balloon dilatation should be performed after sampling the stricture. Several reports show some improvement or even resolution of cholangiographic abnormalities with antimicrobial therapy, usually in conjunction with HAART.77 This would suggest that, depending on the chronicity of disease, the cholangiographic findings are inflammatory and reversible. As with all AIDS-related opportunistic infections, the prognosis of disease is linked to the severity of immunodeficiency.78
GALLBLADDER DISEASE As with AIDS-associated sclerosing cholangitis, disorders of the gallbladder are caused primarily by infections: Isospora belli, CMV, cryptosporidia, microsporidia and Candida albicans are reported etiologies. The most common manifestation of gallbladder disease is acalculous cholecystitis, although symptomatic cholelithiasis may also be seen. The clinical presentation of acalculous cholecystitis includes right upper quadrant pain and fever; abnormal liver tests suggest concomitant sclerosing cholangitis. As in the normal host, ultrasound or isotopic imaging may be used for diagnosis. Laparoscopic cholecystectomy is curative and is the operative technique of choice.79
APPROACH TO THE DIAGNOSIS OF HEPATOBILIARY DISEASE The evaluation of hepatobiliary disease in HIV-infected patients must be tailored to the presenting symptoms and signs, pattern of liver test abnormalities, and severity of immunodeficiency. Patients with a CD4 lymphocyte count >200/mm3 are unlikely to have an opportunistic infection, although some infections (e.g. TB) and neoplasms, including NHL and KS, may occur at only modest levels of immunodeficiency. Striking ALT elevations suggest hepatitis, possibly from drugs or viral disease, depending on the clinical setting. Alcoholic liver disease can generally be suspected by the history, physical examination and pattern of liver test abnormalities. Disproportionate elevations of the alkaline phosphatase suggest infiltrative disorders or biliary tract disease, but may be present in the absence of any obvious identifiable biliary or hepatocellular disease. Jaundice most commonly results from drug-induced hepatitis, highgrade biliary obstruction or NHL.
Abdominal imaging studies provide valuable diagnostic information. CT is most useful to evaluate for mass lesions, adenopathy, peliosis and peritoneal diseases. Infections resulting in diffuse parenchymal liver disease, such as TB, MAC and parasitic diseases, rarely have focal abnormalities found on these studies. The utility of ultrasound compared to CT in various clinical settings has not been well studied. We employ ultrasound in the anicteric patient in whom AIDS cholangiopathy is suspected, reserving CT for those with marked hepatomegaly, jaundice, or suspected mass lesions or intra-abdominal processes. Invasive techniques should be used judiciously. ERCP is most appropriate in the patient with biliary ductal dilation where endoscopic therapy for papillary stenosis, choledocholithiasis or ductal strictures is likely. As mentioned previously, because the majority of opportunistic infections involve the liver secondarily, liver biopsy rarely uncovers disorders not identified in other tissues. Both liver biopsy and bone marrow biopsy may be helpful in the patient with fever of unknown origin when other modalities are non-diagnostic.55,80 As blood cultures may take at least 2 weeks to become positive, liver and/or bone marrow biopsy may expedite the diagnosis. Several studies document the high yield of liver biopsy in HIV-infected patients with fever and abnormal liver tests. Many of these studies are from developing countries where TB is a common histologic finding. When MAC is suspected and blood and bone marrow cultures are negative, liver biopsy may establish the diagnosis. At many centers, however, empiric therapy for MAC, rather than liver biopsy, is often administered pending culture results, given the efficacy and tolera-
Table 39-1. Overview of the Evaluation of Hepatobiliary Disease in Patients with HIV infection 1. The extent and rapidity of the evaluation should be tailored to the clinical presentation and pattern of liver tests. 2. The CD4 lymphocyte count is essential in formulating the differential diagnosis. Opportunistic infections and neoplasms are most prevalent when the CD4 count is <100/mm3. 3. The liver is an innocent bystander during lymphohematogenous dissemination of opportunistic infections and neoplasms. Thus, evaluation of other organs, such as blood or bone marrow, rather than liver biopsy, may establish the diagnosis. 4. Careful review of all medications, both prescription and over-thecounter, and obtaining viral hepatitis serologies should be performed in the patient with marked elevations of the transaminases. Hepatic imaging studies and liver biopsy rarely alter management in this setting. 5. Ultrasound or CT should be performed in the patient with marked elevation of the alkaline phosphatase to exclude mass lesions, infiltrative disorders, or biliary disease. ERCP may be necessary if bile duct dilation is found, especially if the patient has right upper quadrant pain. Liver biopsy may be considered if all studies are negative depending on the clinical setting. 6. In the jaundiced patient, CT is the test of choice and will exclude hepatic mass lesions and biliary dilation; ERCP may be necessary for diagnosis if bile duct dilation is found, and possibly endoscopic therapy. 7. Mild hepatomegaly is common in AIDS and usually requires no specific evaluation. Marked hepatomegaly is best evaluated with CT; liver biopsy may be useful, depending on the clinical presentation and pattern of liver tests.
789
Section V. Liver Diseases Due to Infectious Agents
bility of newer regimens for the treatment of this pathogen. Ultrasound- or CT-guided biopsy can be employed for focal abnormalities rather than blind percutaneous biopsy. The role of liver biopsy for HIV/HCV co-infected patients in whom interferon therapy is anticipated remains controversial. A summary of factors guiding the approach to the evaluation of hepatobiliary disease in HIV infected patients is provided in Table 39-1.
REFERENCES 1. San-Andrés FJ, Rubio R, Castilla J, et al. Incidence of acquired immunodeficiency syndrome-associated opportunistic diseases and the effect of treatment on a cohort of 1115 patients infected with human immunodeficiency virus, 1989–1997. Clin Infect Dis 2003;36:1177–1185. 2. Monkemuller KE, Call SA, Lazenby AJ, et al. Declining prevalence of opportunistic gastrointestinal disease in the era of combination antiretroviral therapy. Am J Gastroenterol 2000;95:457–468. 3. Montessori V, Press N, Harris M, et al. Adverse effects of antiretroviral therapy for HIV infection. Can Med Assoc J 2004;170:229–238. 4. Gebo KA, Diener-West M, Moore RD. Hospitalization rates differ by hepatitis C status in an urban HIV cohort. J AIDS 2003;34:165–173. 5. Bica I, McGovern B, Dhar R, et al. Increasing mortality due to end-stage liver disease in patients with human immunodeficiency virus infection. Clin Infect Dis 2001;32:492–497. 6. Viriyavejakul P, Rojanasunan P, Viriyavejakul A, et al. Opportunistic infections in the liver of HIV-infected patients in Thailand: a necropsy study. Southeast Asian J Trop Med Publ Health 2000;31:663–667. 7. Floris-Moore MA, Amodio-Groton MI, Catalano MT. Adverse reactions to trimethoprim/sulfamethoxazole in AIDS. Ann Pharmacother 2003;37:1810–1813. 8. Becker S. Liver toxicity in epidemiological cohorts. Clin Infect Dis 2004;38(Suppl 2):S49–55. 9. Sulkowski MS, Thomas DL, Chaisson RE, et al. Hepatotoxicity associated with antiretroviral therapy in adults infected with human immunodeficiency virus and the role of hepatitis-C or B virus infection. JAMA 2000;283:74–80. 10. Wit FW, Weverling GJ, Weel J, et al. Incidence of and risk factors for severe hepatotoxicity associated with antiretroviral combination therapy. J Infect Dis 2002;186:23–31. 11. Puoti M, Torti C, Ripamonti D, et al. HIV-HCV Co-Infection Study Group. Severe hepatotoxicity during combination antiretroviral treatment: incidence, liver histology, and outcome. J AIDS 2003;32(3):259–267. 12. Ogedegbe AEO, Thomas DL, Diehl AM. Hyperlactataemia syndromes associated with HIV therapy. Lancet 2003;3:329–337. 13. Yee D, Valiquette C, Peletier M, et al. Incidence of serious side effects from first-line antituberculous drugs among patients treated for active tuberculosis. Am J Respir Crit Care Med 2003;167:1472–1477. 14. Henning KJ, Bell E, Braun J, et al. A community-wide outbreak of hepatitis A: risk factors for infection among homosexual and bisexual men. Am J Med 1995;99:132–136. 15. Fainboim H, Gonzalez J, Fassio E, et al. Prevalence of hepatitis viruses in an anti-human immunodeficiency virus-positive population from Argentina. A multicentre study. J Viral Hepatol 1999;6:53–57. 16. Ida S, Tachikawa N, Nakajima A, et al. Influence of human immunodeficiency virus type 1 infection on acute hepatitis A infection. Clin Infect Dis 2002; 34:379–385.
790
17. Costa-Mattioli M, Allavena C, Poirer AS, et al. Prolonged hepatitis A infection in an HIV-1 seropositive patient. J Med Virol 2002;68:7–11. 18. Hess G, Clemens R, Bienzle U, et al. Immunogenicity and safety of an inactivated hepatitis A vaccine in anti-HIV positive and negative homosexual men. J Med Virol 1995;46:40–42. 19. Rodriguez-Mendez ML, Gonzalez-Quintela A, Aguilera A, et al. Association of HCV and HBV markers in Spanish HIVseropositive patients in relation to risk practices. Hepatogastroenterology 2003;50:2093–2097. 20. Homann C, Krogsgaard K, Pedersen C, et al. High incidence of hepatitis B infection and evolution of chronic hepatitis B infection in patients with advanced HIV infection. J AIDS 1991;4:416–420. 21. Kellerman SC, Hanson DL, McNaghten AD, et al. Prevalence of chronic hepatitis-B and incidence of acute hepatitis-B infection in human immunodeficiency virus-infected subjects. J Infect Dis 2003;188:571–577. 22. Lugoboni F, Migliozzi S, Mezzelani P, et al. Progressive decrease of hepatitis B in a cohort of drug users followed over a period of 15 years: the impact of anti-HBV vaccination. Scand J Infect Dis 2004; 36:131–133. 23. Rodriguez-Mendez ML, Gonzalez-Quintela A, Aguilera A, et al. Prevalence, patterns, and course of past hepatitis B virus infection in intravenous drug users with HIV-1 infection. Am J Gastroenterol 2000;95:1316–1322. 24. Sheng WH, Chen MY, Hsieh SM, et al. Impact of chronic hepatitis B virus (HBV) infection on outcomes of patients infected with HIV in an area where HBV infection is hyperendemic. Clin Infect Dis 2004;38:1471–1477. 25. Puoti M, Airoldi M, Bruno R, et al. Hepatitis B virus coinfection in human immunodeficiency virus-infected subjects. AIDS Rev 2002 4:27–35. 26. Benhamou Y. Antiretroviral therapy and HIV/hepatitis B virus coinfection. Clin Infect Dis 2004;38(Suppl 2):S98–103. 27. Piroth L, Grappin M, Buisson M, et al. Hepatitis B virus seroconversion in HIV/HBV coinfected patients treated with highly active antiretroviral therapy. J AIDS 2000;23:356–357 28. Tedaldi EM, Baker RK, Moorman AC, et al. HIV Outpatient Study (HOPS) Investigators. Hepatitis A and B vaccination practices for ambulatory patients infected with HIV. Clin Infect Dis 2004;38:1478–1484. 29. Santana Rodriguez OE, Male Gil ML, Hernandez Santana JF, et al. Prevalence of serologic markers of HBV, HDV, HCV and HIV in non-injection drug users compared to injection drug users in Gran Canaria, Spain. Eur J Epidemiol 1998;14:555–561. 30. Farci P. Delta hepatitis: an update. J Hepatol 2003;39:Suppl 1:S212–219. 31. Farci P, Roskams T, Chessa L, et al. Long-term benefit of interferon alpha therapy of chronic hepatitis D: regression of advanced hepatic fibrosis. Gastroenterology 2004;126:1740–1749. 32. Verucchi G, Calza L, Manfredi R, et al. Human immunodeficiency virus and hepatitis C virus coinfection: Epidemiology, natural history, therapeutic options and clinical management. Infection 2004;33:33–46. 33. Greub G, Ledergerber B, Battegay M, et al. Clinical progression, survival and immune recovery during antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV cohort study. Lancet 2000;356:1800–1805. 34. Mohsen AH, Easterbrook PJ, Taylor C, et al. Impact of human immunodeficiency virus (HIV) infection on the progression of liver fibrosis in hepatitis C virus infected patients. Gut 2003;52:1035–1040. 35. Yee TT, Griffioen A, Sabin CA, et al. The natural history of HCV in a cohort of haemophilic patients infected between 1961 and 1985. Gut 2000;47:845–851.
Chapter 39 HIV-ASSOCIATED HEPATOBILIARY DISEASE
36. Darby SC, Ewart DW, Giangrande PLF, et al. Mortality from liver cancer and liver disease in haemophilic men and boys in UK given blood products contaminated with hepatitis C. Lancet 1997;350:1425–1431. 37. Garcia-Samaniego J, Soriano V, Castilla J, et al. Influence of hepatitis C genotypes and HIV on histological severity of chronic hepatitis C. Am J Gastroenterol 1997;92: 1130–1134. 38. Sterling RK, Wilson MS, Sanyal AJ, et al. Impact of highly active antiretroviral therapy on the spectrum of liver disease in HCV-HIV coinfection. Clin Gastroenterol Hepatol 2004;2:432–439. 39. Poynard T, Mathurin P, Lai CL, et al., PANFIBROSIS Group. A comparison of fibrosis progression in chronic liver disease. J Hepatol 2003;38:257–265. 40. Martin-Carbonero L, Benhamou Y, Puoti M, et al. Incidence and predictors of severe liver fibrosis in human immunodeficiency virus-infected patients with chronic hepatitis C: a European collaborative study. Clin Infect Dis 2004;38:128–133. 41. Vierling JM, Villamil FG, Rojter SE, et al. Morbidity and mortality of recurrent hepatitis C infection after orthotopic liver transplantation. J Viral Hepatitis 1997;4(Suppl 1): 117–124. 42. Qurishi N, Kreuzberg C, Luchters G, et al. Effect of antiretroviral therapy on liver-related mortality in patients with HIV and hepatitis C virus coinfection. Lancet 2003;362:1708–1713. 43. Puoti M, Gargiulo F, Quiros-Roldan E, et al. Liver damage and kinetics of hepatitis C virus and human immunodeficiency virus replication during the early phases of combination antiretroviral therapy. J Infect Dis 2000;181:2033–360. 44. Gavazzi G, Bouchard O, Leclercq P, et al. Change in transaminases in hepatitis C virus and HIV-coinfected patients after highly active antiretroviral therapy: differences between complete and partial virologic responders. AIDS 2000;16:1021–1023. 45. Tural C, Fuster D, Tor J, et al. Time on antiretroviral therapy is a protective factor for liver fibrosis in HIV and hepatitis C virus (HCV) co-infected patients. J Viral Hepatol 2003;10:118–125. 46. Morsica G, De Bona A, Foppa CU, et al. Ribavirin therapy for chronic hepatitis C does not modify HIV viral load in HIV-1 positive patients under antiretroviral treatment. AIDS 2000;14:1656–1658. 47. Torriani FJ, Rodriguez-Torres M, Rockstroh JK, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. N Engl J Med 2004;351;438–450. 48. Chung RT, Andersen J, Volberding P, et al. Peginterferon alfa-2a plus ribavirin versus interferon alfa-2a plus ribavirin for chronic hepatitis C in HIV-coinfected patients. N Engl J Med 2004;351:451–459. 49. Fainbom H, Gonzalez J, Fassio E, et al. Prevalence of hepatitis viruses in an anti-human immunodeficiency virus-positive population from Argentina. A multicentre study. J Viral Hepatol 1999;6:53–57. 50. Tillmann HL, Heiken H, Knapik-Botor A, et al. Infection with GB virus C and reduced mortality among HIV-infected patients. N Engl J Med 2001;345:715–724. 51. Hoda SA, White JE, Gerber MA. Immunohistochemical studies of human immunodeficiency virus-1 in liver tissues of patients with AIDS. Mod Pathol 1991;4:578–581. 52. Cao YZ, Dieterich D, Thomas PA, et al. Identification and quantitation of HIV-1 in the liver of patients with AIDS. AIDS 1992;6:65–70. 53. Hofman P, Saint-Paul MC, Battaglione V, et al. Autopsy findings in acquired immunodeficiency syndrome (AIDS). A report of
54.
55.
56.
57.
58.
59.
60.
61. 62.
63.
64.
65.
66.
67.
68.
69.
70. 71.
72.
73.
395 cases from the south of France. Pathol Res Pract 1999;195:209–217. Horsburgh CR. The pathophysiology of disseminated Mycobacterium avium complex disease in AIDS. J Infect Dis 1999;179(Suppl 3):S461–465. Prego V, Glatt AE, Roy V, et al. Comparative yield of blood culture for fungi and mycobacteria, liver biopsy, and bone marrow biopsy in the diagnosis of fever of undetermined origin in human immunodeficiency virus-infected patients. Arch Intern Med 1990;150:333–336. Vanhoenacker FM, De Backer AI, Op de Beeck B, et al. Imaging of gastrointestinal and abdominal tuberculosis. Eur Radiol 2004;14:E103–115. Kirk O, Reiss P, Uberti-Foppa C, et al. European HIV Cohorts. Safe interruption of maintenance therapy against previous infection with four common HIV-associated opportunistic pathogens during potent antiretroviral therapy. Ann Intern Med 2002;20:239–250. Rana FS, Hawken MP, Mwachari C, et al. Autopsy study of HIV-1 positive and HIV-1 negative adult medical patients in Nairobi, Kenya. J AIDS 2000;24:23–29. Gordin F, Chaisson RE, Matts JP. Rifampin and pyrazinamide vs. isoniazid for prevention of tuberculosis in HIV-infected persons. An international randomized trial. JAMA 2000;283:1445–1450. Lamps LW, Molina CP, West AB, et al. The pathologic spectrum of gastrointestinal and hepatic histoplasmosis. Am J Clin Pathol 2000;113:64–72. Butt AA, Michaels S, Greer D, et al. Serum LDH level as a clue to the diagnosis of histoplasmosis. AIDS Read 2002;12:317–321. Plettenberg A, Lorenzen T, Burtsche BT, et al. Bacillary angiomatosis in HIV-infected patients – an epidemiological and clinical study. Dermatology 2000;201:326–331. Cury PM, Pulido CF, Furtado VM, et al. Autopsy findings in AIDS patients from a reference hospital in Brazil: analysis of 92 cases. Pathol Res Pract 2003;199:811–814. Hofman V, Marty P, Perrin C, et al. The histological spectrum of visceral leishmaniasis caused by Leishmania infantum MON-1 in acquired immune deficiency syndrome. Hum Pathol 2000;3:75–84. Tirelli U, Spina M, Gaidano G, et al. Epidemiological, biological and clinical features of HIV-related lymphomas in the era of highly active antiretroviral therapy. AIDS 2000;14:1675–1688. Dal Maso L, Franceschi S. Epidemiology of non-Hodgkin lymphomas and other haemolymphopoietic neoplasms in people with AIDS. Lancet 2003;4:110–119. Masliah E, DeTeresa RM, Mallory ME, et al. Changes in pathological findings at autopsy in AIDS cases for the last 15 years. AIDS 2000;14:69–74. Rizzi EB, Schinina V, Cristofaro M, et al. Non-Hodgkin’s lymphoma of the liver in patients with AIDS: sonographic, CT, and MRI findings. J Clin Ultrasound 2001;29:125–129. Gerard L, Galicier L, Maillard A, et al. Systemic non-Hodgkin lymphoma in HIV-infected patients with effective suppression of HIV replication: persistent occurrence but improved survival. J AIDS 2002;30:478–484. Cannon MJ, Laney AS, Pellett PE. Human herpesvirus 8: Current issues. Clin Infect Dis 2003;37:82–87. Jones JL, Hanson DL, Dworkin MS, et al. Incidence and trends in Kaposi’s sarcoma in the era of effective antiretroviral therapy. J AIDS 2000;24:270–274. Bourboulia D, Aldam D, Lagos D, et al. Short- and long-term effects of highly active antiretroviral therapy on Kaposi sarcomaassociated herpesvirus immune responses and viraemia. AIDS 2004;1:485–493. Wilcox CM, Forsmark CE, Darragh T, et al. High-protein ascites in patients with the acquired immunodeficiency syndrome. Gastroenterology 1991;100:745–748.
791
Section V. Liver Diseases Due to Infectious Agents
74. Sheikh RA, Prindiville TP, Yenamandra S, et al. Microsporidial AIDS cholangiopathy due to Encephalitozoon intestinalis: case report and review. Am J Gastroenterol 2000;95:2364–2371. 75. Teare JP, Daly CA, Rodgers C, et al. Pancreatic abnormalities and AIDS related sclerosing cholangitis. Genitourinary Med 1997;73:271–273. 76. Yusuf TE, Baron TH. AIDS Cholangiopathy. Curr Treat Options Gastroenterol 2004;7:111–117. 77. Velásquez J, Gancedo E, Fainboim H, et al. Strategies for the treatment of AIDS-associated sclerosing cholangitis. Am J Med 2004;116:569–570.
792
78. Ko WF, Cello JP, Rogers SJ, et al. Prognostic factors for the survival of patients with AIDS cholangiopathy. Am J Gastroenterol 2003;98:2176–2181. 79. Leiva JI, Etter EL, Gathe J Jr, et al. Surgical therapy for 101 patients with acquired immunodeficiency syndrome and symptomatic cholecystitis. Am J Surg 1997;174:414–416. 80. Kilby JM, Marques MB, Jaye DL, et al. The yield of bone marrow biopsy and culture compared with blood cultures in the evaluation of HIV-infected patients for mycobacterial and fungal infections. Am J Med 1998;104:123–128.
Section VI. Immune Diseases
40
AUTOIMMUNE HEPATITIS AND OVERLAP SYNDROMES Heike Bantel and Michael P. Manns Abbreviations AIC autoimmune cholangiopathy AIH autoimmune hepatitis AMA antimitochondrial antibodies ANA antinuclear antibodies APS1 autoimmune polyendocrine syndrome type 1
CYP HLA NASH PBC PSC
cytochrome P450 human leukocyte antigen non-alcoholic steatohepatitis primary biliary cirrhosis primary sclerosing cholangitis
SLA SMA TNF-a tRNA UGT
soluble liver antigen smooth-muscle antibodies tumor necrosis factor-a transfer RNA uridine diphosphate-glucuronyltransferases
INTRODUCTION
EPIDEMIOLOGY
In 1950 Waldenström described a chronic inflammatory liver disease in a young woman; this condition is now termed autoimmune hepatitis (AIH) and represents a chronic, mainly periportal hepatitis characterized by female predominance, hypergammaglobulinemia, circulating autoantibodies, and a good response to immunosuppressive treatment.1 Serologic detection of autoantibodies is one of the distinguishing features that has led to the subclassification of AIH into three groups (Table 40-1). AIH type 1 represents the commonest form of AIH and is characterized by the presence of antinuclear antibodies (ANA) and/or anti-smooth-muscle antibodies (SMA). The target autoantigen of type 1 AIH is unknown. Characteristic antibodies of AIH type 2 are liver–kidney microsomal antibodies (LKM-1) directed against cytochrome P450 (CYP)2D6 and, with lower frequency, against uridine diphosphate-glucuronyltransferases (UGT). The complex associations of autoantibodies directed against microsomal antigens are summarized in Figures 40-1–40-3. AIH type 3 is characterized by autoantibodies against soluble liver antigen (SLA/LP) directed towards the UGA-suppressor transfer RNA (tRNA)-associated protein.2 The term “overlap syndrome” describes a disease condition in which clinical, biochemical, and serological features of AIH coexist with those of another autoimmune liver disease, most frequently with primary biliary cirrhosis (PBC) but also with primary sclerosing cholangitis (PSC), and, depending on the definition, also with hepatitis C. In adult patients an overlap of PBC and AIH is the most common occurrence, although it remains unclear whether this is a true coexistence of both diseases or an immunoserological overlap characterized by the presence of ANA and antimitochondrial antibodies (AMA). Many ANA-positive patients do not show immunoreactivity against AMA despite a cholestatic liver enzyme profile. This phenomenon has been termed autoimmune cholangiopathy (AIC) or AMA-negative PBC. The coexistence of AIH and PSC has only been conclusively shown in pediatric patients but its existence has been suggested in the adult AIH population. Apart from coexisting, autoimmune liver diseases can also develop into each other (sequential manifestation).3
Originally described in caucasian northern Europeans and North Americans, AIH has a worldwide distribution. This disease affects 100 000–200 000 persons in the USA4 and accounts for 4% of transplant recipients in Europe5 and 5.9% in the USA.6 In northern Europe, the prevalence is estimated at 170 cases per 1 million. Reliable data on the prevalence of autoimmune overlap syndromes are not available. The overlap of AIH and PBC and of AIH and PSC both appear to be present in about 8% of AIH patients.7,8 In contrast, AIH features can be present in about 9% of PBC patients.9 In about 24% of autoimmune diseases of the liver a syndrome of AIC is present.10 Figure 40.2 summarizes overlapping features of AIH type 1 and PBC.
CAUSES, RISK FACTORS, AND DISEASE ASSOCIATIONS AIH and overlap syndromes are often associated with extrahepatic immune-mediated syndromes, including autoimmune thyroiditis, rheumatoid arthritis, and diabetes mellitus (Table 40-2). Most of these autoimmune diseases appear to be inheritable because they are clustered in families. This inherited susceptibility is complex and is likely to rely on a combination of numerous different genes, which most notably include immunogenetic markers (human leukocyte antigen (HLA) genes) and cytokine genes. However, the lack of concordance of most autoimmune diseases in identical-twin pairs implicates that other environmental and host factors, such as bacterial or viral infections, dysregulated apoptosis, or cytokine profile may be relevant for the development of autoimmune diseases.
PATHOGENESIS Autoimmunity is characterized by T-cell-dependent immunopathological responses to auto/neoantigens leading to inflammatory tissue
795
Section VI. Immune Diseases
Table 40-1. Clinical characteristics and distinguishing features of the three subclasses of autoimmune hepatitis (AIH) Clinical features
AIH type 1
AIH type 2
AIH type 3
Diagnostic autoantibodies Target antigen
ANA/SMA Nuclear antigens SMA unknown (??)
SLA/LP UGA-suppressor (tRNA)-associated protein
Prevalence of all AIH types (%)
80%
Common age at presentation
Bimodal (16–30 and >50 years) 41
LKM1 Cytochrome P450 (CYP)2D6 UDP-glucuronosyltransferase UGTIA 20% in Europe 4% in the USA pediatric (2–14 years) 34
58
B14, DR3, C4AQO 82
Unknown 75
Extrahepatic associated diseases (%) HLA association Progression to cirrhosis (%)
B8, DR3, DR4 45
<20% 20–40 years
ANA, antinuclear antibodies; SMA, smooth-muscle actin; LKM1, liver–kidney microsomal antibody type 1; SLA, soluble liver antigen; LP, UGA, UDP, uridine diphosphate; UGTIA,
? in APS1
Chronic hepatitis C
Figure 40-1. Hepatocellular autoantigens. The heterogeneity of hepatocellular autoantigens and their disease specificities. APS1, autoimmune polyendocrine syndrome type 1; CYP, cytochrome P450; UGT, uridine diphosphate-glucuronyltransferase.
Hepatitis in APS1
Autoimmune hepatitis
Dihydralazine hepatitis 2D6
UGT1 Addison disease
2A6
1A6
CYPs and UGTs
21
2E1
Halothane hepatitis
3A 11 17
Adrenal failure in APS1
Alcoholic liver disease
2C9 Anticonvulsant hepatitis
Ovarian failure in APS1
Tienilic acid hepatitis
PBC
AIH
Epitopes
321–351 419–429 257–269 373–389
1 LKM SLA LC, LP
Corticosteroid response
ANA, SMA
(PDH-E2/BCKD-E2)
E1
UDCA response
Bile duct lesions
Figure 40-2. Overlapping syndrome. Serological profiles of an overlapping syndrome between autoimmune hepatitis (AIH) and primary biliary cirrhosis (PBC). LKM, liver–kidney microsomal antibodies; SLA, soluble liver antigen; LC, LP, ANA, antinuclear antibodies; SMA, smooth-muscle antibodies; UDCA,
796
498
AMA E2/3
E4
E5
P450 2D6 -His-Arg-Met-Thr-Trp-Asp-Pro-Ala-Gln-Pro-Pr0-Arg-AspHSV (IE 175)
-Pro-Ala-Gln-Pro-Pro-Arg-Asp-
Figure 40-3. Sequence homologies. Sequence homologies of liver–kidney microsomal antibodies (LKM-1) autoepitopes with herpes virus proteins. Evidence of viral mimicry.
Chapter 40 AUTOIMMUNE HEPATITIS AND OVERLAP SYNDROMES
Table 40-2. Extrahepatic autoimmunologic disease associations of autoimmune hepatitis
Table 40-4. Key points
• Hematologic diseases
Gastrointestinal diseases Rheumatologic diseases
Endocrinologic diseases Others
Autoimmune hemolytic anemia Thrombocytopenic purpura Pernicious anemia Eosinophilia Inflammatory bowel diseasea Celiac disease Synovitisa Rheumatoid arthritis CREST syndrome Systemic sclerosis Sjögren’s syndrome Diabetes mellitus Autoimmune thyroid diseasea Proliferative glomerulonephritis Lichen planus Vitiligo Nail dystrophy Alopecia Uveitis Erythema nodosum
CREST, calcinosis, Raynaud’s phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia. a Most frequently observed.
Table 40-3. Clinical characteristics of autoimmune hepatitis
• •
•
•
About 25% of patients present with acute hepatitis and, rarely, with fulminant hepatitis More commonly, the clinical presentation of autoimmune hepatitis resembles that of other forms of chronic hepatitis and is therefore characterized by non-specific features such as fatigue, right upper quadrant pain, jaundice, mild pruritus, and arthralgias Histologically, there is periportal hepatitis with lymphocytic infiltrates, plasma cells, and piecemeal necrosis. Both patients who present acutely and those who present with chronic hepatitis often show histologic evidence of cirrhosis at the onset of symptoms The diagnosis of autoimmune hepatitis is established by the exclusion of other etiologies of chronic hepatitis and can be aided by using a revised numeric score that describes the probability of having the disease
injury. An autoimmune response can be triggered by the HLA class II-dependent presentation of specific antigenic peptides to T cells via antigen-presenting cells. As a response to cytokines, exposed T cells become activated and differentiate into Th1 or Th2 cells. Proinflammatory cytokines are believed to play an important role in the initiation of the autoimmune response.11 Owing to cytokine activation autoimmune diseases are often observed in associated viral or bacterial infections. Another mechanism is called “molecular mimicry.” As an example, a major B-cell epitope of cytochrome P450, which is targeted by LKM-1 autoantibodies in AIH-2, shares sequence homology with the herpes simplex virus antigen IE 175 (Figure 40.3).12 The immune response is normally tightly regulated. T- and B-cell homeostasis and the removal of autoreactive T cells are regulated by apoptosis. The failure of control by apoptosis may therefore contribute to the initiation and perpetuation of AIH and autoimmune overlap syndromes.13 This could be explained by the inability to kill
• • • • •
•
Autoimmune hepatitis (AIH) is characterized by female predominance, hypergammaglobulinemia, circulating autoantibodies, and a good response to immunosuppressive treatment There are three types of AIH: Type 1 represents the commonest form and is characterized by the presence of antinuclear antibodies (ANA) and/or anti-smooth-muscle antibodies (SMA) Type 2 is characterized by the presence of liver–kidney microsomal antibodies (LKM-1) AIH type 3 is characterized by autoantibodies against soluble liver antigen (SLA/LP) Overlap syndrome describes a disease condition in which clinical, biochemical, and serological features of AIH coexist with those of another autoimmune liver disease (primary biliary cirrhosis, primary sclerosing cholangitis) AIH and overlap syndromes are often associated with extrahepatic immune-mediated syndromes, including autoimmune thyroiditis, rheumatoid arthritis, and diabetes mellitus
Treatment and prevention • The standard initial treatment of AIH consists of prednisone monotherapy (50 mg/day and tapering regimen) or combination therapy with prednisone (30 mg/day) and azathioprine (1-2 mg/kg per day) • Combination therapy is generally preferred because it allows for the reduction of prednisone, frequently to a dose lower than 10 mg • Remission is achieved in 87% of patients within 3 years of treatment. However, a sustained response is only observed in 17% of patients • To prevent relapse episodes, treatment should not be terminated without histological evidence of a complete remission and drug withdrawal should proceed gradually over a 3–6-month period • In cases of treatment failure (~13% of cases), higher doses of prednisone and azathioprine can be considered or alternative drugs can be used (e.g., mycophenolate, MMF, ciclosporin A, cyclophosphamide, tacrolimus)
autoreactive cells, or by inducing autoimmunity against cellular constituents modified by apoptosis. Genetic factors have been implicated in the susceptibility to AIH. In particular, polymorphisms of genes influencing lymphocyte homeostasis, such as the tumor necrosis factor-a promoter gene (TNF-a), the complement factor C4 gene, and the CTLA-4 gene contribute to increased susceptibility to AIH.14 One hypothesis suggests that inheritance of specific HLA class II alleles modified at critical sites provides one of the crucial steps for the development of AIH. The relevance of genetic alterations for AIH is further underlined by the observation that chronic hepatitis occurs in 10–18% of patients with the autoimmune polyendocrine syndrome type 1 (APS1), an autosomal recessive disorder caused by mutations in a single gene (AIRE). APS-1 is characterized by various autoimmune diseases, mainly affecting endocrine glands.15 There has been no report of a genetic predisposition for overlap syndromes.
AUTOIMMUNE HEPATITIS PATHOLOGY AIH is characterized by periportal hepatitis with lymphocytic infiltrates, plasma cells, and piecemeal necrosis. A lobular hepatitis can
797
Section VI. Immune Diseases
Table 40-5. Autoimmune hepatitis: a diagnosis of exclusion 352-379
257-269 Suspected differential diagnosis Hepatitis C infection (HCV) Hepatitis B and D (HBV, HDV)
321-351
410-429
373-389
Figure 40-4. Three-dimensional structure.
be observed, but is only indicative of AIH in the absence of copper deposits or biliary inflammation. The presence of granulomas and iron deposits also argues against AIH.
Hepatitis A virus (HAV) Hepatitis E virus (HEV) Epstein–Barr virus (EBV) Herpes simplex virus (HSV) Cytomegalovirus (CMV) Varicella zoster virus (VZV) Drug-induced hepatitis
Primary biliary cirrhosis (PBC)
Primary sclerosing cholangitis Wilson disease
CLINICAL PRESENTATION About 25% of patients show an acute onset of AIH and rare cases of fulminant progression of AIH leading to acute liver failure have also been reported.16 Commonly, the clinical presentation of AIH resembles that of other forms of chronic hepatitis. AIH is therefore characterized by non-specific features, such as fatigue, right upper quadrant pain, jaundice, mild pruritus, arthralgias, and, when frank cirrhosis has developed, frequently also by spider angiomas and palmar erythema. In later stages, signs of portal hypertension, including ascites, bleeding esophageal varices, and encephalopathy, dominate this chronic progressive liver disease. Both patients who present acutely and those who present with chronic hepatitis often show histologic evidence of cirrhosis at the onset of symptoms, implying that subclinical disease is the diagnostic challenge because it often precedes the onset of disease symptoms. As many as 25% of patients initially show signs of decompensated liver cirrhosis.17 In particular, AIH (especially type 2) is associated with a wide variety of other disorders, most of which are of immunological origin (Table 40-2).
DIFFERENTIAL DIAGNOSIS The clinical presentation of AIH is indistinguishable from other causes of acute or chronic hepatitis, including liver cirrhosis. The clinical picture of AIH can therefore resemble those of viral hepatitis, drug-, or alcohol-induced hepatitis. Other entities in the differential diagnosis of AIH include non-alcoholic steatohepatitis (NASH), genetic hemochromatosis, and a1-antitrypsin deficiency. Cryptogenic hepatitis, an etiologically undefined chronic hepatitis, should also be included in the differential diagnosis of AIH. Patients with cryptogenic hepatitis are negative for viral as well as for autoantibody markers. It remains unknown how many of these patients suffer from AIH without detectable autoantibodies. In about 13% of patients who were initially negatively tested for ANA, SMA, and LKM, detection of SLA autoantibodies contributed to their clarification. These patients clinically resemble those of AIH type 1 with
798
Hemochromatosis
a1-Antitrypsin deficiency
Test performed to exclude Anti-HCV (HCV RNA) HBsAg, anti-HBc (HBV DNA) Anti-HDV, HDV RNA only when HBsAg-positive Antibodies, serology: IgG, IgM Only if suspected Only if suspected Only if suspected Only if suspected Only if suspected History; if applicable, withdrawal of drug LKM-2, LM autoantibody in selected cases Antimitochondrial antibodies (AMA) Specification of reactivity: PDH-E2, BCKD-E2 Liver histology: copper deposition in bile ducts Unresponsive to steroids Cholangiography (ERCP, MRCP) (PSC) Ceruloplasmin, urine copper, eye examination, quantitative copper in liver biopsy Serum ferritin, serum iron, transferrin saturation, HFE gene test (C282Y, H63D?) Liver histology: iron staining, quantitative iron in biopsy Serum a1-antitrypsin (if abnormal isoelectric focusing: PiZZ/PiSS/PiMZ/PiSZ genotype?)
IgG, immunoglobulin G; LKM-2, liver–kidney microsomal type 2; LM, liver microsomal; ERCP, endoscopic retrograde cholangiopancreatography; MRCP, magnetic resonance cholangiopancreatography.
respect to age and sex distribution, HLA antigen profile, inflammatory activity, and response to therapy. If cholestatic signs and immunoserological markers of AIH are present, overlap syndromes have to be included in the differential diagnosis.
DIAGNOSTIC METHODS Clinical characteristics are given in Table 40-3. The diagnosis of AIH is established by the exclusion of other etiologies of chronic hepatitis (Table 40-5). The diagnosis can be aided by the use of a revised numeric score that describes the probability of having the disease (Table 40-6). The sensitivity of the scoring system to establish definite or probable AIH is 89.9%. However, the specificity of the initial version of this score for discriminating AIH from overlapping syndromes such as AIH/PBC or AIH/PSC was low. In the revised AIH score, patients with histologic and cholangiographic evidence of PBC or PSC should be viewed as having variants of cholestatic diseases and not AIH. Specifically, well-defined granulomas, typical bile duct pathology of PSC, and PBC and substantial marginal bile duct proliferation with cholangiolitis and copper accumulation exclude AIH. Cholangiography is recommended for all patients who score as definite or probable AIH but do not respond to steroid treatment. Liver biopsy is generally recommended for grading and
Chapter 40 AUTOIMMUNE HEPATITIS AND OVERLAP SYNDROMES
Table 40-6. International diagnostic criteria for autoimmune hepatitis Parameter
Score
Gender Female +2 Male 0 Serum biochemistry Ratio of elevation of serum alkaline phosphatase versus aminotransferase >3.0 –2 1.53 0 <1.5 +2 Total serum globulin, g-globulin, or IgG Times upper normal limit >2.0 +3 1.5–2.0 +2 1.0–1.5 +1 <1.0 0 Autoantibodies (titers by immunofluorescence on rodent tissues) (adults) ANA, SMA, or LKM1 >1:80 +3 1:80 +2 1:40 +1 <1:40 0 Antimitochondrial antibody Positive –4 Negative 0 Hepatitis viral markers Negative +3 Positive –3 Other etiologic factors History of drug usage Yes –4 No +1 Alcohol (average consumption) <25 g/day +2 >60 g/day –2 Genetic factors: HLA DR3 or DR4 +1 Other autoimmune diseases +2 Response to therapy +2 Complete +3 Relapse +3 Liver histology Interface hepatitis +1 Predominant lymphoplasmacytic infiltrate +1 Rosetting of liver cells –5 None of the above –3 Biliary changes –3 Other changes +2 Seropositivity for other defined autoantibodies ANA, antinuclear antibodies; SMA, smooth-muscle antibody; LKM1, liver–kidney microsomal antibody type 1; HLA, human leukocyte antigen. For details refer to International Autoimmune Hepatitis Group Report. Review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999; 31:929–938.1 Interpretation of aggregate scores: definite AIH, greater than 15 before treatment and greater than 17 after treatment; probable AIH 10–15 before treatment and 12–17 after treatment.
staging of the disease and for decision-making regarding treatment of AIH, although a disease-specific histological feature does not exist for AIH and therefore limits its usefulness in the diagnosis of AIH.
TREATMENT AND PREVENTION The standard treatment of AIH is either prednisone monotherapy (50 mg/day and tapering regimen) or combination therapy with prednisone (30 mg/day) and azathioprine (1–2 mg/kg per day) (Table 40-5). Both are equally effective, although combination therapy is generally preferred because it allows for the reduction of prednisone, frequently to a dose lower than 10 mg, thereby reducing the steroid-associated unwanted side effects. Remission is achieved in 87% of patients within 3 years of treatment. However, a sustained response is only observed in 17% of patients after stopping treatment after an initial treatment period of at least 2 years. To prevent relapse episodes, treatment should not be terminated without histological evidence of a complete remission and drug withdrawal should proceed gradually over a 3–6-month period. Azathioprine monotherapy (2 mg/kg body weight) after prednisone withdrawal is a therapeutic option for the steroid-free maintenance of remission. The induction of remission by azathioprine monotherapy is not effective.
ALTERNATIVE TREATMENTS Treatment failure is marked by deterioration during therapy and occurs in 13% of patients. This situation justifies the termination of conventional therapy and the institution of high-dose regimens or use of alternative drugs. High-dose prednisone alone (60 mg/day) or a lower dose (30 mg/day) in combination with azathioprine (150 mg/day) induces biochemical remission in more than 60% of patients within 2 years. In cases of incomplete response, immunosuppression should be changed to alternative drugs. Drug-related adverse effects can be improved with dose reduction, and a 50% decrease in dose is the first course of action. Alternatively, in the maintenance of remission prednisone can be substituted for azathioprine monotherapy. Thus, side effects of steroids such as psychosis, diabetic decompensation, severe weight gain, and symptomatic osteopenia can be prevented by increasing the azathioprine dose or by azathioprine monotherapy. In order to maintain remission without or with little steroid side effects, topical steroids like budesonide are under investigation. If standard therapy fails, alternative drugs such as ciclosporin A, cyclophosphamide, mycophenolate mofetil, or tacrolimus can be considered. However, since these strategies have not been evaluated in randomized trials, they should only be administered after consultation with specialized hepatological centers. Liver transplantation remains as a therapeutic option for patients with treatment failure who do not reach remission despite therapy for years and who progress to cirrhosis with signs of decompensation.18 This procedure may also be used to rescue patients who present with fulminant hepatic failure secondary to AIH. The long-term outlook after liver transplantation is excellent, with 5-year survival rates of 92%. The recurrence of AIH after liver transplantation is independent of persistent autoantibodies and ranges between 11% and 35%.19 Individual adjustments of immunosuppressive therapy after transplantation in patients with AIH may be necessary to prevent or control the recurrence of AIH.
799
Section VI. Immune Diseases
OVERLAP SYNDROMES PATHOLOGY In addition to lymphocytic interface hepatitis, a characteristic feature of AIH, florid lesions of middle-sized bile ducts with portal inflammation and formation of granulomas can be observed in AIH/PBC overlap syndrome. The AIH/PSC overlap syndrome is characterized by lymphocytic interface hepatitis and fibrous obliterative cholangitis, the histologic hallmark of PSC.
CLINICAL PRESENTATION In an overlap syndrome, the classical appearance of the individual disease component is mixed with features of another autoimmune liver disease. Thus, in addition to the above-discussed non-specific symptoms such as chronic fatigue associated with AIH, clinical signs of cholestasis, including pruritus and jaundice, can occur with overlap syndromes.
DIFFERENTIAL DIAGNOSIS Depending on the leading symptoms, the differential diagnosis includes all forms of cholestatic and non-cholestatic liver diseases such as PBC, PSC, AIC, AIH, viral hepatitis, Wilson’s disease, hemochromatosis, and a1-antitrypsin deficiency.
DIAGNOSTIC METHODS The diagnosis of an overlap syndrome relies on the biochemical profile (either cholestatic with elevated alkaline phosphatase, g-glutamyltransferase, and bilirubin levels, or hepatocellular, with elevated aspartate aminotransferase and alanine aminotransferase levels in addition to elevated gammaglobulins), the histology, showing portal inflammation with or without the involvement of bile ducts, and the autoantibody profile, showing AMA directed against antigens of the oxoacid dehydrogenase complex (PDH-E2, BCKD-E2, OADC-E2 (PBC)) or pANCA (PSC), and autoantibodies associated primarily with AIH such as LKM, SLA/LP, or ANA. In cholestatic cases cholangiography detects sclerosing cholangitis. Immunglobulins are elevated in all autoimmune liver diseases; in PBC the elevation of immunoglobulin M is more pronounced.
of etiology once the presence of advanced cirrhosis has been ensured.3
PROGNOSIS WITH AND WITHOUT TREATMENT The natural history and prognosis of AIH are largely defined by the inflammatory activity present at diagnosis and, more importantly by the presence or development of cirrhosis. Patients with periportal hepatitis develop cirrhosis in 17% within 5 years. However, when bridging necrosis or necrosis of multiple lobules is present, cirrhosis develops in 82%. The presence of cirrhosis indicates a mortality of 58% in 5 years. However, the presence of cirrhosis at the beginning of treatment does not influence response or short-term outcome. The course of AIH is also significantly influenced by the HLA antigen profile of the affected individual. In this way HLA B8 antigen profile is associated with severe inflammation at presentation and a higher likelihood of relapse after treatment. Patients with HLA DR3 have a lower probability of reaching remission, show a higher relapse rate, and require transplantation more often. HLA DR4-positive individuals have a higher age of onset (or diagnosis) and a more benign outcome. In overlap syndromes with dominating features of PSC an increased risk (~20%) of developing cholangiocarcinoma exists. In PBC-dominating overlap syndromes the prognosis depends on the stage of cirrhosis according to the Child–Pugh criteria.
CONCLUSION AIH and overlap syndromes are chronic non-infectious diseases that require long-term treatment. If therapy is adequate, the prognosis of both AIH and overlap syndromes is excellent. Compared with other chronic liver diseases, AIH under remission is less frequently associated with the development of liver cirrhosis. During pregnancy close monitoring of disease activity in a hepatological center is recommended. If liver transplantation is required, the results are excellent. Treatment should not be terminated without histological evidence of complete remission and erratic modifications of the therapeutic regimen should not be performed during therapy.
REFERENCES FIRST TREATMENT AND PREVENTION As a general rule, the leading disease component is treated. In an overlap syndrome presenting as hepatitis, immunosuppression with prednisone (or combination therapy with azathioprine) is initiated. In cholestatic disease ursodeoxycholic acid (13–15 mg/kg body weight per day) is administered. Both treatments can be combined when biochemistry and histology suggest a relevant additional disease component.5
COMPLICATIONS AND THEIR MANAGEMENT It has been suggested that corticosteroid-resistant patients with AIH/PBC overlap syndrome benefit from ciclosporin A therapy. However, validated therapeutic guidelines for overlap syndromes and their complications are not yet available because of their low prevalence. As discussed above, liver transplantation is the treatment of choice in end-stage autoimmune liver diseases irrespective
800
1. International Autoimmune Hepatitis Group Report. Review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999; 31:929–938. 2. Manns MP, Gerken G, Kyriatsoulis A, et al. Characterization of a new subgroup of autoimmune chronic hepatitis by autoantibodies against a soluble liver antigen. Lancet 1987; 1:292–294. 3. Vogel A, Wedemeyer H, Manns MP, Strassburg CP. Autoimmune hepatitis and overlap syndromes. J Gastroenterol Hepatol 2002; 17 (suppl 3):S389–S398. 4. Jacobson DL, Gange SJ, Rose NR, Graham NMH. Epidemiology and estimated population burden of selected autoimmune diseases in the United States. Clin Immunol Immunopathol 1997; 84:223–243. 5. European Liver Transplant Registry 2001. www.ELTR.com. 6. Wiesner RH, Demetris AJ, Belle SH, et al. Acute hepatic allograft rejection: incidence, risk factors, and impact on outcome. Hepatology 1998; 28:638–645. 7. Czaja AJ. The variant forms of autoimmune hepatitis. Ann Intern Med 1996; 125:588–598.
Chapter 40 AUTOIMMUNE HEPATITIS AND OVERLAP SYNDROMES
8. Van Buuren HR, van Hoogstraten HJE, Terkivatan T, et al. High prevalence of autoimmune hepatitis among patients with primary sclerosing cholangitis. J Hepatol 2000; 33:543–548. 9. Chazouilleres O, Wendum D, Serfaty L, et al. Primary biliary cirrhosis–autoimmune hepatitis overlap syndrome: clinical features and response to therapy. Hepatology 1998; 28:296–301. 10. Czaja AJ, Carpenter HA. Autoimmune hepatitis with identical histologic features of bile duct injury. Hepatology 2001; 34:659–665. 11. Vergani D, Mieli-Vergani G. 2000. The role of T cells in autoimmune hepatitis. In: Manns MP, Paumgartner G, Leuschner U, eds. Immunology and liver. Dordrecht: Kluwer Academic Publishers; 2000:133–136. 12. Manns MP, Griffin KJ, Sullivan KF, Johnson EF. LKM-1 autoantibodies recognize a short linear sequence in P450IID6, a cytochrome P-450 monooxygenase. J Clin Invest 1991; 88:1370–1378. 13. Chervonsky AV. Apoptotic and effector pathways in autoimmunity. Curr Opin Immunol 1999; 11:684–688.
14. Agarwal K, Czaja AJ, Jones DE, Donaldson PT. Cytotoxic T lymphocyte antigen-4 (CTLA-4) gene polymorphisms and susceptibility to autoimmune hepatitis. Hepatology 2000; 31:49–53. 15. Obermayer-Straub P, Strassburg CP, Manns MP. Autoimmune polyglandular syndrome type 1. Clin Rev Allergy Immunol 2000; 18:167–183. 16. Nikias GA, Batts KP, Czaja AJ. The nature and prognostic implications of autoimmune hepatitis with acute presentation. J Hepatol 1994; 19:225–232. 17. Roberts SK, Therneau TM, Czaja AJ. Prognosis of histological cirrhosis in type 1 autoimmune hepatitis. Gastroenterology 1996; 110:848–857. 18. Tillmann HL, Jackel E, Manns MP. Liver transplantation in autoimmune liver disease: selection of patients. Hepatogastroenterology 1999; 46:3053–3059. 19. Manns MP, Bahr MJ. Recurrent autoimmune hepatitis after liver transplantation: when non-self becomes self. Hepatology 2000; 32:868–870.
801
Section VI: Immune Diseases
41
PRIMARY BILIARY CIRRHOSIS Jayant A. Talwalkar, Keith D. Lindor Abbreviations AFP a-fetoprotein ALT alanine aminotransferase AMA antimitochondrial antibody ANA antinuclear antibody AST aspartate aminotransferase BCOADC branched-chain ketoacid dehydrogenase complex BEC bile duct epithelial cells CT computed tomography ELISA enzyme-linked immunosorbent assay FDA US Food and Drug Administration FISH fluorescence in situ hybridization GTP g-glutamyltransferase HCC hepatocellular carcinoma
HDL HRT IAD IBD ICAM IIF IL LDL Lp MHC MRCP MRI
high-density lipoprotein hormone replacement therapy idiopathic adulthood ductopenia inflammatory bowel disease intracellular adhesion molecules indirect immunofluorescence interleukin low-density lipoprotein lipoprotein major histocompatibility magnetic resonance cholangiopancreatography magnetic resonance imaging
NFkB OGDC PBC PCR PDC PPAR-a PSC SMA SNP TNF UC UDCA
nuclear factor kB oxaloglutaric dehydrogenase complex primary biliary sclerosis polymerase chain reaction pyruvate dehydrogenase complex peroxisome proliferator-activated receptor-a primary sclerosing cholangitis smooth muscle antibody single nucleotide polymorphisms tumor necrosis factor ulcerative cholangitis ursodeoxycholic acid
INTRODUCTION
EPIDEMIOLOGY
Primary biliary cirrhosis (PBC) is a chronic cholestatic liver disease of unknown etiology, characterized histologically by the presence of portal inflammation and necrosis of the interlobular and septal bile ducts. Progressive bile duct destruction is associated with the development of portal and periportal inflammation, subsequent fibrosis, and eventual cirrhosis associated with liver failure and the complications of portal hypertension. It remains one of the major indications for liver transplantation worldwide. Evidence to date suggests that immunologic factors, which include T-cell activation directed at bile ducts, are involved in its pathogenesis. Highly specific autoantibodies reactive with biliary epithelial cell surface antigens and associated autoimmune disorders observed among patients with PBC have also been cited as further evidence for an immunologic cause. Patients affected by PBC are often asymptomatic middleaged women who are commonly diagnosed after evaluation for abnormal serum hepatic biochemical tests. Fatigue, pruritus, and/or unexplained hyperlipidemia may also be found at initial presentation, and may suggest a diagnosis of PBC. Antimitochondrial antibody (AMA) positivity is nearly diagnostic of PBC when present. The identification of PBC is important, as effective medical therapy with ursodeoxycholic acid (UDCA) has been demonstrated to halt disease progression and improve survival independent of liver transplantation. Therapeutic options for the medical complications of PBC, such as fatigue and metabolic bone disease, however, are not currently available. Mathematical models have been developed which accurately characterize and predict the natural history of PBC and assist in determining the optimal timing for liver transplantation if indicated.
PBC affects all races and has no specific geographic predilection. Women are primarily affected, with a female to male ratio of 9:1. The median age of disease onset is 50 years, but varies between 20 and 90 years. Previously thought to be a rare condition, recent investigations have observed that PBC is not so uncommon (Table 41-1). Estimates of annual incidence have been reported between 2 and 24 cases per million population. Prevalence estimates range from 19 to 240 cases per million population.2–4 Recent data from Olmsted County, Minnesota, suggest a stable incidence rate over the last 25 years but a higher prevalence than described in Canada. Overall incidence and prevalence rates of 27 and 402 cases per million population, respectively, are comparable to results from northern England. Differences in methodology and case definitions have made comparisons between investigations difficult.1,2 Recent interest has emerged regarding the seroepidemiology of AMA among patients without recognized PBC. In general populations, the frequency of AMA detection is estimated at 0.6%. In this study, all individuals were asymptomatic yet 50% were eventually found to have abnormal liver test results.3 The frequency of serum AMA positivity among patients evaluated for liver disease in a referral-based institution is 6.2%.4 In this setting, however, less than 40% of unselected patients will fulfill accepted diagnostic criteria for PBC.5 Notably, this prevalence rate is similar to the 5.7% rate of asymptomatic women with raised g-glutamyltransferase (GTP) levels who eventually were diagnosed with PBC.6 The variation in disease prevalence worldwide suggests that environmental risk factors are required for the phenotypic expression of PBC. From a population-based study of 100 PBC cases and 223 con-
803
Section VI. Immune Diseases
Table 41-1. Prevalence of Primary Biliary Cirrhosis by Geographic Region Region Australia Canada Northern England Olmsted County, USA
Prevalence (per million persons) 19 25 240 402
trols,7 the presence of tobacco use (odds ratio (OR) = 2.4) and history of psoriasis (OR = 4.6) were associated with PBC compared to controls. A history of eczema (OR = 0.13) was observed to be protective. The odds of having extrahepatic autoimmune disease among first-degree relatives of PBC cases were twice as great as in controls, but this difference was not statistically significant. A similar epidemiologic investigation8 examined 241 PBC cases against contemporary control groups (261 siblings and 225 friends, respectively). The prevalence rate of having at least one family member with PBC among identified cases was 6%. Mothers and sisters of PBC patients were more likely to have the disease than brothers, daughters, or sons. Coexisting autoimmune diseases among PBC patients included Sjøgren’s syndrome (17.4%), Raynaud’s phenomenon (12.5%), and autoimmune thyroid disease (11.5%), with significantly lower frequencies among siblings and friends. Compared to siblings, PBC patients were more likely to be female (OR = 4.2) and to have one or more autoimmune diseases (OR = 2.3), a history of shingles (OR = 2.7), previous tonsillectomy (OR = 2.5) and previous cholecystectomy (OR = 2.3). Similar comparisons between friends revealed strong associations with one or more autoimmune diseases (OR = 4.9), smoking (OR = 2.0), previous abdominal surgery (OR = 2.7) and previous tonsillectomy (OR = 1.9) among PBC cases. An increased rate of urinary tract infections among smokers with PBC has raised the possibility of an infectious etiology for PBC, manifested by the concept of molecular mimicry (see below).9
GENETICS The genetic predisposition to autoimmunity in PBC has been associated with alleles from the major histocompatibility (MHC) loci. No association between class I MHC loci and PBC has been found to date. A number of class II MHC loci, including DR8, DQA1*0102 and DQ/b1*0402, have been observed in patients with PBC.10 Similar findings are not observed in patients with AMA-negative PBC.11 Class III complement antigens C4 null and c4B2 alleles have also been described in association with PBC.10 The haplotypes DR3, DR8, and DR4 appear more prominent in Caucasian populations, in contrast to DR2 and DR8 haplotypes among Japanese subjects. The association between HLA-DR8 and PBC is most frequently observed, but is present in less than 40% of reported cases. Disease resistance has been associated with the DQA1*0102 haplotype. The use of linkage analysis methodologies to accurately identify susceptibility genes for PBC requires the prior identification of affected patients and families. The existence of mother–daughter cases of PBC suggests the involvement of non-MHC genes in PBC suscepti-
804
bility.12 Recently, a polymorphism involving exon 1 of the CTLA-4 gene has been identified as the first non-MHC susceptibility locus in PBC.13 However, this is also observed in patients with autoimmune thyroid disease and type 1 diabetes mellitus. Patients of Italian descent more commonly have single nucleotide polymorphisms (SNP) of the interleukin-10 gene promoter region, and this SNP occurs more frequently among patients with PBC of Italian nationality than in Japanese counterparts.14 No increased frequency of HLA-DR/DQ antigens and CTLA-4 genotypes among PBC patients from Brazil is observed. Tumor necrosis factor alleles were also not shown to influence disease progression.15 The presence of an increased familial risk for disease has been suggested as an indirect link to a genetic component for PBC susceptibility. Initial studies estimated the prevalence rate of PBC among first-degree relatives of index cases as low as 1.3%. Using serum AMA as a screening test among family members of PBC patients, a 2–4% prevalence rate of seropositivity has been observed among asymptomatic individuals. Criticisms regarding the generalizability of these investigations, however, have included the strict use of referred patients and the absence of appropriate control groups from similar populations. From a recent population-based investigation estimating the familial risk among 157 patients with definite or probable PBC, a positive family history was identified in 6.4% of instances. All familial cases were identified in women with PBC. An increased rate among first-degree relatives is due to a predominance of mother–daughter relationships. The overall prevalence rate among offspring of PBC patients was 1.2%, with a 2.3% rate among females. Given a population-based prevalence rate of 0.039% in this investigation, the relative risk for PBC among siblings was calculated at 10.5. Other relative risks (RR) for PBC among first-degree relatives (RR = 18.4), offspring (RR = 30.6), and daughters of women with PBC (RR = 58.7) have also been determined. For a populationbased prevalence rate of 0.05% among women over 40 years of age with PBC, the relative risk for PBC among their daughters was 15. Among six identified mother–daughter pairs, an earlier age of presentation (39.2 years for daughters vs 53.6 years for mothers, p < 0.05) was also observed.16
PATHOGENESIS IMMUNE-MEDIATED MECHANISMS Evidence to date strongly suggests that PBC is an immune-mediated process. The mechanisms that result in bile duct inflammation and fibrosis, however, remain incompletely understood. Both cellular and humoral abnormalities have been observed in patients affected by PBC. Aberrant expression of class II HLA phenotypes on bile duct epithelia have been reported but are non-specific, as similar findings occur in primary sclerosing cholangitis and, less often, biliary obstruction. Immunohistochemical staining of T lymphocytes located in areas of portal and periportal inflammation reveal a mixture of CD4+ and CD8+ T cells. Damage within bile ducts is also mediated by direct cytotoxic reactions from CD4+ and CD8+ T cells. Abnormal suppressor T-cell activity has also been observed in asymptomatic first-degree relatives of PBC subjects. B lymphocytes, however, are not commonly observed in areas of biliary injury. Specific lymphocyte subsets in peripheral blood, including autoreactive
Chapter 41 PRIMARY BILIARY CIRRHOSIS
T cells as well as abnormal concanavalin-A immunosuppression profiles, have been identified.17 Natural killer cells have also been identified to participate in the development of PBC.18 Intracellular adhesion molecules (e.g. ICAM-1) are expressed in areas of epithelial cell damage by lymphocytes and may play a role in the pathogenesis of PBC.19 Recent investigations have examined the role of serum cytokines in PBC, demonstrating the presence of increased levels of tumor necrosis factor-a (TNF-a), interleukin-8 (IL-8), and interleukin-12 (IL-12) in advanced as opposed to early histologic stages.20 The correlation between these findings and disease pathogenesis remains unknown. The major finding associated with humoral immunity resides with recognition of the antimitochondrial antibody (AMA). These autoantibodies have been associated with the E2 subunit of the pyruvate dehydrogenase complex (PDC-E2), the E2 unit of the branched-chain ketoacid dehydrogenase complex (BCOADC-E2), the E2 subunit of two oxaloglutaric dehydrogenase complexes (OGDC-E2), and E1 subunits of PDC and protein X. AMA is directed against the PDC E2 subunit in 90–95% of PBC patients. The reasons for AMA development against proteins along the inner surface of mitochondrial membranes and the role they play in injury to small intrahepatic bile ducts (rather than other tissues) is unknown. Recent data have demonstrated that these proteins share a common epitope with antigens in the cytoplasmic region of bile duct cells in PBC. Of note, there have been similar findings observed in PBC patients without AMA seropositivity, which remains incompletely understood.21 Arguments against direct cytotoxic activity from AMA include the persistence of antibody after liver transplantation without immediate disease recurrence; the absence of correlation between serum antibody titer and hepatic involvement; the absence of AMA in some patients with histologic confirmation of PBC; and the induction of AMA after recombinant PDC-E2 protein in animal models without resulting PBC.22 PBC is characterized by the breakdown of normal immune selftolerance to pyruvate dehydrogenase complex (PDC). Autoimmune responses to PDC have recently been studied in animal models. Oral dosing with PDC in SJL/J mice at both high and low doses was associated with a shift from Th1 to Th2 responses in a majority of animals.23 Sensitization with a covalently modified murine PDC was associated with high-affinity antibodies, suggesting a breakdown in tolerance to self PDC after antigenic exposure.24 Administration of bacterial DNA in this mouse model is also observed to enhance Th1 cytokine responses.25 Immunization with PDC is augmented in the presence of interferon-g. However, the finding of lymphocytic cholangitis on liver histology may be non-specific, given similar observations in control animals.26 The predominant immunoglobulin subtypes of AMA have been identified as IgG1 and IgG3, yet peripheral blood elevations of IgM appear most commonly in PBC. Errors in activating IgG synthesis or abnormal T-cell suppressor activity have been hypothesized as reasons for this discrepancy.27 Evidence supporting a potential contribution from IgA antibodies in the pathogenesis of PBC has emerged.28 Co-infection of cell lines with IgA monoclonal antibody to PDC shows great affinity to biliary epithelial cells of patients with PBC.29 An IgA-type AMA has also been detected in the urine of female patients.30 Serum IgA antibodies have been associated with bile duct loss and interface hepatitis on liver histology.31
The recent identification of apoptosis in association with chronic liver disease has also been extended to PBC.32 An increase in cholangiocyte Fas receptor expression33 with reductions in Bcl-2 from small bile ducts34 supports the involvement of apoptosis in disease pathogenesis. Evidence of antiapoptotic effects from tauroursodeoxycholic acid in human hepatocytes includes mitochondrial membrane stabilization35 and nuclear factor kB (NFkB) activation.36 In PBC, apoptosis of bile duct epithelial cells (BEC) (as detected by the TUNEL assay) was significantly increased compared to PSC or control livers.37 Proteolytic caspase enzymes subsequently create immunogenic fragments after apoptosis that contribute to the production of antimitochondrial antibodies.38 Release of PDC onto the cell surface during apoptosis may allow this antigen to stimulate immune system responses observed in PBC.39
NON-IMMUNE MEDIATED MECHANISMS The absence of self-tolerance to host antigens or proteins suggests that infection may be a potential trigger of PBC.40 A number of infectious agents including Escherichia coli, Helicobacter sp., and retroviruses have been implicated.41 The concept of ‘molecular mimicry’ by microbial antigens is proposed as the unifying mechanism to explain the cross-reactivity with self antigens that occurs via the immune system.9 The activation of T-cell clones involved in PBC using peptides from E. coli analogous to sequences within the PDC-E2 subunit has been demonstrated.42 In addition, serum antibodies to E. coli protease have been observed in 30% of PBC cases compared to 4% of controls.43 Anti-sp100 reactivity strongly correlates with AMA seropositivity in women with recurrent urinary tract infections, regardless of whether they have PBC, suggesting a pathogenic role for E. coli.44 Helicobacter pylori DNA has also been exclusively identified within hepatic tissue from PBC patients compared to controls.45 Similar findings among patients with primary sclerosing cholangitis (PSC), ulcerative colitis (UC), and other chronic liver diseases, however, limit the specificity of this observation. Colonization of biliary epithelium with H. pylori has not been observed in PBC.46 The significance of an increased prevalence of Gram-positive cocci DNA within the gallbladder bile of PBC subjects compared to controls (75% vs 5%) remains unknown.47 Recent attention has focused on the role of Chlamydia pneumoniae and the development of PBC. C. pneumoniae has been implicated in other chronic autoimmune disorders, such as multiple sclerosis. Among patients with early and advanced PBC, the frequency of seropositivity to C. pneumoniae was 100%, compared to 8% of liver disease controls. Eight explanted PBC livers were tested for C. pneumoniae 16S RNA by in situ hybridization, and all were positive.48 In contrast, PCR amplification of the Chlamydia-specific 16S rRNA gene was negative in 25 liver samples from patients with PBC.49 Evidence for an underlying viral infection is supported by electron microscopy of cholangiocytes and the increased frequency of serum antibodies to retroviral antigens in PBC. The triggering of abnormal PDC-E2 expression in cholangiocytes by an enveloped retrovirus has also recently been demonstrated.50 Cloning of exogenous retroviral nucleotide sequences from patients with primary biliary cirrhosis has been reported.51
805
Section VI. Immune Diseases
Additional support for the role of infection comes from recent work that has identified the first major histocompatibility complex (MHC) class I restricted epitope of PDC-E2 (amino acids 159–167 on E2 components).52 Antilipoic acid-specific antibodies were detected in 81% (79 of 97) of antimitochondrial antibody (AMA)positive patients with PBC but not in controls. This suggests that native or xenobiotic lipoic acid may stimulate and contribute to autoimmunity.53 Novel insights about the pathogenesis of PBC have also been reported. Enhanced intrahepatic NO synthetase expression with nitrotyrosine accumulation is associated with bile duct injury and progressive histologic fibrosis in PBC.54 Evidence for oxidative stress and lipid peroxidation as contributing features in PBC have recently been observed. Common features include reduction of intracellular glutathione and perinuclear expression of 4-hydroxynonenal in biliary epithelial cells.55 Intermediate filament cytoskeletal alterations and Mallory body formation in PBC appears to be related to bile acid-induced cell stress.56 The canalicular multidrug resistance proteins that mediate ATP-dependent biliary excretion of organic anions such as bilirubin diglucuronide may also influence disease pathogenesis in PBC.57 Impaired biliary secretion of bicarbonate in cholangiocytes isolated from patients with PBC involves transepithelial H(+)/HCO3(-) transport, providing a molecular basis for the impaired bicarbonate secretion in this cholestatic syndrome.58 Fetal microchimerism, defined as the persistence of cells in the maternal circulation after delivery, has also been hypothesized as an underlying mechanism for the increased association between female gender and autoimmune disease in general. In one study, 42% of PBC patients showed evidence for fetal microchimerism within liver tissue without any evidence in control tissue.59 Despite the use of polymerase chain reaction (PCR) and/or fluorescence in situ hybridization (FISH) techniques, no significant difference in Y chromosome-specific sequences in females with PBC and other liver diseases was found.60
CLINICAL FEATURES ASYMPTOMATIC PBC Individuals with asymptomatic disease comprise 20–25% of all PBC diagnoses, with prevalence rates as high as 60% based on the increased use of screening liver biochemistry profiles.61 Asymptomatic patients tend to be older than symptomatic counterparts at the time of diagnosis.62 The majority of asymptomatic patients with PBC, however, do appear to develop both symptoms and progression of hepatic disease over time.63 No specific features to predict the development of symptomatic disease, however, have been identified.62,63
SYMPTOMATIC PBC Symptoms attributed to PBC, including fatigue, pruritus, and keratoconjuctivitis xerostomia are the most commonly reported (Table 41-2). Other symptoms include right upper quadrant abdominal pain, nausea, and anorexia. A greater prevalence of symptomatic PBC in female as opposed to male individuals has been observed, without a clear explanation. Physical examination findings may
806
Table 41-2. Symptoms and Signs at Initial Presentation of Primary Biliary Cirrhosis Feature Asymptomatic Sicca syndrome Fatigue Pruritus Hepatomegaly Splenomegaly Jaundice Xanthelasma
Prevalence (%) 40 70 60 35 25 15 10 10
include the presence of hyperpigmentation, hepatosplenomegaly, and xanthelasmas. Though uncommon, the presence of steatorrhea in PBC is usually due to bile salt malabsorption, pancreatic exocrine insufficiency, or concomitant celiac disease. Osteopenia resulting in bone pain and/or spontaneous fractures is associated with metabolic bone disease, which is common in PBC. Jaundice is usually a late symptom that heralds the onset of advanced histologic disease, but was seen in 20% of cases from earlier series at the time of presentation. Complications from cirrhosis and portal hypertension, such as ascites, variceal bleeding, and hepatic encephalopathy, occur late in the course of disease.61
DIAGNOSIS BIOCHEMICAL FEATURES The most characteristic biochemical abnormality in PBC is an elevated serum alkaline phosphatase (usually three to four times the upper limit of normal).61 Subjects with a positive serum AMA and histology compatible with PBC may rarely have normal serum alkaline phosphatase levels.64 Modestly increased values for alanine aminotransferase (ALT) and aspartate aminotransferase (AST) are common. The finding of more significant elevations (>200 U/l) requires the exclusion of superimposed viral or drug-induced hepatic injury. Serum total bilirubin levels often rise during disease progression but are commonly within normal limits at diagnosis. Levels reaching 20 mg/dl are unusual, but can be associated with advanced hepatic disease. Elevations in serum total bilirubin, hypoalbuminemia, and prolongations in prothrombin time are associated with poor clinical outcome and often justify consideration for liver transplantation. Hypercholesterolemia is observed in up to 85% of cases at diagnosis. Serum IgM levels may also be elevated in patients with PBC.61
SEROLOGIC FEATURES Between 90% and 95% of patients with PBC will test positive for serum antimitochondrial antibody (AMA) in titers greater than or equal to 1:40. The finding of AMA seropositivity is not organ specific but remains highly sensitive (98%) as a diagnostic test.61 Reduced specificity and sensitivity for AMA detection by indirect immunofluorescence (IIF) is partially resolved with the use of enzyme-linked immunosorbent assays (ELISA).65 Conversely, three recent investigations describe a 13–17% prevalence rate for AMAnegative PBC, which is higher than previously accepted values.66,67
Chapter 41 PRIMARY BILIARY CIRRHOSIS
PBC patients may also exhibit serum antinuclear antibody (ANA) and/or smooth muscle antibody (SMA) in 35–66% of cases. The finding of serum ANA and clinical features suggestive of PBC in the absence of a positive AMA has been termed AMA-negative PBC. No evidence for cross-reactivity with ANA and the anti-M2 subunit of AMA has been observed.68,69 Serum anticentromere antibodies in PBC patients with the CREST syndrome in the absence of scleroderma have also been noted between 10% and 15% of the time. Other autoantibodies found in PBC include rheumatoid factor (70%) and antithyroid (antimicrosomal, antithyroglobulin) antibodies (40%).61,69 Antigp210 antibodies, which are formed against nuclear envelope proteins, have been declared to be highly specific and diagnostic of PBC. The prevalence rate is estimated at 25%. Anti-p62 antibodies were recently observed in association with advanced fibrosis in PBC.68,70
Stage I
Stage II
RADIOLOGIC FEATURES Ultrasonography or cross-sectional imaging with computed tomography (CT) or magnetic resonance imaging (MRI) is useful to exclude biliary tract obstruction in patients suspected to have PBC. Increased hepatic echogenicity and/or features of portal hypertension (splenomegaly, intra-abdominal varices, reversal of portal vein flow) may also be observed but are generally absent at diagnosis. Non-progressive periportal adenopathy detected by CT has been described in up to 88% of individuals with PBC.71 Large or bulky adenopathy warrants the exclusion of malignancy such as lymphoma or metastatic disease when present.
Stage III
HISTOLOGIC FEATURES Liver biopsy is needed to determine the stage of histologic disease at diagnosis and to confirm the existence of PBC in patients with persistently negative serum AMA testing. In the presence of serum alkaline phosphatase levels >1.5 times the upper limit of normal and AST values less than fivefold elevated, a liver biopsy is not essential for diagnosing PBC when serum AMA is positive.3 Histologic classification schemes developed by Ludwig and colleagues72 as well as Scheuer and associates73 are the most widely employed for staging PBC. Both systems describe the characteristic progression of liver injury in PBC that includes both focal and segmental destruction of intralobular bile ducts resulting in cholestasis and eventual biliary cirrhosis (Figure 41-1). Stage I PBC is associated with portal tract inflammation due to predominantly lymphoplasmacytic infiltrates, resulting in the destruction of septal and interlobular bile ducts up to 100 mm in diameter. Focal duct obliteration with granuloma formation has been termed the ‘florid duct lesion’ and is considered almost pathognomonic for PBC when present (Figure 41-2). Hepatic lobular involvement is uncommon at this stage of disease, but rare microgranulomas are seen in some cases. Most subjects with stage I PBC are clinically asymptomatic. Stage II PBC is consistent with the descriptions of periportal hepatitis by Ludwig et al.72 or ductular proliferation by Scheuer et al.73 An extension of the portal infiltrates observed in stage I disease to periportal areas is most commonly observed with associated interface hepatitis (piecemeal necrosis). Eosinophils may also be present within the inflammatory reaction. The histologic findings of chol-
Stage IV
Figure 41-1. Schematic representation of the staging system of primary biliary cirrhosis (Ludwig’s classification). Stage I is inflammation within the portal space, focused on the bile duct. Stage II is the inflammation extending into the hepatic parenchyma (interface hepatitis or piecemeal necrosis). Stage III is fibrosis, and stage IV is cirrhosis with regenerative nodules.
angitis, granulomas, and ductular proliferation are most commonly observed in stage II disease. Lobular involvement is similar to stage I disease. Individuals with stage II disease may be clinically asymptomatic, but at a lower frequency than in stage I involvement. Stage III PBC is dominated by the existence of septal or bridging fibrosis. The inflammatory features described with stage II disease are often seen in association with fibrosis spanning portal tracts. Ductopenia (defined as the loss of >50% of interlobular bile ducts)
807
Section VI. Immune Diseases
Figure 41-2 Florid duct lesion in stage I primary biliary cirrhosis.
becomes more common, resulting in cholestasis within periportal and paraseptal hepatocytes. Increased hepatic copper deposition beginning in stage II becomes more apparent. The majority of patients with stage III disease are clinically symptomatic. Stage IV disease in both classification systems is consistent with biliary cirrhosis. Nodular regeneration in association with features of stage III disease is observed, with a ‘garland’ shaped appearance that is characteristic of advanced PBC. Portal tract involvement with cholangitis can also be seen in the remaining bile ducts. Cholestatic abnormalities involving the lobular parenchyma, as seen in stage III, remain present. Most subjects considered for liver transplantation have stage IV disease. When histologic disease in PBC is staged, the most advanced finding must be used to determine the extent of involvement. The presence of both non-advanced and advanced histologic features in patients undergoing liver transplantation for PBC provides further evidence of the great sampling variability that is observed from liver biopsy.74
Table 41-3. Differential Diagnosis of Primary Biliary Cirrhosis Extrahepatic biliary tract obstruction Choledocholithiasis Strictures Malignancy Primary sclerosing cholangitis Drug-induced cholestasis (e.g. estrogens, phenothiazines) Granulomatous hepatitis Autoimmune hepatitis Chronic hepatitis C Alcoholic hepatitis Sarcoidosis Celiac disease
syndrome compared to patients with typical PBC has not been reported.
DIFFERENTIAL DIAGNOSIS OVERLAP SYNDROME WITH AUTOIMMUNE HEPATITIS Selected patients with PBC may also have clinical and histologic features compatible with autoimmune hepatitis. This situation has been described as an ‘overlap’ syndrome between the two conditions. No consensus diagnostic criteria have been agreed upon for overlap syndrome. When patients have at least two of three serologic and histologic features of each condition, the estimated frequency can be as high as 20%.75 Refinement of diagnostic criteria using the International Autoimmune Hepatitis Group revised classification system, however, has reduced the prevalence rate of this condition to <5%.76 Clinical experience is noted for similar responses to therapy with ursodeoxycholic acid as first-line therapy.77 Successful adjuvant therapy with corticosteroids in selected patients has been reported.75,78 Long-term data regarding prognosis with overlap
808
The patient who is diagnosed with PBC is typically a woman in the fifth or sixth decade of life with complaints of fatigue and/or pruritus. An increasing number of asymptomatic patients are diagnosed with PBC after the systematic evaluation of abnormal serum liver biochemistries. A number of other conditions to be excluded, however, require further investigation (Table 41-3). Extrahepatic mechanical biliary obstruction is the primary alternative diagnosis that must be excluded. Potential etiologies include choledocholithiasis, strictures, and neoplasms. Ultrasonography and cross-sectional imaging (CT, MRI) are sufficiently accurate in detecting biliary obstruction that invasive cholangiography is infrequently required. The technique of magnetic resonance cholangiopancreatography (MRCP) is over 90% accurate for assessing bile duct patency.79 Histologic findings of mechanical biliary obstruction, including edema, ductular proliferation, and neutrophilic inflamma-
Chapter 41 PRIMARY BILIARY CIRRHOSIS
tion rather than a lymphoplasmacytic infiltrate, assist in making the correct diagnosis. In primary sclerosing cholangitis (PSC) the finding of a positive serum AMA is rare. The presence of inflammatory bowel disease (IBD) is also highly suggestive of PSC, although rare instances of IBD with PBC have been reported. Histologic assessment is usually not specific enough to distinguish PSC from PBC. The finding of periductal fibrosis in PSC or granulomatous bile duct destruction in PBC, although pathognomonic for both conditions when present, occurs in only 10% of cases, respectively. Cholangiography is required to confirm the existence of PSC and is considered the diagnostic gold standard. Small duct PSC is considered a variant of PSC and is found in 5% of affected patients with IBD and evidence of chronic cholestasis.80 Similar clinical and biochemical findings associated with PBC can be observed from drug-induced hepatotoxicity. Serum hepatic biochemical profiles are usually consistent with a cholestatic hepatitis. An idiosyncratic mechanism of action is the most common etiology for liver injury. Offending medications include estrogens, androgenic steroids, phenothiazines, amoxicillin–clavulanate, itraconazole, trimethoprim–sulfamethoxazole, and phenytoin. In autoimmune hepatitis, the presence of significant inflammatory changes that involve adjacent bile ducts has been reported in up to 20% of cases and can mimic cholestatic liver disease. Serum AMA detection, however, occurs in only 8–25% of cases of autoimmune hepatitis, compared with 95% in PBC.81 Hepatic sarcoidosis is associated with a similar cholestatic biochemical profile as observed in PBC. Occasionally, portal hypertension resulting from presinusoidal fibrosis has also been observed. Histologic findings in hepatic sarcoidosis are characterized by extensive involvement with inflammatory granulomas in portal and periportal distributions. Chest radiographic abnormalities that suggest pulmonary involvement with sarcoidosis are found in 90% of cases.82 In rare instances, the identification of granulomatous bile duct inflammation in patients with chronic hepatitis C is observed. This occurs in the absence of fibrosing cholestatic disease, which typically occurs after liver transplantation. Serum AMA status is usually negative.83 Idiopathic adulthood ductopenia (IAD) is a disease affecting men with normal cholangiographic features in the absence of IBD. Serum autoantibodies (ANA, SMA, and/or AMA) are rarely positive. In contrast to small duct PSC, IAD is associated with a more rapid course of disease progression.84 AMA-negative PBC is observed among patients with cholestatic biochemical profiles and liver histology compatible with PBC in the absence of detectable serum AMA. Serum autoantibodies, including ANA, SMA, and anticarbonic anhydrase, are usually present. Of note, there appears to be no difference in natural history or responsiveness to ursodeoxycholic acid (UDCA) therapy in AMA-negative patients compared to AMA-positive patients with PBC. 85,86
EXTRAHEPATIC ASSOCIATED CONDITIONS As many as 70% of individuals with PBC have coexisting extrahepatic autoimmune disease states (Table 41-4). Secondary
Table 41-4. Systemic Conditions Associated with Primary Biliary Cirrhosis Feature Sjøgren’s syndrome Renal tubular acidosis Gallstones Arthritis Thyroid disease Scleroderma Raynaud’s phenomenon CREST syndrome Celiac disease
Prevalence (%) 70 50 30 20 20 15 10 5 4
Sjøgren’s syndrome resulting in keratoconjunctivitis sicca and xerostomia is the most prevalent autoimmune disease, occurring in about 70% of cases. Arthritis, including inflammatory joint disorders, has been observed in 10–40% of instances. Scleroderma or any component of the CREST syndrome (calcinosis, Raynaud’s phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasias) may be found in up to 10% of patients. Thyroid disease affects 15–20% of individuals and consists primarily of lymphocytic (Hashimoto’s) thyroiditis. Graves’ disease and hyperthyroidism are uncommon.154 Proximal or distal renal tubular acidosis occurs in as many as 50% of patients yet is often without clinical significance. Cholelithiasis is present in 30% of patients. Idiopathic pulmonary fibrosis and inflammatory bowel disease are each observed in less than 5% of individuals.61,80 A number of recent investigations have studied the frequency of asymptomatic celiac disease in patients with PBC. Seroepidemiologic data found prevalence rates between 1 and 4% for both antiendomysial and tissue transglutaminase antibodies. Using duodenal biopsy as the gold standard, less than 1% of these individuals are found to have histologic evidence of celiac disease. Given these data, the recommendation for universal screening for celiac disease in asymptomatic patients with PBC is controversial.87–89
DISEASE-RELATED COMPLICATIONS A number of systemic complications associated with PBC have been documented that represent disease progression and impair healthrelated quality of life in some individuals.
FATIGUE The prevalence of fatigue in PBC has been inconsistently reported. Natural history and controlled trial investigations over the past 25 years have reported the presence of fatigue in 0–76% of affected patients.90 Referral bias and systematic evaluation to document fatigue in these trials are probably responsible for this discrepancy in prevalence. For patients asymptomatic at the time of diagnosis, the cumulative risk for developing fatigue over a 5-year period is substantial, at nearly 50%.63 Recent studies have documented prevalence rates between 60% and 80% using multi-item validated questionnaires.91–94 Further investigation reveals that fatigue is independent of the severity of hepatic disease, sleep disturbance, or depression. Fatigue
809
Section VI. Immune Diseases
is not significantly improved with ursodeoxycholic acid therapy.91–94 Alterations in central neurotransmission and impaired corticotrophin-releasing hormone release are hypothesized as mechanisms of fatigue in PBC.95 A recent study that correlated fatigue with increased blood manganese levels and globus pallidus abnormalities on MRI requires confirmation.96 Identifying effective medical treatment specific to fatigue in PBC has received attention only recently. The empirical use of antioxidant therapy had no effect on fatigue scores in a randomized, crossover trial.97 Fatigue severity in PBC was not significantly improved with the use of ondansetron (a serotonin receptor antagonist) versus placebo.98 Despite implying that altered serotoninergic neurotransmission may influence fatigue, the use of selective serotonin reuptake inhibitors was ineffective in two small randomized trials.99,100
PRURITUS Pruritus is reported in 25–70% of patients affected by PBC.101 Recent data, however, suggest a decline in frequency that coincides with a greater proportion of asymptomatic and early-stage diagnoses.102 The annual risk for developing pruritus is estimated at 20%.63,102 The underlying pathogenesis of pruritus in PBC remains unknown. Recent hypotheses include the accumulation of serum bile acids and an increased release of endogenous opioids.101 Most patients report more severe symptoms at night than in daylight hours. In most cases, pruritus gradually resolves with progression of the hepatic disease. Serum alkaline phosphatase level and Mayo risk score are independent risk factors for the development of pruritus.102 Symptoms can be adequately controlled in the majority of patients if appropriate therapeutic measures are applied. Initial therapeutic attempts with antihistamines are often complicated by residual sedation. Cholestyramine (a bile acid-binding resin) may decrease the intensity of pruritus but rarely leads to complete resolution. In subjects with an intact gallbladder the use of 4 g cholestyramine before and/or after breakfast is thought to maximize bile acid sequestration, resulting in symptom improvement. The use of divided doses over a 24-hour period post cholecystectomy is more effective than once- or twice-daily dosing. Administration of cholestyramine several hours apart from other medications (including ursodeoxycholic acid) is recommended to prevent reductions in gastrointestinal absorption of these other drugs.61,101 Hepatic enzyme inducers, including phenobarbital and rifampin, have been associated with improvements in bile acid flow experimentally and are used for the treatment of pruritus. Excessive sedation at higher doses has limited the clinical utility of phenobarbital.101 Rifampin, in doses of 150–450 mg daily, is associated with a rapid onset of action and symptom relief. Although well tolerated, this medication is also associated with liver injury in 15% of cases and bone marrow aplasia on rare occasions.103 The drugs flumenicol and stanazolol have been reported to be of some benefit in PBC. 101 Parenteral naloxone and oral nalmefene have also been associated with the symptomatic improvement of pruritus in pilot and controlled trial settings. Intravenous propofol has also improved pruritus in selected patients. Initial reports of symptom improvement with use of the 5HT3 antagonist ondansetron await further
810
confirmation. S-adenosylmethionine, an antioxidant compound involved with glutathione synthesis, has been suggested to impart antipruritic effects, yet no clinical testing in PBC has so far taken place.101 From a secondary analysis of clinical trial data, a beneficial effect from sertraline has been observed. Prospective validation of these findings is awaited.104 In patients with pruritus refractory to medical therapy, liver transplantation is the most effective therapeutic option.
SICCA SYNDROME Despite receiving less attention in the literature than fatigue and pruritus, the occurrence of problematic symptoms due to keratoconjunctivitis sicca and xerostomia is reported in over 70% of patients (unpublished data). The majority of patients, however, do not satisfy criteria for the diagnosis of primary Sjøgren’s syndrome. In contrast, patients with classic features of Sjøgren’s syndrome and high-titer serum AMA have been recognized to develop PBC at a later date.105 Nonetheless, the secondary involvement of salivary and lacrimal glands with inflammation is thought to be the underlying cause. Treatment is directed at symptomatic improvement using artificial tears and oral sialagogues as first-line agents. Topical ciclosporin for keratoconjunctivitis sicca has not been tested in PBC.106 Oral pilocarpine and cemeviline, which have been used in patients with primary Sjøgren’s syndrome, are without extensive clinical experience in PBC.107,108
DYSLIPIDEMIA Hypercholesterolemia and hyperlipidemia are present in up to 85% of patients with PBC.61 This may be the initial serum biochemical abnormality in the asymptomatic patient with PBC. In early stage disease, lipoprotein (Lp) abnormalities are commonly found, including reduced Lp (a) levels. Initially, there is a marked elevation in high-density lipoprotein (HDL) rather than low-density lipoprotein (LDL), but this ratio is reversed with histologic disease progression. A reduction in serum cholesterol levels (specifically low-density lipoprotein) and an improvement in xanthelasma formation has been associated with ursodeoxycholic acid therapy.109 There is no clear correlation between xanthelasma formation and serum cholesterol levels in PBC. Statin-based therapy is effective for dyslipidemia in PBC, yet no large experience has been reported to date. No increased risk of death from cardiovascular disease is observed compared to the general population based on the severe hypercholesterolemia associated with PBC.110
METABOLIC BONE DISEASE Metabolic bone disease in PBC is related to osteopenia rather than osteomalacia among North American patients. Although a number of chronic liver disease states have recently been associated with osteopenia, the greater involvement in PBC is probably associated with subsequent cholestasis and its predilection for females who are independently at risk for metabolic bone disease. Decreased bone formation (osteoporosis) rather than increased bone resorption is felt to be the most important cause of the ostopenia. In premenopausal women with PBC, the defective formation of new bone (osteoblast activity) is also predominant. Increased osteoclastic
Chapter 41 PRIMARY BILIARY CIRRHOSIS
activity (bone resorption) can be present in premenopausal PBC patients and is accentuated by the postmenopausal state. Calcium and vitamin D metabolism is often normal in anicteric patients with PBC.111 Cigarette smoking is also associated with reduced serum vitamin D levels and may explain the elevated risk for osteopenia in these individuals.112 Approximately one-third of patients with PBC have osteopenia and 11% have osteoporosis, i.e. a Z score less than -2.5 by lumbar spine bone mineral densitometry. Among patients undergoing liver transplantation, a 20% worsening in bone mineral density may occur up to 6 months following surgery, which can significantly increase the risk for osseous fractures (particularly involving vertebral bodies).111 The increased risk for osteoporosis in PBC remains controversial, however, as other studies observe similar rate of bone loss as in age-matched healthy postmenopausal women.113 Risk factors for metabolic bone disease in PBC include age, body mass index, and stage 3 or 4 histologic disease.114 Among premenopausal women the risk for osteoporosis is significantly correlated with serum albumin and Mayo risk score.115 Ensuring adequate calcium intake (1000–1200 mg daily) and weightbearing activity is recommended as initial treatment for all patients. Measurement of serum vitamin D levels is also an essential part of the preventative management in PBC. The presence of vitamin D deficiency is variable among patients with metabolic bone disease but is based on malabsorption of fat-soluble vitamins. Oral replacement therapy is indicated if measured serum vitamin D levels are reduced. The 25-hydroxylation of vitamin D is intact in PBC patients; therefore, the use of vitamin D rather than 1,25-dihydroxyvitamin D or 25-hydroxyvitamin D is appropriate. Dosing is generally between 25 000 and 50 000 IU two to three times per week. The use of calcitonin, a drug that inhibits bone resorption, is not of proven benefit in PBC-related osteoporosis.116 Hormone replacement therapy (HRT) has been used for the prevention of osteoporosis in patients with PBC. Despite the cholestatic potential of higher-dose estrogens, HRT was found to be safe and appeared effective in several investigations, including a recently published randomized controlled trial.116,117 The potential risk of worsening jaundice and liver failure was not observed. Repeat biochemical evaluation at 2-week intervals for a duration of 6–8 weeks is advised if HRT treatment is to be initiated. Sodium fluoride has been associated with improved lumbar spine bone mineral density and increased bone formation in osteoporosis. Among subjects with PBC, the use of sodium fluoride in a placebocontrolled trial setting was not found to conclusively restore bone mineral density or impede the progression of osteoporosis. Escalating doses of therapy have dose-dependent gastrointestinal side effects, which limit its clinical usefulness. The use of bisphosphonates (including alendronate and etidronate) has been successfully tolerated in patients with osteoporosis. Among PBC patients, the existence of slight improvements in lumbar spine bone mineral density with etidronate have been reported but not confirmed in a randomized trial.116 Two recent placebo-controlled trials also confirm that oral alendronate safely increases bone mass in patients with PBC.118,119 For all studies to date in PBC, no data are available regarding the effect of treatment on incident of vertebral fractures.
STEATORRHEA Steatorrhea is a common finding in patients with advanced hepatic disease due to PBC. This finding has been attributed to a number of potential causes.61,80 Impairment of bile acid delivery and low critical micellar concentrations in the small intestine are the most common etiologies. The coexistence of untreated celiac disease has been reported as a cause of steatorrhea in PBC patients. Exocrine pancreatic insufficiency as a manifestation of overall glandular dysfunction is seen in selected individuals, causing diarrhea and fat malabsorption. Bacterial overgrowth syndrome is a potential cause of steatorrhea in PBC when associated with the presence of scleroderma and its variant forms. Determining the exact cause is important, as a variety of specific therapeutic options can be employed for symptom relief. In patients with decreased bile acid concentrations the oral replacement of medium-chain triglycerides with long-chain compounds coupled with an overall reduced fat intake is usually of benefit. Adherence to a gluten-free diet in celiac disease should improve symptoms as well. Pancreatic enzyme replacement therapy and rotating empiric antibiotic use are also of benefit with pancreatic insufficiency and bacterial overgrowth, respectively.
FAT-SOLUBLE VITAMIN DEFICIENCY Malabsorption of fat-soluble vitamins is common in PBC patients, especially those with advanced hepatic disease. The cause is related to intrahepatic cholestasis and impaired bile acid delivery to the small intestine. Vitamin A deficiency, observed in 20% of cases, is often clinically asymptomatic. When symptomatic, the presence of night blindness is observed but can be subtle. Oral replacement therapy should be initiated with 25 000–50 000 IU two to three times a week. As discussed previously, vitamin D deficiency can also occur and is the next most common fat-soluble vitamin deficiency. Vitamin E deficiency is rarely observed in individuals with PBC. When present, it may be associated with ataxia due to abnormalities of the posterior vertebral columns of the spinal cord. The alleviation of neurologic symptoms with parenteral vitamin E replacement is not universally effective, however. Oral replacement therapy is indicated for asymptomatic patients. Prolongation of the serum prothrombin time is most commonly associated with vitamin K deficiency. Significant hepatic dysfunction resulting in coagulopathy is observed in end-stage PBC as well. The initiation of water-soluble oral vitamin K using 5 mg tablets with repeat measurement of serum prothrombin time is effective in determining the extent of malabsorption. If correction of prothrombin time is achieved, daily oral doses between 5 and 10 mg/day should be initiated.61,80
CANCER The development of hepatocellular carcinoma (HCC) in cirrhoticstage PBC has been increasingly recognized worldwide. Age, male gender, and advanced histologic stage are associated with the development of HCC.120 The clinical effectiveness of HCC surveillance employing abdominal ultrasound with serum a-fetoprotein (AFP) levels every 6–12 months in end-stage PBC patients, however, remains unknown. An increased risk of extrahepatic malignancy such as breast cancer is also controversial.80 No association between
811
Section VI. Immune Diseases
PBC and endometrial carcinoma appears to exist.121 Although no evidence for an increased risk of colorectal neoplasia is associated with PBC, a recent study observed that the probability of recurrence was significantly lower in patients treated with UDCA than in controls (7% vs 28% at 3 years) following removal of colonic adenomatous polyps.122
DISEASE-MODIFYING THERAPIES Limitations in knowledge about the pathogenesis underlying PBC have been associated with few therapies that reverse or halt disease progression. Of all medical therapies attempted to date (Table 415), only ursodeoxycholic acid (UDCA) has been shown to be effective in patients with PBC, resulting in US Food and Drug Administration (FDA) approval in 1998.
IMMUNOSUPPRESSIVE AGENTS Corticosteroids Improvements in symptomatology, serum hepatic biochemistries, and histology occurred with the use of corticosteroids in 36 patients in a 1-year placebo-controlled trial.123 No difference in mortality was observed after 2 additional years, however.124 A primary limitation of long-term use is worsening osteopenia leading to osteoporosis. The successful use of bisphosphonate agents in PBC may renew interest in systemic corticosteroid therapy. Larger randomized controlled trials will be required to determine treatment safety and efficacy. In patients with
Table 41-5. Medical Therapies Evaluated in Primary Biliary Cirrhosis Medication
Primary action
Therapeutic benefit
Controlled trials D-Penicillamine Colchicine
Cupruretic Antifibrotic
Corticosteroids Cyclosporine Azathioprine Methotrexate UDCA
Immunosuppressive Immunosuppressive Immunosuppressive Immunosuppressive Immunomodulatory
None Biochemical, histologic (?) Clinical, biochemical Biochemical None Biochemical Biochemical, histologic
Uncontrolled trials Budesonide Malotilate Chlorambucil Thalidomide Simvastatin Pyruvate dehydrogenase Combination therapy Corticosteroids + colchicine UDCA + methotrexate UDCA + budesonide UDCA + silymarin UDCA + bezafibrate UDCA + sulindac UDCA + tetracycline UDCA + combivir
812
Immunosuppressive Immunomodulatory Immunomodulatory Immunomodulatory Immunomodulatory Immunomodulatory
Biochemical Biochemical None None Biochemical None
Compared to Historical controls
Biochemical
UDCA monotherapy No control group No control group UDCA monotherapy UDCA monotherapy No control group No control group
Similar to UDCA alone Biochemical None Biochemical Biochemical None Biochemical
AIH-PBC overlap syndrome there is evidence from anecdotal reports for biochemical and histological improvement using corticosteroids and/or combination therapy with ursodeoxycholic acid (UDCA).75
Azathioprine The successful use of azathioprine in autoimmune hepatitis has not been met with similar results in PBC. No improvements in symptoms, serum hepatic biochemistries, histology, or survival have been reported in two studies.125,126 A suggestion of improved survival after statistical recalculation from one trial,126 however, has not been met with widespread clinical use of this therapy.
Cyclosporine Pilot investigations of ciclosporin in PBC showed improvements in symptoms and biochemical features. A randomized trial including 19 subjects receiving cyclosporine (4 mg/kg/day) and 10 subjects receiving placebo was noted for improved symptoms and hepatic biochemistries after 1 year.127 Further examination in a large, randomized control trial of 346 patients with a median follow-up of 2.5 years,128 however, failed to reveal any histologic benefit despite biochemical improvement. Significant renal toxicity and hypertension was noted in all studies.
Methotrexate The initial use of oral methotrexate at 15 mg/week in two PBC patients was associated with symptomatic and biochemical improvement.129 Subsequent experience in nine patients in an openlabel investigation revealed histologic improvement in five and stability in four others after 17 months of therapy.130 One placebo-controlled trial has been reported involving a lower dose (7.5 mg/week), which resulted in biochemical improvement (not including total bilirubin) without an increase in survival. Of note, a significant number of patients with stage I–II histologic disease were enrolled in this study.131 Interstitial pneumonitis has been observed in association with methotrexate in 15% of PBC patients (compared to 3–5% in patients with rheumatologic disease), raising safety concerns for long-term clinical use.132 The risk of hepatic fibrosis commonly associated with cumulative doses of methotrexate in psoriatic arthritis has not been reported in PBC patients.
ANTIFIBROTIC AGENTS D-Penicillamine Based on abnormalities in copper excretion and the presence of significant concentrations in hepatic tissue, a number of clinical trials using D-penicillamine for PBC have been performed without recognizing substantial benefits. In the largest study reported, of 312 patients,133 as many as 20% developed serious drug-related side effects, including membranous glomerulonephritis in a few instances.
Colchicine A number of investigations have used colchicine in the treatment of PBC. Improvements in hepatic biochemical parameters were noted at doses of 1.2 mg/day. An increase in liver-related survival was also observed at 4 years of follow-up, but only after all placebo-treated patients were crossed over to colchicine.134 To date, no long-term
Chapter 41 PRIMARY BILIARY CIRRHOSIS
benefits have been demonstrated with the use of colchicine in spite of minimal toxicity. A recent meta-analysis failed to confirm either a beneficial or a detrimental effect from colchicine based on limitations in study methodological quality. 135
URSODEOXYCHOLIC ACID (UDCA) The mechanism of action for UDCA therapy in PBC is multifactorial. In addition to promoting endogenous bile acid secretion, there is evidence to suggest that UDCA is associated with membrane stabilization, reduced aberrant HLA type I expression on hepatocytes, and decreased cytokine production. The inhibition of apoptosis and mitochondrial dysfunction caused by exposure to hydrophobic bile acids are also prevented by UDCA.136
Responders Five randomized controlled trials of adequate size and duration have provided extensive information regarding the effectiveness of UDCA in PBC.137–141 Three studies employed doses of 13–15 mg/kg/day,137–139 one employed doses of 14–16 mg/kg/day,140 and the remaining investigation used 10–12 mg/kg/day.141 Improvements in symptoms and hepatic biochemical parameters were demonstrated in all five studies. UDCA may also stabilize or improve histopathological lesions in early-stage PBC.142 A combined analysis of three studies using UDCA (13–15 mg/kg/day) revealed improvements in transplant-free survival in patients receiving the active drug.143 Long-term (10-year) survival with UDCA has also been observed to exceed Mayo PBC model predictions in selected populations.144 The positive effects of UDCA on disease progression and survival free of liver transplantation have been questioned in a recent metaanalysis.145 By combining eight placebo-controlled trials involving 1114 patients, no differences in death or the need for liver transplantation between UDCA and placebo-treated patients was reported. The majority of identified studies, however, had followup periods of 2 years or less. A number of trials also used doses of UDCA at less than 13 mg/kg/day. With the exclusion of these investigations, a 32% risk reduction in death or need for liver transplantation with UDCA compared to placebo remains.146 Similar findings have been documented by independent methods for the effect on liver transplantation.147 Recent investigations also suggest that a long-term survival benefit is not observed in all populations receiving UDCA.148,149 Doses of UDCA above 20 mg/kg/day, however, do not appear to uniformly improve serum liver biochemistries despite optimal bile enrichment.150,151
Incomplete Responders Incomplete response is defined as the failure to normalize serum hepatic biochemistries and/or the development of cirrhosis while on UDCA. Between 60 and 70% of individuals will be categorized as incomplete responders after 6–12 months of therapy.152 A recent Markov model described a 17–27% risk for cirrhosis over 10 years in patients with early-stage histological disease. Predictors of incomplete response leading to cirrhosis include higher levels of serum alkaline phosphatase, serum total bilirubin, and lower serum albumin levels.153
Among patients experiencing a suboptimal response to UDCA, there are a number of potential extrahepatic causes that must be excluded. In addition to medication non-compliance or inappropriate dosing, the concomitant use of cholestyramine for pruritus may result in a lack of UDCA absorption. The diagnoses of autoimmune hypothyroidism and celiac disease must be eliminated as causes of elevated serum hepatic biochemistries. Finally, a lack of complete response to UDCA may suggest the presence of true overlap syndrome with autoimmune hepatitis. Evidence of overlap includes unexplained increased elevations in serum aminotransferases and significant interface hepatitis on liver histology.
COMBINATION THERAPIES UDCA plus Corticosteroids Reports from two randomized controlled trials showed reductions in serum hepatic biochemistry values, with mixed results relative to histologic improvement.154,155 Follow-up in both trials was short, ranging from 9 to 12 months. One study included the use of azathioprine at 50 mg/day.155 The use of oral budesonide in combination with UDCA has been variably associated with biochemical improvements.156,157 Conflicts between reported short-term histologic improvement156 and significant worsening of osteopenia157 with budesonide, however, have diminished interest in a long-term study of this medication. A recent investigation, however, demonstrated preliminary safety and efficacy in patients with early-stage rather than advanced PBC given budesonide.158
UDCA plus Colchicine No significant benefit from combination therapy with UDCA and colchicine has been demonstrated. Short durations of treatment (less than 2 years) and low doses of colchicine (approximately 1 mg/day) have been proposed as limitations. A recent investigation, however, reported reductions in the number of treatment failures, slower progression of Mayo risk score, and improvement in hepatic histology from UDCA with colchicine compared to patients receiving UDCA monotherapy.159 Long-term results from a previously reported study do not show an improved survival with combination therapy.160
UDCA plus Methotrexate A number of pilot investigations examining UDCA plus methotrexate have demonstrated no overall benefit. Data from recently published experiences fail to demonstrate greater efficacy with methotrexate and UDCA compared to UDCA alone.161 Results of a long-term multicenter randomized controlled trial comparing UDCA with combination therapy showed no additional benefit.162
NOVEL AGENTS Malotilate Malotilate is a compound associated with improved hepatic protein metabolism. In a randomized multicenter trial of 101 PBC patients given malotilate (n = 52) or placebo (n = 49), improvements in biochemical parameters occurred in the active treatment arm compared to placebo.163 No impact on disease progression or survival at the end of 2 years’ treatment was noted, however.
813
Section VI. Immune Diseases
Chlorambucil
Sulindac
In a randomized trial in 24 PBC patients using doses between 0.5 and 4 mg/day164 there was no significant improvement in biochemical or histologic parameters after 2–6 years of follow-up. Significant bone marrow suppression in four subjects required discontinuation of chlorambucil.
In 11 patients without complete response to UDCA, the combination of sulindac (100–300 mg/day) and UDCA (10–15 mg/kg/day) resulted in significant biochemical improvement compared to 12 patients remaining on UDCA monotherapy. An imbalance in baseline features with more advanced histological stage among UDCA monotherapy patients was observed.169
Thalidomide Thalidomide is a derivative of glutamic acid that selectively inhibits TNF-a production by monocytes. Among PBC patients given thalidomide (n = 10) or placebo (n = 8) in a small pilot trial,165 no improvements in biochemical or histologic parameters were observed. Side effects, including sedation and fatigue, caused two patients to discontinue active treatment.
Silymarin Silymarin is a drug that has been associated with hepatoprotective effects in experimental and clinical studies, especially in patients with chronic hepatitis C and alcoholic liver disease. In patients with an incomplete response to UDCA monotherapy, the use of silymarin with UDCA was not associated with significant benefits in an openlabel pilot investigation.166
Bezafibrate Bezafibrate is a hypolipidemic medication that stimulates the canalicular phospholipid pump MDR3 via binding and activation of transcription factor peroxisome proliferator-activated receptor-a (PPAR-a). This results in an increased biliary secretion of phospholipids, which are cytoprotective against bile salts. Bezafibrate, either alone or in combination with UDCA,167,168 in patients with PBC has been associated with similar improvements in serum hepatic biochemical values. Long-term studies, however, are required to confirm these early positive results.
Patients who became symptomatic Patients who remained asymptomatic Matched control population
Simvastatin From an open-label pilot investigation focusing on the lipid-lowering abilities of simvastatin (a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor), six patients were treated for 2 months with doses varying between 5 and 20 mg daily. Significant reductions in serum alkaline phosphatase and immunoglobulin M levels were observed, in addition to improved serum lipid levels. Further study is required to determine whether this class of agent will be useful for adjuvant therapy.170
Tetracycline Based on recent data suggesting a potential role for Chlamydia pneumoniae in the development of PBC,48 the use of 1 g tetracycline twice a day for 3 weeks was not associated with significant biochemical improvement in 15 patients with PBC. Liver histology in this pilot investigation was not a study endpoint.109
Pyruvate Dehydrogenase The concept of oral tolerance induction was tested using purified bovine pyruvate dehydrogenase complex (PDC) in six patients with early-stage PBC. After 6 months of therapy, no significant change in liver biochemistries or serum AMA titers was observed.172
Combivir Given the recognition of retrovirus particles in biliary epithelial cells of patients with PBC, a small pilot investigation of ziduvidine and lamivudine (Combivir) was performed in seven patients. In the absence of significant toxicity, some patients achieved complete normalization of serum alkaline phosphatase levels. A subsequent randomized controlled trial is being planned to confirm these findings.173
Proportion Surviving
1
NATURAL HISTORY AND PROGNOSIS
0.8 0.6
ASYMPTOMATIC PBC
0.4 0.2 0 0
5
10
15
20
Years of Followup Figure 41-3. Survival of asymptomatic patients and those who developed symptoms compared with an age- and gender-matched control population. (From Springer J, Cauch-Dudek K, O’Rourke K, et al. Asymptomatic primary biliary cirrhosis: a study of its natural history and prognosis. Am J Gastroenterol 1999;94:47, with permission.)
814
For asymptomatic patients at diagnosis, the cumulative risk for developing symptoms is 50% at 5 years and 95% at 20 years of follow-up.63 The median time from AMA positivity to persistent serum liver test elevations is 5.6 years.64 Among 10 of 29 patients who underwent repeat liver biopsy, a 40% rate of histologic disease progression was observed over a median of 11.4 years. Death from liver disease is less common in asymptomatic versus symptomatic patients. However, a lower overall median survival for asymptomatic patients is observed compared to an age- and sex-matched healthy population.62,63 Disease progression to end-stage liver failure is more likely to occur in symptomatic patients. From a large survey study
Chapter 41 PRIMARY BILIARY CIRRHOSIS
of asymptomatic Japanese patients with PBC, the variables of serum total bilirubin, albumin, cholesterol, histological stage, and the presence or absence of UDCA treatment were found to be useful prognostic indicators.174 Serum AMA status does not influence prognosis.175
SYMPTOMATIC PBC The development of cirrhosis and complications from portal hypertension occurs in symptomatic individuals with PBC. Prognostic indicators of poor outcome include older age, elevated serum total bilirubin levels, depressed hepatic synthetic function, and advanced histologic stage. Among PBC patients followed for a median of 3.4 years there is a cumulative risk of 40% for developing esophageal varices.176 Bleeding from esophageal varices occurs in 20% of cases and is associated with an increased risk of mortality. The identification of esophageal varices in precirrhotic PBC results from presinusoidal hepatic fibrosis caused by granulomatous bile duct inflammation and portal edema.
patients179 and includes independent adjusted predictor variables such as patient age, serum total bilirubin, albumin, prothrombin time, and the presence or absence of edema or ascites. The calculation of a risk score from these variables can be translated into a survival function to estimate mortality risk in an individual patient. Increases in risk score are associated with decreases in survival. The absence of histologic information for predicting survival offers the advantage of avoiding invasive procedures such as liver biopsy. Additional capabilities for the model include the monitoring of efficacy from experimental drug therapy in clinical trials and determining the optimal timing for liver transplantation. Subsequent model refinement using time-dependent methods based on serial assessments of clinical status has improved the accuracy for predicting short-term survival from PBC within 2 years from time of assessment.183 The Mayo risk score has also been shown to retain its ability for predicting survival in UDCA-treated patients. The prolonged survival free of liver transplantation associated with UDCA appears to remain effective for 10 years or more. 184
PROGNOSTIC SURVIVAL MODELS In the majority of patients with PBC a progressive clinical course resulting in fibrosis and eventual cirrhosis is often observed.177–180 Estimates of overall median survival range between 10 and 15 years from the time of diagnosis, whereas advanced histologic disease (stage 3 or 4) imparts a median survival approaching 8 years.181 Elevations in total bilirubin above 8–10 mg/dl have been associated with a median life expectancy of 2 years.182 To account for these clinical variables as determinants of survival, a number of mathematical models simulating the natural history of PBC have been developed and refined for clinical use. Among the prognostic models reported to date (Table 41-6),179–182 the Mayo Clinic formula has undergone the most extensive crossvalidation and is widely referred to for predicting long-term survival in PBC. This model is based on the serial follow-up of 312 PBC
LIVER TRANSPLANTATION The most effective therapeutic alternative for patients with endstage PBC is liver transplantation. Indications for referral include accepted criteria required for all hepatic disease etiologies, as well as refractory complications of portal hypertension (variceal bleeding, ascites requiring frequent large-volume paracentesis, hepatic encephalopathy, spontaneous bacterial peritonitis, hepatorenal syndrome, and hepatopulmonary syndrome). Severe fatigue, intractable pruritus, and severe pain from vertebral body compression fractures are also considered indications for liver transplantation. Increasing disparities in donor organ availability and recipient need have significantly hampered the ability to perform liver transplantation for
Table 41-6. Prognostic Models in Primary Biliary Cirrhosis Authors
Predictive variables
Formula
Extramural validation
Dickson et al.179
Age Total bilirubin Serum albumin Prothrombin time Edema score Bleeding varices Bilirubin Bilirubin Ascites Albumin Age GI bleeding Central cholestasis Cirrhosis IgM UDCA treated Bilirubin Prothrombin time Procollagen III Hyaluronic acid
R = 0.871 loge (age in years) -2.53 loge (albumin in g/dl) + 0.039 loge (bilirubin in mg/dl) + 2.38 loge (prothrombin time in seconds) + 0.859 (edema) loge R = 1.68 (bleeding -0.25) + 2.03 loge (bilirubin -30.3) Calculated from pocket chart/tables
Yes
Not stated
No
Rydning et al.180 Christensen et al.181
Poupon et al.182
No Yes
815
Section VI. Immune Diseases
impairments in health-related quality of life such as fatigue and pruritus. PBC remains among the most frequent indications for liver transplantation in the US. Patient and graft survival rates from liver transplantation are reported to approach 92% and 85% at 1-year and 5-year intervals, respectively.185 Recent data examining the impact of liver transplantation in PBC patients showed an association between improved clinical and economic outcomes with a pretransplant Mayo risk score <7.8.186 Hepatic retransplantation occurs in less than 10% of PBC subjects. Survival after retransplantation is significantly reduced when performed more than 30 days after initial surgery in PBC, and is associated with increases in resource utilization.187 No difference in outcome after liver transplantation in those transplanted during the UDCA era compared to their pre-UDCA counterparts has been observed.188 Previously considered a controversial topic, the recurrence of PBC following liver transplantation has now been demonstrated. Initial reports of allograft histologic features consistent with stage I PBC were limited by the absence of explicit criteria for recurrent disease, including the exclusion of acute cellular rejection. It has now been estimated that the rate of recurrent PBC may be as high as 15% at 3 years and 30% at 10 years with prospective follow-up. Serum AMA status appears to be independent of recurrence risk. This antibody may disappear soon after transplantation, only to return later with or without recurrent disease. The tapering of corticosteroids from immunosuppression regimens following liver transplantation does not appear to be a clear risk factor for recurrent PBC. Recent published studies document a shorter time to recurrence with the use of tacrolimus rather than ciclosporin as primary immunosuppression. No information is available regarding the efficacy of UDCA therapy in halting disease progression from early-stage recurrent PBC.185
CONCLUSIONS Primary biliary cirrhosis is an important hepatic disease that affects middle-aged women. Although data regarding the pathogenesis of disease continue to emerge, much remains unknown about the interaction between host factors and immune system dysregulation that results in hepatic disease. The differential diagnosis is broad and PBC is often associated with other extrahepatic autoimmune conditions. Disease-specific complications, including fatigue, pruritus, and metabolic bone disease, are important to recognize and treat appropriately. Among various medical therapies intended to halt disease progression, only ursodeoxycholic acid (UDCA) has been associated with an increase in transplant-free survival. Prognostic models are available to provide accurate predictions of survival even with UDCA therapy. The natural history of PBC is usually indolent, but often accelerates once symptoms develop. Liver transplantation is an effective therapeutic modality for patients with end-stage liver disease from PBC, with excellent patient and graft survival rates.
REFERENCES 1. Prince MI, James OF. The epidemiology of primary biliary cirrhosis. Clin Liver Dis 2003;7:795–819.
816
2. Kim WR, Lindor KD, Locke GR, et al. Epidemiology and natural history of primary biliary cirrhosis in a U.S. community. Gastroenterology 2000;119:1631. 3. Shibata M, Onozuka Y, Morizane T, et al. Prevalence of antimitochondrial antibody in Japanese corporate workers in Kanagawa prefecture. J Gastroenterol 2004;39:255–259. 4. Zein CO, Angulo P, Lindor KD. When is liver biopsy needed in the diagnosis of primary biliary cirrhosis? Clin Gastroenterol Hepatol 2003;1:89–95. 5. Zuber MA, Recktenwald C. Clinical correlation of antimitochondrial antibodies. Eur J Med Res 2003;8:61–70. 6. Sakugawa H, Nakasone H, Nakayoshi T, et al. Epidemiology of primary biliary cirrhosis among women with elevated gammaglutamyl transpeptidase levels in Okinawa, Japan. Hepatol Res 2003;26:330–336. 7. Howel D, Fischbacher CM, Bhopal RS, et al. An exploratory population-based case–control study of primary biliary cirrhosis. Hepatology 2000;31:1055. 8. Parikh-Patel A, Gold EB, Worman H, et al. Risk factors for primary biliary cirrhosis in a cohort of patients from the United States. Hepatology 2001;33:16. 9. Bogdanos DP, Baum H, Grasso A, et al. Microbial mimics are major targets of crossreactivity with human pyruvate dehydrogenase in primary biliary cirrhosis. J Hepatol 2004;40:31–39. 10. Gershwin ME, Ansari A, Mackay IR, et al. Primary biliary cirrhosis: an orchestrated immune response against biliary epithelial cells. Immunol Rev 2000,174:210. 11. Stone J, Wade JA, Cauch-Dudek K, et al. Human leukocyte antigen class II associations in serum antimitochondrial antibodies (AMA)-positive and AMA-negative primary biliary cirrhosis. J Hepatol 2002;36:8–13. 12. Mackay IR. Autoimmunity and primary biliary cirrhosis. Baillière’s Best Prac Res Clin Gastroenterol 2000;14:519. 13. Agarwal K, Jones DE, Daly AK, et al. CTLA-4 polymorphism confers susceptibility to primary biliary cirrhosis. J Hepatol 2000;32:538. 14. Matsushita M, Tanaka A, Kikuchi K, et al. Association of single nucleotide polymorphisms of the interleukin-10 promoter gene and susceptibility to primary biliary cirrhosis: immunogenetic differences in Italian and Japanese patients. Autoimmunity 2002;35:531–536. 15. Bittencourt PL, Palacios SA, Farias AQ, et al. Analysis of major histocompatibility complex and CTLA-4 alleles in Brazilian patients with primary biliary cirrhosis. J Gastroenterol Hepatol 2003;18:1061–1066. 16. Jones DEJ, Watt FE, Metcalf JV, et al. Familial primary biliary cirrhosis reassessed: a geographically-based population study. J Hepatol 1999;30:402. 17. Ichiki Y, Shimoda S, Ishibashi H, et al. Is primary biliary cirrhosis a model autoimmune disease? Autoimmun Rev 2004;3:331–336. 18. Kita H, Naidenko OV, Kronenberg M, et al. Quantitation and phenotypic analysis of natural killer T cells in primary biliary cirrhosis using a human CD1d tetramer. Gastroenterology 2002;123:1031–1043. 19. Yokomori H, Oda M, Yoshimura K, et al. Expression of intercellular adhesion molecule-1 and lymphocyte functionassociated antigen-1 protein and messenger RNA in primary biliary cirrhosis. Intern Med 2003;42:947–954. 20. Neuman M, Angulo P, Malkiewicz I, et al. Tumor necrosis factoralpha and transforming growth factor-beta reflect severity of liver damage in primary biliary cirrhosis. J Gastroenterol Hepatol 2002;17:196–202. 21. Jones DE. Autoantigens in primary biliary cirrhosis. J Clin Pathol 2000;53:813–821. 22. Krams SM, Surh CD, Coppel RI, et al. Immunization of experimental animals with dehydrolipoamide acyltransferase, as
Chapter 41 PRIMARY BILIARY CIRRHOSIS
23.
24.
25.
26.
27.
28. 29.
30.
31.
32. 33. 34.
35.
36.
37.
38.
39.
40. 41. 42.
43.
a purified recombinant polypeptide, generates mitochondrial autoantibodies but not primary biliary cirrhosis. Hepatology 1989;9:411. Jones DE, Palmer JM, Robe A, et al. Oral tolerisation to pyruvate dehydrogenase complex as a potential therapy for primary biliary cirrhosis. Autoimmunity 2002;35:537–544. Palmer JM, Robe AJ, Burt AD, et al. Covalent modification as a mechanism for the breakdown of immune tolerance to pyruvate dehydrogenase complex in the mouse. Hepatology 2004;39:1583–1592. Jones DE, Palmer JM, Burt AD, et al. Bacterial motif DNA as an adjuvant for the breakdown of immune self-tolerance to pyruvate dehydrogenase complex. Hepatology 2002;36: 679–686. Sasaki M, Allina J, Odin JA, et al. Autoimmune cholangitis in the SJL/J mouse is antigen non-specific. Dev Immunol 2002;9:103–11. Gershwin ME, Ansari AA, Mackay IR, et al. Primary biliary cirrhosis: an orchestrated immune response against epithelial cells. Immunol Rev 2000;174:210–225. Floreani A, Baragiotta A, Pizzuti D, et al. Mucosal IgA defect in primary biliary cirrhosis. Am J Gastroenterol 2002;97:508–510. Fukushima N, Nalbandian G, Van De Water J, et al. Characterization of recombinant monoclonal IgA anti-PDC-E2 autoantibodies derived from patients with PBC. Hepatology 2002;36:1383–1392. Tanaka A, Nalbandian G, Leung PS, et al. Mucosal immunity and primary biliary cirrhosis: presence of antimitochondrial antibodies in urine. Hepatology 2000;32:910. Masuda J, Omagari K, Ohba K, et al. Correlation between histopathological findings of the liver and IgA class antibodies to 2-oxo-acid dehydrogenase complex in primary biliary cirrhosis. Dig Dis Sci 2003;48:932–938. Rust C, Gores GJ. Apoptosis and liver disease. Am J Med 2000;108:567. Matsunaga Y, Terada T. Mast cell subpopulations in chronic inflammatory hepatobiliary diseases. Liver 2000;20:152. Iwata M, Harada K, Kono N, et al. Expression of Bcl-2 familial proteins is reduced in small bile duct lesions of primary biliary cirrhosis. Hum Pathol 2000;31:179. Benz C, Angermuller S, Otto G, et al. Effect of tauroursodeoxycholic acid on bile acid-induced apoptosis in primary human hepatocytes. Eur J Clin Invest 2000;30:203. Rust C, Karnitz LM, Paya CV, et al. The bile acid taurochenodeoxycholate activates a phosphatidylinositol 3-kinasedependent survival signaling cascade. J Biol Chem 2000;275:20210. Tinmouth J, Lee M, Wanless IR, et al. Apoptosis of biliary epithelial cells in primary biliary cirrhosis and primary sclerosing cholangitis. Liver 2002;22:228–234. Matsumura S, Van De Water J, Kita H, et al. Contribution to antimitochondrial antibody production: cleavage of pyruvate dehydrogenase complex-E2 by apoptosis-related proteases. Hepatology 2002;35:14–22. Macdonald P, Palmer J, Kirby JA, et al. Apoptosis as a mechanism for cell surface expression of the autoantigen pyruvate dehydrogenase complex. Clin Exp Immunol 2004;136:559–567. Haydon GH, Neuberger J. PBC: an infectious disease? Gut 2000;47:586. Trauner M, Boyer JL. Cholestatic syndromes. Curr Opin Gastroenterol 2001;17:242. Shigematsu H, Shimoda S, Nakamura M, et al. Fine specificity of T cells reactive to human PDC-E2 163-176 peptide, the immunodominant autoantigen in primary biliary cirrhosis: implications for molecular mimicry and cross-recognition among mitochondrial autoantigens. Hepatology 2000;32:901. Nilsson I, Lindgren S, Eriksson S, et al. Serum autoantibodies to Helicobacter hepaticus and Helicobacter pylori in patients with chronic liver disease. Gut 2000;46:410.
44. Bogdanos DP, Baum H, Butler P, et al. Association between the primary biliary cirrhosis specific anti-sp100 antibodies and recurrent urinary tract infection. Dig Liver Dis 2003;35:801–805. 45. Nilsson HO, Taneera J, Castedal M, et al. Identification of Helicobacter pylori and other Helicobacter species by PCR, hybridization, and partial DNA sequencing in human liver samples from patients with primary sclerosing cholangitis or primary biliary cirrhosis. J Clin Microbiol 2000;38:1072. 46. Myung SJ, Kim MH, Shim KN, et al. Detection of Helicobacter pylori DNA in human biliary tree and its association with hepaticolithiasis. Dig Dis Sci 2000;45:1405. 47. Hiramatsu K, Harada K, Tsuneyama K, et al. Amplification and sequence analysis of partial bacterial 16S ribosomal RNA gene in gallbladder bile from patients with primary biliary cirrhosis. J Hepatol 2000;33:9. 48. Abdulkarim AS, Petrovic LM, Kim WR, et al. Primary biliary cirrhosis: an infectious disease caused by Chlamydia pneumoniae? J Hepatol 2004;40:380–384. 49. Leung PS, Park O, Matsumura S, et al. Is there a relation between Chlamydia infection and primary biliary cirrhosis? Clin Dev Immunol 2003;10:227–233. 50. Xu L, Shen Z, Guo L, et al. Does a betaretrovirus infection trigger primary biliary cirrhosis? Proc Natl Acad Sci USA. 2003;100:8454–8459. 51. Xu L, Sakalian M, Shen Z, et al. Cloning the human betaretrovirus proviral genome from patients with primary biliary cirrhosis. Hepatology 2004;39:151–156. 52. Matsumura S, Kita H, He XS, et al. Comprehensive mapping of HLA-A0201-restricted CD8 T-cell epitopes on PDC-E2 in primary biliary cirrhosis. Hepatology 2002;36:1125–1134. 53. Bruggraber SF, Leung PS, Amano K, et al. Autoreactivity to lipoate and a conjugated form of lipoate in primary biliary cirrhosis. Gastroenterology 2003;125:1705–1713. 54. Sanz-Cameno P, Medina J, Garcia-Buey L, et al. Enhanced intrahepatic inducible nitric oxide synthase expression and nitrotyrosine accumulation in primary biliary cirrhosis and autoimmune hepatitis. J Hepatol 2002;37:723–729. 55. Tsuneyama K, Harada K, Kono N, et al. Damaged interlobular bile ducts in primary biliary cirrhosis show reduced expression of glutathione-S-transferase-pi and aberrant expression of 4hydroxynonenal. J Hepatol 2002;37:176–183. 56. Fickert P, Trauner M, Fuchsbichler A, et al. Mallory body formation in primary biliary cirrhosis is associated with increased amounts and abnormal phosphorylation and ubiquitination of cytokeratins. J Hepatol 2003;38:387–394. 57. Kullak-Ublick GA, Baretton GB, Oswald M, et al. Expression of the hepatocyte canalicular multidrug resistance protein (MRP2) in primary biliary cirrhosis. Hepatol Res 2002;23:78–82. 58. Melero S, Spirli C, Zsembery A, et al. Defective regulation of cholangiocyte Cl-/HCO3(-) and Na+/H+ exchanger activities in primary biliary cirrhosis. Hepatology 2002;35:1513–1521. 59. Fanning PA, Jonsson JR, Clouston AD, et al. Detection of male DNA in the liver of female patients with primary biliary cirrhosis. J Hepatol 2000;33:690. 60. Schoniger-Hekele M, Muller C, Ackermann J, et al. Lack of evidence for involvement of fetal microchimerism in pathogenesis of primary biliary cirrhosis. Dig Dis Sci 2002;47:1909–1914. 61. Talwalkar JA, Lindor KD. Primary biliary cirrhosis. Lancet 2003;362:53–61. 62. Springer J, Cauch-Dudek K, O’Rourke K, et al. Asymptomatic primary biliary cirrhosis: a study of its natural history and prognosis. Am J Gastroenterol 1999;94:47. 63. Prince MI, Chetwynd A, Craig WL, et al. Asymptomatic primary biliary cirrhosis: clinical features, prognosis, and symptom progression in a large population based cohort. Gut 2004;53:865–870.
817
Section VI. Immune Diseases
64. Mitchison HC, Bassendine MF, Hendrick A, et al. Positive antimitochondrial antibody but normal alkaline phosphatase: is this primary biliary cirrhosis? Hepatology 1986;6:1279. 65. Tanaka A, Miyakawa H, Luketic VA, et al. The diagnostic value of anti-mitochondrial antibodies, especially in primary biliary cirrhosis. Cell Mol Biol 2002;48:295–299. 66. Kananathan R, Suresh RL, Merican I. Primary biliary cirrhosis at Hospital Kuala Lumpur: a study of 17 cases seen between 1992 and 1999. Med J Malaysia 2002;57:56–60. 67. Golovnova EV, Il’chenko LIu, Tsaregorodtseva TM, et al. Primary biliary cirrhosis: 12-year experience in observation. Ter Arkh 2003;75:26–30. 68. Muratori P, Muratori L, Ferrari R, et al. Characterization and clinical impact of antinuclear antibodies in primary biliary cirrhosis. Am J Gastroenterol 2003;98:431. 69. Heathcote EJ. Management of primary biliary cirrhosis. Hepatology 2000;31:1005–1013. 70. Miyachi K, Hankins RW, Matsushima H, et al. Profile and clinical significance of anti-nuclear envelope antibodies found in patients with primary biliary cirrhosis: a multicenter study. J Autoimmun 2003;20:247–54. 71. Blachar A, Federle MP, Brancatelli G. Primary biliary cirrhosis: clinical, pathologic, and helical CT findings in 53 patients. Radiology 2001;220:329–36. 72. Ludwig J, Dickson ER, McDonald GS. Staging of chronic nonsuppurative destructive cholangitis (syndrome of primary biliary cirrhosis). Virchows Arch 1978;379:103. 73. Scheuer PJ. Primary biliary cirrhosis: Chronic non-suppurative destructive cholangitis. Am J Pathol 1965;46:387. 74. Garrido MC, Hubscher SG. Accuracy of staging in primary biliary cirrhosis. J Clin Pathol 1996;49:556–559. 75. Chazouilleres O, Wendum D, Serfaty L, et al. Primary biliary cirrhosis–autoimmune hepatitis overlap syndrome: clinical features and response to therapy. Hepatology 1998;28:296. 76. Talwalkar JA, Keach JC, Angulo P, et al. Overlap of autoimmune hepatitis and primary biliary cirrhosis: an evaluation of a modified scoring system. Am J Gastroenterol 2002;97:1191–1197. 77. Joshi S, Cauch-Dudek K, Wanless IR, et al. Primary biliary cirrhosis with additional features of autoimmune hepatitis: response to therapy with ursodeoxycholic acid. Hepatology 2002;35:409–413. 78. Gunsar F, Akarca US, Ersoz G, et al. Clinical and biochemical features and therapy responses in primary biliary cirrhosis and primary biliary cirrhosis–autoimmune hepatitis overlap syndrome. Hepatogastroenterology 2002;49:1195–200. 79. Romagnuolo J, Bardou M, Rahme E, et al. Magnetic resonance cholangiopancreatography: a meta-analysis of test performance in suspected biliary disease. Ann Intern Med 2003;139:547–557. 80. Angulo P, Lindor KD. Primary biliary cirrhosis and primary sclerosing cholangitis. Clin Liver Dis 1999;3:529–570. 81. Czaja AJ, Carpenter HA. Autoimmune hepatitis with incidental histologic features of bile duct injury. Hepatology 2001;34:659. 82. Keeffe EB. Sarcoidosis and primary biliary cirrhosis. Am J Med 1987;83:977. 83. Farges O, Bismuth H, Sebagh M, et al. Granulomatous destruction of bile ducts after liver transplantation: primary biliary cirrhosis recurrence or hepatitis C virus infection? Hepatology 1995;21:1765–1767. 84. Angulo P, Maor-Kendler Y, Lindor KD. Small-duct primary sclerosing cholangitis: a long-term follow-up study. Hepatology 2002;35:1494–1500. 85. Inverizzi P, Crosignani A, Battezzati PM, et al. Comparison of the clinical features and clinical course of antimitochondrial antibody-positive and -negative primary biliary cirrhosis. Hepatology 1997;25:1090. 86. Kim WR, Poterucha JJ, Jorgensen RA, et al. Does antimitochondrial antibody status affect response to treatment in
818
87.
88.
89.
90. 91. 92.
93.
94.
95. 96.
97.
98.
99.
100.
101.
102. 103.
104.
105.
106.
patients with primary biliary cirrhosis? Outcomes of ursodeoxycholic acid therapy and liver transplantation. Hepatology 1997;26:22. Volta U, Rodrigo L, Granito A, et al. Celiac disease in autoimmune cholestatic liver disorders. Am J Gastroenterol 2002;97:2609–2613. Floreani A, Betterle C, Baragiotta A, et al. Prevalence of coeliac disease in primary biliary cirrhosis and of antimitochondrial antibodies in adult coeliac disease patients in Italy. Dig Liver Dis 2002;34:258–261. Bizzaro N, Villalta D, Tonutti E, et al. IgA and IgG tissue transglutaminase antibody prevalence and clinical significance in connective tissue diseases, inflammatory bowel disease, and primary biliary cirrhosis. Dig Dis Sci 2003;48:2360–2365. Leuschner U. Primary biliary cirrhosis – presentation and diagnosis. Clin Liver Dis 2003;7:741–758. Cauch-Dudek K, Abbey S, Stewart DE, et al. Fatigue in primary biliary cirrhosis. Gut 1998;43:705–710. Huet PM, Deslauriers J, Tran A, et al. Impact of fatigue on the quality of life of patients with primary biliary cirrhosis. Am J Gastroenterol 2000;95:760–767. Prince MI, James OF, Holland NP, et al. Validation of a fatigue impact score in primary biliary cirrhosis: towards a standard for clinical and trial use. J Hepatol 2000;32:368–373. Goldblatt J, Taylor PJ, Lipman T, et al. The true impact of fatigue in primary biliary cirrhosis: a population study. Gastroenterology 2002;122:1235–1241. Swain MG. Fatigue in chronic disease. Clin Sci 2000;99:1–8. Forton DM, Patel N, Prince M, et al. Fatigue and primary biliary cirrhosis: association of globus pallidus magnetisation transfer ratio measurements with fatigue severity and blood manganese levels. Gut 2004;53:587–592. Prince MI, Mitchison HC, Ashley D, et al. Oral antioxidant supplementation for fatigue associated with primary biliary cirrhosis: results of a multicentre, randomized, placebocontrolled, cross-over trial. Aliment Pharmacol Ther 2003;17:137–143. Theal J, Toosi MN, Girlan LM, et al. A randomized, controlled crossover trial of ondansetron in patients with primary biliary cirrhosis and fatigue. Hepatology 2005;41:1305–1312. Ter Borg PC, Van Os E, Van Den Broek WW, et al. Fluvoxamine for fatigue in primary biliary cirrhosis and primary sclerosing cholangitis: a randomised controlled trial. BMC Gastroenterology 2004;4:13. Talwalkar JA, Jorgensen RA, Keach JC, et al. Fluoxetine for the treatment of fatigue in primary biliary cirrhosis. Dig Dis Sci 2004; (in press). Bergasa NV, Mehlman JK, Jones EA. Pruritus and fatigue in primary biliary cirrhosis. Best Prac Res Clin Gastroenterol 2000;14:643. Talwalkar JA, Souto E, Jorgensen RA, et al. Natural history of pruritus in primary biliary cirrhosis. Clin Gastroenterol Hepatol 2003;1:297–302. Prince MI, Burt AD, Jones DE. Hepatitis and liver dysfunction with rifampicin therapy for pruritus in primary biliary cirrhosis. Gut 2002;50:436–439. Browning J, Combes B, Mayo MJ. Long-term efficacy of sertraline as a treatment for cholestatic pruritus in patients with primary biliary cirrhosis. Am J Gastroenterol 2003;98:2736–2741. Csepregi A, Szodoray P, Zeher M. Do autoantibodies predict autoimmune liver disease in primary Sjogren’s syndrome? Data of 180 patients upon a 5 year follow-up. Scand J Immunol 2002;56:623–629. Sall K, Stevenson OD, Mundorf TK, et al. Two multicenter, randomized studies of the efficacy and safety of ciclosporin ophthalmic emulsion in moderate to severe dry eye disease. Ophthalmology 2000;107:631–639.
Chapter 41 PRIMARY BILIARY CIRRHOSIS
107. Vivino FB, Al-Hashimi I, Khan Z, et al. Pilocarpine tablets for the treatment of dry mouth and dry eye symptoms in patients with Sjogren syndrome: a randomized, placebo-controlled, fixed-dose, multicenter trial. Arch Intern Med 1999;159: 174–181. 108. Petrone D, Condemi JJ, Fife R, et al. A double-blind, randomized, placebo-controlled study of cevimeline in Sjogren’s syndrome patients with xerostomia and keratoconjunctivitis sicca. Arthritis Rheum 2002;46:748–754. 109. Del Puppo M, Galli Kienle M, Crosignani A, et al. Cholesterol metabolism in primary biliary cirrhosis during simvastatin and UDCA administration. J Lipid Res 2001;42:437–441. 110. Longo M, Crosignani A, Battezzati PM, et al. Hyperlipidaemic state and cardiovascular risk in primary biliary cirrhosis. Gut 2002;51:265–269. 111. Hay JE. Osteoporosis in liver diseases and after liver transplantation. J Hepatol 2003;38:856–865. 112. Brot C, Jorgensen NR, Sorensen OH. The influence of smoking on vitamin D status and calcium metabolism. Eur J Clin Nutr 1999;53:920. 113. Newton J, Francis R, Prince M, et al. Osteoporosis in primary biliary cirrhosis revisited. Gut 2001;49:282–287 114. Menon KV, Angulo P, Weston S, et al. Bone disease in primary biliary cirrhosis: independent indicators and rate of progression. J Hepatol 2001;35:316–323. 115. Solerio E, Isaia G, Innarella R, et al. Osteoporosis: still a typical complication of primary biliary cirrhosis? Dig Liver Dis 2003;35:339–346. 116. Vleggaar FP, van Buuren HR, Wolfhagen FH, et al. Prevention and treatment of osteoporosis in primary biliary cirrhosis. Eur J Gastroenterol Hepatol 1999;11:617–621. 117. Pereira SP, O’Donohue J, Moniz C, et al. Transdermal hormone replacement therapy improves vertebral bone density in primary biliary cirrhosis: results of a 1-year controlled trial. Aliment Pharmacol Ther 2004;19:563–570. 118. Guanabens N, Pares A, Ros I, et al. Alendronate is more effective than etidronate for increasing bone mass in osteopenic patients with primary biliary cirrhosis. Am J Gastroenterol 2003;98:2268–2274. 119. Zein CO, Jorgensen RA, Clarke B, et al. Alendronate improves bone mineral density in patients with primary biliary cirrhosis: randomized placebo-controlled trial. Gastroenterology 2004;126:A671. 120. Shibuya A, Tanaka K, Miyakawa H, et al. Hepatocellular carcinoma and survival in patients with primary biliary cirrhosis. Hepatology 2002;35:1172–1178. 121. Floreani A, Paternoster D, Mega A, et al. Sex hormone profile and endometrial cancer risk in primary biliary cirrhosis: a case–control study. Eur J Obstet Gynecol Reprod Biol 2002;103:154–157. 122. Serfaty L, De Leusse A, Rosmorduc O, et al. Ursodeoxycholic acid therapy and the risk of colorectal adenoma in patients with primary biliary cirrhosis: an observational study. Hepatology 2003;38:203–209. 123. Mitchison HC, Bassendine MF, Malcolm AJ, et al. A pilot, double-blind controlled 1-year trial of prednisolone treatment in primary biliary cirrhosis. Hepatic improvement but greater bone loss. Hepatology 1989;10:420. 124. Mitchison HC, Palmer JM, Bassendine MF, et al. A controlled trial of prednisolone treatment in primary biliary cirrhosis: Three-year results. J Hepatol 1992;15:336. 125. Crowe J, Christensen E, Smith M, et al. Azathioprine in primary biliary cirrhosis: A preliminary report of an international trial. Gastroenterology 1980;78:1005. 126. Christensen E, Neuberger J, Crowe J, et al. Beneficial effect of azathioprine and prediction of prognosis in primary biliary cirrhosis: Final results of an international trial. Gastroenterology 1985;89:1034.
127. Wiesner RH, Ludwig J, Lindor KD, et al. A controlled trial of cyclosporine in the treatment of primary biliary cirrhosis. N Engl J Med 1990;322:1419. 128. Lombard M, Portmann B, Neuberger J, et al. Cyclosporin A treatment in primary biliary cirrhosis: Results of a long-term placebo controlled trial. Gastroenterology 1993;104:519. 129. Kaplan MM, Knox TA, Arora S. Primary biliary cirrhosis treated with low-dose oral pulse methotrexate. Ann Intern Med 1988;109:429. 130. Kaplan MM, Knox TA. Treatment of primary biliary cirrhosis with low-dose weekly methotrexate. Gastroenterology 1991;101:1332. 131. Hendrickse MT, Rigney E, Giaffer MH, et al. Low-dose of methotrexate is ineffective in primary biliary cirrhosis: long-term results of a placebo-controlled trial. Gastroenterology 1999;117:400. 132. Sharma A, Provenzale D, McKusick A, et al. Interstitial pneumonitis after low-dose methotrexate therapy in primary biliary cirrhosis. Gastroenterology 1994;107:266. 133. Dickson ER, Fleming TR, Wiesner RH, et al. Trial of penicillamine in advanced primary biliary cirrhosis. N Engl J Med 1985;312:1011. 134. Kaplan MM, Alling DW, Zimmerman HJ, et al. A prospective trial of colchicine for primary biliary cirrhosis. N Engl J Med 1986;215:1448. 135. Gong Y, Gluud C. Colchicine for primary biliary cirrhosis. Cochrane Database Syst Rev 2004;2:CD004481. 136. Lazaridis KN, Gores GJ, Lindor KD. Ursodeoxycholic acid mechanisms of action and clinical use in hepatobiliary disorders. J Hepatol 2001;35:134–146. 137. Lindor KD, Dickson ER, Baldus WP, et al. Ursodeoxycholic acid in the treatment of primary biliary cirrhosis. Gastroenterology 1994;106:1284. 138. Poupon RE, Balkan B, Eschwege E, et al. A multicenter, controlled trial of Ursodiol for the treatment of primary biliary cirrhosis. N Engl J Med 1991;324:1548. 139. Heathcote EJ, Cauch-Dudek K, Walker V, et al. The Canadian multicenter, double-blind, randomized controlled trial of ursodeoxycholic acid in primary biliary cirrhosis. Hepatology 1994;19: 1149. 140. Pares A, Caballeria L, Rodes J, et al. Long-term effects of ursodeoxycholic acid in primary biliary cirrhosis: results of a double-blind, controlled multicentric trial: the UDCACooperative Group from the Spanish Association for the Study of the Liver. J Hepatol 2000;32:561. 141. Combes B, Carithers RL, Maddrey WC, et al. A randomized, double-blind, placebo-controlled trial of ursodeoxycholic acid in primary biliary cirrhosis. Hepatology 1995;22:759. 142. Poupon RE, Lindor KD, Pares A, et al. Combined analysis of the effect of treatment with ursodeoxycholic acid on histologic progression in primary biliary cirrhosis. J Hepatol 2003;39:12–16. 143. Poupon RE, Lindor KD, Cauch-Dudek K, et al. Combined analysis of randomized controlled trials of ursodeoxycholic acid in primary biliary cirrhosis. Gastroenterology 1997;113:884. 144. Poupon RE, Bonnand AM, Chretien Y, et al. Ten-year survival in ursodeoxycholic acid treated patients in primary biliary cirrhosis. Hepatology 1999;29:1668. 145. Goulis J, Leandro G, Burroughs AK. Randomised controlled trials of ursodeoxycholic acid therapy for primary biliary cirrhosis: a meta-analysis. Lancet 1999;354:1053. 146. Poupon RE. Ursodeoxycholic acid for primary biliary cirrhosis: lessons from the past–issues for the future. J Hepatol 2000;32:689. 147. Gluud C, Christensen E. Ursodeoxycholic acid for primary biliary cirrhosis. Cochrane Database Syst Rev 2002;1:CD000551. 148. Papatheodoridis GV, Hadziyannis ES, Deutsch M, et al. Ursodeoxycholic acid for primary biliary cirrhosis: final results of
819
Section VI. Immune Diseases
149.
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
820
a 12-year, prospective, randomized, controlled trial. Am J Gastroenterol 2002;97:2063–2070. Combes B, Luketic VA, Peters MG, et al. Prolonged follow-up of patients in the U.S. multicenter trial of ursodeoxycholic acid for primary biliary cirrhosis. Am J Gastroenterol 2004;99:264–268. Angulo P, Dickson ER, Therneau TM, et al. Comparison of three doses of ursodeoxycholic acid in the treatment of primary biliary cirrhosis: a randomized trial. J Hepatol 1999;30:830. Roda E, Azzaroli F, Nigro G, et al. Improved liver tests and greater biliary enrichment with high dose ursodeoxycholic acid in early stage primary biliary cirrhosis. Dig Liver Dis 2002;34:523–527. Jorgensen R, Angulo P, Dickson ER, et al. Results of long-term ursodiol treatment for patients with primary biliary cirrhosis. Am J Gastroenterol 2002;97:2647–2650. Corpechot C, Carrat F, Poupon R, et al. Primary biliary cirrhosis: incidence and predictive factors of cirrhosis development in ursodiol-treated patients. Gastroenterology 2002;122:652–658. Leuschner M, Gultdutuna S, You T, et al. Ursodeoxycholic acid and prednisolone versus ursodeoxycholic acid and placebo in the treatment of early stages of primary biliary cirrhosis. J Hepatol 1996;25:39. Wolfhagen FHJ, van Hooganstraten HJF, van Buuren HR, et al. Triple therapy with ursodeoxycholic acid, prednisone, and azathioprine in primary biliary cirrhosis: a 1-year, randomized placebo-controlled study. J Hepatol 1998;29:736. Leuschner M, Maier K-M, Schlichting J, et al. Oral budesonide and ursodeoxycholic acid for the treatment of primary biliary cirrhosis: results of a prospective, double-blind trial. Gastroenterology 1999;117:918. Angulo P, Smith C, Jorgensen R, et al. Budesonide in the treatment of patients with primary biliary cirrhosis with suboptimal response to ursodeoxycholic acid. Hepatology 1999;20:471. Hempfling W, Grunhage F, Dilger K, et al. Pharmacokinetics and pharmacodynamic action of budesonide in early- and late-stage primary biliary cirrhosis. Hepatology 2003;38:196–202. Almasio PL, Floreani A, Chiaramonte M, et al. Multicenter randomized placebo-controlled trial of ursodeoxycholic acid with or without colchicine in symptomatic primary biliary cirrhosis. Aliment Pharmacol Ther 2000;14:1645. Kaplan MM, Cheng S, Price LL, et al. A randomized controlled trial of colchicine plus ursodiol versus methotrexate plus ursodiol in primary biliary cirrhosis: ten-year results. Hepatology 2004;39:915–923. Bach N, Bodian C, Bodenheimer H, et al. Methotrexate therapy for primary biliary cirrhosis. Am J Gastroenterol 2003;98:187–193. Combes B, Emerson SS, Flye NL. The primary biliary cirrhosis ursodiol plus methotrexate or its placebo study – a multicenter randomized trial. Hepatology 2003;38(S1):210A. European Muticentre Study Group. The results of a randomized double blind controlled trial evaluating malotilate in primary biliary cirrhosis. J Hepatol 1993;17:227. Hoofnagle JH, Davis GL, Schafer DF, et al. Randomized trial of chlorambucil for primary biliary cirrhosis. Gastroenterology 1986;91:1327. McCormick PA, Scott F, Epstein O, et al. Thalidomide as therapy for primary biliary cirrhosis: a double-blind placebo controlled pilot study. J Hepatol 1994;21:496. Angulo P, Patel T, Jorgensen RA, et al. Siymarin in the treatment of patients with primary biliary cirrhosis with a suboptimal response to ursodeoxycholic acid. Hepatology 2000;32:897.
167. Miyaguchi S, Ebinuma H, Imaeda H, et al. A novel treatment for refractory primary biliary cirrhosis? Hepatogastroenterology 2000;47:1518. 168. Yano K, Kato H, Morita S, et al. Is bezafibrate histologically effective for primary biliary cirrhosis? Am J Gastroenterol 2002;97:1075–1077. 169. Leuschner M, Holtmeier J, Ackermann H, et al. The influence of sulindac on patients with primary biliary cirrhosis that responds incompletely to ursodeoxycholic acid: a pilot study. Eur J Gastroenterol Hepatol 2002;14:1369–1376. 170. Ritzel U, Leonhardt U, Nather M, et al. Simvastatin in primary biliary cirrhosis: effects on serum lipids and distinct disease markers. J Hepatol 2002;36:454–458. 171. Maddala YK, Jorgensen RA, Angulo P, et al. Open-label pilot study of tetracycline in the treatment of primary biliary cirrhosis. Am J Gastroenterol 2004;99:566–567. 172. Suzuki A, Van de Water J, Gershwin ME, et al. Oral tolerance and pyruvate dehydrogenase in patients with primary biliary cirrhosis. Dev Immunol 2002;9:55–61. 173. Mason A, Nair S. Primary biliary cirrhosis: new thoughts on pathophysiology and treatment. Curr Gastroenterol Rep 2002;4:45–51. 174. Nakano T, Inoue K, Hirohara J, et al. Long-term prognosis of primary biliary cirrhosis (PBC) in Japan and analysis of the factors of stage progression in asymptomatic PBC (a-PBC). Hepatol Res 2002;22:250–260. 175. Joshi S, Cauch-Dudek K, Heathcote EJ, et al. Antimitochondrial antibody profiles: are they valid prognostic indicators in primary biliary cirrhosis? Am J Gastroenterol 2002;97:999–1002. 176. Gores GJ, Wiesner RH, Dickson ER, et al. A prospective evaluation of esophageal varices in primary biliary cirrhosis. Development, natural history, and influence on survival. Gastroenterology 1989;95;1552. 177. Christensen E, Crowe J, Doniach D, et al. Clinical pattern and course of disease in primary biliary cirrhosis based on an analysis of 236 patients. Gastroenterology 1980;78:236. 178. Shapiro JM, Smith H, Schaffner F. Serum bilirubin: a prognostic factor in primary biliary cirrhosis. Gut 1979;20:137. 179. Dickson E, Grambsch PM, Fleming TR, et al. Prognosis in primary biliary cirrhosis: model for decision making. Hepatology 1989;10:1–7. 180. Rydning A, Schrumpf E, Abdelnoor M, et al. Factors of prognostic importance in primary biliary cirrhosis. Scand J Gastroenterol 1990;25:119. 181. Christensen E, Altman DG, Neuberger J, et al. Updating prognosis in primary biliary cirrhosis using a time-dependent Cox regression model. Gastroenterology 1993;105:1865. 182. Poupon RE, Balkau B, Guechot J, et al. Predictive factors in ursodeoxycholic acid treated patients with primary biliary cirrhosis: Role of serum markers of connective tissue. Hepatology 1994;19:635. 183. Murtaugh PA, Dickson ER, van Dam GM, et al. Primary biliary cirrhosis: prediction of short-term survival based on repeated patient visits. Hepatology 1994;20:126. 184. Angulo P, Lindor KD, Therneau TM, et al. Utilization of the Mayo risk score in patients with primary biliary cirrhosis receiving ursodeoxycholic acid. Liver 1999;19:115. 185. Neuberger J. Liver transplantation for primary biliary cirrhosis. Autoimmun Rev 2003;2:1–7. 186. Kim WR, Wiesner RH, Therneau TM, et al. Optimal timing of liver transplantation for primary biliary cirrhosis. Hepatology 1998;28:33. 187. Kim WR, Wiesner RH, Poterucha JJ, et al. Hepatic retransplantation in cholestatic liver disease: impact of the interval to retransplantation on survival and resource utilization. Hepatology 1999;30:395.
Section VI: Immune Diseases
42
PRIMARY SCLEROSING CHOLANGITIS Ulrika Broomé, Annika Bergquist Abbreviations ANA antinuclear antibody ANCA antineutrophil cytoplasmic antibodies BEC bile duct epithelial cells CCR5 chemokine receptor 5 CTFR cystic fibrosis transmembrane conductance regulator gene ERC endoscopic retrograde cholangiography EUS endosonography-guided FISH fluorescence in situ hybridization
HCC IBD ICAM-1 MHC MIC MMP-3 MRI NO pANNA
hepatocellular carcinoma inflammatory bowel disease intercellular adhesion molecule-1 major histocompatibility major histocompatibility class I chain matrix metalloproteinase-3 magnetic resonance imaging nitric oxide antineutrophil nuclear antibodies
INTRODUCTION Before 1980 primary sclerosing cholangitis (PSC) was a medical rarity. In 1980 three large studies from various parts of the world presented data on the natural history of this fascinating disease. Today, PSC has become one of the most important cholestatic liver diseases in the western world. PSC is a chronic cholestatic liver disease characterized by diffuse inflammation, obstruction, and fibrosis of the intra- and extrahepatic bile ducts. In a large number of patients the disease will lead to the development of liver cirrhosis and its complications. The disease process is, however, highly variable in the individual and also between patients. The most feared complication in PSC is the development of cholangiocarcinoma, which affects 10–15% of all patients. The etiology of PSC remains unknown, although growing evidence supports that immune mechanisms are involved in the pathogenesis. PSC is closely associated with inflammatory bowel disease (IBD) and approximately 70–80% have an associated bowel disease. There is at present no treatment that can halt the disease process, and therefore PSC has become a major indication for liver transplantation, with excellent long-term results. Although there is at present no specific treatment that can stop disease progression, a lot can be done to improve the quality of life and to treat complications in these patients. The management of PSC requires a multidisciplinary approach involving specialties in gastroenterology and hepatology, endoscopy, immunology, pathology, and liver transplant surgery.
EPIDEMIOLOGY In 1867, the first paper on obstruction of the bile ducts by thickening of their walls was published by Hoffman. This paper is supposed to be the first report ever on PSC, but the presence of the disease was not presented to the French readers until 1924 (by Delbet). In 1927, Miller described the first case of PSC in the
PET PSC RANTES SIR SMA TNF UC UDCA
positron emission tomography primary sclerosing cholangitis regulated upon activation normal T cellexpressed and secreted standard incidence ratios smooth muscle antibody tumor necrosis factor ulcerative cholangitis ursodeoxycholic acid
English literature. Over the next 50 years the disease remained a medical rarity, and by 1980 only approximately 100 cases had been published in the English literature. With the increased use of endoscopic retrograde cholangiography (ERC) during the 1980s and 1990s more cases were identified. The development of magnetic resonance imaging (MRI) provides a non-invasive diagnostic tool with no known side effects or complications, which further leads to increased identification of PSC. This is especially true among IBD patients, and PSC is a more common disease today than was previously believed. The prevalence of PSC is not known. Epidemiological investigations can be carried out in order to estimate the frequency or distribution of a disease within a population. The reliability of epidemiological studies is dependent on several factors, such as stringent inclusion criteria for the cases, clear definition of the date of disease onset, a well-defined study period, area and population, multiple case-finding methods, and rigorous tracing of all cases. Moreover, the disease should be a distinct entity with no overlap to other diseases. How do these requirements apply to PSC? Today, almost half of all PSC patients are asymptomatic at the time of diagnosis. Active screening procedures are therefore needed to identify all patients with PSC. This could possibly be done in an IBD unit with regular screening of liver function tests. Even if there is a rise in the liver function tests, they can fluctuate in the individual patient. Therefore, a normal test does not exclude PSC, and repetitive measurements are recommended. About 10% of PSC patients have normal liver function tests. This raises the question whether we should screen all IBD patients with magnetic resonance cholangiography (MRC) for a possible underlying PSC despite normal liver function tests. For ethical and economical reasons, however, this cannot be recommended. Moreover, the subgroup of PSC patients without underlying IBD will not be identified this way. There is no single sensitive and specific diagnostic marker for PSC. Diagnosis is made from a combination of clinical, biochemical, radiological and histological features. Each diagnostic tool is hampered
821
Section VI. Immune Diseases
by overlap with other liver diseases. In a majority of PSC cases the onset is insidious and patients have features of general chronic liver disease. The insidious onset may lead to problems in defining the onset of disease. In several studies it has been shown that the time between first abnormal liver function tests to time to definite diagnosis of PSC is around 5 years.1 Thus, there are several major drawbacks concerning epidemiological data in PSC. Because of the nature of the disease it is impossible to fulfil the required criteria for a reliable epidemiological study. Therefore, very few epidemiological studies are performed to evaluate the incidence and prevalence of PSC. The prevalence has been estimated by measuring the prevalence of PSC in patients with IBD, where prevalence and incidence figures are available. The prevalence of PSC in IBD is around 3–5%,2 which results in a prevalence of PSC between 1 and 6/100 000. However, this is probably an underestimation of the ‘true’ prevalence. There are two population-based investigations of incidence and prevalence of PSC reported using accepted criteria, including cholangiography for diagnosis.3,4 The first published study was from Norway and the Aker University Hospital in Oslo.3 The Aker Hospital serves 130 000 inhabitants and 17 new PSC cases were identified over a 10-year period. The mean yearly incidence was estimated to be 1.3/100 000 and the point prevalence at the end of the study (31 December 1995) was 8.5/100 000. Similar results were found in the second study from the United States, in Olmsted County, Minnesota, by Bambha et al.4 Between 1976 and 2000 22 patients met the criteria for PSC. The age-adjusted yearly incidence was 1.25/100 000 for men and 0.54/100 000 for women. The prevalence in year 2000 was 20.9/100 000 for men and 6.3/100 000 for women. Bambha et al. also addressed the question as to whether PSC had an increasing incidence. In their study there was a trend towards increasing incidence for men during the study period (1976–2000). A further analysis of PSC incidence over the most recent 10-year period revealed no trend toward increasing incidence in either men or women. The conclusion is that the trend of increased incidence found in 1976–2000 is most likely explained by a change in physician awareness, and by the increasing use of ERC. There are global differences reported in the prevalence of PSC. In Spain, the point prevalence in 1988 was 2.24 cases per million, indicating that the prevalence of PSC may be lower in the southern part of Europe than in Scandinavia and the US. This can be explained by the lower prevalence of IBD, which at the time was reported to be lower in Spain than in the United States and northern Europe. Very little is known about the prevalence of PSC in Latin America, Asia and Africa. The diagnosis relies on cholangiography and liver biopsy, facilities that are not widely available in many places. Moreover, PSC must be distinguished from several other disorders which are more commonly present outside Europe, such as Clonorchis sinensis infection, cholangiocarcinoma without an underlying PSC, AIDS-related cholangiopathies, complicated cholelithiasis etc. Whether the lack of reports from this part of the world represents a lower prevalence of PSC can only be speculated on. Reasons for a lower prevalence in different geographic areas can depend on differences in race and/or environmental factors. The study by Hurlburt et al.5 investigated the prevalence of autoimmune diseases in a non-white population and speaks in favor of race being
822
an important factor for the development of PSC. These authors investigated native Alaskans through a review of clinical records from 1984 to 2000. The frequency of autoimmune hepatitis and primary biliary cirrhosis was similar to reported rates from other populations, but interestingly no case of PSC was found. In conclusion, PSC is a rare disease, although all epidemiological studies probably underestimate its true incidence and prevalence owing to the nature of the disease and difficulties in making a correct diagnosis.
PATHOGENESIS The aetiology and pathogenesis of PSC have not yet been clearly established. Both autoimmune and non-autoimmune pathogenetic mechanisms have been postulated. The study of etiopathogenesis in PSC is complicated by several disease-related factors, which leads to difficulties in interpretation of the results of such studies: • A number of exogenous factors can cause a histological and cholangiographic picture similar to PSC, such as infection, toxicity, arterial ischemia and neoplasia; therefore, secondary causes to cholangiographic changes must always be excluded in patients with PSC. • The onset of PSC is insidious and etiologic factors may act as triggering events long before a clinical diagnosis is made. It therefore seems less hopeful to find the etiology of PSC than to be able to define the pathogenetic mechanisms causing it. • PSC is a very heterogeneous disease. Patients may be symptomatic or asymptomatic, have early or end-stage disease, have involvement only of the small bile ducts or the intrahepatic ducts, or of the whole biliary tree. PSC may overlap with autoimmune hepatitis and be responsive to steroids, as often is seen in paediatric PSC. Moreover, 10–15% of patients will develop cholangiocarcinoma. The heterogeneity of disease presentation and the clinical course might represent subgroups of PSC with different pathogenesis. In order to compensate for these differences large groups of patients have to be studied, preferably in an early disease course. Because the disease is rare the majority of studies are hampered by the inclusion of only a small number of patients. • The majority of patients have an associated inflammatory bowel disease and the pathogenesis of PSC must explain this close relationship. There is also the risk that pathogenetic studies of PSC are influenced by factors causing IBD. The main target in PSC is the liver, and etiologic studies should focus on the diseased organ. Studies of lymphocytes have shown differences in lymphocyte populations in the liver and blood. However, there are practical and ethical problems in obtaining enough liver tissue from patients with early disease. Moreover, the disease process can have a focal distribution in the liver, leading to a large sampling variability.
ANIMAL MODELS Theoretically, animal models can overcome some of these problems. Several animal models have been found to cause a sclerosing cholangitis-like picture, although in some of them cholangiography has not been performed. Four types of animal model of sclerosing cholangitis have been reported:6
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
1. Models involving bacterial cell wall components or colitis. 2. Models induced by injury to biliary epithelial or endothelial cells of hepatic arterioles or peribiliary capillaries. 3. Models induced by toxic, infectious or intraluminal injury to the biliary tract. 4. Experimental obstruction of extrahepatic bile ducts. The models involving reactions to enteric bacterial cell-wall components or colitis could explain the close association between PSC and IBD. In a rat model of small bowel bacterial overgrowth it was shown that only genetically susceptible rat strains developed the cholangiographic picture of sclerosing cholangitis. Treatment with metronidazole or tetracycline prevented hepatic injury. Hepatobiliary injury correlated significantly with Kupffer cell production of tumor necrosis factor-a (TNF-a). In a bacteria-free model described by Orth et al. the hapten 2, 4, 6-trinitrobenzenesulphonic acid was injected into the bile ducts of Lewis rats, resulting in a cholangiographic picture similar to that of PSC.7 Spontaneous production of interferon-g, interleukin-10 and TNF-a was increased. CD3+ T cells infiltrated the liver, and up-regulation of MHC class II antigens expression on bile duct epithelium was found, as well as the production of antineutrophil cytoplasmic antibodies (ANCA). ANCA showed specificity to myeloperoxidase, catalase and actin, and occurred between 1 and 12 weeks after 2, 4, 6-trinitrobenzenesulphonic acid injections. Recently, an Austrian group described that mice with a targeted disruption of the multidrug resistance gene (Mdr2) developed features of sclerosing cholangitis.8 A careful investigation of the pathophysiologic mechanisms leading to sclerosing cholangitis in Mdr2 -/- mice demonstrated that leaky ducts caused by disrupted tight junctions and base membranes led to bile acid leakage into the portal tracts, showing that this was a multistep process. The bile acids induced a periductal inflammation, followed by activation of periductal fibrogenesis, and finally obliterative cholangitis due to atrophy and death of bile duct epithelial cells. It is of interest to point out that in another study a reduction in hepatocyte tight junction proteins was reported in patients with PSC compared to patients with primary biliary cirrhosis. Whether these findings can cause increased paracellular permeability and bile regurgitation in human PSC is unclear.9 In several of the other animal models a sequence of events was found that showed first an initiation phase with the production of proinflammatory cytokines, especially TNF-a, secreted by activated macrophages (Kupffer cells). Then a subsequent infiltration of CD4+ and CD8+ T cells indicates a role for the adaptive immune response, although the antigen specificity of the T cells remains unknown. Although the animal models in PSC have several drawbacks, we learn from these studies that the pathogenesis of PSC include both non-immune and immune factors. In PSC the male preponderance and the relatively poor response to immunosuppression suggest that PSC is not a classic autoimmune disease. However, there is strong evidence that immune mechanisms play an essential role in the pathogenesis of this disease. Associations with certain HLA haplotypes, the presence of autoantibodies, lymphocytes in the portal tracts, increased levels of total serum immunoglobulins and the association with other autoimmune diseases support this notion. The key question, however, is whether these immune mechanisms are the cause or the consequences of the tissue injury.
NON-IMMUNE MECHANISMS The non-immune mechanisms can be divided into infectious, toxic, ischemic and genetic factors.
Infectious Mechanisms Because of the close relationship between ulcerative colitis (UC) and PSC the first pathogenetic theories for PSC were based on the belief that bacteria or toxins were absorbed across the diseased colonic mucosa and brought to the liver by the portal blood. However, studies of histological findings in the liver in UC patients undergoing colectomy could not be correlated to the presence of portal bacteremia. Moreover, portal phlebitis, the histological hallmark of portal bacteremia, is uncommon in PSC. Long-term antibiotic treatment has not been shown to improve the disease process. In explanted livers from patients with PSC patients (but not from patients with primary biliary cirrhosis) positive bacterial cultures have been found. The most common bacteria were a-haemolytic streptococci. It was suggested that the cause of the bacterial load was contamination from cannulation of the bile duct after ERC, indicating that the bacterial infection was more a secondary than a primary event. Helicobacter species, which may colonize the biliary tract, have also been implicated as a possible cause of PSC. In a study by Nilsson et al. using PCR on liver biopsy specimens Helicobacter was detected in 9 of 12 patients, as well as in the majority of patients with primary biliary cirrhosis, but only in one of 23 patients with non-cholestatic liver disease.10 Antibodies to Helicobacter hepaticus often cross-react with Helicobacter pylori. These antibodies frequently occurred in patients with chronic liver disease, but were not overrepresented in any specific disease group and were not increased compared with healthy individuals. In a more recent study, an increased prevalence of seropositivity for non-gastric Helicobacter species was found in approximately 30% patients with PSC as well as in patients with autoimmune hepatitis.11 The presence of Helicobacter was analyzed in bile obtained at the time of ERC in 122 patients with various hepatobiliary diseases, including four with PSC. Helicobacter was not present in any of these patients.12 At present there are no data to support Helicobacter as being of major pathogenetic importance in PSC.
Virus Reovirus 3 has been shown to cause non-suppurative destructive cholangitis in weaning mice, and cytomegalovirus is known to be able to damage the intrahepatic bile ducts. Both these viruses have previously been suggested as causative agents in PSC. Patients with AIDS and cytomegalovirus or Cryptosporidium have been found to develop a cholangiographic appearance similar to that of PSC. Current evidence does not support any pathogenetic role for these viruses in PSC. In a case control study by Ponsioen et al. the role of previous or persisting viral or atypical bacteria in PSC was investigated.13 Serological screening for antibodies against 22 viruses, Chlamydia and Mycoplasma pneumoniae was carried out. The only difference between PSC patients and the various control groups was a markedly elevated seroprevalence of IgG and IgA Chlamydia LPS antibodies. However, no actual presence of Chlamydia antibodies in liver tissue could be demonstrated. At present there is no virus
823
Section VI. Immune Diseases
or bacterium that can be directly implicated in the pathogenesis of PSC. It has, however, been speculated that autoantibodies to Chlamydia can give rise to autoantibodies that cross-react with certain constituents of the bile duct system in genetically susceptible hosts. Another possibility could be that a bacterial antigen might act as a molecular mimic of an autoantigen precipitating an immune cascade, and the pathogenetic microorganism need no longer to be present in these pathogenetic mechanisms.
IMMUNE MECHANISMS Autoantibodies Antinuclear (ANA) and antismooth muscle (SMA) autoantibodies occur in between 20 and 70% of PSC patients. Chapman et al. reported that 63% of patients with PSC with UC had antibodies reacting with colonic epithelium.89 Antineutrophil cytoplasmic antibodies (ANCA) are the autoantibodies most frequently associated with PSC, occurring in as many as 88% of patients.14 The target antigen for this antibody appears to be localized to the periphery of the nucleus in the neutrophils. It has therefore been suggested that these autoantibodies should be renamed ‘antineutrophil nuclear antibody’ (pANNA). pANNA is not specific for PSC and is found in patients with ulcerative colitis (UC) as well as those with autoimmune hepatitis. The pANNA levels do not seem to correlate with disease-specific clinical parameters. The molecular identity of the antigen(s) recognized by pANNA is unknown. In a study including 69 PSC patients, antigenic specificities of ANCA in PSC were tested. In this study some specificities of ANCA (bactericidal/permeability increasing protein and cathepsin G) were associated with the presence of cirrhosis, and lactoferrin was more frequently detected in PSC in conjunction with ulcerative colitis than in those without IBD.15 The clinical significance of ANCA in PSC has not been established. Most investigators today consider pANNAs in PSC to be an epiphenomenon rather than of primary pathogenetic importance. Angulo et al.16 investigated the profile and significance of 20 different serum autoantibodies in patients with PSC. PSC patients had a significantly greater rate of autoantibodies than did controls for antinuclear, anticardiolipin, antithyroperoxidase, rheumatoid factor and ANCA; 97% were positive for at least one antibody and 81% had at least three positive antibody tests. Anticardiolipin antibodies correlated with disease severity as measured by Mayo risk score and histologic stage. In PSC patients autoantibodies binding to human biliary epithelial cells (BEC) were recently detected.17 Significantly more patients with PSC had anti-BEC antibodies binding to a 40 kDa component of the biliary epithelium membrane than did controls. In 90% of patients with PSC the autoantibodies reacted only with cytokinestimulated target cells. Furthermore, anti-BEC antibodies induced BECs to produce high levels of interleukin-6 and the expression of CD44, showing for the first time that PSC patients may have autoantibodies that are of functional importance.
Cellular Factors There is a T cell-predominant portal tract infiltrate in PSC that consists preferentially of both CD4+ and CD8+ cells. An increase in g/d T cells has also been found in the liver as well as the blood in PSC.
824
Xu et al. investigated liver-derived T cells (LDL) isolated from patients with PSC, primary biliary cirrhosis, autoimmune hepatitis and healthy controls.18 Expression of HLA-DR but not IL-2 was found in PSC. Liver-derived T cells from PSC patients had impaired proliferative and functional capacity. Cytokines analyzed in the supernatants of cultures from the liver-derived T cells showed significantly higher levels of TNF-a and IL-1b. The addition of anti-TNF antibodies restored the diminished proliferative response, indicating that blocking of cytokines can normalize the immune response in PSC. Apoptosis is an important factor in normal cellular homeostasis and it has been suggested that dysregulated apoptosis may be of importance in pathogenesis of cholestatic liver disease. Timmouth et al.19 analyzed samples from explanted livers from patients with PBC and PSC, and biliary epithelial cells were stained for markers of apoptosis. Livers from PBC patients demonstrated greater evidence of apoptosis than did livers from patients with PSC.19 The majority of patients with PSC have an associated IBD. The bowel disease does not seem to influence the activity of liver disease. PSC can develop before onset of IBD, but also several years after liver transplantation, and PSC can occur many years after colectomy. Moreover, not all PSC patients have IBD, and only 5–7% of all IBD patients have PSC. Thus, the interaction between the bowel and PSC suggests that these two diseases are interrelated but not dependent on each other’s activity. Grant et al.20 have proposed an interesting new concept regarding the association between PSC and IBD, suggesting that an enterohepatic circulation of lymphocytes exists.20 Mucosal lymphocytes in the gut can become activated during active inflammatory bowel disease. These lymphocytes can thereafter persist as long-lived memory cells capable of recirculation through the liver, as they express homing receptors that direct migration not only to the liver but also to the gut. If the right conditions exist, these dual-homing lymphocytes can be activated in the liver and cause inflammation and tissue damage, promoting the recruitment of more mucosal lymphocytes. This results in a chronic inflammation. The specific gut homing integrin a4b7 and its ligand MAd-CAM have recently been found in the liver of patients with PSC, and vascular adhesion molecule (VAP-1) constitutively expressed in the liver is up-regulated in the gut in IBD, thereby supporting this notion21 (Figure 42-1). The enlarged lymph nodes found in over 60% of all PSC patients could provide a mechanism for continuous lymphocyte recruitment to the liver.
Cholangiocytes The cholangiocytes have developed mechanisms to respond to infection by recruiting and interacting with leukocytes in order to clear pathogens. When these cells are damaged during immune-mediated attacks the secretion of chemokines and expression of adhesion molecules may promote the inappropriate recruitment of inflammatory cells.22 It has been demonstrated that patients with PSC express class II antigens on the biliary epithelium, and in end-stage disease ICAM-1 is also expressed. Once significant cholangiocyte destruction occurs, the integrity of the biliary system is broken down and a vicious cycle of inflammation and tissue damage ensues. Although the cholangiocytes are able to actively recruit and retain effector cells, it is still unclear whether they can act as non-professional antigen-presenting cells. Professional antigen-presenting cells also
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
Normal
PNAd PLN
L-Selectin
L-Selectin
MAdCAM-1 MLN
47
VAP-1 MAdCAM-1 Portal LN
N
M
VAP-1R
VAP-1R
M VAP-1R
M ad (V CA ap M -1 -1 )
VAP-1 Memory T cells
Figure 42-1. Proposed enterohepatic recirculation of lymphocytes by Grant et al.20 Upper panel: Naïve lymphocytes (N) use the peripheral node addressin (PNAd) to enter peripheral lymph nodes (PLN) and MAdCAM-1 to enter mesenteric lymph nodes (MLN). After activation in MLN, mucosal memory cells (M) are directed back to the gut by their ability to bind MAdCAM-1 on mucosal vessels. Some mucosal effector cells may also bind VAP-1, allowing them to recirculate through the liver. Similarly, lymphocytes activated in portal lymphoid tissue use VAP-1 to bind liver endothelium and may be able to bind to low levels of VAP1 on mucosal endothelium. Thus, a proportion of mucosal T cells and liver T cells can migrate to both the liver and the gut, providing immune surveillance across both sites. Lower panel: The ability of liver endothelium to express MAdCAM-1 under certain inflammatory conditions promotes the recruitment of a4b7+ memory cells originally activated in the gut, and if these cells encounter antigen in the liver, inflammation will be established. Similarly, increased expression of VAP-1 on inflamed mucosal vessels promotes the recruitment of lymphocytes originally activated in the liver.
Inflammation
VAP-1 MAdCAM-1 Portal LN
MAdCAM-1 MLN
47
M ad (V CA ap M -1 -1 )
M
VAP-1R
VAP-1R
M VAP-1R VAP-1 MAdCAM-1
Memory T cells
express co-stimulatory molecules such as members of the B7 family (CD80 and CD84), which are required for T-cell activation. These co-stimulatory molecules seem to be lacking in PSC. The biliary epithelium in PSC expresses CD44, which is a lymphocyte homing receptor.23 CD44 can bind chemokines and growth factors. Antibiliary epithelial antibodies detected in almost 70% of all PSC patients can induce the expression of CD44, and may therefore be of importance in mediating the local inflammatory response.17 Nitric oxide (NO) is formed constitutively in low concentrations by neuronal NO synthases (NOS1) and endothelial NOS (NOS3). Hepatocytes, cholangiocytes, stellate and Kupffer cells can express an inducible NOS (NOS3). Spirli et al. showed that TNF-a and IFN-g could synergistically stimulate the cholangiocytes to generate NO via NOS2 induction.24 NO can cause ductular cholestasis by inhibition of HCO3- and chloride-secretory mechanisms, and also
inhibit cholangiocellular bile formation.25 In liver tissue from patients with PSC interlobular and septal ducts strongly expressed NOS2.25 The increased NO production could be of pathogenetic importance in PSC by facilitating biliary retention of toxins, and the generation of reactive nitrogen oxide species could result in cell damage and bile duct injury. It has also been shown that protein nitrosylation may inactivate DNA repair enzymes, providing a link to cholangiocarcinoma in PSC.26
Genetic Factors Evidence supporting that genetic factors are of pathogenetic importance in PSC came first from a few case reports of familial occurrence of PSC. Recently a paper on familial occurrence in a large group of patients with PSC was published. In this Swedish study it was shown that first-degree relatives with PSC have a prevalence of
825
Section VI. Immune Diseases
0.7%, indicating a nearly 100-fold increased risk of developing PSC compared to the general population.27 The best way to understand the genetic factors in PSC would be to perform twin studies or to find sufficient family members for linkage analysis. Because of the low prevalence of PSC and the low heritability, however, this may never be possible. It has been suggested that PSC is a ‘complex trait,’ i.e. a disease in which one or more genes, acting either alone or in concert, increase or reduce the risk of disease.28 This suggests that disease alleles do not by themselves cause disease, but simply increase the chance that disease will develop under the appropriate conditions. The allele may also have an impact on disease development. The method most used to study the genetic basis of PSC has been association analysis and case–control studies. To date, genetic studies in PSC have concentrated on the major histocompatibility complex (MHC) and a few single-nucleotide molecule polymorphisms in other immunoregulatory genes and genes of importance for fibrosis.29 Today, with our fast-growing knowledge of genetics, it seems to be of less importance to study single-nucleotide molecule polymorphisms or single cytokine polymorphisms than to study other genes that interact with them. Besides the HLA studies the majority of genetic studies are hampered by inclusion of low numbers of PSC patients. As these genes most probably have a relatively small influence, we need much larger studies. It is therefore important to point out that studies that have been found to be negative might have been underpowered to detect small effects. Positive findings should always be replicated in a second set of patients before it can be concluded that the studied gene is of pathogenetic or prognostic importance.
Immunogenetics HLA class I and class II antigens are of critical importance for T-cell immunity. HLA molecules present short antigenic peptides for recognition by antigen-specific T cells. These molecules are highly polymorphic. More than 90% of the inherited variation in the HLA class I and II genes gives rise to alleles that encode different amino acid sequences concentrated in the antigen-binding site of the HLA molecule. HLA genotyping studies suggest that there is a strong genetic component in PSC (Table 42-1). Several studies have investigated this issue and three different HLA haplotypes have been found to be associated with PSC: A1-B8-TNFA*2-DRB3*0101DRB1*0301-DQB1*0201, DRB3*0101-DRB1*1301-DQB1*0603, and DRB1*1501-DQB1*0602. DRB1*0401-DQB1*0302, on the other hand, may be associated with disease resistance.29,30–32 There
Table 42-1. Key HLA Haplotypes Associated with PSC160
826
Haplotype
Significance in PSC
B8-TNF*2-DRB3*0101-DRB1*0301DQA1*0501-DQB1*0201 DRB3*0101-DRB1*1301DQA1*0103-DQB1*0603 DRB5*0101-DRB1*1501DQA1*0102-DQB1*0602 DRB4*0103-DRB1*0401DQA1*03-DQB1*0302
Strong association with disease susceptibility Strong association with disease susceptibility Weak association with disease susceptibility Strong association with protection against disease
are still controversies as to which allele within each haplotype may form the primary association. Donaldsson33 suggested that susceptibility/resistance might be determined by specific amino acids at DQb-87 and DQb-55, respectively. Although these associations have been confirmed and seem reliable, they still do not account for more than 50% of the genetic associations in PSC patients. Linkage with other genes within the HLA region may also account for the observed associations. The second MHC class III genes encode the major histocompatibility class I chain (MIC)-related gene family maps between the HLA B and TNFa. The MIC family consists of five members (A–E), of which only MICA and MICB encode expressed proteins. These molecules are expressed on the gastrointestinal and thymic epithelial cells. They are induced by stress and heat shock, and appear to activate natural killer cells as well as g/d T cells. NK cells and g/d cells have been found in increased numbers in PSC. It has been speculated that up-regulation of MIC molecules leads to persistent immune activation, which leads to auoimmunity, or failed activation causing persistent infection. These molecules are highly polymorphic and have been investigated in two separate studies in PSC. In a British study including two independent PSC populations, allelespecific amplification of 16 MICA alleles was investigated.34 The MICA* 008 allele was more common in PSC than in control subjects (66% vs 48%), whereas the MICA*002 allele had a protective effect from PSC. The MICA*008 association was independent of B8 and other HLA haplotypes associated with PSC. In a Norwegian study including 130 PSC patients 5 MICA and 15 MICB microsatellite alleles were analyzed.35 The prevalence of MICA 5.1 (90% vs 74%) and MICB24 (58% vs 29%) alleles was considerably increased in patients with PSC compared to healthy individuals. It was also shown that HLA B8 and DR3 were associated with PSC only in the presence of MICA5.1 and MICB 24. MICA*008 is the main allele carrying the MICA*5.1 microsatellite allele. The frequency of patients with PSC carrying all four alleles was also substantially greater (49% vs 18%). Although the association with MICA*008 has been confirmed, it still seems unclear whether the MIC alleles are associated with PSC only as a part of an extended B8-DR3 haplotype. The matrix metalloproteinsases (MMPs) are a family of proteolytic enzymes that degrade extracellular matrix and regulate fibrosis formation. MMP-1 and MMP-3 enzymes are believed to play a part in degradation of normal liver extracellular matrix, as well as in the development of liver fibrosis. Moreover, these enzymes are involved in invasion and metastasis in cancer. Both MMP-1 and MMP-3 carries a promotor dimorphism. In a large group of Norwegian PSC and UC patients no differences in MMP-1 and MMP-3 promoter polymorphism frequencies were found.36 However, when subgrouping patients it was found that PSC patients with UC showed an increased frequency of the MMP-3 allele compared to PSC patients without UC. Moreover, all PSC patients with cholangiocarcinoma carried the MMP-1 allele 1G, compared to only 72% of PSC patients without cholangiocarcinoma; these results indicate genetic heterogeneity among PSC patients. In a British study only the influence of matrix metalloproteinase3 (MMP-3) polymorphism was investigated. A functional polymorphism of MMP-3 was shown to be associated with both susceptibility to PSC and progression to portal hypertension.37
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
The immune response is orchestrated by cytokines, and many cytokines have been found to have some degree of polymorphism. Not all of these genetic variations have a functional consequence. The tumor necrosis factor has been implicated in the pathogenesis of PSC, but the 308A promotor nucleotide polymorphism found in PSC is most likely due to the linkage polymorphism with the DRB*103-DQA1*0501-DQB1*02 haplotype.38 Other cyokines investigated include interleukins-1 and -10, and both have been negative.39,40 Intercellular adhesion molecule-1 (ICAM-1) polymorphism has been implicated in the susceptibility to inflammatory bowel disease. ICAM-1 is expressed on the bile ductules and interlobular bile ducts in advanced PSC. Moreover, increased levels of ICAM-1 have been found in the serum of patients with PSC. In a study including 104 PSC patients (69% with IBD) the ICAM-1 polymorphism K469E was associated with protection against PSC.41 When the ICAM-1 genotype was correlated with the HLA genotype, it was shown that the protective effect of E469E homozygosity was statistically significant in the absence of HLA DRB1*0301 DQB1*0201. ICAM-1 polymorphism seemed also to be independent of TNFa-308, MMP-3 and MMP-9 genes. Chemokines are involved in leukocyte trafficking and recruitment and have a crucial role in determining which cells migrate to the injured and/or inflamed organ. Chemokine receptor 5 (CCR5) is the natural ligand for RANTES (regulated upon activation normal T cellexpressed and secreted). During colonic inflammation RANTES is upregulated on the colonic epithelium and CCR5 on the activated T cells in the colon.42 It has been suggested that CCR5 plays a role in lymphocyte recruitment and immune response within the liver. A 32 base-pair deletion in CCR5 (CCR5-D32 mutation), results in a non-functioning receptor. In a study including 71 PSC patients a significantly increased frequency of CCR5-D32 heterozygotes and a significant higher CCR5-D32 allele frequency was found compared to healthy controls.43 The CCR5-D32 allele may skew the immune
response to recurrent or high levels of antigenic stimuli in the liver that favor chronic inflammation over resolution of the inflammatory response. Cystic fibrosis is caused by defects in the cystic fibrosis transmembrane conductance regulator gene (CTFR). Approximately 10–35% of cystic fibrosis patients develop biliary disease due to decreased chloride secretion into the bile canaliculi, leading to obstruction of intrahepatic bile ductules by inspissated secretions. A recent analysis of the CTFR gene in patients with various liver diseases revealed that PSC patients demonstrated a significantly increased number of mutations/variants of the CTFR gene compared to healthy controls, as well as to patients with other liver disorders.44 Moreover, function analysis showed that PSC patients had a disturbed CTFR chloride channel function.44 The polymorphism of the CTFR gene, however, could not be confirmed in an Italian study including 64 PSC patients.32 Thus, PSC patients seem to be genetically susceptible individuals with an immune response that favors chronic inflammation and lacks the possibilities for resolution. Both the cellular and humoral immune responses are involved in the pathogenesis of PSC (Figure 42-2). The cholangiocytes themselves seem to play an active role in the disease process, and are not only targets of an immune attack but also active participants.
CLINICAL FEATURES DEFINITION OF DISEASE The diagnosis of PSC is based on characteristic clinical, biochemical, histologic and – most importantly – radiologic features, with irregularity and beading of the intra- and/or extrahepatic bile ducts (Figure 42-3). Secondary causes of sclerosing cholangitis, such as previous biliary surgery, biliary stone disease, ischemic bile-duct damage due to treatment with floxouridine, congenital biliary-tree
Figure 42-2. Theory of the pathogenesis of PSC presented by JM Vierling in Primary sclerosing cholangitis, edited by M. Manns. Dordrecht: Kluwer Academic, 1998: p. 43.
Immunogenetically susceptible host
Portoenteric bacterial induction of TNF secretion with activation of BEC and endothelial cells
IL-8, MCP-1, IL-6, secretion by BEC Ag shedding, processing and presentation
Periductular fibrosis
Chemoattraction, activation of neutrophils, monocytes T and B cells cytokine secretion activation of fibroblasts
Ischemic atrophy of BEC
Progressive obstruction and cholestasis
Fibroductular interface hepatitis
Fibrosing obliteration of bile ducts
Biliary cirrhosis (cholangiocarcinoma)
Progressive portal fibrosis
827
Section VI. Immune Diseases
PSC
Clinical features
Biochemical features
Histological features
Radiological features
Male predominance Concomitant IBD
Cholestatic liver function tests (ALP)
Localized or multifocal strictures
Classical symptoms: abdominal pain fever pruritus jaundice
Fluctuations (10% normal)
Bile duct proliferation Periportal inflammation and fibrosis Obliteration and loss of bile ducts
Segments of dilatation (‘beading’)
Figure 42-3. The diagnosis of PSC is based on clinical, biochemical, histological and, most importantly, cholangiographic features.
828
abnormalities, cholangiopathy associated with acquired immunodeficiency syndrome, or bile duct neoplasms must be excluded. Cholangiographic features simulating intrahepatic PSC can also be seen in patients with cirrhosis, hepatocellular carcinoma, polycystic liver disease, sub-massive hepatic necrosis, amyloidosis, intrahepatic portal thrombosis, liver metastases, leukaemia and lymphoma, pointing to the importance of combining clinical, histological and cholangiographic features when diagnosing PSC. Some patients with IBD have cholestatic biochemical abnormalities and liver histology consistent with PSC but a normal cholangiogram. These patients are considered to have small bile duct sclerosing cholangitis, which is part of the spectrum of PSC.45–47 Clinical features of small bile duct PSC are discussed below.
extrahepatic biliary strictures, stones, sludge formation, or infections occur. Positive ANA or SMA titers have been detected in 20–70% of all PSC patients. The rationale for the autoantibody screen in a patient with suspected PSC is the exclusion of the differential diagnoses primary biliary cirrhosis and autoimmune hepatitis. The antineutrophil antibody (ANCA) is raised in 65–95% of PSC patients and may have a fluctuating course. It has been detected in 25% of first-degree relatives of patients with PSC. ANCA is not specific for PSC and is present in 50–80% of patients with ulcerative colitis without liver disease, as well as in systemic vasculitides, in particular Wegener’s granulomatosis. The use of ANCA in clinical practice is of limited or no value (see section about pathogenesis).
DIAGNOSIS – BLOOD TESTS
DIAGNOSIS – IMAGING
The classic laboratory finding in patients with PSC is an increased alkaline phosphatase level (ALP). Most patients also have raised aminotransaminase (AST/ALT) levels. In many studies, a twofold increase in ALP level is used as one of the diagnostic criteria for PSC. The range of the increase is for practical clinical use not much help, as a normal ALP level has been reported in approximately 10% of all PSC patients. The liver enzymes often run a fluctuating course. It is therefore important to follow patients with UC with repeated measurements of liver function tests to make sure that they do not suffer from PSC. Accordingly, it is advisable not only to take liver function tests in UC patients regularly, but also to perform MRC on wide indications whenever symptoms or signs indicate liver disease in IBD patients. The level of serum albumin and the prothrombin time are normal in early PSC. In patients suffering from both PSC and UC, a decrease in albumin may be related to an active IBD. Serum bilirubin levels are usually normal in the early stages of PSC, but gradually increase as the disease progresses. Striking fluctuations in bilirubin levels may occur, even in the early stages of the disease, when significant
ERC remains the gold standard for the diagnosis of large duct PSC. The bile ducts show localized or multifocal strictures and intervening segments of normal or diverticulum-like outpouchings or ectatic ducts. MRC is increasingly being used as a non-invasive means of diagnosis. MRC has been tested in patients with PSC and can usually equal ERC. Introducing a new method makes demands on information about its efficacy, safety and cost. MRC seems superior for the visualization of the intrahepatic ducts, and the intraobserver variability is low.48 In a large study by Textor et al. including 150 patients with PSC the sensitivity and specificity of MRC were evaluated49 and showed 88% sensitivity and 99% specificity for the diagnosis of PSC. In a recent study intravenous morphine administration was used prior to MRC, primarily to evaluate candidates for living liver donation. Use of morphine intravenously improved the image quality by a contraction of the sphincter of Oddi and accomplished a distention of the biliary tree.50 Whether this method is applicable in PSC to increase the sensitivity has not been evaluated. Because of its safe, non-invasive nature MRC can be recommended when diagnosis is the major goal of the procedure.51,52 A specific risk
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
assessment was carried out in 106 ERCs in 83 patients. In asymptomatic PSC patients the complication rate was 2% and in symptomatic patients the risk was higher, at 14%. Pancreatitis, cholangitis, and worsening of symptoms were the most common complications within a week after the procedure.53 To prevent bacterial complications prophylaxis with antibiotics prior to ERC is recommended.54 The effectiveness of MRC compared with ERC based on overall cost has recently been evaluated. Talwalkar et al. compared the costs for MRC versus ERC as initial testing strategy and showed that the accuracy of the methods was similar; however, the cost for ERC exceeded that of MRC, particularly owing to the cost for treatment of complications after ERC. In summary, all these results suggest that MRC is the diagnostic method of choice in PSC. ERC should be reserved for patients with a need for interventions such as dilatation, or brushings of a dominant stricture where an underlying cholangiocarcinoma is suspected. CT and ultrasound are of no or little value to diagnose PSC but have a role in identifying complications of the disease, such as biliary stones, gallbladder disease and cholangiocarcinoma.
DIAGNOSIS – HISTOLOGY The reasons for performing liver biopsy in a patient with a cholangiography showing characteristic changes for PSC are twofold. First, to stage the disease so as to be able to better evaluate the prognosis of the patient, and second to rule out treatable coexisting liver disease, most importantly autoimmune hepatitis. The liver histology in PSC is variable and often unspecific, and is diagnostic in less than half of all cases. Characteristic features in liver biopsies from patients with PSC are: bile duct proliferation, periportal inflammation and fibrosis, and obliteration and finally loss of bile ducts. The pathognomonic periductal fibrosis or ‘onion-skinning’ is a rare finding that is present only in 13% of cases. The extent of fibrosis is scored into four stages, where stage IV represents cirrhosis. Sampling variability does occur, and among 56 PSC patients undergoing two liver biopsies on the same occasion 11% differed by more than one stage, and cirrhosis was missed in 37%. Overlapping features of autoimmune hepatitis are frequently seen in patients with PSC. However, with the revised scoring system for autoimmune hepatitis recently published the possibility of excluding the disease in patients with PSC has improved considerably. Nevertheless, approximately 10% of all PSC patients will fulfill the criteria for definite or probable autoimmune hepatitis. A recent study from the Mayo Clinic evaluated the role for liver biopsy in PSC.55 A review of 138 charts from patients with PSC was carried out. In 79 patients a liver biopsy had been performed after the cholangiography was taken. The result of the biopsy changed the management in only one patient, who was found to have overlap to autoimmune hepatitis and was treated with corticosteroids and azathioprine. This patient had biochemical features consistent with autoimmune hepatitis. One patient developed a bile leak after liver biopsy, requiring hospitalization for pain management and observation. This study clearly shows good arguments for not performing liver biopsy routinely in patients with large duct PSC. The nonspecific findings on liver biopsy, the sampling variability, and the new prognostic model, which does not require histology, further support this belief. Liver biopsy was therefore only recommended in selected cases. Another reason for performing liver biopsy in
patients with PSC is to search for biliary dysplasia. It has been shown that biliary dysplasia in bile ducts indicates the presence of cholangiocarcinoma in another part of the liver. It has also been proposed that biliary dysplasia can precede the development of cholangiocarcinoma and therefore be an early marker of malignancy.56 Further studies are needed to evaluate the role of biliary dysplasia in predicting patients who will develop cancer. However, whenever a liver biopsy is performed in a patient with PSC the pathologist should look for the presence of dysplasia, and if it is found the patient should be carefully evaluated for the presence of cholangiocarcinoma.
CLINICAL PRESENTATION OF LARGE BILE DUCT PSC The typical PSC patient today is an asymptomatic patient with UC having a rise in liver function tests, including alkaline phosphatase. The clinical presentation of PSC can, however, vary greatly. The patient may be either asymptomatic or symptomatic. The presentation of symptoms and signs in patients with PSC has changed during the last 10–20 years. In the early 1980s PSC was commonly seen as a symptomatic disease, with jaundice, pruritus and abdominal pain present in about two-thirds of all patients at the time of diagnosis, and asymptomatic patients were rare (7–10%). As the use of ERC has increased and our knowledge of PSC has increased, the number of patients with asymptomatic disease at diagnosis has risen. Today, 15–45% of all PSC patients are asymptomatic at time of diagnosis. Many asymptomatic patients remain completely well despite progression of cholangiographic and histology changes. 17% of the asymptomatic patients and 50% of the symptomatic ones are reported to have cirrhosis at the time of diagnosis. Among symptomatic patients, the disease is characterized by periodic remissions and exacerbations. The most common symptoms in patients with PSC are fatigue, jaundice, pruritus and abdominal pain, whereas ascites, bleeding from esophageal varices and acute cholangitis are much less frequent (Table 42-2).
DIFFERENTIAL DIAGNOSIS PSC must be distinguished from secondary sclerosing cholangitis, such as previous biliary surgery, biliary stone disease, ischemic bile-duct damage, congenital biliary-tree abnormalities, or bile duct
Table 42-2. Symptoms in Patients with Large Duct Primary Sclerosing Cholangitis Symptoms Asymptomatic disease Fatigue Pruritus Jaundice Abdominal pain Hepatomegaly Weight loss Acute cholangitis Ascites Variceal bleeding
Frequency (%) 15–44 65 25–59 30–69 16–37 34–62 10–34 5–28 2–10 2–14
829
Section VI. Immune Diseases
carcinoma. The risk of cholangiocarcinoma is increased in patients with PSC and is seen in 13% of all cases.57 There is an inherent problem in diagnosing PSC when a cholangiocarcinoma is present. To rule out that the cholangiographic changes are due solely to the cholangiocarcinoma, the localization of the tumor in relation to other biliary stricures, the presence of IBD and a previous history of elevated liver function tests has to be considered. Cholangiopathy mimicking PSC is also seen in patients with AIDS, especially in conjunction with cytomegalovirus infection. Other conditions that present with cholestasis and ductopenia in liver biopsies must also be excluded, e.g. primary biliary cirrhosis, sarcoidosis, liver allograft rejection, chronic graft-versus-host disease, and drug-induced reactions. Biochemical cholestatic patterns can rarely also be seen in autoimmune, alcohol-related and viral hepatitis, and these conditions must also be excluded. In most cases patient history, biochemistry and liver histology provide information that contributes to correct diagnosis. Important differential diagnoses to PSC are other hepatobiliary disorders seen in UC. Seventy to 90% of patients with PSC have UC, but almost 15% of patients with UC have been reported with an associated hepatobiliary disorder. PSC is the most common hepatobiliary disease in UC, with a prevalence around 3–7%. Liver biopsies from patients with UC show a spectrum of abnormalities, including fatty changes, portal inflammation and fibrosis, chronic active hepatitis, pericholangitis, cirrhosis and cholangiocarcinoma. Clinical, biochemical, and histological features of autoimmune hepatitis are reported in patients with PSC. In particular, children are usually considered to have more autoimmune features than adults and respond to immunosuppressive treatment with corticosteroids.58 Adults with PSC are increasingly being seen with features of both PSC and autoimmune hepatitis.59–61 The prevalence of overlapping features of autoimmune hepatitis to PSC is reported to be between 1.4%60 and as much as 24%.61 Younger adults seem to have more overlapping features than older patients.58 Recently Abdo et al.59 published six patients with autoimmune disease and no evidence of biliary disease at diagnosis who after an average of 4.6 years became resistant to immunosuppression and developed cholangiographic features of PSC. Patients with a pre-existing PSC can also change its biochemical pattern and express more autoimmune hepatitis features, with, for example, a rise in aminotransferase levels and ANA titers, and respond well to steroids. Thus, when a patient with a well-established PSC changes the biochemical appearance it is important to consider the occurrence of another autoimmune disease. These patients may benefit from steroid treatment and should therefore be carefully investigated.
do not include the presence of IBD as a criterion.46,47 As only 70–80% of patients with large duct PSC suffer from concomitant IBD, and the liver histology is characteristic in only approximately 50% of all cases with typical large duct disease, there is a risk of missing the diagnosis of small duct PSC with the use of these strict criteria. On the other hand, with less strict criteria it can be difficult to distinguish between small bile duct PSC and other cholestatic disorders. We suggest the following definition of small bile duct PSC: • • • •
Elevated liver enzymes with a cholestatic profile (rise in ALP) A liver biopsy consistent with PSC A normal cholangiogram Exclusion of other underlying chronic liver diseases, such as viral hepatitis, primary biliary cirrhosis, autoimmune hepatitis, alcoholic liver disease and drug-induced liver injury.
In the largest three studies, including altogether 83 patients with small duct disease, the natural history of this entity has been described. The Swedish, Mayo Clinic and Oxford studies included 32, 18, and 33 patients, respectively. Whether small bile duct disease should be regarded as an early form of PSC or a specific subgroup of PSC is not clear. Eleven of the 83 patients developed cholangiographic changes. This indicates that development of small duct disease to large duct disease is rare. The clinical presentation in small bile duct PSC seems similar to that in patients with classic PSC, regarding symptoms and associated conditions. The classic symptoms of PSC, namely fever, itching, fatigue and abdominal pain, are also present in patients with small bile duct PSC. The most common symptom is abdominal pain, which was present in 34% of the cases from Sweden. This is interesting, as the reason for the abdominal pain in PSC is unknown. It has been believed that the strictures of the bile duct tree, gallstones and bile duct stones could be the reason for the abdominal pain in large duct PSC. However, the fact that those PSC patients with only small duct disease also have the same type of upper right quadrant abdominal pain speaks against this hypothesis and raises the question as to whether other mechanisms also are involved. All three studies show comparable results, with a favorable prognosis for patients with small duct disease. Two of them (Mayo and Oxford/Oslo) compared patients with small duct PSC with patients having large duct PSC and showed better survival for the former. This can be partly explained by the fact that no case of cholangiocarcinoma has been described during a follow-up of 1–10 years in the small bile duct group. In summary, small bile duct PSC seems to be part of a spectrum of PSC and have a more favorable prognosis. The progression to large duct disease is rare, and the risk of developing cholangiocarcinoma is low.
SMALL BILE DUCT PSC In 1985 Wee and Ludwig introduced the conception of small bile duct PSC. They defined this entity as a patient with similar biochemical and histological features as PSC but a normal cholangiogram. The definition of the disease varies in the different studies. There are three recent studies investigating the clinical presentation and prognosis of small bile duct PSC. Some authors include only patients with an associated IBD, chronic cholestasis, liver histology consistent with PSC and a normal cholangiogram,45 whereas others
830
ASSOCIATED DISEASES INFLAMMATORY BOWEL DISEASE (IBD) PSC is associated with a number of diseases, the most important being IBD. The majority of patients with PSC who have an associated IBD suffer from UC.47a Although Atkinson and Caroll had already suggested an association with Crohn’s disease in 1964, this association has been less well recognized. In more recent studies,
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
1.3–13% of patients with Crohn’s disease also have PSC. It was suggested in a study from Norway that PSC might be seen as commonly in colonic Crohn’s disease as in UC. IBD can be diagnosed at any time during the course of PSC, and PSC can occur at any time during the course of IBD. PSC may also develop after proctocolectomy in colitis. Proctocolectomy for UC appears to have little or no effect on the progression of PSC, and there seems to be no correlation between the severity of UC and that of PSC. There is wide global range in the prevalence of UC in patients with PSC, varying from 23% (reported from Japan) to as much as 100% (in Norway). The prevalence of PSC in patients with IBD can vary between 2.4% and 7.5%. The reason for this variation can possibly be explained by the fact that PSC can be asymptomatic with only a mild elevation of liver function tests, and that these abnormalities may fluctuate, so that episodes of normal liver function tests may occur. Therefore, the prevalence of PSC in patients with IBD will depend on the number of patients being screened with liver function tests, and how frequently these tests are made. Other causes of elevation of liver function tests besides PSC, such as autoimmune hepatitis and fatty changes, may be present in IBD patients. Furthermore, elevated liver enzymes may be seen during colonic disease activity, treatment with steroids, and total parenteral nutrition. Most UC patients having PSC suffer from substantial colitis. The prevalence of PSC is 5.5% in patients with substantial and 0.5% in patients with distal colitis. The colitis in PSC generally runs a quiescent course compared to that in UC patients without PSC. In addition, patients require less immunosuppression independently of disease extent.62 Furthermore, the subclinical time span of IBD in PSC patients may be as long as 7 years before the onset of clinical symptoms. Thus, the quiescent course of the colitis in PSC may lead to an underestimation of the colitis in PSC unless all patients undergo colonoscopic investigation with multiple biopsies. Furthermore, proctosigmoidoscopy as the sole means of investigation is insufficient in diagnosing UC in patients with PSC, as the rectum often in spared in these patients. In PSC patients with UC requiring a proctocolectomy, an ileal pouch–anal anastomosis has proved to be the most appropriate definitive operation for the majority of cases.63 Non-specific inflammation of the pouch (pouchitis) is the most frequent long-term complication in UC patients operated on with an ileal pouch–anal anastomosis. Patients with PSC are more likely to have chronic pouchitis (60%) than UC patients without PSC (15%). The risk of pouchitis is not related to the severity of the liver disease. In PSC patients having pouchitis before liver transplantation, most continue to have pouchitis afterwards in spite of triple immunosuppression treatment, suggesting that antibiotic, rather than immunemodifying, therapy is the treatment of choice for pouchitis. In studies investigating the relationship between PSC and UC after liver transplantation, conflicting data have been found. In some studies the severity of UC improved significantly after transplantation. In contrast, some other reports have shown that the colitis worsened after transplantation.64 In a recent study from Pittsburg, the evolution of inflammatory bowel disease in PSC patients who had been transplanted was evaluated. Among 303 PSC patients who underwent liver transplantation 68% had inflammatory bowel disease. The risk of colectomy because of intractable disease was increased significantly after liver transplantation.64 All patients who
had an active colitis prior to transplantation developed active colitis afterwards, and exacerbation of the colitis occurred after steroid withdrawal. The reason for this is not known; however, it was hypothesised that cytomegalovirus, which has recently been implicated in some patients with steroid-refractory IBD65 could be part of the explanation.
PSC AND THE RISK OF COLORECTAL CARCINOMA IN PATIENTS WITH ULCERATIVE COLITIS Ulcerative colitis (UC) is a well-known risk factor for the development of colorectal carcinoma. The two major risk factors are long duration of the disease and extensive colitis. The cause of neoplastic change in UC remains unexplained. The majority of the patients with PSC who have an associated inflammatory bowel disease have UC with pancolitis. The presence of PSC has been shown to increase the risk of colorectal cancer or dysplasia in patients with UC.66,67 The first study to illustrate this was from Sweden and showed that the absolute cumulative risk of developing colorectal dysplasia/cancer in the PSC/UC group was 9%, 31%, and 50%, respectively, after 10, 20, and 25 years of disease duration. In the group with UC only, the corresponding risk was 2%, 5%, and 10%, respectively. An almost identical cumulative incidence of colorectal neoplasia was found both in a Finnish case–control study including 45 patients with UC and concomitant PSC and 45 pair-matched control patients with UC only, and in population-based Swedish study including 125 PSC patients. However, some studies have not shown an increased risk for UC patients with PSC to develop colorectal dysplasia/cancer. The discrepancies may be explained by different study designs. Because the published data on the risk of colorectal dysplasia/cancer in patients with PSC are conflicting, a metaanalysis has been performed.68 Eleven studies met all the eligibility criteria for the meta-analysis. Altogether, 16 844 patients with UC were included in the analysis; 560 had colorectal carcinoma. Of the 564 PSC patients included 60 had carcinoma of the colon. It was shown that PSC patients with UC are at an increased risk of colorectal dysplasia/cancer compared with patients with UC alone, OR 4.79 (CI 3.58, 6.41). The absolute cumulative risk of developing colorectal neoplasia in PSC and UC compared with UC only is shown in Figure 42-4. This increased risk is still present when the risk of colorectal cancer alone is considered. Because studies with null results have not been published, it was calculated how many studies with null results would be required to change the conclusion. It was found that 163 studies with null results would need to be done in order to change these results. Thus, it can be concluded that the presence of PSC is an independent risk factor for the development of colorectal dysplasia/cancer in patients with ulcerative colitis. The reason for the increased risk of colorectal neoplasia in PSC patients is obscure. Patients with cholestatic liver disease have decreased bile acid excretion and a relatively high proportion of secondary bile acids. It has been speculated that secondary bile acids play a role in the carcinogenesis of the colorectal mucosa in PSC. This
831
Section VI. Immune Diseases
Table 42-3. Associated Conditions Reported in PSC75 p<0.001
Cumulative risk (%)
40
30
20
10
UC + PSC UC
10
20
30
Duration Years Figure 42-4. Absolute cumulative risk of developing colorectal neoplasia in patients with ulcerative colitis with primary sclerosing cholangitis compared with patients with ulcerative colitis without primary sclerosing cholangitis (p < 0.001).68a
is supported by the fact that right-sided cancers seem to be more common in patients with PSC than in patients with UC alone.69 Moreover, two recent reports clearly demonstrated that treatment with ursodeoxycholic acid reduced the risk for developing colorectal dysplasia in patients with PSC and UC.70,71 This further supports the belief that bile acids do play a role in the development of colorectal neoplasia in PSC. Treatment with sulfasalazin/5-ASA has been shown to decrease the risk for colorectal carcinoma in patients with UC. In a study from Sweden this finding could not be confirmed, however, and the number of patients treated with sulfasalazin/5ASA did not differ in UC patients with and without PSC.69 PSC patients with UC remain at increased risk of developing colon cancer/dysplasia even after they have undergone liver transplantation.72 Vera et al. found that the cumulative risk of developing colorectal cancer was 14% and 17%, 5 and 10 years after liver transplantation.73 In a multivariate analysis it was shown that colorectal dysplasia after liver transplantation, duration of colitis >10 years, and pancolitis were independent risk factors for the development of colorectal carcinoma after liver transplantation. Patients developing colorectal cancer have significantly reduced survival. It is therefore important to emphasize that all patients with PSC and UC should be included in colonoscopy surveillance programs, probably earlier than UC patients without PSC. Also after transplantation, annual colonoscopies of these patients are suggested. The preferred colectomy procedure in patients with PSC has been to create an ileal pelvic pouch with an ileoanal anastomosis. A recent study investigated if patients with UC and PSC also have an increased risk of neoplastic transformation or development of atrophy in the ileal pouch mucosa after construction of an ileal pelvic pouch with an ileoanal anastomosis or construction of a continent Kock ileostomy.74 Flexible video endoscopic examinations of the ileal pouch were performed in 16 patients with UC and PSC and
832
Ulcerative colitis (80–100%) Crohns disease (7%) Pancreatitis (15–50%) Diabetes mellitus Thyroid disorders Psoriasis Polymyositis Celiac disease Rheumatoid arthritis Multiple sclerosis Sjögren’s syndrome Vitiligo Nephritis Systematic lupus erythematosus Vasculitis Sarcoidosis Fibrosing alveolitis Sacroileitis Idiopathc thrombocytopenic purpura Pyoderma gangrenosum Retroperitoneal fibrosis
in 16 UC patients without PSC, matched regarding type of reservoir, indication for surgery, pouch duration, age at onset of UC, age at follow-up, and UC-duration at time of colectomy. Multiple biopsies were sampled from different locations in the pouch for histological assessment of the degree of mucosal atrophy and dysplasia, and for flow cytometric DNA-analysis. The PSC patients developed moderate or severe atrophy in the pouch significantly more often than UC patients without PSC (p < 0.01). Persistent severe mucosal atrophy, a risk factor for neoplastic transformation, was revealed in eight PSC-patients and only in two controls. One of the patients with PSC had high-grade dysplasia in multiple locations. Low-grade dysplasia was assessed in three patients in the PSC group and in two of the controls. UC patients with PSC with an ileal reservoir are more prone to develop mucosal atrophy in the pouch and seem to have a higher risk of neoplastic transformation in the pouch mucosa than UC patients without PSC.
ASSOCIATED DISEASES BESIDES INFLAMMATORY BOWEL DISEASE A multitude of case reports on associated conditions to PSC have been published.75,76 Whether there are true associations with PSC itself or just parallel or secondary phenomena is not clear. A list of reported associated conditions is given in Table 42-3. Only one study has systematically evaluated the presence of autoimmune diseases in patients with PSC.76a Saarinen et al. studied 119 patients with PSC and found that 25% had an associated autoimmune disorder, most frequently insulin-dependent diabetes mellitus (10%) and thyroid disease (8.4%). In view of the close association between PSC and UC it is difficult to distinguish between disease associations related to IBD or PSC. Because of this fact 119 patients with PSC were matched for comparison against 119 controls with UC without liver disease. It was shown that patients with PSC had significantly
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
more associated autoimmune diseases than did patients with UC only. Therefore, many of the autoimmune conditions are in fact associated with PSC and not primarily with IBD. Besides IBD, pancreatitis appears to be the most common disorder associated with PSC, affecting 15–50% of all patients with PSC. Asymptomatic pancreatic changes are often seen on ERCP, although clinically important pancreatic exocrine failure is rare. A recent study has evaluated pancreatic abnormalities at magnetic resonance imaging (MRI). A comparison between 29 patients with PSC and 29 age- and gender-matched controls with other liver diseases (the majority with cirrhosis) was made. In this study 35% of the PSC patients and 48% of the control patients had pancreatic abnormalities.77 However, in another MRI study of PSC patients, changes in the pancreas, such as signal on T2-weighted images, decreased signal on T1-weighted images, enlargement of pancreas and decreased contrast enhancements were found to be specific findings in PSC compared with controls.78 Thus, pancreatic changes in patients with PSC seem common and can be identified both with ERCP and MRI. The diagnosis of chronic pancreatitis in PSC should be considered in patients with inappropriate steatorrhea or weight loss, especially in PSC patients who are not jaundiced and who have an early stage of the disease. The diagnosis is important as pancreatic enzyme supplements may be of benefit. The cause of pancreatitis in PSC remains unclear. It can be hypothesized: • That the biliary stricture can cause obstruction problems from the pancreas • That the inflammatory process directed against the biliary epithelium in some PSC patients also can be targeted against the epithelium in the pancreatic ducts • Changes in the pancreatic juice can cause secondary inflammatory changes.
It is, however, noteworthy that it has been shown that PSC patients have an increased risk of developing pancreatic cancer, which is also seen in patients with chronic pancreatitis.57 However, whether the patient with PSC and pancreatic cancer has had an underlying pancreatitis is unknown. Autoimmune pancreatitis or sclerosing pancreatitis is a recently described type of chronic pancreatitis not associated with alcohol and is responsive to treatment with corticosteroids. In a recent study from Japan including 388 patients with PSC the association between autoimmune pancreatitis and PSC was evaluated.79 In this study only 37% had a concomitant IBD, which was present in younger patients. Of all patients with PSC 7.2% were reported as having autoimmune pancreatitis, which was only seen in older patients. None of the patients had both IBD and autoimmune pancreatitis. Therefore, it seems as if the PSC patients with autoimmune pancreatitis represent a separate entity of PSC, which may be explained by geographical or environmental differences. However, similar cases have been reported in Europe.80 Speaking against sclerosing pancreatitis as a spectrum of PSC affecting just the pancreatic duct is a recent study from Japan, showing that patients with sclerosing pancreatitis have increased levels of serum IgG4, in contrast to patients with PSC, who had low levels.81 The reason for the relationship between celiac disease and PSC remains unknown, although an immunologic connection is suspected. The clinical awareness of possible coexistence is important, and if patients with PSC present with steatorrhea and weight loss, celiac disease must be considered.
DISEASE COMPLICATIONS Complications in PSC can be divided into four main categories (Figure 42-5):
Figure 42-5. Complications in PSC can be divided into four main categories.
Complications in PSC
Related to chronic cholestasis
Cirrhosis Liver failure
Extra hepatic disease
Specific for PSC
Pruritus
Peristomal varices
IBD
Fever
Fatigue
Pancreatitis
Abdominal pain
Fat soluble vitamin deficiency
Celiac disease
Dominant stricture
Diabetes
Gallstones (bladder, tract)
Thyroid disease
Cholangiocarcinoma
Osteoporosis
833
Section VI. Immune Diseases
1. Complications associated with chronic cholestasis 2. Complications associated with end-stage disease and portal hypertension 3. Complications specific for PSC 4. Complications related to associated disorders (see section about associated disorders). Health-related quality of life has emerged as an important outcome in the management of chronic diseases. Impairment of quality of life in PSC patients compared to the general population was confirmed in two independent studies.82,83 Patients with noncholestatic liver disease had lower health-related quality of life scores than did patients with cholestatic liver disease. Liver diseasespecific questionnaires showed similar declines in physical as well as mental function among patients with PSC, and the patients experienced a substantial improvement in all aspects of quality of life after liver transplantation.84
Complications and Symptoms Associated with Chronic Cholestasis These include pruritus, fat malabsorption, osteoporosis and fatsoluble vitamin deficiency. Fatigue is also a common complaint in patients with PSC, but owing to its subjectivity is often not included as a symptom. The cause of fatigue in cholestatic liver disease is unknown, although it has been suggested that altered serotoninergic transmission in the brain could account for this symptom.85 There is a problem when evaluating fatigue in general, as this symptom is highly subjective, can vary within the patient, and is difficult to measure objectively. Sometimes it is also difficult to differentiate between fatigue and minimal liver encephalopathy. Björnsson et al. compared the occurrence of fatigue and its impact on quality of life in patients with PSC, IBD and irritable bowel syndrome.86 Total fatigue score did not significantly differ between patients with PSC and IBD. Depression and general health were independent predictors for fatigue. The fatigue in PSC patients is often difficult to treat and leads to sick leave. A recent placebocontrolled study has evaluated the effect of treatment of fatigue in patients with PSC and primary biliary cirrhosis.87 Fluvoxamine, a selective serotonin receptor inhibitor (SSRI), was used in 17 patients with fatigue and either PSC or PBC. No effect was found after 6 weeks of treatment compared to the placebo group. Another troublesome symptom in PSC is pruritus, present in approximately one-third of patients, the severity of which is not closely linked to the severity of the underlying liver disease. Evidence has emerged to support the role of central nervous system in the pathogenesis of pruritus in cholestatic liver diseases. The treatment of pruritus in PSC is the same as for other cholestatic patients having this complication.88 The bile acid-binding resin cholestyramine is usually the first choice therapy for patients with cholestatic pruritus. It is important to point out that this medication also binds several other drugs, including ursodeoxycholic acid (UDCA), and needs to be taken 2–4 hours before other medications. The starting dose should be 4 g and can be escalated to 16 g. If the patient does not respond to cholestyramine, treatment with rifampicin can be initiated at a dose of 150–300 mg twice a day. As rifampicin can cause liver hepatotoxicity, liver function tests should be checked after starting this treatment. Treatment with opiate antagonists has been evaluated in
834
cholestatic liver diseases. Oral nalmefen (2 mg twice a day) improved pruritus in some patients. Symptoms of opioid withdrawal can occur during the initial treatment, possibly due to an increased opioidergic tone in the central nervous system in these patients.85 Osteoporosis is a common complication of cholestatic liver disease. Angulo et al. studied 81 PSC patients with bone mineral density of the lumbar spine and found significantly less value than expected compared with normal individuals. Seven patients (8.6%) had a bone mineral density below the fracture threshold. These patients were older, had a longer duration of IBD and a more advanced PSC. Pathologic fractures after liver transplantation occur in one-third of patients with PSC, and almost exclusively in those patients who already had ostopenia at the time of OLT. The cause of osteoporosis in PSC is multifactorial, and pathophysiological mechanisms specifically related to cholestatic disease have not been identified. The general principles for handling osteoporosis in PSC are similar to those followed for postmenopausal osteoporosis. In all patients optimal conditions for preventing bone loss should be established. Attention should focus on adequate calcium and vitamin D intake and physical exercise. It has been suggested that PSC patients should receive prophylactic therapy with calcium and vitamin D in a daily oral dose of 800 mg vitamin D and 1 g of calcium.89 Antiresorptive agents such as biphosphonates appear to be the most rational choice if treatment of osteoporosis is needed in PSC patients. Fat-soluble vitamin deficiency is a common feature in PSC patents with advanced disease. Among 56 PSC patients 40% had vitamin A deficiency, 14% vitamin D deficiency and 2% vitamin E deficiency.88 In PSC patients undergoing evaluation for liver transplantation the deficiencies were even more prominent. The patient’s fat-soluble vitamin level should therefore be monitored, and those found to have decreased levels are recommended replacement therapy. In patients having malabsorption it is important to exclude chronic pancreatits as well as celiac disease, as both are overrepresented in PSC.
Complication of Portal Hypertension Complications such as oesophageal varices, ascites and encephalopathy do not differ from other liver disorders. However, a special complication of portal hypertension in PSC patients with an ileal stoma is the development of peristomal varices. The majority of the patients developing peristomal varices also suffer from oesophageal varices. Patients bleeding from peristomal varices often present with recurrent hemorrhagic episodes. Local treatments to control bleeding are usually unsuccessful in the long run, and liver transplantation should be considered. It has been suggested that ileostomy should be avoided in PSC patients requiring proctocolectomy, and that ileoanal, ileorectal or ileal pouch–anal anastomosis, which are less susceptible to variceal formation and bleeding, are to be preferred in these patients. The presence of esophageal varices in patients with cirrhosis ranges from 30 to 80%. High sampling variability in liver biopsies hampers the diagnosis of cirrhosis in patients with PSC, and almost one-third of all cirrhosis is missed if only one biopsy is performed. Moreover, PSC patients present with portal hypertension already in a precirrhotic stage due to a significant presinusoidal component. In a study from the Mayo Clinic 283 PSC patients who had been investigated with upper gastrointestinal
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
endoscopy were included:90 102 (36%) had esophageal varices, 57 of which were moderate to large, and 28 patients had a history of variceal bleeding. Only 53% (54 of 102) of patients with varices had cirrhosis on liver biopsy. Among patients without previous bleeding the presence of a low platelet count (<150 ¥ 103/dl) was the best predictor of the presence of varices. Therefore, not only PSC patients with cirrhosis should undergo upper endoscopy to search for varices, but also patients with a low platelet count.
Complications Specific for PSC These include fever, bacterial cholangitis, gallbladder and biliary stones, abdominal pain, dominant biliary tract strictures, and cholangiocarcinoma. Episodes of fever may occur without an association with bacterial cholangitis, and disappear spontaneously without antibiotic treatment.91 Ascending cholangitis, secondary to bacterial infections of the biliary system, is uncommon unless the biliary tree has been surgically manipulated. There are obvious problems when trying to distinguish between episodes of idiopathic fever and chills, and those produced by acute bacterial cholangitis. The actual benefit from prophylactic antibiotic therapy to prevent recurrent bouts of cholangitis noticed in these patients has not been properly documented. If acute cholangitis does occur in patients who have not been operated upon, the presence of cholelithiasis, a dominant stricture or a bile duct cancer should be considered, prompting a cholangiography. Chronic recurrent cholangitis may become difficult to treat and can be an indication for liver transplantation.92 Chronic cholestasis is associated with the development of cholesterol gallstones, and recurrent cholangitis may lead to the development of pigment gallstones. Most stones in PSC patients are confined to the gallbladder, and up to one-third of patients with PSC are reported having gallbladder stones. Moreover, some patients with PSC also have concomitant biliary tract calculi that develop after the PSC diagnosis. Patients with PSC and biliary tract calculi have more symptoms, especially ascending cholangitis, than those with PSC unassociated with biliary tract calculi. Small biliary stones and pigment debris, found in up to one-third of patients, can be effectively treated endoscopically. Dominant strictures located in the common bile duct or common hepatic duct in the liver hilum affect 10–40% of all PSC patients at some time during the course of the disease.93 It is usually difficult to differentiate between a dominant benign stricture and a malignant stricture due to cholangiocarcinoma. Whenever a dominant stricture occurs brush cytology and fine needle biopsy should be performed in order to exclude a malignant change in the biliary system. Although the sensitivity of brushings may be low, the specificity is usually high (see section on diagnosis of cholangiocarcinoma).
HEPATIC MALIGNANCIES IN PRIMARY SCLEROSING CHOLANGITIS PREVALENCE Parker first described an association between UC and cholangiocarcinoma in 1954.93a Later, in 1971, Converse et al. found that bile duct carcinoma in UC most commonly occurs in patients with pre-
existing PSC.93b Today, we know that PSC can be complicated by cholangiocarcinoma, gallbladder and hepatocellular carcinoma (HCC), as well as fibrolamellar carcinoma. The increased risk for cholangiocarcinoma in PSC is well established, the prevalence varying in different studies between 5% and 20%.57,94,95 The main reasons for variation in the number of PSC patients found to have cholangiocarcinoma are probably differences in patient selection, diagnostic ambitions and autopsy rates. This is well reflected by data from a Swedish study including 305 PSC patients. The overall prevalence of cholangiocarcinoma in this study was 8%. However, among PSC patients with a follow-up of more than 5 years, 16% developed cholangiocarcinoma. Seventy-nine (26%) patients in this study died or underwent liver transplantation, and 30% of those were found to have cholangiocarcinoma. In the group of PSC patients who died, only 69% underwent autopsy. It has been recognized in several studies that cholangiocarcinoma in some PSC patients will only be revealed at autopsy. The largest study that has evaluated the risk of cholangiocarcinoma in PSC includes 604 Swedish PSC patients identified between 1970 and 1998.57 Since Sweden has approximately 8.9 million inhabitants, this cohort accounts for a considerable proportion of all Swedish PSC patients. The diagnosis of PSC was based on biochemical, clinical and cholangiographic features. Median time of follow-up was 5.7 years (range 0–27.8), 79% had concomitant inflammatory bowel disease; 74% of all dead patients without a diagnosis of hepatobiliary carcinoma before death underwent autopsy. Because all Swedes are assigned a unique 10-digit national registration number, a follow-up was provided though links to the Swedish Cancer and Death Registries. Cumulative incidences of malignancies and standard incidence ratios (SIR) were calculated with the incidence rates in the Swedish population. The risk for hepatobiliary malignancy was increased 161 times. The prevalence of cholangiocarcinoma in this study was 13%. Of the patients with cholangiocarcinoma 37% developed the tumor within 1 year after PSC diagnosis. The incidence rate of cholangiocarcinoma was 1.5% per year in PSC patients with disease duration of more than 1 year. A recent study from the Mayo Clinic confirms these data:94 174 patients with PSC were followed for a median of 11.4 years, and the incidence of cholangiocarcinoma was 0.6% per year, the highest incidence being within 2.5 years after PSC diagnosis, with a 1.34% incidence rate per year. The prevalence of cholangiocarcinoma in this study was approximately 7%. In 30–50% of all PSC patients with hepatobiliary malignancy, cholangiocarcinoma is diagnosed at the same time as PSC or within a year of the diagnosis.57,96 There is a risk that some of these cancer patients may not have an underlying PSC, as cholangiocarcinoma without PSC may present with cholangiographic changes similar to those seen in PSC patients without cholangiocarcinoma. Therefore, it is important to evaluate the clinical characteristics in such patients, including concomitant IBD, previous rise in liver function tests, and the presence of cirrhosis. Cholangiocarcinoma in PSC can be intra- or extrahepatic. In a study by Ahrendt et al.171 including 25 PSC patients with biliary malignancy, 76% of the tumors were located in the perihilar region, 16% were intrahepatic and 8% were located in the gallbladder. Cholangiocarcinoma is a leading cause of mortality in patients with PSC. The prognosis is dismal, with a medium survival time of 5 months after diagnosis (Figure 42-6). Cholangiocarcinoma in PSC
835
Section VI. Immune Diseases
1.0 0.9 0.8
Surviving
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0
1
2
3 4 5 6 7 8 Time since diagnosis (years)
9
10
11
Figure 42-6. Survival of 72 patients with primary sclerosing cholangitis and cholangiocarcinoma or gallbladder carcinoma. Eleven of the patients were transplanted. In three of them the cholangiocarcinoma was diagnosed prior to the transplantation and in the remaining eight the tumors were diagnosed after transplantation.57
develops when patients are in their 40s, i.e. about 20 years earlier than cholangiocarcinoma in patients without PSC.
PATHOGENESIS OF CHOLANGIOCARCINOMA IN PSC Cholangiocarcinoma is a malignant proliferation of bile duct epithelial cells that can arise at any point in the biliary tree. The factors responsible for the malignant bile duct change in PSC are not known, and cholangiocarcinoma in PSC can arise at any stage of PSC. PSC is characterized by proliferation and fibrosis of the bile ducts, caused by a chronic inflammation. The fact that patients with chronic Clonorchis sinensis and Opisthorchis viverini (liver fluke) infection also run an increased risk of developing cholangiocarcinoma may suggest that long-standing inflammation of the bile ducts enhances the risk for malignancy. When the cholangiocytes are continuously exposed to the concerted actions of chronic inflammatory agents and hydrophobic bile acids, the cells may become predisposed to oncogenic mutations and further progression to the malignant state. Recently, it was suggested that NO-mediated inhibition of 8-oxodG base excision DNA repair processes could be a mechanism potentiating mutagenic DNA damage in patients with chronic inflammation such as PSC in the biliary tract.26 Activation of onc genes,98,99 functional loss of tumor suppressor genes,98,100 dysregulation of bile duct cell proliferation and apoptosis100 are all mechanisms operating in the development of cholangiocarcinoma in PSC. Ras genes encode proteins involved in signal transduction, affecting cell proliferation and differentiation. Activation of these genes leads to uncontrolled cell growth. Boberg et al. found K-ras mutations in 33% of cholangiocarcinoma from patients with PSC.98 Similar figures have been reported in cholangiocarcinoma without PSC.101 Inactivation of the tumor suppressor gene p53 is present in a variety of human cancers. Loss of function of p53 leads to insufficient initiation of cell cycle arrest, or apoptosis if excessive DNA damage has occurred. Another important function of p53 is suppression of Bcl-2 gene expression, which is a major gene involved in
836
apoptosis. p53 inactivation seems important for the development and progression of cholangiocarcinoma in PSC. Mutation of p53 in PSC-related cholangiocarcinoma varies from 30 to 80%.98,100 However, p53 mutation seems to be a late event in tumor development, as no p53 accumulation was seen in areas of biliary dysplasia in a study on PSC patients.100 It has been shown that loss of chromosome 9p21, and inactivation of the p16 tumor suppressor gene – both having critical roles in the cell cycle machinery – are common events in PSC associated cholangiocarcinoma.102 Whether or not p16 inactivation and loss of chromosome 9p21 can serve as predictors of cholangiocarcinoma in PSC remains to be determined. Increased bile duct proliferation and bile duct dysplasia may be morphological steps in the transition from benign to malignant bile duct epithelium in PSC. This is suggested by the fact that cholangiocarcinoma is often multifocal, and biliary dysplasia is observed in non-tumor liver tissue distant from the tumor.56,100 Moreover, biliary dysplasia may precede the development of cholangiocarcinoma. In a study of criteria for bile duct dysplasia in PSC it was shown that with the adoption of strict definitions of this histological finding, an acceptable agreement could be achieved among three liver pathologists.56 Moreover, it was shown that bile duct dysplasia occurred mostly in PSC patients with cholangiocarcinoma and in patients who later will develop this complication.103 In a study by Ludwig et al. biliary dysplasia was however only found in one patient among 60 PSC patients who underwent liver transplantation.103a None of the patients included in the study by Ludwig suffered from cholangiocarcinoma.
DIAGNOSIS OF CHOLANGIOCARCINOMA IN PSC Clinical Symptoms or Signs of Cholangiocarcinoma Cholangiocarcinoma in the setting of PSC is difficult to diagnose and is often diagnosed at an advanced stage of tumor growth and spread, or accidentally at liver transplantation in end-stage PSC. Clinically, biliary malignancy is often suspected when a PSC patient shows rapid, progressive liver disease with increasing bilirubin levels, weight loss and abdominal pain. However, end-stage PSC without cholangiocarcinoma can also present with rapid disease progression, and it is not possible to distinguish clinically between end-stage PSC and PSC complicated by cholangiocarcinoma. Abdominal pain is a symptom that has been shown to be especially associated with cholangiocarcinoma in PSC. In a study from the Mayo Clinic including 174 patients with PSC the opposite was found, with decreased risk of cholangiocarcinoma on a univariate analysis (RR = 0.27) for patients with symptoms (fatigue, pruritus, abdominal pain).94 This study also showed that a history of variceal bleeding was associated with an increased risk for the development of cholangiocarcinoma.94 These conflicting data demonstrate the overlap between clinical features in PSC patients with and without cholangiocarcinoma. There is a problem in differentiating between malignant and nonneoplastic strictures cholangiographically, as fibrotic strictures of the bile ducts are already present at the onset of the disease process in PSC (Figure 42-7). It has been suggested that marked dilatation of the ducts or ductal segments and the appearance of a polypoid mass 1 cm or more in diameter are mostly found in PSC patients with malignancy. In a retrospective study encompassing PSC patients
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
Figure 42-7. A cholangiography from a patient with PSC complicated by cholangiocarcinoma. The arrow indicates the location of the tumor.
with cholangiocarcinoma, the benefit of computed tomography (CT), cholangiography, ultrasound and MRI in demonstrating cholangiocarcinoma has been evaluated. Polypoid bile duct masses were seen only in two patients on cholangiography. The most sensitive method was MRI, although this was only performed in a minority of the patients. It was possible to detect cholangiocarcinoma in PSC in 80% of the patients with a combination of the various radiological methods. However, in studies in which PSC patients both with and without cholangiocarcinoma have been included, lesions highly suspicious of being malignant were also noted in PSC patients not suffering from cholangiocarcinoma. In a controlled study CT and cholangiography were evaluated for the detection of cholangiocarcinoma in PSC. The sensitivity and specificity for the detection of cholangiocarcinoma with CT were 82% and 80%, respectively.104 Controlled studies for the sensitivity of MRI for the detection of cholangiocarcinoma in PSC are lacking. The first report of positron emission tomography (PET) using a radiolabeled glucose analog (FDG-PET) accumulating in malignancies for detecting cholangiocarcinoma in PSC was published in 1998
by Keiding et al.104a The study showed promising results regarding the possibility of revealing small cholangiocarcinomas in PSC. Since then, FDG-PET has been shown to be very sensitive for the detection of distant metastases, but less so for the detection of regional lymph node metastases in bile duct cancers.105 Furthermore, FDGPET has been evaluated for the purpose of staging patients with biliary tract cancer. In 36 patients with cholangiocarcinoma (seven had concomitant PSC) the overall sensitivity for nodular cholangiocarcinoma with a mass greater than 1 cm was 85% and the specificity 50%. For infiltrating cholangiocarcinoma the sensitivity was only 18% and specificity 100%. In 58% of patients with biliary stents FDG uptake was seen along stents in the bile duct. In the small group of patients with concomitant PSC (n = 7) one false positive result was identified.106 This patient had an ongoing bacterial cholangitis. In summary, FDG-PET is specific for the detection of distant metastases but in patients with PSC there are false positives. There is a need to identify PSC patients before manifest cancer occurs, as well as those at increased risk of developing cholangiocarcinoma. False positive diagnosis of malignancy must always be avoided, as cholangiocarcinoma may disqualify a patient from a life-saving liver transplantation. To predict whether a PSC patient has developed cholangiocarcinoma at the time of evaluation for liver transplantation is a challenge. The Mayo Model risk score has been evaluated in PSC patients undergoing liver transplantation, and a marked increase in the incidence of biliary malignancy was shown at a score above 4.4. The Mayo PSC Natural History Model in that study was based on a formula that included bilirubin level, histological stage, age, and the presence of splenomegaly. Brandsaeter et al.107 recently presented data from all Nordic PSC patients listed for transplantation during 1990–2001. In this study hepatobiliary malignancy was found in 20% (52/255) of the patients. Recent diagnosis of PSC, no UDCA treatment, clinical suspicion of cholangiocarcinoma and previous colorectal cancer were shown to be predictors of malignancy. ‘Clinical suspicion of cancer’ in this study was based on one or more of the following parameters: clinical course, findings on CT or ERC, and raised levels of CA 19-9. A suspicion of malignancy prior to liver transplantation was raised in 67% (35/52) of the patients. Only 20 of these proved to have cholangiocarcinoma in the explanted liver. Among patients with no suspicion of malignancy prior to liver transplantation 10% (17/166) were actually found to have cholangiocarcinoma. This further illustrates the difficulty of predicting cholangiocarcinoma in PSC. Interestingly, in the study by Brandsaeter, the outcome for the patients with PSC and cholangiocarcinoma after transplantation was rather good, with a 1-, 3- and 5-year survival of 65%, 35% and 35%, respectively. With the increased use of MRI, increased awareness of the risk of cancer with active and repeated brushings in the biliary tree via ERC, the number of patients diagnosed accidentally at liver transplantation will probably decrease. In a study by Burak et al.94 including 174 PSC patients and 11 with cholangiocarcinoma, no patient was identified accidentally at liver transplantation.
Histology and Brush Cytology Bile duct carcinomas in PSC are often scirrous in nature, and even if malignant changes are suspected at CT or ultrasonography, it is often difficult to obtain a representative biopsy specimen. Moreover,
837
Section VI. Immune Diseases
even if representative material is gathered, it can still be difficult to differentiate between non-neoplastic and malignant changes in needle biopsies. Brush cytology from strictures obtained at ERC has a good specificity but a relatively low sensitivity, and the method is therefore of limited value in diagnosing cholangiocarcinoma in PSC.108 Repeated brushings can increase sensitivity and are recommended when there is a strong suspicion of cholangiocarcinoma and the initial material is negative. Despite the high specificity there are false positive cases. It is difficult to differentiate between cellular atypia generated by chronic inflammation and neoplasia, and the intraobserver variability is high. Ponsioen et al.110 evaluated brush cytology of dominant strictures in PSC. This study included 47 brush samples from 43 PSC patients and the sensitivity and specificity for cholangiocarcinoma diagnosis were 60% and 89%, respectively. Immunohistochemical analyses of p53 and K-ras mutations were also carried out, but showed no additional value in detecting malignancy. The use of endosonography-guided (EUS) fine needle aspiration can further improve the sensitivity. This method is safe and provides additional information on the nature of hilar strictures as well as a good sample for cytologic evaluation.111 A high prevalence of DNA aneuploidy in cholangiocarcinoma (80%) from patients with PSC and low prevalence of DNA aneuploidy in benign strictures (12%) has been shown.100 Using DNA measurements from brush samples in the biliary tree might improve the diagnostic yield of malignancy.112,113 In a prospective study by Lindberg on biliary strictures (20 PSC), DNA analysis with flow cytometry of brush samples showed a sensitivity of 52% and a specificity of 96%.112 This is in line with another prospective study of biliary strictures in 110 patients undergoing ERC. This study compared the accuracy of digital image analysis and routine cytology, the sensitivities of which were 39.3% and 17.9%, and the specificities 77.3% and 97.7%, respectively.114 DNA analysis, either by flow cytometry or digital image analysis, is a useful adjunct to routine cytology for the detection of malignancy in biliary strictures. Fluorescence in situ hybridization (FISH) for detection of malignancy in biliary strictures has recently been evaluated in 131 patients.109 The sensitivity was 34% and specificity 91% using FISH, which was significantly better than cytology. However, in the subgroup of patients with PSC the sensitivity and specificity were shown to be no better than cytology. FISH as an additional tool to routine cytology in PSC needs to be further evaluated. As mentioned above, biliary duct dysplasia has been suggested to precede cholangiocarcinoma in PSC. The presence of biliary duct dysplasia in a liver biopsy strengthens the suspicion of cholangiocarcinoma in another part of the liver.100 An example of a bile duct from a patient with PSC with biliary dysplasia is shown in Figure 42-8. A majority of cholangiocarcinomas in PSC show overexpression of the p53 protein, which is not seen in bile ducts with benign proliferation. Should overexpression of p53 be found, it speaks in favor of bile duct malignancy; however, this is not yet part of clinical routine.
Serum Tumor Markers The best-evaluated and most commonly used tumor markers in clinical practice for the detection of bile duct carcinoma in PSC are CA 19-9 and CEA. Two additional tumor markers (CA 50 and CA 242)
838
were evaluated. It was concluded that these tumor markers could not serve as early markers for the detection of cholangiocarcinoma in PSC.117 CEA is a glycoprotein normally found in the foetal gut, pancreas and liver. CEA was the first tumour marker described to be associated with colorectal carcinoma. CEA has its greatest application in the follow-up of patients with colorectal carcinoma after tumour resection. After curative surgery this marker returns to normal within weeks after resection. CA 19-9 is a mucin type glycoprotein, which is increased in the serum in malignant pancreatic and biliary tumours. In 1995 Ramage et al. published a study, including 74 PSC patients, in which a combination of the two tumour markers of CA 19-9 and CEA was evaluated in detecting cholangiocarcinoma. Cut off values of 5 ng/mL for CEA and 200 U/ml for CA 19-9 resulted in a sensitivity of 53% and 60% for CEA and CA 19-9, respectively. The specificity was 86% for CEA and 91% for CA 19-9. Using the formula CA 19-9 + (CEA ¥ 40) with a cut off level of 400 a sensitivity of 66%, a specificity of 100% and an accuracy of 86% in diagnosing cholangiocarcinoma was achieved. Later, the Ramage formula 400 has been evaluated by others.116 The sensitivity and specificity has not been shown to be as high as in the original study by Ramage.116 In a study by Björnsson et al. 72 PSC patients were followed with multiple serum measurements of CA 19-9 and CAE.116 Seven of the 72 PSC patients developed cholangiocarcinoma. Among the PSC patients with cholangiocarcinoma only half had CA 19-9 values above the upper reference value (37 U/ml), and one third had values above 200 U/ml. The sensitivity and specificity of CA 19-9 was 38% and 81% respectively when the cut off level 200 U/ml was used. In a recent study by Siqueira et al. the sensitivity and specificity of serum levels of CA 19-9 or CEA on diagnosing cholangiocarcinoma in PSC was investigated.116 Of the 333 PSC retrospectively studied patients, the sensitivity and specificity for CEA in 144 patients was 68% and 81% and for CA 19-9 in 55 patients was 67% and 98% respectively. The results of this study clearly demonstrate that the cut off level for CA 19-9 for the diagnosis of cholangiocarcinoma is important. When a cut off level of 37 U/ml was used the specificity decreased from 98% to 40%. It was also shown that the use of both the markers CEA and CA 199 was superior to either of them alone. PSC patents with concomitant cholangitis or with an active inflammation in the bile ducts frequently have high levels of serum tumour markers. Björnsson et al. found a correlation between high CA 19-9 values and serum alkaline phosphatase levels.116 There is, however, no correlation between the levels of the serum tumour markers CEA and CA 19-9 and bilirubin levels, which is a consistent finding in several studies. The possibility of early detection of tumours or premaligancy by the use of serum CEA and CA 19-9 is very low. In the study by Ramage, eight patients had incidental cholangiocarcinoma diagnosed first at examination of the explanted liver. Two of the eight patients had a tumour marker score of less than 400 and were the only patients in this group with a more than 6 months survival without tumour recurrence. In the study by Hultcrantz et al. an effort to diagnose cholangiocarcinoma at an early phase was made.116 Four tumour markers (CA 19-9, CEA, CA 50, and CA 242) were evaluated in 75 PSC patients who were prospectively followed for 3
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS Figure 42-8. Normal interlobular bile ducts (A) compared with a dysplastic bile duct (B) from a patient with PSC.56
A
B
years. Two patients developed cholangiocarcinoma during the study; one had normal and one increased serum tumour marker levels. Transient, non-cholangiocarcinoma associated non cholangitisassociated elevations of CA 19-9 above 35 U/ml were seen in five patients. However, none of these patients had CA 19-9 values above 200 U/ml, which was the cut off level in the study by Ramage et al. In the study by Hultcrantz, two additional patients developed cholangiocarcinoma at follow up (8 yr.) and none of them had earlier shown elevated tumour markers. In summary, the CEA and CA 19-9 serum tumor markers have a role in diagnosing cholangiocarcinoma in patients with PSC. The use of both can improve sensitivity and specificity. A mild rise in CA 199 values is non-specific and should be interpreted cautiously. Awareness of the presence of false negative as well as false positive results is important. In most transplant centers PSC patients with cholangiocarcinoma will not be accepted for liver transplantation. A rise in
CA 19-9 and/or CEA cannot be used to exclude a PSC patient from a potentially life-saving liver transplantation. A summary of the tools for diagnosing cholangiocarcinoma in PSC is given in Table 42-4.
Tumor Markers in Bile The lack of sensitive and specific clinical, radiologic and serologic markers for the detection of cholangiocarcinoma in PSC has led to the search for markers of malignancy in bile. Two recent studies have evaluated CEA and CA 19-9 as diagnostic tools for cholangiocarcinoma in PSC in the bile.112,115 The results from these studies both show that measurement of these tumor markers in bile is of no clinical value in diagnosing cholangiocarcinoma in PSC. Therefore, analyses of CEA and CA 19-9 in bile seem to have no advantage over serum analyses. Mutations of the K-ras onc gene have been described in pancreas and bile duct carcinomas. K-ras mutations have also been detected
839
Section VI. Immune Diseases
Table 42-4. Sensitivity and Specificity of Methods for Detecting Cholangiocarcinoma in PSC (Message: It is important to combine diagnostic tools to increase the sensitivity and specificity) Method
Sensitivity
Specificity
Clinical features/rapid disease progression Mayo Model Risk Score ERC CT MRI* PET106 Brush cytology110,112 Tumor markers (CA 19-9, CEA)115,117 DNA measurement112 FISH109 Biliary dysplasia*
Poor
Poor
Good Poor Moderate Good Good Poor Good Moderate Moderate ?
Good Poor Moderate Good Good Excellent Good Good Excellent ?
*Controlled studies are lacking.
in stools from patients with cholangiocarcinoma and pancreatic cancer. In a study by Saurin et al118 bile specimen were obtained from 117 non-PSC patients with biliary strictures to determine whether the detection of K-ras mutations could differentiate between strictures of benign and malignant origin. K-ras gene point mutation was detected in bile from 31% (22/90) patients with primary malignancy in the bile ducts and in 4% (1/25) of those with benign strictures. It was concluded that detection of K-ras mutations in bile appears to be specific for differentiating between benign and malignant biliary strictures. However, this does not seem to be the case in patients with an underlying PSC. In PSC, K-ras point mutations seem to be present in bile at an early stage of malignancy, and are frequent in PSC patients without signs of malignancy. This has been demonstrated by Kubicka et al.,99 who studied the occurrence of K-ras mutations in the bile fluids obtained by ERC in 56 patients with PSC and 20 with other cholestatic liver disease. None of the patients without PSC had K-ras mutations in the bile fluid, whereas 17 (30%) of the PSC patients revealed K-ras mutations in bile fluid. At follow-up no patient with PSC without K-ras mutations developed cholangiocarcinoma. Among the 17 patients with Kras mutations in the bile two developed cholangiocarcinoma 14 and 34 months after first detection of the gene alteration. Liver transplantation in 19 of the 56 PSC patients revealed two additional patients with incidental cholangiocarcinoma and two with highgrade biliary dysplasia. All four of these patients had previous K-ras mutations in the bile. Thus, the presence of K-ras mutations in bile from non-PSC patients is specific for biliary malignancy. In patients with PSC, however, K-ras mutations in bile might be interpreted as an early sign of malignancy or even a risk factor for the later development of cholangiocarcinoma. The relatively low frequency of K-ras mutations (30%) in cholangiocarcinoma tissue from PSC patients may, however, limit the value of such analyses.119
RISK FACTORS FOR CHOLANGIOCARCINOMA IN PSC The identification of risk factors for cholangiocarcinoma development in PSC could lead to the possibility of interfering with the
840
malignant transformation. In addition, knowledge of the possible risk factors for cancer may be of value in identifying PSC patients in whom early liver transplantation could be recommended. Duration of disease is not a risk factor for cholangiocarcinoma development.94 This is supported by the fact that many patients develop cholangiocarcinoma within the first year of PSC diagnosis, and that cholangiocarcinoma in PSC occurs both in patients with and without cirrhosis.120 This is in contrast to the risk of developing hepatocellular carcinoma, which mostly is seen in patients with underlying cirrhosis. However, Burak et al. showed that bleeding from oesophageal varices, occurring only in advanced PSC, was associated with an increased risk of cholangiocarcinoma.94 PSC can involve the intrahepatic or the extrahepatic biliary systems, most commonly both. In patients with UC it is well known that the colonic distribution of the colitis is an independent risk factor for the development of colorectal cancer. UC patients with pancolitis have a significantly higher risk of colorectal carcinoma than patients with left-sided colitis. In PSC it has not been shown that patients with intra- and extrahepatic involvement of the disease have an increased risk of developing cholangiocarcinoma compared to those having involvement of the intrahepatic ducts only. Small bile duct PSC represents patients with abnormal liver function tests, a characteristic liver histology but a normal cholangiography, and no other cholestatic disorder. Three recent studies present data on small bile duct PSC, including altogether 83 patients.45–47 As described above, none of these patients was found to acquire cholangiocarcinoma, although one developed hepatocellular carcinoma and 13% progressed to involvement of the large bile ducts. Patients with primary biliary cirrhosis, a biliary disease only affecting the small bile ducts, do not have an increased risk for cholangiocarcinoma development. It therefore seems justified to conclude that only patients having chronic inflammation of the large bile ducts are at an increased risk for cholangiocarcinoma development, regardless of whether only the intra- or the extrahepatic biliary tree is affected. Bergquist et al.121 have shown that PSC patients who are previous or current smokers are at increased risk for the development of cholangiocarcinoma. In another study this finding could not be confirmed, but another association between alcohol consumption and the risk for cholangiocarcinoma development was found. Also these findings need confirmation, although it seems wise to recommend PSC patients to abstain from smoking as well as overconsumption of alcohol. The majority of PSC patients have an associated inflammatory bowel disease. In a European multicenter study including 394 PSC patients from five countries cholangiocarcinoma was diagnosed in 48 (12.2%).96 Inflammatory bowel disease was found in 325 (82%) of the patients. It was suggested that a long-standing duration of inflammatory bowel disease was a risk factor for cholangiocarcinoma development. Patients who developed cholangiocarcinoma were diagnosed with inflammatory bowel disease at least 1 year before PSC more often than in the group without cholangiocarcinoma (90% vs 65%, respectively; p = 0.001). Moreover, the duration of inflammatory bowel disease before diagnosis of PSC was significantly longer in the cancer group than in the group without cancer (17.4 years vs 9.0 years; p = 0.009). The presence of colorectal dysplasia/cancer in patients with concomitant UC seems to be another risk for cholangiocarcinoma
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
Table 42-5. Risk Factors for Cholangiocarcinoma in PSC Factors associated with an increased risk for cholangiocarcinoma
Factors not associated with an increased risk of cholangiocarcinoma
Smoking Alcohol95 Long duration of IBD96 Cororectal cancer/dysplasia122 Variceal bleeding94 Biliary dysplasia56,100 K-ras mutations in bile99
Cirrhosis Long duration of PSC Extension of large duct PSC Gender
1.0 0.9
No HB-malignancy
0.8
development in PSC.122 In a case–control study comparing 40 PSC patients to 80 age- and sex-matched controls with UC but without PSC, 10 PSC patients developed cholangiocarcinoma. Among the 10 PSC patients with cholangiocarcinoma nine also had inflammatory bowel disease. Seven of the patients with cholangiocarcinoma also had colorectal dysplasia/cancer. Thus, cholangiocarcinoma was significantly more common in UC patients with PSC and colorectal neoplasia than in those with UC and PSC without colorectal dysplasia/carcinoma (p <0.02). The finding that colorectal dysplasia/cancer should be a risk factor for the future development of cholangiocarcinoma is confirmed by Brandsaeter et al.,122 who showed that previous colorectal carcinoma was a risk factor for the development of hepatobiliary carcinoma. In the recent study by Burak et al.94 proctocolectomy in patients with IBD and PSC was found to be associated with the development of cholangiocarcinoma. However, no association between colorectal cancer/dysplasia and cholangiocarcinoma was found, but data were lacking concerning the reason for colectomy in many patients. As discussed previously, K-ras mutations in bile, biliary dysplasia and DNA aneuploidy are suggested as predictors for the later development of cholangiocarcinoma. Table 42-5 summarizes the risk factors for cholangiocarcinoma in PSC.
TREATMENT FOR CHOLANGIOCARCINOMA IN PSC Liver transplantation is a well-established treatment in end-stage PSC, with excellent survival rates. Liver transplantation, however, is usually not recommended for cholangiocarcinoma, as tumor recurrence is frequent. Incidentally detected cholangiocarcinoma of less than 1 cm in diameter discovered at the time of gross examination of the explanted liver did not influence the chance of survival, but cholangiocarcinoma diagnosed prior to liver transplantation predicted a poor prognosis. In the Nordic study by Brandsaeter et al.122 the 1-, 3- and 5-year survival for PSC patients with cholangiocarcinoma was 65, 35 and 35%, respectively (Figure 42-9). Cancerpositive hilar lymph nodes and the presence of tumor tissue at the resection margin have a negative impact on the outcome of transplanted PSC patients with cholangiocarcinoma. Ahrendt et al. studied the outcome for 50 PSC patients who had been managed with extrahepatic biliary resection between 1980 and 1995.122a The operative complication rate was 32% and the operative mortality was 6% in all patients; 28% (14/50) of patients have died since 1980, and none have developed cholangiocarcinoma. Owing to high com-
n=192
Survival probability
0.7 0.6 0.5
HB-malignancy
n=30
0.4 0.3 0.2 p<0.001
0.1 0.0 0
1
2
3
4
5
6
7
8
9
10 11 12 13
Years post-transplantation Figure 42-9. Survival following liver transplantation. Patients with malignancy versus patients without malignancy.122
plication rates extrahepatic biliary resection is not a therapeutic alternative in general, but might have a role in highly selected patients. Resection of cholangiocarcinoma in PSC is hampered by the fact that the tumor is often multifocal. The Mayo Clinic has developed a management protocol for highly selected PSC patients with perihilar cholangiocarcinoma.123 Neoadjuvant chemoirradiation and 5-fluorouracil are given, followed by oral capecitabine while awaiting liver transplantation. A staging laparotomy is performed, with biopsy of lymph nodes, examination of the tumor and routine biopsy of regional lymph nodes. Only patients with negative staging operations remain eligible for liver transplantation. The survival following liver transplantation in 28 patients, who were obviously highly selected and who underwent this management, was 88% at 1 year and 82% at 5 years. A few uncontrolled studies indicate that treatment with ursodeoxycholic acid protects against the development of cholangiocarcinoma.122,124 The fact that treament with UDCA reduces the risk of developing colorectal dysplasia/cancer in PSC patients with UC certainly makes this drug attractive for further studies of cancer prevention in PSC.
PSC AND GALLBLADDER CARCINOMA The gallbladder epithelium in PSC patients has features of chronic inflammation similar to that found in the liver. Moreover, it has been shown that the frequency of gallbladder polyps in PSC patients is 4%. It therefore seems logical to suppose that patients with PSC also are prone to develop gallbladder carcinoma. In several studies investigating the incidence of cholangiocarcinoma in PSC, patients also having gallbladder carcinomas have been included in the cohort of cholangiocarcinoma patients, as there are difficulties in separat-
841
Section VI. Immune Diseases
ing the two entities. In a recent study the prevalence of gallbladder cancer was assessed in PSC patients with a gallbladder mass who had undergone cholecystectomy.125 Among 102 PSC patients who underwent cholecystectomy at the Mayo Clinic between 1977 and 1999, 14 (13.7%) had a gallbladder mass. In eight of the 14 the mass was caused by adenocarcinoma. In the six patients with benign masses 33% had epithelial cell dysplasia. It is noteworthy that the mass lesions were not recognized preoperatively at ultrasound in five (42%). It is important to point out that five of the malignant gallbladder masses were £ 10 mm in size. Thus, gallbladder polyps in patients with PSC should be removed earlier than in other patients, where often a size >10 mm is required before a cholecystectomy is recommended, and ultrasound of the gallbladder yearly is recommended.
PSC AND HEPATOCELLULAR CARCINOMA (HCC) There have been several case reports on HCC in PSC. In a study from the Mayo Clinic including 134 PSC patients with cirrhosis undergoing liver transplantation HCC was found in three (2%). None of the PSC patients with HCC had evidence of hepatitis B or C infection or other risk factors for the development of HCC. There were no significant differences in clinical characteristics in PSC patients with and without HCC. Thus, HCC can occur in end-stage PSC, and PSC patients with cirrhosis should be included in surveillance programs with ultrasonography. Case reports have also described the association of PSC with fibrolamellar carcinoma.
PANCREATIC CARCINOMA In a Swedish study including 604 PSC patients the risk for developing extrahepatic malignancies was also estimated.57 It was found for the first time that PSC patients had a 10–14-fold increased risk of developing pancreatic carcinoma compared to the general population. None of these patients had a known history of chronic pancreatitis. There might be a risk of detection bias, as it might sometimes be difficult to differentiate pancreatic carcinoma arising from caput pancreaticus from cholangiocarcinoma arising from the large bile ducts. However, Isaksson et al.125a found PSC to be a risk factor for the development of pancreatic carcinoma. The authors identified 15 000 patients with pancreatic carcinoma, together with all associated diseases in these patients. The presence of PSC was significantly overrepresented among patients with pancreatic carcinoma. Of PSC patients 15–50% have changes at cholangiography that indicate chronic pancreatitis. Chronic pancreatitis not associated with PSC is a known risk factor for the development of pancreatic carcinoma.
SUMMARY Patients with PSC are at increased risk of developing cancer at the sites exposed to chronic inflammation, such as the biliary tree, the colon and the pancreas. The risk for colorectal carcinoma in PSC is discussed in the section on IBD. Factors responsible for the malignant development remain unknown, and it is great challenge for future research to identify these risk factors in order to intervene with the development of this dreadful complication.
842
TREATMENT The treatment of PSC has been limited by lack of knowledge about the pathogenic process. Treatment can be divided into specific therapy for the underlying disease process and treatment of symptoms and complications (for treatments of complications see section on complications). Moreover, treatment can be separated into medical, endoscopic and surgical, including liver transplantation. There are few prospective randomized controlled treatment studies in PSC. In many studies patients with advanced disease are included. As the major target in PSC is the bile ducts, and as they lack the capacity to regenerate, the disease process will lead to the development of ductopenia and a progressive irreversible failure of hepatic biliary excretion. Therefore, specific treatment would be more effective early in the disease course. There are certain problems that have to be focused on when evaluating the efficacy of a treatment in patents with PSC. • PSC is a rare disease, leading to problems in recruiting sufficient numbers of patients to achieve statistical significance. In order to reach statistical power to assess significant differences on hard endpoints, such as survival or need for liver transplantation, almost 350 patients have to be treated for 5 years. • The disease progression is usually slow, leading to a need for long duration of therapy studies, and long-term placebo controlled studies will lead to huge costs. • PSC has a fluctuating course and symptoms and biochemical abnormalities may change due to the natural course of the disease. There are no good surrogate markers for outcome in PSC. • The histological assessment of disease progression is hampered by the high rate of sampling variability in liver biopsies. • A high number of PSC patients develop cholangiocarcinoma and die from this complication without having end-stage disease. The pathophysiological mechanisms underlying the carcinogenic process in PSC are probably different from factors causing the non-malignant biliary destructive process, and other therapies affecting the neoplastic development are needed. Studies evaluating the efficacy of treatment in PSC should therefore not only be prospective randomized and controlled, but should also include a large group of patients, preferably in an early stage of disease, who should be followed for many years. At present effective medical treatment for PSC is lacking. Several medical treatments with different mechanisms of action, including immunosuppressants, anti-inflammatories, antifibrotics and bile acids have been investigated in PSC. None have shown convincing evidence of benefit. UDCA is a hydrophilic bile acid having multiple effects such as choleretic, hepatoprotective and immunmodulatory properties. As UDCA has proved to be effective in primary biliary cirrhosis, it has come to be the most tested drug in this context (Table 42-6). The largest study, including 105 patients with a mean follow-up period of 2.2 years, did not demonstrate any clinical benefit and did not alter disease progression or time to transplantation. The authors suggested that this lack of effect could be explained by the inclusion of patients with advanced histological stage, short duration of follow-up and inadequate dosage of UDCA. In a Cochrane Systematic Review of bile acid treatment for PSC to
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
Table 42-6. Summary of Studies of UDCA in PSC Investigator Hayashi90 Chazouillèrs90 O’Brien 91 Beuers92 Stiehl94
DeMaria96 Lindor97 Van Hoog straten98 Mitchell01 Harnois01 Olsson04 Okolicsanyi03 Schramm99a Färkkilä01b Sterling04c Stiehl97d
n
RCT
Duration (yrs)
Dose (mg/day)
Lab tests
Histology
ERC
Survival
1 15 12 14 27 20 12 40 102 48 26 30 198 86 15 80 25 65
– – – + – + – + + – + – + – – + + –
2 0.5 1.5 1 1 0.25 1 2 2.2 2 2 1 5 ≈4 4.5 3 2 ≈4
600 750–1250 10/kg 13–15/kg 750 750 750 600
+ + + + + + + – + + + + – + – +
– NE NE + NE NE + NE – – + NE NE + + NE –
– NE NE NE NE NE e – NE – NE NE NE NE
NE NE NE – NE NE NE – – – NE e – NE e – – e
13–15/kg 20/kg 25–30/kg 20/kg 8–13/kg 500–750 15/kg 13–15/kg 750
+ –
a
Combined with prednisolon and azathioprine. Comparison with UDCA + metronidazole. Combined with mycophenolate. d Combined with endoscopic dilatations. e Compared with expected survival according to Mayo score. RCT, randomized controlled study; +, improvement; –, no effect; NE, not evaluated. b c
May 2002, six randomized controlled studies were identified, including 223 patients (113 placebo) treated for a median of 2 years.126 It was concluded that none of the trials was of high methodological quality; UDCA did have a marked effect on liver biochemistry, such as bilirubin concentrations, serum ALP and serum AST/ALT levels, but no effect on serum albumin concentration. UDCA did not reduce the risk of death, need for liver transplantation, the development of ascites or encephalopathy. It was concluded that UDCA was safe and free from serious adverse events in PSC patients. Biliary enrichment of UDCA increases with increasing dose and reaches a plateau at 22–25 mg/kg without an increase in toxic hydrophobic bile acids.127 Furthermore, the in vitro immunomodulatory effects of UDCA are enhanced with increasing concentrations. It is therefore interesting to treat PSC patients with higher doses of UDCA. Two published studies have evaluated the use of high-dose UDCA for patients with PSC.128,129 Mitchell et al.128 randomized 26 patients to UDCA 20 mg/kg or placebo for 2 years. The UDCAtreated patients had significantly less progression in disease stage and an improvement in cholangiographic appearance. Harnois et al.129 treated 30 patients with 25–30 mg UDCA/kg with 1 year’s followup. The prognoses for the patients were calculated using the Mayo prognostic model and the expected mortality at 4 years was considerably improved in patients given high-dose UDCA than in those who received placebo in another study performed by the same group. In a Nordic study, 198 PSC patients were included in a randomized double-blind placebo-controlled study for 5 years’ treatment with UDCA 20 mg/kg.130 Despite being the largest controlled prospective study in PSC ever performed, the study could not demonstrate a beneficial effect of UDCA on survival or need for liver transplantation. It is, however, of great interest that two studies have shown that
treatment with UDCA clearly decreases the risk of developing colorectal dysplasia in patients with PSC and ulcerative colitis,70,71 suggesting that UDCA should be used in PSC and UC as a prophylactic cancer preventive. Controlled studies with methotrexate, colchicine, penicillamin and ciclosporin have not demonstrated any favorable effect. In a small 1-year study including 10 patients, tacrolimus (FK506) was used, showing an improvement in liver biochemistry. Corticosteroids have only been examined in small studies without showing convincing results, although some favorable effects have been reported. Boberg et al.131 showed that 3.7% of patients with PSC will have a complete or partial response on steroid treatment, and some more patients may show some reduction of transaminase or bilirubin level. Patients who responded to steroids had a better long-term survival than non-responders. However, glucosteroids may accelerate the onset and progression of osteoporosis. Budesonide, a corticosteroid with extensive first-pass hepatic metabolism, has been suggested to have fewer side effects. However, in a 1-year open study it was found that oral budesonide did not show any clinical, biochemical or histological benefit: instead, it was associated with significant worsening of osteoporosis in patients with PSC.132 There is some evidence to suggest that smoking may be protective of PSC. In two small trials the efficacy of nicotine patches was studied and showed no effect on clinical or biochemical tests.133,134 Pentoxifylline prevents the production of tumor necrosis factor, which has been implicated in the pathogenesis of PSC. In a pilot study pentoxifylline was given to 20 PSC patients showing no improvement in liver function tests or alleviation of symptoms.135 Etanercept is also an inhibitor of tumor necrosis factor and was given to 10 PSC patients.136 In two patients pruritus disappeared during treatment, but no effect was seen on cholangiographic appearance, and it is noteworthy that mean biliru-
843
Section VI. Immune Diseases
bin increased significantly. Cladribine, an antilymphocytic agent, was used in four patients with PSC and found to decrease periportal inflammation but not liver function tests, cholangiography or biochemical tests.137 UDCA has also been tested in combination with other drugs, such as prednisolone, azathioprine and methotrexate.138,139 In the study by Schramm et al.138 evaluating combination treatment with UDCA, prednisolone and azathioprine, six patients who had follow-up liver biopsies showed histological improvement. However, changes in liver histology in such a small group of patients should be interpreted cautiously. Worsening of ERC was seen in only one of 10 patients. Sterling et al. could not show that mycophenolate mofetil combined with UDCA provided any additional benefit compared to UDCA alone.140 The value of combining UDCA with bile duct stricture dilation was evaluated in a non-randomized study, indicating that the combination was more effective than the predicted survival calculated from their baseline status. Maybe, the most effective future therapy in PSC will be to combine different treatment modalities. However, future studies will not only have to assess the efficacy of the treatments, but as all therapies will most likely be given for many years, an acceptable benefit/risk ratio must be proved.
ENDOSCOPIC AND SURGICAL TREATMENT A dominant stricture in PSC is found in approximately 10–40% of patients and is defined as a tight stricture impeding normal bile flow in either the common bile duct or the common hepatic duct, or the left or right main hepatic ducts.141 Dominant bile duct strictures may cause rapid and clinical detoriation in some patients with PSC, and can be treated endoscopically or surgically. Biliary tract reconstructive procedures have been used to bypass dominant strictures not amenable to endoscopic or percutaneous drainage. Nonrandomized studies suggest that bilioenteric bypass may reduce symptoms in non-cirrhotic PSC patients and prolong survival. However, biliary manipulations increase the risk of stricturing and bacterial cholangitis. Moreover, previous abdominal operations may also have a negative influence on the outcome after liver transplantation.122 Thus, extrahepatic biliary resection is not a therapeutic alternative in general, but might have a role in highly selected patients. An endoscopic approach is therefore the treatment of choice in compensated patients with symptomatic dominant strictures (Figure 42.10). The role of endoscopic therapy in PSC is still not properly evaluated, however. It is indicated for selected patients with dominant strictures that can be reached with the guide wire, usually within 2 cm of the bifurcation. If changes are mainly intrahepatic, treatment is less effective. Endoscopic treatment in PSC involves balloon dilatation of the strictures and/or placement of a biliary stent, and in some centres nasobiliary drainage with or without lavage is also used. Many of these reports also include sphincterotomy to facilitate further interventions. Complications occur in 10–15% of cases. Among 123 patients with PSC who underwent endoscopic treatment a total of 30 ERC-related complications were described: cholangitis (30), pancreatitis (8), postsphincterotomy bleed (1) and aminoglycoside-induced tubular necrosis (1).53 Endoscopic interventions should only be performed in symptomatic patients with dominant strictures. Patients most likely to benefit from endoscopic therapy are those with a progressive increase in liver enzymes, jaundice or cholangitis.142 Indeed,
844
findings in several uncontrolled studies suggest that these patients may have a prolonged biochemical and clinical response for several years after endoscopic therapy. However, end-stage patients in need of a liver transplant should preferably not undergo endoscopic interventions, as detoriation of liver function has been described after ERC.143 Placement of a stent is associated with recurrent occlusion and bacterial cholangitis, and therefore the stents must be changed regularly. Cholangitis can be a serious complication in PSC because of the presence of poorly drained ducts. Giving antibiotics before endoscopic interventions can reduce the risk of cholangitis. Because of the risk of cholangitis, it has been suggested that stents should be avoided in PSC and only balloon dilatation should be undertaken.143 In a study by Ponsioen et al.141 32 patients were treated with short-term stenting (mean 11 days). In 83% of the patients the cholestatic complaints improved, and during follow-up only 20% needed new interventions during the first year. Baluyat et al. evaluated survival in 63 consecutive patients with PSC who underwent endoscopic therapy and were followed for a medium of 34 months.144 The survival in the endoscopically treated groups was compared to that of the predicted Mayo risk score and found to be significantly higher (83% vs 65%). It is, however, noteworthy that the bilirubin value was measured just before endoscopic treatment, which may overestimate the severity of the underlying liver disease. Stiehl et al.124 prospectively studied 52 patients treated with UDCA and repeated balloon dilatations (five were also stented). Five years after the first dilatation the Kaplan–Meier survival rates free of liver transplantation were 100% in stage 2, 72% in stage 3 and 56% in stage 4 disease. Actuarial survival free of transplantation was significantly better than with the Mayo Model risk score. It is important to point out that only a skilled endoscopist should perform endoscopic therapy in patients with PSC. Several cases of bile duct perforation during ERC have been described in such patients, which may be due to a weakened bile duct wall or to technical reasons.141 Although it seems as if endoscopic therapy in selected PSC patients is of clinical importance, optimal regimens have not been defined and it must be emphasized that endoscopic interventions should never delay referral for liver transplantation.
LIVER TRANSPLANTATION As there is no current treatment that will effectively halt the progression of PSC, liver transplantation remains the only option for patients who progress to end-stage disease. In the US PSC is the fifth most common indication for transplantation, accounting for approximately 10% of liver transplants. In the Nordic countries PSC is the most common indication. If the patient does not have cholangiocarcinoma the results of liver transplantation in PSC are excellent, and several recent reports have reported 1-year survival of 90–97% and 5-year rates of 80–85%.107,145–147 The variable course of PSC and the unpredictable risk of cholangiocarcinoma make the decision for transplantation extremely difficult. It is important to find tools to predict the survival for the individual patient in order to find the optimal time for transplantation. Several studies have developed sophisticated prognostic models based on independent clinical and histologic variables. Despite this, the timing of liver transplantation in PSC remains difficult, and none of the prognos-
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
tic models is able to predict the appearance of cholangiocarcinoma. It is also important to know that in all the prognostic models the confidence intervals are wide, and so extrapolation to an individual patient must be cautious. The indication for liver transplantation in PSC is similar to those for patients with other chronic liver diseases with signs of decompensated end-stage disease. For patients without end-stage disease who suffer from severe recurrent cholangitis, intractable pruritus or progressive osteopenia, liver transplantation can also be considered. The ideal time for listing for liver transplantation is also dependent on the waiting time.148 In a Nordic study 253 PSC patients were accepted on the waiting list between 1990 and 2000;107 87% received a first liver allograft and 13% died without transplantation. It was shown that a high bilirubin and a high MELD score were associated with an increased risk of dying or being withdrawn from the waiting list, indicating that for these patients referral was too late. An English study149 studied pretransplant variables associated with survival after liver transplantation. It was shown that the presence of inflammatory bowel disease, previous upper abdominal surgery, ascites, increased creatinine and the presence of malignancy all had a negative impact on survival after liver transplantation. Data from the Mayo Clinic have demonstrated that post-transplant survival rate is clearly related to pretransplant Child–Pugh stage. One-, 2- and 5-year survival rates were respectively 98.1, 97 and 91% for Child–Pugh A, 89.1, 81 and 55% in Child–Pugh B, and 73, 53 and 16% in Child–Pugh C. An increased resource utilization was also associated with higher Child–Pugh score. It seems therefore advisable not to wait too long to refer PSC patients for evaluation for transplantation. With the new allocation system in the USA, PSC patients with intractable pruritus, recurrent bleeding from peristomal varices, bile duct dysplasia on biliary cytology, and recurrent bouts of cholangitis will not have a high enough MELD score to be accepted for transplantation. These patients will need special consideration by the regional board in order to have a good outcome. Ten to 20% of all PSC patients develop cholangiocarcinoma; for these patients liver transplantation is not an effective treatment because of the high probability of tumor recurrence, and therefore cholangiocarcinoma is usually a contraindication for transplantation. However, an incidental finding of cholangiocarcinoma less than 1 cm in the explanted liver may have no impact on outcome after transplantation. Brandsaeter et al.122 described the outcome of 31 patients with PSC and cancer who were transplanted. The 1- and 5-year survival was 65 and 35%, respectively. In a study from the Mayo Clinic it was demonstrated that a small subgroup of patients with cholangiocarcinoma, meeting strict selection criteria, who undergo radiation and chemotherapy prior to liver transplantation appear to have acceptable long-term results.123 There is accumulating evidence that PSC recurs after liver transplantation. The diagnosis of post-transplant PSC has not been widely accepted, largely because of problems with non-anastomotic biliary strictures that occur after liver transplantation in several other conditions. Therefore, other causes of biliary strictures after liver transplantation must be excluded, such as hepatic artery thrombosis, established ductopenic disease, anastomotic strictures alone, non-anastomotic strictures before post-transplant day 90, and ABO incompatibility between donor and recipient. With the use of strict criteria Graziadei150 reported that 24 of 120 PSC patients (20%)
Table 42-7. Criteria for Recurrence of PSC after Liver Transplantation150 Diagnosis Confirmed diagnosis of PSC prior to liver transplantation and Cholangiography showing intrahepatic and/or extrahepatic biliary stricturing, beading and irregularity > 90 days post transplant or Histology with fibrous cholangitis and/or fibro-obliterative lesions with or without ductopenia, biliary fibrosis, or biliary cirrhosis Exclusion critera Hepatic artery thrombosis/stenosis Established ductopenic rejection Anastomotic strictures alone Non-anastomotic strictures before post-transplant day 90 ABO incompatibility between donor and recipient
Table 42-8. Prevalence of PSC Recurrence and Predictors of Recurrence in Studies Using Cholangiography and Histology as a Basis for Diagnosis of Recurrence Reference Jeyarajah Goss Kubota161 Graziadei147 Yusoff 162 Vera153 Kuglemas163 Khettry164
Number of patients (% recurrence) 100 (16) 127 (9) 53 (6) 120 (20) 12 (17) 152 (37) 71 (21) 40 (14)
Predictors of recurrence CMV infection
None found Intact colon, male sex OKT-3 Recipient donor gender mismatch
had recurrent PSC after liver transplantation (Table 42-7). The median time for the development of non-anastomotic biliary strictures was 360 days. The survival was no different in the group with recurrent disease compared to those without. However, according to more recent reports it seems as if recurrence of PSC might affect outcome in the long term.151,152 In a Birmingham study PSC recurred in 37% of 132 patients.153 Multivariate analysis showed that being male (RR 1.2) and having an intact colon before transplantation (RR 8.7) were associated with recurrence. Taken together, it seems as if the recurrence rate of PSC varies from 6 to 37%, depending on the definition of recurrent disease. Predictive factors for PSC recurrence have also been investigated in studies showing somewhat conflicting results (Table 42-8). PSC patients have an increased incidence of both acute and chronic rejection and hepatic artery thrombosis.148,154 A study from Birmingham addressed the question of whether the presence of IBD affects the outcome after liver transplantation in patients with PSC.154a The study included 55 PSC patients, 31 of whom had UC. After liver transplantation there were significantly more episodes of acute rejection in the patients with IBD compared to those without IBD. Narumi et al. found that those PSC patients having IBD had a greater risk for severe rejection and a greater need for retransplantation.154b Van de Vire et al.72 did not find that the presence of IBD had an impact on outcome after liver transplantation. Welsh et al.155 investigated biliary complications and graft and patient survival in all patients who underwent liver transplantation
845
Section VI. Immune Diseases
in the United Kingdom between 1994 and 2003. A comparison was made between patients receiving a Roux-en-Y anastomosis (n = 264) and those with a duct-to-duct anastomosis (n = 98). A significantly higher stricture rate and a worse graft and patient survival were found in patients with duct-to-duct anastomosis, clearly showing that Roux-en-Y choledochojejunostomy is the method of choice for biliary reconstruction in patients with PSC undergoing liver transplantation. Vascular problems, chronic rejections and probably also ascending cholangitis might account for some of the biliary problems seen in these patients after transplantation. The United Network for Organ Sharing (UNOS) database between 1987 and 2001 includes 3309 PSC patients and 3254 patients with primary biliary cirrhosis.152 The outcome for these two patient groups was compared. After adjusting for other risk factors PSC patients had a significantly lower graft and patients survival than did PBC patients. Late retransplantion rate was significantly higher in the PSC group. Although the PSC patients were younger they had a lower survival rate, which became apparent 7 years after liver transplantation. Solano et al.,151 in a retrospective Canadian study including 132 PSC patients, showed that the overall patient and graft survival was similar to that in non-PSC patients, although the survival curves begin to diverge after 5 years. Retransplantation was required in 19 patients. Recurrence of PSC and chronic rejection were the major determinants of graft loss. The less favorable long-term outcome in these patients speaks against early referral for liver transplantation in order to avoid the risk of cholangiocarcinoma. Because PSC is closely associated with IBD it is important never to forget the colon after liver transplantation. The colitis usually runs a quiescent course before transplantion. In several studies, however, it has been shown that 30–50% of patients have exacerbation of IBD symptoms after liver transplantation. This is noteworthy, as the immunosuppressive agents commonly used after liver transplantation have been found to be of clinical benefit in patients with active colitis. However, some studies have not confirmed the more agressive course of the colitis after liver transplantation. Whether this more aggressive course can be explained by more rapid steroid withdrawal at some centers remains to be evaluated in prospective studies. Haagsma et al.156 found that tacrolimus is more frequently associated with post-transplant exacerbation of IBD. It is also important to remember that patients with UC remain at increased risk for developing colorectal dysplasia/cancer. In a report from the Mayo Clinic including 57 patients with an intact colon transplanted for PSC, the risk of colorectal cancer was 1% per person per year. The cumulative incidence of dysplasia was 15% at 5 years and 21% at 8 years. The majority of cancers were diagnosed within 30 months after transplant. Regardless of the duration of colitis, PSC patients with ulcerative colitis should undergo colonoscopy on a yearly basis after liver transplantation.
PROGNOSIS – NATURAL HISTORY OF PSC As described previously, the clinical course of patients with PSC is variable, and PSC is frequently a progressive disease leading to complications due to chronic cholestasis, cirrhosis, hepatic failure,
846
portal hypertension or cholangiocarcinoma. Some patients have a long-standing silent disease without progression, lasting for decades, whereas others have a rapid progression of symptoms and signs ending with manifest liver failure and a need for transplantation within a few years. Many patients have exacerbations and remissions. Both histologic and cholangiographic progression of the hepatic changes may occur without the development of new symptoms or signs. Therefore, the natural history has been difficult to describe, with conflicting reports of the prognosis appearing in the literature. The natural history of PSC is described in a number of studies: the median survival time from diagnosis to death or transplantation is reported to be around 12 years, but may be as high as 21 years. Differences in survival time between the various studies may have many reasons, one being variations in the definition of disease onset. It is difficult to determine the onset of PSC accurately because it is usually insidious. Early detection can only be achieved with prospective regular follow-ups of liver function tests in patients with IBD. Whether the onset of the disease is defined as the time of the first symptom consistent with PSC, the time of the first abnormal liver function test, the time of first cholangiography, or the time of referral to a tertial center, there will be differences in the overall survival time. Furthermore, whether termination of the disease is defined as the time of death or time for liver transplantation will also influence survival time. One of the largest studies, including 305 PSC patients, showed that 44% were asymptomatic at diagnosis; 22% of the asymptomatic patients developed symptoms during a median of 5 years’ follow-up.156a The study also shows that patients with asymptomatic PSC at diagnosis have a more favorable prognosis than patients with symptoms. A comparison of survival in symptomatic and asymptomatic patients is shown in Figure 42-11. Asymptomatic patients with PSC are diagnosed with increasing frequency. Porayko et al. analyzed 45 asymptomatic PSC patients for evaluation of progression and natural history of asymptomatic PSC disease.156b The majority of these patients had minimal histological changes (stage I–II) at liver biopsy, none with cirrhosis. During a mean follow-up of 6 years, 76% showed progression of the liver disease, i.e. development of signs or symptoms, and 31% of the asymptomatic patients died or were transplanted because of liver failure. The histological progression is also variable and the evaluation is difficult owing to a high degree of sampling variability. A study of histological progression over time was made in 107 PSC patients showing that stage II PSC progressed in 66% at 2 years and in 93% at 5 years. Stage III progressed in 25% at 2 years and in 52% at 5 years.157 Even though asymptomatic patients seems to have a more favorable prognosis, and there are reports of patients with long-term survival, PSC must be regarded as a progressive disease frequently leading to complications of cholestasis, cirrhosis, liver failure and cholangiocarcinoma. A study of the spontaneous clinical course of PSC has been performed.91 Sixty-five patients with PSC kept symptom diaries and recorded the symptoms of itch, fever and pain daily for a 3-year period. Symptoms occurred in 84% of the patients, were usually intermittent, and itch only or pain only were the most frequently reported episodes. No correlation was found between serological markers, histological features and symptoms, except between itch and ALP levels. Although there was considerable variation
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
A
B
Figure 42-10. Dilation procedure of a dominant stricture in a patient with PSC (A) before dilation, (B) and (C) balloon dilation (D) after dilation.
847
Section VI. Immune Diseases
D
C
Figure 42-10, cont’d.
Table 42-9. Factors Found to be of Prognostic Importance in PSC Dickson et al. (n = 426) Age Bilirubin Histological stage Splenomegaly Variceal bleeding Albumin
848
Kim et al. (n = 405)97 Age Bilirubin Albumin AST
Broome et al. (n = 305) Age Bilirubin Histological stage
Shetty et al. (n = 208) Age Child–Pugh score Bilirubin
Boberg et al. (n = 330)1 Age at diagnosis of PSC Bilirubin Albumin
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
CONCLUSION
1
Cumulated survival
0.8 0.6 0.4 0.2 Symptomatic at diagnosis Asymptomatic at diagnosis 0 0
24
48
72
96 120 144 Time (months)
168
192
216
Figure 42-11. Kaplan–Meier estimated survival curves of symptomatic and asymptomatic PSC patients (p < 0.001). (Gut 1996;38:610–615.)156a
in symptom load between patients, it was concluded that most symptom periods in PSC are monosymptomatic and brief. Factors found to be of prognostic importance in PSC are presented in Table 42-9. A time-dependent model for prediction of the prognosis of PSC patients has recently been presented: 330 PSC patients from five centers were included, and log bilirubin, albumin and age at diagnosis were found to be independent prognostic variables. Time-fixed and time-dependent models were compared. The time-dependent model was shown to be superior to the time-fixed variant in assigning low 1-year survival probabilities to patients who actually survived less than one year.1 None of the prognostic models is able to predict cholangiocarcinoma development. Thus, the chances of predicting cholangiocarcinoma in the setting of PSC are very small, and it is important to stress that so far no factor has been singled out to identify patients who will later develop cholangiocarcinoma. Once the cholangiocarcinoma has developed, the prognosis is poor, as currently no curative treatment is available. This is discussed further in the section on hepatobiliary malignancies. The prognostic value of cholangiography has also been investigated. In a study by Olsson et al. high-grade intrahepatic strictures on the first diagnostic cholangiogram indicate early jaundice, whereas short and high-grade extrahepatic strictures indicate early pruritus, abdominal pain, and fever.157a Ponsioen158 evaluated cholangiograms in 174 patients with PSC, and using a radiologic classification system for the severity of sclerosis it was shown that this actual scoring was inversely correlated with survival. As PSC is a disease of young people the question sometimes arises concerning the risk of pregnancy in these patients. This matter was addressed in one study of 13 pregnancies in 10 patients with PSC. The pregnancy did not seem to have a negative effect on the progression of the disease, and the outcome for both babies and mothers was favorable. However, in two pregnant patients the pruritus was so intense as to bring on premature delivery. In a more recent study one mother with PSC developed a dominant stricture during pregnancy that required dilatation shortly after delivery. The outcome for the baby was good.159
Although numerous papers have been published during the last two decades about PSC, many questions remain unanswered. The cause of PSC is still unknown, although there are many data supporting that PSC is not a classic autoimmune disease. Immune mechanisms are, however, assumed to be of major importance for the disease process. The disease may be heterogeneous not only in clinical presentation, but also from a pathogenetic perspective. Medical therapy to halt disease progression and prevent cirrhosis is lacking. Therapy should therefore focus on the treatment of complications and symptoms in order to improve the patient’s quality of life. The only curative treatment remains liver transplantation, which has excellent survival rates, and PSC has become one of the most important indications. Data have, however, emerged that the long-term results after liver transplantation are worse for PSC than for primary biliary cirrhosis. This is partly explained by the recurrence of PSC after liver transplantation, which seems to have an impact on long-term outcome. It is still unclear why PSC is a premalignant disease with an increased risk for cancer development at all sites where a chronic inflammation is present, such as the biliary tree, the gallbladder, the colon and the pancreas. Treatment with UDCA decreases the risk of developing colorectal dysplasia/cancer in PSC patients with UC, which is certainly a glimmer of hope for these patients. The role of UDCA in the prevention of liver cancer needs to be evaluated in the future. The major message today is that the medical fraternity must gain forces in the battle to unravel knowledge about PSC. Only when large cohorts of patients are gathered can we hope to find answers for all the questions concerning this disease.
REFERENCES 1. Boberg KM, Rocca G, Egeland T, et al. Time-dependent Cox regression model is superior in prediction of prognosis in primary sclerosing cholangitis. Hepatology 2002;35:652–657. 2. Parlak E, Kosar Y, Ulker A, et al. Primary sclerosing cholangitis in patients with inflammatory bowel disease in Turkey. J Clin Gastroenterol 2001;33:299–301. 3. Boberg KM, Aadland E, Jahnsen J, et al. Incidence and prevalence of primary biliary cirrhosis, primary sclerosing cholangitis, and autoimmune hepatitis in a Norwegian population. Scand J Gastroenterol 1998;33:99–103. 4. Bambha K, Kim WR, Talwalkar J, et al. Incidence, clinical spectrum, and outcomes of primary sclerosing cholangitis in a United States community. Gastroenterology 2004;125: 1364–1369. 5. Hurlburt KJ, McMahon BJ, Deubner H, et al. Prevalence of autoimmune liver disease in Alaska Natives. Am J Gastroenterol 2002;97:2402–2407. 6. Vierling JM. Animal models for primary sclerosing cholangitis. Best Pract Res Clin Gastroenterol 2001;15:591–610. 7. Orth T, Neurath M, Schirmacher P, et al. A novel rat model of chronic fibrosing cholangitis induced by local administration of a hapten reagent into the dilated bile duct is associated with increased TNF-alpha production and autoantibodies. J Hepatol 2000;33:862–872. 8. Fickert P, Fuchsbichler A, Wagner M, et al. Regurgitation of bile acids from leaky sclerosing cholangitis in Mdr2 (Abcb4)
849
Section VI. Immune Diseases
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
850
bile ducts causes knockout mice. Gastroenterology 2004;127:261–274. Sakisaka S, Kawaguchi T, Taniguchi E, et al. Alterations in tight junctions differ between primary biliary cirrhosis and primary sclerosing cholangitis. Hepatology 2001;33:1460–1468. Nilsson HO, Taneera J, Castedal M, et al. Identification of Helicobacter pylori and other Helicobacter species by PCR, hybridization, and partial DNA sequencing in human liver samples from patients with primary sclerosing cholangitis or primary biliary cirrhosis. J Clin Microbiol 2000;38:1072–1076. Nilsson I, Lindgren S, Eriksson S, Wadstrom T. Serum antibodies to Helicobacter hepaticus and Helicobacter pylori in patients with chronic liver disease. Gut 2000;46:410–414. Fallone CA, Tran S, Semret M, et al. Helicobacter DNA in bile: correlation with hepatobiliary diseases. Aliment Pharmacol Ther 2003;17:453–458. Ponsioen CY, Defoer J, Ten Kate FJ, et al. A survey of infectious agents as risk factors for primary sclerosing cholangitis: are Chlamydia species involved? Eur J Gastroenterol Hepatol 2002;14:641–648. Terjung B, Spengler U, Sauerbruch T, Worman HJ. ‘Atypical pANCA’ in IBD and hepatobiliary disorders react with a 50kilodalton nuclear envelope protein of neutrophils and myeloid cell lines. Gastroenterology 2000;119:310–322. Roozendaal C, Horst G, Pogany K, et al. Prevalence and clinical significance of anti-lactoferrin autoantibodies in inflammatory bowel diseases and primary sclerosing cholangitis. Adv Exp Med Biol 1998;443:313–319. Angulo P, Peter JB, Gershwin ME, et al. Serum autoantibodies in patients with primary sclerosing cholangitis. J Hepatol 2000;32:182–187. Xu B, Broome U, Ericzon BG, Sumitran-Holgersson S. High frequency of autoantibodies in patients with primary sclerosing cholangitis that bind biliary epithelial cells and induce expression of CD44 and production of interleukin 6. Gut 2002;51:120–127. Xu B, Broome U, Remberger M, Sumitran-Holgersson S. Tumor necrosis factor alpha impairs function of liver derived T lymphocytes and natural killer cells in patients with primary sclerosing cholangitis. Gut 2001;49:131–141. Tinmouth J, Lee M, Wanless IR, et al. Apoptosis of biliary epithelial cells in primary biliary cirrhosis and primary sclerosing cholangitis. Liver 2002;22:228–234. Grant AJ, Lalor PF, Salmi M, et al. Homing of mucosal lymphocytes to the liver in the pathogenesis of hepatic complications of inflammatory bowel disease. Lancet 2002;359:150–157. Grant AJ, Lalor PF, Hubscher SG, et al. MAdCAM-1 expressed in chronic inflammatory liver disease supports mucosal lymphocyte adhesion to hepatic endothelium (MAdCAM-1 in chronic inflammatory liver disease). Hepatology 2001;33:1065–1072. Adams DH, Afford SC. The role of cholangiocytes in the development of chronic inflammatory liver disease. Front Biosci 2002;7:276–285. Cruickshank SM, Southgate J, Wyatt JI, et al. Expression of CD44 on bile ducts in primary sclerosing cholangitis and primary biliary cirrhosis. J Clin Pathol 1999;52:730–734. Spirli C, Fabris L, Duner E, et al. Cytokine-stimulated nitric oxide production inhibits adenylyl cyclase and cAMPdependent secretion in cholangiocytes. Gastroenterology 2003; 124:737–753. Jaiswal M, LaRusso NF, Shapiro RA, et al. Nitric oxidemediated inhibition of DNA repair potentiates oxidative DNA damage in cholangiocytes. Gastroenterology 2001; 120:190–199. Jaiswal M, LaRusso NF, Burgart LJ, Gores GJ. Inflammatory cytokines induce DNA damage and inhibit DNA repair in
27.
28. 29. 30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42. 43.
44.
45.
46.
cholangiocarcinoma cells by a nitric oxide-dependent mechanism. Cancer Res 2000;60:184–190. Bergquist A, Lindberg G, Saarinen S, Broome U. Increased prevalence of primary sclerosing cholangitis among first degree relatives. J Hepatology 2005;42:252–256. Donaldson PT, Norris S. Immunogenetics in PSC. Best Pract Res Clin Gastroenterol 2001;15:611–627. Donaldson PT. Genetics of autoimmune and viral liver diseases; understanding the issues. J Hepatol 2004;41:327–332. Moloney MM, Thomson LJ, Strettell MJ, et al. Human leukocyte antigen-C genes and susceptibility to primary sclerosing cholangitis. Hepatology 1998;28:660–662. Spurkland A, Saarinen S, Boberg KM, et al. HLA class II haplotypes in primary sclerosing cholangitis patients from five European populations. Tissue Antigens 1999;53:459–469. Neri TM, Cavestro GM, Seghini P, et al. Novel association of HLA-haplotypes with primary sclerosing cholangitis (PSC) in a southern European population. Dig Liver Dis 2003;35:571–576. Donaldson PT, Norris S. Evaluation of the role of MHC class II alleles, haplotypes and selected amino acid sequences in primary sclerosing cholangitis. Autoimmunity 2002;35:555–564. Norris S, Kondeatis E, Collins R, et al. Mapping MHCencoded susceptibility and resistance in primary sclerosing cholangitis: the role of MICA polymorphism. Gastroenterology 2001;120:1475–1482. Wiencke K, Spurkland A, Schrumpf E, Boberg KM. Primary sclerosing cholangitis is associated to an extended B8-DR3 haplotype including particular MICA and MICB alleles. Hepatology 2001;34:625–630. Wiencke K, Louka AS, Spurkland A, et al. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients. J Hepatology 2004;41:209–214. Satsangi J, Chapman RW, Haldar N, et al. A functional polymorphism of the stromelysin gene (MMP-3) influences susceptibility to primary sclerosing cholangitis. Gastroenterology 2001;121:124–130. Bernal W, Moloney M, Underhill J, Donaldson PT. Association of tumor necrosis factor polymorphism with primary sclerosing cholangitis. J Hepatol 1999;30:237–241. Donaldson PT, Norris S, Constantini PK, et al. The interleukin1 and interleukin-10 gene polymorphisms in primary sclerosing cholangitis: no associations with disease susceptibility/resistance. J Hepatol 2000;32:882–886. Mitchell SA, Grove J, Spurkland A, et al. Association of the tumor necrosis factor alpha -308 but not the interleukin 10 627 promoter polymorphism with genetic susceptibility to primary sclerosing cholangitis. Gut 2001;49:288–294. Yang XS, Cullen SN, Li JH, et al. Susceptibility to primary sclerosing cholangitis is associated with polymorphisms of intercellular adhesion molecule-1. J Hepatol 2004;40:375–379. Dwinell MB, Eckmann L, Leopard JD, et al. Chemokine receptor expression by human intestinal epithelial cells. Gastroenterology 1999;117:359–367. Eri R, Jonsson JR, Pandeya N, et al. CCR5-D32 mutation is strongly associated with primary sclerosing cholangitis. Genes Immun 2004;5:444–450. Sheth S, Shea JC, Bishop MD, et al. Increased prevalence of CFTR mutations and variants and decreased chloride secretion in primary sclerosing cholangitis. Hum Genet 2003;113:286–292. Angulo P, Maor-Kendler Y, Lindor KD. Small-duct primary sclerosing cholangitis: a long-term follow-up study. Hepatology 2002;35:1494–1500. Bjornsson E, Boberg KM, Cullen S, Fleming K, Clausen OP, Fausa O, et al. Patients with small duct primary sclerosing
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
47a. 47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
cholangitis have a favorable long term prognosis. Gut 2002;51:731–735. Atkinson A. Carroll W. Sclerosing cholangitis associated with regional enteritis. JAMA 1964;188:183–186. Broome U, Glaumann H, Lindstom E, et al. Natural history and outcome in 32 Swedish patients with small duct primary sclerosing cholangitis (PSC). J Hepatol 2002;36:586–589. Vitellas KM, Enns RA, Keogan MT, et al. Comparison of MR cholangiopancreatographic techniques with contrast-enhanced cholangiography in the evaluation of sclerosing cholangitis. Am J Roentgenol 2002;178:327–334. Textor HJ, Flacke S, Pauleit D, et al. Three-dimensional magnetic resonance cholangiopancreatography with respiratory triggering in the diagnosis of primary sclerosing cholangitis: comparison with endoscopic retrograde cholangiography. Endoscopy 2002;34:984–990. Silva AC, Friese JL, Hara AK, Liu PT. MR cholangiopancreatography: Improved ductal distention with intravenous morphine administration. Radiographics 2004;24:677–687. Angulo P, Pearce DH, Johnson CD, et al. Magnetic resonance cholangiography in patients with biliary disease: its role in primary sclerosing cholangitis. J Hepatol 2000;33:520–527. Fulcher AS, Turner MA, Franklin KJ, et al. Primary sclerosing cholangitis: evaluation with MR cholangiography–a case–control study. Radiology 2000;215:71–80. van den Hazel SJ, Wolfhagen EH, van Buuren HR, et al. Prospective risk assessment of endoscopic retrograde cholangiography in patients with primary sclerosing cholangitis. Endoscopy 2000;32:779–782. Bjornsson ES, Kilander AF, Olsson RG. Bile duct bacterial isolates in primary sclerosing cholangitis and certain other forms of cholestasis – a study of bile cultures from ERCP. Hepato-Gastroenterology 2000;47:1504–1508. Burak KW, Angulo P, Lindor KD. Is there a role for liver biopsy in primary sclerosing cholangitis? Am J Gastroenterol 2003;98:1155–1158. Fleming KA, Boberg KM, Glaumann H, et al. Biliary dysplasia as a marker of cholangiocarcinoma in primary sclerosing cholangitis. J Hepatol 2001;34:360–365. Bergquist A, Ekbom A, Olsson R, et al. Hepatic and extrahepatic malignancies in primary sclerosing cholangitis. J Hepatol 2002;36:321–327. Feldstein AE, Perrault J, El-Youssif M, et al. Primary sclerosing cholangitis in children: a long-term follow-up study. Hepatology 2003;38:210–217. Abdo AA, Bain VG, Kichian K, Lee SS. Evolution of autoimmune hepatitis to primary sclerosing cholangitis: A sequential syndrome. Hepatology 2002;36:1393–1399. Kaya M, Angulo P, Lindor KD. Overlap of autoimmune hepatitis and primary sclerosing cholangitis: an evaluation of a modified scoring system. J Hepatol 2000;33:537–542. van Buuren HR, van Hoogstraten HJF, Terkivatan T, et al. High prevalence of autoimmune hepatitis among patients with primary sclerosing cholangitis. J Hepatol 2000;33:543–548. Florin THJ, Pandeya N, Radford-Smith GL. Epidemiology of appendicectomy in primary sclerosing cholangitis and ulcerative colitis: its influence on the clinical behaviour of these diseases. Gut 2004;53:973–979. Poritz LS, Koltun WA. Surgical management of ulcerative colitis in the presence of primary sclerosing cholangitis. Dis Colon Rectum 2003;46:173–178. Dvorchik I, Subotin M, Demetris AJ, et al. Effect of liver transplantation on inflammatory bowel disease in patients with primary sclerosing cholangitis. Hepatology 2002;35:380–384. Cottone M, Pietrosi G, Martorana G, et al. Prevalence of cytomegalovirus infection in severe refractory ulcerative and Crohn’s colitis. Am J Gastroenterol 2001;96:773–775.
66. Lindberg BU, Broome U, Persson B. Proximal colorectal dysplasia or cancer in ulcerative colitis. The impact of primary sclerosing cholangitis and sulfasalazine: results from a 20-year surveillance study. Dis Colon Rectum 2001;44: 77–85. 67. Shetty K, Rybicki L, Brzezinski A, et al. The risk for cancer or dysplasia in ulcerative colitis patients with primary sclerosing cholangitis. Am J Gastroenterol 1999;94:1643–1649. 68. Soetikno RM, Lin OS, Heidenreich PA, et al. Increased risk of colorectal neoplasia in patients with primary sclerosing cholangitis and ulcerative colitis: a meta-analysis. Gastrointest Endosc 2002;56:48–54. 68a. Vierling JM. Aetiopatnogenesis of primary sclerosing cholangitis. In: Manns MP, Chapman RW, Stiehl A. et al., eds. Primary Sclerosing Cholangitis. Dordrecht Netneulands. Kluwer Academic Publishers; 1998:37–45. 69. Lindberg BU, Broome U, Persson B. Proximal colorectal dysplasia or cancer in ulcerative colitis. The impact of primary sclerosing cholangitis and sulfasalazine – results from a 20-year surveillance study. Dis Colon Rectum 2001;44:77–83. 70. Tung BY, Emond MJ, Haggitt RC, et al. Ursodiol use is associated with lower prevalence of colonic neoplasia in patients with ulcerative colitis and primary sclerosing cholangitis. Ann Intern Med 2001;134:89–95. 71. Pardi DS, Loftus EV, Kremers WK, et al. Ursodeoxycholic acid as a chemopreventive agent in patients with ulcerative colitis and primary sclerosing cholangitis. Gastroenterology 2003;124:889–893. 72. van de Vrie W, de Man RA, van Buuren HR, et al. Inflammatory bowel disease and liver transplantation for primary sclerosing cholangitis. Eur J Gastroenterol Hepatol 2003;15:657–663. 73. Vera A, Gunson BK, Ussatoff V, et al. Colorectal cancer in patients with inflammatory bowel disease after liver transplantation for primary sclerosing cholangitis. Transplantation 2003;75:1983–1988. 74. Stahlberg D, Veress B, Tribukait B, Broome U. Atrophy and neoplastic transformation of the ileal pouch mucosa in patients with ulcerative colitis and primary sclerosing cholangitis – a case–control study. Dis Colon Rectum 2003;46:770–778. 75. Al-Osaimi AMS, Berg CL. Association of primary sclerosing cholangitis and celiac disease: A case report and review of the literature. Dig Dis Sci 2004;49:438–443. 76. Lohse AW, Schramm C, Boberg KM. Extrahepatic and extraintestinal manifestations of primary sclerosing cholangitis. Freiburg: Kluwer Academic, 2003. 76a. Saarinen S, Olerup O, Broomé U. Increased freqnency of autoimmune diseases in patients with primary sclerosing cholangitis. Am J Gastroenterol 2000;95:3195–3199. 77. Resnick D, Krinsky GA, Lavelle MT, et al. Do patients with primary sclerosing cholangitis have a greater frequency of pancreatic abnormalities at MRI than patients with other liver diseases? J Comput Assist Tomogr 2002;26:994–999. 78. Ito K, Mitchell DG, Outwater EK. Pancreatic changes in primary sclerosing cholangitis: evaluation with MR imaging. Am J Roentgenol 1999;173:1535–1539. 79. Takikawa H, Takamori Y, Tanaka A, et al. Analysis of 388 cases of primary sclerosing cholangitis in Japan – presence of a subgroup without pancreatic involvement in older patients. Hepatol Res 2004;29:153–159. 80. Eerens I, Vanbeckevoort D, Vansteenbergen W, Van Hoe L. Autoimmune pancreatitis associated with primary sclerosing cholangitis: MR imaging findings. Eur Radiol 2001;11:1401–1404. 81. Hamano H, Kawa S, Horiuchi A, et al. High serum IgG4 concentrations in patients with sclerosing pancreatitis. N Engl J Med 2001;344:732–738.
851
Section VI. Immune Diseases
82. Younossi ZM, Boparai N, Price LL, et al. Health-related quality of life in chronic liver disease: the impact of type and severity of disease. Am J Gastroenterol 2001;96:2199–2205. 83. Kim WR, Lindor KD, Malinchoc M, et al. Reliability and validity of the NIDDK-QA instrument in the assessment of quality of life in ambulatory patients with cholestatic liver disease. Hepatology 2000;32:924–929. 84. Gross CR, Malinchoc M, Kim WR, et al. Quality of life before and after liver transplantation for cholestatic liver disease. Hepatology 1999;29:356–364. 85. Swain M. Pathogenesis of pruritus and fatigue in cholestatic liver disease. Dordrecht: Kluwer Academic, 2003. 86. Bjornsson E, Simren M, Olsson R, Chapman RW. Fatigue in patients with primary sclerosing cholangitis. Scand J Gastroenterol 2004;39:961–968. 87. ter Borg PCJ, van Os E, van den Broek WW, et al. Fluvoxamine for fatigue in primary biliary cirrhosis and primary sclerosing cholangitis: a randomised controlled trial. BMC Gastroenterology 2004;4:13. 88. Levy C, Lindor KD. Current management of primary biliary cirrhosis and primary sclerosing cholangitis. J Hepatol 2003;38(Suppl 1):S24–S37. 89. Chapman RW. The management of primary sclerosing cholangitis. Curr Gastroenterol Rep 2003;5:9–17. 90. Zein CO, Lindor KD, Angulo P. Prevalence and predictors of esophageal varices in patients with primary sclerosing cholangitis. Hepatology 2004;39:204–210. 91. Olsson R, Broome U, Danielsson A, et al. Spontaneous course of symptoms in primary sclerosing cholangitis: relationships with biochemical and histological features. HepatoGastroenterology 1999;46:136–141. 92. Angulo P, Lindor KD. Primary sclerosing cholangitis. Hepatology 1999;30:325–332. 93. Stiehl A, Rudolph G, Kloters-Plachky P, et al. Development of dominant bile duct stenoses in patients with primary sclerosing cholangitis treated with ursodeoxycholic acid: outcome after endoscopic treatment. J Hepatology 2002;36:151–156. 93a. Parker RGF, Kendall EJC. The liver in ulcerative colitis. Brit Med J 1954;2:1030–1032. 93b. Converse CF, Reagan JW, DcCosse JJ. Ulcerative colitis and carcinoma of the bile ducts. Am J Surg 1971;121: 39–45. 94. Burak K, Angulo P, Pasha TM, et al. Incidence and risk factors for cholangiocarcinoma in primary sclerosing cholangitis. Am J Gastroenterol 2004;99:523–526. 95. Chalasani N, Baluyut A, Ismail A, et al. Cholangiocarcinoma in patients with primary sclerosing cholangitis: A multicenter case–control study. Hepatology 2000;31:7–11. 96. Boberg KM, Bergquist A, Mitchell S, et al. Cholangiocarcinoma in primary sclerosing cholangitis: risk factors and clinical presentation. Scand J Gastroenterol 2002;37:1205–1211. 97. Kim WR, Therneau TM, Wiesner RH, et al. A revised natural history model for primary sclerosing cholangitis. Mayo Clin Proc 2000;75:688–694. 98. Boberg KM, Schrumpf E, Bergquist A, et al. Cholangiocarcinoma in primary sclerosing cholangitis: K-ras mutations and Tp53 dysfunction are implicated in the neoplastic development. J Hepatol 2000;32:374–380. 99. Kubicka S, Kuhnel F, Flemming P, et al. K-ras mutations in the bile of patients with primary sclerosing cholangitis. Gut 2001;48:403–408. 100. Bergquist A, Glaumann H, Stal P, et al. Biliary dysplasia, cell proliferation and nuclear DNA-fragmentation in primary sclerosing cholangitis with and without cholangiocarcinoma. J Intern Med 2001;249:69–75. 101. Kang Y, Kim WR, Lee H, Kim Y. Mutation of p53 and K-ras, and loss of heterozygosity of APC in intrahepatic cholangiocarcinoma. Lab Invest 1999;79:477–483.
852
102. Ahrendt SA, Eisenberger CF, Yip L, et al. Chromosome 9p21 loss and p16 inactivation in primary sclerosing cholangitisassociated cholangiocarcinoma. J Surg Res 1999;84:88–93. 103. Fleming KA, Boberg KM, Glaumann H, et al. Biliary dysplasia as a marker of cholangiocarcinoma in primary sclerosing cholangitis. J Hepatol 2001;34:360–365. 103a. Ludwig J, Wahlström H, Batts K, et al. Papillary bile ducts dysplasia in primary sclerosising cholangitis. Gastroenterology 1992;102:2134–2138. 104. Campbell WL, Peterson MS, Federle MP, et al. Using CT and cholangiography to diagnose biliary tract carcinoma complicating primary sclerosing cholangitis. Am J Roentgenol 2001;177:1095–1100. 104a. Keiding S, Hansen SB, Rasmussen HH, et al. Detection of cholangiocarcinoma in primary sclerosing cholangitis by positron emission tomography. Hepatology 1998;28:700–706. 105. Kluge R, Schmidt F, Caca K, et al. Positron emission tomography with [F-18]fluoro-2-deoxy-D-glucose for diagnosis and staging of bile duct cancer. Hepatology 2004;33:1029–1035. 106. Anderson CD, Rice MH, Pinson W, et al. Fluorodeoxyglucose PET imaging in the evaluation of gallbladder carcinoma and cholangiocarcinoma. J Gastrointest Surg 2004;8:90–97. 107. Brandsater B, Broome U, Isoniemi H, et al. Liver transplantation for primary sclerosing cholangitis in the Nordic countries: Outcome after acceptance to the waiting list. Liver Transpl 2003;9:961–969. 108. Ponsioen CY, Vrouenraets SM, van Milligen de Wit AW, et al. Value of brush cytology for dominant strictures in primary sclerosing cholangitis. Endoscopy 1999;31:305–309. 109. Kipp BR, Stadheim LM, Halling SA, et al. A comparison of routine cytology and fluorescence in situ hybridization for the detection of malignant bile duct strictures. Am J Gastroenterol 2004;99:1675–1681. 110. Ponsioen CY, Vrouenraets SME, de Wit A, et al. Value of brush cytology for dominant strictures in primary sclerosing cholangitis. Endoscopy 1999;31:305–309. 111. Fritscher-Ravens A, Broering DC, Knoefel WT, et al. EUSguided fine-needle aspiration of suspected hilar cholangiocarcinoma in potentially operable patients with negative brush cytology. Am J Gastroenterol 2004;99: 45–51. 112. Lindberg B, Arnelo U, Bergquist A, et al. Diagnosis of biliary strictures in conjunction with endoscopic retrograde cholangiopancreaticography, with special reference to patients with primary sclerosing cholangitis. Endoscopy 2002;34:909–916. 113. Rumalla A, Baron T, Leontovich O, et al. Improved diagnostic yield of endoscopic biliary brush cytology by digital image analysis. Mayo Clin Proc 2001;76:29–33. 114. Baron TH, Harewood GC, Rumalla A, et al. A prospective comparison of digital image analysis and routine cytology for the identification of malignancy in biliary tract strictures. Clin Gastroenterol Hepatol 2004;2:214–219. 115. Bjornsson E, Kilander A, Olsson R. CA 19-9 and CEA are unreliable markers for cholangiocarcinoma in patients with primary sclerosing cholangitis. Liver 1999;19:501–508. 116. Siqueira E, Schoen RE, Silverman W, et al. Detecting cholangiocarcinoma in patients with primary sclerosing cholangitis. Gastrointest Endosc 2002;56:40–47. 117. Hultcrantz R, Olsson R, Danielsson A, et al. A 3-year prospective study on serum tumor markers used for detecting cholangiocarcinoma in patients with primary sclerosing cholangitis. J Hepatol 1999;30:669–673. 118. Saurin J, Joly-Pharaboz M, Pernas P, et al. Detection of Ki-ras gene point mutations in bile specimens for the differential diagnosis of malignant and benign biliary strictures. Gut 2000;47:357–361.
Chapter 42 PRIMARY SCLEROSING CHOLANGITIS
119. Boberg KM, Schrumpf E, Bergquist A, et al. Cholangiocarcinoma in primary sclerosing cholangitis: K-ras mutations and Tp53 dysfunction are implicated in the neoplastic development. J Hepatol 2000;32: 374–380. 120. Ahrendt SA, Pitt HA, Nakeeb A, et al. Diagnosis and management of cholangiocarcinoma in primary sclerosing cholangitis. J Gastrointest Surg 1999;3:357–367;discussion 367–368. 121. Chalasani N, Baluyut A, Ismail A, et al. Cholangiocarcinoma in patients with primary sclerosing cholangitis: a multicenter case–control study. Hepatology 2000;31:7–11. 122. Brandsaeter B, Isoniemi H, Broome U, et al. Liver transplantation for primary sclerosing cholangitis: predictors and consequences of hepatobiliary malignancy. J Hepatol 2004;40:815–822. 122a. Ahrendt SA, Pitt HA, Nakeeb A, et al. Diagnosis and management of cholangiocarcinoma in primary sclerosing cholangitis. J Gastrointest Surg 1999;4:357–367. 123. De Vreede I, Steers JL, Burch PA, et al. Prolonged disease-free survival after orthotopic liver transplantation plus adjuvant chemoirradiation for cholangiocarcinoma. Liver Transpl 2000;6:309–316. 124. Stiehl A, Rudolph G, Kloters-Plachky P, et al. Development of dominant bile duct stenoses in patients with primary sclerosing cholangitis treated with ursodeoxycholic acid: outcome after endoscopic treatment. J Hepatol 2002;36:151–156. 125. Buckles DC, Lindor KD, Larusso NF, et al. In primary sclerosing cholangitis, gallbladder polyps are frequently malignant. Am J Gastroenterol 2002;97:1138–1142. 125a. Jonsson F, Isaksson B, Feychting M, et al. Pancreato-biliary inflammation and hyperinsulinemia are risk factors of pancreatic cancer. Pancreatology 2003;3:209–210. 126. Chen W, Gluud C. Bile acids for primary sclerosing cholangitis. Cochrane Database Syst Rev 2003:CD003626. 127. Rost D, Rudolph G, Kloeters-Plachky P, Stiehl A. Effect of high-dose ursodeoxycholic acid on its biliary enrichment in primary sclerosing cholangitis. Hepatology 2004;40:693–698. 128. Mitchell SA, Bansi DS, Hunt N, et al. A preliminary trial of high-dose ursodeoxycholic acid in primary sclerosing cholangitis. Gastroenterology 2001;121:900–907. 129. Harnois DM, Angulo P, Jorgensen RA, et al. High-dose ursodeoxycholic acid as a therapy for patients with primary sclerosing cholangitis. Am J Gastroenterol 2001;96:1558–1562. 130. Olsson R, Boberg KM, de Muckadell OS, et al. High-dose ursodeoxycholic acid in primary sclerosing cholangitis: a five year multicenter randomised controlled study. Gastroenterology 2005;129:1464–1472. 131. Boberg KM, Egeland T, Schrumpf E. Long-term effect of corticosteroid treatment in primary sclerosing cholangitis patients. Scand J Gastroenterol 2003;38:991–995. 132. Angulo P, Batts KP, Jorgensen RA, et al. Oral budesonide in the treatment of primary sclerosing cholangitis. Am J Gastroenterol 2000;95:2333–2337. 133. Angulo P, Bharucha AE, Jorgensen RA, et al. Oral nicotine in treatment of primary sclerosing cholangitis – a pilot study. Dig Dis Sci 1999;44:602–607. 134. Vleggaar FP, van Buuren HR, van Berge Henegouwen GP, et al. No beneficial effects of transdermal nicotine in patients with primary sclerosing cholangitis: results of a randomized doubleblind placebo-controlled cross-over study. Eur J Gastroenterol Hepatol 2001;13:171–175. 135. Bharucha AE, Jorgensen R, Lichtman SN, et al. A pilot study of pentoxifylline for the treatment of primary sclerosing cholangitis. Am J Gastroenterol 2000;95:2338–2342. 136. Epstein MP, Kaplan MM. A pilot study of etanercept in the treatment of primary sclerosing cholangitis. Dig Dis Sci 2004;49:1–4.
137. Duchini A, Younossi ZM, Saven A, et al. An open-label pilot trial of cladibrine (2-cholordeoxyadenosine) in patients with primary sclerosing cholangitis. J Clin Gastroenterol 2000;31:292–296. 138. Schramm C, Schirmacher P, Helmreich-Becker I, et al. Combined therapy with azathioprine, prednisolone, and ursodiol in patients with primary sclerosing cholangitis. A case series. Ann Intern Med 1999;131:943–946. 139. van Hoogstraten HJ, Vleggaar FP, Boland GJ, et al. Budesonide or prednisone in combination with ursodeoxycholic acid in primary sclerosing cholangitis: a randomized double-blind pilot study. Belgian–Dutch PSC Study Group. Am J Gastroenterol 2000;95:2015–2022. 140. Sterling RK, Salvatori JJ, Luketic VA, et al. A prospective randomized pilot study of ursodeoxycholic acid combined with mycophenolate mofetil in the treatment of primary sclerosing cholangitis. Aliment Pharmacol Ther 2004;20:943–948. 141. Ponsioen CY, Lam K, de Wit A, et al. Four years experience with short term stenting in primary sclerosing cholangitis. Am J Gastroenterol 1999;94:2403–2407. 142. Enns R, Eloubeidi MA, Mergener K, et al. Predictors of successful clinical and laboratory outcomes in patients with primary sclerosing cholangitis undergoing endoscopic retrograde cholangiopancreatography. Can J Gastroenterol 2003;17:243–248. 143. Kaya M, Petersen BT, Angulo P, et al. Balloon dilation compared to stenting of dominant strictures in primary sclerosing cholangitis. Am J Gastroenterol 2001;96:1059–1066. 144. Baluyut AR, Sherman S, Lehman GA, et al. Impact of endoscopic therapy on the survival of patients with primary sclerosing cholangitis. Gastrointest Endosc 2001;53:308–312. 145. Sekido H, Takeda K, Morioka D, et al. Liver transplantation for primary sclerosing cholangitis. J Hepato-Biliary-Pancreatic Surg 1999;6:373–376. 146. Wiesner RH. Liver transplantation for primary sclerosing cholangitis: timing, outcome, impact of inflammatory bowel disease and recurrence of disease. Best Pract Res Clin Gastroenterol 2001;15:667–680. 147. Graziadei IW, Wiesner RH, Marotta PJ, et al. Long-term results of patients undergoing liver transplantation for primary sclerosing cholangitis. Hepatology 1999;30:1121–1127. 148. Bjoro K, Schrumpf E. Liver transplantation for primary sclerosing cholangitis. J Hepatol 2004;40:570–577. 149. Neuberger J, Gunson B, Komolmit P, et al. Pretransplant prediction of prognosis after liver transplantation in primary sclerosing cholangitis using a Cox regression model. Hepatology 1999;29:1375–1379. 150. Graziadei IW, Wiesner RH, Batts KP, et al. Recurrence of primary sclerosing cholangitis following liver transplantation. Hepatology 1999;29:1050–1056. 151. Solano E, Khakhar A, Bloch M, et al. Liver transplantation for primary sclerosing cholangitis. Transplant Proc 2003;35:2431–2434. 152. Maheshwari A, Yoo HY, Thuluvath PJ. Long-term outcome of liver transplantation in patients with PSC: A comparative analysis with PBC. Am J Gastroenterol 2004;99:538–542. 153. Vera A, Moledina S, Gunson B, et al. Risk factors for recurrence of primary sclerosing cholangitis of liver allograft. Lancet 2002;360:1943–1944. 154. Gow PJ, Chapman RW. Liver transplantation for primary sclerosing cholangitis. Liver 2000;20:97–103. 154a. Miki C, Harrison JD, Gunson BK, et al. Inflammatory bowel disease in primary sclerosing cholangitis: an analysis of patients undergoing liver transplantation. Br J Surg 1995;82:1114–1147. 154b. Narumi S, Roberts JP, Emond JC, et al. Liver transplantation for sclerosing cholangitis. Hepatology 1995;22:451–455. 155. Welsh FKS, Wigmore SJ. Roux-En-Y choledochojejunostomy is the method of choice for biliary reconstruction in liver
853
Section VI. Immune Diseases
156.
156a.
156b.
157.
157a.
158.
854
transplantation for primary sclerosing cholangitis. Transplantation 2004;77:602–604. Haagsma EB, Van den Berg AP, Kleibeuker JH, et al. Inflammatory bowel disease after liver transplantation: the effect of different immunosuppressive regimens. Aliment Pharmacol Ther 2003;18:33–44. Broomé U, Olsson R, Lööf L, et al. Natural history and prognostic factors in 305 Swedish patients with primary sclerosing cholangitis. Gnt 1996;38:610–615. Porayko MK, Wiesner RH, La Russo NF. Patients with asympotomatic primary sclerosing cholangitis frequently have progressive disease. Gastroenterology 1990;98:1594–1597. Angulo P, Larson DR, Therneau TM, et al. Time course of histological progression in primary sclerosing cholangitis. Am J Gastroenterol 1999;94:3310–3313. Olsson RG, Asztely MS. Prognostic value of cholangiography in primary sclerosing cholangitis. Eur J Gastroenterol Heatol 1995;7:251–256. Ponsioen CY, Vrouenraets SME, Prawirodirdjo W, et al. Natural history of primary sclerosing cholangitis and prognostic value of cholangiography in a Dutch population. Gut 2002;51:562–566.
159. Gossard AA, Lindor KD. Pregnancy in a patient with primary sclerosing cholangitis. J Clin Gastroenterol 2002;35:353–355. 160. Cullen S, Chapman R. Primary sclerosing cholangitis. Autoimmun Rev 2003;2:305–312. 161. Kubota T, Thomson A, Clouston AD, et al. Clinicopathologic findings of recurrent primary sclerosing cholangitis after orthotopic liver transplantation. J Hepato-Biliary-Pancreatic Surg 1999;6:377–381. 162. Yusoff IF, House AK, De Boer WB, et al. Disease recurrence after liver transplantation in Western Australia. J Gastroenterol Hepatol 2002;17:203–207. 163. Kugelmas M, Spiegelman P, Osgood MJ, et al. Different immunosuppressive regimens and recurrence of primary sclerosing cholangitis after liver transplantation. Liver Transpl 2003;9:727–732. 164. Khettry U, Keaveny A, Goldar–Najafi A, et al. Liver transplantation for primary sclerosing cholangitis: A long-term clinicopathologic study. Hum Pathol 2003;34:1127-1136.
Section VI: Immune Diseases
43
OVERLAP SYNDROMES E. J. Heathcote Abbreviations AIH autoimmune hepatitis AMA mitochondrial antibodies ANA antinuclear antibodies
ASMA PBC
antismooth muscle antibody primary biliary cirrhosis
INTRODUCTION No criteria have been established that define overlap syndromes. Individuals who appear to have two overt autoimmune chronic liver diseases simultaneously or sequentially are perhaps the most clearcut cases of overlap syndrome. The term ‘overlap’ is not used for those who have two liver diseases that are quite disparate, for example chronic hepatitis B and non-alcoholic fatty liver disease. However, overlap has occasionally been used when referring to individuals infected with hepatitis C who also have serological markers suggesting a background of autoimmunity. Difficulty arises when the term overlap syndrome is applied somewhat loosely to describe patients who clearly present with one autoimmune liver disease but who also have features of another. For example, patients who have all the clinical, biochemical and histologic features of primary biliary cirrhosis (PBC) but test positive for antinuclear antibodies (ANA) rather than the typical hallmark of mitochondrial antibodies (AMA), are described by some as having PBC-AIH overlap, whereas others would simply call this disease AMA-negative PBC. This heterogeneity in nomenclature leads to much confusion in the literature regarding overlap syndromes. Interpreting data about natural history and potential therapeutic interventions is consequently very difficult.
AUTOANTIBODIES IN SERUM OF PATIENTS WITH LIVER DISEASE In a patient with liver disease it is traditional to associate the finding of autoantibodies in serum with the presence of autoimmune liver disease, but many individuals who have liver disease which is clearly not autoimmune in origin, for example alcoholic hepatitis, and chronic hepatitis C, will test positive for the non-organ non-speciesspecific autoantibodies such as ANA.1 These same autoantibodies may be detected in 2–14% of perfectly healthy individuals: the older the individual, the higher this figure.2 So, their presence in the sera of individuals with chronic liver disease does not necessarily indicate an autoimmune basis for the underlying liver disease. Sometimes an autoantibody usually thought typical for one autoimmune disease (e.g. AMA in PBC) may be detected in a patient who otherwise has all the classic features of another liver disease (e.g. autoimmune hepatitis; AIH).3 This may truly reflect an overlap between AIH and PBC; alternatively, it could be AIH with
ULN UDCA
upper limit normal ursodeoxycholic acid
preclinical PBC, or it may in fact be a variant of AIH with no relation to PBC whatsoever.
THE DYNAMIC NATURE OF AUTOIMMUNE LIVER DISEASE The lack of a clear definition for ‘overlap syndrome’ allows for a great deal of variation in the interpretation of the term. There is no single, absolute diagnostic test for any of the autoimmune liver diseases; rather, the diagnosis is made by reviewing a composite of the clinical, biochemical, serologic/immunologic and histologic features of the disease, supported by patient response to therapy. The somewhat complex nature of an individual’s liver disease may not be entirely apparent at the time of their first clinical presentation, and therefore it is probably unwise to make a diagnosis of an overlap syndrome until reasonable follow-up has been achieved.
MORE THAN ONE AUTOIMMUNE LIVER DISEASE PRESENT SIMULTANEOUSLY AUTOIMMUNE HEPATITIS (AIH) AND PRIMARY SCLEROSING CHOLANGITIS (PSC) In Children In 1995 Wilschanski and colleagues4 retrospectively described the clinical presentation and outcome of 32 children with radiological evidence of PSC. Of these 32, nine were originally diagnosed as having autoimmune hepatitis and the others had isolated features typical of AIH. Intrahepatic biliary disease predominated, with only three of 32 having common bile duct involvement. Of the nine children with overt autoimmune hepatitis and PSC, five (55.6%) subsequently required a liver transplant. In contrast, only five of the 22 (22.7%) with PSC and only minor features of AIH progressed to liver transplant. Hence overt AIH and PSC appeared to carry a worse prognosis. Perhaps also relevant was the older age at presentation of the overt AIH/PSC group. It may be that their worse prognosis was related to the fact that their liver disease had been present but clinically silent for many years prior to diagnosis. No specific features reliably predicted which children had the AIH/PSC overlap at the time of initial presentation. Coexisting inflammatory bowel
855
Section VI. Immune Diseases
disease was just as common in those with PSC and mild isolated features of AIH as in those with overt overlap. In addition, despite the presence of biliary tract disease in all patients, half the children had normal levels of serum alkaline phosphatase. HLA typing did not show any consistent pattern, but the sample size was very small. Some 6 years later, Gregorio and colleagues5 described a 16-year prospective study, again in children, who at presentation were all given a primary diagnosis of autoimmune hepatitis. Between 1984 and 1997, 76 children were referred with clinical and/or biochemical evidence of liver disease associated with the presence of serum autoantibodies. Of these 76, 21 (27.6%) were considered too ill to undergo radiologic investigation of their biliary tree and were excluded from the study. The remaining 55 were evaluated and met the revised autoimmune hepatitis scoring system criteria for a diagnosis of AIH. In addition, they all underwent radiologic investigation of the biliary tree at the time of presentation (median time between diagnosis of AIH and ERCP 1.1 months). Twenty-three (41.8%) of these children were found to have abnormal cholangiography, but the typical stricturing and beading pattern well described in adults with PSC was only observed in two (9%). The most common abnormality was mucosal irregularity of the common bile duct, which was present in 15 (65%) patients. An abnormal pancreatic duct was observed in three (13%), and stones were detected in the common duct/cystic duct/gallbladder in two (9%). The authors referred to those with AIH and an abnormal biliary tree as having autoimmune sclerosing cholangitis (ASC). Follow-up cholangiography was performed from 1 to 9 years later in 36 patients, 17 of whom had previously been diagnosed with ASC. Nine of these 17 (52.9%) were found to have static disease and eight (47.1%) had progressive intra- or extrahepatic bile duct abnormalities. Of the 28 who did not have ASC at initial presentation, 17 (60.7%) underwent repeat cholangiography and all but one remained with a normal endoscopic retrograde cholangiopancre-
atography (ERCP) a median of 6 years later. The one child who was shown to change from having a normal duct to having advanced intra- and extrahepatic disease did so 8 years after her original presentation with AIH, ulcerative colitis and urticaria pigmentosa. All 55 children were alive at the time of publication, but four had required a liver transplant, three of whom had ASC and the fourth ductopenia on liver biopsy. Thus, just as in the retrospective study by Wilschanski et al.,4 overlap of AIH with cholangitis appears to have a poorer prognosis than AIH alone. In both these pediatric studies there were very few clinical features at presentation that distinguished those children with the overlap of AIH plus biliary tract disease from those with isolated AIH. Gregorio et al.5 found that inflammatory bowel disease was more common in those with ASC (12, i.e. 44%) than in those with AIH alone (5, i.e. 18%), whereas a family history of autoimmune disease was found more commonly (20, or 71%) in children with AIH alone than in those with ASC (10, or 37%). Similarly, there were very few laboratory features that distinguished these groups of children. Although the presence of the autoantibody ANCA was significantly more common in those with ASC (74%), ANCA was also positive in 33% of those with only AIH. Of great practical relevance is that neither the height of the serum alkaline phosphatase nor the g-glutamyltransferase (GGT) levels at presentation distinguished these children in either study. Thus coexisting biliary tract disease must be sought in all children given a diagnosis of AIH.
In Adults The coexistence of AIH and PSC appears to much be less frequent in adults than in children. There have been several case reports6-8 describing patients who appear to have both AIH and PSC simultaneously (Figures 43-1, 43-2).
Figure 43-1. Liver biopsy.
856
Chapter 43 OVERLAP SYNDROMES Figure 43-2. Magnetic resonance cholangiogram (MRC).
Recently Abdo et al.9 published a case series of six adults with AIH in whom only later (mean 4.6 years) was PSC first documented on ERCP. These patients were reinvestigated, as they had all become resistant to the treatment which in the past had controlled their AIH. Three of these individuals had undergone ERCP at the time of initial presentation, and in all three the biliary tree had appeared normal. None of the six had evidence of intrahepatic bile duct disease on histologic evaluation of their initial liver biopsy. In followup, all six were found to have changes typical of PSC on ERCP. Three of the six subsequently required a liver transplant. Because the ‘gold standard’ hallmark for PSC – namely beading and stricturing of the intra- and/or extrahepatic biliary tree on cholangiography – is a late manifestation of the disease and may not be demonstrated even when examination of the interlobular bile ducts on liver histology suggests intrahepatic PSC, it is not possible to say whether the overlap of AIH and PSC described above was truly sequential. In this series of six adult cases (just as in the children with AIH and PSC) none of the usual laboratory markers for cholestasis, namely the serum levels of GGT and alkaline phosphatase (ALP), were indicative of biliary tract pathology. Recent studies employing the cytokeratin 7 stain show that the use of this stain may facilitate the identification of early intrahepatic biliary obstruction.10 In the children with overlap, retention of copper in the periportal hepatocytes was the first clue to suggest longstanding cholestasis.
These case series indicate that clinicians should maintain a high index of suspicion, particularly in children and young adults who present with autoimmune hepatitis, that they may also have concurrent PSC and vice versa. Now that a reliable non-invasive method of examining the biliary tree is available, namely MRI, the full extent of the overlap of AIH with PSC may become realized.11 Unfortunately, in individuals who present with PSC there is no single noninvasive test to indicate the coexistence of AIH.
MANAGEMENT OF AIH/PSC OVERLAP SPECIFIC TREATMENT Once it has been recognized that a true overlap exists, each individual autoimmune liver disease requires the appropriate therapy. It is often the presence of a greater than tenfold elevation in serum aminotransferase levels in the patient with PSC that prompts the clinician to consider overlapping autoimmune hepatitis. However, a transaminitis, at least in the early phases, acute cholestasis (as is seen with stones in the common bile duct), is not unusual in PSC. Thus in individuals who present with PSC it may be optimal to initiate treatment that improves the cholestasis prior to the introduction of any immunosuppressive therapy. Ursodeoxycholic acid (UDCA),
857
Section VI. Immune Diseases
although not shown to improve survival in PSC,12 does improve the biochemical picture of cholestasis, liver histology, and even the cholangiographic features of this disease – particularly if given in a dose >20 mg/kg/day.13,14 In those individuals who first present with features of autoimmune hepatitis who are also shown to have PSC, it is probably optimal for immunosuppressive therapy to be the first line of treatment, particularly in patients who have florid symptoms and/or bridging necrosis on liver histology. UDCA may be added once the AIH has settled. Reports of the long-term follow-up of both the pediatric studies of AIH/PSC (ASC) overlap indicate that the outcome was worse than in those with either AIH or PSC alone, and that those with the overlapping diseases frequently required liver transplantation. There are insufficient data to indicate whether this is also true for adults with AIH and PSC.
PREVENTATIVE TREATMENT There are two significant non-hepatic complications of the AIH/PSC syndrome that may be adversely affected by the primary treatments for this disease.
Metabolic Bone Diseases All forms of chronic cholestasis, particularly in the presence of jaundice, are complicated by the development of progressive osteopenia/osteoporosis. The need to administer corticosteroid therapy in those with an AIH/PSC overlap syndrome further increases the risk of thin bones in such individuals. Thus even in the young it is important that adequate calcium and vitamin D supplementation be provided. Recent studies in patients with PBC suggest that the bisphosphonate alendronate is effective in the prevention and management of osteoporosis in individuals with chronic cholestasis.15 Prophylactic bisphosphonate therapy has also been shown to be beneficial in individuals who require long-term corticosteroids for any reason.16 In those with additional pancreatic involvement or celiac disease, osteomalacia caused by malabsorption of calcium and vitamin D may also be a feature of the complicating metabolic bone disease.
Sepsis The need for corticosteroid therapy to control autoimmune hepatitis requires that all patients be well educated about the need for prompt treatment of any focus of infection. In those individuals with additional sclerosing cholangitis it is particularly important to avoid maneuvers that could introduce infection into a poorly draining biliary system. Appropriate antibiotic coverage needs to be provided for all invasive procedures.
ONE AUTOIMMUNE LIVER DISEASE WITH ADDITIONAL FEATURES OF ANOTHER AUTOIMMUNE LIVER DISEASE AIH WITH FEATURES OF PBC Before AMA were identified as a hallmark for PBC, it had been recognized that certain individuals thought to have ‘postnecrotic
858
cirrhosis’ had additional features of cholestasis. Once ERCP became available, some of these individuals were found to have PSC. Others were found to test positive for AMA despite features of coexistent ‘hepatitis’. In 1976, Geubel et al.17 from the Mayo Clinic described 125 individuals given a diagnosis of severe chronic active hepatitis, 15 of whom had had a poor response to immunosuppressive therapy. Other features that distinguished these poor responders were a high ALP:AST, high IgM, the presence of pruritus, and AMA in serum. In addition, the histologic analysis indicated that the patients who failed to respond to immunosuppressive therapy never had evidence of bridging necrosis. The degree of bile duct damage, cholestasis or granulomas did not distinguish ‘responders’ from ‘non-responders’, although the non-responders did have significantly fewer interlobular bile ducts in the portal tracts than the responders. The poor responders to corticosteroids represented 12% of their population with postnecrotic cirrhosis, and such patients were thought to represent an overlap with PBC.
PBC WITH FEATURES OF AIH There is rarely any confusion about making a diagnosis of PBC. The combination of positive AMA and an elevated ALP in a woman with or without symptoms of cholestasis is sufficient to make the diagnosis. Liver histology confirms the diagnosis and assists with establishing a prognosis, but a liver biopsy is not considered essential to make a diagnosis of PBC.18 The concept of AMA-positive PBC with overlapping features of AIH has arisen because some patients given a primary diagnosis of PBC have a greater elevation of serum transaminase levels than ‘expected’. Chazouilleres et al.19 examined 130 patients given an initial diagnosis of PBC and found that 12% had features of autoimmune hepatitis. The autoimmune features these authors chose were a composite of ALT (>fivefold ULN), serum IgG levels (>twofold ULN) or positive ASMA and a liver biopsy which showed moderate to severe periportal and periseptal lymphocytic piecemeal necrosis. Subsequently Talwalkar20 from the Mayo Clinic scrutinized a similar number of cases of PBC for features of AIH by applying the revised AIH scoring system.21 He was unable to find any patients with PBC who scored in the ‘definite’ range for AIH, though he did find that 19% of their PBC population scored in the ‘probable’ range.
IS IT PRIMARILY PBC OR AIH? The introduction of AMA testing greatly facilitated the diagnosis of PBC.22 However, AMA may be detected in other situations, and there are reports of patients with apparent autoimmune hepatitis who test positive for AMA.3 The report by Kenny et al.23 noted that AMA, when detected in the sera of patients thought to have AIH, tended to be of low titer and sometimes were really microsomal antibodies rather than mitochondrial antibodies. This confusion arises because anti-LKM1 antibodies (markers of type 2 AIH) stain similarly by immunofluorescence and may therefore be confused with AMA. Lohse et al.24 reported a series of 20 patients diagnosed with PBC, 20 patients diagnosed with autoimmune hepatitis, and 20 patients given a diagnosis of an overlap of PBC and AIH, and compared the biochemical, immunologic and histologic features of the three groups. The designation of ‘overlap’ was applied in 14 patients because they tested positive for AMA but had an ALT more than
Chapter 43 OVERLAP SYNDROMES
twice the upper limit of normal, and in six because they tested strongly positive for AMA and their liver histology showed features of both AIH and PBC. In their 20 cases of overt AIH, one tested positive for AMA but repeat testing by ELISA was negative, whereas 16 of the 20 ‘overlaps’ tested AMA positive by both immunofluorescence and ELISA, and all 20 patients given a primary diagnosis of PBC were AMA positive. From this heterogeneous population the authors concluded that those with ‘overlap’ in their series had more features in common with PBC, particularly in terms of elevation of serum alkaline phosphatase and serum IgM values. However, they did note that the HLA pattern was more typical of that seen in patients with AIH, namely more cases with HLA, DR3 and/or DR4, and so concluded that these patients indeed had a PBC/AIH overlap. In this retrospective study the therapy given to each patient was not standardized. Patients thought to have predominant AIH responded well to immunosuppressive therapy. Sixteen of the 20 patients diagnosed with an ‘overlap’ syndrome were also given immunosuppressive therapy for a minimum of 2 years. In followup, serum aminotransferase levels improved, as did the alkaline phosphatase levels in 12 of these 16, but all were also given additional UDCA. Of the 20 individuals given a primary diagnosis of PBC, four received immunosuppressive therapy, two with some biochemical benefit. The rest of this group was treated with UDCA, seven of whom had liver biochemical tests that returned to normal. No data beyond 2 years were presented and the numbers under study were small. Thus it cannot be said whether these treatments were appropriate or effective in the long term; neither do we know the natural history of these so-called overlaps. As both corticosteroid therapy and UDCA can be associated with an improvement in liver biochemical tests in almost all forms of liver disease, response to these therapies cannot be considered specific. It is interesting to note that none of the reports describing AIH/PBC overlap discuss any clinical features that distinguish overlapping PBC/AIH versus PBC or AIH alone – rather, the reports all focused on laboratory and histological features (Table 43-1).
AMA NEGATIVE PBC OR AIH/PBC OVERLAP? In 1987 Brunner and Klinge25 described three women who had clinical, histologic and biochemical criteria for PBC but who all tested AMA negative and instead had high-titer ANA. All were given treat-
ment with immunosuppressive therapy, and at least in the short term an improvement was seen. They were given a diagnosis of ‘immune cholangitis.’ Michielleti et al.26 later described 17 patients referred for a randomized controlled trial of UDCA for PBC who consistently tested negative for AMA by both immunofluorescence and immunoblotting, but who otherwise had all the clinical, biochemical and histologic features of PBC. These patients also all tested positive for ANA, generally in high titer. Subsequently several other series of AMA-negative individuals, all of whom appeared to have PBC but tested positive in high titer for ANA, were reported. These cases all had higher levels of IgG and lower levels of IgM, and somewhat higher serum aminotransferase levels than their AMA-positive counterparts, but in all, liver histology showed the typical histologic features of PBC, and such patients are now recognized as having AMA-negative PBC.27 Subsequently, a study by Kim28 reported that the response to UDCA in terms of changes in liver biochemistry was no different in AMA-negative PBC than in AMA-positive PBC. The natural history of AMA-negative PBC is similar to that of AMA-positive PBC, and to date the only difference appears to be their HLA associations. Whereas class 2 HLA DR8 is predominant in AMA-positive PBC, this is not the case in AMA-negative PBC.29 Thus although AMA-negative PBC is associated with certain features of AIH, namely high-titer ANA and higher IgG and aspartate aminotransferase (AST) levels, they probably do not represent a PBC/AIH overlap syndrome and should be treated as cases of PBC.
PSC WITH FEATURES OF AIH A large study, Boberg et al.30 reviewed cases of proven PSC and calculated their AIH score. The results suggested that features of AIH were indeed common in their population with PSC. However, although the AIH score may be helpful in confirming a case of AIH, it has not been validated in individuals with mixed forms of autoimmune liver disease. To rectify this confusion, this AIH score was revised. With the updated criteria21 the Mayo Clinic found that only 1.4% and 6% of patients with proven PSC had scores qualifying for ‘definite’ and ‘probable’ AIH, respectively.31 However, using the same revised scoring system, a group from the Netherlands in a somewhat younger cohort of patients with PSC found 8% with a ‘definite’ AIH score.32
PSC WITH FEATURES OF PBC Table 43-1. AIH, PBC, AIH/PBC Overlap—Laboratory and Histologic Features Laboratory
AIH
PBC
AIH/PBC overlap
ALP ALT AMA ANA SMA IgG IgM
+ +++ – +++ ++ +++ –
+++ + +++ + + + ++
++ ++ ++ ++ + ++ +/–
Liver histology Piecemeal necrosis Bile duct loss
+++ –
– +++
++ ++
There have been a few reported cases of radiologically documented PSC in whom the serum also tested positive for AMA and who had evidence of granulomatous bile duct destruction on liver biopsy.33 It is most likely that this represents two individual autoimmune liver diseases. Although there is a fairly wide differential diagnosis for the radiologic pattern of PSC, it does not include PBC. The ERCP findings in end-stage PBC may simulate PSC because external compression of the nodular liver may give an irregular appearance to the biliary tree; however, it should be easily distinguishable from true PSC. Such cases probably represent true PSC with superimposed early PBC. Liver histology in subjects who test positive for AMA but who have entirely normal liver biochemistry indicates that early lesions, including granulomatous bile duct destruction, are nearly always found on liver histology.34
859
Section VI. Immune Diseases
MANAGEMENT OF AIH WITH FEATURES OF PBC OR PBC WITH FEATURES OF AIH AND PSC WITH FEATURES OF PBC It has been known for 30 years that immunosuppressive therapy effectively improves survival in patients with severe AIH, but immunosuppressive therapy is not always recommended for those with mild AIH.35 The current standard of care for PBC is UDCA 15 mg/kg/day.36 This treatment leads to slowed progression of fibrosis37 and liver failure.38 Some would argue that treatment with UDCA increases survival free of liver transplantation,39 but none would acknowledge that this treatment cures PBC. UDCA has also been used effectively – at least in the short term – in patients with mild AIH, but it is not advised in the long term or for individuals with severe AIH. Thus there is much debate as to the appropriate management of AIH with overlapping features of PBC, and vice versa. Corpechot et al.40 have shown that in their patients with PBC outcome is worse in those with lymphocytic piecemeal necrosis on liver histology. They have advocated additional treatment with corticosteroid therapy in this patient population, but this has not been evaluated in a prospective manner. In a retrospective study of patients with PBC and features of overlapping AIH no difference in survival was observed in those patients with features of AIH, compared to those with PBC alone, when randomized to treatment with UDCA or placebo.41 Patients with PBC are at increased risk for osteoporosis, both because of their chronic cholestasis and because many are postmenopausal. Thus the major concern if steroid therapy is prescribed to those with PBC with or without overlapping features of AIH is that it may promote further osteoporosis. The same concern pertains to the treatment of AIH with features of PBC. For those patients in whom corticosteroid therapy is considered essential it may be advisable to introduce bisphosphonate therapy simultaneously with the corticosteroids. The management of patients with PSC who appear also to have PBC does not present any controversy as all forms of chronic cholestasis respond to treatment with UDCA, with an improvement in liver biochemistry.42 Recent evidence suggests that patients with PSC may benefit from a high dose >20 mg/kg/day UDCA. Similarly, the management of symptomatic cholestasis, namely pruritus, is the same for all, and appropriate management and prevention of osteoporosis does not vary according to the cause of cholestasis.
SEQUENTIAL PBC AND AIH There are a few individual case reports of patients who were first given a diagnosis of AMA-positive PBC with typical biochemical and histologic features that responded to UDCA treatment, and who subsequently had a complete change in their symptomatology, biochemistry and histology, changing to autoimmune hepatitis, even with loss of AMA.43,44 Only when treatment with immunosuppressive therapy was introduced were the features of the AIH adequately controlled. These cases are so rare that no definitive
860
conclusions can be drawn, except to suggest that the immune response governed by the individuals’ genetic make-up may evolve in response to ongoing exposure to inciting antigen(s).
DIFFERENTIAL DIAGNOSIS IN OVERLAPPING SYNDROMES In any patients with a chronic liver disease it is important to consider numerous possible explanations for ‘atypical’ features before making a diagnosis of an overlapping syndrome. In the case of cholestasis in AIH, a common cause is the introduction of exogenous hormones such as the contraceptive pill, hormone replacement therapy or testosterone. Treatment with hormones may induce symptomatic or asymptomatic cholestasis, but importantly without duct injury. Duct injury but not duct loss is reported to be present in 30% of cases of otherwise typical AIH.45 True duct loss may occasionally be the result of a drug reaction, most commonly to antibiotics.46 Although patients with a background of liver disease are not known to be at increased risk of such untoward drug reactions, all patients with liver disease are more susceptible to the cholestatic effect of estrogen therapy simply because the liver disease itself impairs canalicular transport of bile, and in addition the effect of estrogen on bile flow promotes even more cholestasis. In individuals with PSC it is not unusual to see fluctuating levels of serum aminotransferases. This may be related to common duct stones or to the presence of sludge. Thus overlap with AIH should not be considered until the appropriate investigations have been performed and a longer-term perspective is available.
SUMMARY Despite ongoing rigorous investigation into the pathogenesis of the autoimmune liver diseases, their etiology remains obscure. The pathogenesis is clearly multifactorial, with both exogenous influences and genetics playing a role. There are so many potential targets within the liver, and thus it is not surprising that mixed pictures are sometimes observed. The rarity and the heterogeneous nature of overlap syndromes unfortunately means that there are no clear diagnostic criteria and little or no evidence on which to recommend appropriate management.
REFERENCES 1. Lenzi M, Bellentani S, Saccoccio G, et al. Prevalence of nonorgan-specific autoantibodies and chronic liver disease in the general population: a nested case–control study of the Dionysos cohort. Gut 1999;45:435–441. 2. Hooper B, Whittingham S, Mathews JD, et al. Autoimmunity in a rural community. Clin Exp Immunol 1972;12:79–87. 3. Shibata M, Momizne T, Tanaka A, et al. A case of autoimmune hepatitis with a high titre of antimitochondrial antibody and normal gamma-globulinemia. J Gastroenterol Hepatol 2001;16:830–831. 4. Wilschanski M, Chait P, Wade JA, et al. Primary sclerosing cholangitis in 32 children: clinical, laboratory, and radiographic features, with survival analysis. Hepatology 1995;22:1415–1422. 5. Gregorio GV, Portmann B, Karani J, et al. Autoimmune hepatitis/sclerosing cholangitis overlap syndrome in childhood: A 16 year prospective study. Hepatology 2001;33:544–553.
Chapter 43 OVERLAP SYNDROMES
6. Gohlke F, Lohse AW, Dienes HP, et al. Evidence for an overlap syndrome of autoimmune hepatitis and primary sclerosing cholangitis. J Hepatol 1996;24:699–705. 7. Luketic VAC, Gomez AG, Sanyal AJ, Shiffman ML. An atypical presentation for primary sclerosing cholangitis. Dig Dis Sci 1997;42:2009–2016. 8. Hatzis GS, Vassilious VA, Delladetsima JK. Overlap syndrome of primary sclerosing cholangitis and autoimmune hepatitis. Eur J Gastroenterol Hepatol 2001;13:203–206. 9. Abdo A, Bain VG, Kichian K, Lee SS. Evolution of autoimmune hepatitis to primary sclerosing cholangitis: a sequential syndrome. Hepatology 2002;36:1393–1399. 10. Goldstein NS, Soman A, Gordon SC. Portal tract eosinophils and hepatocyte cytokeratin 7 immunoreactivity helps distinguish early-stage, mildly active primary biliary cirrhosis and autoimmune hepatitis. Am J Clin Pathol 2001;116:846–853. 11. Talwalkar JA, Angulo P, Johnson CD, et al. Cost-minimization analysis of MRC versus ERCP for the diagnosis of primary sclerosing cholangitis. Hepatology 2004;40:39–45. 12. Lindor KD for the Mayo PSC–Ursodeoxycholic Acid Study Group. Ursodiol for primary sclerosing cholangitis. N Engl J Med 1997;336:691–695. 13. Mitchell SA, Bansi DS, Hunt N, et al. A preliminary trial of high-dose ursodeoxycholic acid in primary sclerosing cholangitis. Gastroenterology 2001;121:900–907. 14. Harnois DM, Angulo P, Jorgensen RA, et al. High-dose ursodeoxycholic acid as a therapy for patients with primary sclerosing cholangitis. Am J Gastroenterol 2001;96:1558–1562. 15. Guanabens N, Pares A, Ros I, et al. Alendronate is more effective than etidronate for increasing bone mass in osteopenic patients with primary biliary cirrhosis. Am J Gastroenterol 2003;98:2268–2274. 16. Adachi JD, Bensen WG, Brown J, et al. Intermittent etidronate therapy to prevent corticosteroid-induced osteoporosis. N Engl J Med 1997;337:382–387. 17. Geubel AP, Baggenstoss AH, Summerskill WHJ. Responses to treatment can differentiate chronic active liver disease with cholangitic features from the primary biliary cirrhosis syndrome. Gastroenterology 1976;71:444–449. 18. Zein CO, Angulo P, Lindor KD. When is liver biopsy needed in the diagnosis of primary biliary cirrhosis? Clin Gastroenterol Hepatol 2003;1:89–95. 19. Chazoullier O, Wendum D, Serfaty L, et al. Primary biliary cirrhosis – autoimmune hepatitis overlap syndrome: clinical features and response to therapy. Hepatology 1998;2002:296–301. 20. Talwalkar JA, Keach JC, Angulo P, Lindor KD. Overlap of autoimmune hepatitis and primary biliary cirrhosis: an evaluation of a modified scoring system. Am J Gastroenterol 2002;97:1191–1197. 21. Alverez F, Berg PA, Bianchi FB, et al. International Autoimmune Hepatitis Group Report: review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999;31:929–938. 22. Walker JG, Doniach D, Roitt M, et al. Serological tests in diagnosis of primary biliary cirrhosis. Lancet 1965;1:827. 23. Kenny RP, Czaja AJ, Ludwig J, Dickson ER. Frequency and significance of automitochondrial antibodies in severe chronic active hepatitis. Dig Dis Sci 1986;31:705–711. 24. Lohse AW, Meyer zum Buschenfelde KH, Franz B, et al. Characterization of the overlap syndrome of primary biliary cirrhosis (PBC) and autoimmune hepatitis: evidence for it being a hepatitic form of PBC in genetically susceptible individuals. Hepatology 1999;29:1078–1084. 25. Brunner G, Klinge O. A chronic destructive non-suppurative cholangitis-like disease picture with antinuclear antibodies (immunocholangitis). Dtsch Med Wochenschr 1987;112:1454–1458. 26. Michieletti P, Wanless IR, Katz A, et al. Antimitochondrial antibody negative primary biliary cirrhosis: a distinct syndrome of autoimmune cholangitis. Gut 1994;35:260–265.
27. Invernizzi P, Crosignani A, Battezzati PM, et al. Comparison of the clinical features and clinical course of antimitochondrial antibody-positive and -negative primary biliary cirrhosis. Hepatology 1997;25:1090–1095. 28. Kim WR, Poterucha JJ, Jorgensen RA, et al. Does antimitochondrial antibody status affect response to treatment in patients with primary biliary cirrhosis? Outcomes of ursodeoxycholic acid therapy and liver transplantation. Hepatology 1997;26:22–26. 29. Stone J, Wade JA, Cauch-Dudek K, et al. Human leukocyte antigen class II associations in serum antimitochondrial antibodies (AMA)-positive and AMA negative primary biliary cirrhosis. J Hepatol 2002;36:8–13. 30. Boberg KM, Fausa O, Haaland T, et al. Features of autoimmune hepatitis in primary sclerosing cholangitis: an evaluation of 114 primary sclerosing cholangitis patients according to a scoring system for the diagnosis of autoimmune hepatitis. Hepatology 1996;23:1369–1376. 31. Kaya M, Angulo P, Lindor KD. Overlap of autoimmune hepatitis and primary sclerosing cholangitis: an evaluation of a modified scoring system. J Hepatol 2000;33:537–542. 32. van Buuren HR, van Hoogstraten HJF, Terkivatan T, et al. High prevalence of autoimmune hepatitis among patients with primary sclerosing cholangitis. J Hepatol 2000;33:543–548. 33. Burak KW, Urbanski SJ, Swain MG. A case of coexisting primary biliary cirrhosis and primary sclerosing cholangitis: a new overlap of autoimmune liver diseases. Dig Dis Sci 2001;46:2043–2047. 34. Mitchison HC, Bassendine MF, Hendrick A, et al. Positive antimitochondrial antibody but normal alkaline phosphase: is this primary biliary cirrhosis? Hepatology 1986;6:1279–1284. 35. Czaja AJ, Freese DK. Diagnosis and treatment of autoimmune hepatitis. Hepatology 2002;36:479–497. 36. Heathcote EJ. Management of primary biliary cirrhosis. Hepatology 2000;31:1005–1013. 37. Pares A, Caballeria L, Rodes J. Long-term ursodeoxycholic acid treatment delays progression of mild primary biliary cirrhosis. J Hepatol 2001;34(Suppl 1):187–188. 38. Gluud C, Christensen E. Ursodeoxycholic acid for primary biliary cirrhosis (Review). Cochrane Database Syst Rev 2004 Vol. 2. 39. Poupon RE, Lindor KD, Cauch-Dudek K, et al. Combined analysis of randomized controlled trials of ursodeoxycholic acid in primary biliary cirrhosis. Gastroenterology 1997;113:884–890. 40. Corpechot C, Carrat F, Poupon R, Poupon RE. Primary biliary cirrhosis: incidence and predictive factors of cirrhosis development in ursodiol-treated patients. Gastroenterology 2002;122:652–658. 41. Joshi S, Cauch-Dudek K, Wanless IR, et al. Primary biliary cirrhosis with additional features of autoimmune hepatitis: response to therapy with ursodeoxycholic acid. Hepatology 2002;35:409–413. 42. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: mechanisms of action and therapeutic use revisited. Hepatology 2002;36:525–531. 43. Colombato LA, Alvarez F, Cote J, Huet PM. Autoimmune cholangiopathy: the result of consecutive primary biliary cirrhosis and autoimmune hepatitis? Gastroenterology 1994;107:1839–1843. 44. Weyman RL, Voigt M. Consecutive occurrence in primary biliary cirrhosis and autoimmune hepatitis: a case report and a review of the literature. Am J Gastroenterol 2001;96:585–587. 45. Czaja AJ, Carpenter HA. Autoimmune hepatitis with incidental histologic features of bile duct injury. Hepatology 2001;34: 659–665. 46. Al Traif I, Lilly L, Wanless IR, Heathcote J. Cholestatic liver disease with ductopenia (vanishing bile duct syndrome) following clindamycin and trimethoprim–sulfamethoxazole administration. Am J Gastroenterol 1994;89:1230–1234.
861
Section VI. Immune Diseases
GRAFT-VERSUS-HOST DISEASE AND THE LIVER
44
Daniel Shouval and Oren Shibolet Abbreviations aGvHD acute GvHD APC antigen presenting cells BMT bone marrow transplantation cGvHD chronic GvHD CMV cytomegalovirus GI gastrointestinal GvHD graft-versus-host disease GvL graft-versus-leukemia
HCV HLA HPCT IL-1 ICAM-1 LFA-1 MHC
hepatitis C virus human leukocyte antigens hematopoietic progenitor cell transplantation interleukin-1 intracellular adhesion molecule-1 alb2 integrin major histocompatibility complex
INTRODUCTION Bone marrow transplantation (BMT) and peripheral blood stem cell transplantation (PBSCT) may induce cholestatic as well as hepatocellular liver injury. Hepatic injury in these conditions may be the result of veno-occlusive disease (VOD), hepatotoxic chemotherapy, irradiation and infection, or it may be caused by graft-versus-host disease (GvHD). GvHD is an immunologically mediated clinical syndrome induced by immunocompetent cells from an organ donor against human leukocyte antigens (HLA) of the recipient. It may involve a single or multiple organs.1 It was first observed 50 years ago, when Barnes and colleagues recognized a wasting syndrome in mice following allogeneic BMT (“Runt disease”).2 Ten years later, Billingham defined the immunological requirements for the development of GvHD.3 These include: (1) the graft must contain immune competent cells; (2) the graft must recognize the host as non-self; (3) the recipient must be unable to mount an effective immune response against the graft; (4) the graft must survive in the host long enough to become sensitized and mount an immunological response against the host. If these requirements are met, donor lymphocytes may attack recipient organs, causing tissue injury, and resulting in significant morbidity and mortality. GvHD most commonly occurs following allogeneic BMT or hematopoietic progenitor cell transplantation (HPCT) for hematological malignancies4 and seldom following solid organ transplantation.5 It is rarely reported following blood transfusion in immune-competent patients.6 Sporadic cases were reported in patients receiving immune suppression for other causes.7 GvHD has two distinct forms, acute and chronic, which differ in their time of onset (with some overlap), organ involvement, and presentation. The disease involves the skin, liver, gastrointestinal (GI) tract, and the hematopoietic and immune systems, with epithelial injury being the most common histopathological manifestation. GvHD is graded as mild (grade I and II), moderate (grade III), and severe (grade IV) according to the severity of organ involve-
PBSCT TCR TOR TNF-a VCAM-1 VLA-4 VOD
peripheral blood stem cell transplantation T-cell receptor target of rapamycin molecule tumor necrosis factor-a vascular cell adhesion molecule-1 a4b1 integrin veno-occlusive disease
ment, with mortality rates ranging between <10% and 90% respectively.8 GvHD may have a beneficial role in the host’s response against tumors, designated as graft-versus-leukemia (or tumor) (GvL) effect, with an inverse correlation between GvHD occurrence and leukemia recurrence.9 Due to the immune nature of the disease, treatment of GvHD involves immunosuppressive medications. Mortality is commonly the consequence of treatment adverse effects or infection and not due to organ failure. In this chapter we will describe the pathogenesis, immunology, clinical manifestations, treatment, and prognosis of the two forms of GvHD, with a focus on liver involvement.
EPIDEMIOLOGY The incidence of acute GvHD (aGvHD) varies from less then 10% to more then 80% in recipients of hematopoietic cells, depending on HLA disparity between donor and recipient, the pretransplantation induction regimen, the amount of competent T lymphocytes in the graft, and the patient’s age.10 There is an unresolved controversy concerning the incidence of aGvHD following PBSCT or BMT, with reports suggesting higher or similar incidence respectively.11,12 Chronic GvHD (cGvHD) has a variable incidence of 10–80%.13 The incidence of cGVHD seems to increase in patients with PBSCT when compared to BMT.14 Hepatic involvement in GvHD occurs in up to 70% of patients, with isolated hepatic involvement occurring in approximately 20%.15 The incidence of GvHD following liver transplantation is unknown but seems to be around 10–30%.16,17
IMMUNOPATHOGENESIS HPCT involves ablation of the patient’s original hematopoietic system and replacing it with the donor’s hematopoietic system containing immune-competent cells. Following engraftment, donor-
863
Section VI. Immune Diseases
derived T cells can react against recipient-displayed antigens leading to GvHD. The major histocompatibility complex (MHC) antigens play a central role in the pathogenesis of GvHD. MHC antigens are encoded by the MHC gene locus, which is a closely linked highly polymorphic multigene and multiallelic complex located on the short arm of chromosome 6. The antigens encoded by these gene loci form distinct groups, known as the HLA. There are two classes of HLA, each comprised of two chains. Class I is made up of a heavy chain, and a light chain (b2-microglobulin), and is further subdivided into groups A, B, and C. Class II contains two chains, and is comprised of groups DR, DQ, and DP. Both classes are cell surface molecules responsible for T-lymphocyte recognition and histocompatibility. HLA class I antigens are found on all nucleated cells while class II antigens are mostly restricted to immune cells such as lymphocytes. During HPCT efforts must be made to minimize HLA disparity between donor and recipient in order to improve engraftment and lessen the severity of GvHD. Minor histocompatibility antigens are epitopes from intracellular proteins presented together with MHC molecules to donor T cells. The presentation of these antigens to T cells causes their activation, leading to GvHD.18,19 There are three distinct phases to aGvHD. Phase I consists of tissue injury during the conditioning regimen, even prior to infusion of donor cells. These regimens are toxic to rapidly dividing cells such as the intestinal, biliary epithelia, and skin. The damaged tissues release multiple proinflammatory cytokines such as interleukin-1 (IL-1), interferon-a, and tumor necrosis factor-a (TNF-a), causing up-regulated presentation of adhesion molecules and enhancement of recipient MHC antigens. This increased presentation of tissue antigens augments recognition of recipient antigens by donor lymphocytes. It was previously shown that intensive conditioning regimens, i.e., high-dose chemotherapy or total-body irradiation, increase the risk of GvHD.20 Phase II of aGvHD consists of recognition of recipient antigens as foreign by donor T cells. Following transplantation, host peptides are ingested and processed by antigen-presenting cells (APC) and presented to T lymphocytes. Donor CD4+T cells induce GvHD to antigens presented via the MHC class II system, while CD8+T cells induce GvHD to antigens presented via the MHC class I system. The specific characteristics of the HLA disparity between donor and host have a major impact on the severity of the GvH reaction. Studies show that a single different amino acid at an HLA site may induce GvHD, while larger disparities in other antigens may go unnoticed.21 Adhesion molecules play a major role in controlling the contact between T cells and their targets. CD62E (E-selectin), a4b1 integrin (VLA-4), alb2 integrin (LFA-1), intracellular adhesion molecule-1 (ICAM-1), and vascular cell adhesion molecule-1 (VCAM-1) are some of the adhesion molecules involved in the process.22 Following adhesion, T cells are activated, through interactions between the T-cell receptor (TCR) and MHC peptides and co-stimulatory signals (such as CD80, CD86, CD40/cd40L) provided by APCs. The interactions can cause T-cell activation but may also induce Tcell hyporesponsiveness. The direction of the response depends on the complicated interplay between TCR-binding and the effect of stimulatory or inhibitory signals.23 Phase III of aGvHD consists of the response phase when tissues are damaged. It was suggested that TH1 lymphocytes, which
864
produce IL-2 and interferon-g, could lead to activation of cytotoxic lymphocytes, thereby promoting GvHD, and that TH2 cells, which secrete IL-4 and IL-10, prevent it.24 Experimental data from various models suggest that this paradigm may be too simplistic, showing that both TH1 and TH2 cells are important for the development of liver GvHD.25–27 During the third phase of aGvHD there is a marked increase in proinflammatory cytokine production, which may lead to a “cytokine storm.”28 Cytokines taking part in this phase include IL-1 and TNF-a. It has been suggested that lipopolysaccharide-induced TNF-a secretion occurs following damage to mucosal tissue and pathogen invasion. IL-2, IL-6, IL-8, and IL-11 serum levels also increase; however; the exact role of these cytokines is still unknown.29 The cellular component of injury in aGvHD is induced by the cytotoxic action of effector cells such as natural killer cells that induce apoptosis via the perforin and the FAS/FAS-ligand pathways.30 Inhibition of both of these pathways was shown to delay the onset and severity of GvHD but not to prevent it, suggesting that perforin/FAS independent pathways also exist.31,32 Data suggest that FasL is essential for the induction of hepatic GvHD.33 In contrast to the well-established immunopathogenesis of aGvHD, less is known about the pathogenesis of cGvHD. It is suggested that target cell injury is mediated by autoimmune donor T cells, which recognize antigens shared by donor and host.34 Activation of these cells causes an autoimmune-like disease resembling the histopathologic manifestations of primary biliary cirrhosis and autoimmune cholangiopathy.35,36 Immune-regulatory T cells such as CD4+CD25+T cells and NKT lymphocytes were shown in animal models to have an important role in inhibiting GvHD. It is postulated that Th2 cytokines play a role in cGvHD. Elevations of IL-5 serum levels with eosinophils and IL-4 with gammopathy have been observed.37 Damage to the thymus during aGvHD may cause immune dysregulation with an inability to delete autoreactive T cells.38 There is also evidence of B-cell dysregulation leading to a high prevalence of autoimmune autoantibodies to cell surface and intracellular antigens.39 In addition to the immune dysregulation of cGvHD there is marked immune deficiency because of T- and B-cell dysfunction and reduced production.40 A different mechanism underlies the development of transfusionassociated GvHD, a disease that is almost invariably fatal but is beyond the scope of this chapter.
CLINICAL MANIFESTATIONS OF GRAFT-VERSUS-HOST DISEASE GvHD is clinically divided into aGvHD and cGvHD. aGvHD occurs within 100 days of BMT, while cGvHD occurs after 100 days post-transplantation. This division is arbitrary and is not always easily distinguishable (For a brief comparison between aGvHD and cGvHD, see Table 44-1). Hyperacute GvHD is a severe form of aGvHD, occurring within the first week post-transplantation. It usually manifests as fever, diffuse skin rash, severe hepatitis, inflammation, and capillary leak. The syndrome is almost always fatal but
Chapter 44 GRAFT-VERSUS-HOST DISEASE AND THE LIVER
Table 44-1. Comparison of acute and chronic graft versus host disease Acute GvHD
Chronic GvHD
Onset Clinical manifestations
<100 days of BMT Biliary/hepatic manifestations, skin rash, diarrhea, thrombocytopenia
incidence Cytokine pattern Risk factors
10%–80% TH1 (IFNg, IL-12, TNFa) HLA disparity, increasing age, underlying disease, intensity of conditioning regimen. Corticosteroids, methotrexate, ursodecholic acid, (UDCA) ciclosporin, tacrolimus, OKT3, ATG, anti-TNF ~30% of patients with moderate to severe aGVHD achieve cure, about 70% progress to cGVHD
>100 days following BMT Autoimmune manifestations i.e. sclerosing cholangitis and biliary cirrhosis; sicca syndrome; scleroderma; pulmonary disease (BOOP); esophagitis with stricture; immunoglobulin deficiency 10%–80% TH2 (IL-4, IL-5) HLA disparity, prior acute GVHD, use of non-T-cell depleted marrow, female to male donation Corticosteroids, Thalidomide (increased incidence of chronic GvHD when used early following BMT), hydroxychloroquine, tacrolimus, UDCA, photophoresis ~60% of the patients with more extensive disease respond to treatment. The remaining patients have a poor prognosis and will either die from infections or organ failure or will need prolonged immunosuppressive treatment
Treatment (partial list)
Prognosis
is rare.41,42 An acute GvH reaction may occur within 2–3 weeks following organ transplantation in the presence of ABO blood group mismatch. Organs involved in aGvHD include liver, skin, GI tract and mucosal surfaces, hematopoietic system, lacrimal and salivary glands, and bronchioles.
LIVER INVOLVEMENT IN GVHD The liver is the second most commonly affected organ in aGvHD. Initially patients complain of increasing fatigue, anorexia, dark urine, and acholic stools. Pruritus may result from liver or skin involvement or both. The earliest clinical signs are the development of jaundice with increasing levels of conjugated bilirubin and alkaline phosphatase.43,44 Currently two distinct forms of hepatic aGvHD are recognized in association with BMT, the classical and the hepatitic forms. In the classical form, abnormal liver function tests include increased serum alkaline phosphatase, g-glutamyltranspeptidase and bilirubin levels accompanied by mild to moderate hepatomegaly. Increase in aminotransferase levels is usually delayed and mild. The hepatitic form is characterized by significant hepatocellular injury with a more than 10-fold increase in alanine aminotransferase and aspartate aminotransferase levels.45–48 Usually both skin and GI manifestations are apparent by the time liver involvement is observed; however isolated hepatic involvement may be the presenting feature.49 Clinical features of chronic liver disease, such as hepatic encephalopathy, ascites, coagulopathy, and hypoalbuminemia, are uncommon during aGvHD and their presence signifies advanced and sometimes terminal disease in patients with end-stage GvHD. During cGvHD similar manifestations occur. cGvHD often mimics autoimmune liver disease, leading to fibrosis of the biliary tree, sclerosing cholangitis, or biliary cirrhosis.50,51 During this phase, signs of chronic liver disease may develop, such as ascites, and variceal hemorrhage.44,52,53 A plethora of autoantibodies have been described in both acute and cGvHD, including antinuclear (ANA), anti-smooth-muscle (ASMA), antimitochondria, and antiliver
kidney microsomal (anti-LKM) antibodies.54 In fact, GvHD shares several similarities mainly with PBC, but also with PSC. These include the clinical manifestations of cholestatic liver disease as a result of injury to the biliary tree, scleroderma-like skin lesions, involvement of the parotid and lacrimal glands leading to sicca syndrome, pancreatic insufficiency, gastrointestinal injury, Raynaud phenomenon, and pulmonary involvement.
HISTOPATHOLOGY OF HEPATIC GvHD Acute GvHD can present morphologically as hepatitis, resembling drug or viral induced damage. More specific for the diagnosis is portal inflammation with lymphocytic infiltration, mostly into small caliber bile ducts, causing degenerative changes and sloughing of the epithelial cells. Endothelial damage manifesting as endotheliatis is an uncommon finding. Prominent histopathological findings in the classical form of aGvHD include bile duct injury and portal lymphocytic infiltration, while hepatocytes are relatively preserved. Intrahepatic cholestasis is significantly more common in this form55–57. The hepatitic form is characterized by lobular hepatitis and the presence of acidophilic bodies, but with relatively mild bile duct injury. Chronic GvHD is characterized mainly by portal inflammation and injury to interlobular bile ducts, which show more prominent abnormalities than in the acute form. Common histopathologic manifestations include inflammatory infiltration of the biliary epithelium by mononuclear cells and focal hepatocyte apoptosis. This process of progressive non-suppurative cholangitis leads to gradual disappearance of interlobular bile ducts. Loss of bile ducts is a relatively late phenomenon and may lead to vanishing bile duct syndrome or rarely to fibrosis and cirrhosis35. Lobular changes other than canalicular cholestasis are typically mild although an unusual form of hepatitic cGvHD has been described with significant parenchymal damage resembling viral hepatitis. A rare histopathological finding is an autoimmune-like variant with marked lobular and intraportal plasma cell infiltration.58 The hepatic pathologic manifestations and especially the involvement of the biliary epithelium are not specific to GvHD and
865
Section VI. Immune Diseases
are rather similar to other clinical situations such as primary biliary cirrhosis, primary sclerosing cholangitis, hepatitis C virus (HCV), liver rejection or cytomegalovirus (CMV) infection. (Table 44-2 summarizes the differences between several of these common etiologies). The role of liver biopsy in diagnosing hepatic GvHD is controversial. Although important in differentiating GvHD from vascular, viral, or drug-associated liver injury, iron overload, and other forms of hepatitis, it still poses a small but significant risk. It is however doubtful whether outcome of clinical end-points are significantly altered by the results obtained through a liver biopsy. If a liver biopsy is required, transjugular instead of transcutaneous liver biopsy may be considered in these sick patients with anemia and marked thrombocytopenia.43,59,60
SKIN Skin involvement is the most common manifestation in GvHD. It usually presents as a maculopapular erythematous rash involving the neck, ears, shoulders, and palms of the hands and soles of the feet. The rash may spread to involve the whole body, resembling scleroderma. Histological examination usually reveals perivascular lymphocytic infiltration, dyskeratotic epidermal keratinocytes, exocytosed lymphocytes, and individual cell death with apoptosis at the base of the crypts.61
GASTROINTESTINAL TRACT Clinical presentation is dependent on the area of involvement. Upper GI involvement manifests as nausea, vomiting, dyspepsia, and anorexia due to esophageal stricture while lower GI involvement is characterized by diarrhea and abdominal pain. The involvement can be severe, with voluminous diarrhea, reaching 10 liters a day. The diarrhea is usually bloody, leading to severe anemia. Abdominal pain can also vary in severity from mild cramping to severe abdominal pain and ileus. Histological examination reveals
crypt cell necrosis with cryptitis. In severe disease there is denudation of whole areas of the epithelium.62–65
HEMATOPOIETIC SYSTEM The effect of aGvHD on the hematopoietic system usually manifests as thrombocytopenia, and a profound decrease in serum concentration of immunoglobulins. In transfusion-associated GvHD bone marrow aplasia is common.
STAGING Staging of aGvHD is usually done separately for each organ system involved. The evaluation is later combined to enable overall assessment of the clinical stage.66,67 In the liver, the disease is staged according to bilirubin levels, with stage 1-bilirubin levels at 2– 3 mg/dl, stage 2, 3–6 mg/dl, stage 3, 6–15 mg/dl, and stage 4 with bilirubin levels greater then 15 mg/dl. In skin, stage 1 involves a rash covering less than 25% of the body surface, stage 2 an area between 25 and 50%, and stage 3 generalized erythema. Stage 4 describes diffuse erythema with bulla formation and scleroderma-like changes. GI involvement is measured according to volume of diarrhea, with stage 1 being 500–1000 ml/day, stage 2, 1000– 1500 ml/day, in stage 3 the volume of diarrhea is greater than 1500 ml/day, and in stage 4 there is ileus and bleeding. This overall staging system proposed by Glucksberg et al.66 is based on combining the above scores but may lack sensitivity. A newer system has been proposed by the international BMT registry, grouping patients with patterns of organ involvement associated with treatment complications or failure.67 Other grading systems also exist.68 The current common grading system for cGvHD divides patients into those needing treatment as having “extensive” disease and those who do not need treatment as having “limited” involvement.55 Lately a new system has been proposed grading patients as having favorable, intermediate- or high-risk outcome based on extensive skin involvement, thrombocytopenia, and gradual onset.69
Table 44-2. The differential diagnosis of cGvHD with PBC and chronic hepatitis C
866
cGvHD
PBC
Chronic HCV
Gender Time/age of onset Liver function test pattern
Increased risk in female to male donation. >100 days Following BMT Elevated Alkaline phosphatase, gamma glutamyl transpeptidase and bilirubin levels. Mild elevation of ALT and AST
M>F Highest incidence, 4–5th decade Carriers may develope fluctuating serum ALT levels; One third of patients have persistently normal ALT.
Auto-antibodies
AMA; rheumatoid factor; ASMA (25%), anti-thyroid, ANA (22%).
Histological findings
Sparing of large bile ducts, absence of granulomas, progression to cirrhosis rare.
F>M 4–5th decade Elevated Alkaline phosphatase, gamma-glutamyl transpeptidase, mild elevated AST, ALT. Bilirubin gradual increases over years AMA (95%). Other autoantibodies less frequent: ASMA, anti-thyroid antibodies, ANA. Epithelial injury to small bile ducts; ductopenia; epitheloid granuloma; progression to cirrhosis
Treatment
Immunosuppressive agents (see text); T cell depletion and anti-T-cell antibodies
No effective treatment, UDCA may delay progression.
Mixed cryoglobulins (30–50%). ANA (20%), ASMA (20%), anti-LKM (5%). Common-Steatosis, lymphoid aggregates, bile duct damage. Range-active hepatitis, bridging fibrosis, hepatocyte necrosis, cirrhosis Pegylated interferon alpha and Ribavirin.
Chapter 44 GRAFT-VERSUS-HOST DISEASE AND THE LIVER
DIFFERENTIAL DIAGNOSIS Hepatic injury following BMT or HPCT is common.43,44,46,48,56 Some of the various characteristics of a GvHD and a cGvHD are shown in Table 44-1. Furthermore, cGvHD may occasionally present with clinical, laboratory or histological features resembling PBC or chronic hepatitis C infection as shown in Table 44-2. However, a variety of etiological factors other then GvHD may contribute to post-transplantation liver damage and mandates a careful and systematic approach to the differential diagnosis in these patients.70,71 One such practical approach is to consider etiologies based on the time interval following transplantation.
EARLY HEPATIC COMPLICATIONS FOLLOWING TRANSPLANTATION NON-INFECTIOUS CAUSES OF LIVER DYSFUNCTION FOLLOWING HPCT Veno-occlusive disease (VOD) (see Chapter 46) This is a life-threatening complication of BMT, manifesting as jaundice, tender hepatomegaly, and ascites. It is thought to be secondary to injury to the hepatic sinusoidal epithelial cells, leading to central vein occlusion. The incidence varies from 4% to 50%, depending on the chemotherapy regimen used and genetic susceptibility. The disease usually occurs prior to day 20 posttransplantation but approximately 25% of cases occur later.72 The diagnosis is usually based on clinical signs and symptoms. Imaging modalities may be employed to rule out infiltrative lesions and serve to visualize the hepatic veins as well as the biliary tree. Although Doppler ultrasound examination may occasionally demonstrate portal flow reversal, it is an insensitive modality for diagnosis of VOD.73 Recently it was suggested that measurement of the hepatic artery resistance index might be a more sensitive marker for VOD.74 Liver biopsy plays an important role in establishing the diagnosis. Treatment is based on either antithrombotic or thrombolytic agents (such as tissue plasminogen activator, heparin, defibrotide, and antithrombin III), with a marginal success rate,75 or mechanical relief of obstruction with transjugular portosystemic shunt.76 Liver transplantation was successfully performed in a few selected patients with VOD.77–78
EFFECTS OF HEPATOTOXIC AGENTS Several of the drugs employed in the setting of BMT may cause liver function disturbances. These include antibiotics (i.e., clavulinic acid, trimethoprim-sulfamethoxazole and antifungal agents), chemotherapeutic agents (azathioprine, mycophenolate mofetil, and ciclosporin) and miscellaneous other agents. Total parental nutrition may also lead to cholestatic dysfunction, either by inducing steatosis or by biliary obstruction secondary to sludge formation.
BILIARY DISEASE Cholestatic liver dysfunction may be the presenting manifestation of malignancy recurrence. Acalculous cholecystitis may occur following organ transplantation.79
SYSTEMIC INFECTIONS Infections are common following BMT and HPCT.80–82 Chronic infections with HBV and HCV are the most important infectious agents to consider in diagnosing disturbed liver functions in long-term survivors of BMT.
Hepatitis B Virus (HBV) Infection Mild to severe hepatitis may develop in patients chronically infected with HBV who undergo BMT, regardless of whether they have been asymptomatic carriers or already had active liver disease at the time of transplantation. Disease manifestations can range from mild abnormal cholestatic liver dysfunction through fibrosing cholestatic hepatitis to fulminant hepatitis and death.83 In hepatitis B surface antigen carriers, corticosteroid treatment may enhance viral replication leading to HBV reactivation and hepatitis, which usually occurs during the immune reconstitution phase within 60 days post-transplantations.84 Diagnosis is based on serum HBV-DNA determination as well as immunohistochemistry staining of HBcAg in liver biopsies. Treatment of reactivation may include nucleoside analogues such as lamivudine and adefovir dipavoxil. Evidence has recently been provided justifying pre-emptive nucleoside analogue treatment prior to organ transplantation in patients at risk in order to avoid HBV reactivation. However, the length of posttransplantation treatment is not established.85 Furthermore recipient of nucleoside (tide) analogues and in particular of lamivudine may develop viral escape mutants (i.e., YMDD). HPCT recipients are also at risk of contracting de novo HBV infection. Therefore, pretransplantation active immunization against hepatitis B is highly recommended. There is evidence to suggest that immunity to HBV in bone marrow donors, regardless of whether acquired through natural infection followed by recovery and immunity or acquired through active immunization, may lead to protection of organ transplant recipients against HBV infection.86–88 However, GvHD may abrogate immune memory leading to loss of immunity to HBV following transplantation.88
Hepatitis C Virus (HCV) Infection A mild increase in aminotransferases may be observed in HCVinfected patients following BMT. Acute hepatitis C infection may also rarely occur.89,90 Recent data suggest that fibrosis and cirrhosis are important survival risk factors in HCV-infected BMT patients.81,82 Diagnosis is based on serological and nucleic acid viral markers and occasionally on liver histology. Treatment is currently unsatisfactory as the only partially effective therapy is combined interferon and ribavirin treatment. Administration of interferon following BMT is associated with an increased risk of GvHD.91
Cytomegalovirus (CMV) Infection CMV infection is common in BMT recipients. It usually occurs between 40 and 100 days post-transplantation but can occur later. Hepatic manifestations include fever, jaundice, and disturbed liver functions, but the disease can involve other organ systems (lungs, GI). Diagnosis is based on serological markers and viral nucleic acids, or biopsy of involved organs. Ganciclovir is the most effective treatment. Prophylaxis with this drug is sometimes used.92
867
Section VI. Immune Diseases
Other Viral Infections Adenovirus, hepatitis G virus, herpes simplex virus, human herpes virus-6, human parvovirus B-19, Epstein–Barr virus, and transfusiontransmitted virus have all been reported in BMT recipients. Some can cause severe fulminant hepatitis (herpes simplex virus) or post-transplantation-associated lymphoproliferative disease (Epstein–Barr virus), while the role of others (transfusion-transmitted viruses such as, hepatitis G virus) in inducing disease is currently unknown.93
POST-TRANSPLANTATION LYMPHOPROLIFERATIVE DISEASE Lymphoproliferative disease may affect the liver post-transplantation and is occasionally associated with Epstein–Barr virus infection. It may occur at any time following transplantation, but usually does so within the first 2 years. Reduction in immune suppression and donor T-lymphocyte infusion may result in disease regression. Diagnosis is based on Epstein–Barr virus DNA detection by polymerase chain reaction or detection of viral particles by immunostaining of liver biopsies.97
Fungal Infections Fungal infections usually occur in the first weeks post-transplantation, when immune suppression is most profound. Candida albicans is the most common infecting fungus, but invasive Aspergillus fumigatus is also an important infection in BMT recipients. A raised alkaline phosphatase level may be the earliest sign of infection. Imaging studies are helpful in detecting liver lesions. Diagnosis is based on liver biopsy and culture. Amphotericin, fluconazole, and newer antifungal agents such as caspofungin are useful in eradicating infection.94,95
Mycobacterial Infection Although rare, mycobacterical infection is significantly more prevalent in BMT recipients following transplantation in countries where tuberculosis is prevalent. Diagnosis is based on culture, polymerase chain reaction, imaging, and liver biopsy. Mycobacterial treatment is useful in eradicating the disease.96
Miscellaneous Bacterial Infections The liver may become involved during systemic infection. Diagnosis is usually based on clinical evidence and primary infection is rarely limited to the liver. Ascending cholangitis may be the primary source for septicemia, leading to cholestatic dysfunction early posttransplantation. Treatment is with antibiotics.
LATE POST-TRANSPLANTATION LIVER COMPLICATIONS IRON OVERLOAD Iron overload is a late complication of organ transplantation resulting from intensive transfusion treatment. Manifestations include hepatomegaly, hepatic fibrosis, and cirrhosis. Diagnosis is based on measurements of body iron stores and assessment of iron content. Treatment includes an aggressive phlebotomy regimen and chelation.53
RECURRENT DISEASE IN BMT RECIPIENTS TRANSPLANTED FOR HEMATOPOIETIC MALIGNANCY The original hematologic disorder may recur and invade the liver, as may be proven by a liver biopsy.
868
TREATMENT ACUTE GVHD Prevention and Treatment The development of moderate or severe GvHD is associated with significant morbidity and reduced survival. It was shown that, once GvHD develops, treatment modalities to prevent progression are limited.98 Therefore major efforts have been invested into preventing GvHD. The two basic strategies of GvHD prophylaxis are donor (graft) lymphocyte depletion and pharmacological therapy. T-lymphocyte depletion of the graft is an effective means of eliminating or reducing the risk of GvHD.99 However, most lymphocyte separation methods may cause loss of stem cells needed for successful engraftment, thereby reducing engraftment rates. Other problems encountered are increased leukemia recurrence rates and higher rates of infection.100 Initial trials of T-cell depletion used exvivo incubation of donor bone marrow with broadly reactive antilymphocyte antibodies such as anti-TCR antibodies (i.e., anti-CD3, anti-CD2, anti-CD5, anti-CD8 and anti-CD52 (Campath 1)). It was hoped that complement fixation combined with direct antibody-dependent cellular toxicity would cause donor T-cell destruction. However, most of these trials were unsuccessful.101 Newer methods include antibodies bound to toxins, antibodies conjugated to magnetic beads, lectins, agglutination, and E-rosette formation or addition of donor lymphocytes following recovery from the acute phase of induction regimen toxicity.21 Although these techniques are effective in considerably reducing the numbers of T cells in the graft, they are ineffective in improving disease-free survival because of increased leukemia recurrence rates and death secondary to infectious complications.102–104 Therefore, these treatments are usually reserved for high-risk patients (i.e., mismatched transplants).
Pharmacological Treatment Depending on clinical presentation and severity, treatment of GvHD may require a systemic as well as an organ directed approach. The most commonly used method for prevention and treatment of GvHD employs immunosuppressive agents. This arm of treatment acts either specifically or non-specifically to inhibit T cell activation. Non-specific immunosuppressive drugs. Corticosteroids have been the cornerstone of anti-GvHD treatment, often used in combination with other medications. Their mechanism of action in GvHD remains unknown, but may involve anti-inflammatory effects, direct lymphotoxic properties, and decreasing proinflam-
Chapter 44 GRAFT-VERSUS-HOST DISEASE AND THE LIVER
matory cytokines. Although widely used, a recent report showed no added105 or only limited benefit to corticosteroids when added to methotrexate for patients with moderate to severe GvHD.106 Trials assessing the use of newer steroids which undergo extensive first pass in the liver, such as budesonide,107 or directly administrating medications to the intestine and the liver via a catheter placed in the splanchnic vasculature are currently underway.108 Methotrexate. This drug is an antimetabolite analog of aminopterin. It interferes with critical metabolic pathways via its action as a folic acid antagonist. Methotrexate is used in GvHD patients in various combinations with steroids, ciclosporin, tacrolimus, sirolimus, sacrolimus, and mycophenolate mofetil. Combination therapy was shown to improve survival.109 Recently, the administration of lowdose methotrexate intra-arterially to the hepatic artery was shown to be beneficial for patients with severe hepatic GvHD.110 Side effects of methotrexate include bone marrow suppression and renal, hepatic, and GI toxicity. Long-term administration of the drug can cause dose-dependent hepatic fibrosis. This is an uncommon side effect in the setting of BMT because of the relatively short duration of administration and the low cumulative dose. Elevations of bilirubin and transaminases are commonly seen during methotrexate treatment. Thalidomide. Thalidomide is a glutamic acid derivative. The mechanism of action of the drug is unknown. It may act as an antiangiogenic, mild anti-TNF agent, or an inhibitor of lymphocyte–APC interactions. Various studies have shown decreased overall survival and increased incidence of cGvHD when the drug was initiated early after BMT. In contrast, its use in cGVHD in combination with steroids and other immunosuppressive drugs has been shown to relieve cGvHD effectively in up to 50% of patients. Major side effects of the drug include sedation and constipation, myelosuppression, skin rashes and ulceration, and toxic peripheral neuritis that may be irreversible.111 Azathioprine. This drug has not been specifically used as an antiGVHD medication, although it is mentioned in several immunosuppressive regimens. Topical azathioprine may be beneficial in the treatment of severe oral mucosal involvement of GvHD.112 Ursodeoxycholic acid (UDCA). Several reports have shown that the incidence and severity of aGvHD, especially grade III and IV liver and intestinal GvHD, was reduced with UDCA treatment. Administration of UDCA showed an improvement in hepatic enzyme serum levels in patients with hepatic GvHD.52 The most common side effect is GI irritation.113 UDCA may be beneficial in improving biochemical markers of cholestasis and pruritus, irrespective of the etiology.114,115 Hydroxychloroquine. This drug has been used in cGvHD with promising initial results.116
T-cell-specific Immunosuppressive Drugs Ciclosporin. Ciclosporin is currently the mainstay treatment for the prevention of GvHD, in combination with other immunosuppressive drugs.117,118 It is a fungal-derived hydrophobic peptide with potent immunosuppressive activity through inhibition of increased
expression of high affinity IL-2R by activated T-cells. It inhibits the intracellular activation molecule calcineurin, which is essential for T-cell activation. Major side effects include nephrotoxicity, elevated bilirubin levels, hypertension, and seizures. Rarely the development of a thrombocytopenic thrombotic purpura-like syndrome has been described. The drug is extensively metabolized and cleared mainly in the bile, with lesser amounts in the urine. It has multiple interactions with other drugs. Tacrolimus. Tacrolimus is a fungal macrolide antibiotic. Similarly to ciclosporin, it interferes with T-cell activation by inhibiting the calcineurin pathway. In clinical trials it reduced the occurrence of aGvHD, although overall survival was unchanged. There are no data to suggest that tacrolimus is more efficacious than ciclosporin; however, several reports suggest that tacrolimus can be used as a salvage drug for patients who continue to have progressive GvHD while being treated with ciclosporin. The side effect profile of this drug is very similar to the one described for ciclosporin.119 Rapamycin (sirolimus). Rapamycin is a fungal macrolide devoid of antibiotic activity. Its action is mediated via the target of rapamycin molecule (TOR), and its activation leads to arrested cell cycle maturation. It also inhibits several intracellular transduction pathways essential for T-cell activation. Preliminary studies with rapamycin have shown encouraging results in patients who were unresponsive to steroids. One side effect is GI irritation, with diarrhea and vomiting. Liver enzyme abnormalities, hyperlipidemia, and blood dyscrasias, including thrombocytopenia and neutropenia, have been reported.120 Mycophenolate mofetil. Mycophenolate mofetil is a fungal-derived morpholinethyl ester that possesses antibiotic, antifungal, and antiviral properties. It has a synergistic effect with ciclosporin and has been shown to enhance engraftment. Its main toxicity is myelosuppression and GI irritation.121 Antibodies. In view of the importance of T cells in inducing GvHD, several regimens using specific monoclonal or polyclonal antibodies have been assessed. Their use is currently under evaluation. Some molecules currently in use are described below: Muromonab-CD3 (OKT3). This is a mouse monoclonal antibody directed at the CD3 molecule present on T lymphocytes. Administration of the antibody causes activation and depletion of human T cells. Despite its anti-T-lymphocyte activity, results in clinical trials of this antibody in BMT patients and in preventing GvHD have been disappointing.122 Antithymocyte globulin. These are polyclonal antilymphocyte antibodies derived from various sources. Administration of antithymocyte globulin induced destruction of human lymphocytes. Several studies in BMT recipients have shown either no benefit or mild beneficial effect. The main side effects relate to lymphocyte destruction and include cytokine release, infections, and serum sickness.123 Intravenous immunoglobulin. These are polyclonal antibodies derived from human serum. Initial trials with intravenous immunoglobulin have shown promise in decreasing the incidence of GvHD, especially in younger patients (under 20 years of age), with
869
Section VI. Immune Diseases
a decrease in overall mortality. A recent trial has not found benefit for intravenous immunoglobulin over placebo in BMT recipients, with an increased incidence of VOD.124 Anti-TNF. TNF has been implicated in the pathogenesis of aGvHD. Several reports assessed the efficacy of a chimeric human–mouse anti-TNF agent (infliximab) in the treatment of aGvHD following failure of other conventional treatment modalities, with promising results.125 Anti-IL-2-receptor antibody (daclizumab). Daclizumab is a humanized monoclonal antibody against the IL-2 receptor expressed on activated T lymphocytes. Initial studies showed it to be well tolerated in patients with acute steroid-resistant GvHD. However, a recently published report suggested that the combination of daclizumab and steroids might decrease survival.126 Anti-co-stimulatory agents. CTLA4Ig is a fusion protein designed to block co-stimulatory signals important for T-cell activation, such as CD86 and CD80. Anti-CD40L antibodies are also important in blocking signal transduction through this pathway. These treatments have been successfully used in autoimmune diseases and have been shown to be effective in animal models of GvHD. Clinical trials assessing these agents are currently under way.127–129
New and Experimental Agents Monoclonal antibodies or receptor antagonists. These include anti-CD5, IL-1-receptor antagonist, murine anti-LFA-1 antibody, and anti-IL-2a receptor antibody (humanized and chimeric). Clofazimine, psoralen and photopheresis, radiation, peptides, and polymers. These drugs and treatment modalities were successfully tried in a limited number of patients with GvHD. Initial results are encouraging but further validation is needed.101,129
PREVENTION AND TREATMENT OF CHRONIC GVHD (NON-SPECIFIC AND T-CELL-SPECIFIC) A multidisciplinary approach with the involvement of dentists, ophthalmogists, gastroenterologists, and physiotherapists is important in managing the chronic manifestations of GvHD.13,102,103 Treatment of cGvHD is similar to that of aGvHD, with various modifications. Generally, patients who develop cGvHD require reinstitution of immunosuppressive medication (if already discontinued), increase in dosing, or addition of medication if still on such treatment. Common regimens include steroids, either systemically or topically, or in combination with other agents, such as ciclosporin. Other options include tacrolimus, mycophenolate mofetil, rapamycin, thalidomide, anti-T-cell antibodies, and photophoresis, all of which have been described for aGvHD (see above). Close attention should be given to infectious complications, as patients with cGvHD are severely immunodeficient. This should include routine assessment for CMV infection by monitoring CMV antigenemia. Patients suffering from recurrent respiratory infections secondary to immunoglobulin deficiency may benefit
870
from intravenous immunoglobulin infusions. Routine prophylaxis against Pneumocystis carinii is recommended and prophylactic antibiotic treatment against encapsulated bacterial agents has also been suggested. The US Centers for Disease Control recommend a rigorous vaccination program, although patients with cGvHD may be poorly responsive to vaccination.130 Live attenuated vaccines are prohibited until at least 2 years following transplantation.
LIVER TRANSPLANTATION FOR SEVERE GvHD Liver failure is a rare complication of GvHD. A recent literature review identified 80 patients who underwent liver transplantation for GvHD. Data collected by the US United Network for Organ Sharing (UNOS) revealed a 1 and 5 year actuarial survival rate in 73 patients who underwent liver transplantation of 72.4% and 62.9% respectively. Thus, long-term disease free survival is an obtainable goal in BMT patients with hepatic failure as a result of GvHD.131
PROGNOSIS GvHD, whether acute or chronic, has a detrimental effect on survival. Approximately 30% of patients with moderate to severe aGvHD achieve cure, and about 70% progress to cGvHD.98 Approximately 50% of patients with cGvHD have limited disease and a good prognosis. Sixty percent of the patients having a more extensive disease respond to treatment and will eventually also have a good outcome. The remaining patients have a poor prognosis and will either die from infections or organ failure or will need prolonged immunosuppressive treatment.132
CONCLUSION GvHD is a systemic complication of BMT or PBSCT, which frequently involves the liver. Immunological mediated hepatic involvement is usually manifested clinically by cholestatic liver dysfunction and less frequently as hepatocellular injury. Although often presenting in the context of a life-threatening situation, liver failure is relatively rare, while immune suppression and systemic infection are the more frequent detrimental prognostic factors. Better technology for depletion of donor lymphocytes pretransplantation as well as pharmacological intervention through immunosuppressive agents have contributed to improved survival, while UDCA may be beneficial in ameliorating cholestasis. We would like to thank Dr O. Pappo for her comments on the histopathology section.
REFERENCES 1. Goker H, Haznedaroglu IC, Chao NJ. Acute graft-vs-host disease: pathobiology and management. Exp Hematol 2001; 29:259–277. 2. Barnes DWH, Corp MJ, Loutit JF, et al. Treatment of murine leukaemia with x-rays and homologous bone marrow: preliminary communication. Br J Med 1956; 2:626–627. 3. Billingham RE. The biology of graft-versus-host reactions. Harvey Lect 1966; 62:21–78.
Chapter 44 GRAFT-VERSUS-HOST DISEASE AND THE LIVER
4. Basara N, Kiehl MG, Fauser AA. New therapeutic modalities in the treatment of graft-versus-host disease. Crit Rev Oncol Hematol 2001; 38:129–138. 5. Gulbahce HE, Brown CA, Wick M, et al. Graft-vs-host disease after solid organ transplant. Am J Clin Pathol 2003; 119:568–573. 6. Schroeder ML. Transfusion-associated graft-versus-host disease. Br J Haematol 2002; 117:275–287. 7. Leitman SF, Tisdale JF, Bolan CD, et al. Transfusion-associated GVHD after fludarabine therapy in a patient with systemic lupus erythematosus. Transfusion 2003; 43:1667–1671. 8. Vogelsang GB, Lee L, Bensen-Kennedy DM. Pathogenesis and treatment of graft-versus-host disease after bone marrow transplant. Annu Rev Med 2003; 54:29–52. 9. Kolb HJ, Schmid C, Barrett AJ, Schendel DJ. Graft-versusleukemia reactions in allogeneic chimeras. Blood 2004; 103:767–776. 10. Ferrara JL, Deeg HJ. Graft-versus host disease. N Engl J Med 1991; 324:667–674. 11. Bensinger WI, Martin PJ, Storer B, et al. Transplantation of bone marrow as compared with peripheral-blood cells from HLAidentical relatives in patients with hematologic cancers. N Engl J Med 2001; 344:175–181. 12. Schmitz N, Beksac M, Hasenclever D, et al. Transplantation of mobilized peripheral blood cells to HLA-identical siblings with standard-risk leukemia. Blood 2002; 100:761–767. 13. Farag SS. Chronic graft-versus-host disease: where do we go from here? Bone Marrow Transplant 2004; 33:569–577. 14. Heldal D, Brinch L, Tjonnfjord G, et al. Fewer relapses and increased chronic GVHD in patients transplanted with blood stem cells: a 5-year follow-up in a single centre study. Bone Marrow Transplant 2003; 32:257–264. 15. El-Sayed MH, El-Haddad A, Fahmy OA, et al. Liver disease is a major cause of mortality following allogeneic bone-marrow transplantation. Eur J Gastroenterol Hepatol 2004; 16:1347–1354. 16. Smith DM, Agura E, Netto G, et al. Liver transplant-associated graft-versus-host disease. Transplantation 2003; 75:118–126. 17. Taylor AL, Gibbs P, Bradley JA. Acute graft versus host disease following liver transplantation: the enemy within. Am J Transplant 2004; 4:466–474. 18. Goulmy E, Schipper R, Pool J, et al. Mismatches of minor histocompatibility antigens between HLA-identical donors and recipients and the development of graft-versus-host disease after bone marrow transplantation. N Engl J Med 1996; 334:281–285. 19. Chao NJ. Minors come of age: Minor histocompatibility antigens and graft-versus-host disease. Biol Blood Marrow Transplant 2004; 10:215–223. 20. McSweeny PA, Niederwieser D, Shizuru JA, et al. Hematopoietic cell transplantation in older patients with hematologic malignancies: replacing high-dose cytotoxic therapy with graft-versus-tumor effects. Blood 2001; 97: 3390–3400. 21. Zeiser R, Marks R, Bertz H, Finke J. Immunopathogenesis of acute graft-versus-host disease: implications for novel preventive and therapeutic strategies. Ann Hematol 2004; 83:551–565. 22. Butcher EC, Picker LJ. Lymphocyte homing and homeostasis. Science 1996; 272:60–66. 23. Shlomchik WD. Antigen presentation in graft-vs-host disease. Exp Hematol 2003; 31:1187–1197. 24. Reddy P, Ferrara JL. Immunobiology of acute graft-versus-host disease. Blood Rev 2003; 17:187–194. 25. Zeng D, Lan F, Hoffmann P, Strober S. Suppression of graftversus-host disease by naturally occurring regulatory T cells. Transplantation 2004; 77:S9–S11. 26. Jung U, Foley JE, Erdmann AA, et al. CD3/CD28-costimulated T1 and T2 subsets: differential in vivo allosensitization generates distinct GVT and GVHD effects. Blood 2003; 102:3439–3446.
27. Rifle G, Herve P. Regulatory (suppressor) T cells in peripheral allograft tolerance and graft-versus-host reaction. Transplantation 2004; 77:S5. 28. Antin JH, Ferrara JL. Cytokine dysregulation and acute graftversus-host disease. Blood 1992; 80:2964–2968. 29. Teshima T, Ordemann R, Reddy P, et al. Acute graft-versus-host disease does not require alloantigen expression on host epithelium. Nat Med 2002; 8:575–581. 30. Maeda Y, Levy RB, Reddy P, et al. Both perforin and Fas ligand are required for the regulation of alloreactive CD8+ T cells during acute graft-versus-host disease. Blood 2005; 105:2023–2027. 31. Graubert TA, DiPersio JF, Russell JH, Ley TJ. Perforin/granzyme-dependent and independent mechanisms are both important for the development of graft-versus-host disease after murine bone marrow transplantation. J Clin Invest 1997; 100:904–911. 32. Marks L, Altman NH, Podack ER, Levy RB. Donor T cells lacking Fas ligand and perforin retain the capacity to induce severe GvHD in minor histocompatibility antigen mismatched bone-marrow transplantation recipients. Transplantation 2004; 77:804–812. 33. van Den Brink MR, Moore E, Horndasch KJ, et al. Fas-deficient lpr mice are more susceptible to graft-versus-host disease. J Immunol 2000; 164:469–480. 34. Higman MA, Vogelsang GB. Chronic graft versus host disease. Br J Haematol 2004; 125:435–454. 35. Demetris AJ. Immune cholangitis: liver allograft rejection and graft-versus-host disease. Mayo Clin Proc 1998; 73:367–379. 36. Wakae T, Takatsuka H, Seto Y, et al. Similarity between hepatic graft-versus-host disease and primary biliary cirrhosis. Hematology 2002; 7:305–310. 37. Kansu E. The pathophysiology of chronic graft-versus-host disease. Int J Hematol 2004; 79:209–215. 38. Weinberg K, Blazar BR, Wagner JE, et al. Factors affecting thymic function after allogeneic hematopoietic stem cell transplantation. Blood 2001; 97:1458–1466. 39. Shustov A, Luzina I, Nguyen P, et al. Role of perforin in controlling B-cell hyperactivity and humoral autoimmunity. J Clin Invest 2000; 106:R39–R47. 40. Maury S, Mary JY, Rabian C, et al. Prolonged immune deficiency following allogeneic stem cell transplantation: risk factors and complications in adult patients. Br J Haematol 2001; 115:630–641. 41. Sullivan KM, Deeg HJ, Sanders J, et al. Hyperacute graft-v-host disease in patients not given immunosuppression after allogeneic marrow transplantation. Blood 1986; 67:1172–1175. 42. Kim DH, Sohn SK, Kim JG, et al. Clinical impact of hyperacute graft-versus-host disease on results of allogeneic stem cell transplantation. Bone Marrow Transplant 2004; 33:1025– 1030. 43. Ho GT, Parker A, MacKenzie JF, et al. Abnormal liver function tests following bone marrow transplantation: aetiology and role of liver biopsy. Eur J Gastroenterol Hepatol 2004; 16: 157–162. 44. Hogan WJ, Maris M, Storer B, et al. Hepatic injury after nonmyeloablative conditioning followed by allogeneic hematopoietic cell transplantation: a study of 193 patients. Blood 2004; 103:78–84. 45. Chiba T, Yokosuka O, Kanda T, et al. Hepatic graft-versus-host disease resembling acute hepatitis: additional treatment with ursodeoxycholic acid. Liver 2002; 22:514–517. 46. Strasser SI, Shulman HM, Flowers ME, et al. Chronic graftversus-host disease of the liver: presentation as an acute hepatitis. Hepatology 2000; 32:1265–1271. 47. Akpek G, Boitnott JK, Lee LA, et al. Hepatic variant of graftversus-host disease after donor lymphocyte infusion. Blood 2002; 100:3903–3907.
871
Section VI. Immune Diseases
48. Fujii N, Takenaka K, Shinagawa K, et al. Hepatic graft-versushost disease presenting as an acute hepatitis after allogeneic peripheral blood stem cell transplantation. Bone Marrow Transplant 2001; 27:1007–1010. 49. Yeh KH, Hsieh HC, Tang JL, et al. Severe isolated acute hepatic graft-versus-host disease with vanishing bile duct syndrome. Bone Marrow Transplant 1994; 14:319–321. 50. Knapp AB, Crawford JM, Rappeport JM, Gollan JL. Cirrhosis as a consequence of graft-versus-host disease. Gastroenterology 1987; 92:513–519. 51. Stechschulte DJ Jr, Fishback JL, Emami A, Bhatia P. Secondary biliary cirrhosis as a consequence of graft-versus-host disease. Gastroenterology 1990; 98:223–225. 52. Fried RH, Murakami CS, Fisher LD, et al. Ursodeoxycholic acid treatment of refractory chronic graft-versus-host disease of the liver. Ann Intern Med 1992; 116:624–629. 53. Tomas JF, Pinilla I, Garcia-Buey ML, et al. Long-term liver dysfunction after allogeneic bone marrow transplantation: clinical features and course in 61 patients. Bone Marrow Transplant 2000; 26:649–655. 54. Quaranta S, Shulman H, Ahmed A, et al. Autoantibodies in human chronic graft-versus-host disease after hematopoietic cell transplantation. Clin Immunol 1999; 91:106–116. 55. Shulman HM, Sullivan KM, Weiden PL, et al. Chronic graftversus-host syndrome in man. A long-term clinicopathologic study of 20 Seattle patients. Am J Med 1980; 69:204–217. 56. Shulman HM, Sharma P, Amos D, et al. A coded histologic study of hepatic graft-versus-host disease after human bone marrow transplantation. Hepatology 1988; 8:463–470. 57. Ma SY, Au WY, Lie AK, et al. Liver graft-versus-host disease after donor lymphocyte infusion for relapses of hematologic malignancies post allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant 2004; 34:57–61. 58. Rouquette-Gally AM, Boyeldieu D, Prost AC, Gluckman E. Autoimmunity after allogeneic bone marrow transplantation. A study of 53 long-term-surviving patients. Transplantation 1988; 46:238–240. 59. Ma SY, Au WY, Ng IO, et al. Role of liver biopsy in the management of liver dysfunction after hematopoietic stem-cell transplantation in a hepatitis B virus-prevalent patient population. Transplantation 2003; 76:169–176. 60. Snover DC, Weisdorf SA, Ramsay NK, et al. Hepatic graft versus host disease: a study of the predictive value of liver biopsy in diagnosis. Hepatology 1984; 4:123–130. 61. Hofmeister CC, Quinn A, Cooke KR, et al. Graft-versus-host disease of the skin: life and death on the epidermal edge. Biol Blood Marrow Transplant 2004; 10:366–372. 62. McDonald GB, Shulman HM, Sullivan KM, Spencer GD. Intestinal and hepatic complications of human bone marrow transplantation. Part I. Gastroenterology 1986; 90:460–477. 63. McDonald GB, Shulman HM, Sullivan KM, Spencer GD. Intestinal and hepatic complications of human bone marrow transplantation. Part II. Gastroenterology 1986; 90:770–784. 64. Socie G, Mary JY, Lemann M, et al. Prognostic value of apoptotic cells and infiltrating neutrophils in graft-versus-host disease of the gastrointestinal tract in humans: TNF and Fas expression. Blood 2004; 103:50–57. 65. Akpek G, Chinratanalab W, Lee LA, et al. Gastrointestinal involvement in chronic graft-versus-host disease: a clinicopathologic study. Biol Blood Marrow Transplant 2003; 9:46–51. 66. Glucksberg H, Storb R, Fefer A, et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HL-A-matched sibling donors. Transplantation 1974; 18:295–304. 67. Rowlings PA, Przepiorka D, Klein JP, et al. IBMTR severity index for grading acute graft-versus-host disease: retrospective
872
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
comparison with Glucksberg grade. Br J Haematol 1997; 97:855–864. Atkinson K, Horowitz MM, Gale RP, et al. Consensus among bone marrow transplanters for diagnosis, grading and treatment of chronic graft-versus-host disease. Committee of the International Bone Marrow Transplant Registry. Bone Marrow Transplant 1989; 4:247–254. Akpek G, Lee SJ, Flowers ME, et al. Performance of a new clinical grading system for chronic graft-versus-host disease: a multicenter study. Blood 2003; 102:802–809. Arai S, Lee LA, Vogelsang GB. A systematic approach to hepatic complications in hematopoietic stem cell transplantation. J Hematother Stem Cell Res 2002; 11:215–229. Liastos C, Mehta AB, Potter M, Burroughs AK. The hepatologist in the hematologist’s camp. Br Haematol 2001; 113:567–578. Kumar S, DeLeve LD, Kamath PS, Tefferi A. Hepatic venoocclusive disease (sinusoidal obstruction syndrome) after hematopoietic stem cell transplantation. Mayo Clin Proc 2003; 78:589–598. Lassau N, Auperin A, Leclere J, et al. Prognostic value of doppler-ultrasonography in hepatic veno-occlusive disease. Transplantation 2002; 74:60–66. Yoshimoto K, Ono N, Okamura T, Sata M. Recent progress in the diagnosis and therapy for veno-occlusive disease of the liver. Leuk Lymphoma 2003; 44:229–234. Coppell JA, Brown SA, Perry DJ. Veno-occlusive disease: cytokines, genetics, and haemostasis. Blood Rev 2003; 17:63–70. Rajvanshi P, McDonald GB. Expanding the use of transjugular intrahepatic portosystemic shunts for veno-occlusive disease. Liver Transpl 2001; 7:154–157. Rapoport AP, Doyle HR, Starzl T, et al. Orthotopic liver transplantation for life-threatening veno-occlusive disease of the liver after allogeneic bone marrow transplant. Bone Marrow Transplant 1991; 8:421–424. Kim ID, Egawa H, Marui Y, et al. A successful liver transplantation for refractory hepatic veno-occlusive disease originating from cord blood transplantation. Am J Transplant 2002; 2:796–800. Jacobson AF, Teefey SA, Lee SP, et al. Frequent occurrence of new hepatobiliary abnormalities after bone marrow transplantation: results of a prospective study using scintigraphy and sonography. Am J Gastroenterol 1993; 88:1044–1049. Locasciulli A, Alberti A, de Bock R, et al. Impact of liver disease and hepatitis infections on allogeneic bone marrow transplantation in Europe: a survey from the European Bone Marrow Transplantation (EBMT) Group – Infectious Diseases Working Party. Bone Marrow Transplant 1994; 14:833–837. Locasciulli A, Alberti A. Hepatitis B and hepatitis C virus infections in stem cell transplantation. Leuk Lymphoma 1999; 35:255–260. Locasciulli A, Testa M, Valsecchi MG, et al. The role of hepatitis C and B virus infections as risk factors for severe liver complications following allogeneic BMT: a prospective study by the Infectious Disease Working Party of the European Blood and Marrow Transplantation Group. Transplantation 1999; 68:1486–1491. Liang R, Lau GK, Kwong YL. Chemotherapy and bone marrow transplantation for cancer patients who are also chronic hepatitis B carriers: a review of the problem. J Clin Oncol 1999; 17:394–398. Seth P, Alrajhi AA, Kagevi I, et al. Hepatitis B virus reactivation with clinical flare in allogeneic stem cell transplants with chronic graft-versus-host disease. Bone Marrow Transplant 2002; 30:189–194. Picardi M, Selleri C, De Rosa G, et al. Lamivudine treatment for chronic replicative hepatitis B virus infection after allogeneic
Chapter 44 GRAFT-VERSUS-HOST DISEASE AND THE LIVER
86. 87.
88.
89.
90.
91.
92.
93.
94. 95.
96. 97.
98.
99.
100.
101.
102. 103. 104.
105.
106.
107.
bone marrow transplantation. Bone Marrow Transplant 1998; 21:1267–1269. Shouval D, Ilan Y. Immunization against hepatitis B through adoptive transfer of immunity. Intervirology 1995; 38:41–46. Lau GK, Suri D, Liang R, et al. Resolution of chronic hepatitis B and anti-HBs seroconversion in humans by adoptive transfer of immunity to hepatitis B core antigen. Gastroenterology 2002; 122:614–624. Shouval D, Ilan Y. Transplantation of hepatitis B immune lymphocytes as means for adoptive transfer of immunity to hepatitis B virus. J Hepatol 1995; 23:98–101. Strasser SI, Myerson D, Spurgeon CL, et al. Hepatitis C virus infection and bone marrow transplantation: a cohort study with 10-year follow-up. Hepatology 1999; 29:1893–1899. Ljungman P, Johansson N, Aschan J, et al. Long-term effects of hepatitis C virus infection in allogeneic bone marrow transplant recipients. Blood 1995; 86:1614–1618. Giardini C, Galimberti M, Lucarelli G, et al. Alpha-interferon treatment of chronic hepatitis C after bone marrow transplantation for homozygous beta-thalassemia. Bone Marrow Transplant 1997; 20:767–772. Zaia JA, Forman SJ. Cytomegalovirus infection in the bone marrow transplant recipient. Infect Dis Clin North Am 1995; 9:879–900. Walter EA, Bowden RA. Infection in the bone marrow transplant recipient. Infect Dis Clin North Am 1995; 9:823–847. Wingard JR. Fungal infections after bone marrow transplant. Biol Blood Marrow Transplant 1999; 5:55–68. Marr KA, Bowden RA. Fungal infections in patients undergoing blood and marrow transplantation. Transpl Infect Dis 1999; 1:237–246. Yuen KY, Woo PC. Tuberculosis in blood and marrow transplant recipients. Hematol Oncol 2002; 20:51–62. Loren AW, Porter DL, Stadtmauer EA, Tsai DE. Post-transplant lymphoproliferative disorder: a review. Bone Marrow Transplant 2003; 31:145–155. Weisdorf D, Haake R, Blazar B, et al. Treatment of moderate/severe acute graft-versus-host disease after allogeneic bone marrow transplantation: an analysis of clinical risk features and outcome. Blood 1990; 75:1024–1030. Prentice HG, Blacklock HA, Janossy G, et al. Depletion of T lymphocytes in donor marrow prevents significant graft-versushost disease in matched allogeneic leukaemic marrow transplant recipients. Lancet 1984; 1:472–476. Ho VT, Soiffer RJ. The history and future of T-cell depletion as graft-versus-host disease prophylaxis for allogeneic hematopoietic stem cell transplantation. Blood 2001; 98:3192–3204. Bruner RJ, Farag SS. Monoclonal antibodies for the prevention and treatment of graft-versus-host disease. Semin Oncol 2003; 30:509–519. Lee SJ, Vogelsang G, Flowers ME. Chronic graft-versus-host disease. Biol Blood Marrow Transplant 2003; 9:215–233. Lee SJ. New approaches for preventing and treating chronic graft-versus-host disease. Blood 2005. Lee SJ, Klein JP, Barrett AJ, et al. Severity of chronic graftversus-host disease: association with treatment-related mortality and relapse. Blood 2002; 100:406–414. Ancin I, Ferra C, Gallardo D, et al. Do corticosteroids add any benefit to standard GVHD prophylaxis in allogeneic BMT? Bone Marrow Transplant 2001; 28:39–45. MacMillan ML, Weisdorf DJ, Wagner JE, et al. Response of 443 patients to steroids as primary therapy for acute graft-versushost disease: comparison of grading systems. Biol Blood Marrow Transplant 2002; 8:387–394. Ross WA, Couriel D. Colonic graft-versus-host disease. Curr Opin Gastroenterol 2005; 21:64–69.
108. Shapira MY, Bloom AI, Or R, et al. Intra-arterial catheter directed therapy for severe graft-versus-host disease. Br J Haematol 2002; 119:760–764. 109. Ringden O, Klaesson S, Sundberg B, et al. Decreased incidence of graft-versus-host disease and improved survival with methotrexate combined with ciclosporin compared with monotherapy in recipients of bone marrow from donors other than HLA identical siblings. Bone Marrow Transplant 1992; 9:19–25. 110. Bloom AI, Shapira MY, Or R, et al. Intrahepatic arterial administration of low-dose methotrexate in patients with severe hepatic graft-versus-host disease: an open-label, uncontrolled trial. Clin Ther 2004; 26:407–414. 111. Flowers ME, Martin PJ. Evaluation of thalidomide for treatment or prevention of chronic graft-versus-host disease. Leuk Lymphoma 2003; 44:1141–1146. 112. Epstein JB, Nantel S, Sheoltch SM. Topical azathioprine in the combined treatment of chronic oral graft-versus-host disease. Bone Marrow Transplant 2000; 25:683–687. 113. Ruutu T, Eriksson B, Remes K, et al. Ursodeoxycholic acid for the prevention of hepatic complications in allogeneic stem cell transplantation. Blood 2002; 100:1977–1983. 114. Mela M, Mancuso A, Burroughs AK. Review article: pruritus in cholestatic and other liver diseases. Aliment Pharmacol Ther 2003; 17:857–870. 115. Kowdley KV. Ursodeoxycholic acid therapy in hepatobiliary disease. Am J Med 2000; 108:481–486. 116. Khoury H, Trinkaus K, Zhang MJ, et al. Hydroxychloroquine for the prevention of acute graft-versus-host disease after unrelated donor transplantation. Biol Blood Marrow Transplant 2003; 9:714–721. 117. Ruutu T, Niederwieser D, Gratwohl A, Apperley JF. A survey of the prophylaxis and treatment of acute GVHD in Europe: a report of the European Group for Blood and Marrow, Transplantation (EBMT). Chronic Leukaemia Working Party of the EBMT. Bone Marrow Transplant 1997; 19:759–764. 118. Storb R, Martin P, Deeg HJ, et al. Long-term follow-up of three controlled trials comparing ciclosporine versus methotrexate for graft-versus-host disease prevention in patients given marrow grafts for leukemia. Blood 1992; 79:3091–3092. 119. Lee TJ, Kennedy LA. Tacrolimus: an alternative for graft-versushost disease prevention. Ann Pharmacother 2000; 34:377–381. 120. Cutler C, Antin JH. Sirolimus for GVHD prophylaxis in allogeneic stem cell transplantation. Bone Marrow Transplant 2004; 34:471–476. 121. Vogelsang GB, Arai S. Mycophenolate mofetil for the prevention and treatment of graft-versus-host disease following stem cell transplantation: preliminary findings. Bone Marrow Transplant 2001; 27:1255–1262. 122. Hebart H, Ehninger G, Schmidt H, et al. Treatment of steroidresistant graft-versus-host disease after allogeneic bone marrow transplantation with anti-CD3/TCR monoclonal antibodies. Bone Marrow Transplant 1995; 15:891–894. 123. Remberger M, Aschan J, Barkholt L, et al. Treatment of severe acute graft-versus-host disease with anti-thymocyte globulin. Clin Transplant 2001; 15:147–153. 124. Cordonnier C, Chevret S, Legrand M, et al. Should immunoglobulin therapy be used in allogeneic stem-cell transplantation? A randomized, double-blind, dose effect, placebo-controlled, multicenter trial. Ann Intern Med 2003; 139:8–18. 125. Couriel D, Saliba R, Hicks K, et al. Tumor necrosis factor-alpha blockade for the treatment of acute GVHD. Blood 2004; 104:649–654. 126. Lee SJ, Zahrieh D, Agura E, et al. Effect of up-front daclizumab when combined with steroids for the treatment of acute graftversus-host disease: results of a randomized trial. Blood 2004; 104:1559–1564.
873
Section VI. Immune Diseases
127. Wallace PM, Johnson JS, MacMaster JF, et al. CTLA4Ig treatment ameliorates the lethality of murine graft-versus-host disease across major histocompatibility complex barriers. Transplantation 1994; 58:602–610. 128. Jacobsohn DA. Emerging therapies for graft-versus-host disease. Expert Opin Emerg Drugs 2003; 8:323–338. 129. Jacobsohn DA, Vogelsang GB. Anti-cytokine therapy for the treatment of graft-versus-host disease. Curr Pharm Des 2004; 10:1195–1205. 130. Dykewicz CA, National Center for Infectious Diseases, Centers for Disease Control and Prevention, et al. Guidelines for
874
preventing opporturistic infections among hematopoietic stem cell transplant recipients: focus on community respiratory virus infections. Biol Blood Marrow Transplant 2001;7:19S– 22S. 131. Barshes NR, Myers GD, Lee D, et al. Liver transplantation for severe hepatic graft versus host disease: an anlysis of aggregate survival data. Liver Transpl 2005;11:525–531. 132. Martin PJ, Schoch G, Fisher L, et al. A retrospective analysis of therapy for acute graft-versus-host disease: secondary treatment. Blood 1991; 77:1821–1828.
Section VII: Vascular Disease of the Liver
BUDD–CHIARI SYNDROME Dominique-Charles Valla
Abbreviations BCS Budd–Chiari syndrome IVC inferior vena cava PNH Paroxysmal nocturnal hemoglobinuria
TIPS
45
transjugular intrahepatic portosystemic stent shunt
INTRODUCTION Budd–Chiari syndrome (BCS) is a heterogeneous disorder characterized by an obstacle to the hepatic venous outflow, be it at the level of the small hepatic veins, large hepatic veins, or the suprahepatic portion of the inferior vena cava (IVC).1 Variation in the level of obstruction is one of the factors explaining the heterogeneity of the disease. Therefore, optimal denomination should include the level of predominant occlusion (i.e., BCS due to IVC occlusion, or BCS due to small hepatic vein obstruction).1,2 The definitions for these levels are presented in Table 45-1.2 BCS is further defined as primary when the obstructing process arises from the venous wall (phlebitis or fibrous thickening) or the venous lumen (thrombosis).1 BCS is considered secondary when the veins are compressed or invaded by a lesion arising outside these veins. Veno-occlusive disease is a misleading term that should be replaced either by “sinusoidal obstruction syndrome” for toxic injury to sinusoidal endothelial cells (particularly in the setting of myeloablative therapy or following ingestion of pyrrolizidine alkaloids),3 or by BCS where appropriate.
EPIDEMIOLOGY BCS is a rare disease: surveys performed in the early 1990s in Japan4 and France (unpublished data from the Observatoire National du Syndrome de Budd–Chiari) could identify only a few hundred cases over one to two decades. However, in some places such as Nepal a high prevalence has been recorded.5 Current estimates of the prevalence may be higher due to increased awareness and improved diagnostic methods.
CAUSES PRIMARY BUDD–CHIARI SYNDROME Following a systematic investigation, one or several thrombotic risk factors are found in about 90% of patients with primary BCS; two or more of these factors are found in at least 25% of patients.6–10 Estimates of the frequencies or odds ratios for these risk factors in BCS patients are presented in Table 45-2.
Primary myeloproliferative disorders account for about 50% of cases of primary BCS.6,10,11 Conversely, only 1–3% of patients with primary myeloproliferative disorders are likely to develop BCS.12,13 In most cases the myeloproliferative disorder can be classified as essential thrombocythemia or polycythemia rubra, but forms that are difficult to classify are encountered as well. Most patients are women (80%) and of a young age (30 years on average). It is poorly understood why some patients with myeloproliferative disorders develop hepatic vein or IVC thrombosis. Association with other risk factors for thrombosis (mainly factor V Leiden or antiphospholipid antibodies) can be identified in about 30% of patients. At the time of presentation with BCS, the myeloproliferative disorder has not previously been recognized in 90% of patients. Moreover, in half of the patients, peripheral blood cell counts at presentation are not suggestive for a myeloproliferative disease.6,10,14 These occult or latent forms are explained by hypersplenism, iron deficiency, and dilution of blood element due to a marked increase in plasma volume – all factors related to portal hypertension.15 Diagnosis is made by showing clusters of dystrophic megakaryocytes in a bone marrow biopsy, or by showing formation of so-called spontaneous or endogenous colonies in cultures of erythroid progenitors on erythropoietinpoor media (in the subject without primary myeloproliferative disorder, erythroid colonies only form upon adding erythropoietin to the culture medium).6,14–16 In the author’s experience, 12% of patients have discordant results by bone marrow biopsy compared to the assessment of erythroid colonies. BCS patients with myeloproliferative disorders have, on average, higher platelet counts and a larger spleen than those without myeloproliferative disorders, but a large overlap makes these features of little practical utility. However, in a patient with features of marked portal hypertension, platelet counts over 300 000/fl strongly suggest that a myeloproliferative disorder is present. Except in the setting of a systemic inflammatory syndrome, secondary thrombocytosis has not been reported to cause BCS. Isotopic determination of total red cell mass shows an increased total red cell mass in many patients with a normal hematocrit. Such a finding is strong evidence for a primary myeloproliferative disorder as BCS complicating secondary erythrocytosis appears exceptional. Late transformation into leukemia has been reported in some of these patients17,18 but the incidence remains unclear. No specific information is available on the treatment of primary myeloproliferative disorders complicated by BCS.
877
Section VII. Vascular Disease of the Liver
decision on bone marrow transplantation. Liver transplantation has been successfully performed in anecdotal BCS patients with PNH but recurrence has been observed.22 All known inherited thrombophilias have been implicated in BCS.6–10 As a rule, the thrombophilias are cured by liver transplantation. Activated protein C resistance, generally related to heterozygous or homozygous factor V Leiden mutation, is found in about 25% of patients with BCS, whereas its prevalence is about 5% in the Caucasian population.7,10 When present, factor V Leiden mutation is usually associated with another thrombotic risk factor. Factor V Leiden mutation is particularly frequent in some subgroups of BCS patients: those with progressive IVC obstruction; those with acute BCS related to pregnancy; and among oral contraceptive users.10 Diagnosis is ruled out by a normal test for activated protein C resistance. Confirmation of the mutation is obtained by molecular biology techniques. The G20210A mutation in the prothrombin gene is responsible for increased plasma levels of prothrombin and a moderate increase in the risk of venous thrombosis. This mutation is found in about 5% of BCS patients, whereas the background prevalence in western countries is about 2%.7,10 Diagnosis is made by molecular biology techniques. Antithrombin, protein C, or protein S is produced by the liver and their plasma level is non-specifically decreased in many patients with BCS. Family studies are necessary to establish a low plasma level of these proteins as a primary deficiency. The high number of mutations in different families precludes direct molecular testing at present. Formulas taking into account the degree of liver insufficiency have yielded a prevalence estimate of about 25% for primary protein C deficiency in patients with BCS.7,10 However, these formulas require further validation. These three inherited thrombophilias are cured with liver transplantation from an unaffected donor. Afibrinogenemia may paradoxically predispose to thrombosis and has been implicated as a cause for BCS.23
Paroxysmal nocturnal hemoglobinuria (PNH) accounts for about 5% of cases of primary BCS, a surprisingly high proportion when the extreme rarity of this blood disease is considered. Indeed, retrospective surveys have indicated that up to 12% of patients with PNH will develop BCS.19 In many cases, thrombosis is limited to the small hepatic veins and the patient is asymptomatic.20 Diagnosis of PHN is made with flow cytometry, showing a decreased proportion of CD55- and CD59-positive blood cells. The mechanisms for hepatic vein or IVC thrombosis in PHN are not known. Treatment is not well established. The outcome is poor due to BCS and to the other complications of the blood disease. Resolution of BCS following bone marrow transplantation has been reported.21 The risk of severe deterioration from superimposed sinusoidal obstruction syndrome due to the conditioning regimen use before stem cell transplantation should be strongly considered before making a
Table 45-1. Level of Hepatic Venous Outflow Obstruction2 Designation
Definition
Small hepatic veins
Veins that cannot be shown clearly in hepatic venograms or by ultrasound studies; they include terminal hepatic veins (central veins), intercalated veins, and interlobular veins (collecting veins) Veins that are regularly demonstrable on hepatic venograms and ultrasound studies; segmental hepatic vein branches are generally included A segment of the IVC that involves the entire area that is in contact with the right lobe and the caudate lobe of the liver up to the entry level of the right middle and left hepatic veins A segment of the IVC that extends from the entry level of the right, middle, and left hepatic veins to the junction between the IVC and the right atrium
Large hepatic veins Hepatic IVC
Suprahepatic IVC
IVC, inferior vena cava.
Table 45-2. Thrombotic Risk Factors in Budd–Chiari Syndrome (BCS) Patients Country
Myeloproliferative disorders Occult Classical Antiphospholipid syndrome Paroxysmal nocturnal hemoglobinuria Factor V Leiden mutation Factor II mutation Protein C deficiency Protein S deficiency Antithrombin deficiency Plasminogen deficiency Recent pregnancy Recent oral contraceptive use
Israel6 n positive/ n tested (%)
India9
France10 n positive/ n tested (%)
n positive/ n tested (%)
OR for BCS (95% CI)
n positive/ n tested (%)
OR for BCS (95% CI)
10/22 (45.4%) 8/22 (36.4%) 2/22 (9.0%) 5/22 (22.7%) 1/22 (4.4%)
NA NA 12/43 (27.0%) 2/43 (4.6%) 0/43
NA NA NA NA NA
NA NA NA 6/53 (11.3%) NA
NA NA NA NA NA
31/61 (50.8%) 15/61 (24.6%) 16/61 (26.2%) 9/63 (14.3%) 1/63 (1.6%)
6/22 (27.2%) NA 7/22 (31.8%) 4/22 (18.2%) 5/22 (22.7%) NA NA NA
11/43 (25.6%) 2/43 (4.7%) 4/43 (9.3%) 0/43 0/43 NA NA 12/20 (60%)
11.3 (4.6–26.5) 2.1 (0.4–9.6) 6.8 ( 1.9–24.4) — — — — 2.4 (0.9–6.2)
14/53 (26.4%) 0/53 7/53 (13.2%) 3/53 (5.7%) 2/53 (3.8%) NA 2/17 (11.7%) 1/17 (5.9%)
14.4 (11.9–35.7) — 8.3 (2.1–10.9) 4.4 (2.2–20.6) 4.4 (2.1–10.9) NA NA NA
20/63 (31.7%) 3/47 (6.4%) 4/21 (19.0%) 2/30 (6.6%) 0/47 1/21 (4.7%) 3/47 (6.4%) 23/47 (48.9%)
OR, odds ratio; CI, confidence index; NA, not available.
878
Netherlands7,9
Chapter 45 BUDD-CHIARI SYNDROME
Increased plasma levels of factor VII, factor VIII, or homocysteine are risk factors for thrombosis.24 Plasma levels of all three substances are altered in patients with liver disease. Therefore, the significance of their plasma level in patients with BCS is unclear. The antiphospholipid syndrome associated with BCS is the primary type, i.e., the American Rheumatism Association criteria for the diagnosis of systemic lupus were not fulfilled.25,26 Anticardiolipin antibodies have been reported in about 25% of primary BCS patients.6–11 Antibeta-2 glycoprotein I antibodies, lupus anticoagulant, and association with other autoantibodies are more convincing evidence for the diagnosis of antiphospholipid syndrome than the mere presence of low-titer anticardiolipin antibodies. Behçet’s disease, an established cause for venous thrombophlebitis, accounts for less than 5% of BCS patients in western countries.27 In countries where the disease is prevalent, such as Turkey, the corresponding figure is close to 40%.28,29 Conversely, the incidence of BCS in patients with Behçet’s disease is 5–10%. Over 80% of affected patients have IVC involvement with or without extension to the hepatic veins. The outcome of BCS patients with Behçet’s disease seems to be particularly poor. Liver transplantation has been performed successfully in a few patients. Idiopathic hypereosinophilic syndrome was associated with a few cases of idiopathic BCS.30 The endothelial toxicity of certain eosinophil constituents has been incriminated.31 Eosinophilia associated with 5q deletion has also been reported.32 Idiopathic granulomatous venulitis has been described in several case reports.33–35 The criteria for a diagnosis of sarcoidosis were fulfilled in some,33 but not in all, cases. A good response to corticosteroid therapy has been reported.34 Various gastrointestinal diseases have been anecdotally reported in association with BCS, including celiac disease36,37 and ulcerative colitis.38,39 The mechanisms involved in celiac disease are unknown. As to ulcerative colitis, thrombosis has been attributed to the hypercoagulable state related to the inflammatory syndrome, but reports of an association with an underlying myeloproliferative disorder cast doubts on this simple explanation.40 The link between pregnancy and BCS is based on the time relationship with the onset of acute, and generally severe, clinical manifestations of hepatic vein thrombosis.10,41–43 However many pregnant patients have other thrombotic risk factors.11 Similar to deep vein thrombosis at other sites, BCS related to pregnancy is strongly associated with factor V Leiden mutation.10 Drug-related BCS appears to be exceptional.44 It is of note that, in drug- and plant-related sinusoidal obstruction syndrome, the obstructive process begins in the sinusoids and may affect central veins but does not progress to involve the major hepatic veins.3 A case of hepatic vein thrombosis attributed to a slimming drug preparation was recently reported.45 A case–control study has established oral contraceptives as a risk factor for BCS.46 However, this study was performed in the era of oral contraceptives with a high estrogen content. In a more recent study, the increased risk in oral contraceptive users fell short of statistical significance.7 In both studies, oral contraceptive users generally had other risk factors for thrombosis. Therefore, the role of current oral contraceptives in BCS should be re-evaluated.
SECONDARY BUDD–CHIARI SYNDROME Compression by Space-Occupying Lesions Compression alone is unlikely to induce thrombosis, unless an underlying prothrombotic state is present.11 The local and systemic prothrombotic state associated with inflammation is probably involved in BCS cases reported in association with amebic45 and pyogenic48 liver abscess. Cystic hydatic disease due to Echinococcus granulosus45 rarely produces BCS. By contrast, alveolar hydatid disease due to E. multilocularis commonly obstructs of the hepatic veins or IVC via direct invasion.49 Following blunt abdominal trauma, hepatic veins or IVC can be compressed by an intrahepatic hematoma50 or by a ruptured right hemidiaphragm, allowing herniation of the liver into the thorax51 and compression of the hepatic veins. Hepatic vein ligation during hepatic resection can be followed by hepatic decompensation when asymptomatic obstruction of the remaining veins is overlooked.52,53 Following liver transplantation, misplacement of the graft resulting in a torsion of the hepatic veins or IVC, or stenosis of the hepatic venous anastomosis can result in hepatic venous outflow block.54 Secondary liver cancer, the most common cause of malignant liver tumor, is an uncommon cause of hepatic venous outflow block.55 Compression by a thoracoabdominal aortic aneurysm has been reported.56 In polycystic dominant kidney disease, compression of the hepatic veins by large-sized or infected liver cysts is the cause for portal hypertension, occasionally found in this disease.57 A solitary simple cyst of the liver or a nodule of focal nodular hyperplasia may compress one or several of the three major hepatic veins, albeit without giving rise to signs or symptoms. This occurs when the cyst or nodule is large-sized and located in the upper central part of the liver. A large collateral circulation usually develops, circumscribing the space-occupying lesion.
Endoluminal Invasion by a Tumor This complication is specific to those malignant tumors which tend to progress inside the lumen of their venous outflow tract, toward the IVC, up to the point where they obstruct the hepatic vein ostia, producing an acute form of BCS. These malignant tumors include Wilm’s tumor58 and renal cell carcinoma,59 hepatocellular carcinoma,60 adrenocortical carcinoma,61 and leiomyosarcoma of IVC,62 and hepatic angiosarcoma.63 Right atrial myxoma can obstruct the IVC by a similar but retrograde mechanism.64 The endoluminal tumor can usually be cleaved from the venous walls. This feature allows an urgent operation to be performed to relieve the obstruction. However, prognosis is dismal because lung metastasis is almost always present. Protracted remission or cure is likely only when an effective systemic therapy is available.
EVALUATION FOR ETIOLOGY Whether or not there is evidence for an underlying disorder from clinical data and routine laboratory tests, a comprehensive investigation is recommended in order to allow the recognition of the associated risk factors for thrombosis which then allows for better monitoring, earlier identification of subsequent complications of the underlying disease, and counseling relatives with inherited
879
Section VII. Vascular Disease of the Liver
A
B
Figure 45-1. Pathology of venous lesions in a 31-year-old female patient with essential thrombocythemia. Cross-section of the ostium of the median hepatic vein in the inferior vena cava (IVC) along the axis of the hepatic vein (on the right) and across the axis of the IVC (on the left). (A) Gross aspect. There is ostial stenosis but, upstream, the vein is patent. (B) Mounted slide at the level of the ostium. There is a membrane-like fibrous material obstructing the ostium which is stenosed by fibrous subendothelial thickening. Such lesions are considered the late stage of a thrombus. Similar short-length stenosis in the vicinity of the ostium is encountered on at least one major hepatic vein in about 25% of patients with Budd–Chiari syndrome seen in western countries. The reason why thrombosis develops at this uncommon site is not known. (Courtesy of late Dr. Molas, Pathology Department, Hôpital Beaujon.)
Table 45-3. Initial Workup for Causes of Budd–Chiari Syndrome (BCS)
Figure 45-2. Obstruction of the suprahepatic inferior vena cava (IVC). Inferior vena cavography through the transfemoral route. The catheter in seen in the IVC at the bottom center of the figure. There is complete interruption of the IVC and obstruction of the ostium of the enlarged middle hepatic vein (which remains patent upstream). There is a complex intrahepatic collateral circulation connecting the hepatic veins with the lower portion of the IVC. There are also upward-winding collaterals connecting the veins with the superior vena cava territory. Although the IVC obstruction is membranous-like, it is currently considered the late transformation of a thrombus.
thrombophilia.65 A proposal for the initial workup is presented in Table 45-3.
PATHOLOGY AND PATHOGENESIS OF VENOUS LESIONS THROMBOSIS AS A CAUSE FOR PRIMARY BCS The role of thrombosis in causing primary BCS has long been debated. Primary endophlebitis and congenital malformation have been proposed as alternative hypotheses. Indeed, fresh thrombi are
880
1. For local factors or tumors Abdominal imaging (otherwise needed for BCS diagnosis) 2. For systemic diseases Clinical history and examination 3. For myeloproliferative disorders and paroxysmal nocturnal hemoglobinuria Complete blood cell counts, bone marrow biopsy. Where available, endogenous erythroid colony assays Flow cytometry for CD55- and CD59-deficient cells 4. For antiphospholipid syndrome Lupus anticoagulant, anticardiolipin antibodies, antibeta2glycoprotein 1 and antinuclear factors 5. For inherited coagulopathies Activated protein C resistance (or factor V Leiden mutation) and factor II gene mutation; protein C, protein S, and antithrombin plasma levels in patients without decreased clotting factor levels 6. For hyperhomocysteinemia Blood folate and vitamin B12 levels rather than serum homocysteine level or MTHFR polymorphism
not the most common finding in explanted livers. Rather, fibrous subendothelial thickening is found, involving various length of the hepatic veins or IVC, as illustrated by Figure 45-1, with or without superimposed thrombosis.55 Short-length stenoses can even take the aspect of a membrane, as illustrated by Figure 45-2, suggesting a congenital anomaly. However, short-length stenoses are now considered a sequela of thrombosis, based on the following observations: the transformation of recent thrombi into short-length stenoses;66,67 the presence of venous wall remnants within the stenosed segment;68 the similarly high prevalence of underlying thrombophilias in patients with and without short-length stenoses; and adulthood presentation of most patients with short-length stenoses, dismissing a congenital malformation.68–70 Short-length stenoses are found in about 25% of patients with hepatic vein obstruction and over 60% of patients with IVC obstruction.68–70
Chapter 45 BUDD-CHIARI SYNDROME
Table 45-4. Main Clinical and Laboratory Features of Budd–Chiari Syndrome Patients According to the Area Country Predominant site of venous obstruction Number of patients Ascites (%) Abdominal pain (%) Hepatomegaly (%) Dilated veins over body, trunk (%) Leg edema (%) Jaundice (%) Platelet count (103/mm3) ALT (¥ ULN) Serum bilirubin (mmol/l) Serum albumin (g/l) Serum creatinine (mmol/l) Prothrombin Child–Pugh score Cirrhosis (%)
India41 Mostly IVC
Japan139 Mostly IVC
France, Netherlands, USA85 Mostly hepatic veins
119 86 57 89 49 41 18 NA NA > 17 in 51% < 30 in 37% NA < l70% in 19% NA
157 31 23 55 27 32 6 121 ± 64a NA 27.4 ± 14.8a NA NA NA NA
NA
NA
237 84 NA 79 NA NA NA 265 (10–896)b 1.0 (0.1–86)b 28 (3–301)b 34 (17–57)b 80 (35–469)b INR > 2.3 in 26% A 24%, B 54%, C 22% 11
IVC, inferior vena cava; NA, not available; ALT, alanine aminotransferase; ULN, upper limit of normal; INR, international normalized ratio. a Mean ± SD. b Median, range.
DIFFERENTIATION OF HEPATIC VEIN THROMBOSIS AND IVC THROMBOSIS As compared with pure hepatic vein thrombosis, IVC thrombosis is more common in the Far East; it runs a more indolent course and is more commonly complicated by hepatocellular carcinoma (Table 45-4).70 Thus, the recently proposed distinction of primary hepatic vein thrombosis (“true BCS”) from primary IVC thrombosis (“obliterative hepatocavopathy”) is well grounded for clinical and therapeutic purposes.70 However, causes of these two conditions are similar. Indeed, IVC thrombosis is generally associated with obstruction of a variable length of the hepatic veins, as illustrated in Figure 45-2, or may follow hepatic vein thrombosis.70,71 Moreover, comprehensive investigations show that underlying risk factors are similar for both sites.8,9,72 What remains to be elucidated is why the IVC is so frequently involved in the Far East and relatively spared in the west.
LOCAL FACTORS It is not known why, in the setting of a generalized thrombotic diathesis, thrombosis develops in the hepatic veins or the suprahepatic IVC. A local factor is found in fewer than 5% of cases, whereas such a factor is found in at least 25% of patients with portal vein thrombosis unrelated to cirrhosis or cancer. It has been proposed that the motion of the diaphragm can induce sufficient trauma to the endothelium of the IVC and large hepatic veins and that it can induce local activation of the coagulation process in susceptible patients. Indeed, IVC thrombosis may follow blunt abdominal trauma in patients with an underlying thrombotic risk factor.73 Furthermore, postmortem studies using inflation of the lungs have shown that, during inspiration, the descent of the diaphragm leads to compression of the termination of the large hepatic veins near the edges of the foramen where the IVC penetrates the
diaphragm.69 Indeed, in many patients, the terminal part of the large hepatic veins is the predominant site of involvement.
SIMULTANEOUS OBSTRUCTION VERSUS SEQUENTIAL INVOLVEMENT OF THE MAJOR HEPATIC VEINS Combining the available data from clinical, imaging, and pathological studies, it is clear that the usual scenario for primary BCS due to hepatic vein obstruction is that of the successive, albeit in a random order, progressive involvement of the hepatic veins. This scenario is illustrated in Figure 45-3. By contrast, the simultaneous, abrupt, formation of a thrombus in all three major veins appears to be extremely rare.
PATHOPHYSIOLOGY AND HISTOPATHOLOGY OF LIVER DAMAGE Former autopsy studies47,55 and more recent analyses in explanted liver at transplantation74,75 have allowed a detailed description of the hepatic lesions at an advanced stage of the disease. Liver biopsy provides information on earlier stages of the disease41,76–78 but with considerable sampling variation due to inhomogeneous distribution of the lesions.74,75 The obstruction of a single main hepatic vein is clinically silent.55 As depicted in Figure 45-4, the obstruction of two or three main hepatic veins produces two hemodynamic changes: an increased sinusoidal blood pressure and a reduced sinusoidal blood flow. Raised sinusoidal pressure explains liver enlargement, pain, and ascites through several mechanisms.
881
Section VII. Vascular Disease of the Liver
1
2
3
4
Figure 45-3. Scenario likely accounting for most cases of hepatic vein thrombosis. (1) The first step consists of the rapid or progressive formation of a thrombus, usually in one major hepatic vein, close to its ostium in the inferior vena cava. This event is followed by the formation of a short-length stenosis (as depicted) or by complete obliteration of the vein (not shown). Collaterals develop connecting the obstructed territories to patent hepatic or extrahepatic veins in the vicinity. (2) In the next step, there is a rapid or progressive development of a thrombus in another major hepatic vein. Short-length or diffuse stenosis again develops, together with collaterals. There may be atrophy of the parenchyma in the obstructed territory, with compensatory hypertrophy of the preserved territories (most commonly the caudate lobe). (3) A thrombus can then develop in a vein that had remained partially (as depicted) or completely (not shown) patent. (4) Alternatively a thrombus can develop in the portal vein. Probably depending on the velocity of the obstructive process, steps 1–2 can take place without notice. Symptoms would develop only when obstruction is rapid, or when most of the venous outflow tract is obstructed. Steps 3 and 4 are probably always associated with symptoms. The scenario where complete thrombosis develops simultaneously in all three major hepatic veins appears to be rare.
Hepatic venous outflow obstruction Increased sinusoidal pressure
Congestion Portal hypertension Ascites
Decreased liver perfusion
Centrilobular necrosis Fibrosis hepatic atrophy/failure
Figure 45-4. Pathophysiology of Budd–Chiari syndrome. Increased sinusoidal pressure upstream from the obstruction is almost constant but, due to compensatory mechanisms, not always very marked. Decreased hepatic perfusion occurs when there is abrupt obstruction but, due to compensatory mechanisms, it is usually transient. Congestion is likely to potentiate the effects of ischemia.
882
1. There is sinusoidal dilatation and congestion, predominantly in the central area of the hepatic lobules (Figure 45-4).79 2. The filtration of interstitial fluid increases, which leads to its passage through the liver capsule when the capacity for lymph drainage is exceeded. Usually, the filtrated fluid has a high protein content, due to the high permeability of sinusoidal wall to proteins. 3. Portal pressure increases, leading to transudation of fluid from the splanchnic bed into the abdominal cavity.80 Changes in total hepatic blood flow have not been well characterized. Indirect evidence suggests that, in the microcirculation of the areas with impaired outflow, perfusion with portal blood decreases while perfusion with arterial blood increases.81,82 These alterations are unevenly distributed. Ischemic-type damage to the liver cells (noninflammatory centrilobular cell necrosis as shown in Figure 45-5) is
Chapter 45 BUDD-CHIARI SYNDROME
Figure 45-5. Liver biopsy findings in a patient with recent Budd–Chiari syndrome (Masson’s trichrome). The lesions affect the centrilobular area whereas the immediate periportal area is well preserved. There is a loss of the hepatocytes. The sinusoidal spaces are dilated. There is no fibrosis. These features are only diagnostic of Budd–Chiari syndrome if cardiac or pericardial disease has been ruled out. (Courtesy of Prof. C. Degott, Pathology Department, Hôpital Beaujon.)
(200) (192)
ALT (x ULN)
120
50 20 10 5 2 6
12
20
30
Days Figure 45-6. Course of serum alanine aminotransferase (ALT) changes following presentation in Budd–Chiari syndrome patients with initial ALT levels > 5 ¥ upper limit of normal (ULN). There is a rapid decrease in ALT levels except in 3 patients (red lines) who, within weeks, died or were transplanted. Ischemia–reperfusion injury may explain a rapid decrease in most patients, whereas the absence of reperfusion may explain the progressive increase and poor outcome in some. ALT, alanine aminotransferase; ULN, upper limit of normal. (Denié C, Valla D, personal data.)
found in about 70% of cases.55 Reperfusion injury may participate in liver cell damage, which would explain the rapid return of alanine aminotransferase values to normal in the majority of patients (Figure 45-6).83 Centrilobular necrosis is likely worsened by congestion.84 When hepatic veins and portal veins are both obstructed in the same region, confluent cell loss occurs in regions supplied by preterminal and larger portal tracts.74 Liver failure resulting from ischemic liver cell necrosis is rarely fulminant. Within a few weeks of obstruction of the hepatic veins, fibrosis develops and can predominate either in the centrilobular area (when there is pure hepatic vein obstruction) or in the periportal area (when there is associated obstruction of the
Figure 45-7. Liver biopsy findings in a patient with long-standing Budd–Chiari syndrome (Masson’s trichrome). The parenchymal architecture is destroyed. Still, portal tracts (center bottom) and hepatic veins (upper left) are recognizable. There are regenerative nodules. The area extending from the hepatic veins to the portal tract is fibrotic and devoid of hepatocytes (“parenchymal extinction”). (Courtesy of Prof. C. Degott, Pathology Department, Hôpital Beaujon.)
portal veins).74,79 Within a few months, nodular regeneration may take place, predominantly in the periportal area,55,74,75,79 as illustrated in Figure 45-7. Cirrhosis can eventually develop but it is found in only 10–20% of patients coming to transplantation.55,74,75 Whereas there is evidence for increased sinusoidal pressure being a stable, permanent state, decreased hepatic perfusion appears inconstant. Indeed, centrilobular necrosis is lacking in 20–45% of patients.78,85 Moreover, at the time of presentation, alanine aminotransferase levels are normal in over half of the patients (as illustrated in Figure 45-8) and transiently elevated in most (Figure 45-6). Natural mechanisms able to compensate for decreased perfusion may therefore be involved. These mechanisms are illustrated in Figure 45-9. They include: (1) development of venous collateral channels bypassing the obstructed veins (obvious in Figures 45-2 and 45-10);86,87 (2) redistribution of portal flow from areas where outflow is impaired toward areas where outflow is preserved;81,82 (3) increased portal pressure; and (4) increased arterial flow. Superimposed thrombosis of the intrahepatic portal veins appears to be common at an advanced stage of BCS, which can be explained by the combination of a stagnant flow, and an underlying thrombophilic state. Obstruction of the extrahepatic portal vein is present in 10–20% of patients.85,88 Severe obliteration of intrahepatic portal veins (similar to those illustrated in Figure 45-11) can be found in over one-half of the cases of explanted livers.74,75 The areas where portal and hepatic veins are simultaneously obstructed undergo infarction (as illustrated in Figure 45-12) or parenchymal extinction (i.e., transformation into fibrotic areas devoid of parenchymal cells, as shown in Figure 45-7).74,75 Depending on whether only hepatic veins or both hepatic veins and portal vein are obstructed, bridging fibrosis can be found in a venovenous or in a portovenous or portoportal disposition, respectively.74,75 Because obstruction of the hepatic veins is usually asynchronous, atrophy of the areas of the liver affected early may coexist with the congested or hyperplastic areas that were affected more recently. In
883
Section VII. Vascular Disease of the Liver
85 N = 49
Number of patients
N = 115
Figure 45-8. Distribution of serum alanine aminotransferase (ALT) values at presentation with Budd–Chiari syndrome. The distribution is bimodal with a boundary at a value of 5 times the upper limit of normal (ULN). Most patients present with ALT < 5 ¥ ULN. They are more likely to have no pain, lower serum bilirubin, and higher prothrombin levels. They are less likely to have abdominal pain, fever, and extensive necrosis at liver biopsy. However, the extent of fibrosis is similar. (Denié C, Valla D, personal data.)
0 0
1
2
3
4
5
6
7
8
9
10 20 30 50 100 200 300 400
ALT (xULN)
Hepatic venous outflow obstruction
Decreased low pressure (portal) inflow Redistributed portal flow Increased portal pressure Increased arterial inflow
Hepatic vein collaterals
Restored hepatic perfusion
Figure 45-9. Compensatory mechanisms tend to circumvent the effects of hepatic venous outflow obstruction. The development of the collateral venous circulation is crucial in relieving sinusoidal pressure. Patients with asymptomatic Budd–Chiari syndrome have numerous large-sized collaterals that may reach the size of the normal hepatic veins. The collateral circulation also allows the mechanisms that tend to restore hepatic perfusion to operate, including: (1) intrahepatic redistribution of the portal venous inflow from the obstructed territories to the territories where the outflow is preserved or restored; (2) increased portal pressure which improves portal perfusion; and (3) increased arterial inflow, the “buffer response” to decreased portal inflow.
80% of the cases, the caudate lobe is hypertrophied, causing IVC stenosis.89,90 Caudate lobe enlargement is explained by its veins draining directly into the IVC caudal to the ostia of the main hepatic veins; they are preserved from the thrombotic process affecting the main hepatic veins. Preservation of this drainage allows for compensatory hypertrophy and also for serving as an outflow path for intrahepatic venous collaterals draining the obstructed lobes of the liver. A frequent, and odd, feature of long-standing BCS is the development of multiple large regenerative nodules, some of them resembling focal nodular hyperplasia, as illustrated in Figure 45-13.74,75,91,92 These nodules can be viewed as a response to a focal loss of portal perfusion and hyperarterialization in areas with preserved hepatic venous outflow.74,75
884
Figure 45-10. Short-length stenosis demonstrated at retrograde hepatic venography in a young female patient with occult polycythemia rubra. The intrahepatic collateral circulation is well seen, connecting the median hepatic vein with the right part of the liver.
There are several reports of hepatocellular carcinoma developing in BCS patients.93–98 However there are few data to quantify the risk of malignancy. It is noteworthy that in most of the reported cases, there was long-standing, asymptomatic occlusion of the IVC; the patients originated from Asia, Africa, or Jamaica; and cirrhosis was documented in all of them. The growth of cancer seems to be much slower in patients with BCS than in patients with hepatitis B virus-related chronic liver disease.95 There is no evidence, at present, that hepatocellular carcinoma resulted from the transformation of benign regenerative macronodules.75,91
MANIFESTATIONS AND COURSE Affected patients are 35 years on average, and most are female.85 The cardinal features of BCS in various geographical areas are presented in Table 45-1. Presentation varies from a picture of acute hepatic failure to an asymptomatic condition recognized fortuitously.86 Several classifications into acute, subacute, and chronic
Chapter 45 BUDD-CHIARI SYNDROME
A
B
Figure 45-11. Obstruction of the intrahepatic portal veins in patients with Budd–Chiari syndrome (BCS). (A) Medium-sized portal vein. There is marked subendothelial thickening. (B) Small-sized portal vein. The arrows point to the original outline of the portal vein. The original lumen is completely obstructed by cellular and fibrillar material. There are several recanalization channels. Such lesions affect about half the intrahepatic portal veins in explanted livers from BCS patients.
A
C
B
Figure 45-12. Intrahepatic portal vein thrombosis occurring in a Budd–Chiari syndrome patient. (A) After contrast injection, lack of opacification of the segmental portal vein indicates recent thrombosis (arrow). There is an associated infarct in the corresponding territory, seen as an unenhanced area (asterix). Rapid deterioration in this patient led to emergency liver transplantation. (B) Gross examination of the sliced explant confirmed recent portal vein thrombosis and hepatic infarction (arrows). (C) There is a well-circumscribed area of infarction (arrows) in the vicinity of a thrombosed portal vein (arrowhead). (Reproduced from Valla DC. The diagnosis and management of the Budd– Chiari syndrome: consensus and controversies. Hepatology 2003; 38:793–803, with permission.)
885
Section VII. Vascular Disease of the Liver
A
B
C
D
E
F
Figure 45-13. Focal nodular regenerative hyperplasia-like lesions in a Budd–Chiari syndrome liver. (A) T1-weighted magnetic resonance imaging (MRI) showing multiple hyperintense nodules (arrowhead). (B) T2-weighted MRI showing that the nodules are hypointense, some of them with a central scar (arrowhead); the inhomogeneous area of hyperintensity is congestive (asterix). (C) Hepatic arteriography showing arterialization: the hepatic artery (arrow) is larger than in the splenic artery; there are multiple hypervascular areas. (D) The sliced native liver at transplantation shows multiple nodules in an otherwise congestive parenchyma, some of them with a central scar (arrow). (E) The fixed liver slice shows multiple pale nodules, some of them harboring a central scar (arrow). (F) Low-power view of a nodule showing the central scar devoid of portal vein (arrow); the neighboring liver parenchyma is congestive with two thrombosed hepatic veins. (Reproduced from Valla DC. The diagnosis and management of the Budd–Chiari syndrome: consensus and controversies. Hepatology 2003; 38:793–803, with permission.)
886
Chapter 45 BUDD-CHIARI SYNDROME
Table 45-5. Presenting Forms of Budd–Chiari Syndrome (BCS): Definitions and Distributions Reference: n
Fulminant
Acute
Subacute
Chronic
115
n = 22
Acute portal hypertension. Massive increase in AST or ALT. Death within days
Ascites +++. Moderate increase in AST/ALT. Death within months
Ascites ++. Splenomegaly +. AST and ALT near normal. Moderate progression but variable
117
0 Acute symptoms with fulminant hepatic failure
27% Acute symptoms without fulminant hepatic failure
72% Duration of illness < 6 months. No evidence of cirrhosis
11% Severe pain. Distension. Jaundice. Ascites. Hepatomegaly. Encephalopathy. Severe hepatocellular dysfunction. In general, < 4 weeks’ duration 7% —
11% Pain. Distention. Tender hepatomegaly. Ascites. No encephalopathy
39% —
28% ALT or AST > 5 ¥ ULN. No combination of lobar atrophy/hypertrophy, no gross irregularity of the liver surface 7%b
— —
Ascites +. Malnutrition. AST and ALT increased. Cirrhosis. Death inevitable 0 >6 months’ duration. Evidence of portal hypertension and cirrhosis 39% Symptoms and signs of portal hypertension. Large nodular liver. Distended veins. Relatively preserved hepatocellular function 65%a° ALT or AST £ 5 ¥ ULN. Combination of lobar atrophy/hypertrophy, or gross irregularity of the liver surface 46%b
n = 44
41
n = 177
99
n = 72
—
—
AST, aspartate aminotransferase; ALT, alanine aminotransferase. a 20% of the patients with chronic BCS could recall an episode of acute BCS in the past. b An additional form, so-called acute-on-chronic, was defined by the following features: AST or ALT > 5 ¥ ULN (upper limit of normal), and combination of atrophy/hypertrophy of liver lobes, or gross irregularity of the liver surface. This form accounted for 47% of the patients.
presentation have been proposed because prognosis and management differ accordingly. These classifications are presented in Table 45-5. Schematically, the acute presentation is characterized by a short illness, abdominal pain and fever, ascites, marked elevation in serum aminotransferases and markedly decreased coagulation factors; whereas the chronic presentation is characterized by the indolent development of ascites, or portal hypertensive bleeding, with normal or mildly increased serum aminotransferases, and moderately decreased plasma coagulation factors; jaundice is uncommon. The chronic presentation is more frequent than the acute presentation. Except for descriptive purposes, however, the utility of these classifications is uncertain for several reasons.1 First, definitions have differed among the proposed classifications. Second, these classifications combine duration and severity while these two features can be dissociated. Third, the relationship between clinical presentation and hepatic parenchymal or hepatic venous lesions is not straightforward. Indeed, extensive parenchymal fibrosis is commonly found in patients with an acute presentation.41 However, combining clinical and pathological features may be of prognostic value. It was recently shown that an acute presentation in patients with anatomic features of long-standing disease is associated with a poor outcome as compared with patients presenting acutely but without evidence of long-standing disease or patients with a chronic presentation.99 Presentation appears to depend both on the extent and on the speed of the obstructive process (Figure 45-3).55 Thus, obstruction of only one major hepatic vein usually develops without symptoms; slow obstruction of two or three major veins produces a chronic presentation or, when accompanied with extensive collaterals, no symptoms at all;86 both rapid obstruction of at least two major veins,
and a fresh thrombus superimposed on a long-standing but partial obstruction, gives rise to an acute presentation.
DIAGNOSIS BCS should be suspected when ascites, liver enlargement, and upper abdominal pain are simultaneously present, or when intractable ascites contrasts with moderate alteration of liver tests; when liver disease occurs in a patient with known risk factors for thrombosis; or when liver disease remains unexplained after other common or uncommon causes have been excluded.
DIRECT EVIDENCE The diagnosis of BCS is established when an obstructed hepatic venous outflow tract is demonstrated. X-ray venography remains a reference for evaluation of the hepatic veins.100 Three patterns of opacification are regarded as specific at retrograde catheterization: 1. spreading out from the catheter tip wedged into a blocked vein, a fine “spider-web” network pattern without filling of venous radicals; 2. when there is incomplete occlusion of the hepatic veins, a coarse network of collateral veins which arch outward from the catheter tip and then come together again near the site of entry of the hepatic vein into the IVC (Figure 45-10); 3. a patent vein upstream from a stricture (Figure 45-10). Diagnostic pitfalls include failure to cannulate the hepatic vein ostia, and a distorted appearance of hepatic veins. These two features are
887
Section VII. Vascular Disease of the Liver
encountered in patients with cirrhosis of other origin. Direct percutaneous venography can show a localized obstruction in the vicinity of the ostia when the hepatic veins cannot be entered using retrograde cannulation. Inferior venacavography allows for demonstration of caval stenosis or occlusion (Figure 45-2). In many patients with pure hepatic vein thrombosis, the IVC appears narrowed at its intrahepatic portion, because of the enlargement of the caudate lobe. An increased pressure gradient across the IVC stenosis or the presence of a collateral cavacaval circulation are better indications of the impact of intrahepatic stenosis on caval hemodynamics than the apparent degree of stenosis. Sonography must be combined with color Doppler imaging and pulse Doppler analysis of hepatic vein wave form. The following features can be considered specific for hepatic vein obstruction: 1. a large hepatic vein appearing void of flow signal, or with a reversed, or turbulent flow; 2. large intrahepatic or subcapsular collaterals with continuous flow connecting to the hepatic veins or the diaphragmatic or intercostal veins (as illustrated in Figure 45-14); 3. a spider-web appearance usually located in the vicinity of hepatic vein ostia, together with the absence of a normal hepatic vein in the area; 4. an absent or flat hepatic vein wave form without fluttering; 5. a hyperechoic cord replacing a normal vein.82,101 Absence of visualization, or tortuosity of the hepatic veins by grayscale real-time sonography, albeit with flow signals at Doppler imaging, are common in BCS but not specific, being also observed in advanced cirrhosis of other causes. A distinctive feature, however, is the association with intrahepatic or subcapsular hepatic venous collaterals. This collateral circulation is the most sensitive feature for the diagnosis, being found in over 80% of cases of BCS.82,101 Magnetic resonance imaging (MRI) with spin-echo and gradientecho sequences, and intravenous gadolinium injection allows visualization of obstructed hepatic veins and IVC, intrahepatic, or subcapsular collaterals (as shown in Figure 45-15), as well as the spider-web network pattern.81,82 MRI, however, is not as effective
A
as sonography in demonstrating the intrahepatic collaterals. MRI does not allow for an easy determination of the flow direction as well. If computed X-ray tomography fails to visualize the hepatic veins, then this suggests hepatic vein obstruction. However, false-positive and indeterminate results are found in approximately 50% of cases.82
INDIRECT EVIDENCE Liver biopsy usually shows the characteristics of congestion of the sinusoids, liver cell loss, or fibrosis in the centrilobular area (Figures 45-5 and 45-7).2 The differential diagnoses for hepatic congestion are heart failure and constrictive pericarditis; and for centrilobular liver cell loss with mild to marked congestion, circulatory failure whatever its cause, and sinusoidal obstruction syndrome. Isolated perivenular fibrosis is encountered in alcoholic or diabetic patients.
Figure 45-14. Diagnostic features at color Doppler-ultrasonography in a patient with Budd–Chiari syndrome. A large collateral is seen in red (flow directed toward the probe), connecting the obstructed right hepatic vein (with bright walls) to the patent median hepatic vein, seen in blue (flow directed away from the probe).
B
Figure 45-15. Diagnostic features at magnetic resonance imaging in an asymptomatic patient with a fortuitous diagnosis of Budd–Chiari syndrome. T1-weighted sequences. There is an increased number of tubular (black) structures in an abnormal location. These structures are large-sized hepatic vein collaterals. (A) In the left lower corner, the right accessory hepatic vein is seen enlarged, connecting to the patent inferior vena cava (lower left of the panel). (B) In the center, a large subcapsular vein makes up for the left hepatic vein.
888
Chapter 45 BUDD-CHIARI SYNDROME
The absence of congestion in the centrilobular area is a strong argument against the diagnosis of hepatic vein thrombosis. Surprisingly, small vein thrombosis is relatively uncommon in biopsy specimens from most patients with BCS. The rare form of BCS due to involvement of the small hepatic veins with patent large veins is only recognized at liver biopsy.2 Differentiation of this form from sinusoidal obstruction syndrome is not always feasible on the basis of pathological examination.2,3 At a late stage, differentiation of hepatic vein thrombosis complicated by cirrhosis from cardiac cirrhosis, and from cirrhosis complicated by hepatic vein thrombosis, may become difficult on pathological grounds.102 Liver specimens can be obtained through transcapsular needle puncture under sonographic guidance, or during laparoscopy. Serious consideration should be given to the risk of bleeding from the puncture site in these patients who are likely to receive early anticoagulation or, urgently, thrombolytic therapy. Liver imaging can also provide indirect evidence for hepatic vein thrombosis. Based on quantitative measurements, hypertrophy of the caudate lobe is found in about 80% of patients.90 This hypertrophy is explained by the preservation of the multiple hepatic veins draining this lobe directly into the IVC. However, caudate lobe enlargement is also common in many cases of cirrhosis of other causes. Liver scintiscanning is non-specific and insensitive. Altered parenchymal perfusion pattern is seen at computed tomography or magnetic resonance following bolus intravenous contrast injection. The most characteristic pattern is early homogeneous central enhancement (particularly at the level of the caudate lobe) together with delayed patchy enhancement and prolonged retention of the contrast agent in the periphery of the liver.82 This pattern is suggestive but neither sensitive nor specific for BCS, being observed in many other situations where portal venous perfusion is compromised.103 Among the latter situations, constrictive pericarditis deserves a special mention because, clinically, it closely mimics hepatic venous obstruction, and diagnosis can be missed at echocardiography.104
DIAGNOSTIC WORKUP Color Doppler imaging combined with pulsed Doppler should be the first test performed, and magnetic resonance should only be used in patients in whom the diagnosis is unclear. The major advantages of the ultrasound study, beyond sensitivity and specificity, are relatively low cost, wide availability, complete lack of harm, and minimal technical difficulty. Limitations lie in the patient’s body habitus which may preclude complete sonographic evaluation, and in a lack of experience on the part of the examiner in diagnosing BCS. MRI is a minimally invasive investigation with no harm from contrast injection to kidney function. Techniques can be standardized and the results are not examiner-dependent. In a minority of patients, mostly cirrhotics, where uncertainty persists, the third investigation can be liver biopsy because it is expected to give information not only for a diagnosis of hepatic vein thrombosis but also for important differential diagnosis: sinusoidal obstruction syndrome, cirrhosis of other origin, and diffuse spreading of malignant cells within the microcirculation. When coagulation disorders preclude liver biopsy through the transcapsular approach, an attempt at retrograde cannulation of the hepatic veins for
venography and liver biopsy using the transjugular route should be performed. X-ray venography is no longer considered necessary for establishing the diagnosis.1 In planning treatment, however, X-ray venography remains a gold standard, permitting a precise delineation of outflow obstruction, which is facilitated by prior non-invasive imaging. Therefore, venography can be reserved for patients where interventional therapy is deemed necessary, and will allow for the performance of a decompressive intervention in the same session if needed.1 The deleterious effect of iodinated contrast injection in patients with renal insufficiency must be kept in mind.
THERAPY TREATMENT FOR THE UNDERLYING CONDITION Treatment for the underlying conditions is logical. Preventing extension of thrombosis into other hepatic veins, collaterals, and into the intrahepatic or extrahepatic portal venous system is aimed at preventing deterioration, and at keeping feasible all the therapeutic options that require a patent portal vein.65 The evidence for the efficacy of anticoagulation therapy is circumstantial, including: 1. an improved outcome since the introduction of systematic anticoagulation, in non-transplant as well as in transplant patients;78,105–107 2. reports on recanalization of thrombosed hepatic veins, and of thrombosed portal vein associated with thrombosed hepatic veins;108 3. the efficacy in patients with portal vein or other deep vein thrombosis.109 The type and duration of optimal anticoagulation have not been established. In most centers, heparin followed by long-term administration of a vitamin K antagonist have been used, whatever the underlying thrombotic risk factor. Low-molecular-weight heparins are generally preferred to unfractionated heparins because of a lower risk of heparin-induced thrombocytopenia and easier administration. Newer agents and alternative protocols require further evaluation.110
MANAGEMENT OF COMPLICATIONS Treatments recommended for portal hypertension and ascites in cirrhotic patients likely apply to BCS patients, albeit with some caution. For example, anticoagulant therapy increases the risk of bleeding from paracentesis. During active bleeding from esophageal varices, the reduction in splanchnic blood flow induced by exogenous vasoconstrictors might precipitate portal venous thrombosis in the patient with BCS. Lastly, anticoagulation therapy may increase the risk of endoscopic therapy for esophageal varices.
LIVER DECOMPRESSION As illustrated in Table 45-6, decompression aims to decrease sinusoidal pressure by restoring the outflow of blood from the liver by means of recanalizing the obstructed venous outflow, or by side-toside portacaval shunting. Recanalization can be attempted using thrombolytic therapy for recent thrombosis, or percutaneous angioplasty, or surgery for more long-standing disease. The benefit/risk ratio of thrombolysis is still unclear. From the limited experience thus far reported, in situ thrombolysis combined with angioplasty
889
Section VII. Vascular Disease of the Liver
Table 45-6. Therapy for Budd–Chiari Syndrome Recanalization of obstructed outflow tract
Decompression through side-to-side portosystemic shunting
Recent thrombosis Short-length stenosis Stenosis + recent thrombosis Patent IVC
Obliterated or compressed IVC
Compressed IVC Replacement of the liver and its vessels
Thrombolysis (local or systemic) Angioplasty ± stenting Angioplasty ± stenting + local thrombolysis Surgical • Portacaval • Mesocaval TIPS Surgical • Mesoatrial • Mesoinnominate • Portacaval + cavoatrial TIPS Liver transplantation
IVC, inferior vena cava; TIPS, transjugular intrahepatic portosystemic stent shunt. Rationale for the various forms of therapy for primary Budd–Chiari syndrome. Recanalization procedures (thrombolysis, angioplasty, stenting) would relieve the outflow block. Side-to-side portosystemic shunting (surgery or TIPS) would decompress the liver and improve perfusion through the hepatic artery. Transplantation would both correct the block and replace the failing liver.
for a pre-existing stenosis appears superior to thrombolysis alone, whether systemic or in situ.107,111 Percutaneous angioplasty, primarily or secondarily associated with stenting, has achieved long-term patency rates of the order of 80–90% both for IVC and for hepatic vein thrombosis.107,112,113 An immediate relief of symptoms has been reported. Rethrombosis and restenosis are amenable to in situ thrombolysis and reangioplasty, respectively. Data on long-term effects are encouraging for IVC obstruction but are still scarce for hepatic vein obstruction. The indications for primary stenting need to be clarified. Surgical angioplasty and the hepatoatrial anastomosis (Senning’s procedure)114 have been abandoned with the development of percutaneous maneuvers. Side-to-side portacaval shunting aims to transform the portal vein into an outflow tract, with the expectation that loss of portal inflow will be compensated for by an increased arterial flow. When the IVC is patent, porta- or mesocaval shunts have given better results than other types of surgical shunts.76,115–120 Interposition of a venous or prosthetic graft is usually necessary. Reported in-hospital mortality averages 20%, which may be explained by the poor general condition of some patients. In an experimental model, however, studies on sham operation and portacaval shunting suggest that surgery by itself is particularly risky in the setting of BCS.121 Most surviving patients have no ascites and improved liver function at 1 year of follow-up. Reversal of fibrosis has been reported.76,122 Shunt dysfunction, occurring in about 30% of patients, can result from thrombosis, stenosis, or caval compression by the caudate lobe.123 The outcome of the patient with shunt dysfunction is poor. Stenosis can be amenable to percutaneous stenting.124 Hemodynamically significant compression of intrahepatic IVC can be corrected with the insertion of a stent before or after porta- or mesocaval shunting.125 In patients with obliteration of the IVC at its terminal portion, or severe compression at its intrahepatic portion, when percutaneous maneuvers were not possible, long prosthetic interposition grafts have been used to construct porto- or mesoatrial, cavoatrial with portoatrial, or mesoinnominate anastomoses.120,122 Postshunt encephalopathy seems to be rare.76 The impact of surgical shunting on survival remains unclear. Two multivariate analyses failed to show
890
a survival advantage in surgically shunted patients as compared to patients receiving only medical treatment;78,99 however, a benefit was suggested in a third report.85 High operative mortality and late shunt dysfunction could cancel the long-term beneficial impact of surgical decompression. Except when terminal IVC is obliterated, transjugular intrahepatic portosystemic stent shunt (TIPS) appears to be technically successful in most cases of BCS.126–128 The shunt can be constructed through the suprahepatic IVC when no hepatic vein stump is available.129,130 Intrahepatic caval stenosis is bypassed by the TIPS. Insertion of a TIPS has been successful in patients with partial portal vein thrombosis. Secondary thrombosis or shunt dysfunction requiring revision occurs in about 70% of cases by 6 months.128,131 A recent report suggests improved 5-year survival as compared to the reported findings in similarly severe patients managed differently. Interestingly, TIPS dysfunction was not always associated with deterioration of the patient’s condition.128 However, another study found no survival difference between patients treated with surgical shunt or TIPS.85 TIPS insertion is more difficult and hazardous in BCS patients than in cirrhotic patients.127 Intrahepatic hematomas have been reported.132,133 Secondary stenosis of the terminal IVC has required stenting or liver transplantation.134 More experience is needed to evaluate this new technique, particularly in patients in a poor condition.
LIVER TRANSPLANTATION Liver transplantation has been used as an alternative to, or after failure of, surgical shunting.18,105,116,135,136 Ten-year survival rates are about 75%, on average, which is encouraging when compared with a survival rate below 55% in patients with severe BCS in recent reports on large cohorts.85,99 Further cohort studies, including transplant and non-transplant patients, are needed to evaluate the results according to baseline prognostic factors. The risk of recurrence of BCS is acceptably low when anticoagulant therapy is instituted early.18 The apparently low risk of exacerbating malignant transformation of an underlying myeloproliferative disorder because of the use of immunosuppression requires further assessment.18,136
Chapter 45 BUDD-CHIARI SYNDROME
Anticoagulation
Figure 45-16. Algorithm for the management of patients with Budd–Chiari syndrome according to the European group for the Study of Vascular Disorders of the Liver (EN-Vie consortium: www.envie-project.org).
Severe Manifestations
No Yes
Anticoagulation Medical therapy
Angioplasty Thrombolysis Stenting
Tips
Transplantation
MANAGEMENT In this area, where evidence is scarce, there appear to be some areas of consensus.1 First, treatment of the underlying condition and lifelong anticoagulation therapy should be initiated without delay. Second, patients in a stable condition without symptoms should not undergo any interventional therapy. Third, portosystemic shunting or transplantation should not be proposed to patients whose condition improves rapidly on medical therapy, as judged from recovery in liver function and easy control of ascites on low-salt diet and diuretics. In these patients, finding a short-length stenosis in the IVC or a large hepatic vein may prompt angioplasty with or without stenting. Currently, there is no established timeframe to define what a rapid improvement is. Reasonably, in patients with an acute presentation and severe liver disease it should be a matter of days, whereas in patients with a chronic presentation it could be a matter of weeks. Fourth, patients who do not steadily improve, or whose symptoms recur on medical therapy, should be considered for decompression. The order, or the combination, in which the various options for decompression should be proposed remains debated. Still, there appear to be some further areas of consensus.1 First, the possible need for prompt transplantation should be kept in mind because deterioration in the patient’s condition can be rapid. Stent placement should be in a position that will not hamper subsequent transplantation if needed. Preservation of renal function deserves attention so that optimal immune suppression regimens can be used early post-transplant. Patients are better managed at, or in close connection with, centers where all possible treatment options, including transplantation, are readily available. Second, TIPS insertion is currently preferred to surgical shunting because of a low operative mortality and because efficacy is not compromised by caudate lobe enlargement.127,128 Third, pharmacological thrombolysis and angioplasty or TIPS are usually considered together. Indeed, thrombolysis should always be available when angioplasty or TIPS insertion is undertaken because of the risk of early hepatic or portal venous thrombosis. Moreover, angioplasty or TIPS insertion immediately following thrombolysis is currently preferred to thrombolysis alone on the unproved basis that this will permit a high-velocity blood flow to be maintained, which in turn will prevent rethrombosis.111 Therefore, in order to reduce the risk of bleeding induced
by thrombolysis, all non-essential invasive procedures should be avoided when percutaneous intervention is considered. As to patients not responsive to medical therapy alone, there remain areas of dispute including: (1) whether TIPS or angioplasty should be used as a first-line procedure in patients still in a good condition; and (2) whether hepatic transplantation is an alternative to portosystemic shunting, a rescue operation after failed shunting, or a primary operation in patients with severe liver disease. Furthermore, the criteria for severe liver disease need to be clarified. The algorithm recently proposed by the European Group for the Study of Vascular Disorders of the Liver is presented in Figure 45-16.65 Finally, the problem patient is one with associated portal vein thrombosis who, on average, is in poor condition and in whom derivative procedures and transplantation will be impossible or hazardous. When portal vein thrombosis is recent, TIPS combined with thrombolysis may be the only option.111,137
SURVIVAL AND PROGNOSIS Data on the natural history of BCS are limited. In early studies where no therapy was administered, diagnosis was difficult and most cases were diagnosed postmortem. One-year mortality was estimated to be about 60%. Since these early days, however, noninvasive imaging now allows recognition of asymptomatic forms with an excellent spontaneous outcome in up to 20% of patients. Furthermore, various forms of therapy have been implemented. Large cohorts of patients have shown an overall 5 year-survival rate of 65–69%.78,85,99 Figure 45-17 shows the overall survival curve in the largest cohort of patients so far reported.85 There are two portions to the survival curve. The downward slope is steep during the first 2 years, which accounts for half of the mortality. Thereafter, the slope is less steep, but still, does not appear to plateau. The main causes of death in that cohort were liver failure (n = 17), postoperative multiorgan failure (n = 12), sepsis (n = 4), newly developed malignancy (n = 2), cardiovascular disease (n = 3), cerebrovascular accident (n = 2), variceal bleeding (n = 1), combinations of the above (n = 3), or unknown (n = 8). Recent multivariate analyses have assessed the prognostic significance of disease characteristics at the time of diagnosis as well as
891
Section VII. Vascular Disease of the Liver
Table 45-7. Prognostic Factors in Budd–Chiari Syndrome Patients. Results from Multivariate Analyses Reference
78
Encephalopathy Ascites
present versus absent present versus absent Score 1, 2, or 3a INR £ 2.3 versus INR mmol/l
Prothrombin < 2.3) Bilirubin Pugh score Creatinine Acute-on-chronic form ALT Prognostic indexb Portosystemic shunting
P
2.11
0.04
1.33 —
mmol/l present versus absent IU/l yes versus no
77
RR
—
0.005 0.29
0.344
RR
99 P
NA
0.009
NA
0.008
NA
0.008
RC
85 P
2.15
0.006
1.26
0.0001
RR
P
3.58 2.83
< 0.001 0.08
2.05 1.004
0.02 0.7
RR, risk ratio; RC, regression coefficient; INR, international normalized ratio; ALT, alanine aminotransferase. a Ascites score: 1, absent without diuretics; 2, absent on diuretics; 3, refractory. b Prognostic index (PI) is according to Zeitoun et al.78: PI = 0.75 ascites score + 0.28 Pugh score + 0.037 age + 0.0036 creatinine.
Transplant-free survival
1
Table 45-8. Prognostic Scores for Budd-Chiari Syndrome
82%
1984-2000 (n=237) Langlet et al.99
69% 0.75
62%
59% Murad et al.85
0.50
INR, international normalized ratio. a Ascites score: 1, absent without diuretics; 2, absent on diuretics; 3, refractory. b 0, absent; 1, present.
0.25
0 0 1
5
10
15
20
Years Figure 45-17. Transplant-free survival in Budd–Chiari syndrome (BCS) patients. This multicenter cohort of western patients with a diagnosis of BCS made between 1984 and 2000 was followed up until January 2001. (After Murad SD, Valla DC, de Groen PC, et al. Determinants of survival and the effect of portosystemic shunting in patients with Budd–Chiari syndrome. Hepatology 2004; 39:500–508.)
treatment-related variables. Their results, and the prognostic indices that were derived, are presented in Table 45-7. Several points should be kept in mind before extrapolating these data. First, most of the patients included in these studies were from western countries. Occlusion of the suprahepatic IVC is less common than hepatic vein thrombosis in the west than in Africa or in Asia. The clinical features of BCS related to IVC occlusion differ from those of hepatic vein obstruction. However, the level of obstruction was not found to affect survival in western studies.78,85 Moreover, causal factors may differ in western and eastern patients, although this view has not been substantiated. Second, the impact of causal factors on prognosis has not been adequately assessed due to a lack of systematic investigations. Some studies have suggested that patients with Behçet’s disease138 have a worse outcome than those with the
892
0.95 ascites score (1, 2, or 3)a + 0.35 Pugh score + 0.047 age (years) +0. 0045 creatinine (mmol/l) +2.2 acute-onchronic form (0 or 1)b -0.26 1.27 encephalopathy (0 or 1)b +1.04 ascites (0 or 1)b +0.72 prothrombin (INR) +0.004 bilirubin (mmol/l)
other, more common, causes for BCS. The excess mortality could be related to extrahepatic involvement. Third, the cohorts have been constituted over periods of 20 years or more, a lapse of time during which many aspects of the management have evolved. Fourth, the two most recent studies85,99 are partly redundant with two others,77,78 since a part of their patients had been included in the latter reports. Redundancy may partially explain the similarities of their results. These multivariate analyses brought significant information for therapy (discussed above) and prognosis (presented in Tables 45-7 and 45-8). All studies have identified Child–Pugh score or its components as the major prognostic variables. None of the four studies found liver biopsy data to be of independent prognostic value once adjustment for Child–Pugh score was performed. Other data suggest that portal vein thrombosis is a factor for a poor outcome in BCS patients.74,75,88 However, in multivariate analysis adjusting for the Child–Pugh score component, portal vein thrombosis fell far from having an independent prognostic value (P = 0.9).85 This does not mean that superimposed portal vein thrombosis is not a factor in deterioration. Rather, this finding can be interpreted as indicating that portal vein thrombosis, by inducing deterioration in the Child–Pugh score components, loses its independent prognostic value. Further studies addressing this issue are needed. In the study that evaluated a clinicopathological classification, the acute-onchronic form had an extremely poor outcome as compared to the purely acute or purely chronic forms. The independent value
Chapter 45
Transplant-free survival
BUDD-CHIARI SYNDROME
1
9.
0.75
10. 11.
0.50
12. 0.25
Class I (n = 55) Class II (n = 95) Class III (n = 55)
13.
0 0
5
10
15
20
Years Figure 45-18. Survival in Budd–Chiari syndrome patients according to the prognostic classification of Murad et al. See Table 45-7 for a description of the score. Class I includes patients with a score between 0 and 1.1; class II, patients with a score 1.1–1.5; and class III, 1.5 and higher. (After Murad SD, Valla DC, de Groen PC, et al. Determinants of survival and the effect of portosystemic shunting in patients with Budd–Chiari syndrome. Hepatology 2004; 39:500–508.)
14.
15.
16.
persisted after adjustment for other prognostic variables.99 Most importantly, these studies have allowed a better appreciation of the extreme heterogeneity of BCS with regard to outcome. This is illustrated in Figure 45-18. A subgroup of patients with a 15-year survival of about 90% can be identified at the time of diagnosis; and, at the other end of the spectrum, a subgroup with a 5-year survival only 49% can also be identified.85
17. 18. 19.
20.
REFERENCES 1. Janssen HL, Garcia-Pagan JC, Elias E, et al. Budd–Chiari syndrome: a review by an expert panel. J Hepatol 2003; 38:364–371. 2. Ludwig J, Hashimoto E, McGill D, Heerden JV. Classification of hepatic venous outflow obstruction: ambiguous terminology of the Budd–Chiari syndrome. Mayo Clin Proc 1990; 65:51–55. 3. DeLeve LD, Shulman HM, McDonald GB. Toxic injury to hepatic sinusoids: sinusoidal obstruction syndrome (venoocclusive disease). Semin Liver Dis 2002; 22:27–42. 4. Okuda H, Yamagata H, Obata H, et al. Epidemiological and clinical features of Budd–Chiari syndrome in Japan. J Hepatol 1995; 22:1–9. 5. Shrestha SM, Okuda K, Uchida T, et al. Endemicity and clinical picture of liver disease due to obstruction of the hepatic portion of the inferior vena cava in Nepal. J Gastroenterol Hepatol 1996; 11:170–179. 6. Hirshberg B, Shouval D, Fibach E, et al. Flow cytometric analysis of autonomous growth of erythroid precursors in liquid culture detects occult polycythemia vera in the Budd–Chiari syndrome. J Hepatol 2000; 32:574–578. 7. Janssen HL, Meinardi JR, Vleggaar FP, et al. Factor V Leiden mutation, prothrombin gene mutation, and deficiencies in coagulation inhibitors associated with Budd–Chiari syndrome and portal vein thrombosis: results of a case-control study. Blood 2000; 96:2364–2368. 8. Bhattacharyya M, Makharia G, Kannan M, et al. Inherited prothrombotic defects in Budd–Chiari syndrome and portal vein
21.
22.
23.
24. 25.
26.
27.
28.
thrombosis: a study from North India. Am J Clin Pathol 2004; 121:844–847. Mohanty D, Shetty S, Ghosh K, et al. Hereditary thrombophilia as a cause of Budd–Chiari syndrome: a study from western India. Hepatology 2001; 34:666–670. Deltenre P, Denninger MH, Hillaire S, et al. Factor V Leiden related Budd–Chiari syndrome. Gut 2001; 48:264–268. Denninger MH, Chait Y, Casadevall N, et al. Cause of portal or hepatic venous thrombosis in adults: the role of multiple concurrent factors. Hepatology 2000; 31:587–591. Anger BR, Seifried E, Scheppach J, Heimpel H. Budd–Chiari syndrome and thrombosis of other abdominal vessels in the chronic myeloproliferative diseases. Klin Wochenschr 1989; 67:818–825. Wanless IR, Peterson P, Das A, et al. Hepatic vascular disease and portal hypertension in polycythemia vera and agnogenic myeloid metaplasia: a clinicopathological study of 145 patients examined at autopsy. Hepatology 1990; 12:1166–1174. De Stefano V, Teofili L, Leone G, Michiels JJ. Spontaneous erythroid colony formation as the clue to an underlying myeloproliferative disorder in patients with Budd–Chiari syndrome or portal vein thrombosis. Semin Thromb Hemost 1997; 23:411–418. Valla D, Casadevall N, Lacombe C, et al. Primary myeloproliferative disorder and hepatic vein thrombosis. A prospective study of erythroid colony formation in vitro in 20 patients with Budd–Chiari syndrome. Ann Intern Med 1985; 103:329–334. Dayal S, Pati HP, Pande GK, et al. Multilineage hemopoietic stem cell defects in Budd Chiari syndrome. J Hepatol 1997; 26:293–297. Vons C, Smadja C, Bourstyn E, et al. Results of portal systemic shunts in Budd–Chiari syndrome. Ann Surg 1986; 203:366–370. Srinivasan P, Rela M, Prachalias A, et al. Liver transplantation for Budd–Chiari syndrome. Transplantation 2002; 73:973–977. Socie G, Mary JY, de Gramont A, et al. Paroxysmal nocturnal haemoglobinuria: long-term follow-up and prognostic factors. French Society of Haematology. Lancet 1996; 348:573–577. Valla D, Dhumeaux D, Babany G, et al. Hepatic vein thrombosis in paroxysmal nocturnal hemoglobinuria. A spectrum from asymptomatic occlusion of hepatic venules to fatal Budd– Chiari syndrome. Gastroenterology 1987; 93:569–575. Graham ML, Rosse WF, Halperin EC, et al. Resolution of Budd–Chiari syndrome following bone marrow transplantation for paroxysmal nocturnal haemoglobinuria. Br J Haematol 1996; 92:707–710. Bahr MJ, Schubert J, Bleck JS, et al. Recurrence of Budd–Chiari syndrome after liver transplantation in paroxysmal nocturnal hemoglobinuria. Transpl Int 2003; 16:890–894. Offenstadt G, Desaint B, Conard J, et al. (Budd–Chiari syndrome with massive thrombosis of the subdiaphragmatic venous system and major fibrinopenia (author’s translation).) Nouv Presse Med 1982; 11:2139–2141. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999; 353:1167–1173. Espinosa G, Font J, Garcia-Pagan JC, et al. Budd–Chiari syndrome secondary to antiphospholipid syndrome: clinical and immunologic characteristics of 43 patients. Medicine (Baltimore) 2001; 80:345–354. Pelletier S, Landi B, Piette JC, et al. Antiphospholipid syndrome as the second cause of non-tumorous Budd–Chiari syndrome. J Hepatol 1994; 21:76–80. Bismuth E, Hadengue A, Hammel P, Benhamou JP. Hepatic vein thrombosis in Behçet’s disease. Hepatology 1990; 11:969–974. Bayraktar Y, Balkanci F, Bayraktar M, Calguneri M. Budd–Chiari syndrome: a common complication of Behçet’s disease. Am J Gastroenterol 1997; 92:858–862.
893
Section VII. Vascular Disease of the Liver
29. Saatci I, Ozmen M, Balkanci F, et al. Behçet’s disease in the etiology of Budd–Chiari disease. Angiology 1993; 44:392–398. 30. Walker M. Idiopathic hypereosinophilia associated with hepatic vein thrombosis. Arch Intern Med 1987; 147:2220–2221. 31. Elouaer-Blanc L, Zafrani ES, Farcet JP, et al. Hepatic vein obstruction in idiopathic hypereosinophilic syndrome. Arch Intern Med 1985; 145:751–753. 32. Zylberberg H, Valla D, Viguie F, Casadevall N. Budd–Chiari syndrome associated with 5q deletion and hypereosinophilia. J Clin Gastroenterol 1996; 23:66–68. 33. Russi EW, Bansky G, Pfaltz M, et al. Budd–Chiari syndrome in sarcoidosis. Am J Gastroenterol 1986; 81:71–75. 34. Young ID, Clark RN, Manley PN, et al. Response to steroids in Budd–Chiari syndrome caused by idiopathic granulomatous venulitis. Gastroenterology 1988; 94:503–507. 35. Saito T, Harada K, Nakanuma Y. Granulomatous phlebitis of small hepatic vein. J Gastroenterol Hepatol 2002; 17:1334–1339. 36. Martinez F, Berenguer M, Prieto M, et al. Budd–Chiari syndrome caused by membranous obstruction of the inferior vena cava associated with coeliac disease. Dig Liver Dis 2004; 36:157–162. 37. El Younsi S, Nassif T, Kuoch V, et al. (Association of Budd–Chiari syndrome and celiac disease.) Gastroenterol Clin Biol 2003; 27:236–237. 38. Junge U, Wienke J, Schuler A. Acute Budd–Chiari syndrome, portal and splenic vein thrombosis in a patient with ulcerative colitis associated with antiphospholipid antibodies and protein C deficiency. Z Gastroenterol 2001; 39:845–852. 39. Kraut J, Berman JH, Gunasekaran TS, et al. Hepatic vein thrombosis (Budd–Chiari syndrome) in an adolescent with ulcerative colitis. J Pediatr Gastroenterol Nutr 1997; 25:417–420. 40. Decaens T, Maitre S, Marfaing A, et al. (Inflammatory bowel disease and latent thrombocythemia: a novel cause of hepatic vein thrombosis.) Gastroenterol Clin Biol 2004; 28:394–397. 41. Dilawari JB, Bambery P, Chawla Y, et al. Hepatic outflow obstruction (Budd–Chiari syndrome). Experience with 177 patients and a review of the literature. Medicine (Baltimore) 1994; 73:21–36. 42. Khuroo MS, Datta DV. Budd–Chiari syndrome following pregnancy. Report of 16 cases, with roentgenologic, hemodynamic and histologic studies of the hepatic outflow tract. Am J Med 1980; 68:113–121. 43. Singh V, Sinha SK, Nain CK, et al. Budd–Chiari syndrome: our experience of 71 patients. J Gastroenterol Hepatol 2000; 15:550–554. 44. Valla D. Drug-induced vascular and sinusoidal lesions of the liver: Baillière’s Clin Gastroenterol 1988; 2:481–500. 45. Coche JC, Horsmans Y, Greer Y, Geubel AG. Budd–Chiari syndrome likely related to a slimming drug preparation. J Hepatol 1995; 23:626–627. 46. Valla D, Le MG, Poynard T, et al. Risk of hepatic vein thrombosis in relation to recent use of oral contraceptives. A case-control study. Gastroenterology 1986; 90:807–811. 47. Aikat BK, Bhusnurmath SR, Chhuttani PN, Datta DV. Hepatic vein obstruction – a retrospective analysis of 72 autopsies and biopsies. Indian J Med Res 1978; 67:128–144. 48. Gupta S, Blumgart LH, Hodgson HJ. Budd–Chiari syndrome: long-term survival and factors affecting mortality. Q J Med 1986; 60:781–791. 49. Fleiner-Hoffmann AF, Pfammatter T, Leu AJ, et al. Alveolar echinococcosis of the liver: sequelae of chronic inferior vena cava obstructions in the hepatic segment. Arch Intern Med 1998; 158:2503–2508. 50. Markert DJ, Shanmuganathan K, Mirvis SE, et al. Budd–Chiari syndrome resulting from intrahepatic IVC compression secondary to blunt hepatic trauma. Clin Radiol 1997; 52:384–387.
894
51. Fernandez-Gonzalez AL, Llorens R, Herreros JM, et al. Blunt traumatic rupture of the right hemidiaphragm and Budd–Chiari syndrome. Ann Thorac Surg 1994; 58:559–561. 52. Ambrosetti P, Widmann JJ, Robert J, Rohner A. (Acute Budd–Chiari’s syndrome after surgical treatment of polycystic liver disease.) Gastroenterol Clin Biol 1992; 16:894–896. 53. Benesch M, Urban C, Deutschmann H, et al. Management of Budd–Chiari syndrome by hepatic vein stenting after extended right hepatectomy. J Pediatr Surg 2002; 37:1640–1642. 54. Dousset B, Legmann P, Soubrane O, et al. Protein-losing enteropathy secondary to hepatic venous outflow obstruction after liver transplantation. J Hepatol 1997; 27:206–210. 55. Parker RGF. Occlusion of the hepatic veins in man. Medicine (Baltimore) 1959; 38:369–402. 56. Sigal E, Pogany A, Goldman IS. Marked hepatic congestion caused by a thoracoabdominal aneurysm. Gastroenterology 1984; 87:1367–1371. 57. Uddin W, Ramage JK, Portmann B, et al. Hepatic venous outflow obstruction in patients with polycystic liver disease: pathogenesis and treatment. Gut 1995; 36:142–145. 58. Schraut WH, Chilcote RR. Metastatic Wilms’ tumor causing acute hepatic-vein occlusion (Budd–Chiari syndrome). Gastroenterology 1985; 88:576–579. 59. Ciancio G, Soloway M. Renal cell carcinoma invading the hepatic veins. Cancer 2001; 92:1836–1842. 60. Saisse J, Hardwigsen J, Castellani P, et al. Budd–Chiari syndrome secondary to intracardiac extension of hepatocellular carcinoma. Two cases treated by radical resection. Hepatogastroenterology 2001; 48:836–839. 61. Carbonnel F, Valla D, Menu Y, et al. Acute Budd–Chiari syndrome as first manifestation of adrenocortical carcinoma. J Clin Gastroenterol 1988; 10:441–444. 62. Gowda RM, Gowda MR, Mehta NJ, et al. Right atrial extension of primary venous leiomyosarcoma: pulmonary embolism and Budd–Chiari syndrome at presentation – a case report. Angiology 2004; 55:213–216. 63. Rademaker J, Widjaja A, Galanski M. Hepatic hemangiosarcoma: imaging findings and differential diagnosis. Eur Radiol 2000; 10:129–133. 64. Anagnostopoulos GK, Margantinis G, Kostopoulos P, et al. Budd–Chiari syndrome and portal vein thrombosis due to right atrial myxoma. Ann Thorac Surg 2004; 78:333–334. 65. Janssen HL, Garcia-Pagan JC, Elias E, et al. Budd–Chiari syndrome: a review by an expert panel. J Hepatol 2003; 38:364–371. 66. Terabayashi H, Okuda K, Nomura F, et al. Transformation of inferior vena caval thrombosis to membranous obstruction in a patient with the lupus anticoagulant. Gastroenterology 1986; 91:219–224. 67. Sevenet F, Deramond H, Hadengue A, et al. Membranous obstruction of the inferior vena cava associated with a myeloproliferative disorder: a clue to membrane formation? Gastroenterology 1989; 97:1019–1021. 68. Kage M, Arakawa M, Kojiro M, Okuda K. Histopathology of membranous obstruction of the inferior vena cava in the Budd–Chiari syndrome. Gastroenterology 1992; 102:2081– 2090. 69. Valla D, Hadengue A, el Younsi M, et al. Hepatic venous outflow block caused by short-length hepatic vein stenoses. Hepatology 1997; 25:814–819. 70. Okuda K. Inferior vena cava thrombosis at its hepatic portion (obliterative hepatocavopathy). Semin Liver Dis 2002; 22:15–26. 71. Valla DC. Hepatic vein thrombosis (Budd–Chiari syndrome). Semin Liver Dis 2002; 22:5–14. 72. Dayal S, Pati HP, Pande GK, et al. Multilineage hemopoietic stem cell defects in Budd Chiari syndrome. J Hepatol 1997; 26:293–297.
Chapter 45 BUDD-CHIARI SYNDROME
73. Balian A, Valla D, Naveau S, et al. Post-traumatic membranous obstruction of the inferior vena cava associated with a hypercoagulable state. J Hepatol 1998; 28:723–726. 74. Tanaka M, Wanless IR. Pathology of the liver in Budd–Chiari syndrome: portal vein thrombosis and the histogenesis of venocentric cirrhosis, veno-portal cirrhosis, and large regenerative nodules. Hepatology 1998; 27:488–496. 75. Cazals-Hatem D, Vilgrain V, Genin P, et al. Arterial and portal circulation and parenchymal changes in Budd–Chiari syndrome: a study in 17 explanted livers. Hepatology 2003; 37:510–519. 76. Henderson JM, Warren WD, Millikan WJ, et al. Surgical options, hematologic evaluation, and pathologic changes in Budd–Chiari syndrome. Am J Surg 1990; 159:41–48; discussion 48–50. 77. Tang TJ, Batts KP, de Groen PC, et al. The prognostic value of histology in the assessment of patients with Budd–Chiari syndrome. J Hepatol 2001; 35:338–343. 78. Zeitoun G, Escolano S, Hadengue A, et al. Outcome of Budd–Chiari syndrome: a multivariate analysis of factors related to survival including surgical portosystemic shunting. Hepatology 1999; 30:84–89. 79. Akiyoshi H, Terada T. Centrilobular and perisinusoidal fibrosis in experimental congestive liver in the rat. J Hepatol 1999; 30:433–439. 80. Beattie C, Sitzmann JV, Cameron JL. Mesoatrial shunt hemodynamics. Surgery 1988; 104:1–9. 81. Menu Y, Sebag G, Vigrain V, et al. Budd–Chiari syndrome: MR evaluation. Diagn Interv Radiol 1990; 2:23–28. 82. Miller WJ, Federle MP, Straub WH, Davis PL. Budd–Chiari syndrome: imaging with pathologic correlation. Abdom Imaging 1993; 18:329–335. 83. Gonzalez-Flecha B, Reides C, Cutrin JC, et al. Oxidative stress produced by suprahepatic occlusion and reperfusion. Hepatology 1993; 18:881–889. 84. Henrion J. Ischemia/reperfusion injury of the liver: pathophysiologic hypotheses and potential relevance to human hypoxic hepatitis. Acta Gastrenterol Belg 2000; 63:336–347. 85. Murad SD, Valla DC, de Groen PC, et al. Determinants of survival and the effect of portosystemic shunting in patients with Budd–Chiari syndrome. Hepatology 2004; 39:500–508. 86. Hadengue A, Poliquin M, Vilgrain V, et al. The changing scene of hepatic vein thrombosis: recognition of asymptomatic cases. Gastroenterology 1994; 106:1042–1047. 87. Cho KJ, Geisinger KR, Shields JJ, Forrest ME. Collateral channels and histopathology in hepatic vein occlusion. AJR Am J Roentgenol 1982; 139:703–709. 88. Mahmoud AEA, Helmy AS, Billingham S, Elias E. Poor prognosis and limited therapeutic options in patients with Budd–Chiari syndrome and portal venous system thrombosis. Eur J Gastroenterol Hepatol 1997; 9:485–489. 89. Gupta S, Barter S, Phillips GW, et al. Comparison of ultrasonography, computed tomography and 99mTc liver scan in diagnosis of Budd–Chiari syndrome. Gut 1987; 28:242–247. 90. Mathieu D, Kracht M, Zafrani E, et al., eds. Advances in hepatobiliary radiology. St Louis: CV Mosby; 1990:3–28. 91. Vilgrain V, Lewin M, Vons C, et al. Hepatic nodules in Budd–Chiari syndrome: imaging features. Radiology 1999; 210:443–450. 92. Ibarrola C, Castellano VM, Colina F. Focal hyperplastic hepatocellular nodules in hepatic venous outflow obstruction: a clinicopathological study of four patients and 24 nodules. Histopathology 2004; 44:172–179. 93. Kage M. Budd–Chiari syndrome and hepatocellular carcinoma. J Gastroenterol 2004; 39:706–707. 94. Takamura M, Ichida T, Yokoyama J, et al. Recurrence of hepatocellular carcinoma 102 months after successful eradication and removal of membranous obstruction of the inferior vena cava. J Gastroenterol 2004; 39:681–684.
95. Shin SH, Chung YH, Suh DD, et al. Characteristic clinical features of hepatocellular carcinoma associated with Budd–Chiari syndrome: evidence of different carcinogenic process from hepatitis B virus-associated hepatocellular carcinoma. Eur J Gastroenterol Hepatol 2004; 16:319–324. 96. Jang JW, Yoon SK, Bae SH, et al. Rapidly progressing Budd–Chiari syndrome complicated by hepatocellular carcinoma. Korean J Intern Med 2003; 18:191–195. 97. Havlioglu N, Brunt EM, Bacon BR. Budd–Chiari syndrome and hepatocellular carcinoma: a case report and review of the literature. Am J Gastroenterol 2003; 98:201–204. 98. Matsui S, Ichida T, Watanabe M, et al. Clinical features and etiology of hepatocellular carcinoma arising in patients with membranous obstruction of the inferior vena cava: in reference to hepatitis viral infection. J Gastroenterol Hepatol 2000; 15:1205–1211. 99. Langlet P, Escolano S, Valla D, et al. Clinicopathological forms and prognostic index in Budd–Chiari syndrome. J Hepatol 2003; 39:496–501. 100. Wilson MW, Ring EJ, LaBerge JM, et al. Percutaneous transhepatic hepatic venography in the delineation and treatment of Budd–Chiari syndrome. J Vasc Interv Radiol 1996; 7:133–138. 101. Ohta M, Hashizume M, Tomikawa M, et al. Analysis of hepatic vein waveform by Doppler ultrasonography in 100 patients with portal hypertension. Am J Gastroenterol 1994; 89:170–175. 102. Wanless I, Liu J, Butany J. Role of thrombosis in the pathogenesis of congestive hepatic fibrosis (cardiac cirrhosis). Hepatology 1995; 21:1232–1237. 103. Mathieu D, Vasile N, Grenier P. Portal thrombosis: dynamic CT features and course. Radiology 1985; 154:737–741. 104. Solano FX, Young E, Talamo TS, Dekker A. Constrictive pericarditis mimicking Budd–Chiari syndrome. Am J Med 1986; 80:113–115. 105. Halff G, Todo S, Tzakis AG, et al. Liver transplantation for the Budd–Chiari syndrome. Ann Surg 1990; 211:43–49. 106. Min AD, Atillasoy EO, Schwartz ME, et al. Reassessing the role of medical therapy in the management of hepatic vein thrombosis. Liver Transpl Surg 1997; 3:423–429. 107. Barrault C, Plessier A, Valla D, Condat B. (Non surgical treatment of Budd–Chiari syndrome: a review.) Gastroenterol Clin Biol 2004; 28:40–49. 108. Schmets L, Hagege H, Merlet C, et al. (Porto-hepatic thrombosis, revealing paroxysmal nocturnal hemoglobinuria, followed by regression induced by heparin therapy.) Gastroenterol Clin Biol 1993; 17:955–958. 109. Condat B, Pessione F, Hillaire S, et al. Current outcome of portal vein thrombosis in adults: risk and benefit of anticoagulant therapy. Gastroenterology 2001; 120:490–497. 110. D’Amico EA, Villaca PR, Gualandro SF, et al. Successful use of Arixtra in a patient with paroxysmal nocturnal hemoglobinuria, Budd–Chiari syndrome and heparin-induced thrombocytopenia. J Thromb Haemost 2003; 1:2452–2453. 111. Sharma S, Texeira A, Texeira P, et al. Pharmacological thrombolysis in Budd Chiari syndrome: a single centre experience and review of the literature. J Hepatol 2004; 40:172–180. 112. Fisher NC, McCafferty I, Dolapci M, et al. Managing Budd–Chiari syndrome: a retrospective review of percutaneous hepatic vein angioplasty and surgical shunting. Gut 1999; 44:568–574. 113. Xu PQ, Dang XW. Treatment of membranous Budd–Chiari syndrome: analysis of 480 cases. Hepatobiliary Pancreat Dis Int 2004; 3:73–76. 114. Vogt PR, Andersson LC, Jenni R, et al. Dorsocranial liver resection and direct hepatoatrial anastomosis for hepatic venous outflow obstruction: long-term outcome and functional results. Am J Gastroenterol 1996; 91:539–544.
895
Section VII. Vascular Disease of the Liver
115. Bismuth H, Sherlock DJ. Portasystemic shunting versus liver transplantation for the Budd–Chiari syndrome. Ann Surg 1991; 214:581–589. 116. Shaked A, Goldstein RM, Klintmalm GB, et al. Portosystemic shunt versus orthotopic liver transplantation for the Budd–Chiari syndrome. Surg Gynecol Obstet 1992; 174:453–459. 117. Mahmoud AEA, Mendoza AEA, Meshokhes AN, et al. Clinical spectrum, investigations and treatment of Budd–Chiari syndrome. Q J Med 1996; 89:37–43. 118. Ahn SS, Yellin A, Sheng FC, et al. Selective surgical therapy of the Budd–Chiari syndrome provides superior survivor rates than conservative medical management. J Vasc Surg 1987; 5:28–37. 119. Pisani-Ceretti A, Intra M, Prestipino F, et al. Surgical and radiologic treatment of primary Budd–Chiari syndrome. World J Surg 1998; 22:48–53; discussion 53–54. 120. Kohli V, Pande GK, Dev V, et al. Management of hepatic venous outflow obstruction. Lancet 1993; 342:718–722. 121. Orloff MJ, Johansen KH. Treatment of Budd–Chiari syndrome by side-to-side portacaval shunt: experimental and clinical results. Ann Surg 1978; 188:494–512. 122. Orloff MJ, Daily PO, Orloff SL, et al. A 27-year experience with surgical treatment of Budd–Chiari syndrome. Ann Surg 2000; 232:340–352. 123. Panis Y, Belghiti J, Valla D, et al. Portosystemic shunt in Budd–Chiari syndrome: long-term survival and factors affecting shunt patency in 25 patients in western countries. Surgery 1994; 115:276–281. 124. Pelage JP, Denys A, Valla D, et al. Budd–Chiari syndrome due to prothrombotic disorder: mid-term patency and efficacy of endovascular stents. Eur Radiol 2003; 13:286–293. 125. Pisani-Ceretti A, Intra M, Prestipino F, et al. Surgical and radiologic treatment of primary Budd–Chiari syndrome. World J Surg 1998; 22:48–53; discussion 53–54. 126. Rossle M, Olschewski M, Siegerstetter V, et al. The Budd–Chiari syndrome: outcome after treatment with the transjugular intrahepatic portosystemic shunt. Surgery 2004; 135:394–403. 127. Mancuso A, Fung K, Mela M, et al. TIPS for acute and chronic Budd–Chiari syndrome: a single-centre experience. J Hepatol 2003; 38:751–754. 128. Perello A, Garcia-Pagan JC, Gilabert R, et al. TIPS is a useful long-term derivative therapy for patients with Budd–Chiari
896
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
syndrome uncontrolled by medical therapy. Hepatology 2002; 35:132–139. Soares GM, Murphy TP. Transcaval TIPS: indications and anatomic considerations. J Vasc Interv Radiol 1999; 10:1233–1238. Gasparini D, Del Forno M, Sponza M, et al. Transjugular intrahepatic portosystemic shunt by direct transcaval approach in patients with acute and hyperacute Budd–Chiari syndrome. Eur J Gastroenterol Hepatol 2002; 14:567–571. Cejna M, Peck-Radosavljevic M, Schoder M, et al. Repeat interventions for maintenance of transjugular intrahepatic portosystemic shunt function in patients with Budd–Chiari syndrome. J Vasc Interv Radiol 2002; 13:193–199. Fickert P, Trauner M, Hausegger K, et al. Intra-hepatic haematoma complicating transjugular intra-hepatic portosystemic shunt for Budd–Chiari syndrome associated with antiphospholipid antibodies, aplastic anaemia and chronic hepatitis C. Eur J Gastroenterol Hepatol 2000; 12:813–816. Hasegawa S, Eisenberg LB, Semelka RC. Active intrahepatic gadolinium extravasation following TIPS. Magn Reson Imaging 1998; 16:851–853. Turnes J, Garcia-Pagan JC, Gonzalez-Abraldes J, et al. Stenosis of the suprahepatic inferior vena cava as a complication of transjugular intrahepatic portosystemic shunt in Budd–Chiari patients. Liver Transpl 2001; 7:649–651. Ringe B, Lang H, Oldhafer KJ, et al. Which is the best surgery for Budd–Chiari syndrome: venous decompression or liver transplantation? A single-center experience with 50 patients. Hepatology 1995; 21:1337–1344. Saigal S, Norris S, Srinivasan P, et al. Successful outcome of orthotopic liver transplantation in patients with preexisting malignant states. Liver Transpl 2001; 7:11–15. Opitz T, Buchwald AB, Lorf T, et al. The transjugular intrahepatic portosystemic stent-shunt (TIPS) as rescue therapy for complete Budd–Chiari syndrome and portal vein thrombosis. Z Gastroenterol 2003; 41:413–418. Orloff LA, Orloff MJ. Budd–Chiari syndrome caused by Behçet’s disease: treatment by side-to-side portacaval shunt. J Am Coll Surg 1999; 188:396–407. Okuda H, Yamagata H, Obata H, et al. Epidemiological and clinical features of Budd–Chiari syndrome in Japan. J Hepatol 1995; 22:1–9.
Section VII: Vascular Disease of the Liver
46
SINUSOIDAL OBSTRUCTION SYNDROME (Hepatic Veno-occlusive Disease) Laurie D. DeLeve Abbreviations GSH glutathione MMP-9 matrix metalloproteinase-9 NO nitric oxide
RILD SOS
radiation-induced liver disease sinusoidal obstruction syndrome
INTRODUCTION Sinusoidal obstruction syndrome (SOS) is a disease that has been rediscovered a few times and more recently renamed. It was first described in cattle1,2 as a complication of pyrrolizidine ingestion and the first publication about the human disease came from South Africa in 1920.3 The liver toxicity from pyrrolizidine alkaloids was rediscovered in Jamaica4 and named hepatic veno-occlusive disease based on the easily recognizable feature of fibrous obliteration of the small hepatic venules in some cases.5 The iatrogenic form of SOS was first noted with the advent of chemotherapy,6,7 but the disease remained a sporadic complication of chemotherapy until the introduction of stem cell transplantation (the current terminology for what was formerly referred to as bone marrow transplantation).8–10 Early on this liver disease was already postulated to be of primary vascular origin.6,11 In recent years it has become clear that the disease is initiated in the hepatic sinusoids12,13 and that it may be present in the absence of hepatic venular involvement,14 prompting the change in name to sinusoidal obstruction syndrome.15 In addition to SOS, hepatic veno-occlusive lesions may be seen in radiation-induced liver disease (RILD), alcoholic liver disease, and, rarely, after liver transplantation.16–19 Some case reports of liver disease with veno-occlusive lesions after liver transplantation may be due to azathioprine-induced SOS, but there are case reports of “hepatic veno-occlusive disease” occurring in patients who did not receive azathioprine. These other forms of “hepatic veno-occlusive disease” have different pathophysiology from SOS due to chemotherapy and pyrrolizidine alkaloids, but ascites is a prominent feature of each of these diseases. This chapter will focus on SOS due to chemotherapy, with a brief discussion of RILD and SOS due to ingestion of plant alkaloids.
EPIDEMIOLOGY
TIPS
transjugular intrahepatic portosystemic shunt
pyrrolizidine alkaloids. In some non-western nations there are still sporadic cases of SOS due to ingestion of so-called bush teas containing pyrrolizidine alkaloids and epidemics in regions where inadequately winnowed wheat is contaminated by plants containing pyrrolizidine alkaloids. In western countries infrequent cases occur due to ingestion of pyrrolizidine alkaloids present in teas from Crotalaria, Sennecio, and Heliotropium.
CHEMOTHERAPY AND IMMUNOSUPPRESSIVE AGENTS Table 46-1 lists the chemotherapeutic or immunosuppressive agents most closely associated with SOS unrelated to myeloablative preparative regimens for stem cell transplantation. Gemtuzumab ozogamicin, used to treat acute myeloid leukemia, comprises the toxin calicheamicin coupled to a humanized monoclonal antibody to CD33, an antigen present on myeloblasts. Gemtuzumab ozogamicin has been linked to SOS, particularly in patients who have been exposed to the drug either before or after stem cell transplantation.20,21 The incidence has been reported to be particularly high when the interval between gemtuzumab ozogamicin and stem cell transplantation is less than 3.5 months.21 The most common setting for SOS due to actinomycin D has been in patients with nephroblastoma (Wilms tumor), in particular in patients with right-sided tumors who received both actinomycin D and abdominal irradiation.22,23 The thiopurines, azathioprine and 6-thioguanine, have been associated with three forms of liver injury caused by damage to hepatic endothelial cells: (1) SOS; (2) peliosis hepatitis; and (3) nodular regenerative hyperplasia.24–31 Azathioprine-associated SOS has been most commonly described after long-term immunosuppression for kidney or liver transplantation. 6-Thioguanine, a metabolite of azathioprine, has been associated with SOS in case reports of exposure to the drug in chemotherapeutic or immunosuppressive regimens.
PYRROLIZIDINE ALKALOIDS
RADIATION-INDUCED LIVER DISEASE
Until the advent of chemotherapy the only recognized cause of SOS was the ingestion of teas or foodstuffs contaminated with
RILD is a form of hepatic veno-occlusive disease that differs from SOS in the clinical presentation, the time course, and some histo-
897
Section VII. Vascular Disease of the Liver
Table 46-1. Standard Chemotherapy or Immunosuppressive Linked to Sinusoidal Obstruction Syndrome—Not in the Setting of Stem Cell Transplantation Actinomycin D ± abdominal irradiation Azathioprine Cytosine arabinoside Dacarbazine Gemtuzumab ozogamicin Mithramycin 6-Thioguanine Urethane
logical characteristics, but that has in common the characteristic veno-occlusive lesion. It is a complication that may occur when the mean liver dose of irradiation is greater than 31 Gy and when a larger portion of the liver is irradiated.32,33
SOS ASSOCIATED WITH STEM CELL TRANSPLANTATION Myeloablative stem cell transplantation for malignancy is the most common setting for SOS in North America and western Europe. The incidence of SOS varies greatly between transplantation centers and depends mainly on the choice of the preparative chemotherapeutic regimen, but also on diagnostic criteria for SOS and patient selection. The reported incidence ranges from 5 to 60%, with an overall case fatality rate of around 30%.34–37 The non-myeloablative stem cell transplantation regimens, which use reduced-intensity preparative chemotherapy, are an alternative approach to the traditional myeloablative regimens used in preparation for stem cell transplantation. SOS has been reported in some small series of non-myeloablative stem cell transplantations,38,39 but one large case series did not find any cases of SOS.40
PATHOGENESIS CLINICAL STUDIES The clinical presentation and histological features of SOS associated with stem cell transplantation have been very well characterized and therefore provide some insights into the mechanisms involved. Several findings demonstrate that the preparative regimen for stem cell transplantation initiates the injury in SOS. In stem cell transplantation a chemotherapeutic preparative regimen is given over several days followed by infusion of stem cells. The onset of symptoms of SOS can be as early as the day of infusion of stem cells, which indicates that it is the chemotherapy rather than the infusion of stem cells that initiates SOS. In cyclophosphamide-based regimens, cyclophosphamide metabolism is one of the risk factors for developing SOS,41 again demonstrating that the chemotherapy itself initiates the injury. The incidence of SOS varies dramatically between transplantation centers and the intensity of the chemotherapeutic regimen seems to be the greatest determinant of this variability. Unlike other intrinsic liver diseases, the manifestations of portal hypertension precede evidence of parenchymal disease. This indicates that SOS is a primary circulatory disorder. Careful analysis of
898
histology has shown that involvement of the hepatic veins is not essential to the development of clinical signs of SOS, although occlusion of the central veins is associated with more severe disease and the presence of ascites.14 Taken together, these findings suggest that the primary circulatory impairment occurs at the level of the sinusoid with exacerbation of the syndrome as the obstructive process extends into the central veins. The role of clotting in the pathogenesis of SOS has been one of the most controversial areas in the elucidation of the pathophysiology (see review by Korte42). Most of the currently available clinical evidence suggests that clotting is not involved. If intrahepatic coagulation contributes to the disease, the expected sequence of events would be damage to the sinusoidal and venular endothelium with initiation of coagulation by tissue factor. Damage to endothelial lining would certainly be expected to predispose to clotting. Studies have demonstrated increases in procoagulants and decreases in natural anticoagulants. The most consistent findings are decreases in protein C and factor VII levels. There is an association with low levels of protein C and factor VII prior to transplantation and the development of SOS and decreases have been seen in a number of studies following stem cell transplantation. However the low levels prior to transplantation may indicate underlying liver disease and the post-transplantation fall likely reflects hepatic dysfunction posttransplant. Increased fibrinogen is observed post-transplant, but this most likely occurs as an acute-phase response. Von Willebrand factor, factor VIII and tissue plasminogen activator increase after transplantation, which likely reflects endothelial damage. Immunohistochemical staining has demonstrated fibrinogen and factor VIII in the wall of the central veins.43 Given the disruption of the venular endothelial barrier and the accumulation of plasma proteins in the vessel wall, the presence of fibrinogen and factor VIII in the vessel wall likely reflects their presence in the blood as an acute-phase reactant and a marker of endothelial damage, respectively. There are reasons to believe coagulation does not play a role in the development of the disease. Fibrin was not observed in monocrotalineinduced SOS in humans44 and platelets could not be demonstrated by immunohistochemistry in SOS in stem cell transplantation patients.43 Prophylactic infusion of heparin or antithrombin III does not prevent SOS and thrombolytic therapy is beneficial in a minority of patients. Thus the preponderance of data in humans does not suggest that clotting is involved.
ANIMAL STUDIES In vitro studies have shown that hepatic sinusoidal endothelial cells are more susceptible than hepatocytes to various drugs that have been implicated in SOS clinically.26,45,46 In vivo studies have used a rat model of SOS induced by monocrotaline.12 Monocrotaline is a pyrrolizidine alkaloid, a class of plant compounds that cause SOS in humans who ingest these compounds in foods or herbal teas. The monocrotaline rat model shares the same characteristic signs, symptoms, and histology as the human disease.12 The earliest anatomic changes of SOS have been characterized by electron microscopy and in-vivo microscopy in this model.12,13 Kupffer cells disappear from the sinusoid, sinusoidal endothelial cells round up, and gaps form within and between sinusoidal endothelial cells. Red blood cells penetrate through the gaps in the endothelial barrier into the space of
Chapter 46 SINUSOIDAL OBSTRUCTION SYNDROME
Disse. With partial obstruction of the sinusoid by the rounded-up sinusoidal endothelial cells, the space of Disse becomes the pathway of least resistance and blood begins to flow within the space of Disse. Sinusoidal endothelial cells, stellate cells, and any remaining Kupffer cells are dissected off the space of Disse by the flow of blood and the sinusoid is largely denuded of lining cells. The sinusoidal cells embolize downstream, further obstructing sinusoidal flow. As the number of perfused sinusoids reaches a nadir, there is a parallel increase in necrosis of centrilobular hepatocytes. Mononuclear cells accumulate in the sinusoid and within the central vein and the mononuclear cell aggregates contribute to the sinusoidal obstruction. Fibrin was not identified by electron microscopy in a rat model of monocrotaline-induced SOS.12 In a rat model of monocrotalineinduced liver injury that is not SOS, fibrin has been demonstrated by immunohistochemistry47 and anticoagulants reduced the severity of hepatocyte necrosis.48 The biochemical underpinning of the changes described in the previous paragraph include changes in levels of nitric oxide (NO), glutathione (GSH), and matrix metalloproteinase-9 (MMP-9). Monocrotaline is P450-activated only within the liver.49–52 Hepatocytes, sinusoidal endothelial cells,26 and Kupffer cells all metabolize monocrotaline to monocrotaline pyrrole, the electrophilic metabolite. GSH is the major detoxification pathway for monocrotaline pyrrole26 and sinusoidal endothelial cells are selectively more susceptible than hepatocytes to toxicity because of weaker GSH defenses. One of the major adducts for monocrotaline pyrrole is the cytoskeletal protein, F-actin.53 Once sinusoidal endothelial glutathione is depleted, monocrotaline pyrrole binds to F-actin, leading to F-actin depolymerization.54 F-actin depolymerization leads to increased synthesis and activity of MMP-9.54 MMP-9 (gelatinase B) is exocytosed and digests extracellular matrix components. The depolymerization of the F-actin cytoskeleton allows sinusoidal endothelial cells to round up and the increased MMP-9 activity on the abluminal side of the sinusoidal endothelial cell digests extracellular matrix and allows the sinusoidal endothelial cell to let loose from the space of Disse. Normally there is tonic release of NO by sinusoidal endothelial cells and, to a lesser degree, by Kupffer cells.55 Basal release of NO reduces MMP-9 expression, whereas inhibition of NO synthesis increases cytokine-stimulated MMP-9 expression.56,57 As the sinusoidal endothelial cells and Kupffer cells disappear from the sinusoid, there is a parallel drop in hepatic vein NO. The decline in NO permits increased synthesis of MMP-9 by sinusoidal endothelial cells.55 MMP expression and activity are also regulated by redox status,58,59 so that the decline in sinusoidal endothelial cell GSH also permits up-regulation of MMP-9 activity. The events described here form a positive-feedback loop: F-actin depolymerization leads to up-regulation of sinusoidal endothelial cell MMP-9, MMP-9 up-regulation loosens the tethering of sinusoidal endothelial cells from the space of Disse, sinusoidal cells disappear from the sinusoid, and there is less NO from the sinusoidal endothelial cells and Kupffer cells to suppress MMP-9 synthesis, leading to the loss of more sinusoidal endothelial cells with further loss of NO production. The central role of the biochemical changes described in the pathogenesis of SOS are confirmed by the ability to prevent experimental SOS by infusion of GSH, of a liver-specific NO donor (V-PYRRO/NO), or of inhibitors of MMP-9.54,55,60
Table 46-2. Clinical Features Used to Diagnose Sinusoidal Obstruction Syndrome after Stem Cell Transplantation Seattle criteria34
Baltimore criteria61
Diagnosis requires two of three findings within 20 days of transplantation: Bilirubin >34.2 mmol/l (2 mg/dl) Hepatomegaly or right upper quadrant pain of liver origin >2% weight gain due to fluid accumulation
Hyperbilirubinemia plus two or more other criteria: Bilirubin >34.2 mmol/l (2 mg/dl) Hepatomegaly, usually painful ≥5% weight gain Ascites
CLINICAL FEATURES OF SOS AFTER MYELOABLATIVE THERAPY CLINICAL PRESENTATION The clinical presentation of SOS after myeloablative therapy for stem cell transplantation has been well characterized. The clinical features are tender hepatomegaly, hyperbilirubinemia, and fluid retention with weight gain. The diagnosis of SOS for patients who have undergone stem cell transplantation is based on these characteristic clinical features (Table 46-2) and diagnostic criteria have been published by investigators in Seattle and Baltimore.34,61 The Seattle criteria require two of three findings occurring within 20 days of transplantation: bilirubin >2 mg/dl, tender hepatomegaly, and >2% weight gain due to fluid accumulation. The Baltimore criteria require hyperbilirubinemia plus two of three other findings: bilirubin >2 mg/dl, (usually painful) hepatomegaly, >5% weight gain, and ascites. In myeloablative regimens that contain cyclophosphamide, features of SOS may present as early as day 0 (i.e., the day after the preparative regimen is completed and when the stem cells are administered). Most commonly, the onset of SOS is 10–20 days after completion of a cyclophosphamide-containing preparative regimen,34 whereas with other myeloablative regimens the symptoms may occur later.37,62,63 Thus the Seattle criteria, developed largely for patients receiving myeloablative regimens containing cyclophosphamide, have a temporal criterion that may not apply to non-cyclophosphamidecontaining regimens. In stem cell transplantation patients the major considerations in the differential diagnosis include sepsis-related cholestasis, hyperacute graft-versus-host disease, tumor infiltration, and cardiac failure. Cholestasis may also be caused by antibiotics, antifungals, ciclosporin, and parenteral nutrition. Although graft-versus-host disease usually occurs later than the typical time frame for SOS due to cyclophosphamide-containing regimens, hyperacute graft-versushost disease may be difficult to differentiate from SOS. Finally, congestive hepatopathy, e.g., due to cardiotoxic anticancer drugs, will also cause hepatomegaly, jaundice, and portal hypertension.
DIAGNOSTIC STUDIES Laboratory Studies Hyperbilirubinemia is a sensitive indicator of SOS, but jaundice post-transplantation may be due to other causes, such as sepsis,
899
Section VII. Vascular Disease of the Liver
acute graft-versus-host disease, ciclosporin, or hemolysis. The level of total serum bilirubin correlates with non-relapse mortality from various causes, including SOS in this population.64 Serum aminotransferases are elevated in SOS and peak weeks after the treatment with the chemotherapeutic regimen. Serum aspartate aminotransferase >750 U/l carries a poor prognosis.65,66 Therapeutic monitoring and dose adjustment of busulfan may decrease the incidence of SOS, although studies have not reported a significant improvement in survival or relapse rate.67–69 Adjustment of cyclophosphamide dosing based on therapeutic monitoring of the metabolite, carboxyethylphosphoramide mustard, may reduce mortality rates.41,70
Imaging Studies Ultrasonography can confirm hepatomegaly and ascites, can exclude tumor infiltration in the liver parenchyma and vasculature, and can detect biliary tract disease, but probably does not improve the diagnostic yield over and above the use of clinical criteria in the early phase of SOS.71–73 Early studies suggested that increased resistive index of the hepatic artery and decrease or reversal of flow in the portal vein by duplex ultrasonography might be indicators of SOS. Two prospective studies that examined patients before and after stem cell transplantation could not confirm that any of these indicators were diagnostic,71,72 although the presence of these ultrasound findings may be supportive of the diagnosis. Liver biopsy may be particularly helpful in distinguishing SOS from graft-versus-host-disease. Percutaneous or laparoscopic liver biopsy carries a high risk in patients with thrombocytopenia posttransplantation. The risk from transvenous liver biopsy is lower than that of percutaneous liver biopsy, but mortality due to delayed bleeding from capsular perforation, subcapsular hematomas, and hemorrhage at the venepuncture site has been reported to be as high as 5% in post-transplantation patients.74 Tissue specimens obtained by the transvenous route are smaller than those obtained percutaneously, but the transvenous approach also allows measurement of the hepatic venous pressure gradient. A hepatic venous pressure gradient >10 mmHg has a specificity of 90%. Histological Features. The characteristic histologic features of SOS are found in the centrilobular region of the liver. These features are sinusoidal congestion, hepatocyte necrosis, subendothelial or periadventitial fibrosis of the central vein, and sinusoidal fibrosis. Not all features need to be present to make the diagnosis, but the presence of more of the histologic abnormalities correlates with more severe clinical SOS.14 As mentioned earlier, involvement of the central vein was the hallmark feature first identified in this disease. However a significant number of patients who develop SOS after myeloablative therapy do not have involvement of the central vein.14
DIFFERENTIAL DIAGNOSIS Patients with SOS present with jaundice, hepatomegaly, and weight gain. As described above, there are various causes of hyperbilirubinemia in the transplant setting that lead to jaundice, but in isolation these will not cause weight gain and will not usually be associated with hepatomegaly. The symptoms of SOS may be mim-
900
icked by jaundice due to sepsis or cholestatic liver disease in association with intravenous hydration for hypotension causing weight gain or in association with congestive heart failure causing weight gain and hepatomegaly.
CLINICAL COURSE AND PROGNOSIS SOS is frequently classified as mild, moderate, or severe. Clinically, mild SOS is defined as SOS that requires no treatment and that resolves completely; moderate SOS requires medical therapy but resolves completely; and severe SOS requires treatment but does not resolve by day 100 or lead to the death of the patients. This classification is of course after the fact. Mortality from all causes by day 100 after infusion of stem cells in patients with mild, moderate, and severe SOS is 3, 20, and 98%, respectively.34 Using mathematical modeling, the severity of SOS due to cyclophosphamide-containing regimens can be predicted based on bilirubin levels and weight gain,75 but these models may not apply to regimens without cyclophosphamide. Poor prognostic features include higher serum transaminases, portal vein thrombosis, higher wedged hepatic venous pressure gradient, renal insufficiency, and hypoxia. The commonest cause of death from severe SOS is multiorgan failure, usually from renal and cardiopulmonary failure.
CLINICAL FEATURES OF SOS UNRELATED TO MYELOABLATIVE THERAPY SOS may also occur due to ingestion of pyrrolizidine alkaloids present in food sources, teas, or herbal supplements. Pyrrolizidine alkaloids are present in various botanically unrelated plant species such as Senecio, Crotalaria, and Heliotropium. In non-western nations, these pyrrolizidine alkaloids may contaminate inadequately winnowed wheat or may be present in bush teas. SOS after ingestion of pyrrolizidine alkaloids has a different time course than the chemoirradiation-induced syndrome: onset may be after 1–2 months of ongoing exposure and evidence of liver injury can persist for several months to years after the exposure. The features of SOS in this setting are hepatomegaly, ascites and, sometimes, abdominal pain, but jaundice is absent or mild. The histological features of SOS in this setting are sinusoidal congestion, hepatocyte necrosis, fibrotic occlusion of the central vein, and sinusoidal fibrosis. Another form of hepatic veno-occlusive disease is RILD. RILD occurs in adults after a mean liver dose of 31 Gy of conventional fractionation of irradiation.32,33 After partial hepatectomy the threshold dose of irradiation is lower. Three-dimensional radiation therapy treatment planning allows much higher doses of radiation to be delivered to the liver with a low incidence of RILD.76 Onset of RILD is usually 1–2 months after irradiation and the signs and symptoms of liver injury can persist for months. The clinical manifestations of RILD are hepatomegaly, ascites, and weight gain, but bilirubin elevations are minimal and the right upper quadrant pain is much less pronounced than in SOS after chemoirradiation. The histological features of RILD in the first 2 months after irradiation are centrilobular dropout of hepatocytes, congestion, and hemor-
Chapter 46 SINUSOIDAL OBSTRUCTION SYNDROME
rhage. In later months the atrophy of the hepatic cords persists, but congestion is minimal and there may be fibrotic occlusion of the central veins.77 Fibrin has been demonstrated by electron microscopy in the central veins after RILD,78 but not after SOS due to chemoirradiation or pyrrolizidine alkaloids.
PREVENTIVE STRATEGIES AFTER MYELOABLATIVE THERAPY The single most important strategy for preventing severe SOS is to avoid the use of high-risk myeloablative regimens in patients with major risk factors. The major risk factors for fatal SOS are: (1) SOS during a previous exposure to chemoirradiation; (2) a second stem cell transplant with myeloablative conditioning; (3) a short interval between gemtuzumab ozogamicin and stem cell transplantation;21 (4) viral hepatitis with elevated serum transaminases;34,79 and (5) extensive hepatic fibrosis or cirrhosis. The risk of SOS is substantially lower after non-myeloablative regimens.38,40,80,81 The commonest causes of hyperbilirubinemia after non-myeloablative regimens are graft-versus-host disease, sepsis, or a combination of both.40 In patients with hepatic fibrosis the risk for fatal liver decompensation remains high after non-myeloablative regimens.40 Although non-myeloablative regimens reduce overall hepatic toxicity, long-term studies will need to determine the effect on tumor relapse, graft-versus-host-disease, and infection. Use of prophylactic heparin to prevent SOS is standard practice in some stem cell transplantation centers. Prophylactic lowdose heparin can be used safely with careful monitoring of partial thromboplastin time;82,83 low-molecular-weight heparin is also safe in this setting.84,85 Two randomized studies showed a decrease in overall SOS with prophylactic low-dose heparin, but the studies were not powered to determine whether the incidence of fatal SOS was reduced.83,86 Four other studies did not find a reduction in overall SOS in patients treated with prophylactic heparin.82,87–89 Defibrotide is a large single-stranded polydeoxyribonucleotide that has numerous poorly understood effects. Defibrotide has an anticoagulant effect by reducing plasminogen activator inhibitor, increasing endothelial tissue plasminogen activator, and reducing endothelial expression of tissue factor. Defibrotide reduces leukocyte recruitment by reducing leukocyte rolling and adherence to endothelial cells. In two uncontrolled trials, prophylactic use of defibrotide reduced the incidence of SOS compared to historical controls,90,91 but this will need to be confirmed in randomized, controlled studies. Pentoxifylline, ursodeoxycholic acid, and prostaglandin E1 each initially showed promise as prophylactic agents. Several randomized controlled studies have found no effect of pentoxifylline.92,93 A large randomized study showed no benefit of ursodeoxycholic acid.94 Two studies found no effect of prostaglandin E1 in preventing fatal SOS.95,96 In summary, the best preventive strategy at present is to avoid high-risk myeloablative regimens in individuals at the greatest risk for SOS. There are currently no other prophylactic strategies that have been shown by randomized controlled studies to prevent fatal SOS.
TREATMENT FOR SOS AFTER MYELOABLATIVE THERAPY Some 70–80% of patients with SOS will survive without intervention or with conservative management. Fluid retention and ascites are treated with sodium restriction, diuretics, and therapeutic paracentesis as needed. In patients with multiorgan failure, hemodialysis and mechanical ventilation may be used, but these are unlikely to alter the outcome. Several interventions have been tried for severe SOS. Tissue plasminogen activator heparin infusion may improve the outcome in fewer than 30% of patients with severe SOS, but should not be used in patients at increased risk for pulmonary or intracerebral hemorrhage or in patients with renal or pulmonary failure. Defibrotide (see previous section) has been used therapeutically in uncontrolled trials of adults and children with severe SOS with multiorgan failure. The survival to day 100 post-transplantation in patients with severe SOS treated with defibrotide therapy was 35% (note: survival of severe SOS is historically around 5%).97,98 There have been case reports of improvement with treatment with Nacetylcysteine, prostaglandin E1, prednisone, topical nitrate, and vitamin E/glutamine, but these drugs have not been evaluated in formal studies. Transjugular intrahepatic portosystemic shunt (TIPS) has been used in SOS after stem cell transplantation to reduce portal pressure and to treat ascites.99–105 However, TIPS does not appear to alter the prognosis,101–103,106 and has been complicated by fatal acute respiratory distress syndrome.107 Two children who developed SOS underwent surgical portosystemic shunts that were effective in the treatment of ascites, but in both of these cases the surgical shunts were placed after resolution of liver dysfunction.108,109 It is uncommon for SOS to occur in the setting of stem cell transplantation for a benign disorder. When severe SOS occurs in a patient transplanted for a benign condition or in a patient with a malignancy with a good prognosis, liver transplantation may be an option.
CONCLUSIONS In North American and western Europe SOS most commonly occurs due to myeloablative chemotherapy prior to stem cell transplantation for malignancy. SOS is therefore an iatrogenic disease with a high case fatality rate. The incidence of the disease has declined in recent years as chemotherapy regimens, such as the nonmyeloablative regimens, have been chosen to avoid the risk of SOS. However it remains to be seen whether overall survival will improve or whether an increase in graft-versus-host disease, infection, and relapse mortality will offset the gains made by avoiding regimens with a high risk for SOS. Prophylactic medical interventions have not shown benefit and the main strategy to prevent SOS remains avoidance of high-risk regimens in individuals with risk factors for SOS. Management remains largely conservative. The most successful medical interventions that have been studied have shown efficacy in about one-third of patients in uncontrolled trials. Thus the challenge for the future
901
Section VII. Vascular Disease of the Liver
will be to use our improved understanding of the pathophysiology to devise novel prophylactic and/or therapeutic strategies.
REFERENCES 1. Pethick WH. Special report on Pictou cattle disease. Department of Agriculture, Canada, Health of Animals Branch. Ottawa: Canadian Government Printing Bureau; 1907. 2. Cushny AR. On the action of senecio alkaloids and the causation of the hepatic cirrhosis of cattle (Pictou, Molteno, or Winton disease). J Pharmacol Exp Ther 1911; 2:531–548. 3. Willmot FC, Robertson GW. Senecio disease, or cirrhosis of the liver due to senecio poisoning. Lancet 1920; 2: 848–849. 4. McFarlane AL, Branday W. Hepatic enlargement with ascitic children. Br Med J 1945; 1:838–840. 5. Bras G, Jeliffe DB, Stuart KL. Veno-occlusive disease of the liver with non-portal type of cirrhosis occurring in Jamaica. Arch Pathol 1954; 57:285–300. 6. Meacham GC, Tillotson FW, Heinle RW. Liver damage after prolonged urethane therapy. Am J Clin Pathol 1952; 22:22–27. 7. Ohler RL, Houghton JD, Moloney WC. Urethane toxicity; report of a case of hepatic necrosis apparently due to urethane. N Engl J Med 1950; 243:984–988. 8. Vowels MR, Lam Po Tang R, Zagars G, Ewing DP. Total body irradiation and Budd–Chiari syndrome. Pathology 1979; 11:306. 9. Berk PD, Popper H, Krueger GR, et al. Veno-occlusive disease of the liver after allogeneic bone marrow transplantation: possible association with graft-versus-host disease. Ann Intern Med 1979; 90:158–164. 10. Jacobs P, Miller JL, Uys CJ, Dietrich BE. Fatal veno-occlusive disease of the liver after chemotherapy, whole-body irradiation and bone marrow transplantation for refractory acute leukaemia. S Afr Med J 1979; 55:5–10. 11. Selzer G, Parker RGF. Senecio poisoning exhibiting as Chiari’s syndrome: a report on twelve cases. Am J Pathol 1950; 27:885–907. 12. DeLeve LD, McCuskey RS, Wang X, et al. Characterization of a reproducible rat model of hepatic veno-occlusive disease. Hepatology 1999; 29:1779–1791. 13. DeLeve LD, Ito I, Bethea NW, et al. Embolization by sinusoidal lining cell obstructs the microcirculation in rat sinusoidal obstruction syndrome. Am J Physiol Gastrointest Liver Physiol 2003; 284:G1045–G1052. 14. Shulman HM, Fisher LB, Schoch HG, et al. Veno-occlusive disease of the liver after marrow transplantation: histological correlates of clinical signs and symptoms. Hepatology 1994; 19:1171–1180. 15. DeLeve LD, Shulman HM, McDonald GB. Toxic injury to hepatic sinusoids: sinusoidal obstruction syndrome (venoocclusive disease). Semin Liver Dis 2002; 22:623–638. 16. Izaki T, Inomata Y, Asonuma K, et al. Early graft failure due to a veno-occlusive disease after a pediatric living donor liver transplantation. Pediatric Transplant 2004; 8:301–304. 17. Bernstine H, Mor E, Ben Ari Z, et al. Scintigraphic patterns of veno-occlusive disease in liver transplantation. Clin Liver Med 2004; 29:292–295. 18. Nakazawa Y, Chisuwa H, Mita A, et al. Life-threatening venoocclusive disease after living-related liver transplantation [see comment]. Transplantation 2003; 75:727–730. 19. Dhillon AP, Burroughs AK, Hudson M, et al. Hepatic venular stenosis after orthotopic liver transplantation. Hepatology 1994; 19:106–111. 20. Rajvanshi P, Shulman HM, Sievers EL, McDonald GB. Hepatic sinusoidal obstruction following gemtuzumab ozogamicin (Mylotarg®). Blood 2002; 99:2310–2314.
902
21. Wadleigh M, Richardson PG, Zahrieh D, et al. Prior gemtuzumab ozogamicin exposure significantly increases the risk of veno-occlusive disease in patients who undergo myeloablative allogeneic stem cell transplantation. Blood 2003; 102:1578–1582. 22. Tornesello A, Piciacchia D, Mastrangelo S, et al. Veno-occlusive disease of the liver in right-sided Wilms’ tumours. Eur J Cancer 1998; 34:1220–1223. 23. Czauderna P, Katski K, Kowalczyk J, et al. Veno-occlusive liver disease (VOD) as a complication of Wilms’ tumour management in the series of consecutive 206 patients. Eur J Pediatr Surg 2000; 10:300–303. 24. Haboubi NY, Ali HH, Whitwell HL, Ackrill P. Role of endothelial cell injury in the spectrum of azathioprine-induced liver disease after renal transplant: light microscopy and ultrastructural observations. Am J Gastroenterol 1988; 83:256–261. 25. Zafrani ES, Cazier A, Baudelot AM, Feldmann G. Ultrastructural lesions of the liver in human peliosis. A report of 12 cases. Am J Pathol 1984; 114:349–359. 26. DeLeve LD, Wang X, Kuhlenkamp JF, Kaplowitz N. Toxicity of azathioprine and monocrotaline in murine sinusoidal endothelial cells and hepatocytes: the role of glutathione and relevance to hepatic venooclusive disease. Hepatology 1996; 23:589–599. 27. DeLeve LD. Glutathione defense in non-parenchymal cells. Semin Liver Dis 1998; 18:403–413. 28. Satti MB, Weinbren K, Gordon-Smith EC. 6-thioguanine as a cause of toxic veno-occlusive disease of the liver. J Clin Pathol 1982; 35:1086–1091. 29. D’Cruz CA, Wimmer RS, Harcke HT, et al. Veno-occlusive disease of the liver in children following chemotherapy for acute myelocytic leukemia. Cancer 1983; 52:1803–1807. 30. Larrey D, Freneaux E, Berson A, et al. Peliosis hepatis induced by 6-thioguanine administration. Gut 1988; 29:1265–1269. 31. Dubinsky MC, Vasiliauskas EA, Singh H, et al. 6-thioguanine can cause serious liver injury in inflammatory bowel disease patients. Gastroenterology 2003; 125:298–303. 32. Dawson LA, Normolle D, Balter JM, et al. Analysis of radiationinduced liver disease using the Lyman NTCP model. [See comment.] Int J Radiat Oncol Biol Phys 2002; 53:810–821. Erratum appears in Int J Radiat Oncol Biol Phys 2002; 53:1422. 33. Dawson LA, Ten Haken RK, Lawrence TS. Partial irradiation of the liver. Semin Radiat Oncol 2001; 11:240–246. 34. McDonald GB, Hinds MS, Fisher LD, et al. Veno-occlusive disease of the liver and multiorgan failure after bone marrow transplantation – a cohort study of 355 patients. Ann Intern Med 1993; 118:255–267. 35. Bearman SI. The syndrome of hepatic veno-occlusive disease after marrow transplantation. Blood 1995; 85:3005–3020. 36. Richardson PG, Elias AD, Krishnan A, et al. Treatment of severe veno-occlusive disease with defibrotide: compassionate use results in response without significant toxicity in a high-risk population. Blood 1998; 92:737–744. 37. Lee JL, Gooley T, Bensinger W, et al. Veno-occlusive disease of the liver after busulfan, melphalan, and thiotepa conditioning therapy: incidence, risk factors, and outcome. Biol Blood Marrow Transplant 1999; 5:306–315. 38. Schetelig J, Kroger N, Held TK, et al. Allogeneic transplantation after reduced conditioning in high risk patients is complicated by a high incidence of acute and chronic graft-versus-host disease. Haematologica 2002; 87:299–305. 39. Slavin S, Nagler A, Naparstek E, et al. Nonmyeloablative stem cell transplantation and cell therapy as an alternative to conventional bone marrow transplantation with lethal cytoreduction for the treatment of malignant and nonmalignant hematologic diseases. Blood 1998; 91:756–763. 40. Hogan WJ, Maris M, Storer B, et al. Hepatic injury after nonmyeloablative conditioning followed by allogeneic
Chapter 46 SINUSOIDAL OBSTRUCTION SYNDROME
41.
42.
43.
44. 45.
46.
47.
48.
49. 50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
hematopoietic cell transplantation: a study of 193 patients. Blood 2004; 103:78–84. McDonald GB, Slattery JT, Bouvier ME, et al. Cyclophosphamide metabolism, liver toxicity, and mortality following hematopoietic stem cell transplantation. Blood 2003; 101:2043–2048. Korte W. Veno-occlusive disease of the liver after bone marrow transplantation: is hypercoagulability really part of the problem? Blood Coagul Fibrinolysis 1997; 8:367–381. Shulman HM, Gown AM, Nugent DJ. Hepatic veno-occlusive disease after bone marrow transplantation – immunohistochemical identification of the material within occluded central venules. Am J Pathol 1987; 127:549–558. Brooks SEH, Miller CG, McKenzie K, et al. Acute venoocclusive disease of the liver. Arch Pathol 1970; 89:507–520. DeLeve LD. Dacarbazine toxicity in murine liver cells: a novel model of hepatic endothelial injury and glutathione defense. J Pharmacol Exp Ther 1994; 268:1261–1270. DeLeve LD. Cellular target of cyclophosphamide toxicity in the murine liver: role of glutathione and site of metabolic activation. Hepatology 1996; 24:830–837. Copple BL, Banes A, Ganey PE, Roth RA. Endothelial cell injury and fibrin deposition in rat liver after monocrotaline exposure. Tox Sci 2002; 65:309–318. Copple BL, Woolley B, Banes A, et al. Anticoagulants prevent monocrotaline-induced hepatic parenchymal cell injury but not endothelial cell injury in the rat. Toxicol Appl Pharmacol 2002; 180:186–196. Mattocks AR. Toxicity of pyrrolizidine alkaloids. Nature 1968; 217:723–728. Lafranconi WM, Huxtable RJ. Hepatic metabolism and pulmonary toxicity of monocrotaline using isolated perfused liver and lung. Biochem Pharmacol 1984; 33:2479–2484. Mattocks AR, White INH. The conversion of pyrrolizidine alkaloids to n-oxides and to dihydropyrrolizidine derivatives by rat-liver microsomes in vitro. Chem-Biol Interactions 1971; 3:383–396. Hilliker KS, Garcia CM, Roth RA. Effects of monocrotaline and monocrotaline pyrrole on 5-hydroxytryptamine and paraquat uptake by lung slices. Res Commun Chem Pathol Pharmacol 1983; 40:179–197. Lamé MW, Jones AD, Wilson DW, et al. Protein targets of monocrotaline pyrrole in pulmonary artery endothelial cells. J Biol Chem 2000; 275:29091–29099. DeLeve LD, Wang X, Tsai J, et al. Prevention of sinusoidal obstruction syndrome (hepatic veno-occlusive disease) in the rat by matrix metalloproteinase inhibitors. Gastroenterology 2003; 125:882–890. DeLeve LD, Wang X, Kanel GC, et al. Decreased hepatic nitric oxide production contributes to the development of rat sinusoidal obstruction syndrome. Hepatology 2003; 38: 900–908. Eberhardt W, Beeg T, Beck KF, et al. Nitric oxide modulates expression of matrix metalloproteinase-9 in rat mesangial cells. Kidney Int 2000; 57:59–69. Eberhardt W, Huwiler A, Beck KF, et al. Amplification of IL-1 beta-induced matrix metalloproteinase-9 expression by superoxide in rat glomerular mesangial cells is mediated by increased activities of NF-kappa B and activating protein-1 and involves activation of the mitogen-activated protein kinase pathways. J Immunol 2000; 165:5788–5797. Tyagi SC, Ratajska A, Weber KT. Myocardial matrix metalloproteinase(s): localization and activation. Mol Cell Biochem 1993; 126:49–59. Tyagi SC, Kumar S, Borders S. Reduction–oxidation (redox) state regulation of extracellular matrix metalloproteinases and tissue inhibitors in cardiac normal and transformed fibroblast cells. J Cell Biochem 1996; 61:139–151.
60. Wang X, Kanel GC, DeLeve LD. Support of sinusoidal endothelial cell glutathione prevents hepatic veno-occlusive disease in the rat. Hepatology 2000; 31:428–434. 61. Jones RJ, Lee KSK, Beschorner WE, et al. Veno-occlusive disease of the liver following bone marrow transplantation. Transplantation 1987; 44:778–783. 62. Hasegawa S, Horibe K, Kawabe T, et al. Veno-occlusive disease of the liver after allogeneic bone marrow transplantation in children with hematologic malignancies: incidence, onset time and risk factors. Bone Marrow Transplant 1998; 22:1191–1197. 63. Toh HC, McAfee SL, Sackstein R, et al. Late onset venoocclusive disease following high-dose chemotherapy and stem cell transplantation. Bone Marrow Transplant 1999; 24:891–895. 64. Gooley TA, Rajvanshi P, Schoch HG, McDonald GB. Serum bilirubin levels and mortality after myeloablative allogeneic hematopoietic stem cell transplantation. Hepatology 2005; 41:345–352. 65. Shulman HM, McDonald GB, Matthews D, et al. An analysis of hepatic venocclusive disease and centrilobular hepatic degeneration following bone marrow transplantation. Gastroenterology 1980; 79:1178–1191. 66. Strasser SI, McDonald SJ, Schoch HG, McDonald GB. Severe hepatocellular injury after hematopoietic cell transplant: incidence and etiology in 2136 consecutive patients. Hepatology 2000; 299A (abstract). 67. Penta J, Von Hoff DD, Muggia F. Hepatotoxicity of combination chemotherapy for acute myelocytic leukemia. Ann Intern Med 1977; 87:247–248. 68. Demirer T, Buckner CD, Appelbaum FR, et al. Busulfan, cyclophosphamide and fractionated total body irradiation for autologous or syngeneic marrow transplantation for acute and chronic myelogenous leukemia: phase I dose escalation of busulfan based on targeted plasma levels. Bone Marrow Transplant 1996; 17:491–495. 69. Grochow LB. Busulfan disposition: the role of therapeutic monitoring in bone marrow transplantation induction regimens. Semin Oncol 1993; 20:18–25. 70. Qiu R, Yao A, Vicini P, et al. Diminishing the risk of nonrelapse mortality in hematopoietic stem cell transplantation: prediction of exposure to the cyclophosphamide metabolite carboxyethylphosphoramide mustard. Clin Pharmacol Ther 2004; 76:270–280. 71. Hommeyer SC, Teefey SA, Jacobson AF, et al. Venocclusive disease of the liver: prospective study of US evaluation. Radiology 1992; 184:683–686. 72. Teefey SA, Brink JA, Borson RA, Middleton WD. Diagnosis of veno-occlusive disease of the liver after bone marrow transplantation: value of duplex sonography. AJR 1995; 164:1397–1401. 73. McCarville MB, Hoffer FA, Howard SC, et al. Hepatic venoocclusive disease in children undergoing bone-marrow transplantation: usefulness of sonographic findings. Pediatr Radiol 2001; 31:102–105. 74. Shulman HM, Gooley T, Dudley MD, et al. Utility of transvenous liver biopsies and wedged hepatic venous pressure measurements in sixty marrow transplant recipients. Transplantation 1995; 59:1015–1022. 75. Bearman SI, Anderson GL, Mori M, et al. Veno-occlusive disease of the liver: development of a model for predicting fatal outcome after marrow transplantation. J Clin Oncol 1993; 11:1729–1736. 76. McGinn CJ, Ten Haken RK, Ensminger WD, et al. Treatment of intrahepatic cancers with radiation doses based on a normal tissue complication probability model. J Clin Oncol 1998; 16:2246–2252. 77. Ingold JA, Reed GB Jr, Kaplan HS, Bagshaw MA. Radiation hepatitis. AJR 1965; 93:200–208.
903
Section VII. Vascular Disease of the Liver
78. Fajardo LF, Colby TV. Pathogenesis of veno-occlusive liver disease after radiation. Arch Pathol Lab Med 1980; 104:584–588. 79. Locasciulli A, Testa M, Valsecchi MG, et al. The role of hepatitis C and B virus infections as risk factors for severe liver complications following allogeneic BMT: a prospective study by the Infectious Disease Working Party of the European Blood and Marrow Transplantation Group. Transplantation 1999; 68:1486–1491. 80. Nagler A, Or R, Naparstek E, et al. Second allogeneic stem cell transplantation using nonmyeloablative conditioning for patients who relapsed or developed secondary malignancies following autologous transplantation. Exp Hematol 2000; 28:1096– 1104. 81. Feinstein L, Sandmaier B, Maloney D, et al. Nonmyeloablative hematopoietic cell transplantation. Replacing high-dose cytotoxic therapy by the graft-versus-tumor effect. Ann NY Acad Sci 2001; 938:328–337. 82. Bearman SI, Hinds MS, Wolford JL, et al. A pilot study of continuous infusion heparin for the prevention of hepatic venoocclusive disease after bone marrow transplantation. Bone Marrow Transplant 1990; 5:407–411. 83. Attal M, Huguet F, Rubie H, et al. Prevention of hepatic venoocclusive disease after bone marrow transplantation by continuous infusion of low-dose heparin: a prospective, randomized trial. Blood 1992; 79:2834–2840. 84. Or R, Nagler A, Shpilberg O, et al. Low molecular weight heparin for the prevention of veno-occlusive disease of the liver in bone marrow transplantation patients. Transplantation 1996; 61:1067–1071. 85. Forrest DL, Thompson K, Dorcas VG, et al. Low molecular weight heparin for the prevention of hepatic veno-occlusive disease (VOD) after hematopoietic stem cell transplantation: a prospective phase II study. Bone Marrow Transplant 2003; 31:1143–1149. 86. Rosenthal J, Sender L, Secola R, et al. Phase II trial of heparin prophylaxis for veno-occlusive disease of the liver in children undergoing bone marrow transplantation. Bone Marrow Transplant 1996; 18:185–191. 87. Marsa-Vila L, Gorin NC, Laporte JP, et al. Prophylactic heparin does not prevent liver veno-occlusive disease following autologous bone marrow transplantation. Eur J Haematol 1991; 47:346–354. 88. Hagglund H, Remberger M, Klaesson S, et al. Norethisterone treatment, a major risk-factor for veno-occlusive disease in the liver after allogeneic bone marrow transplantation. Blood 1998; 92:4568–4572. 89. Carreras E, Bertz H, Arcese W, et al. Incidence and outcome of hepatic veno-occlusive disease after blood or marrow transplantation: a prospective cohort study of the European Group for Blood and Marrow Transplantation. European Group for Blood and Marrow Transplantation Chronic Leukemia Working Party. Blood 1998; 92:3599–3604. 90. Chalandon Y, Roosnek E, Mermillod B, et al. Prevention of veno-occlusive disease with defibrotide after allogeneic stem cell transplantation. Biol Blood Marrow Transplant 2004; 10:347–354. 91. Versluys B, Bhattacharaya R, Steward C, et al. Prophylaxis with defibrotide prevents veno-occlusive disease in stem cell transplantation after gemtuzumab ozogamicin exposure. Blood 2004; 103:1968. 92. Clift RA, Bianco JA, Appelbaum FR, et al. A randomized controlled trial of pentoxifylline for the prevention of regimenrelated toxicities in patients undergoing allogenic marrow transplantation. Blood 1993; 82:2025–2030. 93. Attal M, Huguet F, Rubie H, et al. Prevention of regimen-related toxicities after bone marrow transplantation by pentoxifylline – a prospective, randomized trial. Blood 1993; 82:732–736.
904
94. Ruutu T, Eriksson B, Remes K, et al. Ursodeoxycholic acid for the prevention of hepatic complications in allogenic stem cell transplantation. Blood 2002; 100:1977–1983. 95. Gluckman E, Joviet I, Scrobohaci ML, et al. Use of prostaglandin E1 for prevention of liver veno-occlusive disease in leukaemic patients treated by allogenic bone marrow transplantation. Br J Haematol 1990; 74:277–281. 96. Bearman SI, Shen DD, Hinds MS, et al. A phase-I phase-II study of prostaglandin E1 for the prevention of hepatic venoocclusive disease after bone marrow transplantation. Br J Haematol 1993; 84:724–730. 97. Richardson PG, Murakami C, Jin Z, et al. Multi-institutional use of defibrotide in 88 patients after stem cell transplantation with severe veno-occlusive disease and multisystem organ failure: response without significant toxicity in a high-risk population and factors predictive of outcome. Blood 2002; 100:4337–4343. 98. Corbacioglu S, Greil J, Peters C, et al. Defibrotide in the treatment of children with veno-occlusive disease (VOD): a retrospective multicentre study demonstrates therapeutic efficacy upon early intervention. Bone Marrow Transplant 2004; 33:189–195. 99. de la Rubia J, Carral A, Montes H, et al. Successful treatment of hepatic veno-occlusive disease in a peripheral blood progenitor cell transplant patient with a transjugular intrahepatic portosystemic stent-shunt (TIPS). Haematologica 1996; 81:536–539. 100. Smith FO, Johnson MS, Scherer LR, et al. Transjugular intrahepatic portosystemic shunting (TIPS) for treatment of severe hepatic veno-occlusive disease. Bone Marrow Transplant 1996; 18:643–646. 101. Azoulay D, Castaing D, Lemoine A, et al. Transjugular intrahepatic portosystemic shunt (TIPS) for severe veno-occlusive disease of the liver following bone marrow transplantation. Bone Marrow Transplant 2000; 25:987–992. 102. Zenz T, Rössle M, Bertz H, et al. Severe veno-occlusive disease after allogeneic bone marrow or peripheral stem cell transplantation – role of transjugular intrahepatic portosystemic shunt (TIPS). Liver 2001; 21:31–36. 103. Fried MW, Connaghan DG, Sharma S, et al. Transjugular intrahepatic portosystemic shunt for the management of severe veno-occlusive disease following bone marrow transplantation. Hepatology 1996; 24:588–591. 104. Annaloro C, Robbiolo L, Pozzoli E, et al. Four-year survival after trans-jugular intrahepatic porto-systemic shunt for venoocclusive disease following autologous bone marrow transplantation. Leukem Lymphoma 2004; 45:1485–1487. 105. Shen-Gunther J, Walker JL, Johnson GA, Mannel RS. Hepatic veno-occlusive disease as a complication of whole abdominopelvic irradiation and treatment with the transjugular intrahepatic portosystemic shunt: case report and literature review. Gynecol Oncol 1996; 61:282–286. 106. Rajvanshi P, McDonald GB. Expanding the use of transjugular intrahepatic portosystemic shunts for veno-occlusive disease. Liver Transplant 2001; 7:154–159. 107. Meacher R, Venkatesh B, Lipman J. Acute respiratory distress syndrome precipitated by transjugular intrahepatic portosystemic shunting for severe hepatic veno-occlusive disease. Is it due to pulmonary leucostasis? Intensive Care Med 1999; 25:1332–1333. 108. Jacobson BK, Kalayoglu M. Effective early treatment of hepatic veno-occlusive disease with a central splenorenal shunt in an infant. J Pediatr Surg 1992; 27:531–533. 109. Murray JA, LaBrecque DR, Gingrich RD, et al. Successful treatment of hepatic venocclusive disease in a bone marrow transplant patient with side-to-side portacaval shunt. Gastroenterology 1987; 92:1073–1077.
Section VII: Vascular Disease of the Liver
47
PORTAL AND SPLENIC VEIN THROMBOSIS Hector Rodriguez-Luna and Hugo E. Vargas Abbreviations AT III Antithrombin III DUS Doppler ultrasound FVL Factor V Leiden HVPG Hepatic venous-portal gradient IPVT Isolated portal vein thrombosis
LT MRA MTHFR C677?T PTHRA20210
Liver transplantation Magnetic resonance angiogram Methylenetetrahydrofolate reductase Factor II prothrombin
PORTAL VEIN THROMBOSIS Portal vein thrombosis (PVT) was first reported by GW Balford and TG Stewart in 1869 in a patient who presented with ascites, splenomegaly, and varices.1 PVT is a rare condition affecting both children and adults with equal gender distribution, and is typically associated with myriad precipitating factors and subtle acute clinical manifestations.2 PVT represents the classic form of presinusoidal (infrahepatic) portal hypertension. In western countries this entity is the leading cause of extrahepatic portal hypertension in noncirrhotic patients.3 The incidence of PVT is not clearly defined and varies depending on the group of patients studied and the diagnostic methods used. In the United States, the overall incidence ranges from 0.05% to 0.5% in autopsy studies. Reported prevalence in candidates for liver transplantation (LT) with cirrhosis is between 0.6% and 26%.7,10 In patients with cirrhosis, the incidence of PVT at the time of LT has been reported to range from 10% to 21%.4–6 In Japan, Okuda et al. reported an incidence of 0.6% by angiography of 708 cirrhotic patients.7 After LT, the incidence of PVT varies from 1% to 2%.8,9
PATHOPHYSIOLOGY Under normal circumstances the portal vein (PV) contributes twothirds of the hepatic blood supply. However, PV occlusion with thrombosis often produces few acute clinical consequences or laboratory manifestations.2,12,13 Two mechanisms account for this ability of the liver to survive the loss of portal perfusion. The first consists of an arterial ‘buffer’ response manifested by immediate vasodilatation of the hepatic arterial bed in response to decreased portal vein flow.12,14,15 The second is the relatively rapid development of collateral veins that bypass the thrombosed portion of the PV (cavernous transformation).16 The latter process may take up to 12 months, although it has been reported as early as 5 weeks after the thrombotic event.16,17 Other collateral veins may also develop within the walls or at the periphery of the structures adjacent to the obstructed portion of the portal vein, such as the bile ducts, gallbladder, pancreas, gastric antrum, and duodenum.18 These collateral veins may
PV PVT SMV
Portal vein Portal vein thrombosis Superior mesenteric vein
alter the appearance of these surrounding structures during imaging, and occasionally lead to erroneous diagnoses of bile duct or pancreatic tumor, pancreatitis, or cholecystitis. In some instances these cavernous vessels can have clinical consequences: bile duct varices have been reported to cause obstructive jaundice.18 As a result of these hemodynamic compensations, the total hepatic blood flow is only minimally reduced, hepatic venous pressure gradient (HVPG) is initially preserved at normal levels, and the portal pressure is elevated.15 The increase in portal pressure allows portal perfusion to be maintained through the collateral veins. This initial state is not static, and portal pressure will increase further over time. The risk for bleeding esophageal varices develops when the HVPG rises to a threshold value of 10–12 mmHg.19 PVT patients can also experience the development of a compensatory hyperdynamic circulatory state akin to that seen in cirrhosis.15 The overall consequences of PVT are related to thrombus extension. Below the thrombus, there is little effect on the intestines so long as the mesenteric venous arches remain patent. Mesenteric ischemia results from extension of the thrombus into the mesenteric venous arches.12 When the ischemia is prolonged, intestinal infarction will ensue. In 20–50% of cases intestinal infarction is responsible for death due to peritonitis and multiple organ failure, even when the infarcted gut is resected.20,21 Above the thrombus, the consequences of PVT to the liver are hardly discernible and there are minimal laboratory abnormalities.22 Clinically, acute signs of liver disease are absent or transient unless the PVT occurs in a patient with cirrhosis.12 Concomitant PVT may be seen in 20% of patients with Budd–Chiari syndrome, and this may worsen their liver disease.23 PVT can be classified anatomically into four grades according to where the thrombus extends.24 These grades are reflective of the clinical consequences of the thrombus and have an impact on the selection of medical and surgical management options. 1. Grade 1: Minimally or partially thrombosed PV, in which the thrombus is minimal or, at most, confined to <50% of the vessel lumen, with or without limited extension into the superior mesenteric vein (SMV).
905
Section VII. Vascular Disease of the Liver
2. Grade 2: More than 50% occlusion of the PV, including total occlusions, with or without limited extension into the SMV. 3. Grade 3: Complete thrombosis of the PV and proximal SMV with open distal SMV. 4. Grade 4: Complete thrombosis of the PV and SMV.
ETIOLOGY The etiology of PVT is highly diverse and, as in other thrombotic disorders, its development depends on the interaction of many factors.25,26 Surprisingly, our understanding of the risk factors for PVT follow very closely those originally postulated by Virchow in the 19th century for the development of venous thrombosis.25 The so-called Vichow’s triad includes flow abnormalities resulting in blood stasis; imbalance between pro and anticoagulant proteins, with resultant activation of clotting proteins; and defects in the blood vessel wall, resulting in a shift to a procoagulant endothelium.25 Of the many disorders that lead to the development of PVT, most can cause disruption of more than one of the elements of this triad. Current opinion favors the postulate that the development of PVT is triggered by the interaction of many factors and is not due to an isolated event.27 However, despite the availability of an everwidening array of diagnostic techniques, 8–15% of PVT cases are classified as idiopathic, as no underlying predisposing condition can be identified at the time of diagnosis.28–30
Cirrhosis Predisposing factors for PVT can be classed as inherited or acquired. Acquired factors are further subdivided into local or systemic conditions (Table 47-1). Among the acquired causes, cirrhosis has been long considered a major cause in adults and is present in 24–32% of patients with PVT.2 The pathogenesis of PVT in cirrhosis is uncertain, and it is still difficult to link cirrhosis per se to the development of PVT. It has been suggested that several concurrent factors in cirrhotic patients, including decreased portal blood flow, the presence of periportal lymphangitis with fibrosis, and a possible thrombophilia, promote the formation of thrombi. Several studies have reported the association of a prothrombotic state and thrombophilia in the setting of cirrhosis.31,32 In two studies, 62–69.5% of cirrhotic patients had a deficiency of one or more natural anticoagulant proteins.32,33 Whether the prothrombotic state is primary or secondary is controversial.
Neoplasia Neoplastic disorders are the second most common cause of PVT in adults and are found in 21–24% of patients with PVT.2 Pancreatic cancer tops the list and is responsible for 11–12% of adult cases, followed by hepatocellular carcinoma, which accounts for 5–6% of cases.2 Other cancers implicated include pulmonary, gastric, prostate, uterine, renal, biliary, malignant carcinoid, and hepatic lymphoma.2,22 Neoplasia can lead to PVT through a combination of systemic and sometimes local factors. A hypercoagulable state of malignancy is believed to be related to increased activity of the coagulation system, as evidenced by markers of accelerated thrombin generation and increased platelet reactivity.25 In addition, tissue factor and cancer procoagulant (cysteine protease) have been implicated in specific types of solid and hematologic tumor. Tissue factor
906
TABLE 47-1. Causes of Portal Vein Thrombosis in Adults Common causes Cirrhosis Neoplasm Pancreatic CA Hepatocellular carcinoma Intra-abdominal malignancy Infection Appendicitis Diverticulitis Cholecystitis Inflammation Pancreatitis Myeloproliferative disorders Polycythemia vera Thrombocytosis PNH Agnogenic myeloid metaplasia Idiopathic Uncommon causes Inherited hypercoagulable states –High risk for thrombosis (low prevalence < 0.04%) Antithrombin III deficiency Protein S deficiency Protein C deficiency –Low risk for thrombosis (high prevalence > 2%) FVL mutation PTHR20210 mutation (Factor II prothrombin) MTHFR C677ÆT mutation Antiphospholipid syndrome Acquired hypercoagulable states Inflammatory bowel disease Pregnancy Oral estrogens Miscellaneous Non-cirrhotic portal hypertension Abdominal surgery, shunt surgery Splenectomy Liver transplant
acts in conjunction with factors VII/VIIa to activate factor X. The enzymatic function of cancer procoagulant is the activation of factor X. Another postulated procoagulant mechanism in cancer patients is impaired fibrinolysis, with a subsequent increase in plasminogen activator inhibitor.25,34
Myeloproliferative Disorders The hypercoagulable state due to myeloproliferative disorders accounts for 3–12% of adult patients with PVT.2,30,35 Some patients with idiopathic PVT have a latent myeloproliferative disorder that becomes evident only years after the diagnosis of PVT. Valla and coworkers found that 48% of adult patients with non-malignant PVT originally classified as idiopathic had either overt or latent myeloproliferative disorders.30
Infection In the adult population, infection accounts for 10–25% of PVT cases in non-cirrhotic, non-cancer patients.2 Septic PVT (pyelophlebitis) is usually related to appendicitis, cholecystitis, or diverticulitis.1,29,31
Chapter 47 PORTAL AND SPLENIC VEIN THROMBOSIS
However, PVT as a result of infection is infrequent in the adult population, with a decreasing incidence because of earlier diagnosis and earlier initiation of effective antibiotic therapy.36 Interestingly, Bacteroides species bacteremia of unknown origin is so strongly associated with PVT that culture of this organism from the blood should prompt a search for portal or mesenteric vein thrombosis.12,37 In children, infection is the most common etiologic factor for PVT, accounting for 43–52% of all cases.31 Neonatal umbilical sepsis, the single most frequent infectious cause, is present in 10–26% of children with PVT.1,2,12,31 Neonatal thrombosis is well documented after omphalitis or umbilical vein cannulation complicated by septic pyelophlebitis. However, infants with infection of the umbilical vein in the absence of prothrombotic disorder infrequently go on to develop PVT.38 The first clinical manifestations of neonatal PVT are frequently delayed until adulthood.
intrahepatic portosystemic shunt (TIPS) carries a risk for PVT of approximately 10%.55 PVT following endoscopic variceal ligation or sclerotherapy was a matter of controversy in the past, as most published studies did not document the patency of portal vessels prior to the initiation of therapy.56,57 In an effort to further clarify this issue, Politoske et al. studied the incidence of PVT in patients treated with sclerotherapy and band ligation for variceal bleeding in cases without pre-existing PVT; the study found no significant difference between the groups after therapy.58 Finally, several authors have suggested an association between PVT and congenital cardiovascular abnormalities, such as atrial septal defect, ventricular septal defect, and deformed inferior vena cava.59
CLINICAL MANIFESTATIONS Thrombophilias Inherited or acquired prothrombotic states may predispose to the development of PVT. The presence of more than one deficiency seems to be the rule rather than the exception.26,27,33 Inherited prothrombotic disorders are subclassified into two groups according to the prevalence in the population. The first group includes deficiencies in protein C, protein S, and antithrombin III (AT III). The prevalence of these deficiencies is very low in the general Caucasian population (<0.04%), with a high associated risk of thrombosis in heterozygotes (~10%). The second group includes gene mutations in factor V Leiden (FVL)41, factor II prothrombin (PTHR A20210),42,43 and methylenetetrahydrofolate reductase (MTHFR C677T).44 Both FVL gene mutation and the PTHR A20210 mutation are associated with a lower relative risk of thrombosis despite being more prevalent (>2%) in the general population. FVL gene mutation and deficiencies of anticoagulant proteins have been associated with PVT.33,45–47 Janssen et al. reported that the relative risk of PVT for individuals with FVL mutation was 2.7, 1.4 for those with PTHR A20210, and 4.6 for those with protein C deficiency. Protein C and S deficiency has been reported in up to 30% of cirrhotic patients with PVT.45 Antithrombin III deficiency has been less frequently associated with PVT.27,45,47 Antiphospholipid syndrome has been reported in up to 11% of patients with PVT.27
Other Inflammatory disorders such as pancreatitis and inflammatory bowel disease (IBD) have also been implicated in PVT.30,39 Pancreatitis accounts for 3–5% of cases of PVT, via either a contiguous inflammatory process, direct compression of the PV by a pseudocyst, or a combination of both. Chronic pancreatitis can also lead to splenic vein thrombosis and a unique form of ‘left-sided’ segmental portal hypertension with the development of isolated gastric varices.30,40 Other associated factors include pregnancy and oral intake of estrogens.48,49 PVT can also be seen in the setting of blunt abdominal trauma, surgery in the absence of septic complications,2 or non-surgical treatment for hepatocellular carcinoma, such as radiofrequency ablation or microwave coagulation therapy.50 Splenectomy carries a PVT risk ranging from 0.7 to 8%.51–54 In patients with underlying myeloproliferative disorder or cirrhosis, splenectomy carries a particularly increased risk of PVT, ranging from 13–18%.51,54 Transjugular
Patients with PVT typically present with abdominal pain, increased abdominal girth, or, more dramatically, hematemesis.2,12,13 Neonatal PVT can present many years later with complications of the ensuing portal hypertension, such as ruptured gastroesophageal varices or splenomegaly.1,2,59 Most patients with PVT diagnosed before the index gastrointestinal bleed eventually go on to bleed within a mean of 4 years from diagnosis.1 However, 10% of patients with PVT never bleed.1 The presence of acute-onset abdominal pain can be an ominous sign and should prompt aggressive work-up for bowel ischemia.2 Bowel ischemia patients may also present with gastrointestinal bleeding. Other common complaints of patients with PVT include nausea, vomiting, diarrhea, anorexia, weight loss, and abdominal distention.28,31,59 Some patients may also experience lowgrade fever.51 Portosystemic encephalopathy is rare unless the patient has underlying liver disease. A rare presentation of PVT is the occurrence of bile duct compression due to collateral veins which develop in the hepatoduodenal ligament.18,60 On physical examination, splenomegaly is seen in 75–100% of cases. Hepatomegaly may also be present. Ascites is an uncommon finding in PVT, and when present is usually mild and transient. Ascites typically develops immediately after the thrombotic event, before the patient has had time to develop a collateral circulation.22,28,59 A rare presentation of PVT is bile duct compression due to collateral veins which develop in the hepatoduodenal ligament.18,60 Some patients may also experience low-grade fever.51 Hepatic enzymes and liver injury tests are usually within normal limits in patients without underlying liver disease.2,12 However, mild elevations in transaminases, alkaline phosphatase, and bilirubin have been reported.12 A mild decrease in red cell count, white cell count, and platelets due to hypersplenism may also be seen.2,13,28,59 Histologically, there is little alteration in the hepatic architecture when the obstruction is limited to the extrahepatic portal vein and its largest intrahepatic branches. Non-cirrhotic patients typically show normal histology, with increased reticulin around the portal tracts. Experimentally, apoptosis of the liver cells can be demonstrated in rats with graded portal vein ligation.61 The degree of apoptosis is related to the grade of portal vein obstruction. There is a simultaneous increase in mitotic activity in the remaining well-perfused liver.61,62 Wanless and colleagues have postulated a mechanism by which the vascular changes seen with chronic PVT may lead to the development of venoportal bridging fibrosis and eventual cirrhosis.63
907
Section VII. Vascular Disease of the Liver
NATURAL HISTORY The natural history of PVT remains shrouded in mystery, as the patient population is heterogeneous and the management is predicated by the time of diagnosis and the underlying etiology. The clinical course is characterized by repeated bouts of variceal hemorrhage, with an average of 2.5–5 episodes per patient.64,65 In an analysis of non-cirrhotic non-neoplastic PVT, the incidence of gastrointestinal bleeding was found to be approximately 12.5 per 100 patient-years. In this study, the only independent predictive factor for bleeding was the size of the esophageal varices.66 In cases of neonatal PVT, the bleeding episodes tend to increase in severity and frequency at puberty, followed by abatement after the development of spontaneous splenorenal or splenogastric shunts in 10–20% of patients.60 The overall prognosis for patients with chronic PVT and recurrent gastrointestinal bleeding in the absence of cirrhosis or malignancy is good, with a mortality rate of approximately 10%.2,67 In a Dutch study, non-cirrhotic patients with PVT had an overall survival rate of 70% at 1 year and 63% after 5 years.68 In the case of acute PVT due to intra-abdominal sepsis, pre-existing liver disease, or abdominal surgery, the mortality rate approaches 50%.31 In children the prognosis is much better, with a 10-year survival rate greater than 70%. Extensive PVT with mesenteric venous thrombosis with bowel infarction is invariably fatal without prompt surgical intervention. Mortality in this scenario can approach 20% even with expedient bowel resection.69
High index of suspicion for portal vein thrombosis
Portal vein Doppler ultrasound
PVT confirmed
PVT ruled out
Evaluate patient for risk factors Refer to Table 1 Screen for portal hypertension Establish age of thrombus
Non-diagnostic ultrasound
MRI or CT scan of the hepatic vessels to confirm or rule out PVT
Determine therapy: thrombolytics vs. chronic anticoagulation vs. no treatment Treat portal hypertension according to manifestations Figure 47-1. Suggested diagnostic work-up for patients with suspected PVT. Refer to text for further details on diagnostic work-up.
DIAGNOSIS The key to the diagnosis of PVT is a high index of suspicion (Figure 47-1). For confirmation, a variety of radiologic techniques can be used to investigate the suspected thrombosis. Invasive angiographic techniques, such as ‘indirect’ portography (venous phase of superior mesenteric artery angiogram) and ‘direct’ portal venography (transhepatic or transjugular) are the time-honored diagnostic techniques for PVT.70 However, a variety of non-invasive techniques, such as color Doppler ultrasound (DUS), computed tomographic angiography (CT angiography), and magnetic resonance angiography (MRA), have become available for the screening of patients suspected of having PVT.70–75 Nowadays, ultrasonography is the first-line diagnostic modality because of its accuracy, affordability, and non-invasiveness. An echogenic thrombus within the portal lumen is the key finding for the ultrasonographic diagnosis of PVT.76 Other signs include dilatation of the proximal vessel, the presence of collateral vessels (best seen near the porta hepatis), or an unidentifiable portal vein.74–76 The lack of variation in portal venous diameter with respiration, coupled with a portal vein diameter greater than 13–15 mm, is also highly indicative of portal vein occlusion. These hallmarks may be less reliable when the thrombus is long-standing. The sensitivity of ultrasonography ranges from 70 to 90%, with a specificity of 99%.24,75 The presence of arterial flow signal in the thrombus typically correlates with a malignant thrombus.75 Major limitations to ultrasonography include obesity, fatty liver, and non-visualization secondary to bowel gas. In addition, ultrasonography is operator dependent, and PVT might be missed if the examiner is not specifically asked to look for it.74–76 CT scans can be used to confirm and follow the course of PVT. On CT scan, the thrombus within the portal vein shows decreased
908
intraluminal density (filling defect in the contrast-enhanced lumen) or as total portal occlusion with or without the development of periportal collaterals creating a ‘train-track’ appearance on enhancement. This ‘train-track’ appearance is associated with proliferation of the vasa vasorum and is associated with old thrombus. A nonenhanced CT scan of the liver will show a high luminal density within the portal vein when the thrombus is less than 10 days old.77 The presence of periportal collaterals suggests that the occlusion is chronic and the thrombus is organized, also known as ‘cavernous transformation’ of the portal vein77 (Figure 47-2). This process may take up to 12 months to occur, although it has been demonstrated as early as 5 weeks after the thrombotic event.12 Contrast-enhanced CT has the advantage over ultrasound of displaying varices (sensitivity 65–85%) and parenchymal hepatic abnormalities.71 The falsepositive rate in one small series was 16%, possibly owing to poor bolus injection. CT is not operator dependent; however, the radiation dose, cost, and need for intravenous contrast make it a less than ideal test.71,78,79 Contrast-enhanced MR angiography, spin-echo MR, and gradientecho MR have been introduced for the diagnosis of PVT. Spin-echo MR images usually shows PVT as an area of abnormal signal within the lumen of the portal vein. PVT appears hyperintense on T1 and T2 images when the thrombus has been formed recently. Old thrombus appears isointense on T1 images. Gradient-echo MR gives a sharper delineation of vascular structures and helps clarify any confusion on spin-echo images. MR angiography shows flow patterns and patency of the portal vein. Tumor thrombi can be differentiated from bland thrombi because they appear more hyperintense on T2weighted images and enhance with gadolinium. The sensitivity of MRI is 85% and the specificity is 90–95%.72,80
Chapter 47 PORTAL AND SPLENIC VEIN THROMBOSIS Figure 47-2. High-resolution CT scan coronal reconstruction of portal anatomy in a patient with ductal adenocarcinoma of the body of the pancreas. Note the portal vein occlusion with cavernous collateral transformation (narrow arrow). In addition, the patient has liver metastases with bile duct obstruction and malignant ascites. (Photo courtesy of Dr Joseph Collins.)
More invasive radiographic techniques are currently reserved for cases when non-invasive testing is inconclusive, immediately before anticipated percutaneous interventional treatment, or when a meticulous preoperative assessment is necessary.70
ISOLATED SPLENIC VEIN THROMBOSIS Isolated splenic vein thrombosis (ISVT) deserves special attention as the etiology and clinical manifestations differ from those of PVT. ISVT usually results in left-sided portal hypertension and isolated gastric fundal varices.81 The most common cause of ISVT is chronic pancreatitis, with a reported incidence up to 45%.82 Pancreatitis-associated ISVT is believed to result from perivenous inflammation.82,83 The prevalence of splenic vein complications in relation to the CT scan severity index of pancreatitis has shown an inversely proportional significant increase in the prevalence of thrombosis.83 Other known factors include pancreatic masses and cancer, splenectomy,84 portal hypertension, renal disorders, and inflammatory disorders.81 The diagnostic test of choice to assess the presence of ISVT is late-phase celiac angiography.82 However, endoscopic ultrasonography (EUS) has emerged as a fairly sensitive and non-invasive diagnostic tool.85,86 Splenoportography was previously used to make this diagnosis (Figure 47-3).
The natural history of ISVT is uncertain and literature reports are few. Most cases of ISVT are asymptomatic and require no treatment.82 In an effort to further define the natural history of pancreatitis-induced ISVT, Heider and co-workers87 studied 53 patients with a history of pancreatitis and ISVT and found that 77% of isolated gastric varices were evident on CT scanning, 31% by esophagogastroduodenoscopy (EGD), and 28% by the combined modalities. The risk of variceal bleeding was 5% for patients with CT-identified varices and 18% for EGD-identified varices.87 Of those patients, only 4% had a gastric variceal bleeding episode and required splenectomy.87 Other investigators have reported a gastric variceal bleeding risk of approximately 10%.88 The treatment of ISVT is conservative, given the low risk of associated gastric variceal bleeding.87 Once there is an index episode of gastrointestinal bleeding, splenectomy is the treatment of choice.82 The surgical and medical teams need to ensure that hepatic fibrosis has been carefully evaluated. Splenectomy in the setting of cirrhosis should be avoided if possible, as the development of PVT post splenectomy may complicate or eliminate the opportunity for liver transplantation in the future.
LIVER TRANSPLANTATION AND PVT In the past, PVT has been considered a relative contraindication to liver transplantation because of the technical difficulties it added to
909
Section VII. Vascular Disease of the Liver
Figure 47-3. Splenoportogram of a patient with splenic vein thrombosis. The needle is within the spleen (closed arrow). Contrast material can be seen to flow from the spleen to the portal vein (PV) via large collaterals (C). The pool of contrast material (open arrow) is within numerous gastric varices.
the procedure. However, in recent years innovative surgical techniques have been introduced and many technical obstacles have been overcome.89–94 As a result, patients with non-neoplastic PVT routinely undergo LT. The first successful LT in a patient with PVT using venous conduits to bypass the thrombotic segment was reported in 1985.93 Since then, many centers have reported several techniques to tackle the problem of PVT in liver transplant candidates. The surgical technique to re-establish portal blood flow in liver transplant recipients with PVT depends on the extent of the thrombosis and the experience of the transplant team. Partial PVT with less than 25% luminal obstruction has no clinical repercussions because it can be treated with resection, whereas more than 25% luminal obstruction requires extensive thrombectomy.95 Thromboendovenectomy and/or direct venous anastomosis has been performed in patients when the thrombosis involves the portal vein with or without extension into the SMV. If the portal vein is completely thrombosed and the proximal SMV is occluded but the distal part is patent, then the preferred method is a graft to the proximal SMV using donor iliac vein.6,24,94 In cases where the portal vein is not amenable to thrombectomy and the SMV is thrombosed, the coronary vein or any large accessible collateral vein can be used to join the donor portal vein to the SMV.94 For patients with extensive and complete occlusion of the portal vein and SMV, cavoportal hemitransposition (a procedure that diverts caval blood to the liver as replacement of portal inflow and without compromise of hepatic function) has been described as an innovative and successful procedure.90 De novo PVT after liver transplantation is rare and usually occurs at the anastomotic site in the early postoperative period. In patients with pretransplant PVT the early outcome seems to be satisfactory, although there remains a risk of the portal vein rethrombosing which ranges from 4.2 to 38.5%.24,89 The greater risk for rethrombosis has been reported in grafts that were not preserved with University of Wisconsin solution, or when venovenous bypass was used.95 In an effort to prevent recurrence of PVT in this group of patients, therapeutic or prophylactic anticoagulation for 3 months has been advocated.6,95,96
910
Liver transplant candidates with PVT, especially those who have more than 50% of the portal vein occluded with or without superior mesenteric vein (SMV) occlusion, are more prone to develop severe perioperative complications, and have a high mortality rate and decreased long-term survival.89 The mortality rate is influenced by the extent of thrombosis before liver transplantation. Mortality has been reported to be greater in patients with PVT and splanchnic vein involvement than in those with only PVT (45.5% vs 36.1%) or associated periphlebitis (83.3% vs 9.4%). Nevertheless, the vast majority of patients with PVT can be technically transplanted, with a survival comparable to that of patients without PVT.89 Acute graft failure due to early occurrence of PVT after liver transplantation, bleeding from esophageal varices, and massive ascites can be serious complications from PVT after liver transplantation.89
MANAGEMENT A thorough etiological investigation and assessment of thrombus chronicity is paramount in the management of PVT in order to identify those conditions amenable to treatment and to tailor treatment. Investigation of the local factors is carried out with Doppler ultrasonography, MRI scan, abdominal CT, or endoscopic ultrasound. When portal cavernous transformation is recognized, portal hypertension can be assumed. Investigation of general thrombophilic factors must be comprehensive because in most patients there are usually several factors that contribute to the hypercoagulable state. Factors such as myeloproliferative disorders should be systematically investigated as they can have subtle presentations before becoming fully apparent on hematological grounds. One feature of a myeloproliferative disorder that may be present before the disease is clinically manifest is the spontaneous formation of erythroid colonies on culture of the circulating or bone marrow precursors in the absence of erythropoietin added to the culture medium. A similar test has also been developed for spontaneous colonies of megakaryocytes.30 Where these tests are not easily available, diagnostic information can also be obtained using isotopic determination of the total red cell volume
Chapter 47 PORTAL AND SPLENIC VEIN THROMBOSIS
coupled with determination of serum erythropoietin levels, provided that iron deficiency has been corrected. Bone marrow biopsy is another means to demonstrate primary myeloproliferative disorder when the peripheral blood picture is not suggestive, but this procedure is too invasive to serve as a screening procedure.12 Other considerations include coagulation factor gene mutations, coagulation inhibitor deficiencies, and the antiphospholipid syndrome. Interpretation of the results of antithrombin, protein C and protein S is particularly difficult in the context of PVT because their plasma levels may be non-specifically decreased whenever there is underlying liver disease or coagulation activation. Therefore, comparisons with the results of prothrombin determination and familial studies are necessary before the conclusion of a primary (inherited) deficiency can be reached.12 Factor V Leiden mutations can be assessed directly using molecular techniques, or indirectly by evaluation of the resistance to activated protein C. Identification of factor II G20210A mutation requires molecular techniques. The antiphospholipid syndrome is diagnosed when high titers of antiphospholipid antibodies are found on two separate occasions, or when a lupus anticoagulant is demonstrated. However, determination of anti-/32 glycoprotein-1 antibodies may be both more sensitive and more specific than the first two tests.98 Hyperhomocysteinemia is difficult to ascertain once PVT has developed because the plasma level is dependent on normal hepatic function. The C677T mutation of the methylene tetrahydrofolate reductase gene is associated with an increased plasma homocysteine; however, it is not clear whether this genetic marker alone is as good a marker for the increased risk of thrombosis as is the plasma homocysteine level.26,44 Gastrointestinal lesions that may be a source of bleeding need to be identified for adequate prophylactic measures to be taken. To date, there has been no controlled study specifically addressing bleeding in the setting of PVT. However, the available uncontrolled data indicate that the measures of established efficacy in patients with cirrhosis in good condition, such as propranolol and endoscopic therapy, can be applied to patients with PVT.26,44,99–101 Therapy for active gastrointestinal bleeding should, likewise, follow the guidelines for patients with intrahepatic portal hypertension regarding sclerotherapy and band ligation.101 There is, however, a matter of concern about the use of vasoconstrictive agents. Theoretically, the profound decrease in splanchnic blood flow induced by bleeding and by the therapeutic vasoconstrictive agents may trigger recurrence, or favor the extension of thrombosis in the portal venous system and precipitate intestinal ischemia. Indeed, peripheral vasopressin infusion has been reported to cause portal and mesenteric vein thrombosis, leading to intestinal ischemia in bleeding cirrhotic patients.102
SURGICAL MEASURES The place of surgery and the optimal type of operation is still being debated. A shunting procedure that would efficiently and permanently decompress the portal venous system with a low risk of encephalopathy would appear ideal. Some authors report a success rate in excess of 80% in shunt procedures, with a rebleeding rate as low as 4%.11,64 Unfortunately, the risk of shunt thrombosis or stenosis is predictably high, between 8 and 24%.11,103 Indeed, several precipitating factors are often present: underlying thrombophilia, surgery for portal hypertension, and splenectomy. Only the largest
veins (superior or inferior mesenteric veins or splenic veins) should be used because of the high risk of shunt thrombosis when using smaller veins.104 However, veins as small as 4 mm can be used.103 Because it leaves the spleen in place and preserves portal perfusion with a lower risk of encephalopathy, distal splenorenal shunt appears most suited for cirrhotic patients.11 Unfortunately, the splenic vein is frequently involved in the thrombotic process. TIPS has been used for the control of intractable bleeding as a bridge to liver transplantation or in patients with non-cavernous PVT as an adjunct to thrombolysis.70,105,106 The Sugiura procedure (transthoracoabdominal esophageal transection) has also been used to manage PVT, but it carries a surgical mortality as high as 20%.107 Splenectomy, only indicated for the management of gastrointestinal bleeding from gastric varices associated with splenic vein thrombosis, is contraindicated in patients with PVT because it may preclude the option of splenorenal shunt surgery at a later stage if needed.109
MEDICAL MEASURES The role of anticoagulant therapy for PVT is not well understood, although a large body of data continues to accumulate. Ideally, recent and old PVT must be differentiated before instituting therapy, as the recent PVT might benefit from thrombolytic therapy.70 In a retrospective study, Condat et al. showed a beneficial role of anticoagulation in recanalization and prevention of thrombus extension in patients with cavernous transformation of the portal vein. A major observation was that anticoagulant therapy reduced the risk of thrombotic events by two-thirds without an increase in the risk or severity of bleeding. Therefore, they suggest anticoagulation in those patients with a demonstrable prothrombotic state, absent or small varices that have never bled, and no predictable bleeding sites outside the gastrointestinal tract.66,110 To what extent spontaneous recanalization can be expected is not known. Current experience suggests that it is possible but uncommon, whereas complete or extensive recanalization can be achieved with anticoagulant therapy in more than 80% of patients.12,111 Recanalization prevents ischemic intestinal injury in the short term and extrahepatic portal hypertension in the long term. Malkowski et al. reported on the efficacy of thrombolytic agents in 28 patients and concluded that if it is administered early after the diagnosis of PVT, 82% of patients will have restitution of portal vein flow. Of those, 36% will have complete recanalization and 46% partial recanalization, with normal hepatopedal flow. The greatest benefit was seen in those patients with PVT of less than 4 weeks’ duration.112 Septic pyelophlebitis represents a special case in which recanalization can follow effective antibiotic therapy even in the absence of anticoagulant therapy. Drainage of associated hepatic, pancreatic, or splenic abscesses to achieve faster control of infection is recommended to allow recanalization by removing the inflammatory process.70 In the case of established PVT with portal hypertension due to cavernous transformation, anticoagulant therapy increased neither the risk of gastrointestinal bleeding nor the severity of bleeding. Valla et al. found that there were no deaths due to bleeding with anticoagulant therapy and no recurrent thrombosis.30 Therefore there is mounting evidence of a positive benefit–risk ratio with anticoagulant therapy. Some investigators only recommend anticoagula-
911
Section VII. Vascular Disease of the Liver
tion in specific clinical scenarios, such as a demonstrable prothrombotic state, concomitant mesenteric vein thrombosis, or portosystemic shunt (to prevent thrombosis).68 Successful thrombolytic therapies, with or without mechanical thrombectomy, have been reported by several investigators in case reports and small series. Thrombolytic agents can be infused via selective SMA or via the transhepatic route. The reported complication rate is low, ranging from none to rectal bleeding.70,113,114 There have been many reports of successful treatment of postliver transplantation PVT with percutaneous portal vein thrombolysis, angioplasty, and endovascular stent placement. Immediate retransplantation is required when serious deterioration of liver function occurs after early PVT.89 TIPS is not recommended because it reduces the effective portal flow and may deteriorate liver function further over the long term.97 In summary, PVT should be considered a clue to the presence of one or several prothrombotic disorders, whether or not a local precipitating factor is identified. Acute PVT can and probably should be treated with anticoagulation or thrombolytic agents in an effort to prevent extension of the thrombus, mesenteric vessel occlusion, and portal hypertension. On the other hand, chronic PVT should be treated conservatively with measures to control major consequences related to portal hypertension. The duration of anticoagulation therapy should be tailored to the identified predisposing factors.
REFERENCES 1. Webb LJ, Sherlock S. The aetiology, presentation and natural history of extrahepatic portal venous obstruction. Q J Med 1979;48:627–639. 2. Cohen J, Edelman RR, Chopra S. Portal vein thrombosis: a review. Am J Med 1992;92:173–182. 3. Belli L, Romani F, Riolo F, et al. Thrombosis of portal vein in absence of hepatic disease. Surg Gynecol Obstet 1989;169:46–49. 4. Nonami T, Yokoyama I, Iwatsuki S, Starzl TE. The incidence of portal vein thrombosis at liver transplantation. Hepatology 1992;16:1195–1198. 5. Lerut JP, Mazza D, van Leeuw V, et al. Adult liver transplantation and abnormalities of splanchnic veins: experience in 53 patients. Transpl Int 1997;10:125–132. 6. Stieber AC, Zetti G, Todo S, et al. The spectrum of portal vein thrombosis in liver transplantation. Ann Surg 1991;213:199–206. 7. Okuda K, Ohnishi K, Kimura K, et al. Incidence of portal vein thrombosis in liver cirrhosis. An angiographic study in 708 patients. Gastroenterology 1985;89:279–286. 8. Langnas AN, Marujo W, Stratta RJ, et al. Vascular complications after orthotopic liver transplantation. Am J Surg 1991;161:76–82;discussion 82–83. 9. Lerut J, Tzakis AG, Bron K, et al. Complications of venous reconstruction in human orthotopic liver transplantation. Ann Surg 1987;205:404–414. 10. Orozco H, Takahashi T, Mercado MA, et al. Postoperative portal vein obstruction in patients with idiopathic portal hypertension. J Clin Gastroenterol 1990;12:607. 11. Warren WD, Henderson JM, Millikan WJ, et al. Management of variceal bleeding in patients with noncirrhotic portal vein thrombosis. Ann Surg 1988;207:623–634. 12. Valla DC, Condat B. Portal vein thrombosis in adults: pathophysiology, pathogenesis and management. J Hepatol 2000;32:865–871.
912
13. Valla DC, Condat B, Lebrec D. Spectrum of portal vein thrombosis in the West. J Gastroenterol Hepatol 2002;17 (Suppl 3):S224–S227. 14. Henderson JM, Gilmore GT, Mackay GJ, et al. Hemodynamics during liver transplantation: the interactions between cardiac output and portal venous and hepatic arterial flows. Hepatology 1992;16:715–718. 15. Lebrec D, Bataille C, Bercoff E, Valla D. Hemodynamic changes in patients with portal venous obstruction. Hepatology 1983;3:550–553. 16. Ohnishi K, Okuda K, Ohtsuki T, et al. Formation of hilar collaterals or cavernous transformation after portal vein obstruction by hepatocellular carcinoma. Observations in ten patients. Gastroenterology 1984;87:1150–1153. 17. Zalcman M, Van Gansbeke D, Matos C, et al. Sonographic demonstration of portal venous system thromboses secondary to inflammatory diseases of the pancreas. Gastrointest Radiol 1987;12:114–116. 18. Perlemuter G, Bejanin H, Fritsch J, et al. Biliary obstruction caused by portal cavernoma: a study of 8 cases. J Hepatol 1996;25:58–63. 19. Garcia-Tsao G, Groszmann RJ, Fisher RL, et al. Portal pressure, presence of gastroesophageal varices and variceal bleeding. Hepatology 1985;5:419–424. 20. Clavien PA, Huber O, Rohner A. Venous mesenteric ischaemia: conservative or surgical treatment? Lancet 1989;2:48. 21. Clavien PA, Durig M, Harder F. Venous mesenteric infarction: a particular entity. Br J Surg 1988;75:252–255. 22. Macpherson AI. Portal hypertension due to extrahepatic portal venous obstruction. A review of 40 cases. J Roy Coll Surg Edin 1984;29:4–10. 23. Vogelzang RL, Anschuetz SL, Gore RM. Budd–Chiari syndrome: CT observations. Radiology 1987;163:329–333. 24. Yerdel MA, Gunson B, Mirza D, et al. Portal vein thrombosis in adults undergoing liver transplantation: risk factors, screening, management, and outcome. Transplantation 2000;69:1873–1881. 25. Falanga A, Rickles FR. Pathophysiology of the thrombophilic state in the cancer patient. Semin Thromb Hemost 1999;25:173–182. 26. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999;353:1167–1173. 27. Denninger MH, Chait Y, Casadevall N, et al. Cause of portal or hepatic venous thrombosis in adults: the role of multiple concurrent factors. Hepatology 2000;31:587–591. 28. Brown KM, Kaplan MM, Donowitz M. Extrahepatic portal venous thrombosis: frequent recognition of associated diseases. J Clin Gastroenterol 1985;7:153–159. 29. Thompson EN, Sherlock S. The aetiology of portal vein thrombosis with particular reference to the role of infection and exchange transfusion. Q J Med 1964;33:465–480. 30. Valla D, Casadevall N, Huisse MG, et al. Etiology of portal vein thrombosis in adults. A prospective evaluation of primary myeloproliferative disorders. Gastroenterology 1988;94:1063–1069. 31. Witte CL, Brewer ML, Witte MH, Pond GB. Protean manifestations of pylethrombosis. A review of thirty-four patients. Ann Surg 1985;202:191–202. 32. Belli L, Romani F, Sansalone CV, et al. Portal thrombosis in cirrhotics. A retrospective analysis. Ann Surg 1986;203:286–291. 33. Amitrano L, Brancaccio V, Guardascione MA, et al. Inherited coagulation disorders in cirrhotic patients with portal vein thrombosis. Hepatology 2000;31:345–348. 34. Arkel YS. Thrombosis and cancer. Semin Oncol 2000; 27:362–374. 35. Lang R, Inbal A, Jutrin I, Ravid M. Recurrent venous thrombosis: the sole manifestation of an occult myeloproliferative disease. Isr J Med Sci 1982;18:705–797.
Chapter 47 PORTAL AND SPLENIC VEIN THROMBOSIS
36. Sobhonslidsuk A, Reddy KR. Portal vein thrombosis: a concise review. Am J Gastroenterol 2002;97:535–541. 37. Ni YH, Wang NC, Peng MY, et al. Bacteroides fragilis bacteremia associated with portal vein and superior mesentery vein thrombosis secondary to antithrombin III and protein C deficiency: a case report. J Microbiol Immunol Infect 2002;35:255–258. 38. Schwartz DS, Gettner PA, Konstantino MM, et al. Umbilical venous catheterization and the risk of portal vein thrombosis. J Pediatr 1997;131:760–762. 39. Reh TE, Srivisal S, Schmidt EH 3rd. Portal venous thrombosis in ulcerative colitis: CT diagnosis with angiographic correlation. J Comput Assist Tomogr 1980;4:545–547. 40. Bernades P, Baetz A, Levy P, et al. Splenic and portal venous obstruction in chronic pancreatitis. A prospective longitudinal study of a medical–surgical series of 266 patients. Dig Dis Sci 1992;37:340–346. 41. Bertina RM. Factor V Leiden and other coagulation factor mutations affecting thrombotic risk. Clin Chem 1997;43:1678–1683. 42. Margaglione M, Brancaccio V, Giuliani N, et al. Increased risk for venous thrombosis in carriers of the prothrombin GÆA20210 gene variant. Ann Intern Med 1998; 129:89–93. 43. Brown K, Luddington R, Williamson D, et al. Risk of venous thromboembolism associated with a G to A transition at position 20210 in the 3¢-untranslated region of the prothrombin gene. Br J Haematol 1997;98:907–909. 44. Margaglione M, D’Andrea G, d’Addedda M, et al. The methylenetetrahydrofolate reductase TT677 genotype is associated with venous thrombosis independently of the coexistence of the FV Leiden and the prothrombin A20210 mutation. Thromb Haemost 1998;79:907–911. 45. Janssen HL, Meinardi JR, Vleggaar FP, et al. Factor V Leiden mutation, prothrombin gene mutation, and deficiencies in coagulation inhibitors associated with Budd–Chiari syndrome and portal vein thrombosis: results of a case–control study. Blood 2000;96:2364–2368. 46. Mahmoud AE, Elias E, Beauchamp N, Wilde JT. Prevalence of the factor V Leiden mutation in hepatic and portal vein thrombosis. Gut 1997;40:798–800. 47. Egesel T, Buyukasik Y, Dundar SV, et al. The role of natural anticoagulant deficiencies and factor V Leiden in the development of idiopathic portal vein thrombosis. J Clin Gastroenterol 2000;30:66–71. 48. Capron JP, Lemay JL, Muir JF, et al. Portal vein thrombosis and fatal pulmonary thromboembolism associated with oral contraceptive treatment. J Clin Gastroenterol 1981; 3:295–298. 49. Donaldson LB, Plant RK. Pregnancy complicated by extrahepatic portal hypertension: review of literature and report of two cases. Am J Obstet Gynecol 1971;110:255–264. 50. Kojima Y, Suzuki S, Sakaguchi T, et al. Portal vein thrombosis caused by microwave coagulation therapy for hepatocellular carcinoma: report of a case. Surg Today 2000;30:844–848. 51. Broe PJ, Conley CL, Cameron JL. Thrombosis of the portal vein following splenectomy for myeloid metaplasia. Surg Gynecol Obstet 1981;152:488–492. 52. Eguchi A, Hashizume M, Kitano S, et al. High rate of portal thrombosis after splenectomy in patients with esophageal varices and idiopathic portal hypertension. Arch Surg 1991;126: 752–755. 53. Fujita F, Lyass S, Otsuka K, et al. Portal vein thrombosis following splenectomy: identification of risk factors. Am Surg 2003;69:951–956. 54. Rossi P, Passariello R, Simonetti G. Portal thrombosis: high incidence following splenectomy for portal hypertension. Gastrointest Radiol 1976;1:225–227.
55. Freedman AM, Sanyal AJ, Tisnado J, et al. Complications of transjugular intrahepatic portosystemic shunt: a comprehensive review. Radiographics 1993;13:1185–1210. 56. Leach SD, Meier GH, Gusberg RJ. Endoscopic sclerotherapy: a risk factor for splanchnic venous thrombosis. J Vasc Surg 1989;10:9–12;discussion 12–13. 57. Deboever G, Elegeert I, Defloor E. Portal and mesenteric venous thrombosis after endoscopic injection sclerotherapy. Am J Gastroenterol 1989;84:1336–1337. 58. Politoske D, Ralls P, Korula J. Portal vein thrombosis following endoscopic variceal sclerotherapy. Prospective controlled comparison in patients with cirrhosis. Dig Dis Sci 1996;41:185–190. 59. Alvarez F, Bernard O, Brunelle F, et al. Portal obstruction in children. I. Clinical investigation and hemorrhage risk. J Pediatr 1983;103:696–702. 60. Dilawari JB, Chawla YK. Spontaneous (natural) splenoadrenorenal shunts in extrahepatic portal venous obstruction: a series of 20 cases. Gut 1987;28:1198–1200. 61. Harada H, Imamura H, Miyagawa S, Kawasaki S. Fate of the human liver after hemihepatic portal vein embolization: cell kinetic and morphometric study. Hepatology 1997;26:1162–1170. 62. Bilodeau M, Aubry MC, Houle R, et al. Evaluation of hepatocyte injury following partial ligation of the left portal vein. J Hepatol 1999;30:29–37. 63. Tanaka M, Wanless IR. Pathology of the liver in Budd–Chiari syndrome: portal vein thrombosis and the histogenesis of venocentric cirrhosis, veno-portal cirrhosis, and large regenerative nodules. Hepatology 1998;27:488–496. 64. Grauer SE, Schwartz SI. Extrahepatic portal hypertension: a retrospective analysis. Ann Surg 1979;189:566–574. 65. Maddrey WC, Sen Gupta KP, Mallik KC, et al. Extrahepatic obstruction of the portal venous system. Surg Gynecol Obstet 1968;127:989–998. 66. Condat B, Pessione F, Hillaire S, et al. Current outcome of portal vein thrombosis in adults: risk and benefit of anticoagulant therapy. Gastroenterology 2001;120:490–497. 67. Voorhees AB Jr, Price JB Jr. Extrahepatic portal hypertension. A retrospective analysis of 127 cases and associated clinical implications. Arch Surg 1974;108:338–341. 68. Janssen HL. Changing perspectives in portal vein thrombosis. Scand J Gastroenterol Suppl 2000:69–73. 69. Zimmerman D, Bell WR. Venous thrombosis and splenic rupture in paroxysmal nocturnal hemoglobinuria. Am J Med 1980;68:275–279. 70. Uflacker R. Applications of percutaneous mechanical thrombectomy in transjugular intrahepatic portosystemic shunt and portal vein thrombosis. Tech Vasc Intervent Radiol 2003;6:59–69. 71. Taylor CR. Computed tomography in the evaluation of the portal venous system. J Clin Gastroenterol 1992;14:167–172. 72. Taylor CR, McCauley TR. Magnetic resonance imaging in the evaluation of the portal venous system. J Clin Gastroenterol 1992;14:268–273. 73. Parvey HR, Raval B, Sandler CM. Portal vein thrombosis: imaging findings. AJR Am J Roentgenol 1994;162:77–81. 74. Tanaka K, Numata K, Okazaki H, et al. Diagnosis of portal vein thrombosis in patients with hepatocellular carcinoma: efficacy of color Doppler sonography compared with angiography. AJR Am J Roentgenol 1993;160:1279–1283. 75. Ueno N, Tomiyama T, Tano S, et al. Color Doppler ultrasonography in the diagnosis of portal vein invasion in patients with pancreatic cancer. J Ultrasound Med 1997;16:825–830. 76. Van Gansbeke D, Avni EF, Delcour C, et al. Sonographic features of portal vein thrombosis. AJR Am J Roentgenol 1985;144:749–752.
913
Section VII. Vascular Disease of the Liver
77. Haddad MC, Clark DC, Sharif HS, et al. MR, CT, and ultrasonography of splanchnic venous thrombosis. Gastrointest Radiol 1992;17:34–40. 78. Mathieu D, Vasile N, Grenier P. Portal thrombosis: dynamic CT features and course. Radiology 1985;154:737–741. 79. Mathieu D, Vasile N, Dibie C, Grenier P. Portal cavernoma: dynamic CT features and transient differences in hepatic attenuation. Radiology 1985;154:743–748. 80. Zirinsky K, Markisz JA, Rubenstein WA, et al. MR imaging of portal venous thrombosis: correlation with CT and sonography. AJR Am J Roentgenol 1988;150:283–288. 81. Koklu S, Koksal A, Yolcu OF, et al. Isolated splenic vein thrombosis: an unusual cause and review of the literature. Can J Gastroenterol 2004;18:173–174. 82. Weber SM, Rikkers LF. Splenic vein thrombosis and gastrointestinal bleeding in chronic pancreatitis. World J Surg 2003;27:1271–1274. 83. Mortele KJ, Mergo PJ, Taylor HM, et al. Splenic and perisplenic involvement in acute pancreatitis: determination of prevalence and morphologic helical CT features. J Comput Assist Tomogr 2001;25:50–54. 84. Petit P, Bret PM, Atri M, et al. Splenic vein thrombosis after splenectomy: frequency and role of imaging. Radiology 1994;190:65–68. 85. Lewis JD, Faigel DO, Morris JB, et al. Splenic vein thrombosis secondary to focal pancreatitis diagnosed by endoscopic ultrasonography. J Clin Gastroenterol 1998;26:54–56. 86. Wiersema MJ, Chak A, Kopecky KK, Wiersema LM. Duplex Doppler endosonography in the diagnosis of splenic vein, portal vein, and portosystemic shunt thrombosis. Gastrointest Endosc 1995;42:19–26. 87. Heider TR, Azeem S, Galanko JA, Behrns KE. The natural history of pancreatitis-induced splenic vein thrombosis. Ann Surg 2004;239:876–880;discussion 880–882. 88. Makowiec F, Riediger H, Emmrich J, et al. [Prophylactic splenectomy for splenic vein thrombosis in patients undergoing resection for chronic pancreatitis]. Zentralbl Chir 2004;129:191–195. 89. Manzanet G, Sanjuan F, Orbis P, et al. Liver transplantation in patients with portal vein thrombosis. Liver Transpl 2001;7: 125–131. 90. Tzakis AG, Kirkegaard P, Pinna AD, et al. Liver transplantation with cavoportal hemitransposition in the presence of diffuse portal vein thrombosis. Transplantation 1998;65:619–624. 91. Orlando G, De Luca L, Toti L, et al. Liver transplantation in the presence of portal vein thrombosis: report from a single center. Transplant Proc 2004;36:199–202. 92. Molmenti EP, Roodhouse TW, Molmenti H, et al. Thrombendvenectomy for organized portal vein thrombosis at the time of liver transplantation. Ann Surg 2002;235:292–296. 93. Shaw BW Jr, Iwatsuki S, Bron K, Starzl TE. Portal vein grafts in hepatic transplantation. Surg Gynecol Obstet 1985;161:66–68. 94. Seu P, Shackleton CR, Shaked A, et al. Improved results of liver transplantation in patients with portal vein thrombosis. Arch Surg 1996;131:840–844;discussion 844–845. 95. Moreno Gonzalez E, Garcia Garcia I, Gomez Sanz R, et al. Liver transplantation in patients with thrombosis of the portal, splenic or superior mesenteric vein. Br J Surg 1993;80:81–85.
914
96. Figueras J, Torras J, Rafecas A, et al. Extra-anatomic venous graft for portal vein thrombosis in liver transplantation. Transpl Int 1997;10:407–408. 97. Fujimoto M, Moriyasu F, Someda H, et al. Recovery of graft circulation following percutaneous transluminal angioplasty for stenotic venous complications in pediatric liver transplantation: assessment with Doppler ultrasound. Transpl Int 1995;8:119–125. 98. Greaves M. Antiphospholipid antibodies and thrombosis. Lancet 1999;354:1031. 99. Kahn D, Terblanche J, Kitano S, Bornman P. Injection sclerotherapy in adult patients with extrahepatic portal venous obstruction. Br J Surg 1987;74:600–602. 100. Kahn D, Krige JE, Terblanche J, et al. A 15-year experience of injection sclerotherapy in adult patients with extrahepatic portal venous obstruction. Ann Surg 1994;219:34–39. 101. Abraldes JG, Bosch J. Novel approaches to treat portal hypertension. J Gastroenterol Hepatol 2002;17(Suppl 3):S232–S241. 102. Brearley S, Hawker PC, Dykes PW, Keighley MR. A lethal complication of peripheral vein vasopressin infusion. Hepatogastroenterology 1985;32:224–225. 103. Bismuth H, Franco D. Portal diversion for portal hypertension in early childhood. Ann Surg 1976;183:439–446. 104. Galloway JR, Henderson JM. Management of variceal bleeding in patients with extrahepatic portal vein thrombosis. Am J Surg 1990;160:122–127. 105. Opitz T, Buchwald AB, Lorf T, et al. The transjugular intrahepatic portosystemic stent–shunt (TIPS) as rescue therapy for complete Budd–Chiari syndrome and portal vein thrombosis. Z Gastroenterol 2003;41:413–418. 106. Blum U, Haag K, Rossle M, et al. Noncavernomatous portal vein thrombosis in hepatic cirrhosis: treatment with transjugular intrahepatic portosystemic shunt and local thrombolysis. Radiology 1995;195:153–157. 107. Orozco H, Takahashi T, Mercado MA, et al. The Sugiura procedure for patients with hemorrhagic portal hypertension secondary to extrahepatic portal vein thrombosis. Surg Gynecol Obstet 1991;173:45–48. 108. Harnar T, Johansen K, Haskey R, Barker E. Left-sided portal hypertension from pancreatic pseudotumor. Am J Gastroenterol 1982;77:639–641. 109. Condat B, Valla D. [Portal vein thrombosis: is anticoagulation indicated?]. Gastroenterol Clin Biol 2001;25:507–508. 110. Baril N, Wren S, Radin R, et al. The role of anticoagulation in pylephlebitis. Am J Surg 1996;172:449–452;discussion 452–453. 111. Malkowski P, Pawlak J, Michalowicz B, et al. Thrombolytic treatment of portal thrombosis. Hepatogastroenterology 2003;50:2098–2100. 112. Walser EM, NcNees SW, DeLa Pena O, et al. Portal venous thrombosis: percutaneous therapy and outcome. J Vasc Interv Radiol 1998;9:119–127. 113. Rosen MP, Sheiman R. Transhepatic mechanical thrombectomy followed by infusion of TPA into the superior mesenteric artery to treat acute mesenteric vein thrombosis. J Vasc Interv Radiol 2000;11:195–198.
Section VII. Vascular Disease of the Liver
48
LIVER INVOLVEMENT IN OSLER–WEBER–RENDU DISEASE (HEREDITARY HEMORRHAGIC TELANGIECTASIA, HHT) Martin Caselitz, Siegfried Wagner, and Michael P. Manns Abbreviations ALK-1 activin receptor-like kinase-1 AVMs arteriovenous malformations BMPs bone morphogenetic proteins
HCC HHT MRI
hepatocellular carcinoma hereditary hemorrhagic telangiectasia magnetic resonance imaging
GENERAL ASPECTS AND DIAGNOSTIC CRITERIA Hereditary hemorrhagic telangiectasia (HHT) or Osler–Weber– Rendu disease is a rare hereditary autosomal dominant disorder of blood vessels. Mucocutaneous and visceral fibrovascular dysplasia leads to various arteriovenous malformations (AVMs) and telangiectasia in different organs. These manifestations predominantly involve the skin, mucosa, liver, GI tract, lung, and brain (Figure 481). In particular angiodysplasia of the mucosa often leads to recurrent bleeding. These features are summarized in the classical triad of Osler’s disease: 1. multiple mucocutaneous telangiectasia (Figure 48-2) 2. epistaxis 3. positive family history The association of these features was described by Rendu in 1896 and independently by Osler in 1901 and Weber in 1907.1–3 The names of these authors appear in various orders in the common eponymous labels for this condition. In 1909 Hanes coined the term “hereditary hemorrhagic telangiectasia” in acknowledgment of the three features that by then defined the disorder.4 Initially it was suspected that visceral involvement in HHT was a rare condition, but interpretation was based on the frequency of symptomatic presentation to an astute physician. With the onset of modern imaging methods and asymptomatic screening programs a much higher visceral involvement was seen.4 Moreover, visceral involvement of HHT is responsible for the high mortality rate of affected patients.5,6 With regard to the importance of visceral involvement, an actual diagnostic score was developed based on the criteria described in Table 48-1.5 These criteria permit a high level of clinical suspicion with limited diagnostic procedures. However, symptoms of HHT are progressive over time. Family members of a
PAVMs TGF-b TGF-bR
pulmonary AVMs transforming growth factor-b TGF-b receptors
patient with HHT can only be informed that they do not have HHT if a molecular diagnosis is applied (see below).
EPIDEMIOLOGY HHT occurs in many ethnic groups with a wide geographic distribution, including Asia, Africa, and the Middle East.7,8 The prevalence of HHT is in the range between 1 in 2000 and 1 in 40 000. There are considerable differences in the geographic distribution of HHT. HHT has been found to occur in 1 in 2351 persons in the French department of Ain, 1 in 3500 on the Danish island Fünen,9 1 in 5155 in the Leeward Islands (West Indies), 1 in 16 500 in the state of Vermont, USA,10 and 1 in 39 000 in northern England.11
GENETIC BACKGROUND HHT is inherited in an autosomal dominant manner, with variable expression of clinical symptoms even among family members.12 However, penetrance of HHT is high (>95%), with an agedependent phenotype that is nearly complete by the age of 40–45 years.11,13 Up to 20% of patients lack a family history of HHT. The homozygous state appears to be lethal.14 Studies of families with HHT have identified two genes whose defects are believed to be responsible for the majority of cases of HHT. The phenotype has thus been classified as HHT-1 or HHT-2, depending on the mutated gene underlying the disorder. Both genes code for two receptors of the transforming growth factor-b (TGF-b) superfamily: endoglin and activin receptor-like kinase-1. TGF-b is a member of a supergene family of polypeptide growth factors, which include activins, inhibins, and bone morphogenetic proteins (BMPs). These growth factors share homology at a group of cysteine residues that are held together by intramolecular disul-
915
Section VII. Vascular Disease of the Liver
Table 48-1. Diagnostic criteria (Curaçao criteria), as presented by Shovlin and Tarte5 The hereditary hemorrhagic telangiectasia (HHT) diagnosis is: • “Definite” if three or four criteria are present • “Possible” or “suspected” if two criteria are present • “Unlikely” if fewer than two criteria are present Criteria Multiple telangiectasia at characteristic sites – Lips (Figure 48-2) – Oral cavitiy – Fingers – Nose • Epistaxis: spontaneous, recurrent nose bleeds • Family history: a first-degree relative with HHT according to these criteria • Visceral manifestation – Gastrointestinal telangiectasia (with or without bleeding) – Pulmonary arteriovenous malformations (AVMs) – Hepatic AVMs – Cerebral AVMs – Spinal AVMs
•
Figure 48-1. Distribution of affected organs in hereditary hemorrhagic telangiectasia.
Figure 48-2. Telangiectasia on lips and tongue.
fide bonds. TGF-b is produced by nearly every cell and each cell expresses the corresponding receptors. This factor plays a significant role in the regulation of cell proliferation and differentiation, wound-healing, angiogenesis, and embryonic development. In endothelium, TGF-b modulates endothelial cell functions (e.g., migration, proliferation, adhesion) interactions between endothelium and smooth-muscle cell layers, and vascular tone. Three highly conserved isoforms of TGF-b exist, each encoded by a separate gene: TGF-b1, TGF-b2, and TGF-b3. The cellular action of TGF-b is mediated through cell surface receptors that have intrinsic serine/threonine kinase activity. Several TGF-b receptors
916
(TGF-bR) have been identified so far: TGF-bRI, TGF-bRII, and TGF-bRIII (also known as g-glycan) and endoglin (see below). Endoglin is homologous to TGF-bRIII in its transmembrane and cytoplasmic tail regions.14 HHT-1 is caused by mutations of the endoglin gene on chromosome 9q3. Confirmation that endoglin (CD 105, ENG) mutations are causative is available from experiments in transgenic mice. Some mice carrying one normal and one mutated copy of the endoglin gene display features of HHT.15 Endoglin is the most abundant TGFb-binding protein found on endothelial cells. It consists of 658 amino acids and is an integral transmembrane protein that associates with TGF-b ligand-binding receptors and modulates cellular responses. Endoglin interacts as a binding protein with affinity to several proteins, including activin-A and several BMPs (BMP-2 and BMP-7).16 Endoglin interacts with, and becomes a component of, the TGF-b1 and TGF-b3 receptor complexes involved in endothelial cell signaling. More than 70 mutations have been identified to date and disease severity has not been correlated with type of mutation. The mutant endoglin products seen in HHT-1 are transient intracellular molecules that show no cell surface expression. Measurable levels of normal endoglin in HHT patients are thus reduced by 50%, even in “normal” vessels, suggesting that a single copy of the gene confers susceptibility to the disease but that a “second hit” or modifier genes likely contribute to the development of vascular abnormalities.17 HHT-2 – the second genetic defect – is mapped on chromosome 12q13. This results in a mutation in the activin receptor-like kinase1 (ALK-1) gene, which has a similar affinity to the TGF-b complex as endoglin. Significant expression of this gene product only occurs in endothelium, but it may also be found in peripheral blood leukocytes. ALK-1 is a type 1 cell surface receptor in the TGF-b superfamily. It modulates TGF-b signaling during the regulation of angiogenesis and plays an important role in controlling blood vessel development and repair. ALK-1 has the properties of a type I serine/threonine kinase receptor that binds to TGF-b1 only in asso-
Chapter 48 LIVER INVOLVEMENT IN OSLER-WEBER-RENDU DISEASE
ciation with TGF-b receptor type II in vitro; however the ALK-1 ligand in vivo remains unknown. About 50 mutations are identified so far. Similar to endoglin, reduced levels of functional ALK-1 are seen, suggesting a haploinsufficiency mechanism of disease.18 The presence of two disease loci provided the basis for genotype/phenotype studies of the disease.19 Recently, a large questionnaire-based study led Berg et al. to the conclusion that the HHT-1 phenotype is distinct from, and more severe than, the HHT-2 phenotype.20 In this study, an earlier onset of epistaxis and telangiectasis was present in patients with HHT-1. Additionally, pulmonary AVMs (PAVMs) were only seen in HHT-1,20 a fact that confirmed earlier observations.21 However, PAVMs can also be encountered in some patients with HHT-2.22,23 Additionally, ALK-1 mutations were also identified in kindreds with pulmonary hypertension and HHT.24
GENETIC BACKGROUND OF LIVER INVOLVEMENT Nikolopoulos et al.25 and Piantanida et al.26 described HHT families with an accumulation of liver involvement, suggesting a genotype/ phenotype correlation of hepatic manifestation in HHT. With respect to liver involvement in HHT, Olivieri et al. hypothesized from their observations on the presence of intrahepatic arteriovenous shunts in six of 10 patients with HHT-2 that mutations in the ALK-1 gene may be associated with a higher risk of liver AVMs.27 A significantly higher liver involvement had also previously been reported for two large HHT-2 families.2,28 These observations prompted another group to screen systematically for mutations in the ENG and ALK-1 genes in a group of HHT patients with and without liver involvement from Germany. The researchers found that hepatic manifestation in HHT patients is associated with mutations in the ALK-1 gene, but rarely also caused by ENG mutations.29
PATHOPHYSIOLOGY In HHT the malformations consist of aberrant vascular development with multiple dilated vessels that are lined by a single layer of endothelium that is attached to a continuous basement membrane. The smallest of the hallmark telangiectases are focal dilatations of postcapillary venules. In fully developed telangiectases the venules are markedly dilated and convoluted with excessive layers of smooth muscle without elastic fibers. These venules often connect directly to dilated arterioles. This aberrant development is also associated with a perivascular mononuclear cell infiltrate. However, no single pathognomonic histological characteristic for the telangiectasia in HHT exists. Various explanations for the characteristic bleeding of these vessels include insufficient smooth-muscle contractile elements, endothelial cell junction defects, perivascular connective tissue weakness, and endothelial cell degeneration.14 Changes of the liver in HHT are described below.
ORGAN MANIFESTATIONS The diverse manifestations of HHT involve mostly vascular abnormalities of the nose, skin, lung, brain, and gastrointestinal tract.
Table 48-2. Clinical manifestations of hereditary hemorrhagic telangiectasia Affected organ or system
Type of lesion
Frequency
Nose, skin, and oral cavity Lung Gastrointestinal tract
Telangiectases Arteriovenous malformations Telangiectases, angiodysplasia, arteriovenous malformations Telangiectases, arteriovenous malformations, arterial aneurysm, cavernous angioma Arteriovenous malformations telangiectases
80–100% 15–30% 11–44%
Central nervous system (brain and spinal cord)
Liver
8–31%
8–30%
Table 48-2 summarizes the most important organ manifestations of persons with HHT. However, angiodysplasia may occur in every organ. Cases of urogenital,11,30,31 ophthalmological,32,33 and splenic34 involvement of HHT have been reported in the literature. Furthermore, aneurysms of the coronary artery35 and the aorta36 are described.
NOSE Epistaxis caused by spontaneous bleeding from telangiectases of the nasal mucosa is the most common (95%) and earliest manifestation of HHT occurring in the majority of affected persons, but not in all. It may be so severe as to require multiple transfusions and oral iron supplementation,37 and on the other hand so mild that HHT is never suspected. Recurrent epistaxis begins by the age of 10 years in many patients and by the age of 21 years in most (> 90%), becoming more severe in the later decades in about two-thirds of affected persons.
SKIN Telangiectases of the skin typically present later in life than epistaxis. By the age of 40, most affected persons (70%) have multiple telangiectasis of the lips, tongue, palate, fingers, face, nail beds, or combination of these.4,38 There may be bleeding from cutaneous telangiectasis, but it is rarely clinically important. In these cases or for cosmetic concern laser ablation can be effective.
LUNG PAVMs are thin-walled abnormal vessels that replace normal capillaries between the pulmonary arterial and venous circulations, often resulting in bulbous sac-like structures.5,39 They are often multiple and appear in both lungs, with a predilection for the lower lobes.40 These “capillary-free” shunts provide three main clinical consequences:41 1. Pulmonary arterial blood cannot be oxygenated, leading to hypoxemia. 2. The absence of a filtering capillary bed allows embolic particles to reach the systemic circulation where it impacts on other capillary beds (e.g., central nervous system). 3. The fragile vessels may lead to hemorrhage into a bronchus or the pleural cavity. Embolic cerebral events (cerebral abscess and embolic stroke) occur in patients regardless of the degree of respiratory symptoms and still carry significant morbidity and mortality.5
917
Section VII. Vascular Disease of the Liver
Complications of PAVMs can be limited if this condition is diagnosed and treated with transcatheter embolization, which offers the safest method of treatment. Whether asymptomatic patients should be treated is a matter of debate but excellent safety profiles in experienced centers supported a trend towards earlier treatment using transcatheter embolization. Surgical management of PAVMs may be an alternative option in selected patients.5 Long-term follow-up of treated patients is important, because the growth of malformations may require further treatment. In addition, prophylactic antibiotics are recommended at the time of dental and surgical procedures to reduce the risk of brain abscess. Screening methods are based on non-invasive procedures to image PAVMs (thoracal radiography, computed tomography) or detection of right-to-left shunt (e.g., contrast echocardiography, radionuclide perfusion, arterial blood-gas measurements).
CENTRAL NERVOUS SYSTEM About 15% of HHT patients may have cerebral involvement with telangiectases, cerebral AVMs, aneurysms, or cavernous angiomas.5 However, asymptomatic HHT patients are not routinely investigated with regard to cerebral involvement, so a higher incidence (up to 23%) is presumed.42 Cerebral involvement can lead to migraine headache, seizures, ischemia of the surrounding tissue due to a steal effect, hemorrhage, and, less commonly, paraparesis.43 An important cause for neurological complications is pulmonary embolism due to PAVMs. It is estimated that in up to two-thirds of those in whom neurological symptoms develop, PAVMs are the source of the symptoms. In the remaining third complications arise from cerebral AVMs. HHT patients with neurological symptoms suggestive of cerebral involvement or pulmonary embolism deserve further assess-
ment, as in the non-HHT population, by experienced neurointerventional centers. Cerebral magnetic resonance imaging (MRI) is currently the most sensitive non-invasive test. The question of whether asymptomatic HHT patients should be screened for cerebral AVM remains debated.
GASTROINTESTINAL TRACT The gastrointestinal tract is the second most common organ system involved in HHT following the respiratory system. Gastrointestinal manifestation of HHT can be found in every section of the gut.38,44 Characteristically, gastrointestinal symptoms do not appear until the fifth or sixth decade of life.4 Telangiectasia or AVM may be seen in about 60% of patients with HHT (Figure 48-3). Hemorrhage, the most common gastrointestinal manifestation, is seen in 10–45% of patients.5,45 In a retrospective study up to 40% of patients had an upper gastrointestinal source of bleeding, up to 10% had a lower gastrointestinal source, and 50% had an indeterminate bleeding side.45 So far no studies of capsular endoscopy are available in HHT patients. Requirements for the transfusion of more than 100 units of blood due to gastrointestinal bleeding have been documented.46 Hemobilia from hepatic telangiectasia has been proposed as the cause of gastrointestinal bleeding as well.47 The basic principles for the management of acute gastrointestinal bleeding in HHT patients are the same as in any other lesion in the first line. Early recognition of HHT is important for proper management. Endoscopy remains the main important tool in diagnosing and treating this condition. Endoscopic findings include AVMs or telangiectases similar to those seen in the oral or nasal mucosa.48 Tagged red-cell scans may diagnose the origin of a subacute bleeding, but are only of limited use because of the often intermittent
Figure 48-3. Angiodysplasia in the stomach.
918
Chapter 48 LIVER INVOLVEMENT IN OSLER-WEBER-RENDU DISEASE
nature of bleeding. In the setting of severe acute hemorrhage, angiography may demonstrate the origin of bleeding as well.48 Recurrent hemorrhage is most difficult to manage. Endoscopic treatment techniques include the use of laser and argon plasma beamer; all methods have comparable results in the control of acute bleeding telangiectasia, with typical success rates exceeding 90%.49,50 However, the long-term results of endoscopy have been disappointing because of the multifocal nature of the disease, with recurrent episodes of bleeding from other sites in the gastrointestinal tract. Asymptomatic, non-bleeding lesions should not be treated because of the risk of inducing acute and/or delayed bleeding. Patients refractory to endoscopic treatment, or those with lesions not amenable to endoscopic therapy, may require angiography with arterial embolization or surgery after localization of the bleeding source. However, with further developments of endoscopic techniques, the need for surgical intervention has steadily decreased and is now rarely required. Long-term treatment of hemorrhage in HHT, regardless of the source, has been disappointing. In some small trials hormonal treatment with estrogen–progestogen combinations (ethinylestradiol 0.05 mg and norethisterone 1 mg daily, given orally) has been shown to decrease transfusion requirements in patients with gastrointestinal bleeding.46 The exact mechanism of hormonal therapy is not known so far. Epsilon-aminocaproic acid, a potent inhibitor of the fibrinolytic system, has been reported to decrease the frequency of bleeding episodes and the number of transfusions required.48 Patients should be advised to avoid nonsteroidal anti-inflammatory drugs and coumadin, which might increase the risk of bleeding.
HEPATIC MANIFESTATION OF HHT HISTORICAL ASPECTS AND EPIDEMIOLOGY OF LIVER INVOLVEMENT Liver involvement in HHT was originally suspected by Osler in 1901. The first case report was published by van Bogaert in 1935.51 Martini reviewed the literature in 197852 and grouped liver involvement in three histological subtypes: 1. telangiectasia with fibrosis or cirrhosis 2. cirrhosis without telangiectasia 3. telangiectasia without fibrosis or cirrhosis Group 2 often had a superimposed hepatopathy such as posttransfusion hepatitis or iron overload present, which leads to cirrhosis. Vascular dilatation and arteriovenous shunts with highoutput cardiac failure was found in group 1 and 3.14 Currently, it is estimated that liver involvement occurs in 8–31% of patients suffering from HHT.13,22,26,53 However, Reilly and Nostrant,45 who described hepatic manifestations in 31% of affected patients, defined this fact as elevated activity of liver enzymes and hepatomegaly. This definition of liver involvement is not appropriate in the context of Osler’s disease, because diagnosis of vascular malformations is based on imaging methods. Furthermore the prevalence of liver involvement depends on the age of the investigated patients, and patients with hepatic involvement, who remain asymp-
tomatic, may be unconsidered. Thus, the true prevalence of hepatic involvement in HHT is unknown.
HISTOPATHOLOGY AND PATHOPHYSIOLOGY OF HEPATIC INVOLVEMENT The liver has a unique vascular supply. Blood enters the liver from two sources, the portal vein and the hepatic artery, merging at the level of hepatic sinusoids and exiting through the hepatic veins. Therefore hepatic changes in HHT can be complex and multiple, including sinusoidal ectasia, arteriovenous shunts (direct communication between arterioles and sinusoids), arterioportal shunts (direct shunt between the branches of the hepatic artery and the portal vein causing portal hypertension), and portovenous shunts (connections between portal veins and sinusoids). Three forms of angiodysplasia have been identified: 1. Telangiectases are focal dilatations that originate from capillaries and postcapillary venules (or sinusoids in the liver) (Figure 48-4). These dilated vessels are lined by a single layer of endothelium attached to continuous basement membrane. The vessels are often surrounded by a mononuclear cell infiltrate. 2. AVMs are larger dilated tortuous vessels of both arterial and venous elements with interrupted elastica lamina and variable thickness in the smooth-muscle layers. AVMs are devoid of interlinking capillaries; therefore significant shunting occurs. In the liver arteriovenous shunting, arterioportal shunting and portovenous shunting may occur. The different types of shunt may explain the wide variety of clinical symptoms in HHT patients with hepatic involvement. The lesions are embedded in dense fibrous tissue. These fibrous bands may link, entrapping the hepatocytes, leading to a fine or course nodular appearance similar to cirrhosis (pseudocirrhosis) (Figure 485). The hepatocellular architecture is preserved within these nodules, including central veins and portal areas. There may be little or no hepatocellular necrosis or inflammation. Reilly and Nostrant45 performed liver biopsy in 10 HHT patients with suspected liver involvement. They found hepatic telangiectasia in 30%, iron overload in 50%, and periportal fibrosis in 80% of the affected patients. However, no case with fully developed cirrhosis, bridging necrosis, or chronic active hepatitis was found. These findings were confirmed by other authors.54 3. Aneurysms form large vessels secondary to the fragmentation of the elastic lamina and loss of the vessel muscularis. Focal nodular hyperplasia – independent of hormonal therapy – in HHT is presumed to be parenchymal hyperplasia secondary to hyperperfusion by large anomalous hepatic arteries. Hepatomegaly (>15 cm medioclavicular line), sometimes accompanied by splenomegaly, is found as a consequence of portal hypertension or intrahepatic AVMs in 44% of affected patients.55 Primary involvement of the liver has to be distinguished from secondary hepatic complications occurring in patients affected by HHT. Viral hepatitis following transfusion due to iron-deficiency anemia may cause liver cirrhosis in older HHT patients. Patients with unexplained elevated liver enzymes for more than 6 months should be
919
Section VII. Vascular Disease of the Liver
Figure 48-4. Microscopic aspect of hepatic involvement in hereditary hemorrhagic telangiectasia with dilated vessels. (Hematoxylin & eosin 100¥).
Figure 48-5. Nodular pattern of hepatic involvement in hereditary hemorrhagic telangiectasia (pseudocirrhosis).
evaluated with hepatitis B and C serologies and sonographic imaging (see below). HHT patients requiring blood transfusions should be vaccinated against hepatitis B. Hepatocellular carcinoma (HCC) is described in a few patients with hepatic manifestation of HHT.56 However, HHT with liver involvement itself cannot be considered
920
as a risk factor for HCC, but viral hepatitis in the affected patients may lead to HCC. Furthermore, hepatic iron overload is described as a complication of repetitive blood transfusions or iron supplementation in HHT. Secondary involvement of the liver in the form of peliosis hepatis,
Chapter 48 LIVER INVOLVEMENT IN OSLER-WEBER-RENDU DISEASE
hepatic adenoma, and focal nodular hyperplasia may develop as a result of long-term hormonal therapy (see section on gastrointestinal tract, above) as well. Systemic AVMs may lead to a high-output cardiac failure and congestion of the liver. The increased pressure in the sinusoids may induce fibrosis of the liver. Therefore cirrhosis in some HHTaffected patients may be caused by right-sided congestive heart failure (cardiac cirrhosis). Up to 70% of HHT patients suffering from symptomatic liver involvement are female and in their fourth or fifth decade.5,13,57,58 These findings were confirmed by other authors, referring pulmonary40,59 and cerebral5,60 vascular malformations. The increasing magnitude of hepatic vascular malformations36,61 and pulmonary arteriovenous fistulas during pregnancy indicates the potential role of hormones in the pathogenesis of vascular malformations.62,63 This is supported by the observation that gastrointestinal bleedings are treated successfully with estrogen–progestogen combinations46 and recurrent episodes of epistaxis depend on menstruation.5 However, so far the exact mechanism of hormonal influence remains unclear.
CLINICAL PRESENTATION Manifestations of HHT are generally not present at birth, but telangiectases and malformations develop with increasing age. Therefore the clinical course can be divided into three periods: 1. an asymptomatic period during the childhood 2. a hemorrhagic period with episodes of severe and recurrent epistaxis from puberty up to the third decade 3. a period of manifest organ involvement (e.g., pulmonary, hepatic) with clinical symptoms and secondary complications in some patients53 Vascular involvement of the liver may be asymptomatic in up to 50% of patients with HHT.51 However, clinical features of hepatic manifestations in HHT patient show a wide variety, including hepatomegaly, pulsatile hepatic nodules, palpable thrill, abdominal bruit, ascites, and variceal bleeding due to portal hypertension. Right upper quadrant pain is encountered in some patients, presumably because of thrombosis in telangiectasia. Recurrent hepatic encephalopathy is described in patients with liver failure, after gastrointestinal bleeding, and in patients with portovenous shunts.64 The development of encephalopathy depends on the magnitude of the shunt; if it is small, encephalopathy is unusual. Other clinical findings include dyspnea due to pulmonary hypertension, high-output congestive cardiac failure, or a hyperdynamic circulation without heart failure secondary to left-to-right shunting of blood within the liver. Abdominal AVMs may further cause a “steal syndrome” which may lead to symptoms of abdominal angina with consequent loss of weight.57 Laboratory results reveal anicteric cholestasis (elevated gglutamyltranspeptidase and alkaline phosphatase) in up to 73% of patients.57 In cases with advanced cirrhosis an impaired liver function can be observed.
DIAGNOSIS OF HEPATIC INVOLVEMENT IN HHT Diagnosis of liver involvement is based both on clinical diagnostic criteria and on imaging methods. Though anicteric cholestasis is well
described by several authors, laboratory tests are not appropriate to diagnose hepatic manifestation of HHT. Percutaneous liver biopsy may reveal typical features such as telangiectasia, hepatic congestion, and periportal fibrosis.45 However, due to a considerable risk of bleeding following biopsy and improvement of imaging methods, liver biopsies are no longer required in most patients. A transjugular wedge biopsy may be an alternative to a percutaneous biopsy in selected patients. As liver involvement in patients with HHT is a complication with a potentially life-threatening outcome, e.g., due to the massive increase of cardiac output, an early diagnosis is desired. Although a dilated hepatic artery and vascular lesions are present in most HHT patients with liver involvement, the picture of hepatic involvement in HHT is highly variable. It may be confused with other comorbidities (Table 48-3) like liver cirrhosis, focal nodular hyperplasia, and hepatic congestion. The situation is further complicated by the fact that a considerable number of older HHT patients are infected with hepatitis C virus due to a history of blood transfusions. Thus, liver involvement in HHT must be differentiated from fibrotic liver disease due to HHT. Angiography is the “gold standard” for the diagnosis of hepatic AVMs (Figure 48-6). Characteristically, dilatation and tortuosity of the hepatic artery and its branches are seen with numerous telangiectatic lesions throughout the liver and early visualization of the hepatic veins and/or right heart chambers. The angiographic appearance depends on the stage of development of the AVM. The differential diagnosis includes conditions of reduced portal venous blood flow, cavernous hemangiomas, highly vascularized liver tumors or metastatic neoplasms, cirrhosis, and hemangioendothelioma of infancy.51,65,66 Dynamic computed tomography (Figure 48-7) and MRI are very sensitive and can confirm the diagnosis of liver involvement. Typical findings include hepatic artery dilatation, disseminated telangiectasia, early filling of the hepatic vein, and a pseudonodular pattern in the liver. The aspects of differential diagnosis are similar to those mentioned above.51,67–69 One should, however, consider that selective angiography of the hepatic artery – the current gold standard – is expensive, invasive, and not readily available for repeat measurements. Therefore, ultrasound has been proposed as a non-invasive approach to screen HHT patients and to diagnose and monitor hepatic HHT lesions.7,51,65 The current literature, mostly case reports, comprises about 35 papers describing the sonographic findings in about 100 patients with hepatic HHT lesions. The characteristic sonographic picture of liver
Table 48-3. Differential diagnosis of sonographic findings in hepatic involvement of hereditary hemorrhagic telangiectasia (HHT) Differential diagnosis of sonographic findings in hepatic involvement of HHT Cirrhosis of the liver: irregular surface of the liver Tumors of the liver (especially focal nodular hyperplasia): dilated and hypertrophic branches of the hepatic artery Caroli’s syndrome, sclerosing cholangitis: dilated intrahepatic bile ducts (B-mode sonography) Arteriovenous fistulas of other origin
921
Section VII. Vascular Disease of the Liver
Figure 48-6. Angiography of hepatic involvement in hereditary hemorrhagic telangiectasia.
Figure 48-7. Computed tomography scan showing intrahepatic vascular malformations.
922
Chapter 48 LIVER INVOLVEMENT IN OSLER-WEBER-RENDU DISEASE
lesions in HHT consists of an intrahepatic hypervascularization (Figure 48-8) caused by a lack of normal tapering of arterial branches combined with a tortuous course of the hepatic artery (Figure 489). To integrate different ultrasound findings in hepatic involvement of HHT, Caselitz and co-workers described a diagnostic scoring system (Table 48-4).55 The two major criteria (dilated common hepatic artery (Figure 48-10) and intrahepatic hypervascularization (Figure 48-8) detect hepatic HHT lesions with high sensitivity and specificity. Both should be positive to establish the diagnosis of hepatic HHT lesions. If one of the major criteria is negative or not measurable, the use of minor criteria may be suggested.
The minor criteria are highly specific, with a lower sensitivity to hepatic involvement in HHT. Furthermore, flow measurements are susceptible to investigator-dependent variability. This especially applies to the arterial parameters regarding the tortuous course of the hepatic artery in HHT. Therefore one of the major criteria with at least two of the minor criteria for the diagnosis of hepatic manifestation of HHT should be combined. In addition to the major and minor criteria, a group of facultative parameters was defined. These features are characteristic of hepatic involvement in HHT but may also be present in many other conditions. They therefore support the diagnosis of hepatic HHT lesions if major or minor criteria are present.55
Figure 48-8. Intrahepatic hypervascularization due to dilated branches of the hepatic artery. (Color Doppler, 3.75-MHz convex transducer.)
Figure 48-9. Tortuous course of the hepatic artery. (B-mode; 3.75-MHz convex transducer.)
923
Section VII. Vascular Disease of the Liver
According to Caselitz and co-workers, one sonographic feature that appears to be linked to the more advanced stages of liver involvement in HHT is an irregular, nodular surface of the liver, usually addressed as pseudocirrhosis (Figure 48-11). This finding needs to be distinguished from regeneration nodules found in liver cirrhosis.55 In addition to the criteria mentioned above, sonographic findings may reveal right heart failure, portal hypertension, and different types of focal liver lesion, such as hemangiomas and focal nodular hyperplasia.51,65 Further sonographic methods include duplex sonography, which allows analysis of blood flow patterns. Characteristically, this shows a high-velocity signal in the main hepatic artery and a decreased resistance index if relevant arterial shunts are present. A pulsatile flow pattern in the portal vein together with a high velocity may indicate the presence of arterioportal shunts. Color Doppler can
Table 48-4. Sonographic criteria for hepatic involvement in hereditary hemorrhagic telangictasia according to Caselitz et al.55 Major criteria Dilated common hepatic artery >7 mm Intrahepatic arterial hypervascularization Minor criteria Vmax of the proper hepatic artery >110 cm/s Resistance index (RI) of the proper hepatic artery <0.6 Vmax of the portal vein >25 cm/s Tortuous course of the extrahepatic hepatic artery Facultative findings Dilated portal vein >13 mm Dilated liver veins >11 mm Hepatomegaly >15 cm in mid clavicular line Nodular liver margin
prevent the confusion arising from B-mode sonography that may mimic biliary dilatation. Furthermore, extrahepatic vascular malformations can be visualized by Doppler sonography as well. Abnormalities of the bile duct similar to those in Caroli’s disease or those in sclerosing cholangitis are described in the literature.47,54,70,71 The intimate anatomic relationship of the vascular abnormalities to the dilated bile ducts suggests that external vascular compression could have caused their dilatation.70 However, these findings must be distinguished from anicteric cholestasis, found regularly in hepatic manifestation of HHT. Imaging methods such as computed tomography and sonography can show focal biliary dilatation in the liver. In case of inconsistent findings, biliary disease can be confirmed by endoscopic retrograde cholangiography or magnetic resonance cholangiography. It was shown that cardiac output in HHT correlated with both arterial and portal venous diameter. This correlation was even closer when both diameters were summed up, while the diameters did not correlate with each other. This suggests that the hepatic HHT lesions may induce both arterial and portal venous dilation. The proportion to which the supplying hepatic vessels contribute to the hepatic blood flow (and subsequently to cardiac output) differs in individual cases. However, the close correlation of the combined diameters with cardiac output points to the importance of hepatic HHT lesions for the circulatory changes in these patients. This is in contrast to the situation in liver cirrhosis where arterial flow compensates for missing portal venous flow, but the increase in cardiac output is mainly caused by extrahepatic shunts. However, measurement of cardiac output is required to evaluate symptomatic patients with hepatic AVM, especially before and after therapeutic procedures. Right heart catheterization with the use of thermodilution or non-invasive echocardiography can be used to calculate cardiac output.57 In conclusion, ultrasound is a method of high sensitivity and specificity to detect hepatic involvement in HHT. Sonography is a
Figure 48-10. Dilated common hepatic artery. (B-mode; 3.75-MHz convex transducer.)
924
Chapter 48 LIVER INVOLVEMENT IN OSLER-WEBER-RENDU DISEASE Figure 48-11. Sonographic aspect of nodular patern of hepatic involvement in hereditary hemorrhagic telangiectasia (pseudocirrhosis). (B-mode; 3.75-MHz convex transducer.)
low-cost, non-invasive bedside method and does not require the application of contrast media or radiation. Thus, the risk of side effects and complications is minimized. Therefore, ultrasound is proposed as the first-line method to detect and monitor hepatic involvement in HHT patients.
THERAPY Clinical complications of hepatic AVMs may be indications for therapeutic interventions. These complications are grouped as follows:54 ∑ cardiac insufficiency induced by arteriovenous shunts ∑ complications of portal hypertension ∑ hepatic encephalopathy caused by portovenous shunts In addition, other complications of patients with HHT in combination with hepatic manifestation have been documented in the literature. A case of a female patient at the age of 33 years, who presented with symptoms of cardiac insufficiency followed by cholangiosepsis, has been documented by Bauer et al.61 Abdominal pain may be related to different causes. These pains may be caused by a tension of the hepatic capsule. On the other hand the pain may be induced by a special form of angina abdominalis followed by a steal syndrome due to hepatic vascular malformations. Since severe symptomatic hepatic involvement in HHT is quite rare and presented by multiple clinical features, a standard therapy cannot be recommended. The therapy of cardiac insufficiency and portal hypertension caused by increased cardiac output or arterioportal shunts should be a conservative one based on pharmaceutical intervention. Beta-blockers are the primary choice for these patients.14 Analgesics could be used in the case of abdominal complaints. In case of insufficient medical treatment, other therapeutic options must be considered. In general there are three options:
1. surgical ligation of the hepatic artery 2. transcatheter embolization of the hepatic artery or appropriate branches 3. liver transplantation A comparison of the different therapeutic approaches is so far based on case reports and small groups of patients. The empirical data of complications and successful outcome are rather inconsistent, so that general guidelines for therapy cannot be given at the moment. The different therapeutic options are based on different concepts and strategies. Liver transplantation is, by definition, an exchange of a pathologically altered organ by a normal substitute: as such, it may be called gene therapy. Surgical ligation and embolization of the hepatic artery are used to reduce the pathological hepatic blood flow. However, embolization of hepatic vessels is applied in various clinical indications. The method is, among others, indicated for liver tumors, bleeding, fistula, aneurysms of the hepatic artery, and intrahepatic vascular malformations. Given the various indications, the technique of embolization is different. The aim of embolization may be necrosis in cases of malignant liver tumors. In contrast, necrosis should be avoided in vascular malformations. Additional special features of arteriovenous shunts of the liver should be considered: The substrate of embolization could be transmitted into the pulmonary vessels, if shunts present with a large diameter. This may lead to right heart failure, especially in patients with latent right heart insufficiency. Therefore, the embolic particles should not pass the hepatic shunts. On the other hand, large particles may cause severe necrosis of the liver parenchyma. As a consequence, special attention should be paid to the size of the embolic particles. Whiting and coworkers used radioactive particles to avoid transmitting embolic particles to the pulmonary vessels.72 Using this method, Whiting et al. could exclude the presence of embolic material in pulmonary vessels. Ligation of the hepatic artery has been reported in only very select cases in the past 20 years, in comparison to alternative methods of
925
Section VII. Vascular Disease of the Liver
therapy.73 The surgical ligation of the hepatic artery is so far regarded as a second- or third-line option in hepatic manifestations of HHT, if liver transplantation or embolization is not possible. Data published in the literature of embolization in patients with hepatic manifestation of HHT are based on 29 patients so far. Given all data, embolization was technically successful in more than 90%. However, the clinical results after embolization differ considerably. While some authors reported severe side effects or a fatal outcome,54,73,74 other authors described a successful outcome in up to 80% of the treated patients.57 Therefore, the efficiency of this therapy is controversial.14,74 These facts may be due to different techniques of embolization. Furthermore, it cannot be ruled out that different subtypes (e.g., shunt characteristics) of hepatic vascular malformations influence the result and rate of side effects. In a recent study intrahepatic portovenous shunts were characterized by small and multiple shunts in an apparent network on the periphery with or without a large shunt.75 In the presence of marked shunting from the portal into the hepatic vein, only the hepatic artery supplies blood to the liver. Hepatic necrosis after hepatic artery embolization supports the presence of shunting from the portal vein to the hepatic veins. Therefore, especially in patients with significant portovenous shunting in some cases causing encephalopathy, arterial embolization may be harmful to the liver. Since embolization may lead to necrosis of liver parenchyma, embolization should not be used in patients with reduced liver function. If patients with insufficiency of the liver undergo embolization, only (sub-)segmental embolization should be performed in order to support regeneration of the liver parenchyma. However, liver transplantation may be the method of choice for this subgroup.73,76 Trembath and co-workers described a coincidence of HHT type 2 and pulmonary hypertension.24 In our own experience an additional female patient with hepatic vascular malformations presented with pulmonary hypertension and severe dyspnea. Embolization was technically sufficient, but no improvement of clinical symptoms could be observed. The patient died due to respiratory failure. Retrospectively no alternative therapeutic option was available for this patient, since her health status did not permit a combination of lung and liver transplantation. Vasodilators, such as sildefanil, used to treat pulmonary hypertension may aggravate a steal syndrome in cases of severe vascular malformation. The small number of treated patients does not allow enough data for a general conclusion. An analysis of subgroups may be of interest in future embolization studies. If vascular hepatic malformations cannot be effectively treated by embolization, liver transplantation may be an alternative option. So far liver transplantations have been reported in 23 HHT patients with liver involvement. Six of these patients have been primarily treated by embolization, ligation, or banding, followed by severe complications requiring liver transplantation. In another case report new vascular malformations occurred after embolization.73 If liver transplantation is discussed as an option in hepatic vascular malformations one should exclude extrahepatic vascular malformations in other organs, especially in the abdomen. In cases of advanced extrahepatic vascular malformations a liver transplantation may not be sufficient to reduce a hyperdynamic circulatory state.
926
Clinical observations of liver involvement in patients suffering from HHT are reported in most cases in the fifth to sixth decade. Therefore it is necessary to keep in mind that – in contrast to embolization – liver transplantation may be restricted by the age of the patient. Another problem of liver transplantation is a persisting lack of donor organs. Therefore, embolization therapy may be a bridge to liver transplantation in selected patients until a suitable organ becomes available. Different complications may be observed after liver transplantation and embolization. As described above, embolization may induce necrosis of the liver parenchyma, ischemic cholangitis, and cholecystitis, with potentially lethal outcome in some patients. After liver transplantation complications such as infections and insufficiency of biliary or vascular anastomosis may occur, requiring surgical intervention or retransplantation.73,76–78 A fatal outcome after embolization, surgical intervention,79 and liver transplantation was observed in singular cases.54 After successful liver transplantation a recurrence of hepatic vascular malformations should not be expected and this fact could be interpreted as a cure of HHT as regards the liver. Embolization of hepatic malformations may also have a long-lasting effect.79,80 After the acute and intermediate postintervention period (> 6 months), further complications after embolization occur rarely during longterm follow-up. In contrast, liver transplantation may be accompanied by several continuing problems such as infections or renal toxicity as a consequence of immunosuppression. Considering the heterogeneous data, it is difficult to present a proposal for general therapeutic guidelines. However, patients with symptomatic liver involvement in HHT should be presented to specialized centers, where interdisciplinary conferences are provided by hepatologists, transplant surgeons, and interventional radiologists. Multicenter trials are required to determine the efficacy of various treatment modalities for this disease. With the identification of specific gene defects, family investigations can identify members at risk.
SUMMARY HHT, also known as Osler–Weber–Rendu disease, is an autosomal dominant disorder that results in fibrovascular dysplasia with the development of telangiectases and AVMs. Telangiectases cause easy bleeding on skin and mucosal membranes, whereas AVMs may lead to serious complications when they are located in lungs, liver, and brain. Clinical manifestations vary over time, and are generally progressive. Phenotypes of HHT have been classified based on the recently identified mutated genes endoglin (HHT-1) and activin-like kinase receptor (HHT-2). Both genes encode for two receptors of the TGF-b families. Other families with phenotype HHT do not bear these mutations, therefore other genes are probably also involved. Liver involvement due to AVMs and pseudocirrhosis is reported in up to 30% of individuals affected by HHT. Generally, hepatic involvement becomes clinically symptomatic predominantly in women during the fifth to sixth decade. Large hepatic AVMs can lead to significant complications, including high-output congestive
Chapter 48 LIVER INVOLVEMENT IN OSLER-WEBER-RENDU DISEASE
heart failure, portal hypertension, hepatic encephalopathy, and abdominal ischemia. Hepatic malformations can be diagnosed by imaging methods such as Doppler sonography, computed tomography, MRI, and angiography. Ultrasound is a method of high sensitivity and specificity to detect hepatic involvement in HHT and is a low-cost, noninvasive bedside method that does not require the application of contrast media or radiation. Therefore, ultrasound is proposed as the first-line method to detect and monitor hepatic involvement in HHT patients. The therapy of hepatic involvement in symptomatic patients remains controversial. Treatment that includes segmental embolization of branch hepatic arteries may be helpful, but can lead to severe complications. An alternative option is liver transplantation, which eradicates malformations. It is crucial to select the appropriate treatment for the right patient.
REFERENCES 1. Rendu HJ. Epistaxis repetees chez un sujet porteur de petits angiomes cutanes et muqueux. Gaz Hop 1896; 13:731–733. 2. Osler W. On a family from recurrent epitaxis, associated with multiple telangiectases of the skin and mucous membranes. Bull Johns Hopkins Hosp 1901; 12:333–337. 3. Weber FP. Multiple hereditary developmental angiomata (telangiectases) of the skin and mucous membranes associated with recurrent haemorrhages. Lancet 1907; 2:160–162. 4. Guttmacher AE, Marchuk DA, White RIJ. Hereditary hemorrhagic telangiectasia. N Engl J Med 1995; 333:918–924. 5. Shovlin CL, Letarte M. Hereditary haemorrhagic telangiectasia and pulmonary arteriovenous malformations: issues in clinical management and review of pathogenic mechanisms. Thorax 1999; 54:714–729. 6. Kjeldsen AD, Vase P, Green A. Hereditary haemorrhagic telangiectasia: a population-based study of prevalence and mortality in Danish patients. J Intern Med 1999; 245:31–39. 7. Yamaguchi H, Azuma H, Shigekiyo T, et al. A novel missense mutation in the endoglin gene in hereditary hemorrhagic telangiectasia. Thromb Haemostat 1997; 77:243–247. 8. El-Harith EA, Al-Kharasani M, Ahmed MA, Stuhrmann M. Hereditary hemorrhagic telangiectasia: a case report. Saudi Med J 1999; 20:797–799. 9. Gallione CJ, Scheessele EA, Reinhardt D, et al. Two common endoglin mutations in families with hereditary hemorrhagic telangiectasia in the Netherland Antilles: evidence for a founder effect. Hum Genet 2000; 107:40–44. 10. Guttmacher AE, McKinnon WC, Upton MD. Hereditary hemorrhagic telangiectasia: a disorder in search of the genetics community. Am J Med Genet 1994; 52:252–253. 11. Porteous ME, Burn J, Proctor SJ. Hereditary haemorrhagic telangiectasia: a clinical analysis. J Med Genet 1992; 29:527–530. 12. Shovlin CL. Molecular defects in rare bleeding disorders: hereditary haemorrhagic telangiectasia. Thromb Haemostat 1997; 78:145–150. 13. Plauchu H, de Chadarevian JP, Bideau A, Robert JM. Age-related clinical profile of hereditary hemorrhagic telangiectasia in an epidemiologically recruited population. Am J Med Genet 1989; 32:291–297. 14. Larson AM. Liver disease in hereditary hemorrhagic telangiectasia. J Clin Gastroenterol 2003; 36:149–158. 15. Bourdeau A, Dumont DJ, Letarte M. A murine model of hereditary hemorrhagic telangiectasia. J Clin Invest 1999; 104:1343–1351.
16. Barbara NP, Wrana JL, Letarte M. Endoglin is an accessory protein that interacts with the signaling receptor complex of multiple members of the transforming growth factor-b superfamily. J Biol Chem 1999; 274:584–594. 17. Bourdeau A, Faughnan ME, McDonald ML, et al. Potential role of modifier genes influencing transforming growth factor-beta1 levels in the development of vascular defects in endoglin heterozygous mice with hereditary hemorrhagic telangiectasia. Am J Pathol 2001; 158:2011–2020. 18. Abdalla SA, Pece-Barbara N, Vera S, et al. Analysis of ALK-1 and endoglin in newborns from families with hereditary hemorrhagic telangiectasia type 2. Hum Mol Genet 2000; 9:1227–1237. 19. Johnson DW, Berg JN, Gallione CJ, et al. A second locus for hereditary hemorrhagic telangiectasia maps to chromosome 12. Genome Res 1995; 5:21–28. 20. Berg J, Porteous M, Reinhardt D, et al. Hereditary haemorrhagic telangiectasia: a questionnaire based study to delineate the different phenotypes caused by endoglin and ALK1 mutations. J Med Genet 2003; 40:585–590. 21. Berg JN, Guttmacher AE, Marchuk DA, Porteous ME. Clinical heterogeneity in hereditary haemorrhagic telangiectasia: are pulmonary arteriovenous malformations more common in families linked to endoglin? J Med Genet 1996; 33:256–257. 22. McDonald JE, Miller JF, Hallam SE, et al. Clinical manifestations in a large hereditary hemorrhagic telangiectasia (HHT) type 2 kindred. Am J Genet 2000; 93:320 –327. 23. Kjeldsen AD, Brusgaard K, Poulsen L, et al. Mutations in the ALK-1 gene and the phenotype of hereditary hemorrhagic telangiectasia in two large Danish families. Am J Med Genet 2001; 98:298–302. 24. Trembath RC, Thomson JR, Machado RD, et al. Clinical and molecular genetic features of pulmonary hypertension in patients with hereditary hemorrhagic telangiectasia. N Engl J Med 2001; 345:325–334. 25. Nikolopoulos N, Xynos E, Vassilakis JS. Familial occurrence of hyperdynamic circulation status due to intrahepatic fistulae in hereditary hemorrhagic telangiectasia. Hepatogastroenterology 1988; 35:167–168. 26. Piantanida M, Buscarini E, Dellavecchia C, et al. Hereditary haemorrhagic telangiectasia with extensive liver involvement is not caused by either HHT1 or HHT2. J Med Genet 1996; 33:441–443. 27. Olivieri C, Mira E, Delù G, et al. Identification of 13 new mutations in the ACVRL 1 gene in a group of 52 unselected Italian patients affected by hereditary haemorrhagic telangiectasia. J Med Genet 2002; 39:e39. 28. Lin WD, Wu JY, Hsu HB, et al. Mutation analysis of a family with hereditary hemorrhagic telangiectasia associated with hepatic arteriovenous malformation. J Formos Med Assoc 2001; 100:817–819. 29. Kuehl HK, Caselitz M, Hasenkamp S, et al. Hepatic manifestation is associated with ALK1 in hereditary hemorrhagic telangiectasia: identification of five novel ALK1 and one novel ENG mutations. Hum Mutat 2005; 25:320. 30. Hagspiel KD, Christ ER, Schopke W. Hereditary hemorrhagic telangiectasis (Osler–Rendu–Weber disease) with pulmonary, hepatic and renal disease pattern. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1995; 163:190 –192. 31. Cooke DA. Renal arteriovenous malformation demonstrated angiographically in hereditary haemorrhagic telangiectasia (Rendu–Osler–Weber disease). J R Soc Med 1986; 79:744–746. 32. Brant AM, Schachat AP, White RI Jr. Ocular manifestations in hereditary hemorrhagic telangiectasia (Rendu–Osler–Weber disease). Am J Ophthalmol 1989; 107:642–646. 33. Lepori JC, Thisse JY, Danchin N, Saudax E. Rendu–Osler familial hemorrhagic telangiectasia with retinal localization. Bull Soc Ophtalmol Fr 1981; 81:641–644.
927
Section VII. Vascular Disease of the Liver
34. Secil M, Goktay A, Dicle O, Pirnar T. Splenic vascular malformations and portal hypertension in hereditary hemorrhagic telangiectasia: sonographic findings. J Clin Ultrasound 2001; 29:56–59. 35. Kurnik PB, Heymann WR. Coronary artery ectasia associated with hereditary hemorrhagic telangiectasia. Arch Intern Med 1989; 149:2357–2359. 36. Romer W, Burk M, Schneider W. Hereditary hemorrhagic telangiectasia (Osler’s disease). Dtsch Med Wochenschr 1992; 117:669–675. 37. Aassar OS, Friedmann CM, White RI Jr. The natural history of epistaxis in hereditary hemorrhagic telangiectasia. Laryngoscope 1991; 101:977–980. 38. Haitjema T, Westermann CJ, Overtoom TT, et al. Hereditary hemorrhagic telangiectasia (Osler–Weber–Rendu disease): new insights in pathogenesis, complications, and treatment. Arch Intern Med 1996; 156:714–719. 39. Kjeldsen AD, Oxhoj H, Andersen PE, et al. Prevalence of pulmonary arteriovenous malformations (PAVMs) and occurrence of neurological symptoms in patients with hereditary haemorrhagic telangiectasia. J Intern Med 2000; 248:255–262. 40. White RI Jr, Lynch-Nyhan A, Terry P, et al. Pulmonary arteriovenous malformations: techniques and long-term outcome of embolotherapy. Radiology 1988; 169:663–669. 41. Ference BA, Shannon TM, White RI Jr, et al. Life-threatening pulmonary hemorrhage with pulmonary arteriovenous malformations and hereditary hemorrhagic telangiectasia. Chest 1994; 106:1387–1390. 42. Fulbright RK, Chaloupka JC, Putman CM, et al. MR of hereditary hemorrhagic telangiectasia: prevalence and spectrum of cerebrovascular malformations. AJNR Am J Neuroradiol 1998; 19:477–484. 43. Roman G, Fisher M, Perl DP, Poser CM. Neurological manifestations of hereditary hemorrhagic telangiectasia (Rendu–Osler–Weber disease): report of 2 cases and review of the literature. Ann Neurol 1978; 4:130–144. 44. Vase P, Grove O. Gastrointestinal lesions in hereditary hemorrhagic telangiectasia. Gastroenterology 1986; 91:1079–1083. 45. Reilly PJ, Nostrant TT. Clinical manifestations of hereditary hemorrhagic telangiectasia. Am J Gastroenterol 1984; 79:363–367. 46. Van Cutsem E, Rutgeerts P, Vantrappen G. Treatment of bleeding gastrointestinal vascular malformations with oestrogen–progesterone. Lancet 1999; 335:953–955. 47. Costa MT, Maldonado R, Valente A, et al. Hemobilia in hereditary hemorrhagic telangiectasia: an unusual complication of endoscopic retrograde cholangiopancreatography. Endoscopy 2003; 35:531–533. 48. Sharma VK, Howden CW. Gastrointestinal and hepatic manifestations of hereditary hemorrhagic telangiectasia. Dig Dis 1998; 16:169–174. 49. Longacre AV, Gross CP, Gallitelli M, et al. Diagnosis and management of gastrointestinal bleeding in patients with hereditary hemorrhagic telangiectasia. Am J Gastroenterol 2003; 98:59–65. 50. Kitamura T, Tanabe S, Koizumi W, et al. Rendu–Osler–Weber disease successfully treated by argon plasma coagulation. Gastrointest Endosc 2001; 54:525–527. 51. Bernard G, Mion F, Henry L, et al. Hepatic involvement in hereditary hemorrhagic telangiectasia: clinical, radiological, and hemodynamic studies of 11 cases. Gastroenterology 1993; 105:482–487. 52. Martini GA. The liver in hereditary haemorrhagic teleangiectasia: an inborn error of vascular structure with multiple manifestations: a reappraisal. Gut 1978; 19:531–537. 53. Kirchner J, Zipf A, Dietrich CF, et al. Universal organ involvement in Rendu–Osler–Weber disease: interdisciplinary
928
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
diagnosis and interventional therapy. Z Gastroenterol 1996; 34:747–752. Garcia-Tsao G, Korzenik JR, Young L, et al. Liver disease in patients with hereditary hemorrhagic telangiectasia. N Engl J Med 2000; 343:931–936. Caselitz M, Bahr MJ, Bleck JS, et al. Sonographic criteria for the diagnosis of hepatic involvement in hereditary hemorrhagic telangiectasia (HHT). Hepatology 2003; 37:1139–1146. Jameson CF. Primary hepatocellular carcinoma in hereditary haemorrhagic telangiectasia: a case report and literature review. Histopathology 1989; 15:550–552. Caselitz M, Wagner S, Chavan A, et al. Clinical outcome of transfemoral embolisation in patients with arteriovenous malformations of the liver in hereditary haemorrhagic telangiectasia (Weber–Rendu–Osler disease). Gut 1998; 42:123–126. Selmaier M, Cidlinsky K, Ell C, Hahn EG. Liver hemangiomatosis in Osler’s disease. Dtsch Med Wochenschr 1993; 118:1015–1019. Shovlin CL, Winstock AR, Peters AM, et al. Medical complications of pregnancy in hereditary haemorrhagic telangiectasia. Q J Med 1995; 88:879–887. Graf C, Perrett G, Torner J. Bleeding from cerebral arteriovenous malformations as part of their natural history. J Neurosurg 1983; 58:331–337. Bauer T, Britton P, Lomas D, et al. Liver transplantation for hepatic arteriovenous malformation in hereditary haemorrhagic telangiectasia. J Hepatol 1995; 22:586–590. Gammon RB, Miksa AK, Keller FS. Osler–Weber–Rendu disease and pulmonary arteriovenous fistulas. Deterioration and embolotherapy during pregnancy. Chest 1990; 98:1522–1524. Livneh A, Langevitz P, Morag B, et al. Functionally reversible hepatic arteriovenous fistulas during pregnancy in patients with hereditary hemorrhagic telangiectasia. South Med J 1988; 81:1047–1049. Fagel WJ, Perlberger R, Kauffmann RH. Portosystemic encephalopathy in hereditary hemorrhagic telangiectasia. Am J Med 1988; 85:858–860. Buscarini E, Buscarini L, Danesino C, et al. Hepatic vascular malformations in hereditary hemorrhagic telangiectasia: Doppler sonographic screening in a large family. J Hepatol 1997; 26:111–118. Hashimoto M, Tate E, Nishii T, et al. Angiography of hepatic vascular malformations associated with hereditary hemorrhagic telangiectasia. Cardiovasc Intervent Radiol 2003; 26:177–180. Buscarini E, Buscarini L, Civardi G, et al. Hepatic vascular malformations in hereditary hemorrhagic telangiectasia: imaging findings. AJR Am J Roentgenol 1994; 163:1105–1110. Kakitsubata Y, Kakitsubata S, Kiyomizu H, et al. Intrahepatic portal-hepatic venous shunts demonstrated by US, CT, and MR imaging. Acta Radiol 1996; 37:680–684. Saxena R, Hytiroglou P, Atillasoy EO, et al. Coexistence of hereditary hemorrhagic telangiectasia and fibropolycystic liver disease. Am J Surg Pathol 1998; 22:368–372. Hatzidakis AA, Gogas C, Papanikolaou N, et al. Hepatic involvement in hereditary hemorrhagic telangiectasia (Rendu–Osler–Weber disease). Eur Radiol 2002; 12 (suppl 4):S51–S55. Hillert C, Broering DC, Gundlach M, et al. Hepatic involvement in hereditary hemorrhagic telangiectasia: an unusual indication for liver transplantation. Liver Transpl 2001; 7:266–268. Whiting JHJ, Morton KA, Datz FL, et al. Embolization of hepatic arteriovenous malformations using radiolabeled and nonradiolabeled polyvinyl alcohol sponge in a patient with hereditary hemorrhagic telangiectasia: case report. J Nucl Med 1992; 33:260–262.
Chapter 48 LIVER INVOLVEMENT IN OSLER-WEBER-RENDU DISEASE
73. Pfitzmann R, Heise M, Langrehr JM, et al. Liver transplantation for treatment of intrahepatic Osler’s disease: first experiences. Transplantation 2001; 72:237–241. 74. Whiting JH Jr, Korzenik JR, Miller FJ Jr. Fatal outcome after embolotherapy for hepatic arteriovenous malformations of the liver in two patients with hereditary hemorrhagic telangiectasia. J Vasc Interv Radiol 2000; 11:855–858. 75. Matsumoto S, Mori H, Yamada Y, et al. Intrahepatic portohepatic venous shunts in Rendu–Osler–Weber disease: imaging demonstration. Eur Radiol 2004; 14:592–596. 76. Odorico JS, Hakim MN, Becker YT, et al. Liver transplantation as definitive therapy for complications after arterial embolization for hepatic manifestations of hereditary hemorrhagic telangiectasia. Liver Transpl Surg 1998; 4:483–490. 77. Boillot O, Bianco F, Viale JP, et al. Liver transplantation resolves the hyperdynamic circulation in hereditary hemorrhagic
telangiectasia with hepatic involvement. Gastroenterology 1999; 116:187–192. 78. Azoulay D, Precetti S, Emile JF, et al. Liver transplantation for intrahepatic Rendu–Osler–Weber ‘s disease: the Paul Brousse hospital experience. Gastroenterol Clin Biol 2002; 26:828–834. 79. Chavan A, Galanski M, Wagner S, et al. Hereditary hemorrhagic telangiectasia: effective protocol for embolization of hepatic vascular malformations – experience in five patients. Radiology 1998; 209:735–739. 80. Stockx L, Raat H, Caerts B, et al. Transcatheter embolization of hepatic arteriovenous fistulas in Rendu–Osler–Weber disease: a case report and review of the literature. Eur Radiol 1999; 9:1434–1437.
929
Section VIII: Liver Transplantation
PRETRANSPLANT EVALUATION AND CARE
49
Aijaz Ahmed, Emmet B. Keeffe Abbreviations AIDS acquired immunodeficiency syndrome AMA antimitochondrial antibody ANA antinuclear antibody AST aspartate aminotransferase BMI body mass index CAD coronary artery disease CMV cytomegalovirus CT computed tomography CTP Child–Turcotte–Pugh EBV Epstein–Barr virus ERCP endoscopic retrograde cholangiopancreatography FOBT fecal occult blood test
HAV HBV HCC HCV HDV HIV HSV IBD INR LDLT MELD MRI
hepatitis A virus hepatitis B virus hepatocellular carcinoma hepatitis C virus hepatitis D virus human immunodeficiency virus herpes simplex virus inflammatory bowel disease international normalized ratio living donor liver transplantation Model for End-Stage Liver Disease magnetic resonance imaging
INTRODUCTION Over the past 25 years there has been significant improvement in the 1-year patient survival rate following liver transplantation, from 30% in the early 1980s1 to the current rate of 85–90%.2 This is related in large part to judicious pretransplant evaluation and patient selection, improvements in surgical technique, the introduction of safer and more efficacious immunosuppressive regimens, and the prevention of infections historically associated with solid organ transplantation and potent immunosuppression.3
PRETRANSPLANT EVALUATION Prompt pretransplant evaluation is a critically important initial step that may influence the outcome of liver transplantation. In the current environment of an organ shortage, every effort should be made to avoid any delay in referral for liver transplant evaluation and listing. Patients with compensated cirrhosis should be monitored closely for the onset of hepatic decompensation or the development of hepatocellular carcinoma (HCC), which are triggers for referral for consideration for liver transplantation.
DEMOGRAPHICS During the last 5 years, the number of liver transplantations performed annually has remained relatively unchanged in the US, at approximately 5000 per year.4 By contrast, there has been an increase in the number of patients presenting with hepatic decompensation. Over the past few years, the number of patients listed for liver transplantation with the United Network for Organ Sharing (UNOS) has remained stationary, at 17 000–18 000. This imbalance
NIDDK OPOs PAP PBC PPD PSC PT TIPS UNOS VZV
National Institute of Diabetes and Digestive and Kidney Diseases organ procurement organizations pulmonary artery pressure primary biliary cirrhosis purified protein derivative primary sclerosing cholangitis prothrombin time transjugular intrahepatic portosystemic shunt United Network for Organ Sharing varicella-zoster virus
between demand and supply has resulted in a rise in the mortality rate among patients on the waiting list. Statistics from 2002 showed that 1777 listed patients died while awaiting liver transplantation, and 688 were removed from the list based on the development of contraindications to transplant surgery.5 These data demonstrate the limited options available to patients with hepatic decompensation and the importance of prompt pretransplant evaluation. The role of judicious patient selection for liver transplantation is even more important as physicians manage an increasing number of patients with end-stage liver disease, related in large part to the increased number of patients with chronic hepatitis C who have progressed through the natural history of this infection to end-stage liver disease or HCC.3,4
MEDICAL JUSTICE VERSUS UTILITY UNOS is committed to providing a prudent and optimal policy for organ allocation and distribution without compromising the ethical principles of medical justice and utility. In the field of liver transplantation, the concept of medical justice prioritizes the sickest patient with the longest waiting time, whereas medical utility prioritizes the patient with the highest likelihood of a successful post-transplant outcome. The Model for End-Stage Liver Disease (MELD), which is the criterion employed by UNOS for organ allocation since 2002, prioritizes medical justice and may thus predispose recipients to a higher risk of post-transplant complications. The recent surge in the number of patients presenting with hepatic decompensation over the past several years, together with the continued donor shortage, has prompted transplant centers to adhere to even more strict selection criteria in an attempt to identify patients who can withstand a prolonged waiting time and expect a reasonable outcome following transplantation. The donor short-
933
Section VIII. Liver Transplantation
age, major cuts in healthcare budgets, and reduced reimbursement rates for liver transplantation may result in a move by individual transplant selection committees to prioritize medical utility and cost-effectiveness by not listing or by delisting the sickest patients.6,7 It is difficult to predictably identify patients at higher risk for posttransplant complications, but more data are becoming available that may help exclude certain subgroups who are poor candidates for transplant surgery.8–11 For example, patients with recurrent hepatitis C virus (HCV) infection and severe allograft dysfunction who are being considered for retransplantation are at risk for a poor outcome after a second transplant and increased resource utilization.3,4 Thus, it remains controversial whether or not to offer a cadaveric organ for retransplantation to a patient with allograft failure, particularly due to recurrent hepatitis C, while an increasing number of patients wait for a donor organ to undergo their first transplant surgery.
AIMS OF LIVER TRANSPLANTATION The two major aims of transplantation for end-stage liver disease are to improve survival and to increase functional status. It is important that potential candidates for pretransplant evaluation do not have conditions that would prevent fulfillment of these aims. Statistics from UNOS on 24 000 adults who received liver transplantation between 1987 and 1998 showed 1-year, 4-year, and 10-year overall survival rates of 85%, 76%, and 61%, respectively.12 The highest 7year survival rates of 78–79% were noted in patients with the cholestatic liver disease, primary biliary cirrhosis (PBC), and primary sclerosing cholangitis (PSC). On the other hand, patients who underwent liver transplantation for hepatic malignancies demonstrated a low 7-year survival rate of 34%. A more recent similar analysis of 17 044 recipients in the UNOS database transplanted from 1990 to 1996 showed nearly identical survival rates to the earlier analysis cited above: 83.0%, 70.2% and 61.9% at 1 year, 5 years and 8 years, respectively.13 In addition, this analysis showed that survival improved significantly over time: 74.8% in 1990 to 86.2% in 1996. This analysis also demonstrated the impact of chronic hepatitis C on liver transplantation: HCV infection was the reason for 20.0% of transplants in 1990–1992 and 30.8% in 1995–1996.13 The second goal of liver transplantation, to improve quality of life, was confirmed by a meta-analysis of health-related quality of life after liver transplantation that demonstrated a significant improvement in post-transplant cognitive, physical, and psychological functioning.14
INDICATIONS FOR LIVER TRANSPLANTATION The indications for liver transplantation can be divided into four major categories: hepatic decompensation secondary to chronic liver disease; acute liver failure; hepatic malignancy confined to the liver; and inborn errors of metabolism. Patients with metabolic disorders may present with hepatic decompensation (hereditary hemochromatosis, Wilson’s disease, and a1-antitrypsin deficiency) or a histologically intact liver with hepatic metabolic defect that results in extrahepatic organ damage (type I hyperoxaluria or familial homozygous hypercholesterolemia).15 Chronic hepatitis C is now the leading indication for liver transplantation in the US (Table 49-1). Pretransplant evaluation and care can be optimized with timely recog-
934
Table 49-1. Liver Disease of Adult Transplant Recipients in the United States (UNOS Database 1987–1998; n = 24 900) Primary Liver Disease Chronic hepatitis C Alcoholic liver disease Alcoholic liver disease and hepatitis C Chronic hepatitis B Cryptogenic cirrhosis Primary biliary cirrhosis Primary sclerosing cholangitis Autoimmune hepatitis Acute liver failure Hepatic malignancy Metabolic diseases Other Unknown
Number
%
5155 4258 1106 1368 2719 2317 2178 1194 1555 951 923 1050 126
20.7 17.1 4.4 5.5 10.9 9.3 8.7 4.8 6.2 3.8 3.7 4.2 0.5
Adapted and reprinted with permission from: Seaberg EC, Belle SH, Beringer KC, et al. Liver transplantation in the United States from 1987–1998: Updated results from the Pitt-UNOS liver transplant registry. In: Cecka JM, Terasaki PI, eds. Clin Transpl 1998. Los Angeles: UCLA Tissue Typing Laboratory, 1999:17–37.
nition and prompt referral of cirrhotic patients with clinical evidence of hepatic decompensation and/or abnormal biochemical parameters indicative of major impairment of hepatic synthetic function.3,16 Incapacitating fatigue and poor nutritional status are typically noted in association with other clinical or biochemical indications for liver transplantation. Acute liver failure is a relatively uncommon indication for transplantation, accounting for 6.2% of adult liver transplants.12 Acute liver failure must be recognized promptly and patients referred for an expedited evaluation for transplantation. Currently, there are two generally accepted listing criteria for acute liver failure, based on the reports of O’Grady et al.17 at King’s College Hospital and by Bernuau et al.18 at Villejuif. These criteria are useful for the early identification of patients with a poor prognosis who would thus benefit from timely liver transplantation (Table 49-2). The King’s College listing criteria provide a high specificity for poor outcome of acute liver failure, but a low negative predictive value.19 Therefore, patients with acute liver failure who do not meet the King’s College criteria should still undergo evaluation for transplantation as spontaneous recovery cannot be reliably predicted.19,20 Over the last decade, several factors have been identified as reliable predictors of improved allograft and patient survival following transplantation for acute liver failure, including early recognition, prompt referral, and multidisciplinary intensive care management. Hepatic malignancy, an uncommon indication in the past (3.8%),12 is now emerging as a growing indication for liver transplantation owing to the increased number of patients with chronic hepatitis C who develop end-stage liver disease, and the implementation of the MELD scoring system which grants priority to selected patients with HCC in the setting of cirrhosis.21 The selection of patients with HCC for either liver transplantation or hepatic resection has been controversial. However, several studies have demonstrated that liver transplantation provides improved survival and better costeffectiveness than hepatic resection in patients with cirrhosis.22 The long-term tumor-free survival following liver transplantation is over 80% for patients with solitary tumors less than 5 cm, or up to three tumors each less than 3 cm (so-called Milan criteria).23 Preliminary
Chapter 49 PRETRANSPLANT EVALUATION AND CARE
Table 49-2. Criteria for Liver Transplantation in Fulminant Hepatic Failure Criteria of King’s College, London1 Acetaminophen patients pH < 7.3, or Prothrombin time > 6.5 (INR) and serum creatinine > 3.4 mg/dl Non-acetaminophen patients Prothrombin time > 6.5 (INR), or Any three of the following variables: Age < 10 or > 40 years Etiology: non-A, non-B hepatitis; halothane hepatitis; idiosyncratic drug reaction Duration of jaundice before encephalopathy > 7 days Prothrombin time > 3.5 (INR) Serum bilirubin > 17.5 mg/dl Criteria of Hospital Paul-Brousse, Villejuif2 Hepatic encephalopathy, and Factor V level: < 20% in patient younger than 30 years of age, or < 30% in patient 30 years of age or older INR, international normalized ratio. 1 From: O’Grady JG, Alexander GJ, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989; 97:439–445. 2 From: Bernuau J, Samuel D, Durand F, et al. Criteria for emergency liver transplantation in patients with acute viral hepatitis and factor V below 50% of normal: a prospective study [Abstract]. Hepatology 1991;14:49A. Adapted and reprinted with permission from: Yu AS, Ahmed A, Keeffe EB. Liver transplantation: evolving patient selection criteria. Can J Gastroenterol 2001;15:729–738.
data are promising regarding the role of preoperative chemoembolization to retard the progression of HCC and to reduce the risk of intraoperative tumor dissemination as patients await transplantation after listing. In one study, 1-year and 2-year survival rates of 91% and 84%, respectively, without tumor recurrence were noted in 27 recipients undergoing this protocol.24 If HCC is diagnosed incidentally on analysis of the explanted liver, survival is comparable to that in patients without tumor and the recurrence rate is lower than in patients with a pretransplant diagnosis of HCC.25 Fibrolamellar HCC is an uncommon and less aggressive variant, with a tendency to affect younger patients and with equal gender distribution.26 It is usually noted in the absence of pre-existing liver disease and is not associated with elevated a-fetoprotein levels. An aggressive approach toward surgical resection of these tumors and contiguous structures or transplantation for non-resectable tumors can yield acceptable results. The outcome following liver transplantation for other uncommon primary hepatic malignancies, such as epithelioid hemangioendothelioma and hepatoblastoma, is encouraging. However, survival following transplantation is dismal in patients with hemangiosarcoma and metastatic lesions to the liver, with the exception of neuroendocrine tumors. Budd–Chiari syndrome results in thrombotic or non-thrombotic blockage of the major hepatic veins, inferior vena cava, or both.27 The conditions that may cause Budd–Chiari syndrome include myeloproliferative syndrome (50%), tumors (10%), estrogen use and pregnancy (10%), hypercoagulable states (5%), and paroxysmal nocturnal hemoglobinura (5%), with the remaining cases (20%) being idiopathic. Progressive liver disease typically ensues and is characterized by congestion with centrilobular necrosis, fibrosis, and cirrhosis following occlusion of hepatic veins. The decision to proceed with transplantation may be facilitated by the performance of a liver biopsy. Liver transplantation is indicated for bridging fibro-
Table 49-3. Contraindications to Liver Transplantation Compensated cirrhosis without complications (CTP score 5–6) Advanced cardiopulmonary disease Human immunodeficiency (HIV) infection* Uncontrolled sepsis Cholangiocarcinoma* Extrahepatic malignancies Active alcoholism or substance abuse Anatomic abnormality precluding liver transplantation Inability to comply, e.g. psychosocial issues No potential for meaningful lifestyle post transplant, such as with irreversible and debilitating neuropsychiatric diseases CTP, Child–Turcotte–Pugh. *In most, but not all, centers cholangiocarcinoma and HIV infection are considered contraindications to liver transplantation. Transplantation for HIV is currently undergoing study at selected transplant centers. Adapted and reprinted with permission from: Yu AS, Keeffe EB. Orthotopic liver transplantation. In: Boyer T, Zakim D, eds. Hepatology, 4th edn. Philadelphia: Harcourt Health Sciences, 2002: 1617–1656.
sis and cirrhosis, whereas decompression by transjugular intrahepatic portosystemic shunt (TIPS) or shunt surgery, which is less often employed, is sufficient for mild liver disease.27,28 The prognosis following transplantation is excellent, but patients may need long-term anticoagulation. Polycystic liver disease is a rare indication for transplantation.29,30 Multiple liver cysts can be complicated by hemorrhage, infection, abdominal pain, massive cystic hepatomegaly, portal hypertension complicated by ascites, biliary obstruction, and rarely malignant transformation to cholangiocarcinoma despite intact hepatic synthetic function. Liver transplantation is curative, but remains controversial when performed in the absence of hepatic decompensation.30 The post-transplant survival rate is comparable to that for other indications of liver transplantation, with complete resolution of debilitating symptoms and complications.
CONTRAINDICATIONS TO LIVER TRANSPLANTATION: ABSOLUTE AND RELATIVE Contraindications to liver transplantation are listed in Table 49-3. A number of clinical situations, for example irreversible neurologic diseases and metastatic cancer, are obvious absolute contraindications. Other relative and absolute contraindications are discussed in the following section. In patients with one or more comorbid conditions, appropriate consultations and a thorough review by the transplant selection committee are helpful in determining whether or not to proceed with liver transplantation.
Active Alcohol or Substance Abuse Liver transplantation is contraindicated in patients with ongoing alcoholism and/or substance abuse, but can be performed once recovery has been convincingly demonstrated. The 1-year and 5-year actuarial survival rates following transplantation for patients with alcoholic liver disease in the US (82% and 68%, respectively) and in Europe (85% and 70%, respectively) are similar to the outcomes of patients transplanted for other types of chronic liver disease.31 In addition, post-transplant improvements in health-related quality of
935
Section VIII. Liver Transplantation
life are similar in patients transplanted for alcoholic liver disease versus other causes of end-stage liver disease.14 Specific improvements were also seen in employment, marital status, psychological health, and social activity following transplantation for alcoholic liver disease.32 Approximately 20% of patients transplanted for alcoholic liver disease use alcohol post transplant, with one-third of these individuals exhibiting repetitive or heavy drinking.31 However, a detrimental effect on graft or patient survival associated with resumption of drinking has only seldom been demonstrated. There are few reliable predictors of relapse in alcoholic patients after liver transplantation.33 Although not supported by all studies, abstinence of less than 6 months prior to transplantation appears to be a reasonable predictor of recidivism and is widely employed along with other criteria for listing for liver transplantation. Thus, patients are considered for liver transplantation upon completion of 6 months of sobriety, rehabilitation, and proof of compliance based on recommendations made by the transplant selection committee.31 There are no good data to determine whether some patients with sobriety less than 6 months might benefit from liver transplantation.
Older Age Liver transplantation is not contraindicated per se in patients with advanced age.34 However, pretransplant evaluation needs to be comprehensive, with emphasis on screening for comorbid conditions, such as cardiopulmonary or vascular disease and cancer. Review of the UNOS database from 1987 to 1998 showed a 1.5-fold relative risk of death following liver transplantation in recipients who were 50–59 years of age, with a further increase in mortality in those who were 60 or older.12,13 On the other hand, publications from individual liver transplant centers demonstrate comparable survival outcome between older and younger adult recipients. As many as two-thirds of older liver transplant recipients are fully functional in the post-transplant period and have improved quality of life. Paradoxically, older recipients have a senescent immune system, which results in decreased requirements for immunosuppressive drugs, and possibly a lower rate of acute allograft rejection. There is emerging evidence that despite favorable short-term survival in the elderly, long-term survival may be worse because of an increased rate of malignancy and heart disease, based on an age-related risk for these conditions. Thus, although advanced age is a negative risk factor for survival after liver transplantation, per se it should not exclude a patient from transplant surgery. However, a thorough pretransplant evaluation and careful long-term follow-up with attention to usual health maintenance issues in the elderly are mandatory.
Obesity The long-term survival rate following liver transplantation in patients with severe obesity has been controversial in reports from transplant centers. For example, in two studies survival in obese patients and non-obese controls was comparable.35,36 However, a study utilizing the UNOS database demonstrated adverse outcomes in patients with severe obesity.37 In an analysis of 18 172 recipients, 54% were classified as ranging from overweight to morbidly obese. Immediate, 1-year, and 2-year mortality was significantly higher in the morbidly obese group (body mass index (BMI) >40). In addition, 5-year mortality was significantly higher in both the severely obese (BMI 35.1–40) and the morbidly obese recipients, mostly as a result of
936
adverse cardiovascular events. The authors concluded that weight loss should be recommended for obese patients awaiting a liver transplantation, especially if their BMI is more than 35.37 Observations in the post-transplant period are mixed regarding the tendency for patients to gain weight, with some studies demonstrating weight gain and others showing steady weight after transplantation.36 The risk of wound infections, respiratory failure and systemic vascular complications in the post-transplant period can be relatively high in obese patients.36
Cardiovascular Disease The decision regarding the candidacy of a patient with cardiac risk factors is dependent on several factors, and these patients should undergo a comprehensive pretransplant evaluation.16,38 Patients with end-stage liver disease maintain a lower baseline systemic vascular resistance. Following liver transplantation, patients with unremarkable pretransplant cardiac evaluation have been noted to develop congestive heart failure and cardiac arrhythmias, most likely resulting from resolution of low systemic vascular resistance with volume overload produced by blood products and fluids, and possibly the vasoactive actions of immunosuppressive agents resulting in increased afterload.39 Cardiac contraindications for liver transplantation include symptomatic coronary artery disease, severe ventricular dysfunction, advanced cardiomyopathy, severe pulmonary hypertension, severe valvular heart disease, and aortic stenosis with significant pressure gradient and poor ventricular function.16,40 The stress of liver transplant surgery can trigger myocardial ischemia or infarction in patients with significant underlying coronary artery disease.38 In this population, the overall morbidity can be as high as 81% and mortality can approach 50%.41 Patients with coronary artery disease can be listed for liver transplantation following correction by angioplasty or bypass surgery.41,42 It is recommended that patients with compensated cirrhosis should undergo angioplasty, as bypass surgery is associated with significantly higher risk of postoperative hepatic decompensation. The experience with combined coronary bypass grafting and liver transplantation is limited and associated with a high risk of complications; this approach has been employed in only a few circumstances.43
Renal Failure Candidates for liver transplantation with coexisting renal insufficiency should undergo an extensive pretransplant evaluation in consultation with a transplant nephrologist. Kidney biopsy may or may not be needed, and is associated with a significant risk of bleeding due to coagulopathy associated with hepatic decompensation.44 The etiology of renal dysfunction in the setting of end-stage liver disease is often multifactorial. Intrinsic renal disease is not a contraindication for liver transplantation, but may be an indication for combined liver–kidney transplantation, which has a slightly reduced but acceptable outcome. Statistics from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Liver Transplantation Database demonstrates that renal insufficiency associated with acute liver failure, occurring in patients undergoing dialysis, or leading to combined liver and kidney transplantation, results in reduced patient and graft survival rates, prolonged stay in the intensive care unit, and higher costs.45
Chapter 49 PRETRANSPLANT EVALUATION AND CARE
Pulmonary Disease Patients with severe pulmonary hypertension and advanced chronic obstructive pulmonary diseases or pulmonary fibrosis in the setting of end-stage liver disease are poor candidates for liver transplantation. Relative contraindications for liver transplant surgery include reversible pulmonary processes, such as reactive airway disease or asthma, hepatic hydrothorax, muscle wasting, and infectious processes.46 Patients with a1-antitrypsin deficiency have been shown to demonstrate an improvement in pulmonary function following liver transplantation. Patients with active tuberculosis should be treated for at least 2 weeks, and optimally for several months, in the pretransplant period, with continuation of treatment for 1 year after liver transplantation.39 Patients who are found to be PPD positive during pretransplant evaluation should undergo prophylaxis with isoniazid plus pyridoxine or ofloxacin for 6 months.47 Patients with cirrhosis are at risk for developing hepatopulmonary syndrome and portopulmonary hypertension. Patients with hepatopulmonary syndrome develop intrapulmonary vascular dilatation (with right-to-left shunting) and arterial hypoxemia in the setting of portal hypertension.48 Post-transplant mortality can be predicted by the severity of pretransplant hypoxemia, with a rise in the mortality rate from less than 5–30% if pretransplant PaO2 drops below 50 mmHg.49 The rapid progression of hepatopulmonary syndrome with worsening oxygenation may warrant consideration for priority listing and expedited liver transplantation. Following transplantation, hepatopulmonary syndrome can improve and completely resolve over several months.50 Patients may need continued oxygen supplementation for as long as 12–15 months while vascular remodeling takes place. The poor prognostic indicators for post-transplant outcome include refractory hypoxemia with multiorgan failure, intracerebral hemorrhage, sepsis, and portal vein thrombosis. Patients with portopulmonary hypertension in the setting of cirrhosis must undergo risk stratification during pretransplant evaluation.51 Several predictors for pulmonary hypertension have been identified, including systemic arterial hypertension, loud pulmonary component of the second heart sound, right ventricular heave on physical examination, right ventricular dilatation, right ventricular hypertrophy, and estimated systolic pulmonary artery pressure (PAP) higher than 40 mmHg on echocardiography.52 These predictors have greater than 90% specificity but a low sensitivity, and identify less than two-thirds of patients with pulmonary hypertension during pretransplant evaluation. On the other hand, it has been demonstrated that a systolic right ventricular pressure of 50 mmHg or higher is a reliable predictor of moderate to severe pulmonary hypertension, with a sensitivity of 97% and specificity of 77%.53 Suspected patients must undergo right-sided cardiac catheterization during the pretransplant evaluation.54 Patients with a mean PAP of 50 mmHg or higher have a 100% risk of cardiopulmonary mortality following liver transplantation. Therefore, severe pulmonary hypertension (PAP 50 mmHg or higher) is an absolute contraindication for liver transplantation. On the contrary, patients with a mean PAP of less than 35 mmHg are suitable candidates for liver transplant surgery with no added increase in risk. Patients with a mean PAP between 35 and 50 mmHg, or a pulmonary vascular resistance of 250 dynes/s/cm-5 or greater, have a 50% risk of cardiopulmonary mortality post transplant. Long-term epoprostenol infu-
sion may improve PAP in this subgroup of patients and lower the risk of mortality associated with liver transplant surgery.55
Human Immunodeficiency Virus Infection Human immunodeficiency virus (HIV) infection is considered an absolute contraindication for transplantation in most liver transplant programs in the US. However, antiretroviral drug therapy has changed the natural history of HIV infection, and the role of these drugs in improving post-transplant outcomes is under study.56 The risk of developing acquired immunodeficiency syndrome (AIDS) in HIV-positive patients was shown to be higher following liver transplantation than in non-transplanted patients in the early experience with transplantation in patients with HIV infection. In addition, post-transplant immunosuppression resulted in increased risk of progression to AIDS in HIV-positive patients. Similarly, HIVnegative patients who acquired HIV infection from donor organs or blood products at the time of liver transplantation demonstrated a poor outcome. However, recent data are more promising.56 The concerns of HIV progression have not been borne out by a growing anecdotal experience in the era of antiretroviral therapy. In addition, the severity of recurrent hepatitis C, which is almost universal in patients with HIV infection coming to transplantation, does not appear to be different from that in HIV-negative patients transplanted for chronic hepatitis C. Finally, immunosuppression does not appear to increase the risk of infectious complications or the development of AIDS.
Infections Both ongoing and untreated infections are contraindications to liver transplantation and must be treated adequately during the pretransplant period. Sepsis and pneumonia are considered absolute contraindications to liver transplantation. Other serious infections, including osteomyelitis, fungal diseases and deep abscesses, must be treated prior to transplantation.39 It is recommended that spontaneous bacterial peritonitis must be treated for at least 48 hours, with documented eradication of infection on repeat paracentesis before transplant surgery.38
Retransplantation Retransplantation is a consideration in patients with recurrent HCV infection following transplantation when progressive liver disease has developed.57 Recurrent hepatitis C is an independent predictor of poor survival following liver transplantation, based on multivariate logistic regression analysis.58 The outcome after retransplantation in the setting of recurrent hepatitis C is less than optimal, with less than 50% survival rate at 12 months. The poor survival rate was unrelated to the presence or absence of complications, such as graft failure due to hepatic artery thrombosis, biliary complications, or chronic ductopenic rejection.59 Hyperbilirubinemia and renal failure that are progressive in nature were independent predictors of poor outcome following retransplantation in the setting of recurrent hepatitis C.58
Anatomic Abnormalities The presence of isolated portal vein thrombosis was previously an absolute contraindication to liver transplantation. However, with the
937
Section VIII. Liver Transplantation
advent of techniques such as thrombectomy or jump grafts, portal vein thrombosis is categorized as a relative contraindication.60 Recent data have clearly identified portal vein thrombosis as a poor prognostic indicator with a significantly higher risk for acute graft failure and portal hypertensive complications following liver transplantation.61 In addition, post-transplant survival is sharply reduced if portal vein thrombosis extends into the superior mesenteric vein. Therefore, it is important to document the patency of the portal vein during the pretransplant evaluation.
Prior Hepatic Surgery Patients with a previous history of abdominal surgery are at risk for developing adhesions with portal hypertensive collaterals. This is associated with longer operative time and higher risk of blood loss, but comparable results are obtainable.
Cholangiocarcinoma Cholangiocarcinoma can be considered a relative or absolute contraindication, depending on the tumor stage. The post-transplant prognosis is highly variable, with 5-year survival rates of 30% for advanced disease and 70% for patients with stage I and II tumors.62 Preliminary data are promising, with the use of neoadjuvant external-beam radiotherapy, brachytherapy, and chemotherapy with fluorouracil if extrahepatic disease is excluded.63 A comprehensive pretransplant evaluation and staging is recommended, including body CT scan, bone scan and exploratory laparotomy to rule out metastases.
Extrahepatic Malignancy Liver transplantation is contraindicated in patients with extrahepatic malignancy except for those with liver metastases from innocuous neuroendocrine tumors, including gastrinoma, insulinomas, glucagonomas, somatostatinomas, and carcinoid tumors.64 Living donors can be considered to provide timely transplant surgery in this patient population.
Previous Non-Hepatic Malignancies Liver transplantation is not an absolute contraindication in patients with a myeloproliferative disorder or a previous non-hepatic malignancy.65 Data from the renal transplant experience have provided insights into the risk of tumor recurrence in patients with preexisting malignancies at the time of transplantation.66 A low recurrence rate of 10% was noted in the post-transplant period with incidental renal tumors, lymphomas, thyroid cancer, and testicular, uterine and cervical carcinomas. The recurrence rate was intermediate, i.e. 11–25%, in patients with uterine body carcinoma, Wilms’ tumor, and carcinomas of the colon, prostate, and breast. A high recurrence rate of more than 25% was noted with bladder carcinoma, sarcomas, malignant melanomas, symptomatic renal carcinomas, non-melanomatous skin cancers, and myelomas. It has been recommended that patients be monitored for a mandatory tumorfree period of 2 years following curative cancer treatment before they can be considered for liver transplantation. A longer time is advocated for malignancies such as malignant melanomas, and breast and colon carcinomas.
938
TIMING OF LIVER TRANSPLANTATION Liver transplantation should be offered to a patient with cirrhosis at a point when the post-transplant survival rate exceeds the pretransplant life expectancy related to the presence of advanced liver disease. The 1-year mortality following liver transplantation is 10–15% in most transplant centers. Patients with Child’s class B cirrhosis have a 1-year survival rate of 85–90% based on the natural history of their liver disease.16,67 Therefore, it is appropriate to refer a cirrhotic patient with Child’s class B or C (or Child–Turcotte–Pugh (CTP) score 7 or higher) for consideration for liver transplantation. It is important to initiate the pretransplant evaluation as soon as the patient meets the minimal listing criteria. Delayed referral of a cirrhotic patient with severe hepatic decompensation (Child’s class C) characterized by progressive, irreversible complications or multiorgan failure is associated with poor post-transplant outcome. These patients should be precluded from costly, time-consuming evaluation based on their poor candidacy for transplantation. The Mayo model for PBC has been validated as a prognostic survival model and is a reliable tool to estimate the timing and outcome of liver transplantation. The Mayo model consists of five independent variables, including age, serum total bilirubin, serum albumin, prothrombin time, and the presence or absence of peripheral edema.68,69 Serum total bilirubin is a variable closely associated with survival in patients with PBC. Serum total bilirubin levels are within the normal range during the early stable phase of PBC and demonstrate a steady increase in the preterminal phase.70 In patients with serum total bilirubin 10 mg/dl or higher the estimated survival is 17 months. The Mayo model has demonstrated a favorable outcome with liver transplantation in patients with PBC during the progressive phase, and a poor prognosis with no change in natural history with medical management.69 The optimal time for liver transplantation with the best post-transplant survival is reflected by a Mayo risk score of 7.8 or lower. In addition to optimizing patient selection and the timing of liver transplantation, the Mayo model of PBC can predict the cost of resource utilization. The model has been validated and is clearly superior than CTP score in predicting outcome for patients with PBC.71 A modified version of the Mayo model incorporates the original five prognostic variables into a table format similar to that of the CTP score. Using the modified Mayo model, a score of 6 predicts a 1-year survival of 90.6%, which is the minimal listing criterion for liver transplantation.72 Patients with PSC may experience several complications, including recurrent episodes of bacterial cholangitis, jaundice due to a dominant stricture, or cholangiocarcinoma.73 A modified Mayo natural history model for PSC consists of five independent prognostic variables, including age, serum total bilirubin, serum aspartate aminotransferase (AST), serum albumin, and history of variceal bleeding.70,73 In contrast to the Mayo PBC model, the Mayo model for PSC is less reliable than the CTP score in predicting posttransplant survival.74
SELECTION CRITERIA FOR LIVER TRANSPLANT LISTING In 1964, Child and Turcotte proposed a system to categorize cirrhotic patients into three classes based on important clinical parameters, including serum total bilirubin, serum albumin, severity of
Chapter 49 PRETRANSPLANT EVALUATION AND CARE
ascites, degree of hepatic encephalopathy, and nutritional status, to estimate the risk of portosystemic shunt surgery in the setting of end-stage liver disease and variceal bleeding.75 Several years later, in 1973, Pugh and colleagues reevaluated and modified the Child’s classification by adding prothrombin time and eliminating nutritional status (CTP scoring system).76 The five clinical variables in the CTP classification can be scored based on their severity (score 1–3) with an aggregate score varying from 5 to 15. The total score allows classification into Child’s class A (CTP score 5–6), B (CTP score 7–10), or C (CTP score 11–15). The degree of hyperbilirubinemia in PBC and PSC uses a different point system when the CTP score is calculated. In addition, other indications for liver transplantation that may apply to patients with PBC and PSC include intractable pruritus, progressive bone disease with recurrent fractures and, in the case of PSC, recurrent bacterial cholangitis. The usefulness of CTP score in predicting the prognosis of patients with end-stage liver disease has been validated.77 Based on post-transplant 1-year survival of 85–90%, it has been recommended that initial listing of patients for liver transplantation should be pursued when the estimated 1-year survival rate of the underlying chronic liver disease is lower than 90%.16,67 A Child’s score of 7 or greater (Child’s class of B or C) is associated with a 1-year survival rate of 90% or lower. Patients with Child’s class A in the setting of cirrhosis with history of gastrointestinal bleeding caused by portal hypertension, or a single episode of spontaneous bacterial peritonitis, as well as patients with acute liver failure or non-metastatic primary hepatocellular cancer, can also be listed for liver transplantation. Despite the implementation of the MELD score for organ allocation, the previously employed minimal listing criteria for liver transplantation remains in effect.67
DIAGNOSTIC STUDIES AND CONSULTATIONS Patients being evaluated for liver transplantation should have no medical contraindications and meet the minimal listing criteria for transplantation. Patients should be considered for transplantation if there are no other alternative forms of treatment available to manage hepatic decompensation that might delay or prevent the need for transplantation.78 Following the initial screening, the transplant hepatologist and nurse coordinators proceed with a comprehensive evaluation to exclude any medical or psychosocial contraindications to liver transplantation (Table 49-4). Evaluation by a social worker is conducted to assess psychosocial status and the availability of social support, which plays an important role in the pretransplant care and management of the patient. Psychiatric consultation may be requested in patients with a history of psychiatric disorder, substance abuse, or alcoholic liver disease. Routine diagnostic tests include ABO blood typing, complete blood count, hepatic and renal chemistries, a-fetoprotein, viral serologies (hepatitis A, hepatitis B, hepatitis C, HIV, cytomegalovirus), chest X-ray, electrocardiogram, tuberculin skin test, and creatinine clearance (Table 49-4). The imaging tests include Doppler ultrasound, and CT or MRI to document the patency of hepatic vasculature and to screen for HCC. Patients over 50 undergo dobutamine echocardiography or the thallium stress test to rule out underlying silent coronary artery disease. Patients with abnormal stress tests or major risk factors for coronary artery disease are referred for cardiology consultation, and cardiac catheterization may be needed.41,42 Doppler imaging of the carotid
Table 49-4. Pretransplant Evaluation for Liver Transplantation Standard blood tests Complete blood count, liver chemistry, kidney profile, coagulation profile (PT, PTT) ANA, smooth muscle antibody, AMA Iron studies, ceruloplasmin, a1-antitrypsin phenotype CMV, EBV, HSV, VZV, HIV; syphilis; toxoplasmosis HAV-HDV serologies a-Fetoprotein Other standard tests Abdominal ultrasound with Doppler, electrocardiogram, chest X-ray, pulmonary function tests, endoscopic evaluations PPD skin tests Standard consultations Dietary Psychosocial Women’s health (Pap smear, mammogram in women over 35) Financial (insurance clearance must be obtained) Overall assessment of patient (clinical judgment in addition to biochemical parameters) Other optional tests CT or MRI (to exclude HCC); angiography if needed to exclude vascular abnormalities Carotid duplex scanning (for older or cardiovascular patients) Contrast echocardiography (for suspected hepatopulmonary syndrome) Cardiac catheterization (for suspected CAD) Colonoscopy (history of IBD, PSC, polyps, family history of colon cancer, (+) FOBT ERCP (in PSC) Liver biopsy Fungal serologies (in areas endemic for dimorphal fungi) Other optional measures Pretransplant vaccines, if needed (hepatitis A and B, pneumococcal vaccine, influenza vaccine, tetanus booster) PT, prothrombin time; ANA, antinuclear antibody; AMA, antimitochondrial antibody; EBV, Epstein–Barr virus; CMV, cytomegalovirus; HSV, herpes simplex virus; VZV, varicella-zoster virus; HIV, human immunodeficiency virus; HAV, hepatitis A virus; HDV, hepatitis D virus; PPD, purified protein derivative; CTP, Child–Turcotte–Pugh; MRI, magnetic resonance imaging; CT, computed tomography; CAD, coronary artery disease; IBD, inflammatory bowel disease; PSC, primary sclerosing cholangitis; ERCP, endoscopic retrograde cholangiopancreatography; FOBT, fecal occult blood test. Adapted and reprinted with permission from: Yu AS, Keeffe EB. Orthotopic liver transplantation. In: Boyer T, Zakim D, eds. Hepatology, 4th edn. Philadelphia: Harcourt Health Sciences, 2002: 1617–1656.
arteries and peripheral vessels is obtained based on clinical suspicion following an initial comprehensive consultation by the transplant hepatologist. Pulmonary function test is indicated for patients with a history of significant tobacco use or chronic lung disease. Rightsided cardiac catheterization is indicated in patients suspected of pulmonary hypertension. Patients with pulmonary hypertension can be categorized into mild, moderate and severe, based on the results of cardiac catheterization, and an estimation of post-transplant outcome can be established. Cancer screening is conducted based on individual risk factors (colonoscopy for occult fecal blood and in patients over age 50; endoscopic retrograde cholangiography in patients with PSC; and Pap smear and mammogram in female subjects). Upper endoscopy is performed to screen for gastroesophageal varices and other stigmata of portal hypertension. Additional testing and consultations are obtained based on the patient’s medical history.
939
Section VIII. Liver Transplantation
Patients must undergo financial clearance as part of the pretransplant evaluation to document adequate insurance coverage. A dedicated financial coordinator discusses with the patient and family members the available options to obtain adequate insurance coverage in an era of rising costs.78 Following successful completion of the evaluation, the patient is presented to the liver transplant selection committee by the primary hepatologist. The committee consists of transplant hepatologists, liver transplant surgeons, transplant nurse coordinators, social worker, psychiatrist, financial coordinator and a nutritionist/dietician. The selection committee determines the candidacy of a patient and may recommend approval for listing, deferral for further evaluation, or medical management owing to poor candidacy for transplant surgery.
DONOR ALLOCATION AND EVALUATION ORGAN DEMAND VERSUS SUPPLY The disparity between organ demand and supply stresses the need for strict adherence to minimal listing criteria for transplantation. The availability of donor organs in the US is approximately 20 per million population.3 The rate is variable across Europe, with organ donation of 25 per million in Spain and less than 10 per million in Italy.3 The rate of supply of cadaveric donor livers has remained relatively steady for the last 5 years, despite numerous drives to increase donor awareness and the rapidly growing demand for transplantation.4
NEW TECHNIQUES TO IMPROVE THE DONOR POOL Several techniques have been adopted to expand the donor pool.4 These include the use of ‘marginal’ donors, cadaveric split liver transplantation, adult living donor liver transplantation (LDLT), domino transplantation, and xenotransplantation (Table 49-5).4 Domino liver transplantation may be an option in only a small percentage of patients.79 Xenotransplantation is an experimental tool but may become an option in the future.80
Marginal Donors The allocation of cadaveric livers based on the MELD score may result in exceedingly prolonged waiting periods for certain patient
Table 49-5. Current Approaches to Donor Pool Expansion Marginal donors Older donors (age > 55) Donors with fatty infiltration Anti-HBc (+) donors HCV-infected donors Split liver transplants, preferably in vivo technique Living donors Domino transplants Xenografts Adapted and reprinted with permission from: Wiesner RH, Rakela J, Ishitani MB, et al. Recent advances in liver transplantation. Mayo Clin Proc 2003;78:197–210.
940
subgroups. The use of marginal cadaveric livers includes organs from donors older than 55 years, donor livers with significant fatty change, and livers from subjects with hepatitis B virus (HBV) or HCV infection. In the past, a higher risk of primary graft non-function was associated with the use of fatty livers and organs from older subjects.81,82 Measures that have improved allograft and host survival following transplant surgery include pretransplant evaluation of allografts by the surgeon, minimizing cold ischemia time, and improvement in post-transplant care.81,83 Allografts from donors with HBV and HCV infection can be considered for liver transplantation. Allografts from donors positive for anti-HBc but negative for HBsAg and anti-HBs are reasonable to consider for transplantation. The use of hepatitis B immunoglobulin plus lamivudine or lamivudine monotherapy can eliminate the risk of de novo post-transplant HBV infection from 72% in the absence of prophylaxis.84–86 Data from UNOS and individual transplant centers have demonstrated that patients with chronic HCV infection who received HCV-positive organs rather than HCVnegative donor organs have similar allograft survival rates.87–89
Split Liver Transplantation Split liver transplantation is a surgical technique that allows two patients to undergo liver transplantation by judicious use of a single donor allograft. A right trisegment (segments 4 to 8) can be used for an adult recipient and the left lateral segment (segments 2 and 3) can be utilized for a smaller adult or a child.90 The initial experience was associated with a higher risk of biliary complications requiring reoperation, with poor survival rates. In 1998, data from King’s College Hospital showed recipient and allograft survivals of 90% and 80%, respectively.91 These data were supported by experience with split liver transplantation using in vivo split grafts at the University of California at Los Angeles, with overall allograft and patient survival rates comparable to those with whole organs.92
Living Donor Liver Transplantation LDLT is a suitable option for patients who are unable to accrue sufficient MELD score based on the current allocation system and who may never receive a cadaveric liver. A typical example is a patient who carries the diagnosis of PSC with the associated risk of cholangiocarcinoma; LDLT offers the option of elective liver transplantation. Following LDLT, the transplanted liver segment requires a few weeks to regenerate into an adequate volume and provide for the recipient’s metabolic needs.93 The risks of partial hepatic resection must be discussed with the donor, in whom hepatic resection is associated with a postoperative morbidity of 15–20% and mortality of 0.1–0.5% (Tables 49-6 and 49-7). In addition to the routine pretransplant evaluation, an adequate hepatic volume for the recipient must be confirmed. The recommended graft-to-recipient body weight ratio (GBWR) is 0.8%.94 A compromise in GBWR is associated with higher risk of post-transplant allograft failure and host mortality. LDLT recipients are at an increased risk of biliary complications (Table 49-8).94 Adult-to-adult elective LDLT was first attempted in 1994.95 In adults, LDLT typically warrants right hepatic resection, but occasionally the right trisegment for larger recipients or the left hepatic lobe for smaller recipients.96,97 LDLT surgery has been shown to
Chapter 49 PRETRANSPLANT EVALUATION AND CARE
Table 49-6. Complications in Donors for Living Donor Liver Transplantation Complications Biliary tract – leaks, biloma, stenosis Hepatic encephalopathy Wound infection Pressure sores Pulmonary – atelectasis, pneumonia, embolus Bowel obstruction Phlebitis Incisional hernia Aborted donor Cholestasis Nerve palsy Portal vein thrombosis
Table 49-9. UNOS Transplant Listing Status for Patients > 18 Years of Age % 3–8 2 1.5 1.4 1 1 1 1 1 1 1 0.5
Adapted and reprinted with permission from: Shiffman ML, Brown RS Jr, Olthoff KM, et al. Living donor liver transplantation: summary of a conference at The National Institutes of Health. Liver Transpl 2002;8:174–188.
Table 49-7. Causes of Death in Recipients after Living Donor Liver Transplantation %
Sepsis Hemorrhage Vascular complications Recurrence of native disease Other
53 14 14 2–3 16
Adapted and reprinted with permission from: Shiffman ML, Brown RS Jr, Olthoff KM, et al. Living donor liver transplantation: summary of a conference at The National Institutes of Health. Liver Transpl 2002;8:174–188.
Table 49-8. Complications Reported in Recipients of Living Donor Liver Transplantation
Postoperative bleeding Biliary tract complications: leaks, strictures Hepatic artery thrombosis or stenosis Hepatic venous outflow obstruction Intrahepatic hemorrhage
Status 2A Hospitalized in the ICU for chronic liver failure with a CTP score of >10 and one of the following: Unresponsive active variceal hemorrhage Hepatorenal syndrome Refractory ascites/hepatohydrothorax Grade 3 or 4 encephalopathy unresponsive to medical therapy Status 2B CTP score >10 or presence of hepatocellular carcinoma or a CTP score > 7 and one of the following: Unresponsive active variceal hemorrhage Hepatorenal syndrome SBP Refractory ascites Status 3 Child–Pugh score >7 but not meeting the criteria for 2B
Causes of death
Complications
Status 1 Fulminant hepatic failure Primary graft dysfunction or hepatic artery thrombosis < 7 days of transplantation Acute decompensated Wilson’s disease
% 46 15–30 3–10 5 5
Adapted and reprinted with permission from: Shiffman ML, Brown RS Jr, Olthoff KM, et al. Living donor liver transplantation: summary of a conference at The National Institutes of Health. Liver Transpl 2002;8:174–188.
provide 60–70% 1-year post-transplant survivals in patients with UNOS statuses 1 and 2A.94 Adult-to-adult LDLT has not had a major impact in terms of reducing the donor shortage. The screening process for potential LDLT donors is strict, with only 15% of donors successfully completing the pretransplant evaluation.98
MELD SCORING SYSTEM FOR DONOR ALLOCATION The MELD system was formulated based on the natural history of cirrhosis and may be biased against patients with debilitating noncirrhotic liver conditions, such as metastatic neuroendocrine tumors to the liver or polycystic liver disease. Furthermore, patients with
Status 7 Temporarily inactive
end-stage liver disease may develop incapacitating manifestations, such as hepatic coma, which has no favorable impact on their MELD score and may predispose to a higher risk of complications. These patients should be considered for an appeal to a regionalized peer review board, which may or may not grant an exemption for priority listing with higher MELD score.
Previous UNOS Organ Allocation System: CTP Scoring System Recognizing the utility of the CTP criteria to predict <90% 1-year survival, UNOS coupled this information with the known mortality rate following decompensation and developed four categories of patient: status 1, 2A, 2B and 3 (Table 49-9). These status assignments were used to determine the urgency of need for a donor organ.78 Status 1 includes patients with fulminant hepatic failure and patients with graft non-function or hepatic artery thrombosis within 7 days of transplantation. Prior to early 2002, the CTP score was used to assign a UNOS status for patients who were placed on the liver transplant waiting list. There is wide variation in the severity of liver disease within different Child’s classes. The CTP scoring system is limited by its discriminatory ability and marked variability. The limited discriminatory ability restricts further stratification and prioritization of potential candidates within a Child’s class. For example, a patient who is Child’s class B may be living at home and functioning at work, or may be admitted to a hospital in a debilitated state. In addition, one patient might have a bilirubin of 5 mg/dl and a second patient a bilirubin of 15 mg/dl, but both may be classified as Child’s class C despite an obvious difference in the severity of illness. Serum bilirubin level is an important prognostic determinant in patients with chronic liver disease. The CTP scores are unable to further
941
Section VIII. Liver Transplantation
stratify and differentiate disease severity owing to the broad Child’s class system. Therefore, patients within a certain UNOS status were allocated organs based on waiting times.67,78 On the other hand, marked variability limits the ability of CTP scoring system in the assessment of subjective variables of ascites and hepatic encephalopathy. There is a lack of uniform standards for diagnosing and grading the severity of ascites and hepatic encephalopathy. Furthermore, the laboratory tests (serum albumin and prothrombin time) included in the CTP scoring system lack standardization. CTP scores lack the ability to integrate renal dysfunction, an important prognostic indicator of survival in cirrhotic patients.99 There are no prospective studies that validate CTP scoring system as a prognostic indicator of disease severity or risk of mortality following liver transplantation.100 The CTP scoring system was utilized to create only three statuses of disease severity (i.e. 2A, 2B, and 3), which is a limitation. It results in categorizing varying degrees of disease severity within the same status, and increases the dependency on waiting time as a determinant of organ allocation.
UNOS Regions The role of local geography may also play a role in the allocation of organs. Currently, there are 11 UNOS regions in the US. UNOS delegates control of organ allocation to local Organ Procurement Organizations (OPOs), who have their own geographic boundaries within the 11 regions. This arbitrary geographic division results in allocation of organs primarily to patients within a local OPO. Geographic variability in waiting times is also an important factor. Owing to the disparity of the donor population in different regions, some waiting times may be measured in months whereas in another region waiting times may be years.101 Thus, a relatively less sick patient listed in one OPO would receive priority over a sicker patient in a nearby OPO within the same UNOS region. These issues continue to raise the question of implementing a single national waiting list. Local favoritism may benefit smaller centers, which fear that a nationalized scheme might lead to their demise.101 Patients with acute liver failure continue to be listed as status 1 and receive the highest priority for donor organs, whereas patients with chronic end-stage liver disease are allocated organs in order of the highest MELD score. OPOs allocate organs to those patients listed as status 1 in local and then regional areas. Once all patients with acute liver failure have been allocated an organ, the next organ is allocated to local and then regional patients, in descending order of their MELD scores.
New UNOS Organ Allocation System: MELD Scoring System UNOS adopted the MELD scoring system as the donor allocation criterion on 28 February 2002, following a ruling by the Department of Health and Human Services102 and the Institute of Medicine103 to emphasize medical urgency and decrease the use of waiting time for organ allocation. The MELD scoring system was originally developed as a model to assess short-term prognosis in patients undergoing TIPS.104 The model successfully predicted 3-month mortality in cirrhotic patients undergoing TIPS using serum creatinine, total bilirubin, international normalized ratio (INR) for prothrombin time, and underlying etiology for end-stage liver disease (Table 49-
942
Table 49-10. MELD Scoring Equation MELD score for TIPS = 0.957 ¥ loge (creatinine mg/dl) + 0.378 ¥ loge (bilirubin mg/dl) + 1.120 ¥ loge(INR) + 0.643 (cause of liver disease)1 2 MELD score for liver transplantation = 0.957 ¥ loge (creatinine mg/dl) + 0.378 ¥ loge (bilirubin mg/dl) + 1.120 ¥ loge (INR) + 0.643 1
0 if cholestatic or alcoholic liver disease, and 1 if other liver disease. Multiply by 10 and round to the nearest whole number. Laboratory values less than 1.0 are set to 1.0. The maximum serum creatinine considered in the MELD score equation is 4.0 mg/dl.
2
10). The calculations for different Mayo models are available online at http://www.mayoclinic.org/gi-rst/models.html. The laboratory tests are drawn based on a schedule to ensure prompt recertification of UNOS status. The Regional Review Boards of UNOS assess patients with medical issues that are not addressed by the MELD system, such as hepatopulmonary syndrome. Acute liver failure is still classed as status 1 and has the highest priority for donor allocation. The serum creatinine level in the MELD scoring system is capped at 4.0 mg/dl, which is also the level that is assigned to patients on dialysis.105,106 In 2001, the MELD score was re-evaluated as a predictor of mortality in patients listed for liver transplantation. MELD is similar to the CTP scoring system in predicting 3-month mortality.107 The MELD model was unable to demonstrate a significant improvement in the prediction of 3-month mortality with the inclusion of factors such as ascites, hepatic encephalopathy, variceal bleeding, spontaneous bacterial peritonitis, body mass index, and etiology of the underlying liver disease.107 There was a small but insignificant difference in the diagnostic accuracy of MELD when the cause of underlying liver disease was eliminated from the model. Thus, the MELD score for liver transplantation was modified (Table 49-10) to exclude the etiology of liver disease.106 However, MELD score has distinct advantages owing to the use of objective rather than subjective variables.107 Calculation of the MELD score requires a programmable calculator or can be calculated at the UNOS website.5 Variation in nutrition, volume status, and diuretic regimen can significantly influence serum creatinine levels and MELD score. Recent prospective data have demonstrated that the MELD score is superior to the CTP score in ranking patients according to the severity of liver disease and the risk of mortality.108 In addition, the year before implementation of MELD was compared to the first year that it was in use in terms of outcomes.109 In the first year after implementation, there was a 12% reduction in new listings, primarily in patients with low MELD scores. There was also a 3.5% reduction in waiting list deaths, a 10.2% increase in cadaveric transplants, and no difference in early post-transplant survival.109
MELD Prioritizes Hepatocellular Carcinoma Patients with non-metastatic HCC are eligible for priority listing and receive extra MELD points. A patient with a tumor 1.9 cm in diameter (stage 1), or patients with 1 nodule 2–5 cm or two to three nodules all <3 cm (stage 2) are eligible for bonus MELD points. Patients with stage 1 (T1) HCC were originally listed at a MELD score equivalent to a 15% mortality (MELD score of 24), and patients with a stage 2 (T2) tumor were listed at a MELD score equivalent to a 30% mortality (MELD score of 29). All patients
Chapter 49 PRETRANSPLANT EVALUATION AND CARE
must be evaluated for extrahepatic metastases and re-evaluated every 3 months. Additional MELD points (two bonus points) equivalent to a 10% mortality may be added every 3 months in those who continue to meet the criteria for stage 1 or stage 2 disease.5 The data were reviewed 6 months after the implementation of the MELD system and demonstrated a disproportionately higher number of donor organs allocated to patients with HCC. Therefore, from 27 February 2003 the listing MELD scores for T1 and T2 HCC patients were decreased to 20 and 24 for stages T1 and T2, respectively, to reflect mortality risks of 8% and 15%.
CONCLUSION In summary, it is important to select patients for pretransplant evaluation based on the two aims of liver transplantation, namely to improve the survival and functional status of patients with end-stage liver disease. Prompt referral is recommended, as the waiting time prior to transplantation continues to increase. Patients must meet the minimal listing criteria for transplantation and have no absolute contraindications. Patients with relative contraindications should be considered for a preliminary discussion with the liver transplant selection committee before embarking on a costly, time-consuming evaluation. Following approval by the transplant selection committee and successful listing, patients must be followed closely and undergo rigorous HCC screening. LDLT is cautiously being employed by many transplant centers, but has yet to have a major impact on the number of patients awaiting transplantation. Efforts must continue to increase donor awareness and the donor pool.
REFERENCES 1. Keeffe EB. Liver transplantation: current status and novel approaches to liver replacement. Gastroenterology 2001;120:749–762. 2. Smith CM, Davies DB, McBride MA. Liver transplantation in the United States: A report from the Organ Procurement and Transplantation Network. In: Cecka JM, Terasaki PI, eds. Clinical transplants 1998. Los Angeles: UCLA Tissue Typing Laboratory, 2001: 19–30. 3. Yu AS, Ahmed A, Keeffe EB. Liver transplantation: evolving patient selection criteria. Can J Gastroenterol 2001;15:729–738. 4. Wiesner RH, Rakela J, Ishitani MB, et al. Recent advances in liver transplantation. Mayo Clin Proc 2003;78:197–210. 5. US Scientific Registry (UNOS). Richmond, VA: United Network for Organ Sharing, 2003. www.unos.org. 6. Showstack J, Katz PP, Lake JR, et al. Resource utilization in liver transplantation: effects of patient characteristics and clinical practice. JAMA 1999;281:1381–1386. 7. Gilbert JR, Pascual M, Schoenfeld DA, et al. Evolving trends in liver transplantation: an outcome and charge analysis. Transplantation 1999;67:246–253. 8. Onaca NN, Levy MF, Sanchez EQ, et al. A correlation between the pretransplantation MELD score and mortality in the first two years after liver transplantation. Liver Transpl 2003;9:117–123. 9. Onaca NN, Levy MF, Netto GJ, et al. Pretransplant MELD score as a predictor of outcome after liver transplantation for chronic hepatitis C. Am J Transplant 2003;3:626–630. 10. Saab S, Wang V, Ibrahim AB, et al. MELD score predicts 1-year patient survival post-orthotopic liver transplantation. Liver Transpl 2003;9:473–476.
11. Nair S, Verma S, Thuluvath PJ. Pretransplant renal function predicts survival in patients undergoing orthotopic liver transplantation. Hepatology 2002;35:1179–1185. 12. Seaberg EC, Belle SH, Beringer KC, et al. Liver transplantation in the United States from 1987–1998: Updated results from the Pitt-UNOS liver transplant registry. In: Cecka JM, Terasaki PI, eds. Clinical transplants 1998. Los Angeles: UCLA Tissue Typing Laboratory, 1999: 17–37. 13. Roberts MS, Angus DC, Bryce CL, et al. Survival after liver transplantation in the United States: a disease-specific analysis of the UNOS database. Liver Transpl 2004;10:886–897. 14. Bravata DM, Olkin I, Barnato AE, et al. Health-related quality of life after liver transplantation: A meta-analysis. Liver Transpl Surg 1999;5:318–331. 15. McDiarmid SV. Liver transplantation for metabolic disease. In: Busuttil RW, Klintmalm GB, eds. Transplantation of the liver. Philadelphia: WB Saunders, 1996: 198–215. 16. Keeffe EB. Selection of patients for liver transplantation. In: Maddrey WC, Schiff ER, Sorrell MF, eds. Transplantation of the liver, 3rd edn. Philadelphia: Lippincott Williams & Wilkins, 2001: 5–34. 17. O’Grady JG, Alexander GJ, Hayllar KM, et al. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989;97:439–445. 18. Bernuau J, Samuel D, Durand F, et al. Criteria for emergency liver transplantation in patients with acute viral hepatitis and factor V (FV) below 50% of normal: a prospective study [Abstract]. Hepatology 1991;14:49A. 19. Shakil AO, Kramer D, Mazariegos GV, et al. Acute liver failure: clinical features, outcome analysis, and applicability of prognostic criteria. Liver Transpl 2000;6:163–169. 20. Riordan SM, Williams R. Use and validation of selection criteria for liver transplantation in acute liver failure. Liver Transpl 2000;6:170–173. 21. Steinmuller T, Jonas S, Neuhaus P. Review article: liver transplantation for hepatocellular carcinoma. Aliment Pharmacol Ther 2003;17(Suppl 2):138–144. 22. Sarasin FP, Giostra E, Mentha G, et al. Partial hepatectomy or orthotopic liver transplantation for the treatment of resectable hepatocellular carcinoma? A cost-effectiveness perspective. Hepatology 1998;28:436–442. 23. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693–699. 24. Harnois DM, Steers J, Andrews JC, et al. Preoperative hepatic artery chemoembolization followed by orthotopic liver transplantation for hepatocellular carcinoma. Liver Transpl Surg 1999;5:192–199. 25. Penn I. Hepatic transplantation for primary and metastatic cancers of the liver. Surgery 1991;110:726–734. 26. Pinna AD, Iwatsuki S, Lee RG, et al. Treatment of fibrolamellar hepatoma with subtotal hepatectomy or transplantation. Hepatology 1997;26:877–883. 27. Menon KV, Shah V, Kamath PS. The Budd–Chiari syndrome. N Engl J Med 2004;350:578–585. 28. Orug T, Soonawalla ZF, Tekin K, et al. Role of surgical portosystemic shunts in the era of interventional radiology and liver transplantation. Br J Surg 2004;91:769–773. 29. Sherlock S. Cystic diseases of the liver. In: Schiff ER, Sorrell MF, Maddrey WC, eds. Schiff ’s Diseases of the liver, 8th edn. Philadelphia: Lippincott-Raven, 1999: 1083–1090. 30. Pirenne J, Aerts R, Yoong K, et al. Liver transplantation for polycystic liver disease. Liver Transpl 2001;7:238–245. 31. Lim JK, Keeffe EB. Liver transplantation for alcoholic liver disease: current concepts and length of sobriety. Liver Transpl 2004;10(Suppl):S31–38. 32. Bravata DM, Olkin I, Barnato AE, et al. Employment and alcohol use after liver transplantation for alcoholic and
943
Section VIII. Liver Transplantation
33.
34. 35.
36. 37.
38. 39.
40.
41.
42.
43. 44. 45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
944
nonalcoholic liver disease: a systematic review. Liver Transpl 2001;7:191–203. Pageaux GP, Michel J, Coste V, et al. Alcoholic cirrhosis is a good indication for liver transplantation even for cases of recidivism. Gut 1999;45:421–426. Keswani RN, Ahmed A, Keeffe EB. Older age and liver transplantation: a review. Liver Transpl 2004;10:957–967. Nair S, Cohen DB, Cohen MP, et al. Postoperative morbidity, mortality, costs, and long-term survival in severely obese patients undergoing orthotopic liver transplantation. Am J Gastroenterol 2001;96:842–845. Keeffe EB, Gettys C, Esquivel CO. Liver transplantation in patients with severe obesity. Transplantation 1994;57:309–311. Nair S, Verma S, Thuluvath PJ. Obesity and its effect on survival in patients undergoing orthotopic liver transplantation in the United States. Hepatology 2002;35:105–109. Carithers RL. Liver transplantation. American Association for the Study of Liver Diseases. Liver Transpl 2000;6:122–135. Rosen HR, Martin P. Liver transplantation. In: Schiff ER, Sorrell MF, Maddrey WC, eds. Schiff ’s Diseases of the liver, 8th edn. Philadelphia: Lippincott-Raven, 1999: 1589–1615. Keeffe EB. Comorbidities of alcoholic liver disease that affect outcome of orthotopic liver transplantation. Liver Transpl Surg 1997;3:251–257. Plotkin JS, Johnson LB, Rustgi V, et al. Coronary artery disease and liver transplantation: the state of the art. Liver Transpl 2000;6:S53–S56. Keeffe BG, Valantine H, Keeffe EB. Detection and treatment of coronary artery disease in liver transplant candidates. Liver Transpl 2001;7:755–761. Eckhoff DE, Frenette L, Sellers MT, et al. Combined cardiac surgery and liver transplantation. Liver Transpl 2001;7:60–61. Pham PT, Pham PC, Wilkinson AH. The kidney in liver transplantation. Clin Liver Dis 2000;4:567–590. Brown RS Jr, Lombardero M, Lake JR. Outcome of patients with renal insufficiency undergoing liver or liver–kidney transplantation. Transplantation 1996;62:1788–1793. Wiesner RH. Current indications, contraindications, and timing for liver transplantation. In: Busuttil RW, Klintmalm GB, eds. Transplantation of the liver. Philadelphia: WB Saunders, 1996:71–84. Chaparro SV, Montoya JG, Keeffe EB, et al. Risk of tuberculosis in tuberculin skin-positive liver transplant patients. Clin Infect Dis 1999;29:207–208. Krowka MJ. Hepatopulmonary syndrome: recent literature (1997 to 1999) and implications for liver transplantation. Liver Transpl 2000;6:S31–S35. Krowka MJ, Porayko MK, Plevak DJ, et al. Hepatopulmonary syndrome with progressive hypoxemia as an indication for liver transplantation: case reports and literature review. Mayo Clin Proc 1997;72:44–53. Egawa H, Kasahara M, Inomata Y, et al. Long-term outcome of living related liver transplantation for patients with intrapulmonary shunting and strategy for complications. Transplantation 1999;67:712–717. Ramsay MA, Simpson BR, Nguyen AT, et al. Severe pulmonary hypertension in liver transplant candidates. Liver Transpl Surg 1997;3:494–500. Pilatis ND, Jacobs LE, Rerkpattanapipat P, et al. Clinical predictors of pulmonary hypertension in patients undergoing liver transplant evaluation. Liver Transpl 2000;6:85–91. Kim WR, Krowka MJ, Plevak DJ, et al. Accuracy of doppler echocardiography in the assessment of pulmonary hypertension in liver transplant candidates. Liver Transpl 2000;6:453–458. Krowka MJ, Plevak DJ, Findlay JY, et al. Pulmonary hemodynamics and perioperative cardiopulmonary-related mortality in patients with portopulmonary hypertension undergoing liver transplantation. Liver Transpl 2000;6:443–450.
55. Kuo PC, Johnson LB, Plotkin JS, et al. Continuous intravenous infusion of epoprostenol for the treatment of portopulmonary hypertension. Transplantation 1997;63:604–606. 56. Fung J, Eghtesad B, Patel-Tom K, et al. Liver transplantation in patients with HIV infection. Liver Transpl 2004;10(Suppl):S39–S53. 57. Berenguer M, Lopez-Labrador FX, Greenberg HB, et al. Hepatitis C virus and the host: An imbalance induced by immunosuppression? Hepatology 2000;32:433–435. 58. Rosen HR, Martin P. Hepatitis C infection in patients undergoing liver retransplantation. Transplantation 1998;66:1612–1616. 59. Sheiner PA, Schluger LK, Emre S, et al. Retransplantation for recurrent hepatitis C. Liver Transpl Surg 1997;3:130–136. 60. Moreno Gonzalez E, Garcia Garcia I, Gomez Sanz R, et al. Liver transplantation in patients with thrombosis of the portal, splenic or superior mesenteric vein. Br J Surg 1993;80:81–85. 61. Manzanet G, Sanjuan F, Orbis P, et al. Liver transplantation in patients with portal vein thrombosis. Liver Transpl 2001;7:125–131. 62. Iwatsuki S, Todo S, Marsh JW, et al. Treatment of hilar cholangiocarcinoma (Klatskin tumors) with hepatic resection or transplantation. J Am Coll Surg 1998;87:358–364. 63. De Vreede I, Steers JL, Burch PA, et al. Prolonged disease-free survival after orthotopic liver transplantation plus adjuvant chemoirradiation for cholangiocarcinoma. Liver Transpl 2000;6:309–316. 64. Pichlmayr R, Weimann A, Oldhafer KJ, et al. Role of liver transplantation in the treatment of unresectable liver cancer. World J Surg 1995;19:807–813. 65. Saigal S, Norris S, Srinivasan P, et al. Successful outcome of orthotopic liver transplantation in patients with preexisting malignant states. Liver Transpl 2001;7:11–15. 66. Penn I. The effect of immunosuppression on pre-existing cancers. Transplantation 1993;55:742–747. 67. Lucey MR, Brown KA, Everson GT, et al. Minimal criteria for placement of adults on the liver transplant waiting list: a report of a national conference organized by the American Society of Transplant Physicians an the American Association for the Study of Liver Diseases. Liver Transpl Surg 1997;3:628–637. 68. Roll J, Boyer JL, Barry D, et al. The prognostic importance of clinical and histologic features in asymptomatic and symptomatic primary biliary cirrhosis. N Engl J Med 1983;308:1–7. 69. Markus BH, Dickson ER, Grambsch PM, et al. Efficiency of liver transplantation in patients with primary biliary cirrhosis. N Engl J Med 1989;320:1709–1713. 70. Wiesner RH, Porayko MK, Dickson ER, et al. Selection and timing of liver transplantation in primary biliary cirrhosis and primary sclerosing cholangitis. Hepatology 1992;16:1290–1299. 71. Kim WR, Wiesner RH, Therneau TM, et al. Optimal timing of liver transplantation for primary biliary cirrhosis. Hepatology 1998;28:33–38. 72. Kim WR, Wiesner RH, Poterucha JJ, et al. Adaptation of the Mayo primary biliary cirrhosis natural history model for application in liver transplant candidates. Liver Transpl 2000;6:489–494. 73. Kim WR, Therneau TM, Wiesner RH, et al. A revised natural history model for primary sclerosing cholangitis. Mayo Clin Proc 2000;75:688–694. 74. Talwalkar JA, Seaberg E, Kim WR, et al. Predicting clinical and economic outcomes after liver transplantation using the Mayo primary sclerosing cholangitis model and Child–Pugh score. Liver Transpl 2000;6:753–758. 75. Child CG, Turcotte JG. Surgery and portal hypertension. Major Probl Clin Surg 1964;1:1–85. 76. Pugh RN, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding esophageal varices. Br J Surg 1973;60:649–654.
Chapter 49 PRETRANSPLANT EVALUATION AND CARE
77. Propst A, Propst T, Zangerl G, et al. Prognosis and life expectancy in chronic liver disease. Dig Dis Sci 1995;40:1805–1815. 78. Yu AS, Keeffe EB. Orthotopic liver transplantation. In: Boyer T, Zakim D, eds. Hepatology, 4th edn. Philadelphia: Harcourt Health Sciences, 2002: 1617–1656. 79. Dyer PA, Bobrow M. Domino hepatic transplantation using the liver from a patient with familial amyloid polyneuropathy. Unrelated Live Transplant Regulatory Authority (ULTRA). Transplantation 1999;67:1202. 80. Sterling RK, Fisher RA. Liver transplantation. Living donor, hepatocyte, and xenotransplantation. Clin Liver Dis 2001;5:431–460. 81. Bzeizi KI, Jalan R, Plevris JN, et al. Primary graft dysfunction after liver transplantation: from pathogenesis to prevention. Liver Transpl Surg 1997;3:137–148. 82. Marsman WA, Wiesner RH, Rodriguez L, et al. Use of fatty donor liver is associated with diminished early patient and graft survival. Transplantation 1996;62:1246–1251. 83. Detre KM, Lombardero M, Belle S, et al. Influence of donor age on graft survival after transplantation – United Network for Organ Sharing Registry. Liver Transpl Surg 1995;1:311–319. 84. Dodson SF, Bonham CA, Geller DA, et al. Prevention of de novo hepatitis B infection in recipients of hepatic allografts from anti-HBc positive donors. Transplantation 1999;68:1058–1061. 85. Holt D, Thomas R, Van Thiel D, et al. Use of hepatitis B core antibody-positive donors in orthotopic liver transplantation. Arch Surg 2002;137:572–575. 86. Loss GE, Mason AL, Blazek J, et al. Transplantation of livers from HBc Ab positive donors into HBc Ab negative recipients: a strategy and preliminary results. Clin Transpl 2001;15:55–58. 87. Vargas HE, Laskus T, Wang LF, et al. Outcome of liver transplantation in hepatitis C virus-infected patients who received hepatitis C virus-infected grafts. Gastroenterology 1999;117:149–153. 88. Marroquin CE, Marino G, Kuo PC, et al. Transplantation of hepatitis C-positive livers in hepatitis C-positive patients is equivalent to transplanting hepatitis C-negative livers. Liver Transpl 2001;7:762–768. 89. Velidedeoglu E, Desai NM, Campos L, et al. The outcome of liver grafts procured from hepatitis C-positive donors. Transplantation 2002;73:582–587. 90. Busuttil RW, Goss JA. Split liver transplantation. Ann Surg 1999;229:313–321. 91. Rela M, Vougas V, Muiesan P, et al. Split liver transplantation: King’s College Hospital experience. Ann Surg 1998;227:282–288. 92. Ghobrial RM, Yersiz H, Farmer DG, et al. Predictors of survival after in vivo split liver transplantation: analysis of 110 consecutive patients. Ann Surg 2000;232:312–323. 93. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997;276:60–66.
94. Shiffman ML, Brown RS Jr, Olthoff KM, et al. Living donor liver transplantation: summary of a conference at The National Institutes of Health. Liver Transpl 2002;8:174–188. 95. Hashikura Y, Makuuchi M, Kawasaki S, et al. Successful livingrelated partial liver transplantation to an adult patient. Lancet 1994;343:1233–1234. 96. Marcos A, Ham JM, Fisher RA, et al. Single-center analysis of the first 40 adult-to-adult living donor liver transplants using the right lobe. Liver Transpl 2000;6:296–301. 97. Marcos A. Right lobe living donor liver transplantation: a review. Liver Transpl 2000;6:3–20. 98. Trotter JF, Wachs M, Trouillot T, et al. Evaluation of 100 patients for living donor liver transplantation. Liver Transpl 2000;6:290–295. 99. Fernandez-Esparrach G, Sanchez-Fueyo A, Gines P, et al. A prognostic model for predicting survival in cirrhosis with ascites. J Hepatol 2001;34:46–52. 100. Freeman RB Jr, Wiesner RH, Harper A, et al., and UNOS/OPTN Liver Disease Severity Score, UNOS/OPTN Liver and Intestine, and UNOS/OPTN Pediatric Transplantation Committees. The new liver allocation system: moving toward evidence-based transplantation policy. Liver Transpl 2002;8:851–858. 101. Ubel PA. Geographic favoritism in liver transplantation – unfortunate or unfair? N Engl J Med 1998;339:1322–1325. 102. Anonymous. Organ procurement and transplantation network – HRSA. Final rule with comment period. Fed Reg 1998;63:16296–16338. 103. Institute of Medicine. Analysis of waiting time. In: Committee on Organ Transplantation. Assessing current policies and the potential impact of the DHHS final rule. Washington, DC: National Academy Press, 1999: 57–78. 104. Malinchoc M, Kamath PS, Gordon FD, et al. A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology 2000;31:864–871. 105. Forman LM, Lucey MR. Predicting the prognosis of chronic liver disease: an evolution from child to MELD. Mayo End-stage Liver Disease. Hepatology 2001;33:473–475. 106. Wiesner RH, McDiarmid SV, Kamath PS, et al. MELD and PELD: Application of survival models to liver allocation. Liver Transpl 2001;7:567–580. 107. Kamath PS, Wiesner RH, Malinchoc M, et al. A model to predict survival in patients with end-stage liver disease. Hepatology 2001;33:464–470. 108. Wiesner R, Edwards E, Freeman R, et al. The United Network for Organ Sharing Liver Disease Severity Score Committee. Model for end-stage liver disease (MELD) and allocation of donor livers. Gastroenterology 2003;124:91–96. 109. Freeman RB Jr, Wiesner RH, Edwards E, et al. Results of the first year of the new liver allocation plan. Liver Transpl 2004;10:7–15.
945
Section VIII: Liver Transplantation
50
LIVER TRANSPLANTATION SURGERY Carlos O. Esquivel and Linda J. Chen Abbreviations ALT alanine aminotransferase AST aspartate aminotransferase DCD donors after cardiac death
ECD HIV HTLV-1
extended-criteria donors human immunodeficiency virus human T lymphotropic virus-1
INTRODUCTION Liver transplantation began in the mid-1950s in the research laboratories of Dr. Jack Cannon at the University of California Los Angeles. Cannon was the first surgeon to attempt liver transplantation by placing the liver in a heterotopic location. In 1959–1960, Starzl in Denver and Moore in Boston pioneered the orthotopic position of the liver, by far the most common technique currently used in clinical hepatic transplantation.1 The first human liver transplant was performed by Starzl in 1963 on a child with biliary atresia. This transplant, as well as a few others, failed. In 1967, following an intense period of animal experimentation, Starzl reinitiated a more successful clinical liver transplant program. Although the outcomes at the time were poor compared to current standards, the results were encouraging. At the time, the morbidity and mortality were due to the lack of effective immunosuppressive drugs. The immunosuppression consisted of a combination of azathioprine and prednisone, a regimen copied from the experience with kidney transplantation.1 The turning point in liver transplantation took place in 1980 when ciclosporin, a potent immunosuppressive drug, was introduced to clinical transplantation. The patient and graft survival improved threefold compared to that of the pre-ciclosporin era.2 Based on these much-improved results, a National Consensus meeting sponsored by the National Institutes of Health in 1983 concluded that liver transplantation was to be the accepted mode of therapy for patients with acute or chronic end-stage liver disease.3 In 1987, another milestone marked the progress of liver transplantation. A new preservation solution was developed by Southard and Belzer.4 The new solution, containing lactobionate and other components, has become the standard in clinical transplantation. The outcomes after hepatic transplantation continue to improve due to several factors, among them the availability of potent and selective immunosuppressive drugs, improved intensive care of patients before and after transplantation, and refinements of the techniques for liver transplantation.5
DIAGNOSTIC INDICATIONS Table 50-1 lists the most common diagnostic indications for liver transplantation. The majority of these indications are associated
INR PT
international normalized ratio prothrombin time
with irreversible liver injury; however, there may be indications where the hepatic synthetic function and histology are normal, but the indication for transplantation is based on a metabolic defect that leads to injury to other organs such as primary amyloidosis, primary hyperoxaluria, and primary hypercholesterolemia. The most common diagnostic indication for liver transplantation in adult patients is hepatitis C, followed by alcoholic cirrhosis. In children, biliary atresia accounts for almost 50% of the diagnostic indications for hepatic replacement. The second most common indication is a conglomerate of diseases grouped as metabolic disorders, such as alpha1-antitrypsin deficiency, tyrosinemia, and hyperoxaluria, among others.5 Chronic liver injury, regardless of the etiology, predisposes to the development of hepatocellular carcinoma. Table 50-2 lists varying hepatic disorders that have been associated with the development of hepatocellular carcinoma. Thus, patients with chronic liver disease must be periodically screened with determinations of alphafetoprotein and radiologic studies for detection of liver cancer.
EVALUATION OF RECIPIENT Potential candidates for liver transplantation must complete a careful evaluation that includes consultations by a hepatologist, transplant surgeon, social worker, psychiatrist when needed, and a financial counselor. Important parameters for the initiation of the evaluation are the presence of end-stage liver disease, the understanding by the patient (or parents, when the candidate involves a minor) of the process of liver transplantation, including its potential financial repercussions. On completion of the evaluation, cases are discussed in a multidisciplinary meeting, and a consensus is reached in regard to a patient’s candidacy. Liver disease may be associated with comorbid conditions. For example, patients with alcoholic cirrhosis and hemochromatosis may also suffer from cardiomyopathy from alcohol injury and iron deposition, respectively. The risk for recidivism among alcoholic patients must be carefully studied. Patients with Alagille’s syndrome often have concomitant pulmonary hypertension and such patients need a comprehensive cardiac workup, including cardiac catheterization. Pulmonary arteriovenous shunting is a common complication of cirrhosis that also requires careful assessment. Patients with coexistent liver cancer must be studied to rule out extrahepatic involvement.
947
Section VIII. Liver Transplantation
Table 50-1. Indications for Hepatic Transplantation Acute liver injury • Viral hepatitis • Toxic injury (acetaminophen (paracetamol), halothane, mushroom, others) • Fulminant Wilson’s disease • Fulminant tyrosinemia Chronic liver injury Cholestatic disease (primary biliary cirrhosis, primary sclerosing cholangitis, biliary atresia, familial cholestatic syndromes) • Hepatocellular disease (viral hepatitis, alcoholic cirrhosis, autoimmune hepatitis) • Vascular disease (Budd–Chiari, veno-occlusive disease) • Massive steatosis
Table 50-2. Hepatic Disorders That May Predispose to the Development of Hepatocellular Carcinoma Cholestatic liver disease Primary biliary cirrhosis Alagille’s syndrome Primary sclerosing cholangitis Familial cholestatic cirrhosis (Byler’s disease) Secondary biliary cirrhosis Biliary atresia Parenteral nutrition
• • • • • • •
•
Mass-occupying lesions Hepatocellular carcinoma Hepatoblastoma Hemangioendothelioma Metastatic neuroendocrine tumor Polycystic liver disease Multiple adenomatosis
• • • • • •
Metabolic diseases Alpha1-antitrypsin deficiency Wilson’s disease Tyrosinemia Hemochromatosis Glycogen storage disease types I and IV Cystic fibrosis Erythropoietic protoporphyria Crigler–Najjar syndrome Oxalosis Urea cycle enzyme deficiency Protein C deficiency Hemophilia A
• • • • • • • • • • • •
Graft failure Rejection (acute, chronic) Primary graft failure Technical failure
• • •
General comorbid factors that require a careful investigation during the evaluation process are seizure disorder refractory to treatment, arteriosclerotic disease, renal dysfunction, labile diabetes mellitus, recent history of malignancy, and morbid obesity. The contraindications for liver transplantation are listed in Table 50-3. Liver transplantation for patients with coexistent human immunodeficiency virus (HIV) infection is undergoing clinical trials in a few transplant centers throughout the world.
SURGICAL ASPECTS The procedure for liver transplantation may be a formidable task; however, the ultimate outcome may depend more on proper patient selection than the procedure itself. Nevertheless, the surgical techniques for the procurement of the liver as well as its implantation may vary as the result of existing conditions at the time of the operation, such as the presence of vascular anomalies and the physiologic condition of the donor or recipient.
948
Chronic active hepatitis; cirrhosis Hepatitis B infection Hepatitis C infection Autoimmune hepatitis Cryptogenic cirrhosis Alcoholic cirrhosis
• • • • •
Metabolic disease Wilson’s disease Hemochromatosis Hereditary tyrosinemia Alpha1-antitrypsin deficiency Glycogen storage disease types I, III, and IV
• • • • •
Other Cystic fibrosis Congenital hepatic fibrosis Neonatal iron storage disease Fanconi’s anemia (androgen-treated)
• • • •
Table 50-3. Contraindications for Liver Transplantation Absolute Patient unable to understand and comply with immunosuppression Active extrabiliary sepsis Metastatic hepatocellular carcinoma Advanced cardiopulmonary disease Active alcoholism and drug addiction Acquired immunodeficiency syndrome (AIDS) Documented anatomical anomalies precluding transplantation Recent history of extrahepatic cancer
• • • • • • • •
Relative Age above 75 years Active sepsis of hepatobiliary origin Active infection of extrahepatic organs Retransplantation for recurrence of hepatitis C Cholangiocarcinoma Severe malnutrition Diabetes mellitus with coronary artery disease Multiple organ failure requiring cardiopulmonary support Patients in coma stage IV Epilepsy Human immunodeficiency virus (HIV) infection Morbid obesity
• • • • • • • • • • • •
DONOR PROCUREMENT The donor pool includes brain-dead donors, donors after cardiac death (DCD or non-heart-beating donors) and living donors. DCD donors have intact brainstem reflexes but have suffered severe irreversible neurological damage. DCD donors are procured between 2
Chapter 50 LIVER TRANSPLANTATION SURGERY
and 5 minutes after circulatory arrest and thus have longer warm ischemia time. Frequent causes of death among organ donors are cerebrovascular accidents, anoxia, trauma, and brain tumors (Table 50-4). Extended-criteria donors (ECD) are often utilized to address the ongoing problem of scarcity of organs.6–8 Proportionally, there seemed to be a higher percentage of brain tumors among DCD than brain-dead donors.9 After a suitable donor is identified, it is imperative that preoperative donor management is optimized. This is coordinated by the local organ procurement organization. A careful history is obtained from the next of kin as well as from the clinical chart, including review of previous admissions. Serologic analysis must be done for hepatitis B and C, HIV, human T lymphotropic virus-1 (HTLV-1), cytomegalovirus, Epstein–Barr virus, and syphilis. Optimal donor management begins once brain death has been determined: efforts are directed at protecting the organs and no longer the brain. This requires the administration of fluids, maintaining normothermia, adequate blood pressure (systolic blood pressure > 100 mmHg), and urine output (>30 ml/h). Care is taken not to fluid-overload the donor as this produces edema within the organ. Catecholamine deficiency as a result of brain death may necessitate the use of vasopressors such as dopamine once sufficient central venous pressure (>10 mmHg) is obtained. The development
of diabetes insipidus may require the use of vasopressin and hypotonic fluids. Hormonal replacement therapy with steroids and triiodothyronine are also instituted. Coagulopathy is corrected with fresh frozen plasma, cryoprecipitate, or platelets. Severe anemia is also treated with the transfusion of packed red blood cells. Table 50-5 lists several donor factors that can be implicated with poor graft function postoperatively. Livers may be preserved up to 24 hours; however, the incidence of biliary strictures and graft dysfunction increases exponentially after 12 hours of preservation. Short cold preservation times (<12 hours) correlate with improved early graft function and decreased incidence of primary graft non-function and a shorter intensive care unit hospital stay. Viaspan solution is the most commonly used cold preservation storage fluid for the liver. Other solutions that have been used in clinical and experimental transplantation and their characteristics are listed in Table 50-6.10–12 The procurement of the cadaveric organ often involves the coordination and collaborative effort of several separate teams: heart, lungs, abdominal viscera (liver, pancreas, kidneys, intestine), and tissue teams. There are two fairly standardized methods of surgical procurement: conventional versus en-bloc technique. Minor variations may
Table 50-5. Donor Factors Implicated in Poor Graft Function Table 50-4. Causes of Death Among Brain-Dead and Donor After Cardiac Death (DCD) Donors Reported by United Network for Organ Sharing (%)a Cause Anoxia Cerebrovascular accident Head trauma Central nervous system tumor Unspecified Total
Brain-Dead Year 2004
DCD
770 (13) 2501 (43) 2355 (41) 50 (1) 109 (2) 5785 (100)
52 (20) 43 (16) 81 (30) 84 (31) 8 (3) 268 (100)
Extremes of age (<3 months or >80 years) ABO incompatibility Prolonged hospital/intensive care unit stay High-dose pressors, hypotension Hypoxia Long cold preservation time (>24 hours) Long warm ischemia time (>1 hour) Hepatic steatosis (>30%) Incompletely resuscitated donor (sodium >160 mg/dl) Hepatic dysfunction (AST or ALT > 4 ¥ normal, alkaline phosphate >2 ¥ normal, PT >2 ¥ normal, bilirubin >4.0 mg/dl) Metabolic acidosis AST, aspartate aminotransferase; ALT, alanine aminotransferase; PT, prothrombin time.
a
www.optn.org.
Table 50-6. Preservation Solutions for the Liver Used in Clinical and Experimental Transplantation Solution
Proprietary name
Supplier/reference
Basic composition
Pricea
Viscosityb
Bretschneiders-HTK Marshall’s (HOC) Celsior University of Wisconsin (UW) UW-HES Belzer-MPS
Custodiol Soltran Celsior Viaspan
Köhler Chemie, Alsbach, Germany Baxter Healthcare, Thetford, UK Sangstat, CA Barr Laboratories, New York
Low Low Medium High
Low Low Low High
N/A KPS-1
High High
Medium Medium
PBS140 Perfadex
N/A Perfadex
Reference28 Organ Recovery Systems, Chicago, IL Reference29 Vitrolife AB, Gothenburg, Sweden
Histidine-tryptophan-ketoglutarate Hypertonic citrate High sodium lactobionate High potassium actobionate; starch colloid UW without starch colloid High-sodium gluconate Phosphate-buffered sucrose Low-potassium, dextran 40 colloid; THAM/PGE, additive
Low High (including cost of additives)
Low Medium
a
High (> £100/l); medium (£100 > x > £80); low ( < £80). High (> 2.5 cP at 20°C); medium (2.5 > x > 1.5); low ( < 1.5 cP). Reproduced from Wilson CH, Stansby G, Haswell M, et al. Evaluation of eight preservation solutions for endothelial in situ preservation. Transplantation 2004;78:1008–1013, by permission of Lippincott. b
949
Section VIII. Liver Transplantation
occur in these accepted techniques.13 The surgical technique of conventional procurement involves in-situ dissection of individual organs. This technique is often used when the donor is physiologically stable. The sequence for removal of the abdominal organs is as follows: first, the liver, followed by the pancreas, and, lastly, the kidneys. The liver is placed in a plastic bag containing the preservation solution. The organ is then placed into two additional bags (three total) to avoid contamination. Finally, the liver is placed in an ice chest and covered with ice. The liver is now ready to be shipped to the recipient’s hospital. The iliac veins and arteries are then procured and placed in the preservation solution should the need arise for alternative methods of reconstruction for the hepatic artery and/or portal vein. Final preparations for implantation of the liver graft are usually performed at the recipient hospital. The en-bloc technique involves procurement of all the abdominal viscera and separation ex situ on the back table. This technique is carried out when the donor is physiologically unstable. The liver is removed from the abdomen en bloc with the abdominal viscera. Minimal dissection is carried out prior to the perfusion of the organs. Separation of the liver from the other abdominal viscera (pancreas, intestine, kidneys) is completed on the back table (ex situ). A comparison of the in-situ and the en-bloc techniques is listed in Table 50-7. Whichever technique is used, it must not jeopardize any of the organs being procured.
SPLIT LIVER PROCUREMENT The arterial and portal vein blood supply to the liver follows a segmental distribution allowing partial liver or cadaveric segmental transplantation (Figure 50-1). A liver allograft may be used for two recipients and is described in the literature as split-liver transplantation. The most common split is to use an extended right lobe (right lobe plus a segment of the left lobe or segments IV, V, VII, and VIII) for the adult recipient and the left lateral segment (a portion of the left lobe or segments II and III) for a child (Figure 50-2). Infrequently the liver has been split into right (segments V, VI, VII, and VIII) and left (segments I, II, III, and IV) lobes providing the opportunity to do a transplant in two adult recipients.14–16 The partition of the liver may be done in situ or ex situ. In the latter, the liver is procured as a whole organ and then divided at the back table. There are several tools the surgeon may use to divide the liver and create hemostasis. Several biological products based on fibrin materials to coat the raw surface of the liver are often used for hemostasis.
Right lobe
Left lobe
II VIII
Length of time for procurement Risk of vascular injury Length of warm dissection Risk of vasospasm Risk of donor instability Length of cold dissection Length of delay in procurement of thoracic viscera
Conventional technique
En-bloc techniquea
Longer Higher Longer Higher Higher Shorter Longer
Shorter Lower Shorter Lower Lower Longer Shorter
a
Required technique for donors after cardiac death.
Vena cava
V
950
I
III
Left lateral segment
VI
Figure 50-1. Anatomic segments of the liver based on its blood supply. The segmental blood supply to the liver along with its capacity for regeneration allows surgeons to conduct partial live or deceased liver transplantation.
Figure 50-2. Split-liver transplantation. Hepatic artery and bile duct are not shown but the concept of where to divide them is the same as that of the portal vein.
Left hepatic vein
To an adult
To a child
Right portal vein
IV
VII
Table 50-7. Comparison of Conventional Versus En-Bloc Technique of Organ Procurement
Left portal vein
Chapter 50 LIVER TRANSPLANTATION SURGERY
Unlike Europe, the split-liver transplant has not been totally embraced in the USA.16–18 The reason is based on initial reports showing a higher incidence of primary graft non-function after split compared to full-size transplantation. In contrast, the European data showed similar results in terms of patient and graft survival between split- and whole-liver transplantation. Thus, efforts should be made to maximize the number of split-liver transplants in the USA to alleviate, at least in part, the existing problem of organ shortage.19
LIVING-DONOR PROCUREMENT In live adult donor to an adult recipient, it is necessary to use the right lobe, which accounts for about 60–70% of the total liver volume.20–23 From an adult donor to a child, it is the left lateral segment (a portion of the left lobe) which accounts for approximately 20% of the liver volume. The true left lobe may be used in a small adult recipient or a larger child.24–26 The key issue is that the donor must be left with enough liver parenchyma to sustain life. This has been estimated to be at least 30% of the total liver volume. Likewise, the recipient needs enough liver mass, and a minimum of 40% of the estimated total liver volume is recommended.24 Another formula is to use the lean body weight of the recipient. The hepatic lobe must be at least 0.8% of the recipient’s lean body weight. For example, a 70-kg recipient will require at least 560 grams of liver parenchyma.25 In our institution we use a minimum of 1% to increase the margin of safety and to compensate for the potential risk of preservation-induced injury to the liver. The technique to divide the liver is similar to the technique described for the in-situ splitting of the allograft. The blood supply to the lobe (or segment) to be used for transplantation must be maintained to avoid anoxic injury to the liver (Figure 50-3). To minimize the blood loss, a transient Pringle maneuver, controlling the inflow to the liver, has recently been reported.27,28 The vascular structures in the hepatic hilum are intermittently cross-clamped for periods of 10–15 minutes, allowing flow through the liver for periods of 5 minutes in between the cross-clampings. This maneuver is helpful in minimizing the blood loss during the division of the
Right lobe
liver parenchyma. The goal is to avoid exposing the donor to transfusion with banked blood with the potential risk of transmission of viruses and transfusion reactions. Furthermore, living donors are required to donate 1 or 2 units of autologous blood prior to the operation. At the time of the operation, isovolemic hemodilution is also carried out, whereby another unit is removed from the donor and replaced with saline.29 If bleeding is significant enough to require blood transfusions, the blood drawn in the operating room is given first, followed by the autologous blood. Banked blood will be used in extenuating circumstances. The cell-saver is always used during the live donor hepatectomy. The advantages of living donation are having a physiologically stable and healthy donor, short preservation time, elective surgery, and the potential psychological benefit of helping a patient, who is usually a close relative. The disadvantages are the surgical and anesthetic risks to which the donors are exposed. This is problematic since the living donor is a healthy individual who is in no need of having major surgery.30 It is unlikely that surgeons would be doing this type of surgery if there were enough organs for transplantation. The morbidity among living donors is significant, with rates up to 65%.31–37 Table 50-8 lists major and minor complications among living donors reported in the literature.
Table 50-8. Reported Complications Among Living Liver Donors Death Liver failure requiring urgent transplantation Pulmonary embolism Perforated peptic ulcer Splenic injury requiring splenectomy Biliary strictures Portal vein thrombosis Arm paralysis Incisional hernia Wound infection Severe hypophosphatemia
Remaining left lobe
Figure 50-3. The sequence of steps following a live donor right lobectomy is as follows: division of the liver parenchyma, transection of the right hepatic duct, division of the right hepatic artery, division of the right portal vein and, finally, division of the right hepatic vein. The right lobe is flushed, ex situ, with preservation solution via the portal vein.
951
Section VIII. Liver Transplantation
Figure 50-4. Incisions used for adult patients with a wide (A) or narrow (B) chest. A bilateral subcostal incision (C) is often enough for small children.
A
B
C
Table 50-9. Physiologic Changes According to Technique of Transplantation
1. No venous bypass 2. Venous bypass 3. Piggyback
Systemic venous return
Portal vein pressure
Renal vein pressure
Impaired
≠
≠
Slightly impaired Undisturbed
Ø ≠
Normal Normal
Diaphragm
Liver
RECIPIENT OPERATION The following description relates to the situation when the liver is placed in an orthotopic location. Three different techniques have been described for liver transplantation: (1) removal of the liver, including the retrohepatic cava on venous bypass; (2) removal of the liver, including the retrohepatic cava without venous bypass; and (3) removal of the liver without the vena cava and placing the liver in a piggyback position.38,39 The piggyback liver transplants may be done with or without venous bypass. The type of incision depends on the habitus of the patient. Incisions often used in liver transplantation are shown in Figure 50-4. The physiologic changes during the anhepatic state associated with each of these techniques are shown in Table 50-9. The decision as to which type of technique may be used depends on the physiologic condition of the patient at the time of the transplant. In hemodynamically stable patients who tolerate the cross-clamping of the vena cava, the hepatectomy and implantation of the liver can be done expeditiously since the removal of the liver requires less dissection than that of the piggyback transplant and also avoids the time needed for the initiation of the venous bypass (Figure 50-5). The disadvantage with this technique is that bleeding from the raw surfaces may be exacerbated because of the venous hypertension brought about by the crossclamping of the portal vein and vena cava. Prolonged crossclamping of the vena cava may result in postoperative renal dysfunction due to renal vein hypertension.40 The utilization of venous bypass may lengthen the operation, but the potential benefits of the bypass, such as decreased bleeding and protection of the renal function by avoiding venous hypertension, outweigh the extra time needed for setting up the equipment as well as the insertion of the cannulae. The insertion of the cannulae may be done through a venous cutdown or percutaneously (Figure 50-
952
Vena cava
Figure 50-5. View of a conventional hepatectomy with removal of the retrohepatic vena cava. The systemic venous return is impaired.
6). The potential complications of venous bypass are: embolism, hematoma formation at the site of cannulae insertion, lymphoceles, injury to brachial plexus in the axilla or femoral vessels or vein in the groin, and fibrinolysis.41,42 The piggyback transplantation is a more laborious procedure since the liver needs to be meticulously separated from the vena cava. There may be several tributaries from the liver to the vena cava and their dissection makes the procedure tedious. Once the liver is separated from the vena cava, a vascular clamp is placed across the confluence of the hepatic veins and the hepatectomy is completed (Figure 50-7). Since the systemic venous return is maintained intact, the hemodynamics change little compared to those procedures that require the cross-clamping of the vena cava. We have observed decreased blood transfusion requirements with the piggyback transplant compared to the conventional techniques.43 The potential complication is increased ascites during the perioperative period which may be the result of functional outlet obstruction.44 However,
Chapter 50 LIVER TRANSPLANTATION SURGERY
A
Subclavian vein
F
Inflow cannula
F D
B E
Portal vein cannula
Portal vein
C
F Inferior vena cava
Figure 50-8. Completion of a liver transplant. The anastomosis of the suprahepatic vena cava (A) is done first, followed by the anastomosis of the infrahepatic vena cava (B), portal vein (C), hepatic artery (D) and, lastly, common bile duct (E). (F) represents drains strategically placed near the anastomosis. Femoral vein cannula
Figure 50-6. Schematic drawing depicting the venous bypass during the anhepatic phase in liver transplantation. The arrows show the direction of blood flow.
Diaphragm Vena cava Confluence of hepatic veins
allograft vena cava and the recipient hepatic veins and one for the portal vein) are required. The diagnosis of portal vein thrombosis is not a contraindication to transplantation.45 This is dealt with at the time of transplantation by performing a thrombectomy or alternatively by using a “jump” graft from the recipient’s superior mesenteric vein to the portal vein of the allograft. Finally, the arterial reconstruction completes the revascularization of the liver. Arterial anomalies are often encountered and their reconstruction may need more innovative methods for their reconstruction.
BILIARY TRACT RECONSTRUCTION
Liver
Tributary veins Figure 50-7. The separation of the liver from the vena cava in piggyback transplantation. The clamp was placed across the confluence of the hepatic veins. The systemic venous return remains undisturbed.
this problem seems to be temporary. The piggyback technique may also be utilized during retransplantation.
VASCULAR RECONSTRUCTION In the conventional technique, a total of three venous anastomoses (two for the vena cava and one for the portal vein) are necessary. In piggyback transplantation only two anastomoses (one between the
Two methods of bile duct reconstruction may be used: (1) an endto-end anastomosis of the donor common bile duct to the recipient common bile duct with or without a T-tube or (2) end-to-side choledochojejunostomy to a Roux en Y with or without a stent. Fine absorbable sutures are utilized for these anastomoses. A choledochojejunostomy is done in patients whose bile duct is diseased or absent, such as in cases of primary sclerosing cholangitis and biliary atresia, respectively, as well as in those cases where a size mismatch or not enough length exists between the donor and recipient’s common bile ducts. Before the closure of the abdomen, two or three drains are placed in the right and left subhepatic spaces and adjacent to the bile duct reconstruction (Figure 50-8).
RESULTS Based on United Network for Organ Sharing data, the overall 1-, 3-, and 5-year patient actuarial survival is 87.6, 79.9, and 74.6%, respectively (Table 50-10). The actuarial graft survival at 1, 3, and 5 years is 82.5, 73.5, and 67.3%, respectively, as shown in Table 50-11.9 Gender of the recipient does not seem to influence patient or graft survival, as demonstrated in Tables 50-12 and 50-13.9
953
Section VIII. Liver Transplantation
Table 50-10. Organ Procurement and Transplantation Network (OPTN) Liver Kaplan–Meier Patient Survival Rates for Transplants Performed 1995–2002 Region USA USA USA USA USA USA
Transplant type
Years post-transplant
Number functioning/ alive
Survival rate
95% Confidence interval
Primary transplant Repeat transplant Primary transplant Repeat transplant Primary transplant Repeat transplant
1 year 1 year 3 years 3 years 5 years 5 years
11 362 875 11 409 939 8871 734
87.6 70.0 79.9 60.2 74.6 53.0
(87.0, 88.2) (67.5, 72.5) (79.3, 80.5) (57.9, 62.6) (73.8, 75.3) (50.6, 55.5)
Data subject to change based on future data submission or correction. 1-year survival based on 2000–2002 transplants, 3-year survival based on 1997–2000 transplants, 5-year survival based on 1995–1998 transplants. Patient survival was not computed due to n less than 10. Based on OPTN data as of March 4, 2005.
Table 50-11. Organ Procurement and Transplantation Network (OPTN) Liver Kaplan–Meier Graft Survival Rates for Transplants Performed 1995–2002 Region USA USA USA USA USA USA
Transplant type
Years post-transplant
Number functioning/ alive
Survival rate
95% Confidence interval
Primary transplant Repeat transplant Primary transplant Repeat transplant Primary transplant Repeat transplant
1 year 1 year 3 years 3 years 5 years 5 years
11 362 875 11 409 939 8871 734
82.5 62.8 73.5 51.5 67.3 42.9
(81.8, 83.1) (60.2, 65.3) (72.8, 74.2) (49.2, 53.7) (66.6, 68.1) (40.6, 45.2)
Data subject to change based on future data submission or correction. Graft survival was not computed due to n less than 10. 1-year survival based on 2000–2002 transplants; 3-year survival based on 1997–2000 transplants; 5-year survival based on 1995–1998 transplants. Based on OPTN data as of March 4, 2005.
Table 50-12. Organ Procurement and Transplantation Network (OPTN) Liver Kaplan–Meier Patient Survival Rates for Transplants Performed 1995–2002 Region USA USA USA USA USA USA
Recipient gender
Years post-transplant
Male Female Male Female Male Female
1 year 1 year 3 years 3 years 5 years 5 years
Number functioning/ alive 7717 4520 7374 4974 5498 4107
Survival rate
95% Confidence interval
86.4 85.4 77.9 78.0 71.4 73.4
(85.7, 87.1) (84.5, 86.4) (77.1, 78.7) (77.0, 79.0) (70.5, 72.4) (72.3, 74.6)
Data subject to change based on future data submission or correction. Patient survival was not computed due to n less than 10. 1-year survival based on 2000–2002 transplants; 3-year survival based on 1997–2000 transplants; 5-year survival based on 1995–1998 transplants. Based on OPTN data as of March 4, 2005.
Table 50-13. Organ Procurement and Transplantation Network (OPTN) Liver Kaplan–Meier Graft Survival Rates for Transplants Performed 1995–2002 Region USA USA USA USA USA USA
Recipient gender
Years post-transplant
Male Female Male Female Male Female
1 year 1 year 3 years 3 years 5 years 5 years
Number functioning/ alive 7717 4520 7374 4974 5498 4107
Survival rate
95% Confidence interval
81.1 79.9 71.3 70.9 63.7 65.8
(80.3, 81.9) (78.8, 81.0) (70.5, 72.2) (69.9, 72.0) (62.7, 64.6) (64.7, 67.0)
Data subject to change based on future data submission or correction. Graft survival was not computed due to n less than 10. 1-year survival based on 2000–2002 transplants; 3-year survival based on 1997–2000 transplants; 5-year survival based on 1995–1998 transplants. Based on OPTN data as of March 4, 2005.
954
Chapter 50 LIVER TRANSPLANTATION SURGERY
SURGICAL COMPLICATIONS Major surgical complications associated with the liver transplant operation fall into broad categories and these are: graft dysfunction, biliary tract, and vascular complications.
Liver Transplant Patients with Hepatocellular Carcinoma Patient Survival 100 90 Probability
80 70 Survival (%)
Similarly, the type of donor (living versus deceased) bears no influence on the outcomes, as shown in Tables 50-14 and 50-15. In contrast, the analysis of age at the time of transplantation clearly shows that patient survival at 3 and 5 years is lower in the under-1-year of age and over 65 years of age groups, compared to the other groups (Table 50-16). The difference is more pronounced in the analyses of graft survival (Table 50-17). Tables 50-10 and 50-11 demonstrate that patient and graft survival is compromised by repeat transplantation. Other factors that may have an impact on patient and graft survival are the condition of the patient at the time of transplantation and the diagnostic indication for transplantation. Hepatocellular carcinoma is associated with the lowest patient survival (31.8% at 5 years) based on a retrospective analysis of 4000 cases at the University of Pittsburgh.5 However, the survival for this indication has improved greatly by following strict criteria for selection of candidates for transplantation. In the USA, patients with hepatocellular carcinoma considered suitable candidates for transplantation are those with a single lesion measuring less than 5 cm in diameter or with three or fewer lesions measuring no more than 5 cm in the aggregate. Following such criteria, the 5-year patient actuarial survival at Stanford University is 80% (Figure 50-9). In a comparison of the outcomes after hepatic transplantation for hepatitis C and B and alcoholic cirrhosis at Stanford University, a trend for lower patient and graft survival in the hepatitis C cohort was observed, but the differences did not reach statistical significance (Figure 50-10).
60 50 40 30 20 10 0 0
1
2
3
4
5
6
7
8
Years Post Transplant Figure 50-9. Actuarial patient survival at Stanford University, California, for patients who underwent hepatic replacement for hepatocellular carcinoma. No attrition was observed past 3 years after transplantation.
Table 50-14. Organ Procurement and Transplantation Network (OPTN) Liver Kaplan–Meier Patient Survival Rates for Transplants Performed 1995–2002 Region USA USA USA USA USA USA
Donor type
Years post-transplant
Number functioning/ alive
Survival rate
95% Confidence interval
Cadaveric Living Cadaveric Living Cadaveric Living
1 year 1 year 3 years 3 years 5 years 5 years
11 240 997 11 797 551 9433 172
85.8 89.1 77.8 80.2 72.1 80.0
(85.2, 86.3) (87.2, 90.9) (77.2, 78.4) (77.2, 83.2) (71.4, 72.9) (74.8, 85.1)
Data subject to change based on future data submission or correction. Patient survival was not computed due to n less than 10. 1-year survival based on 2000–2002 transplants; 3-year survival based on 1997–2000 transplants; 5-year survival based on 1995–1998 transplants. Based on OPTN data as of March 4, 2005.
Table 50-15. Organ Procurement and Transplantation Network (OPTN) Liver Kaplan–Meier Graft Survival Rates for Transplants Performed 1995–2002 Region USA USA USA USA USA USA
Donor type
Years post-transplant
Number functioning/ alive
Survival rate
95% Confidence interval
Cadaveric Living Cadaveric Living Cadaveric Living
1 year 1 year 3 years 3 years 5 years 5 years
11 240 997 11 797 551 9433 172
80.7 79.5 71.2 70.2 64.4 70.8
(80.1, 81.4) (77.2, 81.8) (70.5, 71.9) (67.0, 73.5) (63.7, 65.2) (65.3, 76.3)
Data subject to change based on future data submission or correction. Graft survival was not computed due to n less than 10. 1-year survival based on 2000–2002 transplants; 3-year survival based on 1997–2000 transplants; 5-year survival based on 1995–1998 transplants. Based on OPTN data as of March 4, 2005.
955
Section VIII. Liver Transplantation
Table 50-16. Organ Procurement and Transplantation Network (OPTN) Liver Kaplan–Meier Patient Survival Rates for Transplants Performed 1995–2002 Region USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA
Recipient age
Years post-transplant
Number functioning/ alive
Survival rate
95% Confidence interval
< 1 year 1–5 years 6–10 years 11–17 years 18–34 years 35–49 years 50–64 years 65 + < 1 year 1–5 years 6–10 years 11–17 years 18–34 years 35–49 years 50–64 years 65 + < 1 year 1–5 years 6–10 years 11–17 years 18–34 years 35–49 years 50–64 years 65 +
1 year 1 year 1 year 1 year 1 year 1 year 1 year 1 year 3 years 3 years 3 years 3 years 3 years 3 years 3 years 3 years 5 years 5 years 5 years 5 years 5 years 5 years 5 years 5 years
367 459 170 345 735 3991 5377 791 434 511 253 317 820 4450 4754 805 342 451 218 253 716 3546 3476 601
85.2 83.3 90.5 91.7 87.6 87.7 85.3 80.5 76.9 78.2 85.9 83.3 81.4 79.8 76.5 69.5 74.5 77.4 83.8 75.8 76.1 74.3 69.6 62.8
(81.7, 88.6) (80.1, 86.5) (85.9, 95.2) (88.7, 94.7) (85.4, 89.9) (86.7, 88.6) (84.4, 86.2) (78.0, 83.0) (73.5, 80.3) (75.1, 81.4) (81.8, 90.0) (79.6, 87.1) (79.0, 83.7) (78.8, 80.8) (75.5, 77.5) (66.8, 72.1) (70.7, 78.3) (74.2, 80.6) (79.3, 88.3) (71.5, 80.2) (73.4, 78.7) (73.1, 75.5) (68.4, 70.8) (59.9, 65.8)
Data subject to change based on future data submission or correction. Patient survival was not computed due to n less than 10. 1-year survival based on 2000–2002 transplants; 3-year survival based on 1997–2000 transplants; 5-year survival based on 1995–1998 transplants. Based on OPTN data as of March 4, 2005.
Table 50-17. Organ Procurement and Transplantation Network (OPTN) Liver Kaplan–Meier Graft Survival Rates for Transplants Performed 1995–2002 Region USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA USA
Recipient age
Years post-transplant
Number functioning/ alive
Survival rate
95% Confidence interval
< 1 year 1–5 years 6–10 years 11–17 years 18–34 years 35–49 years 50–64 years 65+ < 1 year 1–5 years 6–10 years 11–17 years 18–34 years 35–49 years 50–64 years 65+ < 1 year 1–5 years 6–10 years 11–17 years 18–34 years 35–49 years 50–64 years 65+
1 year 1 year 1 year 1 year 1 year 1 year 1 year 1 year 3 years 3 years 3 years 3 years 3 years 3 years 3 years 3 years 5 years 5 years 5 years 5 years 5 years 5 years 5 years 5 years
367 459 170 345 735 3991 5377 791 434 511 253 317 820 4450 4754 805 342 451 218 253 716 3546 3476 601
75.4 75.2 81.7 84.7 80.8 82.2 80.8 76.5 64.9 70.0 76.6 71.1 71.0 72.2 71.6 66.5 62.3 66.2 73.8 63.9 64.2 65.7 63.9 59.7
(71.5, 79.4) (71.6, 78.7) (76.1, 87.3) (81.0, 88.4) (78.1, 83.4) (81.1, 83.2) (79.8, 81.7) (73.9, 79.2) (61.3, 68.5) (66.7, 73.4) (71.9, 81.3) (66.8, 75.3) (68.4, 73.6) (71.1, 73.3) (70.5, 72.6) (63.8, 69.1) (58.4, 66.3) (62.8, 69.6) (68.8, 78.8) (59.4, 68.5) (61.5, 67.0) (64.5, 66.9) (62.7, 65.2) (56.7, 62.6)
Data subject to change based on future data submission or correction. Graft survival was not computed due to n less than 10. 1-year survival based on 2000–2002 transplants; 3-year survival based on 1997–2000 transplants; 5-year survival based on 1995–1998 transplants. Based on OPTN data as of March 4, 2005.
956
Chapter 50 LIVER TRANSPLANTATION SURGERY
metabolic acidosis, hyperkalemia, and the onset of the adult respiratory distress syndrome. The treatment is urgent retransplantation. These patients deteriorate quickly, thus, the outcomes depend on the availability of an organ for a timely transplant. The reported 1-year actuarial patient survival following retransplantation for primary graft non-function ranges from 0 to 50%.49,50
Liver Transplant Patients with Hep B, Hep C, or Alcoholic Cirrhosis Patient Survival 100 90 80 Survival (%)
70 60
BILIARY TRACT COMPLICATIONS
50
The most common surgical complications observed in liver transplant recipients are biliary tract problems. In whole-organ liver transplantation an incidence from 6% to close to 20% has been reported.51,52 Among these, strictures account for two-thirds and bile leaks for the remaining third. Biliary strictures can occur at the level of the anastomosis with the cause likely being ischemia of the distal allograft common bile duct. Although some of these patients may respond to dilatation of the stricture and stent placement, this type of injury will require surgical excision of the stricture and reconstruction to a Roux en Y. Intrahepatic strictures are the result of an ischemic injury to the bile ducts which could be brought about by a prolonged preservation time (>12 hours), stenosis, or thrombosis of the hepatic artery.52 Clinical symptoms and radiological findings resemble those of primary sclerosing cholangitis. The patients may experience right upper quadrant pain, pruritus, and transient episodes of jaundice accompanied by abnormal elevation of the alkaline phosphatase and gammaglutamyl transpeptidase. Management of these patients is difficult, often requiring multiple interventions for percutaneous dilation of the strictures and placement of indwelling catheters for drainage of the biliary tree. Ultimately, these patients do end up needing retransplantation.53 Biliary leaks may originate from the anastomosis of the bile duct or from the raw surface of the liver in cases where split or live donor lobes are used.54 An anastomotic leak should raise the suspicion of hepatic artery thrombosis. Hepatic artery thrombosis results in necrosis of the extrahepatic biliary system leading to sepsis. Such a complication often requires an urgent retransplantation.55 On the other hand, leaks from the raw surface may be managed by drainage and observation. Reoperation for direct control is done for persistent leaks or when the leak results in the formation of a subphrenic abscess. In live donor liver transplantation, particularly among recipients of the right lobe, the rate of biliary tract complication has been as high as 49%. Causes implicated for this high complication rate are multiple ducts, ischemia, thermal injury to the ducts during the transection of the parenchyma, and leaks from the raw surface.54
40 30 20 10
Hep B
Hep C
Alcoholic Cirrhosis
0 0
1
A
2
3
4
5
Years Post Transplant
Liver Transplant Patients with Hep B, Hep C, or Alcoholic Cirrhosis Graft Survival 100 90 80 Survival (%)
70 60 50 40 30 20 10
Hep B
Hep C
Alcoholic Cirrhosis
0 0 B
1
2
3
4
5
Years Post Transplant
Figure 50-10. Comparison of patient survival (A) and graft survival (B) after transplantation for hepatitis B and C and alcoholic cirrhosis. The differences observed did not reach statistical significance.
GRAFT DYSFUNCTION Following transplantation, there is often some degree of graft dysfunction. The clinical presentation of graft dysfunction can be mild, characterized by a transient elevation of the international normalized ratio (INR) accompanied by abnormal levels of the liver enzymes, or severe, requiring urgent retransplantation. The latter is known as primary graft non-function and this complication has been observed in between 2 and 10% of the transplants. The incidence of graft failure increases with the use of non-heart-beating donors and donors with significant fatty infiltration.46–48 Between these two extremes, differing degrees of graft dysfunction may be observed. Suffice it to say that the more severe the graft dysfunction, the greater the potential for morbidity, particularly infections, which in turn lead to prolonged intensive care unit and hospital stays, driving up the costs of transplantation. The clinical manifestations of primary graft non-function are encephalopathy leading to coma, uncorrectable INR in spite of massive transfusions of fresh frozen plasma and cryoprecipitate,
VASCULAR COMPLICATIONS Vascular complications are associated with significant morbidity and mortality. The rate of hepatic thrombosis varies from 2% to 10% over the years.56–59 This complication rate has decreased because of improvements in the surgical technique and better solutions for organ preservation. The clinical presentation of hepatic artery thrombosis is variable. The majority of the patients develop necrosis of the extrahepatic biliary system, liver infarcts, and bilomas. Very few patients may not experience these problems but go on to develop intrahepatic biliary strictures. The definitive treatment for hepatic artery thrombosis is retransplantation. The 1-year actuarial survival following retransplantation for hepatic artery thrombosis is
957
Section VIII. Liver Transplantation
about 50%.50 Stenosis of the hepatic artery may result in the development of intrahepatic biliary strictures. Dilatation of the artery stenosis with stent placement has occasionally been successful.60 The ultrasonographic findings of dampening of the intrahepatic arterial wave or proximal dilation accompanied by turbulence suggest hepatic artery stenosis.61 In whole-liver transplantation, the incidence of portal vein thrombosis is about 2%.62,63 Portal vein thrombosis may occur immediately after transplantation or late. Factors implicated in thrombosis of the portal vein are: pretransplant thrombosis of the portal vein, congenital anomalies of the portal vein such as a preduodenal portal vein, or hypo- or aplasia of the portal vein, commonly seen in cases of extrahepatic biliary atresia.62,63 Segmental liver transplantation has been reported to be associated with a higher incidence of portal vein thrombosis.59 Clinical manifestations of portal vein thrombosis depend on the time of presentation. Early in the postoperative period the most common findings associated with portal vein thrombosis are a progressive increase of the INR value, an elevated serum ammonia, and gastrointestinal bleeding. The diagnosis is made by Doppler ultrasonography. Further studies such as a computed tomography or magnetic resonance imaging angiogram are unnecessary and further delay treatment. Unlike hepatic artery thrombosis, prompt exploration for thrombectomy and reconstruction of the portal veins anastomosis are frequently successful in salvaging the graft. Thrombotic complications of the vena cava in full-size liver transplantation are observed in fewer than 1% of the transplants.62,63 Among the complications of the vena cava reconstruction, outflow obstruction is the most common complication, resulting in symptoms similar to Budd–Chiari syndrome, such as intractable ascites, right pleural effusion, and abnormal levels of the hepatic enzymes. Recipients of the right lobe from living donors are at risk of developing outflow obstruction.64,65 Including the middle hepatic vein in addition to the right hepatic vein for drainage of living donor right lobes has eliminated this complication.66 Besides technical mishaps, hypercoagulable states in patients with malignancies, including myeloproliferative disorders and congenital anomalies, are other factors implicated as causes for thrombosis. Many of the patients with stenosis of the vena cava or vein in cases of living donor or split transplantation are successfully managed by transluminal dilatation of the stricture followed by stent placement.67,68 Once thrombosis ensues, the management becomes more difficult. Treatment with chronic anticoagulation may be indicated. Patients eventually develop collateral drainage through the azygos system with subsequent improvement of their symptoms. Retransplantation is rarely indicated.
RETRANSPLANTATION The rate of retransplantation varies from 10 to 30%.50,69–71 In decreasing order of frequency, the indications for retransplantation in children are: vascular complications with hepatic artery thrombosis being the most common, followed by primary graft nonfunction and, lastly, chronic rejection.71 Among adult patients, the indications in decreasing order of frequency are: primary graft nonfunction, recurrence of the underlying liver disease (particularly
958
Table 50-18. Causes for Retransplantation of the Liver and Patient Survival at the University of Pittsburgh Causes Primary non-function Hepatic artery thrombosis Chronic rejection Recurrent disease Acute rejection Biliary complications Technical Miscellaneous Unknown Total
Number (%)
Mean interval
Currently alive n (%)
249 (32.2) 214 (27.6) 113 (14.5) 44 (5.5) 38 (4.9) 22 (2.8) 12 (1.5) 48 (6.2) 34 (4.3) 774 (19.4)
0.3 + 0.4 5.5 + 14 25.1 + 28.3 31.9 + 32.2 5.4 + 20.4 16.8 + 24 34.9 + 35.2 16.4 + 28 28 + 47 10.7 + 24
97 (38.8) 85 (39.7) 48 (42.4) 15 (34.8) 13 (34.2) 8 (36.3) 6 (50) 12 (25) 14 (40) 298 (38.5)
Reproduced from Kashyap R, Jain A, Reyes J, et al. Causes of retransplantation after primary liver transplantation in 4000 consecutive patients: 2 to 19 years follow-up. Transplant Proc 2001;33:1486–1487, with permission.
Table 50-19. Indications for Use of Vascular Conduits During Primary or Secondary Liver Transplantation Aberrant vasculature in donor or recipient Injury to donor vessel during procurement Hepatic artery thrombosis Inadequate length of vessel for transplant, e.g. reduced size, split livers, thrombosed recipient celiac artery, large size discrepancy (adult donor vessel in pediatric recipient) Portal vein thrombosis Hypoplastic portal vein, e.g. biliary atresia
hepatitis C), chronic rejection and, lastly, vascular complications72 (Table 50-18). The outcomes after retransplantation are significantly worse than those of primary transplantation.70,71 Factors implicated in poor outcomes after retransplantation are the diagnostic indication, abnormal serum creatinine, prolonged INR, prolonged intensive care unit stay, and urgency for retransplantation.70–74 A difficult dilemma involves the decision-making for retransplantation for recurrence of hepatitis C following transplantation.75 The patient survival at 1 year following retransplantation for recurrence of hepatitis C varies from 33 to 50% compared to 65–75% for indications other than recurrence of hepatitis C.76,77 Interferon treatment for recurrence of hepatitis C after liver transplantation has been grossly ineffective.78 Further, the efficacy of pre-emptive treatment with interferon prior to retransplantation for recurrence of hepatitis C is unknown. The technique for retransplantation is roughly similar to primary transplantation; however, the degree of difficulty during retransplantation depends on the timing of the operation in relation to the primary transplant. For example, the operation is simple when retransplantation is carried out within a few days of the primary transplant, but the formation of dense adhesions may render the operation a formidable task should retransplantation be done several months or years after the first operation. In retransplantation, vascular conduits may be needed for reconstruction of the hepatic artery and portal vein (Table 50-19). A ductto-duct bile duct reconstruction may be done when the recipient’s
Chapter 50 LIVER TRANSPLANTATION SURGERY
bile duct appears healthy, but often this is not the case. Thus, a reconstruction to a Roux en Y will be the safest approach.
CONCLUSIONS During the last two decades, the techniques for liver transplantation have been refined and to a greater extent the outcomes following hepatic transplantation are related to the proper selection of donors and recipients. Nonetheless, the transplant surgeon must be familiar with the different surgical techniques to address adverse conditions at the time of the transplantation or retransplantation, such as the presence of portal vein thrombosis, arterial anomalies, and the presence of adhesions. Finally, the transplant surgeon must be aware of the ethical issues surrounding the utilization of living liver donors and must protect the potential living donors from coercion by the potential recipients and their relatives.
REFERENCES 1. Starzl TE. The saga of liver replacement, with particular reference to the reciprocal influence of liver and kidney transplantation (1955–1967). J Am Coll Surg 2002; 195:587–610. 2. Calne RY. Immunosuppression in liver transplantation. N Engl J Med 1994; 331:1154–1155. 3. Millard CE. The NIH Consensus Development Conference on liver transplantation. R I Med J 1984; 67:69–71. 4. Southard JH, Belzer FO. Organ preservation. Annu Rev Med 1995; 46:235–247. 5. Jain A, Reyes J, Kashyap R, et al. Long-term survival after liver transplantation in 4000 consecutive patients at a single center. Ann Surg 2000; 232:490–500. 6. Manzarbeitia CY, Ortiz JA, Rothstein KD, et al. Long-term outcome of controlled, non-heart-beating donor liver transplantation. Transplantation 2004; 78:211–215. 7. D’Alessandro AM, Fernandez LA, Chin LT, et al. Donation after cardiac death: the University of Wisconsin experience. Ann Transplant 2004; 9:68–71. 8. Busuttil RW, Tanaka K. The utility of marginal donors in liver transplantation. Liver Transpl 2003; 9:651–663. 9. http://www.optn.org/latestData/rptData.asp. 10. Janssen H, Janssen PH, Broelsch CE. Celsior solution compared with University of Wisconsin solution (UW) and histidinetryptophan-ketoglutarate solution (HTK) in the protection of human hepatocytes against ischemia-reperfusion injury. Transplant Int 2003; 16:515–522. 11. Pedotti P, Cardillo M, Rigotti P, et al. A comparative prospective study of two available solutions for kidney and liver preservation. Transplantation 2004; 77:1540–1545. 12. Wilson CH, Stansby G, Haswell M, et al. Evaluation of eight preservation solutions for endothelial in situ preservation. Transplantation 2004; 78:1008–1013. 13. Miller CM, Rapaport FT, Starzl TE. Organ procurement. In: Wilmore DW, Cheung LY, Harken AH, et al, eds. ACS surgery – principles and practice 2003. New York: WebMD, 2003:995–1007. 14. Kim JS, Broering DC, Tustas RY, et al. Split liver transplantation: past, present and future. Pediatr Transplant 2004; 8:644–648. 15. Merion RM, Rush SH, Dykstra DM, et al. Predicted lifetimes for adult and pediatric split liver versus adult whole liver transplant recipients. Am J Transplant 2004; 4:1792–1797. 16. Humar A, Khwaja K, Sielaff TD, et al. Split-liver transplants for two adult recipients: technique of preservation of the vena cava with the right lobe graft. Liver Transpl 2004; 10:153–155.
17. Adam R, McMaster P, O’Grady JG, et al. European Liver Transplant Association. Evolution of liver transplantation in Europe: report of the European Liver Transplant Registry. Liver Transpl 2003; 9:1231–1243. 18. Renz JF, Emond JC, Yersiz H, et al. Split-liver transplantation in the United States: outcomes of a national survey. Ann Surg 2004; 239:172–181. 19. Yersiz H, Renz JF, Farmer DG, et al. One hundred in situ splitliver transplantations: a single center experience. Ann Surg 2003; 238:496–505. 20. Marcos A, Fisher RA, Ham JM, et al. Right lobe living donor liver transplantation. Transplantation 1999; 68:798–803. 21. Bak T, Wachs M, Trotter J, et al. Adult-to-adult living donor liver transplantation using right-lobe grafts: results and lessons learned from a single-center experience. Liver Transpl 2001; 7:680–686. 22. Miller CM, Gondolesi GE, Florman S, et al. One hundred nine living donor liver transplants in adults and children: a singlecenter experience. Ann Surg 2001; 234:301–311. 23. Trotter JF, Wachs M, Everson GT, et al. Adult-to-adult transplantation of the right hepatic lobe from a living donor. N Engl J Med 2002; 346:1074–1082. 24. Fan ST, Lo CM, Liu CL, et al. Safety of donors in liver donor live transplantation using right lobe grafts. Arch Surg 2000; 135:336–340. 25. Sakamoto S, Uemoto S, Uryuhara K, et al. Graft size assessment and analysis of donors for living donor liver transplantation using right lobe. Transplantation 2001; 71:1407–1413. 26. Ben-Haim M, Emre S, Fishbein TM, et al. Critical graft size in adult-to-adult living donor liver transplantation: impact of the recipient’s disease. Liver Transpl 2001; 7:948–953. 27. Imamura H, Takayama T, Sugawara Y, et al. Pringle’s manoeuvre in living donors. Lancet 2002; 360:2049–2050. 28. Miller CM, Masetti M, Cautero N, et al. Intermittent inflow occlusion in living liver donors: impact on safety and remnant function. Liver Transpl 2004; 10:244–247. 29. Lutz JT, Valentin-Gamazo C, Gorlinger K, et al. Blood-transfusion requirements and blood salvage in donors undergoing right hepatectomy for living related liver transplantation. Anesth Analg 2003; 96:351–355. 30. Shaw BW Jr. Where monsters hide (editorial). Liver Transpl 2001; 7:928–932. 31. Trotter JF, Talamantes M, McClure M, et al. Right hepatic lobe donation for living donor liver transplantation: impact on donor quality of life. Liver Transpl 2001; 7:485–493. 32. Cronin DC, Millis JM, Siegler M. Transplantation of liver grafts from living donors into adults – too much, too soon. N Engl J Med 2001; 344:1633–1637. 33. Malagó M, Testa G, Marcos A, et al. Ethical considerations and rationale of adult-to-adult living donor liver transplantation. Liver Transpl 2001; 7:921–927. 34. Pomposelli JJ, Pomfret EA, Burns DL, et al. Life-threatening hypophosphatemia after right hepatic lobectomy for live donor adult liver transplantation. Liver Transpl 2001; 7:637–642. 35. Pomfret EA, Pomposelli JJ, Lewis D, et al. Live donor adult liver transplantation using right lobe grafts. Arch Surg 2001; 136:425–433. 36. Akabayashi A, Slingsby BT, Fujita M. The first donor death after living-related liver transplantation in Japan. Transplantation 2004; 77:634. 37. Broering DC, Wilms C, Bok P, et al. Evolution of donor morbidity in living related liver transplantation: a single-center analysis of 165 cases. Ann Surg 2004; 240:1013–1024. 38. Witkowski K, Piecuch J. Liver transplant without a venovenous bypass. Ann Transplant 2001; 6:16–17. 39. Reddy KS, Johnston TD, Putnam LA, et al. Piggyback technique and selective use of veno-venous bypass in adult orthotopic liver transplantation. Clin Transplant 2000; 14:370–374.
959
Section VIII. Liver Transplantation
40. Cabezuelo JB, Ramirez P, Acosta F, et al. Does the standard vs piggyback surgical technique affect the development of early acute renal failure after orthotopic liver transplantation? Transplant Proc 2003; 35:1913–1914. 41. Budd JM, Isaac JL, Bennett J, et al. Morbidity and mortality associated with large-bore percutaneous venovenous bypass cannulation for 312 orthotopic liver transplantations. Liver Transpl 2001; 7:359–362. 42. Scholz T, Solberg R, Okkenhaug C, et al. Veno-venous bypass in liver transplantation: heparin-coated perfusion circuits reduce the activation of humoral defense systems in an in vitro model. Perfusion 2001; 16:285–292. 43. Busque S, Esquivel CO, Concepcion W, et al. Experience with the piggyback technique without caval occlusion in adult orthotopic liver transplantation. Transplantation 1998; 65:77–82. 44. Arcari M, Phillips SD, Gibbs P, et al. An investigation into the risk of air embolus during veno-venous bypass in orthotopic liver transplantation. Transplantation 1999; 68:150–152. 45. Leonardi LS, Boin IF, Leonardi MI, et al. Ascites after liver transplantation and inferior vena cava reconstruction in the piggyback technique. Transplant Proc 2002; 34:3336–3338. 46. Bennett-Guerrero E, Feierman DE, Barclay GR, et al. Preoperative and intraoperative predictors of postoperative morbidity, poor graft function, and early rejection in 190 patients undergoing liver transplantation. Arch Surg 2001; 136:1177–1183. 47. Verran D, Kusyk T, Painter D, et al. Clinical experience gained from the use of 120 steatotic donor livers for orthotopic liver transplantation. Liver Transpl 2003; 9:500–505. 48. Abt PL, Desai NM, Crawford MD, et al. Survival following liver transplantation from non-heart-beating donors. Ann Surg 2004; 239:87–92. 49. Agnes S, Avolio AW, Magalini SC, et al. Should retransplantation still be considered for primary non-function after liver transplantation? Transpl Int 1992; 5 (Suppl 1):S170–S172. 50. Kashyap R, Jain A, Reyes J, et al. Causes of retransplantation after primary liver transplantation in 4000 consecutive patients: 2 to 19 years follow-up. Transplant Proc 2001; 33:1486–1487. 51. Roumilhac D, Poyet G, Sergent G, et al. Long-term results of percutaneous management for anastomotic biliary stricture after orthotopic liver transplantation. Liver Transpl 2003; 9:394–400. 52. Guichelaar MM, Benson JT, Malinchoc M, et al. Risk factors for and clinical course of non-anastomotic biliary strictures after liver transplantation. Am J Transplant 2003; 3:885–890. 53. Sung RS, Campbell DA Jr., Rudich SM, et al. Long-term followup of percutaneous transhepatic balloon cholangioplasty in the management of biliary strictures after liver transplantation. Transplantation 2004; 77:110–115. 54. Gondolesi GE, Varotti G, Florman SS, et al. Biliary complications in 96 consecutive right lobe living donor transplant recipients. Transplantation 2004; 77:1842–1848. 55. Tachopoulou OA, Vogt DP, Henderson JM, et al. Hepatic abscess after liver transplantation: 1990–2000. Transplantation 2003; 75:79–83. 56. Settmacher U, Stange B, Haase R, et al. Arterial complications after liver transplantation. Transpl Int 2000; 13:372–378. 57. Cavallari A, Vivarelli M, Bellusci R, et al. Treatment of vascular complications following liver transplantation: multidisciplinary approach. Hepatogastroenterology 2001; 48:179–183. 58. Stringer MD, Marshall MM, Muiesan P, et al. Survival and outcome after hepatic artery thrombosis complicating pediatric liver transplantation. J Pediatr Surg 2001; 36:888–891.
960
59. Sieders E, Peeters PM, TenVergert EM, et al. Early vascular complications after pediatric liver transplantation. Liver Transpl 2000; 6:326–332. 60. Raby N, Karani J, Thomas S, et al. Stenoses of vascular anastomoses after hepatic transplantation: treatment with balloon angioplasty. AJR Am J Roetgenol 1991; 157:167–171. 61. Stell D, Downey D, Marotta P, et al. Prospective evaluation of the role of quantitative Doppler ultrasound surveillance in liver transplantation. Liver Transpl 2004; 10:1183–1188. 62. Settmacher U, Nussler NC, Glanemann M, et al. Venous complications after orthotopic liver transplantation. Clin Transplant 2000; 14:235–241. 63. Buell JF, Funaki B, Cronin DC, et al. Long-term venous complications after full-size and segmental pediatric liver transplantation. Ann Surg 2002; 236:658–666. 64. Marcos A, Orloff M, Mieles L, et al. Functional venous anatomy for right-lobe grafting and techniques to optimize outflow. Liver Transplantation 2001; 7:845–852. 65. Ghobrial RM, Hsieh CB, Lerner S, et al. Technical challenges of hepatic venous outflow reconstruction in right lobe adult living donor liver transplantation. Liver Transpl 2001; 7:551–555. 66. Scatton O, Belghiti J, Dondero F, et al. Harvesting the middle hepatic vein with a right hepatectomy does not increase the risk for the donor. Liver Transpl 2004; 10:71–76. 67. Borsa JJ, Daly CP, Fontaine AB, et al. Treatment of inferior vena cava anastomotic stenoses with the Wallstent endoprosthesis after orthotopic liver transplantation. J Vasc Interv Radiol 1999; 10:17–22. 68. Yamagiwa K, Yokoi H, Isaji S, et al. Intrahepatic hepatic vein stenosis after living-related liver transplantation treated by insertion of an expandable metallic stent. Am J Transplant 2004; 4:1006–1009. 69. Biggins SW, Beldecos A, Rabkin JM, Rosen HR. Retransplantation for hepatic allograft failure: prognostic modeling and ethical considerations. Liver Transpl 2002; 8:313–322. 70. Azoulay D, Linhares M, Huguet E, et al. Decision for retransplantation of the liver – an experience- and cost-based analysis. Ann Surg 2002; 236:713–721. 71. Sieders E, Peeters PMJG, TenVergert EM, et al. Retransplantation of the liver in children. Transplantation 2001; 71:90–95. 72. Lerut J, Laterre PF, Roggen F, et al. Adult hepatic retransplantation. Acta Gastroenterol Belg 1999; 62:261–266. 73. Postma R, Haagsma EB, Peeters PM, et al. Retransplantation of the liver in adults: outcome and predictive factors for survival. Transpl Int 2004; 17:234–240. 74. Yao FY, Saab S, Bass NM, et al. Prediction of survival after liver retransplantation for late graft failure based on preoperative prognostic scores. Hepatology 2004; 39:230–238. 75. Burton JR Jr, Rosen HR. Retransplantation for hepatitis C: what do we really know? Liver Transpl 2004; 10:1504–1506. 76. Neff GW, O’Brien CB, Nery J, et al. Factors that identify survival after liver retransplantation for allograft failure caused by recurrent hepatitis C infection. Liver Transpl 2004; 10:1497–1503. 77. Berenguer M, Prieto M, Palau A, et al. Severe recurrent hepatitis C after liver retransplantation for hepatitis C virus-related graft cirrhosis. Liver Transpl 2003; 9:228–235. 78. Stravitz RT, Shiffman ML, Sanyal AJ, et al. Effects of interferon treatment on liver histology and allograft rejection in patients with recurrent hepatitis C following liver transplantation. Liver Transpl 2004; 10:850–858.
Section VIII: Liver Transplantation
51
POST LIVER TRANSPLANTATION MANAGEMENT AND COMPLICATIONS Paul J. Gaglio and Robert S. Brown, Jr Abbreviations ACR acute cellular rejection aLDLT adult-to-adult living donor liver transplantation ASHD atherosclerotic heart disease AZA azathioprine CMV cytomegalovirus CNI calcineurin inhibitor CYA ciclosporin DACD donor after cardiac death DD deceased donor
EBV EGD HAT HBV HCC HCV IMPDH INR MELD MMF
Epstein–Barr viral early graft dysfunction hepatic artery thrombosis hepatitis B virus hepatocellular carcinoma hepatitis C virus inosine monophosphate dehydrogenase international normalized ratio model for end-stage liver disease mycophenolate mofetil
INTRODUCTION Liver transplantation, once considered an experimental procedure, has now emerged as the treatment of choice for appropriately selected patients with end-stage liver disease. Based on recently published data by United Network for Organ Sharing (UNOS), 5671 liver transplants (5350 deceased donor (DD) and 321 living donor) were performed in 2004.1 The markedly improved graft and patient survival rates following liver transplantation observed over the last decade are derived from multiple factors. These include advances in surgical technique and immunosuppression, selection of appropriate donors, allografts, and recipients, and improved therapies to prevent and treat postoperative complications. Coincident with enhanced post-transplantation survival rates has been the emergence of complications associated with patient longevity, including non-hepatic disease, complications of immunosuppression, infections, neoplasia, and recurrence of the primary disease for which the liver transplantation was indicated. This chapter will delineate common issues related to post-transplantation management, describe short- and long-term post-transplant complications, and discuss therapies and strategies for prevention.
BACKGROUND Successful liver transplantation involves a complex interplay between the donor, allograft, and recipient. To appreciate fully strategies to enhance management and recognize complications following liver transplantation, donor, allograft, and recipient attributes
MRSA PCR PNF PTLD RAPA TAC US UNOS VRE
Methicillin-resistant Staphylococcus aureus polymerase chain reaction primary graft non-function post-transplant lymphoproliferative disease rapamycin tacrolimus ultrasound united network for organ sharing vancomycin-resistant enterococcus
that negatively affect post-transplantation outcome need to be recognized (Figure 51-1).
DONOR FACTORS It is well established that advanced age, medical comorbidities, and instability in the donor, including requirement for pressors to maintain blood pressure, prolonged hypotension, hypernatremia, and infection, may be associated with diminished recipient and graft survival post-transplantation.2,3 Clearly, when several of these factors are present in a potential donor, the negative impact may be cumulative, and thus, these donors are usually considered unacceptable.4 In stable donors, however, the definition of acceptable age has recently been expanded. Data from various animal models indicate that livers retain the ability to regenerate even in animals of advanced age. With this concept in mind, many transplant centers will accept appropriately selected donors up to age 80; data suggest that these grafts will function well without any negative impact on recipient outcomes.5,6 However, emerging evidence indicates that grafts from older donors should be used with caution in hepatitis C-positive recipients due to poor outcomes, including more severe histologic recurrence of hepatitis C and more rapid progression to fibrosis.7 Other important factors which impact recipient outcomes include donor type, specifically, if the allograft was obtained from a deceased, living, or donor after cardiac death (DACD). DD comprise the majority of liver donors. Either by self-identification while living, or after discussion with “next of kin” when donor brain death has been declared, individuals are acknowledged as potential organ
961
Section VIII. Liver Transplantation
Donor
Graft
Age Hypotension Hypernatremia Infection Medical co-morbidities Pressor requirements Deceased, Living, or Non-heartbeating
Preservation time Steatosis Fibrosis Intrinsic disease
Recipient
MELD score Immune-competence Co-morbid disease
Figure 51-1. Pre-, post-transplantation and donor variables affecting posttransplantation outcome. MELD: model for end-stage liver disease.
donors. Recent data from UNOS indicate that 1- and 3-year patient survival in recipients of DD liver transplant is 81 and 71% respectively. However, despite efforts to maximize utilization of organs acquired from DD, including the use of older donors, steatotic livers, and livers infected with hepatitis C or B, a growing disparity exists between the number of available livers and the number of individuals waiting for transplantation. This critical shortage of organs has resulted in both an increase in the waiting time for liver transplantation and death rate among patients on the waiting list. In response, and as a natural evolution of the pre-existing modality of using a left lateral segment graft for adult-to-child living donor liver transplantation, adult-to-adult living donor liver transplantation (aLDLT) has emerged as an alternative to DD liver transplantation.8,9 This procedure requires that the larger, right lobe of the liver (which accounts for approximately 50–60% of the hepatic mass) be removed from the donor, and implanted into the recipient. Rapid regeneration of the liver remnant in the donor and the partial allograft transplanted into the recipient occurs, resulting in restoration of appropriate liver volume within 1–2 months in both donor and recipient following surgery.10 As the recipient of an aLDLT receives a graft which over time must grow to an appropriate volume, selection of recipients best able to tolerate transplantation of a partial graft is necessary. Current data suggest that the 1-year survival of “sicker” patients formerly identified as UNOS status 2A, or with a model for end-stage liver disease (MELD) score (see below) greater than 25 is approximately 66% following LDLT, compared with 80–90% in less ill patients classified as status 2B or MELD less than 25.11 Thus, in appropriately selected recipients, 1-year graft and patient survival in individuals who undergo LDLT is similar to DD.1 Postoperative complications are similar when comparing DD to LDLT, although recipients of LDLT have a greater rate of biliary complications, including bile leaks and biliary strictures that occur in 15–32% of patients.11 Biliary complications following LDLT are
962
usually managed through non-operative modalities, and do not affect post-transplantation outcomes. In addition, the “small-for-size syndrome,” manifested as prolonged post-transplantation cholestasis may occur following LDLT, particularly in recipients who receive a graft of inadequate size.12 Fortunately, the majority of patients who experience this syndrome recover without the requirement of retransplantation. Recently, significant interest in the utilization of DACD as another modality to increase the pool of available organs has emerged. In contradistinction to DD who are declared brain-dead, DACD are patients who are not declared brain-dead, but are critically ill, without any reasonable expectation of potential recovery, who based on previous stated or families’ wishes are removed from life support at the time of death. There are two types of DACD, controlled and uncontrolled. In the controlled DACD (Maastricht category 3, death anticipated) the patient is removed from life support and death occurs in the operating room. Once death has been declared, organs deemed suitable for transplantation are rapidly perfused with cold preservation solution and removed surgically. The uncontrolled DACD (Maastricht category 1 and 2, death not anticipated) is declared dead after cardiac arrest, rushed to the operating room, and organs are harvested. Uncontrolled DACD are not utilized for liver transplantation due to the high rate of primary non-function (defined below), usually due to prolonged ischemia of the graft.13 When utilizing controlled DACD for transplantation, emerging data indicate that recipient and graft survival are diminished when compared to deceased and LDLT, including a higher incidence of primary non-function, biliary injury, and retransplantation.14 However, several centers have reported acceptable outcomes when utilizing controlled DACD organs, particularly those without significant ischemia in well-selected recipients.15
GRAFT FACTORS Intrinsic to post-transplantation success is the quality of the donor organ. However, due to the rapidly expanding disparity between the number of individuals requiring liver transplantation and organ availability, the concept of “organ quality” is in a constant state of redefinition. By sheer necessity, liver transplant professionals have been required to reassess the limits of acceptable preservation time, degree of steatosis and fibrosis, and the impact of pre-existing disease in the donor related to recipient outcomes, in an attempt to achieve maximal use of available organs. In addition, within the limitations of the present organ allocation system, it is becoming increasingly apparent that appropriate matching of allograft and recipient will be associated with improved outcomes. Many stable liver transplant recipients will do well when transplanted with an allograft formerly defined as marginal, while hospitalized, critically ill patients will not thrive if graft quality is suboptimal.16 The ability to delay transplantation of a liver graft after it is harvested has been achieved by improvements in organ preservation solution. The current limitation in “cold” ischemic time, defined as the number of hours that donor grafts may be in preservation fluid prior to transplantation, is up to 14 hours. Recent data indicate that patient and graft outcomes begin to diminish if cold ischemic time exceeds this, particularly in livers with significant steatosis.17,18 Similar advances in understanding the appropriate degree of steatosis have allowed transplant professionals the ability to maximize use
Chapter 51 POST LIVER TRANSPLANTATION MANAGEMENT AND COMPLICATIONS
of available donor organs. It is now apparent that acceptable posttransplantation outcomes can be achieved when allografts with up to 60% steatosis are used.19 These organs were formerly deemed unacceptable. In addition, perceptions regarding acceptability of grafts with intrinsic disease are being revised, as satisfactory graft and patient survival have been achieved when hepatitis C-infected patients receive an appropriately selected graft from a hepatitis Cinfected donor,20 and when using isolated hepatitis B core antibodypositive, hepatitis B surface antigen-negative grafts.21
RECIPIENT FACTORS Given the relatively stable number of available donor organs in the setting of a rapidly expanding pool of potential recipients, the timing of transplantation is critical. Liver transplantation in a stable patient who is anticipated to do well for many years while waiting for an available organ may not be appropriate, while liver transplantation in a moribund patient with a low probability of post-transplantation survival is similarly inappropriate. Prior to 1997, prioritization for liver transplantation was based on the location where patients received their care (i.e., home, hospital, intensive care unit) and heavily dependent on waiting time. In 2002, UNOS instituted several policies in an attempt to produce a more equitable organ allocation scheme. Waiting time as well as location where patients received their care were eliminated as determinants of prioritization of organ allocation. The MELD score, a numerical calculation based on the recipient’s log transformed renal function (creatinine), total bilirubin, and INR (international normalized ratio for prothrombin time) which has been shown to predict transplantation mortality was adopted by UNOS as a mechanism to prioritize waiting-list candidates. MELD had been validated as a predictor of 3-month survival in diverse groups of patients with various etiologies and manifestations of liver disease.22 Thus, currently a patient’s position on the liver transplantation waiting list is now determined by the MELD score; patients with highest MELD scores are ranked highest on the list. Prospective analysis of the impact of MELD indicates improvement in the rate of transplantation, pretransplantation mortality, and short-term post-transplantation mortality rates.23 However, retrospective analysis has suggested that post-transplantation survival may be reduced in patients with very high pre-transplantation MELD score, particularly in hepatitis C-infected patients.24 Thus, it is clear that careful recipient selection, with attention to pressor and ventilatory requirements, need for dialysis and age are important factors in selecting appropriate candidates for liver transplantation. Another factor associated with post-transplantation outcomes is the innate immunocompetence of the recipient; this may modulate recognition and rejection of the newly transplanted liver. It is well established that the rate of acute cellular rejection (ACR) of the allograft is greater in patients who are “high immunologic responders,” i.e., patients with presumed autoimmune liver diseases such as primary biliary cirrhosis, sclerosing cholangitis, and autoimmune hepatitis.25 In addition, early immune-mediated injury may be a factor associated with primary non-function (see below) of a transplanted liver via mechanisms that are not clearly defined. Selection of immunosuppression as well as strategies to prevent both early and late cellular rejection are predicated on stratifying an individual patient’s risk for immunologic injury.
POST-TRANSPLANTATION MANAGEMENT IMMUNOSUPPRESSIVE MEDICATIONS A cornerstone to post-transplantation management is the ability to prevent rejection of the newly transplanted graft, which, when left untreated, can be associated with graft failure. Strategies to prevent both acute and chronic rejection are predicated on understanding how recognition of the newly engrafted liver as “foreign” occurs, how immune-mediated injury can be modulated, while at the same time avoiding overimmunosuppression, which places individuals at high risk of infection. The various immunosuppressive medications currently utilized in liver transplant recipients and their side effects are listed in Table 51-1 and portrayed schematically in Figure 51-2. Unfortunately, all immunosuppressive therapy is associated with undesired effects, the spectrum of which varies and in some cases overlap. In general, either two or three agents are utilized to prevent allograft rejection in the immediate post-transplant period, utilizing a combination of a calcineurin inhibitor (CNI) such as ciclosporin (CYA) or tacrolimus (TAC), a second agent such as mycophenolate mofetil (MMF) or azathioprine (AZA), and a glucocorticoid such as prednisone. As patients achieve adequate liver function and remain free from rejection beyond 6 months post-transplantation, satisfactory immunosuppression can be achieved in many patients with monotherapy, usually with a CNI.
CORTICOSTEROIDS The immune suppressive effects of corticosteroids and the ability of these agents to reduce the incidence of and treat allograft rejection and to prolong graft survival have been known for decades. Corticosteroids affect multiple aspects of immune function, achieved by the suppression of leukocyte, macrophage, and cytotoxic T-cell activity, and diminution of the effect of cytokines, prostaglandins, and leukotrienes.26 However, coincident with the immunosuppressive effects of corticosteroids is the potential for significant side effects, including hypertension, dyslipidemia, glucose intolerance, bone abnormalities, peptic ulcers, and psychiatric disorders. Thus, many transplant professionals adopt a strategy to taper and discontinue glucocorticoids within the first 6 months to a year following transplantation, while maintaining adequate levels of CNI. This strategy is often altered in patients who undergo liver transplantation secondary to an immunologic disorder such as autoimmune hepatitis, primary biliary cirrhosis, and sclerosing cholangitis due to an enhanced risk of ACR.25 In these patients, either long-term use of corticosteroids, with an attempt to minimize doses, is advocated, or chronic use of MMF or AZA in combination with a CNI is required.
T-CELL-DEPLETING AGENTS In the past, antilymphocyte agents (such as antilymphocyte globulin or antithymocyte globulin) or monoclonal antibody preparations (such as OKT3) directed against specific T-cell antigens have been utilized immediately after liver transplantation to induce rapidly an immune-suppressed state via rapid diminution of T cells.26 However, significant systemic side effects have been associated with the use of these agents, including fevers, allergic reactions, serum sickness,
963
Section VIII. Liver Transplantation
Table 51-1. Immunosuppressive Agents Mechanism of Action
Side effects
Prednisone
Suppression of leukocyte, macrophage, and cytotoxic T-cell activity Decrease cytokines, prostoglandins, and leukotrienes Depletes circulating lymphocytes
Hypertension Dyslipidemia
Antilymphocyte globulin Antithymocyte globulin OKT3
Depletes circulating T cells
Basiliximab daclizumab
IL-2 receptor blockade
Ciclosporin
Inactivates calcineurin, decreases IL-2 production, inhibits T-cell activation
Tacrolimus
Azathioprine
Mycophenolate mofetil
Sirolimus
Inactivates calcineurin, decreases IL-2 production, inhibits T-cell activation Inhibits adenosine and guanine production Inhibits DNA and RNA synthesis in rapidly proliferating T cells Inhibits inosine monophosphate dehydrogenase (IMPDH) Prevents T- and B-cell proliferation Inhibits mTOR (target of Rapamycin) Prevents T-cell replication
Glucose intolerance Bone abnormalities Peptic ulcers Psychiatric disorders Flu-like symptoms Anaphylaxis Lymphoproliferative disorders Flu-like symptoms Anaphylaxis Lymphoproliferative disorders Infections Gastrointestinal distress Pulmonary edema and bronchospasm (rare) Hypertension Renal insufficiency Neuropathy Hyperlipidemia Gingival hyperplasia Hirsutism Insulin resistance Hypertension Renal insufficiency Insulin resistance Neuropathy Hyperlipidemia Leukopenia Anemia Thrombocytopenia Pancreatitis Leukopenia Anemia Thrombocytopenia Gastrointestinal side effects Leukopenia Thrombocytopenia Hyperlipidemia Hepatic artery thrombosis (?) Inhibits wound-healing
Il-2, interleukin-2.
APC
IL-1
Corticosteroids
Cyclo TAC
T-CELL
IL-2
IL2 R
Agent
Sirolimus
Anti-IL2 R Nucleus
Replication MMF Azothiaprine
OKT3 ATG Figure 51-2. Mechanism of action of commonly used immunosuppressive therapies. APC, antigen-presenting cells; IL-1, interleukin-1; TAC, tacrolimus.
interleukin-2 (IL-2). Antibodies directed against the IL-2 receptor are effective for initial immunosuppression, as IL-2 receptor blockade down-regulates IL-2-mediated T-cell proliferation.26 Controlled trials have demonstrated that IL-2 receptor antagonists such as basiliximab and daclizumab, which are given intravenously at the time of transplant and during the first post-transplantation week, can reduce the incidence of acute liver graft rejection when utilized in combination with a CNI, although these agents may not be sufficient to prevent rejection when utilized alone. Side effects of IL-2 inhibitors may include infections, gastrointestinal distress, and rarely, pulmonary edema and bronchospasm. As these agents rarely induce renal dysfunction, many transplant programs utilize IL-2 receptor antibodies as induction therapy in individuals with renal insufficiency at the time of transplantation,28 in an attempt to delay initiation or diminish dose of CNIs which may exacerbate renal insufficiency.
CALCINEURIN INHIBITORS and thrombocytopenia. In addition, the long-term risk of lymphoproliferative disorders (see below) is increased in patients who receive these agents. As a result, at present, these preparations are utilized for the treatment of glucocorticoid-resistant rejection, or less commonly, in patients with renal insufficiency in an attempt to delay the use of either CYA or TAC which may be associated with worsening of renal function.27
IL-2 RECEPTOR BLOCKERS T-cell activation and proliferation following presentation of a foreign antigen requires the induction of several cytokines, including
964
T-cell activation is modulated by multiple cytokines, including IL2. CYA and TAC bind to cytoplasmic receptors, forming complexes which inactivate calcineurin, a key enzyme in T-cell signaling. Multiple randomized controlled trials have demonstrated that both CYA and TAC are effective in reducing acute allograft rejection in liver transplant recipients (reviewed by Conti et al.26). The major adverse events associated with both CYA and TAC include hypertension, renal insufficiency, and neurologic complications. Various investigators have published data which indicate that obesity, hyperlipidemia, hirsutism, and gingival hyperplasia occur more commonly in patients who receive CYA, while a higher rate of
Chapter 51 POST LIVER TRANSPLANTATION MANAGEMENT AND COMPLICATIONS
insulin resistance is seen in patients who receive TAC. Several investigators have reported inconsistent absorption of standard CYA, which has largely been obviated by the use of a microemulsified formulation (e.g., Neoral) which allows more consistent blood levels.29 Despite these limitations, CNIs retain a central role in post-transplant immunosuppression.
ANTIPROLIFERATIVE AGENTS Antiproliferative agents such as AZA and MMF prevent the expansion of activated T cells and B cells and regulate immune-mediated injury. AZA, a purine analogue, is metabolized in the liver to its active compound, 6-mercaptopurine, which inhibits adenosine and guanine production, thus inhibiting DNA and RNA synthesis in rapidly proliferating T cells. MMF is a potent non-competitive inhibitor of inosine monophosphate dehydrogenase (IMPDH), an enzyme necessary for the synthesis of guanine, a purine nucleotide. Randomized controlled trials have demonstrated that MMF is more effective than AZA at reducing the incidence of ACR in solid organ transplant recipients.26 MMF may also reduce the incidence of chronic rejection by inhibiting the proliferation of B lymphocytes. The major toxicities associated with the use of either MMF or AZA are bone marrow suppression with leukopenia, anemia, and thrombocytopenia. MMF has been associated with a greater incidence of dyspepsia, peptic ulcers, and diarrhea when compared to AZA, while pancreatitis may occur in individuals prescribed AZA. These side effects usually abate by dose reduction or discontinuation. A newly released enteric formulation of MMF has been shown in renal transplant recipients to have equivalent efficacy when compared to standard MMF, with fewer gastrointestinal side effects, although this agent has not been well assessed in liver transplant recipients. The majority of transplant centers utilize a combination of a CNI with either MMF or, less commonly, AZA during the first 6 months post-transplantation. As both AZA and MMF are not associated with renal insufficiency, MMF can be utilized with a strategy toward minimizing or avoiding CNI use, particularly in patients with renal dysfunction.30
POST-TRANSPLANTATION COMPLICATIONS The complex nature of the surgical procedure utilized both to explant (remove) the diseased, cirrhotic liver and implant (transplant) the new allograft into the recipient makes it intuitive that the majority of the early complications following liver transplantation are technical and related to the surgical procedure itself. Figures 513 and 51-4 depict the surgical anastamoses required to perform a deceased as well as living donor liver transplantation. However, following the first postoperative days, and as patients progress to the first month post-transplantation and beyond, the nature and variety of complications change. Perhaps overly simplistic, a general approach to categorizing post-transplant complications is “the rule of twos,” i.e., understanding post-transplant complications which occur in the first 2 days, first 2 weeks, first 2 months, and beyond (Table 51-2).
Suprahepatic vena cava
Infrahepatic vena cava Portal vein Hepatic artery Bile duct Figure 51-3. Deceased donor liver transplantation.
OTHER IMMUNOSUPPRESSIVE AGENTS Sirolimus (Rapamycin: RAPA) and its derivative Everolimus represent a new class of compounds, which achieve their immuosuppressive effect by inhibiting mTOR (target of Rapamycin). Inhibition of mTOR diminishes intracellular signaling distal to the IL-2 receptor and prevents T-cell replication. As the lymphoproliferative pathways inhibited by RAPA and Everolimus are distinct from those affected by CNIs, investigators have utilized these agents in combination with CNIs to achieve synergistic effect.26 However, hepatic arterial thrombosis has been reported in patients who receive RAPA in the weeks immediately following transplantation.31 In addition, several investigators have noted problems with woundhealing in patients who receive RAPA, potentially due to impairment of granulation mediated by inhibition of transforming growth factorb. Leukopenia, thrombocytopenia, and hyperlipidemia are the principal toxicities associated with RAPA and Everolimus. A positive attribute of both RAPA and Everolimus is preservation of renal function; post-transplantation renal insufficiency can be reversed when RAPA is initiated and CNIs are withdrawn.32
Vena cava Hepatic vein
Hepatic artery Portal vein Bile duct
Intestine Figure 51-4. Living donor liver transplantation.
965
Section VIII. Liver Transplantation
Table 51-2. “Rule of Twos”: Complications During the first 2 Days, 2 Weeks, 2 Months, and Beyond Following Liver Transplantation Two Days Primary non-function Early graft dysfunction Hepatic artery thrombosis Hepatic and portal vein thrombosis Preservation injury Two Weeks Acute cellular rejection Bacterial and fungal infection Biliary complications Cytomegalovirus infection Two Months and Beyond Hypertension Hyperlipidemia Diabetes Obesity Cardiac disease Renal dysfunction Chronic rejection Fungal infection (Cryptococcus, Aspergillus) Cytomegalovirus Post-transplant lymphoproliferative disorder Malignancy Recurrence of primary disease
COMPLICATIONS IN THE FIRST 2 DAYS Primary Non-Function and Early Graft Dysfunction Primary graft non-function (PNF), defined as acidosis, rising INR, progressive elevation in liver transaminases and creatinine, and decreased mentation occur when the newly transplanted liver allograft fails to function normally. The mechanisms responsible for this phenomenon are incompletely understood but may relate to donor factors, inadequate preservation, prolonged ischemia, extensive steatosis of the graft, hepatic artery thrombosis (HAT: see below), or immune response to the implanted organ.33 In the setting of PNF, a rapid assessment of hepatic artery flow needs to occur, as immediate surgical repair of a thrombosed hepatic artery may reverse PNF. If PNF occurs in the absence of HAT, emergent retransplantation is required. In contrast to PNF, early graft dysfunction (EGD) is manifested by an early rise in serum transaminases to values greater than 2000–3000 IU/l, cholestasis with rising bilirubin levels, without concomitant impairment in mental status, coagulopathy, and renal function. EGD may be secondary to ischemic injury or steatosis in the graft, and typically occurs within the first 24–48 hours after the transplant. Unlike PNF, the manifestations of EGD usually improve, and emergency retransplantation is not necessary.
Hepatic Artery Thrombosis A potentially devastating post-transplantation complication is HAT. It is intuitive that HAT occurs more commonly in pediatric patients when compared to adults due to the technical difficulties associated with the anastomosis of smaller-size vessels.34 HAT in the immediate postoperative period may be associated with graft failure,
966
massive elevation in serum liver transaminases, bile leak, hepatic necrosis, and sepsis. As the blood supply to the biliary tree is exclusively provided by the hepatic artery, it is not surprising that HAT is frequently associated with irreversible injury to the biliary tract.35 In individuals with HAT-associated biliary injury, bile duct necrosis, dehiscence of the bile duct or biliary-enteric anastomosis, and the development of diffuse intra- and extrahepatic biliary strictures may occur. These changes may induce chronic cholangitis, intrahepatic abscesses, and eventually secondary biliary cirrhosis and liver failure. For these reasons, HAT which occurs within the first 7 days after liver transplantation is an indication for emergent artery repair or retransplantation. Due to the potentially devastating consequences of HAT, most transplant centers utilize duplex-ultrasound (US) in the immediate post-transplant period as a mechanism to screen for this complication. If duplex-US suggests HAT, emergent angiography is usually performed to confirm the diagnosis prior to laparotomy and surgical revision of the hepatic artery. At specialized transplant centers with specific expertise in interventional radiology, emergency repair and stenting of a thrombosed hepatic artery may be effective in preventing the requirement for retransplantation.36 However, in the majority of cases, emergent surgical revision with an attempt to restore liver perfusion is required to prevent irreversible bile duct injury, and obviate the requirement for repeat liver transplantation. If this cannot be achieved, liver retransplantation may be necessary.
Portal and Hepatic Vein Thrombosis Thrombosis of portal and/or hepatic veins in the immediate posttransplant period is rare, but may be associated with significant consequences. Acute Budd–Chiari syndrome due to hepatic vein or vena cava thrombosis may occur, with abdominal pain, peripheral edema, and the threat of graft failure, as hepatic congestion in the newly transplanted allograft is poorly tolerated. In this circumstance, emergency surgical revision and repair of thrombosis are required. Acute portal vein occlusion may be associated with exacerbation of pre-existing portal hypertension, associated with gastrointestinal bleeding from portosystemic collateral vessels such as esophageal and gastric varices. Acute portal vein thrombosis is managed by surgical repair, while chronic portal vein thrombosis may be well tolerated. Over the last several years, interventional radiologic techniques have been utilized to successfully manage many vascular complications successfully following both deceased and living donor liver transplantation.37
Ischemic and Preservation Injury The newly transplanted liver can be subjected to multiple varieties of ischemic injury: cold (or hypothermic) and warm (or normothermic).38 Cold ischemia occurs prior to transplantation while the liver is cooled in preservation solution, awaiting implantation. Warm ischemia occurs during the transplantation procedure, when hepatic blood flow is interrupted to minimize blood loss during transplantation, or when the formerly cooled liver is subjected to room or body temperature during transplantation. Morphologically, cold ischemia induces injury to the sinusoidal endothelial cell, although this process is usually well tolerated. In contrast to cold ischemia,
Chapter 51 POST LIVER TRANSPLANTATION MANAGEMENT AND COMPLICATIONS
warm ischemia is tolerated poorly and rapidly leads to the death of hepatocytes, with resultant elevation in serum transaminases, apoptosis, and centrilobular necrosis. In general, ischemic injury is well tolerated, but, if significant, may be associated with graft failure. Several investigators have noted improvement in ischemic injury prior to liver transplantation by employing a technique described as ischemic preconditioning.39 Ischemic preconditioning consists of a brief period of ischemia followed by a short interval of reperfusion before the actual surgical procedure. During liver transplantation, hepatic inflow is occluded by placing a vascular clamp or a loop around the portal triad (i.e., portal vein, hepatic artery, and bile duct), rendering the whole organ ischemic for 10–15 minutes, after which the clamp is removed and the liver is reperfused for 10–15 minutes. This technique may be of particular benefit in allografts with significant steatosis.39
COMPLICATIONS IN THE FIRST 2 WEEKS Acute Cellular Rejection Liver allografts are relatively privileged immunologically, and thus, the incidence and consequences of ACR are diminished when compared to other solid organs utilized for transplantation. The reported incidence of ACR within the first post-transplant year is 30–50%, in most cases, usually occurring within the first 10–14 postoperative days. The clinical presentation is variable; ACR may be asymptomatic, or associated with fever or abdominal pain. Laboratory findings include elevation or failure of normalization of serum transaminases, concurrent with a rising alkaline phosphatase and/or bilirubin. The diagnosis of acute liver graft rejection is confirmed by liver biopsy and examination of liver histology.40 Conventional histologic criteria associated with ACR, now standardized in a numerical score called the BANFF criteria, include the presence of periportal lymphocytic infiltrate, as well as bile duct and vascular endothelial injury. The contribution of each component of rejection, i.e., degree of periportal infiltrate, bile duct injury, and endotheliitis, is tabulated and the degree of rejection, i.e., mild, moderate, or severe, is determined. Most cases of ACR respond to treatment with intravenous glucocorticoids. Approximately 10% of patients with ACR will not improve with intravenous glucocorticoids, requiring the administration of monoclonal or polyclonal anti-T cell antibody therapy. Mild and moderate ACR may also respond to either increasing the dose of the primary immunosuppressive agent, or switching to an alternate CNI. This approach has been used with increasing frequency, particularly in patients transplanted for hepatitis C virus (HCV) and hepatitis B virus (HBV) due to concerns regarding the adverse effects of overimmunosuppression on viral recurrence. It has been well established that HCV-induced graft failure, progression to advanced histologic injury, and cholestatic hepatitis occur more frequently in HCV-infected individuals who receive high-dose intravenous glucocorticoids and antilymphocyte preparations.41 Therefore, modulating immunosuppression in the setting of rejection by either increasing the dose or substituting a CNI, and/or reintroduction of mycophenolic acid may be preferred in the setting rather than the use of bolus glucocorticoids and/or anti-T-cell antibodies.
bidity and mortality in the immediate postoperative period. Both bacterial and fungal infection may be observed. Common sources of bacterial infections include the lungs, urogenital system, and the surgical wound. Unfortunately, infection with multidrug-resistant Gram-positive and Gram-negative organisms has proliferated at many transplant centers, likely coincident with the overall rise in severity of medical illness at the time of transplant, and the overuse of antibiotics. Methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococcus (VRE) are the most prevalent of these organisms, and may be associated with a prolonged intensive care unit course and hospital stay, as well as increased morbidity and mortality.42,43 In addition, the incidence of fungal infection is increasing, and appears to be more common in patients with advanced liver disease who have been treated with repeated courses of antibiotics prior to transplantation.44 As a mechanism to prevent post-transplant infection, many transplant centers have adopted a policy of instituting broad-spectrum antibiotics during the first several days to weeks following liver transplantation. Additionally, a short course of antifungal prophylaxis may be instituted, particularly in patients who have received multiple courses of antibiotics and/or perioperative transfusions, those requiring reoperation in the early period following transplantation, and in patients with renal failure.
Biliary Complications Bile leaks and strictures generally occur at the anastomosis of the donor and recipient bile ducts, recognized by a rise in serum bilirubin and/or alkaline phosphatase or by the presence of bile in surgical drains. The incidence of biliary complications is between 5 and 15% following DD liver transplantation. However, between 15 and 30% of patients who undergo living donor liver transplantation develop biliary complications, due to the number and complexity of the biliary anastamoses required, and due to bile leaks from small biliary radicles from the cut surface of the partial liver graft.45 In both deceased and living donor recipients, the majority of bile leaks resolve spontaneously. As the biliary tree receives the vast majority of its blood supply from the hepatic artery, the adequacy of hepatic artery blood flow needs to be evaluated in the setting of any biliary injury. If spontaneous resolution of the bile leak does not occur, endoscopic or radiologic placement of a biliary stent across the biliary anastamoses is often successful.36,46 In some cases surgical exploration and revision of the biliary anastamoses to a Roux en Y choledochojejunostomy may be required. Anastomotic biliary strictures require careful attention, as, if left untreated, cholangitis, graft dysfunction, and eventually secondary biliary cirrhosis may occur. Techniques for management include dilatation and stenting via biliary endoscopy or percutaneous transhepatic cholangiogram by an interventional radiologist. If these modalities are unsuccessful, surgical revision of the biliary anastamosis, achieved by the performance of a Roux en Y choledochojejunostomy in patients with a choledochocholodochostomy, may be required. In some cases, retransplantation may be necessary.
Cytomegalovirus (CMV) Infection Bacterial and Fungal Infections Infection is the most common complication following liver transplantation and is responsible for the majority of the short-term mor-
CMV infection after liver transplantation remains a significant cause of morbidity, and usually occurs within the first several weeks to months following transplantation. The incidence of CMV infection
967
Section VIII. Liver Transplantation
after liver transplantation ranges from 25% to 85%, although the vast majority of CMV infections remain asymptomatic. However, up to 5–15% of patients develop symptomatic CMV infection associated with fever, leukopenia, or thrombocytopenia, and a minority develops tissue-invasive disease. Tissue-invasive disease after orthotopic liver transplantation most frequently affects the liver graft, although pneumonitis, myocarditis, gastritis, and colitis have been reported.47 CMV infection and, especially, CMV hepatitis may have a role in enhancing chronic allograft rejection, although this remains controversial. In addition, CMV infection post liver transplantation has been implicated as a factor that adversely affects graft survival in HCV-infected patients.48 Risk for CMV infection post liver transplantation relates primarily to previous CMV exposure in both recipient and donor, with the lowest rates occurring when both donor and recipient are CMV IgG antibody-negative, while the highest rates occur when the donor is positive and recipient negative. Another important risk factor for acute CMV infection includes overimmunosuppression, as occurs during the treatment of rejection, particularly if anti-T-cell agents such as OKT3 or antithymocyte globulin are used. Thus, anti-CMV therapy is often initiated concurrently when treating ACR with these agents. The availability of intravenous as well as oral ganciclovir and the more bioavailable oral formulation (valganciclovir) for the prevention of CMV disease has dramatically changed the epidemiology and outcomes of CMV infection. In addition, newer diagnostic tests, including assays for CMV antigenemia and quantification of CMV DNA by polymerase chain reaction (PCR), have markedly improved the ability to diagnose CMV infection rapidly and accurately. All liver transplant programs provide prophylaxis to at least those patients at highest risk for developing active CMV infection. However, the precise treatment, duration of therapy, and patient groups offered extended prophylaxis vary considerably between transplant centers. In general, intravenous ganciclovir is administered for the first 7 days post-transplant, and is converted to oral ganciclovir or valganciclovir, either when the patient is able to take oral medications or after a predefined length of time. The duration of oral ganciclovir use also varies considerably between centers, but typically lasts several weeks to months. In the setting of invasive CMV disease, intravenous ganciclovir is initiated and continued for several weeks. Though it has been used by some centers and provides drug levels similar to intravenous administration, oral valganciclovir has not been approved for treatment of CMV. The risk of CMV relapse is reduced if treatment is continued until CMV-DNA can no longer be detected by a PCR assay.49 CMV-immunoglobulin is generally reserved for those patients who develop active CMV disease despite prophylaxis, or fail to respond to intravenous ganciclovir. Given the sensitivity of the current antigen and PCR techniques to detect viremia, several studies have advocated routine surveillance for CMV post-transplantation and pre-emptive treatment if and when the virus is detected.50
COMPLICATIONS BEYOND 2 MONTHS Improvements in the surgical techniques required to perform transplantation, the treatment of postoperative complications, and prevention of rejection have been associated with significant improvements in short-term morbidity and mortality following
968
transplantation. Coincident with improvements in short-term outcomes has been a rise in long-term complications.
Cardiovascular Disease In general, patients with advanced cardiovascular disease, including atherosclerotic heart disease (ASHD), significant peripheral vascular disease, and poorly controlled hypertension, have been excluded from liver transplantation due to concerns regarding increased perioperative morbidity and mortality. Unfortunately, in patients without cardiovascular disease pretransplant, many of the risk factors for coronary artery disease will develop post liver transplantation and require long-term management to prevent complications.
Hypertension Hypertension is common following liver transplantation, and may be associated with multiple factors, including the direct effects of CNIs, renal insufficiency, and obesity. Pre-existing hypertension may be masked preoperatively due to low systemic vascular resistance associated with the hemodynamic milieu of cirrhosis, thus protecting the patient from developing elevated blood pressure. The incidence of post-transplant arterial hypertension in liver graft recipients ranges from 65% to 85% since CYA has been introduced as an immunosuppressant agent. Various mechanisms have been proposed to explain the hypertensive effect of CYA, including vasoconstriction of afferent renal arterioles with impairment of glomerular filtration rate and sodium excretion, increased sympathetic nervous system activation altering the renin–angiotensin system, increasing intracellular calcium concentration, or synthesis and release of endothelin-1.51 Although TAC may be associated with hypertension post liver transplantation, the effect may be less when compared to CYA.52 Unfortunately, as in non-transplanted patients, untreated hypertension may be associated with multiple complications, including end-organ disease and renal failure. The treatment of hypertension post liver transplantation proceeds in protocolized, stepwise fashion, including limiting salt intake, the use of calcium-channel blockers, and the addition or substitution of a beta-blocker or angiotensinconverting enzyme inhibitor in difficult-to-control patients. The dihyropyridine class of calcium-channel blockers, including nifedipine, israpidine, amlodipine, felodipine, and nicardipine, are preferred as first-line agents as they are long-acting, minimally interact with CNIs, and have limited side effects. Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers have also been utilized, although they may lead to hyperkalemia and renal insufficiency in liver transplant recipients and should be used with caution.
Hyperlipidemia Both increased cholesterol and triglyceride levels occur commonly post liver transplantation. It has been reported that 16–43% of patients have elevated serum cholesterol levels, and 40% have increased triglycerides. Risk factors for hyperlipidemia include the effects of commonly used pharmacologic agents after transplantation, such as CNIs, and intravenous and oral glucocorticoids, as well as RAPA. Many investigators have discussed the potential benefit of altering CNIs after liver transplantation in patients with hyperlipidemia; TAC has been reported by some to be associated with a
Chapter 51 POST LIVER TRANSPLANTATION MANAGEMENT AND COMPLICATIONS
decreased incidence of cholesterol and lipid abnormalities,53 while tapering glucocorticoid dose may also be of benefit. In addition, the use of statin cholesterol-lowering agents has been found to be safe and effective in decreasing serum total and low-density lipoprotein cholesterol, without changing CNI levels.54 Another potential, although as yet unproven, benefit of statin therapy following liver transplantation may be its effect on rejection. It has been observed that statins exhibit anti-inflammatory effects, modulate endothelial function, and may repress the induction of major histocompatibility complex class II complexes. Pravastatin has been shown to reduce acute rejection after cardiac and renal transplantation as well as reduce natural killer cell cytotoxicity.55 The short- and long-term effects of statins in the prevention of post-transplantation rejection are being evaluated in prospective clinical trials.
Diabetes The incidence of diabetes post liver transplantation ranges from 13 to 25%.56 Type 1 diabetes prior to or developing post liver transplantation is associated with a poor prognosis. A recent review of the UNOS database indicated that cryptogenic cirrhosis, hypertension, and coronary artery disease were significantly more common following liver transplantation in patients with type 1 and 2 diabetes compared to non-diabetics. Five-year patient and graft survivals by Kaplan–Meier analysis were significantly lower for type 1 compared with type 2 or non-diabetics.57 Despite these grim statistics, a small subset of patients with insulin-requiring diabetes pretransplant may lose their insulin requirement, likely due to a restoration of insulin sensitivity coincident with liver transplantation and loss of portal hypertension.58 Post-transplantation diabetes is multifactorial and may be related to immunosuppression, including the use of CNIs and glucocorticoids, treatment of rejection, and increased caloric intake concomitant with an enhanced feeling of well-being. Treatment of diabetes post-transplantation includes standard medical and nutritional therapy, including limiting caloric intake, an appropriate diet with weight loss, and agents that induce hypoglycemia, including oral hypoglycemics and/or insulin as necessary. Data is emerging which suggests that choice of immunosuppression may affect the incidence of diabetes. Patients treated with TAC appear to develop diabetes at a greater rate than those treated with CYA preparations, even when controlling for dose, exposure, and glucocorticoid use.59,60
Obesity The incidence of obesity, defined as a body mass index greater than 30 kg/m2, following liver transplantation has been reported to be as high as 30%.61 Various factors, including choice of immunosuppression, pretransplant obesity, and failure to return to work and healthy pretransplant activities following transplantation have been considered causative. The use of intravenous as well as prolonged oral glucocorticoid therapy contributes significantly to post-transplantation obesity. Despite the increasing incidence of obesity both pre- and post-transplantation, long-term outcomes in obese patients appear to be stratified by the degree of obesity. A recent review of the UNOS database found that, after adjusting for confounding variables, patient and graft survival at 1 month and 1, 2, and 5 years postoperatively were not significantly different when comparing obese to non-obese patients.62 However, when post-transplant sur-
vival in patients with morbid obesity was analyzed using the same database, primary graft non-function, as well as immediate, 1- and 2-year mortality was significantly higher in the morbidly obese group, and 5-year mortality was significantly higher in both severely and morbidly obese patients, mainly due to adverse cardiovascular events.63 Thus, morbid obesity appears to be an independent predictor of mortality. Many transplant centers either exclude patients with morbid obesity from consideration for transplantation or encourage weight loss prior to transplantation. Treatment of obesity post liver transplantation is unfortunately difficult: many centers adopt a policy of minimizing and/or rapidly tapering glucocorticoids as appropriate, and encouraging exercise and limiting caloric intake, with unfortunately limited success. Several investigators have studied the effect of changing immunosuppression (substituting TAC for CYA preparations) as there is evidence that rates of obesity post-transplantation are greater in patients treated with CYA preparations compared to TAC,60 even when controlling for effect of glucocorticoid dose.
Cardiac Disease As the incidence of obesity, diabetes, hypertension, and hypercholesterolemia is increased following liver transplantation, it is not surprising that ASHD post liver transplantation occurs commonly. Indeed, when excluding recurrent disease, graft loss due to technical complications, and de-novo malignancy, ASHD represents the most common cause of death post liver transplantation. Liver allograft recipients have a greater risk of cardiovascular deaths and ischemic events than an age- and sex-matched population of nontransplanted patients.64 Thus, as ASHD contributes significantly to post-transplantation morbidity and mortality, appropriately identifying and treating patients with ASHD before liver transplantation is vital. Patients with documented or major risk factors for ASHD should undergo extensive pretransplantation evaluation which includes a cardiac stress test and/or cardiac catheterization. Patients with advanced heart disease not amenable to interventional therapy or with poor cardiac function should be deemed non-candidates for liver transplantation. Aggressive therapies, including pharmacologic agents, and, if necessary, angiographic or surgical correction of cardiac vasculature, should be offered to appropriately selected patients with mild or moderate ASHD. An important issue that must be kept in mind when evaluating any patient prior to liver transplantation is the unique hemodynamic profile associated with portal hypertension. First described nearly half a century ago, the resultant increased cardiac output and reduced systemic vascular resistance may reduce afterload, and mask left ventricular dysfunction. Following liver transplantation, normalization of systemic vascular resistance may not be tolerated by a dysfunctional left ventricle.65 Thus, in addition to careful evaluation of ASHD, a complete evaluation of left and right ventricular function must occur as part of an appropriate pretransplantation cardiac assessment.
Renal Dysfunction A significant source of morbidity and mortality following liver transplantation is renal failure. Renal impairment has been well described in the immediate perioperative period due to hepatorenal syndrome, hypovolemia, and acute tubular necrosis. Although the majority of
969
Section VIII. Liver Transplantation
patients recover, up to 25% of patients develop significant renal dysfunction, defined as a glomerular filtration rate of < 40 ml/min by 5 years after transplantation, and approximately 10% develop endstage renal disease and dialysis dependence by 10 years after transplantation.66 Factors associated with the development of end-stage renal disease following liver transplantation include a higher preoperative serum creatinine level, hepatorenal syndrome, requirement for dialysis in the first 3 months postoperatively, and elevated serum creatinine at 1 year post-transplantation.67 Therapy of renal insufficiency following liver transplantation includes a thorough search for reversible factors, including hypovolemia and renal vascular injury, minimizing nephrotoxic medications such an aminoglycosides, non-steroidal anti-inflammatory medications, trimethoprim/sulfamethoxazole, controlling diabetes and hypertension, and appropriately adjusting CNI levels. It has been well established that CNIs such as CYA and TAC are associated with nephrotoxicity via multiple mechanisms, and are classified as acute or chronic: acute CNI nephrotoxicity occurs in the first weeks after transplantation, is associated with impaired renal blood flow, and is usually reversible with CNI dose reduction. In contrast, chronic CNI nephrotoxicity observed several months after transplantation is associated with parenchymal renal injury, and may occur even in the setting of appropriate CNI blood levels, and unfortunately, lead to permanent renal dysfunction. An alternative approach to diminishing CNI exposure is withdrawal and substitution of an alternative immunosuppresive agent. The effect of withdrawal of CNIs and substitution of MMF has been evaluated in prospective trials. This technique has resulted in a significant decrease in the mean values of serum creatinine 6, 12, and 18 months after beginning therapy with MMF, with no significant effect on graft function or rejection. Using this approach, the withdrawal of CNI may be achieved in the majority of patients by 12–18 months post-transplantation.68 It is important to remember however that immunosuppression with MMF monotherapy may be associated with a small but appreciable increased risk of ACR. Finally, conversion to RAPA, an immunosuppressant agent with minimal nephrotoxicity, has been associated with improvement in renal function in patients with CNI-associated renal insufficiency.32
Chronic Rejection Chronic allograft rejection, also termed vanishing bile duct syndrome, occurs rarely after liver transplantation. Diagnostic criteria for chronic rejection include bile duct atrophy affecting the majority of bile ducts, with or without bile duct loss, and obliterative arteriopathy and/or venopathy in large branches of the hepatic artery or portal vein.69 Risk factors for chronic liver rejection include transplantation for primary sclerosing cholangitis, primary biliary cirrhosis, human leukocyte antigen mismatch between donor and recipient, and CMV infection. Chronic rejection is a harbinger of poor outcomes; altering immunosuppression is rarely associated with improvement, often resulting in the requirement for retransplantation.
foreign, and avoidance of infectious complications associated with overimmunosuppression. Most liver transplant recipients are able to diminish rapidly the doses of immunosuppressants, to the extent that the majority of patients are on a single agent, usually a CNI, by 6–12 months post-transplant. Despite rapid tapering of immunosuppression, infectious complications continue to be observed.
FUNGAL INFECTIONS Invasive fungal infection, particularly with aspergillosis species has long been recognized as one of the most significant opportunistic infections in liver transplant recipients. Although uncommon, occurring in 1–6% of patients, the mortality rate for patients with invasive disease exceeds 90%. Most patients present with either pulmonary or central nervous system infections (such as brain abscess). The majority of Aspergillus infections occur within the first month after transplantation, and the incidence is correlated with the degree of illness of the recipient; the overwhelming majority of liver transplant recipients with invasive aspergillosis had evidence of significant hepatic and/or renal dysfunction prior to transplantation, particularly if they were dialysis-dependent. In addition, up to onefourth of all cases of invasive aspergillosis have occurred after retransplantation. Other factors associated with a greater risk of aspergillosis include concomitant CMV infection, treatment with OKT3, and neutropenia. Despite the historically significant death rate associated with aspergillosis infection, recent improvements in both recognition and treatment of infection have been associated with diminished death rates when compared to historical controls. A recently published manuscript indicates that Aspergillus infections in liver transplant recipients are occurring later in the posttransplantation period, with mortality rates decreasing from 90 to 60%.70
CRYPTOCOCCUS Although cryptococcal meningitis occurs rarely in a non-immunosuppressed patient population, organ transplantation remains one of the major factors for cryptococcosis in non-HIV-infected patients. Cryptoccoccal infection usually presents with central nervous system symptoms associated with meningeal irritation greater than 6 months post orthotopic liver transplantation, although cutaneous lesions in the absence of meningeal signs may occur. Decreased cerebrospinal fluid glucose as well as elevated protein and white blood cell count help to establish the diagnosis, while a positive India ink stain and cryptococcal antigen in cerebrospinal fluid and serum confirms the diagnosis. Treatment of cryptococcal infection usually requires a combination of amphotericin B with or without 5-flucytosine, followed by suppressive therapy with oral fluconazole for up to 6 months. Rarely, intrathecal administration of amphotericin B is required, usually in the setting of persistent symptoms of meningeal irritation, and elevated cryptococcal antigen titers in cerebrospinal fluid despite systemic antifungal therapy.71
CYTOMEGALOVIRUS INFECTION
INFECTIOUS COMPLICATIONS Ensuring long-term patient and graft survival requires an appropriate balance of suppression of immune recognition of the graft as
970
Although the majority of CMV infections occur within the first several months post-transplantation, patients continue to remain at increased risk for CMV infection. The greatest risk factor associated with CMV infection includes overimmunosuppression. As in the
Chapter 51 POST LIVER TRANSPLANTATION MANAGEMENT AND COMPLICATIONS
immediate pretransplant period, this most commonly occurs when treating acute allograft rejection. The management of CMV infection and/or disease in the months to years following transplantation is similar to when CMV occurs in the short term postoperatively and is described in detail above.
POST-TRANSPLANT LYMPHOPROLIFERATIVE DISEASE Epstein–Barr viral (EBV) infection is rarely associated with clinical disease in patients who were previously exposed to EBV before liver transplantation. However, EBV-associated syndromes, including post-transplant lymphoproliferative disease (PTLD), are particularly important in children who experience a primary EBV infection after transplantation, and in adults who develop EBV reactivation. In general, clinical disease develops in the setting of impaired production of EBV-specific cytotoxic lymphocytes due to immunosuppression, with subsequent B-lymphocyte transformation and clonal expansion. Variants of PTLD exist where T cells may also be transformed. Clinical presentation is varied, ranging from a mononucleosis-like syndrome, localized lymphadenopathy, or systemic disease with multiorgan involvement. Therefore, unexplained fever, weight loss, lymphadenopathy, graft dysfunction, neurologic symptoms, diarrhea, or pulmonary symptoms following liver transplantation require a thorough investigation for PTLD.72 Evaluation of possible PTLD requires a biopsy of affected tissue and/or lymph nodes with assessment of cell type, gene rearrangement studies to determine B- versus T-cell origin, and a determination if the involved cell lines are monoclonal or polyclonal, as polyclonal disease has been reported to be more responsive to therapy. Treatment of PTLD includes the withdrawal of all immunosuppression (except low-dose prednisone), and in cases where significant lymphadenopathy or organ involvement is present, a multifaceted treatment approach, including antiviral agents (ganciclovir or aciclovir), anti-B-cell agents (anti-CD20 rituximab) and systemic chemotherapy in refractory cases.73 As primary infection with EBV is associated with PTLD, many pediatric transplant programs have initiated a surveillance program post-transplantation whereby EBV DNA is quantified, and immunosuppression is adjusted and/or EBV-directed antiviral therapy is initiated if titers increase.74
MALIGNANCY As long-term survival rates in patients who undergo liver transplantation have improved, a concomitant increase in late complications including malignancy has been noted. These may be de-novo neoplasia, or a recurrence of cancers detected prior to liver transplantation.
HEPATOCELLULAR CARCINOMA Early experience with liver transplantation in the setting of hepatocellular carcinoma (HCC) was associated with a recurrence rate of up to 80% and dismal long-term survival. This resulted in HCC being considered a contraindication to liver transplantation. However, based on work by Mazzaferro and others, selected patients with limited HCC (one solitary lesion < 5 cm or three lesions each
< 3 cm or or T1–2N0M0) were found to have excellent long-term outcomes, with a 5-year survival rate of 70% and a recurrence rate below 15%.75 These criteria are adopted by many transplantation programs to determine candidacy for transplantation in patients with HCC. As liver transplantation in patients with HCC is hampered by both a shortage of donor organs and increased waiting time, many transplant centers have adopted a strategy of considering living donor liver transplantation for these patients, and/or treating patients with ablative therapies such as hepatic artery chemoembolization or radiofrequency ablation prior to transplantation in an attempt to diminish tumor progression. In addition, the current organ allocation system using the MELD score has been modified to grant added priority to patients with HCC who meet Mazzaferro criteria. Unfortunately, HCC recurrence can occur post-transplantation, particularly in patients with large tumors, macrovascular invasion, and evidence of extrahepatic disease at the time of transplantation.
CUTANEOUS MALIGNANCY Skin cancer is the most common malignancy occurring after solid organ transplantation, with an incidence as high as 35–70% in areas of the world where sun exposure is common.76 The incidence of skin cancer after liver transplantation has been reported at 1.6–2.2%, but may be underreported as patients may be lost to follow-up to the transplant center and treated by local physicians. A recent study evaluated the incidence of cutaneous malignancies in 151 liver transplant recipients; 86 documented skin cancers were found in 34 patients, with the majority being squamous cell, followed by basal cell and melanomas. Predictors of malignancy included male gender, red hair, brown eyes, diagnosis of primary sclerosing cholangitis, and use of CYA.77 Thus, many transplant centers recommend an annual dermatologic evaluation in all patients who have undergone liver transplantation.
OTHER MALIGNANCIES An increased incidence of de-novo non-lymphoid malignancies has been shown in immunocompromised patients. To determine the incidence of malignancy in a post liver transplantation population, investigators from King’s College Hospital (London, UK) analyzed all patients who underwent liver transplantation between January 1988 and December 1999. Factors potentially related to risk of malignancy that were evaluated included etiology of liver disease, choice of immunosuppression, and number of rejection episodes. Of 1140 patients undergoing 1271 liver transplantations, 30 patients (2.6%) developed de-novo non-lymphoid malignancy after transplantation. As anticipated, skin cancer was the most common malignancy, followed by oropharyngeal carcinoma, bladder carcinoma, acute leukemia, breast carcinoma, and various other malignancies (sarcoma, seminoma, small-bowel, colon, renal, pancreas). The mean time of presentation of the malignancy after transplantation was 45.1 months (range, 6–133 months), and mean age at diagnosis was 55 years (range, 34–71 years). Interestingly, 1-, 3-, and 5-year survival was not different in patients with and without malignancy. Of note, the authors found that the incidence of de-novo malignancy was significantly greater in patients who underwent transplantation for alcoholic liver disease compared with other causes of liver disease.78
971
Section VIII. Liver Transplantation
Various investigators have also noted the more rapid progression of diverse premalignant conditions following liver transplantation. A greater incidence of colonic adenomas,79 progression of Barrett’s esophagus to high-grade dysplasia,80 and colonic dysplasia, particularly in patients with pre-existing inflammatory bowel disease,81 have been reported. Therefore, many transplant centers have adopted a policy of performing annual screening for malignancies, including skin exam, mammogram, Pap smear or prostate-specific antigen in the appropriate gender, and recommending colonoscopy every 3 years or more frequently if clinically indicated.
RECURRENCE OF PRIMARY DISEASE FOLLOWING LIVER TRANSPLANTATION A major challenge to the liver transplant community is recurrence of the primary disease which caused the patient’s liver to fail. Diseases that do not recur following liver transplantation include congenital anatomic anomalies (biliary atresia, polycystic liver disease, Caroli’s disease, Alagille’s syndrome, congenital hepatic fibrosis) and metabolic diseases (Wilson’s disease, a1-antitrypsin deficiency: Chapter 53). However, all other causes of liver disease, including primary biliary cirrhosis, primary sclerosing cholangitis, autoimmune hepatitis, non-alcoholic fatty liver disease, hemochromatosis, and alcohol-associated liver disease, have been reported to recur after liver transplantation. In some cases this may lead to liver injury and graft failure.82–86 Disorders most commonly associated with recurrence include HBV and HCV (Chapter 52). Fortunately, recurrence of HBV after liver transplantation can be prevented by administering hepatitis B immune globulin at the time of transplantation and at regular intervals thereafter, with or without the use of antiviral agents such as lamivudine and adefovir.87 Unfortunately, HCV recurrence following liver transplantation represents a significant source of morbidity and mortality. In patients with active HCV replication prior to transplantation, reacquisition of viremia following transplantation is universal, and allograft hepatitis due to HCV occurs in up to 90% of patients followed for 5 years.41 Although histological injury in the allograft due to HCV is exceedingly common, disease progression after the development of hepatitis is variable, with some patients experiencing indolent disease and others rapidly progressing to cirrhosis and liver failure. In patents who develop HCV-associated cirrhosis post-transplantation, rapid decompensation is a common occurrence. It has been reported that up to 42% of patients with HCV-associated cirrhosis post-transplantation develop decompensation, manifested as ascites, encephalopathy, or hepatic hydrothorax, and less than 50% of patients survive more than 1 year after the development of decompensation.88 Thus, both prospective and retrospective data have established that the progression of HCV following orthotopic liver transplantation is accelerated when compared to non-transplanted patients. Several lines of anecdotal evidence suggest that HCV recurrence might be more severe in recipients of LDLT when compared to DD recipients. However, recent reports indicate that the overall incidence and time to HCV recurrence are not different when comparing DD with LDLT, and that severe sequelae of HCV recurrence – cholestatic hepatitis, grade III–IV inflammation, and/or HCVinduced graft failure requiring retransplantation – were also not different when comparing DD to LDLT recipients.89
972
At present, both the optimal timing and therapy for the treatment of recurrent HCV following liver transplantation remain inadequately described. Theoretically, eradication of HCV prior to liver transplantation in patients with decompensated liver disease would be beneficial, although, in practice, this strategy has been associated with exacerbation of encephalopathy, infections, and other serious adverse events, particularly in patients treated with high-dose interferon and ribavirin.90 However, initial therapy with low-dose interferon and ribavirin with slow escalation in dose may be associated with improved tolerability and efficacy.91 Following liver transplantation, pre-emptive therapy prior to the development of histological injury or directed therapy after injury occurs have been attempted, with varying degrees of success. It is important to note however that post-transplantation, toxicities of interferon preparations with ribavirin are greater than for non-immunosuppressed patients and responses are lower (Chapter 50). Significant leukopenia and anemia are common and multifactorial, likely due to drug-induced bone marrow suppression and renal insufficiency potentiating ribavirininduced hemolysis.92
SUMMARY Liver transplantation is the treatment of choice for appropriately selected patients with end-stage liver disease. Advances in the past decade have greatly improved outcomes and enhanced survival rates. These advances include improvement in surgical technique and immunosuppression, appropriate selection of donors and recipients, and improvement in therapies to prevent and treat postoperative complications. Successful liver transplantation has led to the emergence of complications associated with patient longevity, including non-hepatic disease, complications of immunosuppression, infections, neoplasia, and recurrence of the primary disease for which the liver transplantation was indicated. It is thus clear that further advancements in patient and graft survival will be associated with enhanced recognition and treatment of long-term complications, particularly the effect of disease recurrence, and cardiovascular and renal complications.
REFERENCES 1. http://www.UNOS.org. 2. Oh CK, Sanfey HA, Pelletier SJ, et al. Related articles, implication of advanced donor age on the outcome of liver transplantation. Clin Transplant 2000; 14:386–390. 3. Rull R, Vidal O, Momblan D, et al. Evaluation of potential liver donors: limits imposed by donor variables in liver transplantation. Liver Transpl 2003; 9:389–393. 4. Busuttil RW, Tanaka K. The utility of marginal donors in liver transplantation. Liver Transpl 2003; 9:651–663. 5. Cescon M, Grazi GL, Ercolani G, et al. Long-term survival of recipients of liver grafts from donors older than 80 years: is it achievable? Liver Transpl 2003; 9:1174–1180. 6. Zhao Y, Lo CM, Liu CL, Fan ST. Use of elderly donors (> 60 years) for liver transplantation. Asian J Surg 2004; 27:114–119. 7. Machicao VI, Bonatti H, Krishna M, et al. Donor age affects fibrosis progression and graft survival after liver transplantation for hepatitis C. Transplantation 2004; 77:84–92. 8. Russo MW, Brown RS Jr. Adult living donor liver transplantation. Am J Transplant 2004; 4:458–465.
Chapter 51 POST LIVER TRANSPLANTATION MANAGEMENT AND COMPLICATIONS
9. Brown RS Jr, Russo MW, Lai M, et al. A survey of liver transplantation from living adult donors in the United States. N Engl J Med 2003; 348:818–825. 10. Marcos A, Fisher RA, Ham JM, et al. Liver regeneration and function in donor and recipient after right lobe adult to adult living donor liver transplantation. Transplantation 2000; 69:1375–1379. 11. Shiffman ML, Brown RS Jr, Olthoff KM, et al. Living donor liver transplantation: summary of a conference at the National Institutes of Health. Liver Transpl 2002; 8:174–188. 12. Emond JC, Renz JF, Ferrell LD, et al. Functional analysis of grafts from living donors. Implications for the treatment of older recipients. Ann Surg 1996; 224:544–552. 13. Otero A, Gomez-Gutierrez M, Suarez F, et al. Liver transplantation from Maastricht category 2 non-heart-beating donors: a source to increase the donor pool? Transplant Proc 2004; 36:747–750. 14. Abt PL, Desai NM, Crawford MD, et al. Survival following liver transplantation from non-heart-beating donors. Ann Surg 2004; 239:87–92. 15. Reich DJ, Munoz SJ, Rothstein KD, et al. Controlled non-heartbeating donor liver transplantation: a successful single center experience, with topic update. Transplantation 2000; 70:1159–1166. 16. Tisone G, Manzia TM, Zazza S, et al. Marginal donors in liver transplantation. Transplant Proc 2004; 36:525–536. 17. Montalti R, Nardo B, Bertelli R, et al. Donor pool expansion in liver transplantation. Transplant Proc 2004; 36:520–522. 18. Verran D, Kusyk T, Painter D, et al. Clinical experience gained from the use of 120 steatotic donor livers for orthotopic liver transplantation. Liver Transpl 2003; 9:500–505. 19. Urena MA, Moreno Gonzalez E, Romero CJ, et al. An approach to the rational use of steatotic donor livers in liver transplantation. Hepatogastroenterology 1999; 46:1164–1173. 20. Arenas JI, Vargas HE, Rakela J. The use of hepatitis C-infected grafts in liver transplantation. Liver Transpl 2003; 9:S48–S51. 21. Saab S, Chang AJ, Comulada S, et al. Outcomes of hepatitis Cand hepatitis B core antibody-positive grafts in orthotopic liver transplantation. Liver Transpl 2003; 9:1053–1061. 22. Kamath PS, Wiesner RH, Malinchoc M, et al. A model to predict survival in patients with end-stage liver disease. Hepatology 2001; 33:464–470. 23. Freeman RB, Wiesner RH, Edwards E, et al. United Network for Organ Sharing Organ Procurement and Transplantation Network Liver and Transplantation Committee. Results of the first year of the new liver allocation plan. Liver Transpl 2004; 10:7–15. 24. Onaca NN, Levy MF, Sanchez EQ, et al. A correlation between the pretransplantation MELD score and mortality in the first two years after liver transplantation. Liver Transpl 2003; 9:117–123. 25. Jain A, Kashyap R, Marsh W, et al. Reasons for long-term use of steroid in primary adult liver transplantation under tacrolimus. Transplantation 2001; 71:1102–1106. 26. Conti F, Morelon E, Calmus Y. Immuosuppressive therapy in liver transplantation. J Hepatol 2003; 39:664–678. 27. Tector AJ, Fridell JA, Mangus RS, et al. Promising early results with immunosuppression using rabbit anti-thymocyte globulin and steroids with delayed introduction of tacrolimus in adult liver transplant recipients. Liver Transpl 2004; 10:404–407. 28. Liu CL, Fan ST, Lo CM, et al. Interleukin 2 receptor antibody (basiliximab) for immunosuppressive induction therapy after liver transplantation: a protocol with early elimination of steroids and reduction of tacrolimus dosage. Liver Transpl 2004; 10:728–733. 29. Lilly LB, Grant D. Optimization of ciclosporin for liver transplantation. Transplant Proc 2004; 36:267S–270S. 30. Herrero JI, Quiroga J, Sangro B, et al. Conversion of liver transplant recipients on ciclosporin with renal impairment to mycophenolate mofetil. Liver Transpl Surg 1999; 5:414.
31. Trotter JF. Sirolimus in liver transplantation. Transplant Proc 2003; 35:193–200. 32. Nair S, Eason J, Loss G. Sirolimus monotherapy in nephrotoxicity due to calcineurin inhibitors in liver transplant recipients. Liver Transpl 2003; 9:126–129. 33. Schemmer P, Mehrabi A, Kraus T, et al. New aspects on reperfusion injury to liver – impact of organ harvest. Nephrol Dial Transplant 2004; 19:26–35. 34. Sieders E, Peeters PM, TenVergert EM, et al. Early vascular complications after pediatric liver transplantation. Liver Transpl 2000; 6:326–332. 35. Bhattacharjya S, Gunson BK, Mirza DF, et al. Delayed hepatic artery thrombosis in adult orthotopic liver transplantation – a 12-year experience. Transplantation 2001; 71:1592–1596. 36. Denys A, Chevallier P, Doenz F, et al. Interventional radiology in the management of complications after liver transplantation. Eur Radiol 2004; 14:431–439. 37. Vignali C, Cioni R, Petruzzi P, et al. Role of interventional radiology in the management of vascular complications after liver transplantation. Transplant Proc 2004; 36:552–554. 38. Selzner N, Rudiger H, Graf R, Clavien PA. Protective strategies against ischemic injury of the liver. Gastroenterology 2003; 125:917–936. 39. Clavien PA, Yadav S, Sindram D, Bentley RC. Protective effects of ischemic preconditioning for liver resection performed under inflow occlusion in humans. Ann Surg 2000; 232:155– 162. 40. Lefkowitch JH. Diagnostic issues in liver transplantation pathology. Clin Liver Dis 2002; 6:555–570. 41. Berenguer M. Natural history of recurrent hepatitis C. Liver Transpl 2002; 8:S14–S18. 42. Desai D, Desai N, Nightingale P, et al. Carriage of methicillinresistant Staphylococcus aureus is associated with an increased risk of infection after liver transplantation. Liver Transpl 2003; 9:754–759. 43. Singh N, Gayowski T, Rihs JD, et al. Evolving trends in multipleantibiotic-resistant bacteria in liver transplant recipients: a longitudinal study of antimicrobial susceptibility patterns. Liver Transpl 2001; 7:22–26. 44. Singh N. Fungal infections in the recipients of solid organ transplantation. Infect Dis Clin North Am 2003; 17:113–134. 45. Fondevila C, Ghobrial RM, Fuster J, et al. Biliary complications after adult living donor liver transplantation. Transplant Proc 2003; 35:1902–1903. 46. Gopal DV, Pfau PR, Lucey MR. Endoscopic management of biliary complications after orthotopic liver transplantation. Curr Treat Options Gastroenterol 2003; 6:509–515. 47. van der Bij W, Speich R. Management of cytomegalovirus infection and disease after solid-organ transplantation. Clin Infect Dis 2001; 33:S32–S37. 48. Razonable RR, Burak KW, van Cruijsen H, et al. The pathogenesis of hepatitis C virus is influenced by cytomegalovirus. Clin Infect Dis 2002; 35:974–981. 49. Sia IG, Wilson JA, Groettum CM, et al. Cytomegalovirus (CMV) DNA load predicts relapsing CMV infection after solid organ transplantation. J Infect Dis 2000; 181:717–720. 50. Norris S, Kosar Y, Donaldson N, et al. Cytomegalovirus infection after liver transplantation: viral load as a guide to treating clinical infection. Transplantation 2002; 74:527–531. 51. Textor SC, Taler SJ, Canznello VJ, et al. Post transplantation hypertension related to calcineurin inhibitors. Liver Transpl 2000; 6:521–530. 52. Risaliti A, Baccarani U, Vianello V, et al. Cardiovascular and metabolic complications after liver transplantation: Neoral- versus tacrolimus-based immunosuppression. Transplant Proc 2001; 33:3684–3685. 53. Charco R, Bilbao I, Chavez R, et al. Low incidence of hypercholesterolemia among liver transplant patients under
973
Section VIII. Liver Transplantation
54.
55.
56.
57.
58.
59.
60.
61. 62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72. 73.
974
tacrolimus monotherapy immunosuppression. Transplant Proc 2002; 34:1555–1556. Neal DA, Alexander GJ. Can the potential benefits of statins in general medical practice be extrapolated to liver transplantation? Liver Transpl 2001; 7:1009–1014. Mehra MR, Uber PA, Vivekananthan K, et al. Comparative beneficial effects of simvastatin and pravastatin on cardiac allograft rejection and survival. J Am Coll Cardiol 2002; 40:1609–1614. Sheiner PA, Magliocca JF, Bodian CA, et al. Long-term medical complications in patients surviving > 5 years after liver transplant. Transplantation 2000; 69:781–789. Yoo HY, Thuluvath PJ. The effect of insulin-dependent diabetes mellitus on outcome of liver transplantation. Transplantation 2002; 74:1007–1012. Stockmann M, Steinmuller T, Nolting S, Neuhaus P. Posttransplant diabetes mellitus after orthotopic liver transplantation. Transplant Proc 2002; 34:1571–1572. Risaliti A, Baccarani U, Vianello V, et al. Cardiovascular and metabolic complications after liver transplantation: Neoral- versus tacrolimus-based immunosuppression. Transplant Proc 2001; 33:3684–3685. Varo E, Padin E, Otero E, et al. Cardiovascular risk factors in liver allograft recipients: relationship with immunosuppressive therapy. Transplant Proc 2002; 34:1553–1554. Reuben A. Long-term management of the liver transplant patient: diabetes, hyperlipidemia, and obesity Liver Transpl 2001; 7:S13–S21. Yoo HY, Molmenti E, Thuluvath PJ. The effect of donor body mass index on primary graft nonfunction, retransplantation rate, and early graft and patient survival after liver transplantation. Liver Transpl 2003; 9:72–78. Nair S, Verma S, Thuluvath PJ. Obesity and its effect on survival in patients undergoing orthotopic liver transplantation in the United States. Hepatology 2002; 35:105–109. Johnston SD, Morris JK, Cramb R, et al. Cardiovascular morbidity and mortality after orthotopic liver transplantation. Transplantation 2002; 73:901–906. Moller S, Henriksen JH. Cirrhotic cardiomyopathy: a pathophysiological review of circulatory dysfunction in liver disease. Heart 2002; 87:9–15. Cohen AJ, Stegall MD, Rosen CB, et al. Chronic renal dysfunction late after liver transplantation. Liver Transpl 2002; 8:916–921. Fisher NC, Nightingale PG, Gunson BK, et al. Chronic renal failure following liver transplantation. Transplantation 1998; 66:59–66. Moreno JM, Rubio E, Pons F, et al. Usefulness of mycophenolate mofetil in patients with chronic renal insufficiency after liver transplantation. Transplant Proc 2003; 35:715–717. Demetris A, Adams D, Bellamy C, et al. Update of the international Banff schema for liver allograft rejection: working recommendations for the histopathologic staging and reporting of chronic rejection. Hepatology 2000; 31:792–799. Singh N, Avery RK, Munoz P, et al. Trends in risk profiles for and mortality associated with invasive aspergillosis among liver transplant recipients. Clin Infect Dis 2003; 36:46–52. Wu G, Vilchez RA, Eidelman B, et al. Cryptococcal meningitis: an analysis among 5521 consecutive organ transplant recipients. Transpl Infect Dis 2002; 4:183–188. Holmes RD, Sokol RJ. Epstein–Barr virus and post-transplant lymphoproliferative disease. Pediatr Transplant 2002; 6:456–464. Bueno J, Ramil C, Somoza I, et al. Treatment of monomorphic B-cell lymphoma with rituximab after liver transplantation in a child. Pediatr Transplant 2003; 7:153–156.
74. Smets F, Sokal EM. Epstein–Barr virus-related lymphoproliferation in children after liver transplant: role of immunity, diagnosis, and management. Pediatr Transplant 2002; 6:280–287. 75. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334:693–699. 76. Valero JM, Rubio E, Moreno JM, et al. De novo malignancies in liver transplantation. Transplant Proc 2003; 35:709–711. 77. Mithoefer AB, Supran S, Freeman RB. Risk factors associated with the development of skin cancer after liver transplantation. Liver Transpl 2002; 8:939–944. 78. Saigal S, Norris S, Muiesan P, et al. Evidence of differential risk for posttransplantation malignancy based on pretransplantation cause in patients undergoing liver transplantation. Liver Transpl 2002; 8:482–487. 79. Atassi T, Thuluvath PJ. Risk of colorectal adenoma in liver transplant recipients compared to immunocompetent control population undergoing routine screening colonoscopy. J Clin Gastroenterol 2003; 37:72–73. 80. Trotter JF, Brazer SR. Rapid progression to high-grade dysplasia in Barrett’s esophagus after liver transplantation. Liver Transpl Surg 1999; 5:332–333. 81. Loftus EV Jr, Aguilar HI, Sandborn WJ, et al. Risk of colorectal neoplasia in patients with primary sclerosing cholangitis and ulcerative colitis following orthotopic liver transplantation. Hepatology 1998; 27:685–690. 82. Neuberger J. Recurrent primary biliary cirrhosis. Best Pract Res Clin Gastroenterol 2000; 14:669–680. 83. Wiesner RH. Liver transplantation for primary sclerosing cholangitis: timing, outcome, impact of inflammatory bowel disease and recurrence of disease. Best Pract Res Clin Gastroenterol 2001; 15:667–680. 84. Molmenti EP, Netto GJ, Murray NG, et al. Incidence and recurrence of autoimmune/alloimmune hepatitis in liver transplant recipients. Liver Transpl 2002; 8:519–526. 85. Burke A, Lucey MR. Non-alcoholic fatty liver disease, nonalcoholic steatohepatitis and orthotopic liver transplantation. Am J Transplant 2004; 4:686–693. 86. Mackie J, Groves K, Hoyle A, et al. Orthotopic liver transplantation for alcoholic liver disease: a retrospective analysis of survival, recidivism, and risk factors predisposing to recidivism. Liver Transpl 2001; 7:418–427. 87. Marzan A, Salizzoni M, Debernardi-Venon W, et al. Prevention of hepatitis B virus recurrence after liver transplantation in cirrhotic patients treated with lamivudine and passive immunoprophylaxis. J Hepatol 2001; 34:903–910. 88. Berenguer M, Prieto M, Rayon J, et al. Natural history of clinically compensated hepatitis C virus related graft cirrhosis after liver transplantation. Hepatology 2000; 32:852–858. 89. Gaglio PJ, Malireddy S, Levitt BS, et al. Increased risk of cholestatic hepatitis C in recipients of grafts from living versus cadaveric liver donors. Liver Transpl 2003; 9:1028–1035. 90. Crippin JS, McCashland T, Terrault N, et al. A pilot study of the tolerability and efficacy of antiviral therapy in hepatitis C virusinfected patients awaiting liver transplantation. Liver Transpl 2002; 8:350–355. 91. Everson GT. Treatment of chronic hepatitis C in patients with decompensated cirrhosis. Rev Gastroenterol Disord 2004; 4:S31–S38. 92. Gane E. Treatment of recurrent hepatitis C. Liver Transpl 2002; 8:S28–S37.
Section VIII: Liver Transplantation
MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
52
Marina Berenguer and Teresa L. Wright Abbreviations antiantibody to HBV surface antigen HBsAg HBIg hepatitis B immunoglobulins HBsAg hepatitis B surface antigen HBV hepatitis B virus HCC hepatocellular carcinoma
HCIg HDV HIV HLA IFN LDLT
HCV immunoglobulin hepatitis delta virus human immunodeficiency virus human leukocyte antigen interferon living donor liver transplantation
INTRODUCTION With significant improvements in immunosuppressive therapy and surgical techniques over the past two decades, liver transplantation has become the definitive and effective therapy for patients with end-stage liver disease, with survival rates approaching 90–95% and 65–80% after 1 and 5 years of follow-up, respectively.1 Among several circumstances that may pose a threat to long-term survival, the greatest is likely the recurrence of the original liver disease.2,3 Viral hepatitis is the leading indication for liver transplantation in the majority of transplant centers. Post-transplantation outcome in these patients largely depends on the prevention of allograft reinfection. While hepatitis B recurrence has been effectively contained by the use of hepatitis B immunoglobulin and oral antivirals,2 recurrent hepatitis C is becoming an increasingly challenging problem to the transplant community.3 As patients survive longer and enter their second to third decade post-transplantation, it is likely that allograft failure related to recurrent hepatitis C will become an increasingly serious problem. In this chapter, we will summarize the current knowledge on recurrent viral diseases following liver transplantation, with particular emphasis on the natural history, pathogenesis, and treatment of these conditions.
RECURRENT HEPATITIS B VIRUS (HBV) INFECTION HBV-related liver disease represents 5–10% of liver transplantations in most series.1,2 Due to the recent improvements in the management of HBV infection, post-transplantation outcomes are now excellent, similar to those achieved by patients with cholestatic liver diseases.4,5 Historically, these patients had done poorly with transplantation. The 5-year survival rate was reported to be 50%, compared to 70–85% for patients with alcoholic or cholestatic liver diseases.2 This reduced survival was in large part related to the high rate of HBV recurrence in the absence of specific prophylactic therapies. Indeed, in an early study, the reinfection rate was 80%, resulting in graft loss in more than 70% of patients.2 Efforts to improve the outcome of these patients were predominantly focused upon
MELD MMF ORFs PBMC PEG
model for end-stage liver disease mycophenolate mofetil open reading frames peripheral blood mononuclear cells pegylated
strategies to prevent reinfection. In recent years, several new therapies have become available that have led to increased survival of this patient group, lending a sense of optimism to clinicians caring for patients with this disease. Interestingly, the number of publications related to “HBV and liver transplantation,” an indirect measure of the interest and scientific advances, has significantly increased in the last 5 years, from 30 publications reported in Medline from 1980 to 1991, to 429 in the period 1995–2003. Many of the questions of the early 1990s now have an evidence-based answer, so that the debate has currently shifted from whether liver transplantation is an option for this patient subgroup to selecting the best approaches to prevent reinfection, particularly in the long term.
INDICATIONS FOR LIVER TRANSPLANTATION The indications and contraindications for transplantation in these patients do not differ from those applied to other forms of liver diseases and typically include complications from portal hypertension, liver failure, or the development of hepatocellular carcinoma (HCC).1 Indeed, virally infected patients are at high risk for developing HCC. Cirrhosis itself is considered a premalignant state, irrespective of its etiology, with an annual incidence of HCC of 4–5%, ranging from 1 to 8%.6 This risk is 20 times greater in surface antigen-positive patients than in negative controls. Among HBV patients, the risk is minimal in carriers with normal enzymes, it is low in patients with chronic hepatitis, and is highest in those with cirrhosis. Other factors increasing the risk of HCC in HBV-infected patients include old age, male gender, long duration of infection, coinfection with delta or hepatitis C virus, aflatoxin exposure, and high levels of HBV replication. Although the rate of cancer recurrence after liver transplantation was very high in early series, more promising results have recently been obtained with accurate staging of tumors and improved patient selection.7 Careful expansion of traditional criteria can be attempted within experimental boundaries.
NATURAL HISTORY HBV infection post-transplantation typically results from the recurrence of an infection present prior to liver transplantation. In the
975
Section VIII. Liver Transplantation
absence of specific prophylactic measures, both type of disease and HBV DNA level before transplantation are the best predictors to assess the risk of recurrence, with the highest rates reported in hepatitis B surface antigen (HBsAg)-positive cirrhotic patients with evidence of active viral replication (HBV DNA and/or HBeAg-positive) and the lowest in those without detectable HBeAg or HBV DNA, those with fulminant hepatitis, or those coinfected with the delta virus (2-year actuarial risk: 75%, 67%, 17%, and 32%, respectively).2,4,5 Most cases of HBV reinfection occur during the first 3 years post-transplantation. Occasionally, HBV infection post-transplantation is a consequence of de novo infection, despite the use of strict screening measures in blood banks with exclusion of HBsAg-positive and also anti HB corepositive donations (Figure 52-1).8,9 The prevalence of de novo HBV hepatitis ranges from 2 to 8%, and is generally related to transmission from an HBsAg-negative anti-HBc-positive donor. The most significant factor associated with transmission is the serologic status of the recipient, so that the risk is almost null in patients who are antiHBs-positive, minor (@ 10%) in those who are anti-HBs-negative but anti-HBc-positive, and high (@ 50–70%) in those without markers of previous exposure to HBV.8,9 Although there have been reports of severe progression, the natural history of de novo hepatitis B is generally more benign than that described for recurrent hepatitis B. The natural history of recurrent hepatitis B is more aggressive than that observed in the immune-competent population. Two patterns of recurrence have been described. Typically, patients develop acute hepatitis after detection of HBsAg in serum. In these cases there is a significant increase in HBV DNA levels, together with a transaminase rise with mild bilirubin elevation, and features of acute lobular hepatitis on liver histology. Progression to chronic hepatitis and cirrhosis may occur within 2 years of transplantation.2 The second pattern of recurrence, fibrosing cholestatic hepatitis, is an entity initially described in these patients, and later, among HCV-infected recipients. It is characterized histologically by the presence of periportal and perisinusoidal fibrosis, ballooned hepatocytes with cell loss, pronounced cholestasis, and a paucity of inflammatory activ-
ity.2 Immunohistochemical stains show high cytoplasmic expression of viral antigens, which, in conjunction with the lack of inflammatory infiltrate, suggests a direct cytopathic effect of the virus. Clinically, patients present with jaundice, biochemical cholestasis with high bilirubin levels despite mild transaminase elevations, and extremely high levels of serum HBV DNA. The course is rapidly progressive with severe cholestasis, coagulopathy, and liver failure within weeks of onset. High early mortality rate occurs following liver retransplantation and among those surviving the postoperative period, an even more aggressive course of recurrent disease develops.2 Patients at risk for this syndrome include those with high levels of viremia pretransplantation and those infected with precore mutants. With currently available effective therapies, this syndrome is rarely, if ever, seen in patients undergoing liver transplantation for HBV disease.
PATHOGENESIS The mechanisms by which HBV leads to liver injury following liver transplantation are incompletely understood. Increased levels of HBV replication are typically observed, likely related to the use of immuosuppressive drugs, particularly corticosteroids. This enhanced replication with excess production of viral proteins in conjunction with the altered host immune responsiveness probably contributes to the pathogenesis of liver damage.2
PREVENTION OF HBV GRAFT REINFECTION (Figure 52-2) Early Prophylaxis Passive Immunization with High-Dose Hepatitis B Immunoglobulins (HBIg) HBIg consists of polyclonal antibodies directed against the viral envelope, and was originally derived from donors positive for antibody to HBV surface antigen (anti-HBsAg). The presumed mechanism of action of this antibody is to neutralize circulating virus by binding to the viral envelope. Empirical application of HBIg aimed at maintaining serum anti-HBs titers above 100 IU/l was initially shown to reduce the rate of viral recurrence. This was further con-
LT in the post-HBIg/ pre-lamivudine era risk. 25% Preemptive LT in the pre-HBIg/ pre-lamivudine era risk. 80%
LT with lamivudine monotherapy risk. 25%
Populations for treatment of post liver transplantation HBV LT in the post-HBIg/ post-lamivudine era risk. 5%
de novo infection: peri-LT acquisition post -LT reactivation
Figure 52-1. Populations for treatment of post-transplantation hepatitis B virus (HBV) disease. Recurrent HBV disease is uncommon, but five different groups with HBV disease of the allograft exist. Management of these populations depends on their prior treatment exposure and the presence of treatment-associated HBV variants. LT, liver transplantation.
976
Therapeutic
Before histologic recurrence
Before OLT
OLT
After histologic recurrence
Recurrent hepatitis
HBV Nucleoside analogues
HBIg 6 Lamivudine
Nucleoside analogs
Interferonribavirin
Interferonribavirin
Interferonribavirin
HCV Figure 52-2. Approaches to the prevention and treatment of HBV and HCV infection in the setting of liver transplantation. OLT, orthotopic liver transplantation.
Chapter 52 MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
firmed in a European multicenter study2 where the administration of HBIg for more than 6 months was shown to reduce dramatically the rate of HBV recurrence to a median rate of 20–30% after 2 years.2,4,5 Recurrence was 75% in patients receiving no or short-term HBIg versus 33% in those receiving long-term HBIg (P < 0.001).2 Long-term administration of HBIg reduced the rate of recurrence in patients with fulminant HBV hepatitis to less than 10%, in HDV coinfected patients to 10–15%, and in HBV DNA-negative cirrhotic patients to less than 30%. Yet, it did not reduce the rate of recurrence in patients with HBV DNA-positive HBV cirrhosis.2 More recent studies have further confirmed the efficacy of long-term HBIg in patients without active HBV replication, with recurrence ranging from 17 to 38% at 2 years when administered to reach titers over 100 UI/l, but reduced efficacy in those in whom HBV DNA is detected prior to transplantation by hybridization methods (>105–106 copies/ml) with rates of recurrence ranging from 70 to 96%.2,4,5 In a recent analysis of the long-term results of HBIg administration in 271 patients, the actuarial rate of HBV recurrence was 26.8%, 34.6%, 40%, and 41.6% at 1, 2, 5, and 10 years, respectively.5 Substantial differences were again observed between patients who were HBV DNA-positive and those who were negative pretransplantation with 5-year rates of 76.7% and 33%, respectively. The adverse prognostic characteristic of active viral replication pretransplantation may be overcome with more aggressive use of HBIg maintaining titers over 500 IU/l and, particularly, by the use of HBIg in combination with oral antivirals pre- and post-transplantation (see later).10 With the first alternative, at least during the first 6 months, recurrence in HBV DNA-positive patients may be reduced to approximately 16–35%. Various regimens have been described, with most including the administration of 10 000 IU HBIg intravenously during the anhepatic phase and 10 000 IU HBIg daily for the first week posttransplantation. The subsequent dosing is either given on a fixed schedule (generally on a monthly basis) or based on anti-HBs titers (readministration when antiHBs is less than 100–500 IU/l).2,10 Due to the high number of variables, including risk of recurrence, time from transplantation, and use of antivirals, the best target value for anti-HBs level and whether this target level can be reduced over
time in an individual patient are still a matter of debate. It is generally accepted though that HBIg should be given so as to obtain anti-HBs titers greater than 500 IU/l during the first week after transplantation, greater than 250 IU/l between days 8, and 90 and greater than 100 IU/l thereafter.2,10 Due to the good results, lifelong passive immunization with HBIg has been used in most transplant centers and considered, until very recently, the “standard of care.” However, HBIg has several drawbacks, including the cost, the need for parenteral and long-term administration, the need for close monitoring of anti-HBs levels, the issue of availability, the potential for breakthrough, and the reduced efficacy in patients with viral replication. Causes of breakthrough are multifactorial and include inadequate anti-HBs titers following transplantation, HBV overproduction coming from extrahepatic sites, and/or mutations in the region of the surface gene of the HBV genome which encodes the “a” determinant region, the putative region for antibody binding. While the first two circumstances appear to play a major role in the early post-transplant period, viral mutations are probably the major cause of HBIg failure in the long term.10 The most common mutation is a substitution of glycine for arginine at amino acid position 145. Discontinuation of HBIg results in reversion of the mutations to the wild-type virus in the majority of patients. One concerning limitation is the difficulty in discontinuing this product in the long term. Recurrent infection has been documented in patients stopping prophylaxis with HBIg after 1 year or more. Furthermore, HBV DNA has been detected by highly sensitive molecular techniques in the serum, liver, and peripheral blood mononuclear cells (PBMC) of HBsAg-negative patients on HBIg prophylaxis,5 suggesting that indefinite treatment is required.
Pretransplantation Therapy with Oral Antivirals Several alternatives are being evaluated to overcome the limitations of HBIg (Table 52-1). The first one is the use of antiviral treatment prior to transplantation to inhibit viral replication.10–12 Due to the risk of worsening hepatic decompensation and low tolerability of interferon in patients with decompensated liver disease, interferon is not recommended in this situation. Nucleoside analogues, though, have
Table 52-1. Prevention of Recurrent Hepatitis B: Alternatives to High Intravenous Doses of Hepatitis B Immunoglobulin (HBIg) Type and timing of intervention
End-point
Available drugs
Potential problems
Pretransplantation antiviral therapy
Decrease viral replication
• •
• •
Pre-emptive post-transplantation antiviral therapy Pre-emptive post-transplantation antiviral therapy in combination with: • High doses of HBIg • Low doses of HBIg HBIg ± oral antivirals in combination with post-transplantation vaccination against HBV
Prevent HBV recurrent infection Prevent HBV recurrent infection
Nucleos(t)ide analogues
Prevent HBV recurrent infection
Interferon ? Nucleos(t)ide analogues
HBIg in combination with nucleos(t)ide analogues
Double-dose vaccination following discontinuation (versus alternating) of HBIg ± oral antivirals
Tolerance Development of resistant mutants Development of resistant mutants/low efficacy • Same as with high-dose HBIg in monotherapy • Development of resistant mutants Failure of vaccination regimen
HBV, hepatitis B virus.
977
Section VIII. Liver Transplantation
Figure 52-3. Pretransplantation therapy with lamivudine: development of lamivudine-resistant mutants. LAM, lamivudine; R, resistance. Mean follow-up 12 months.
100 Median follow-up: 12 months (3–29)
90 80
Villeneuve 00 (n=35) Perrillo 01 (n=30)
70
Yao 01 (n=23) 60 Seehofer 01 (n=17) 50
Marzano 01 (n=33)
40
Fontana 02 (n=162)
30
Andreone 02 (n=25)
20
Fontana 02 (n=154)
10 0 DNA (–) with LAM
Viral R
a potent antiviral effect, inducing a rapid suppression of HBV DNA in serum, are very well tolerated, orally administered, and, in contrast to interferon, do not precipitate worsening of liver function in patients with advanced disease. Recent experience with these drugs in cirrhotic patients awaiting liver transplantation is encouraging.13–19 With lamivudine (100 mg daily), HBV replication is decreased below detection of hybridization assays in 62.5–100% of treated cirrhotic patients, including both those infected with wild-type virus and the e-minus strain of HBV, allowing liver transplantation to be performed in conditions of low risk of recurrence (Figure 52-3). An additional benefit that may be obtained by some, but not all, patients with decompensated cirrhosis is an improvement in the hepatic synthetic function. Clinical improvement and stabilization of hepatic function are slow and gradual, being more apparent after 6 months of therapy. Although this clinical improvement may be achieved by a subgroup of cirrhotic patients, it is less likely in those with severe hepatic insufficiency, including those with increased serum bilirubin and creatinine levels and elevated Child–Pugh and model for end-stage liver disease (MELD) scores.20 Since progression of the disease and even death tend to occur early after the initiation of therapy, generally within the first 6 months, patients with the above characteristics, who most likely have presented late in their disease course, should be prioritized for urgent liver transplantation, irrespective of the antiviral response to lamivudine. Since viremia recurs in more than 80% of patients following treatment discontinuation, circumstances that may result in a flare of liver disease, treatment should be administered indefinitely in these patients. The major drawback of prolonged lamivudine therapy is the selection of drug-resistant mutants with HBV DNA reappearance (Figure 52-3). Mutations typically occur in the YMDD motif of the HBV DNA polymerase gene. This risk increases significantly after 6 months of therapy, reaching 27% after 1 year. While viremia may be lower in patients with YMDD mutations than before therapy because of the decreased replication fitness of the mutants, flares of liver disease with worsening of liver disease may occur.11,21 In addition, the selection of lamivudine-resistant mutants may increase the risk of HBV recurrence despite the use of high doses of HBIg plus lamivudine post-transplantation.22–25 Adefovir dipivoxil (10 mg daily,
978
with dose reductions based on creatinine clearance) is a potent nucleotide analogue that has been shown to suppress viral replication of the wild-type virus, the e-minus strain, and the lamivudineor famciclovir-resistant mutants. In cirrhotic patients who have failed lamivudine, adefovir leads to a significant reduction of HBV DNA levels and normalization of transaminase levels in 61% of the patients.11 It may either be used as salvage therapy or as a primary option. Recent data from 128 patients with decompensated cirrhosis treated with adefovir demonstrated that this drug is safe and effective in treating lamivudine-associated breakthrough, significantly suppressing serum HBV DNA levels. Clinical improvement was also achieved in a proportion of patients with improvement in serum bilirubin, albumin, and prothrombin time.26 The one concern with this drug is the potential for renal toxicity. Since resistance to adefovir is extremely low (2% after 2 years of continuous use), patients in whom a significant clinical improvement is achieved may even be removed from the waiting list. Tenofovir disoproxil fumarate, structurally similar to adefovir and approved for treatment of human immunodeficiency virus (HIV) is also effective in suppressing replication of YMDD mutants, and, interestingly, appears to have less renal toxicity than associated with adefovir. Data on this drug in the liver transplant setting are still lacking. The best posttransplantation prophylaxis in patients with lamivudine-resistant mutants is at present unknown, but probably should be based on triple therapy. Interestingly, adefovir-resistant variants are sensitive to lamivudine,27 providing scientific rationale for combination therapy with both of these agents in treating patients with advanced liver disease awaiting transplantation or those following transplantation who are infected with lamivudine-resistant strains.
Post-Transplantation Therapy with Oral Antivirals Once liver transplantation is performed, there are several alternatives to long-term HBIg (Table 52-1). The first is to continue the pre-emptive therapy with lamivudine which was begun prior to transplantation.28,29 Although this approach is initially effective and patient compliance is good given the few side effects of this drug, therapy is limited by the emergence of HBV mutants with prolonged treatment, required to avoid rebound of viral replication once
Chapter 52 MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
Table 52-2. Prevention of Recurrent Hepatitis B: Combination with Hepatitis B Immunoglobulin (HBIg) and Lamivudine % DNA + pre-OLT
Reference
Duration of pre-OLT treatment with lamivudine
Markowitz 199830 (n = 14) Yao 199931 (n = 10) McCaughan 1999 (n = 9) Yoshida 199932 (n = 7)
3 (0.7–7.8)
7%
8.6 (1–22)
20%
0
NA
Angus 200033 (n = 32) Marzano 200134 (n = 33) Rosenau 200124 (n = 21) Han 200135 (n = 59)
60% on lamivudine (duration NA) 3.2
0
50%
4.6 (0.6–14.1)
0
4.6 (0.06–14.1)
24%
NA
NA
HBIg regimen
Follow-up (months)
Standard high IV (10 000 day 0–7, then weekly during the first month, then monthly) 10 000 IU IV day 0 (7 days if DNA +), 1111 U i.m./weekly for 3 weeks, 1111 U i.m. every 3 weeks 4000 IU i.m. (daily first week, weekly for 3 weeks, then monthly) 2170 IU i.m. for 14 days, then twice weekly with progressive reduction until once monthly at 1 year post-OLT 400–800 IU i.m. daily the first week and then monthly
13
Recurrence rate 0
15.6
10%
15.6
0
17
0
18.5 (5–45)
3%
46 500 IU IV first month, 5000 IU IV monthly
30 ± 8
4%
45 000 IU IV first week then reinjection to maintain anti-HBs ≥ 500IU/l until day 14, then ≥ 200 IU/l Standard high IV (10 000 day 0–7, then monthly)
21 (2.4–49.1)
9.5%
15 (1–61.8)
0
OLT, orthotopic liver transplantation; HBIg, hepatitis B immunoglobulin; IV, intravenous; i.m., intramuscular.
therapy is stopped. Using this approach, the rate of recurrence is as high as 60%, mainly in patients with high levels of viral replication before treatment initiation.28 It may be sufficient though for nonreplicating patients. Due to the low risk of developing resistant mutants, primary adefovir monotherapy pre- and post-transplantation is an attractive alternative that is currently under evaluation. The second alternative, since treatment failures occur with both HBIg and with lamivudine monotherapy, is the use of combination therapy with HBIg and nucleoside analogues. It is the most promising alternative and is becoming the standard of care in most transplant programs.2,10,11 The advantages over a single agent are the following: (1) possibility of administering lower doses of HBIg (400–2000 IU/monthly) which then leads to a significant reduction in cost; (2) potential reduction of development of resistant mutants, which is a frequent event when each drug is given as a single agent; and (3) synergistic effect with failure rates lower than 10–12% in most series. The higher rates of recurrence are typically found in patients who have developed lamivudine resistance prior to transplantation. The best protocol is still unknown since doses, routes, type, and lengths of administrations vary substantially from center to center. In a preliminary report, lamivudine in combination with high intravenous doses of HBIg was shown to be safe and highly effective, but equally expensive.30 Subsequent studies have explored modifications of this approach using lower doses of HBIg, simpler routes of administration, and shorter durations of therapies31–36 (Table 52-2).
Long-Term Prophylaxis In the long term, two approaches have been investigated, particularly in patients at low spontaneous risk of HBV recurrence: HBIg discontinuation followed by monotherapy with oral antivirals and HBV vaccination. In a recent long-term study, it was shown that almost 91% of recurrences occurred within the first 2 years of transplantation and only 3% after the fifth year,5 hence raising the issue of HBIg discontinuation in the long term.
Active HBsAg Vaccination Post-transplantation vaccination against HBV may be attempted after an initial period of HBIg ± oral antivirals. In a pilot study, HBIg was discontinued in 17 low-risk selected liver transplant recipients after a median of 25 months (19–68 months) from transplantation and a double-dose recombinant HBV vaccine was administered at 0, 1, and 6 months.37 Seroconversion to anti-HBs was obtained in 82% of cases. In an update of this study, seroconversion rates to antiHBs (antiHBs titers higher than 10 IU/l) were less impressive, occurring in 64% of 22 patients.38 Disparate results were however reported in a second study,39 with a seroconversion rate of only 23% despite the use of a reinforced triple course of hepatitis B vaccination. The main differences between these two studies include the study population and the concomitant use of lamivudine following HBIg discontination (100% versus 20% in the first study). In fact, the use of lamivudine was found to be a predictive marker of failure to attain seroconversion in the first study (Table 52-3). More recently, a third group has reported the results of HBV vaccination using a more immunogenic vaccine with promising results.40 In that study, 20 patients received repeated doses of recombinant HBV vaccine (20 mg) in combination with a new adjuvant, monophosphoryl lipid A. Vaccination was performed at least 2 days before HBIg administration and titers of anti-HBs concentrations were determined before HBIg injections. HBIg was then discontinued whenever levels were >500 IU/l. Sixteen out of the 20 patients developed anti-HBs levels ranging from 721 to 83 121 IU/l, thus allowing HBIg withdrawal. After 13.5 months (range 6–22 months) of follow-up, all responders had serum anti-HBs concentrations >900 IU/l. If these results are further confirmed, HBV vaccination will enable a substantial proportion of patients now on HBIg to develop a sustained antibody response without the need for continuous passive immunoprophylaxis. This will have major impacts on costs and quality of life. However, several aspects need to be investigated further, including the best vaccine, the definition of protective anti-HBs titers, the amount of HBsAg in each dose, the number
979
Section VIII. Liver Transplantation
Table 52-3. Prevention of Hepatitis B Virus (HBV) Recurrence using HBV Vaccination Following Liver Transplantation Sanchez-Fueyo 200238 (n = 22)
Angelico 200239 (n = 17)
Bienzle 200340 (n = 20)
8/14
0/17
2/18
14/8
17/0
16/4
33 (18–76) months 1–4 weeks
48 (25–85) months 4.5 months
Lamivudine post HBIg D/C
20%
100%
Dose of HBV vaccine
Double at 0-1-6 months
Double at 0-1-2 months
Anti-HBs > 10 IU/l Anti-HBs > 100 IU/l Anti-HBs > 500 IU/l Maximum anti-HBs IU/l in responders HBV recurrence Follow-up after immunization
63% 23% 9% 47 (10–1000)
18% 12% 6% 258 (10–601)
78 (24–156) months Patients maintained on HBIg through HBV vaccination 4 patients also on oral antiviral therapy while vaccination More immunogenic vaccine at weeks 0, 2 , 4, 16, and 18 following HBIg administration 100% 100% 80% 25 344 (1255–83 121)
0% 41 (31–85) months
0% 66 (25–88) months
0% 13.5 (6–22) months
Cause of LT indication (acute/chronic) Immunosuppression (mono/combination) Time from LT to vaccination Time from HBIg D/C to vaccination
LT, liver transplantation; HBIg, hepatitis B immunoglobulin; D/C, discontinuation.
of doses, whether target titers should be the same for different subsets of patients or not, and finally, the necessity for boosting to maintain protective titers. Evaluation of this approach in high-risk patients should also be investigated.
HBIg Substitution with Lamivudine in the Long Term Successful results after a short follow-up have already been reported in low-risk patients.41,42 In the two reported series, only two recurrences were confirmed among 26 patients switched from HBIg ± lamivudine to lamivudine monotherapy after 1–6 months from transplantation. Longer follow-up is needed to determine the incidence of lamivudine-resistant mutants and the efficacy of this approach in high-risk patients. In these and other long-term studies, it has become apparent that, in a substantial subset of patients (up to 45% at 10 years), HBV DNA continues to be detected in serum, liver, or PBMC by polymerase chain reaction (PCR)-based methods.5 Yet the clinical significance of these findings in both the short and long term are unclear since patients are typically asymptomatic with normal liver enzymes and are HBsAg-negative in serum. These findings raise several issues, particularly the indefinite risk of graft reinfection, at least in some patients, and hence the need for the indefinite use of some type of prophylaxis. Determining in whom prophylaxis can be safely stopped may prove to be a difficult task, relying on sensitive PCR techniques to detect HBV DNA in serum, PBMC, and liver.
TREATMENT OF HBV DISEASE OF THE GRAFT Nucleoside analogues are the cornerstone of therapy due to their potent antiviral effect and lack of side effects.11,43 The need for continuous treatment and resistance remain the main limitations. The selection of the antiviral is likely dependent on the category of
980
patient (Figure 52-1). In those who have undergone liver transplantation in the pre-HBIg and/or lamivudine era or those with apparent de novo HBV acquisition, all known antivirals are potential good candidates. In contrast, for those who have undergone surgery in the post-HBIg/lamivudine era and who have broken through, new antivirals such as adefovir or tenofovir that have activity against resistant variants may be best options. Lamivudine is the most widely used nucleoside analogue with good tolerance, rapid loss of HBV DNA in serum in a substantial proportion of patients (60%), “e” seroconversion (30%), and histologic improvement. Adefovir has resulted in viral suppression of nucleoside analogue-resistant variants.11,44
EMERGENCE OF NUCLEOSIDE ANALOGUE RESISTANCE Monotherapy with lamivudine has resulted in the emergence of HBV-resistant variants. This resistance generally occurs after prolonged therapy (more than 6 months) and is associated with a rise in serum HBV DNA and alanine aminotransference levels, indicating a breakthrough in therapy. Molecular analysis has shown changes in the gene for the viral DNA polymerase. Because of the overlapping nature of the HBV open reading frames (ORFs), nucleotide changes in the polymerase may result in amino acid changes not only in the polymerase protein but also in the surface protein, which could in turn theoretically alter binding of HBIg.2,10 When lamivudine is stopped, the wild-type variant re-emerges as the dominant viral population, but retreatment is again associated with the development of resistant mutants at an accelerated rate. Although some cases of histological and clinical deterioration have been reported when drug-resistant mutants develop, these are not consistently associated with hepatic disease progression. The molecular mecha-
Chapter 52 MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
nisms associated with severe recurrence may be a drug-dependent enhanced replication of lamivudine-resistant HBV mutants.45
PREVENTION AND TREATMENT OF DE NOVO HBV INFECTION In order to avoid de novo HBV infection, two complementary approaches may be undertaken: 1. HBV vaccination prior to liver transplantation of all anti-HBsnegative candidates. Unfortunately, as with other immunosuppressed populations, results have been disappointing, with response rates barely reaching 40%;9 2. anti-HBc determination of the donor with use of organs from anti-HBc-positive donors only in recipients already infected with HBV. In order to obtain maximum benefit from these organs while at the same time reducing the risk of HBV transmission, these organs may be used in special circumstances in recipients not infected with HBV. Prophylaxis is then recommended if the risk of transmission is high, particularly in HBV-naive recipients, but may be obviated in those at low risk of transmission, such as those who are antiHBs/anti-HBcore-positive, antiHBs-positive/anti-HBc-negative, or antiHBs-negative/antiHBc-positive.8,9
LIVING RELATED LIVER TRANSPLANTATION Adequate timing of pretransplantation antiviral therapy is an advantage in this setting. Transplantation can thus be performed when HBV DNA has been cleared in serum but before the development of resistant mutants.
RETRANSPLANTATION The initial results on retransplantation for patients with graft failure due to recurrent hepatitis B were discouraging. Improved outcomes have been achieved with specific interventions, mainly with the use of aggressive immunoprophylaxis in combination with pre- and postretransplantation antiviral therapy.46
TRANSPLANTATION IN PATIENTS COINFECTED WITH HEPATITIS DELTA INFECTION Patients with HBV and hepatitis delta virus (HDV) infection are at low risk of viral recurrence.2,4,5 Delta infection is an uncommon indication for transplantation. In the absence of HBIg, both HBV and delta virus can infect the graft, but HDV is not pathogenic until HBV replication also occurs. Current recommendations for HBV apply to HBV–HDV-coinfected liver transplant candidates.
CONCLUSION HBV-related end-stage liver disease is an excellent indication for liver transplantation. Recurrence is effectively prevented with current therapies, and strategies can be tailored individually based on risk of recurrence. In low-risk patients, that is, cirrhotic patients with no evidence of HBV replication (HBV DNA-negative by sensitive hybridization assays), those with fulminant hepatitis, or those coinfected with the delta virus, oral antivirals are not required prior to transplantation, and HBIg in monotherapy may be sufficient initial prophylaxis. In the long term (> 6 months ?), HBIg may be switched to oral antiviral therapy ± HBV vaccination. In contrast, in
high-risk patients, that is, surface antigen-positive cirrhotic patients with evidence of active viral replication, antiviral therapy is needed before transplantation for at least 1 month, and early post-transplantation prophylaxis should be based on combination regimen with HBIg and antivirals. The best HBIg regimen in these patients is still controversial. Whether HBIg may be stopped in the long term is still a matter of debate. HBIg discontinuation could be proposed to patients with undetectable serum HBV DNA by PCR after at least 1–2 years of HBIg, with continuance of lamivudine. Whether adefovir or tenofovir should only be used as salvage therapy or whether these agents should be used as a primary option is also a matter of debate. Certainly, prior to liver transplantation in patients with HBV cirrhosis and detectable HBV DNA, combination therapy with lamivudine plus adefovir or tenofovir seems to make sense, in order to maximize pretransplantation viral suppression and minimize the development of resistance to either agent, resistance that could lead to life-threatening flares of liver disease.
RECURRENT HEPATITIS C VIRUS INFECTION More than 170 million people worldwide are chronically infected by HCV. HCV-associated end-stage liver disease with or without HCC has become the leading diagnosis in patients undergoing liver transplantation.1 In most centers, more than half of transplanted patients are infected with HCV prior to transplantation. It is likely that that this number will increase in future years, mainly as a consequence of the progressive nature of the disease, the inadequate detection of this largely asymptomatic infection, and the lack of effective treatments.47 Viral recurrence defined by the presence of HCV RNA in serum following transplantation occurs universally. Recurrence of chronic hepatitis C is a frequent event, with progression to allograft cirrhosis occurring in a substantial proportion of patients.3 The full consequences of HCV recurrence ultimately result in reduced graft and patient survival compared with patients transplanted for non-viral causes.48,49 There are several factors that may influence the outcome. In fact, it is likely that a different distribution of these factors accounts for the differences in outcome between centers.3,50,51 The need for an effective treatment derives from the potential seriousness of this disease. Unfortunately, to date, there is no suitable intervention to prevent HCV recurrence, and available treatments, interferon and ribavirin, have limited applicability and efficacy in the transplant setting.52 Due to the magnitude of the problem, a consensus conference was recently held to overview the state of the art concerning liver transplantation and HCV disease.53 The conference participants examined the definition of recurrent HCV, the natural history of HCV after liver transplantation, the potential clinical predictors of adverse outcomes, treatment and management strategies, the contribution of different immunosuppressive regimens to outcome, and the role of retransplantation for recurrent HCV in the face of an overall donor shortage.
INDICATION FOR LIVER TRANSPLANTATION Liver transplantation remains the most effective option in patients with decompensated HCV-related cirrhosis. In contrast, traditional
981
Section VIII. Liver Transplantation
medical management is indicated for compensated cirrhosis. A substantial proportion of HCV-infected candidates have a coincidental HCC.1,3,6 Their post-transplantation outcome does not differ from that of patients without the tumor if strict selection criteria are followed.6,7 In some early reports, the presence of HCC was the only variable associated with reduced patient and graft survival,54 possibly due to the high risk of tumor recurrence when no strict tumor criteria were applied. While indications and contraindications for transplantation in HCV-infected patients do not differ from those applied to other forms of liver diseases, post-transplantation outcome is frequently impaired by HCV recurrence. In these circumstances, decompensated patients and patients with HCC should be aggressively managed with traditional medical and/or surgical therapy in an effort to avoid the premature transplantation of some of these patients, even in the new era of living donor liver transplantation (LDLT). In that sense, alternative therapies of HCC including transarterial chemoembolization and hepatic resection should be explored before consideration of liver transplantation. Several series have reported promising results by using primary resection followed by “salvage transplantation.”6,7 However, resection can only be offered to a minority of patients, such as those with compensated cirrhosis without relevant portal hypertension. The addition of antiviral therapy in these patients may even lead to viral eradication and improvement of hepatic fibrosis, eliminating the future need for “salvage transplantation.”
NATURAL HISTORY (Figure 52-4) While recurrence of HCV infection is based on virological parameters, the recurrence of HCV disease requires protocol and/or clinically indicated liver biopsies that report both grade and stage of disease.53 Two patterns of HCV disease have been described, with differences in clinical presentation, prognosis, pathogenesis, and
therapeutic strategies.3,50,55 The commonest response to persistent HCV infection is the evolution over time to chronic hepatitis in a similar way to what has been described in the non-transplant patient but occurring at a viral set at least one log higher. Disease progression in these patients is typically accelerated compared to that observed in the immune-competent host.3,50,56 Progression in patients with this pattern of recurrence may follow two distinct pathways: (1) a linear rate of fibrosis progression;56 and (2) a delayed onset of rapid progression following an initial period of stabilization.57 Approximately two-thirds of patients with this pattern of recurrence develop an acute lobular hepatitis within the first 6 months post-transplantation. By the fifth postoperative year, over 80% of recipients will have developed chronic recurrent HCV disease, characterized histologically by portal and periportal inflammation with or without the presence of portal and/or periportal fibrosis. Lobular hepatitis can also be part of the pathological picture. Progression to chronic hepatitis and cirrhosis is linear in a substantial proportion of patients with a median rate of fibrosis progression of 0.3, ranging from 0.6 to 0.8 stage/year.3,50,56,59 In a subgroup of patients with initial benign recurrence (@ 30%), delayed hepatitis C-related severe liver damage may occur.57 In these patients, progression to severe disease is not linear and patients develop a sudden acceleration in fibrosis following an initial and sometimes prolonged period of stabilization. The presence of some degree of fibrosis and elevated liver enzymes at 3 years post-transplantation may predict this sudden change in the natural history of recurrent hepatitis C, with 70% of patients with these predictive factors developing this acceleration as opposed to 5% of those without these factors. Cholestatic hepatitis is an infrequent but extremely severe pattern of recurrence that leads to graft failure in 50% of patients within a few months of onset.3,50,55 Clinically, it is characterized by progressive jaundice and biochemical cholestasis, usually beginning
Figure 52-4. Risk of recurrent hepatitis C virus infection and variables associated with post-transplantation liver disease progression. FCH, fibrosing cholestatic hepatitis; LT, liver transplantation.
Viral factors Viral load Genotype Quasispecies HCV RNA (–) 2%
FCH 2–6%
Donor factors Age, sex, steatosis Surgical factors Warming ischemic time
LT
99% HCV RNA +
25–45% acute hepatitis
50–98% chronic hepatitis
8–44% graft cirrhosis in 5–7 years
Minimal injury 30–50%
Host factors HLA, race, gender, age, immune genetic background, immune system
Delayed injury 30%
982
42% decompensated 1 year
External factors Immunosuppression, alcohol, viral coinfection, antiviral therapy
Chapter 52
Fibrosis stage (fitted line, 95% CI)
MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
1996–97
4
1994–95
1992–93
1990–91
1985–89
Table 52-4. Transplant Recipient at High Risk of Severe Hepatitis C Virus (HCV)-Related Liver Disease Following Liver Transplantation
• • • • • • • •
3
2
1
0 0
1
2
3
4
5
6
7
8
Years from transplantation to fibrosis score measurement Figure 52-5. Fibrosis progression after liver transplantation. Effect of year of transplantation. Reproduced with permission from M Berenguer.58
in the first trimester post-transplantation after high levels of immunosuppression. Cholestatic HCV disease can occur as the initial manifestation of recurrent HCV disease or can emerge in the setting of chronic HCV disease. The following criteria need to be fulfilled to define fibrosing cholestatic hepatitis: 1 month posttransplant, serum bilirubin >100 mmol/l, serum alkaline phosphatase, and g-glutamyltransferase >5 times the upper limits of normal, characteristic histology of central ballooning (not necrosis or fallout), a paucity of inflammation ± cholangiolar proliferation without bile duct loss, very high serum HCV RNA levels, and absence of surgical biliary complications or hepatic artery thrombosis. A cytopathic mechanism of allograft damage is thought to be involved given the concurrence of extremely high viral burdens, reduced immune response with intrahepatic non-specific Th2 cytokine response, and unusual histology characterized by little inflammation, and severe centrizonal hepatocyte ballooning. Regardless of the pattern of recurrence, the hepatitis C-driven fibrosis response in the allograft leads to the development of graft cirrhosis in 8–30% of cases after a follow-up of 5–10 years.3,48,50,53–55,59–62 In addition, concerning data suggest that the rate of fibrosis progression, and thus the rate of development of graft cirrhosis due to recurrent hepatitis C is accelerated in patients transplanted in recent years (Figure 52-5).48,56,61,63 While short- and medium-term survivals are unaffected by the HCV status, recent data have confirmed that HCV infection is detrimental for long-term survival, with survival rates approaching 75–80% at 5 years in uninfected patients, but only 60–70% in those who are HCV-positive.48,49 The natural history of post-transplantation hepatitis C is, however, highly variable and, while some patients develop cirrhosis in less than 1 year due to recurrent infection, others show minimal or no injury in their protocol liver biopsies during years of followup, even in the presence of high levels of viremia3,50,51 (Figure 524). Factors influencing this variability are poorly understood, and while there are enough data to support the role of some variables such as viral load pretransplantation and donor age, the data regarding other variables such as genotype are controversial.
• • • •
Genotype 1 b Viral load pretransplantation >1 mmol/ml Viral load at 4 months post-transplantation >10 mmol/ml Donor age >50 years Rewarming ischemic time >60 minutes Early histologic recurrence (<6 months) Cytomegalovirus infection Severe early histologic changes : • Hepatic activity index >3 at 4 months or >8 at 1 year • Fibrosis stage >2 at 1 year Rejection episodes >2 Over-immunosuppression Treatment with antilymphocyte globulin Rejection treatment with >3 g methylprednisolone
VARIABLES ASSOCIATED WITH DISEASE SEVERITY AND/OR DISEASE PROGRESSION Pretransplant or early post-transplant recognition of patients with high risk of severe outcome post-transplantation (Table 52-4) is desirable since these patients can be targeted for intervention, or potentially even denied transplantation if outcomes are considered unacceptably poor. In contrast to HBV, where the availability of new therapies has obviated the need for selection of patients based on predictive factors, the elucidation of these predictive factors remains of paramount importance in HCV-infected patients, given the absence of effective prophylactic or therapeutic agents. Factors influencing the rate of progression relate to the virus, the host, or environmental and/or iatrogenic influences on the infected individual (Figure 52-4, Table 52-4). The high-risk patient likely derives from the interaction between these factors, particularly between the virus, the quality of the graft, and the immune system.
Host-Related Variables Immunosuppression Recent data have implicated the immune system in the pathogenesis of liver injury due to HCV.55 In fact, it is likely that immunosuppression is the most powerful determinant of HCV progression. Several findings highlight the deleterious effect of immunosuppression. These include: 1. a higher rate of fibrogenesis in immunosuppressed patients, both liver transplant recipients and HIV-coinfected patients compared to that observed in immune-competent patients;3,55,56 2. a shortened course to allograft cirrhosis, direct consequence of the above observation. While the timeframe between infection and development of cirrhosis is calculated in decades in the immune-competent individual, this time is reduced to a median of approximately 10 years in liver transplant immunosuppressed patients; 3. a greater risk of clinical decompensation following the establishment of compensated cirrhosis in liver transplant recipients compared to non-transplant patients. Indeed, the rate of decompensation is 42% at 1 year and 63% at 3 years in liver transplant recipients with recurrent HCV cirrhosis compared
983
Section VIII. Liver Transplantation
4.
5.
6.
7.
to 3% by 1 year and 18% at 5 years in immune-competent patients with HCV cirrhosis;64 a worse outcome following decompensation for liver transplant recipients compared with non-transplant patients with decompensated HCV cirrhosis with a 3-year survival less than 10% in the former population compared with 60% in the latter;64 an accelerated course of recurrent hepatitis C in more recent years at a time in which more potent immunosuppressive agents are being utilized for both induction and maintenance immunosuppression;48,56,61,63 an adverse effect of cytomegalovirus infection, a virus with inmunosuppressive properties, on HCV-related disease outcome;63 the development of fibrosing cholestatic hepatitis only in the setting of significant immunosuppression, including after solid organ and bone marrow transplantation as well as in HIV infection.55
While the above indirect findings clearly suggest that immune suppression is strongly linked to the severity of recurrent HCV disease, the effect of specific immunosuppressive drugs on both viral replication and HCV-related fibrosis progression is less evident (Table 52-5).3,65–67 Several studies have demonstrated that the use of corticosteroid boluses to treat acute cellular rejection is harmful to HCV-infected recipients as a result of their association with an increased frequency of acute hepatitis, earlier time to recurrence, and higher risk of progressing to graft cirrhosis or developing cholestatic hepatitis.3,65–66 Corticosteroid boluses are associated with an increase in serum HCV RNA concentrations of four- to 100-fold. In fact, the absence of corticosteroids from induction immunosuppression therapy was found in one study to be associated with a
delay in the increase of HCV viral load following transplantation.68 Similarly, the use of OKT3 to treat rejection is associated with a greater risk of recurrence of hepatitis C, but mainly greater risk of post-transplant fibrosis progression and graft loss.3,66,67 While the adverse effect of additional immunosuppression for the treatment of rejection on hepatitis C is demonstrated, the effect of agents used for induction or maintenance immunosuppression is less clear (Table 52-5).3,66,67 For example, there are data that suggest that complete avoidance of corticosteroids may be beneficial but there is also evidence that abrupt and early withdrawal of corticosteroids may be harmful to these patients. Similarly, the data with mycophenolate mofetil (MMF) and antibody induction are also confusing. In fact, results of the potential association between the type of administered immunosuppression and disease severity are difficult to prove due to the multiplicity of immunosuppressive regimens together with the changes in immunosuppressive drugs in individual patients over time. In addition, most studies are single-center studies, based on small sample sizes, retrospective in nature and hence unable to control for selection biases or for the presence of confounding variables. Data on calcineurin inhibitors are less confusing and overall show similar results with tacrolimus and cyclosporine. It has been suggested that the outcome is probably not related to the use of a specific drug but rather with the dose and drug level achieved, reflecting overall immunosuppression, and the way the drug is modified. It is interesting to speculate that a threshold on immunosuppression may exist that warrants absence of rejection while keeping the immune pressure to HCV at a significant level.
Genetic Background It is likely that it is the interplay between the immune system and the virus, modulated by the immunogenetic background, such as the
Table 52-5. Effect of Different Immunosuppressive Agents on Viral Replication and Disease Progression
Pulse steroids Maintenance steroidsa Ciclosporinb Tacrolimusb Azathioprine OKT3 Anti-IL2 receptor antibodies Sirolimus Mycophenolate mofetil MMF + IL-2 receptor antibodies
Mechanism of action
Effect on viral load
Effect on HCV-disease severity/ progression
Global anti-inflammatory and immunosuppressive actions Global anti-inflammatory and immunosuppressive actions Inhibition of early T-cell signal pathways and IL-2 production and release Same as ciclosporin (more potent) Inhibition of adenosine monophosphate production Antilymphocyte antibodies Inhibition of lymphocyte activation and clonal expansion Inhibition of lymphocyte proliferation, fibrosis and fibroblast proliferation Inhibition of inosine monophosphate dehydrogenase Addition of the two compound effects
Increase
Negative
Increase
Controversial
Decrease
Controversial
Unknown Unknown
Controversial Controversial
Unknown Unknown
Controversial Controversial
Unknown
Unknown
Increase
Controversial
Increase
Negative
HCV, hepatitis C virus; MMF, mycophenolate mofetil; IL-2, interleukin-2. a Greater doses or prolonged exposure to steroids appear to worsen recurrence in some studies. However, some authors have suggested that early abrupt steroid withdrawal may lead to immune-reconstitution, and in doing so, induce worsen graft injury through an immune-mediated attack on the virus. b Most studies on calcineurin inhibitors suggest that the risk of severe hepatitis C is not necessarily related to type of drug, but rather the dose, level, and overall level of immunosuppression.
984
Chapter 52 MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
human leukocyte antigen (HLA) system, that shapes the outcome post-transplantation. In that sense, specific HLA class II alleles, such as HLA B14 and HLA DRB1*04 have emerged as possible modulators of disease severity3 and disease severity has been linked with HLA-B-sharing between the donor and the recipient in some, but not all, studies.3,51 In addition, preliminary reports highlight the potential importance of the immunogenetic background of the donor.51
Race Race has recently been found to influence outcome in patients with recurrent HCV infection, with non-Caucasians doing worse than Caucasians.56,62 However this association deserves further analysis.
Viral-Related Variables Post-Transplantation HCV RNA Levels Circulating HCV RNA levels following liver transplantation are typically 10–20-fold higher than levels prior to transplantation,65 without a clear association between levels of viremia and disease severity. High levels of viremia, however, have been described in the setting of fibrosing cholestatic hepatitis55 as well as during the acute phase of recurrent hepatitis C. These observations suggest that, while in many circumstances liver damage is immune-mediated, in some instances it is due to a direct cytopathic effect of HCV.
Pretransplantation HCV RNA Levels Several studies have shown that, as described for HBV, level of viremia pretransplantation predicts the occurrence and/or severity of recurrent hepatitis C.56,62
HCV Genotype The effect of the infecting genotype on the severity of liver disease post-transplantation is unclear.3,51 Some, but not all, studies have implicated genotype 1b in a more severe post-transplantation disease compared to non-1b genotype. Furthermore, a fast HCVrelated disease progression has been observed in centers with a high prevalence of genotype 1b, thus indirectly implicating this genotype in a more aggressive course of the disease. Preliminary data suggest that different strains belonging to genotype 1b may be involved in the pathogenesis of severe liver injury.69
HCV Diversity HCV heterogeneity may play a role in the pathogenesis of progressive HCV disease. The hypervariable region (HVR)-1, a putative target for neutralizing antibodies, is the most common region analyzed. Results from the few studies published to date are however inconclusive and somewhat discrepant, and may be related to the small number of patients included, the different methodologies applied to assess HCV heterogeneity, the type of end-point chosen, the region of the genome evaluated, and the definition of disease severity.3,55 Overall, severe fibrosis seems to be correlated with changes in HCV quasispecies composition. In one study, immunosuppressed HCV transplant recipients had fewer quasispecies and fewer amino acid sequence changes than untreated HCV patients, and the two patients with the least HCV variation died of HCV graft infection. Published data suggest that preferential replication of few genetically divergent quasispecies is enhanced by routine
immunosuppression, particularly in patients with fibrosing cholestatic hepatitis. The reduction in the number of HCV variants can be interpreted as indicative of a decreased immune pressure to the virus due to overall immunosuppression. In contrast, patients with mild recurrence show more HCV variants in the HVR1, but less genetic divergence. A recent study showed an association between HCV clearance post-transplantation and an increase in HCV HVR1 variants following steroid-tapering.55
Donor-Related Variables The age of the donor has been found to be independently associated with disease severity, disease progression, and survival.3,48,51,61,70,71 In one series, only 14% of the recipients who received an organ from a donor younger than 30 developed recurrent HCV-related cirrhosis. In contrast, 45% and 52% of those receiving the organ from donors aged 31–59 or older than 59 developed graft cirrhosis, respectively (P < 0.0001).48 The changing organ donor profile with increasing age of the donors in recent years may explain the observed worsening in outcomes over the same time period. This observation has important implications for donor liver allocation, in that older donors might be more appropriate for HCVnegative recipients in whom the adverse effects of donor age appear to be less deleterious. In addition, the effect of donor age correlates with data from the immunocompetent population where age at the time of infection is an important and powerful determinant of fibrosis development. Age-related changes in liver response may be the key factor that determines the increased susceptibility of the older liver to HCV-related fibrosis.
Coinfection with Other Viruses Patients who develop cytomegalovirus viremia may be at increased risk of severe HCV recurrence.63 The reasons for this association are unknown but likely relate to induction of immune deficiencies. Coinfection with other hepatotropic viruses such as HBV may influence histologic disease severity, but results are conflicting.3,51
Other Variables Prolonged rewarming time during allograft implantation has been associated with severe recurrent disease.3 If these data were confirmed, special emphasis should be paid to minimizing rewarming time. A lack of association has been found between the rate of fibrosis progression prior to transplantation and that observed after transplantation, suggesting that variables present at the time of transplantation and those related to post-transplantation management are more important in influencing disease progression than genetic or viral variables unique to the individual.56
HCV KINETICS AND PATHOGENESIS A rapid and sharp decline in viral load occurs immediately after removal of the infected liver.65,68 Following reperfusion, HCV RNA levels typically decrease further at a rate that exceeds the decrease observed during the previous anhepatic phase. HCV binding to and/or uptake by hepatocytes may contribute to this early posttransplant decrease in viremia. Following this initial decline, HCV RNA levels either increase exponentially, reaching pretransplantation levels as soon as day 4, or continue to decline in the first post-
985
Section VIII. Liver Transplantation
transplant week to increase thereafter, peaking by the fourth postoperative month. These differences in kinetics appear to be related to the use of corticosteroids, so that HCV RNA levels increase rapidly in patients receiving corticosteroids as part of the immunosuppressive regime, while they continue to decrease in those not receiving this drug. At 1 year post-transplantation, the levels are 10–20-fold higher than pretransplantation. The mechanisms by which HCV leads to liver injury are incompletely understood. Several lines of indirect evidence support a role for the cellular immune response in shaping outcome following transplantation. It is likely that the increased levels of HCV replication in conjunction with the altered host immune responsiveness contribute to the pathogenesis of liver damage, particularly to the severe course of disease of the grafted liver.55 In addition, there is a spectrum of severity of recurrent HCV disease that may differ in pathogenesis and clearly differs in outcome. Recurrent chronic hepatitis C disease, which can lead to cirrhosis, should be distinguished from recurrent cholestatic HCV disease. In the former, it is possible that the relative antiviral control by the immune response prevents cytopathic injury while perpetuating chronic liver injury. In contrast, a direct cytopathic mechanism of liver injury appears to predominate in patients with severe cholestatic hepatitis.
HISTOLOGIC CHANGES: IMPORTANCE OF PROTOCOL LIVER BIOPSIES Liver function tests are not correlated with either viremia or with histologic disease severity3,51,72,73 and protocol liver biopsies are generally needed to identify progression to severe forms of chronic hepatitis. In addition, histological findings may be helpful in selecting the patients at high risk of disease progression. In particular, the degree of activity and fibrosis staging observed in the liver biopsy at 1 year are associated with the subsequent risk of developing cirrhosis, with only 3–10% of those with mild hepatitis developing cirrhosis compared to 30–60% in those with moderate to severe activity in their first-year liver biopsy. 3,50,60 Some specific histological features, including the presence of significant steatosis, ballooning degeneration, cholestasis, and confluent necrosis, are also helpful in predicting progression of disease. Finally, the presence of some degree of fibrosis at 3 years post-transplantation predicts the delayed onset of severe liver damage.57
LIVE-DONOR LIVER TRANSPLANTATION FOR HEPATITIS C-INFECTED PATIENTS Despite all the measures to improve the outcome of HCV-infected patients, the increased organ shortage has led to a dramatic increase in the number of patients on the waiting list and in those dying while waiting. The implementation of LDLT was believed to be a potential solution to this problem. The main unanswered question relates to the outcomes obtained with this new technique, and whether they are the same as achieved in recipients of deceased donors. To date, the question remains unanswered since results between centers are conflicting.74 While some studies have suggested that HCV recurs earlier and is associated with more severe hepatitis, other studies have not confirmed these data. However, the number of HCV-infected LDLT recipients to date is still limited, the posttransplant follow-up interval is short, and the available reports lack, in general, protocol liver biopsies. As a result, the data regarding the
986
impact of LDLT on severity of HCV recurrence are inconclusive, as reviewed by the consensus group.53
TREATMENT Given the increasing number of patients progressing to HCV-related graft cirrhosis and the shortage of organ donors, development of strategies to improve the outcome of these recipients is mandatory. These strategies include an adequate timing of liver transplantation in patients with HCV-related end-stage liver disease, antiviral therapy, and the use of specific immunosuppression regimens. In terms of antiviral therapy, three potential alternative and/or complementary approaches may be attempted (Figure 52-2): (1) pre-emptive antiviral therapy as the patient is awaiting the availability of an organ donor; (2) early post-transplant antiviral therapy before histological damage has occurred; and (3) treatment of disease when and if it occurs. The goals of treatment and end-points for success of therapy may be different in these situations. The major end-point of therapy in patients awaiting transplantation may be the stabilization and/or improvement of hepatic function so that the need for surgery is delayed or even obviated. Alternatively, viral eradication or at least viral suppression is also a relevant goal, so that the risk of post-transplantation HCV recurrence and/or aggressive recurrent HCV disease is reduced. The major goal of pre-emptive post-transplantation therapy is to prevent reinfection of the graft, and in doing so, to reduce the incidence and potentially the severity of recurrent disease. Finally, the primary end-point in patients with established disease is viral eradication, since a sustained clearance of HCV RNA is associated with improvement in liver histology in most patients. Indeed, two long-term studies have shown that loss of HCV after treatment is durable (90–100% after 2–3 years of follow-up), and that the durability of the response is associated with improvement in hepatic inflammation (50 and 60% after 2 and 5 years of follow-up) and regression of fibrosis up to 67% after 3–5 years of follow-up.75,76 While the timing and aim of these alternatives are firmly established, their efficacy and tolerance are less clear. It is well known that current antiviral therapy based on interferon with ribavirin is poorly tolerated in both the pre- and post-transplant setting, therefore limiting its general application.
Prevention of Reinfection and/or HCV-Related Recurrent Disease Currently, there is no available intervention to prevent predictably HCV recurrence. In one study, polyclonal immunoglobulin containing anti-HCV was shown to decrease the incidence of recurrent HCV viremia measured 1 year post-transplantation.77 However, preliminary data analyzing the efficacy of anti-HCV immunoglobulin (HCIg) for prevention of HCV recurrence did not demonstrate clinical or virological benefits. Whether this negative study is a function of the dose of HCIg used, the timing of administration, or the type of preparation used (non-neutralizing antibodies) is unknown. It is also possible – if not likely – that the diverse quasispecies nature of HCV makes this virus inherently more difficult to neutralize than the more stable HBV virus populations present in any individual.
Pretransplantation Antiviral Therapy In theory, the rationale for treating patients with decompensated HCV-related cirrhosis is the same as for hepatitis B, that is: (1) to
Chapter 52 MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
slow down clinical disease progression and improve hepatic synthetic function, and in doing so, to reverse the complications of liver disease and obviate or delay the need for liver transplantation; and (2) to achieve an improvement in post-transplantation outcome by clearing the virus prior to transplantation. While the first aim is achievable in some patients with HBV-related decompensated cirrhosis, it is rarely achieved in patients infected with HCV once hepatic decompensation has occurred. In contrast, clearance of HCV prior to transplantation may prevent viral recurrence, and reduction of HCV RNA levels could potentially improve post-transplantation disease progression. Patients waiting for liver transplantation due to HCV-related liver disease typically include two types of patients: (1) those with compensated cirrhosis and HCC; and (2) those with decompensated cirrhosis. While in the former group it is likely that a complete course of antiviral therapy may be achieved with currently available drugs, it is less so in those with advanced hepatic insufficiency. Treatment of Compensated Cirrhosis Therapy of chronic hepatitis C has improved in recent years with the addition of ribavirin and development of pegylated (PEG) forms of interferon (IFN). Sustained virological responses are achieved in 54–56% of patients compared to 44–47% of those treated with standard interferon and ribavirin. Although tolerance is adequate, the response rate to these therapies appears to be lower in patients with compensated cirrhosis or transition to cirrhosis than in patients with less advanced liver disease (43–50% versus 57–65% in non-cirrhotic patients). Tolerability and side effects are similar to those observed with standard interferon, with the exception of cytopenias, particularly neutropenia, that are more frequently seen with the PEG-IFNs. These side effects may become a limitation when treating cirrhotic patients with marginal counts. The use of growth factors (i.e., erythropoietin and neutrophil-stimulating factor) may be helpful in some cases and avoid the reduction and/or discontinuation of antiviral drugs. Combination therapy with PEG-IFN and ribavirin is recommended in patients with cirrhosis provided that no contraindications are present.78 For genotype 1-infected patients, the optimal length of therapy and dose of ribavirin are probably 48 weeks and 1000–1200 mg; in contrast, for genotypes 2 and 3-infected patients, 24 weeks and 800 mg of ribavirin are probably sufficient. Treatment of Patients with Decompensated Cirrhosis (Table 52-6) Data regarding this alternative are limited to small uncontrolled case series.79 These studies emphasize the advantages and disadvantages of this approach. Although antiviral therapy may be successful in some cases, it should be administered with extreme caution due to
the increased risk of infectious complications and hepatic decompensation. In the first study, HCV-infected patients at or near the top of their respective waiting lists were randomly assigned to one of three treatment arms, two involving therapy with interferon in monotherapy, and one in combination with ribavirin. Less than half the patients screened met entry criteria, with thrombocytopenia and leukopenia being the most common reasons for exclusion. Eventually, only 15 patients from five large transplant centers were treated. Nine patients received interferon monotherapy while six received combination therapy with ribavirin (400 mg twice daily). Most patients had advanced liver disease with a mean Child–Pugh score of 12. While on treatment, loss of detectable HCV RNA was seen in 33%, but recurrence was seen in two transplanted patients. In addition, a significant number of adverse effects occurred (n = 23), many of which were considered severe. While thrombocytopenia was the most frequent adverse event, infection was the most severe one, leading to death in one patient. These side effects, particularly life-threatening infections, ultimately led to the early termination of the study. Several conclusions were drawn from this study: 1. a large proportion of patients awaiting transplantation will not benefit from this approach due to the presence of contraindications, particularly thrombocytopenia and neutropenia; 2. awareness of the potential complications should be kept in mind when treating the small proportion who meet initiation criteria. In the second study, 101 patients, 70% of whom were infected with HCV genotype 1, were treated with low doses of interferon (1.5 mU thrice weekly) and ribavirin (600 mg daily), with slow increases in dose of both drugs every 2 weeks as tolerated. Growth factors were administered as needed. Patients were relatively well compensated (50% were Child’s class A), with most having a low Child–Pugh score of less than 7–8. On treatment virological responses occurred in 38% and sustained virological response in 22% of patients. As for previous studies, sustained responses were more common in patients infected with genotypes other than 1. Interestingly, recurrent infection, which was observed in all patients with detectable HCV RNA at the time of transplantation, was prevented in the eight patients who were HCV RNA-negative at the time of transplantation. Although overall rates of severe adverse events were not reported, 28% of the patients had treatment discontinued because of the development of side effects, and serious complications occurred in 8% of the patients. Results from this study are more optimistic than those of the first published report, the
Table 52-6. Pretransplantation Antiviral Therapy with Interferon or Interferon + Ribavirin Author, year (number of patients/% genotype 1)
Child A (%)
Eligibility
VRa
Adverse events
Treatment D/C (%)
Prevention of recurrence
Crippin 200280 (n = 15/67%) Everson 200281 (n = 91/77%) Forns 200382 (n = 30/83%)
0%* 50% 50%
47% NA 62%
33% 38% 30%
87% NA 63%
100% 28% 20%
No 100% 67%
a
All patients had advanced liver disease (Child C). NA, not available; VR, hepatitis C virus RNA undetectable by polymerase chain reaction; D/C, discontinuation.
987
Section VIII. Liver Transplantation
difference possibly related to the best hepatic function of most patients included in the latter study. In addition, lower doses were used at initiation. In the most recent study, patients on the waiting list for liver transplantation were considered for antiviral therapy if the expected time on the waiting list was shorter than 4 months, if there was not evidence of renal failure, and if patients met minimal blood count criteria. Of 50 patients evaluated, only 62% met entry criteria. The majority of these patients were infected with HCV genotype 1. At the time of inclusion, half of the patients were Child–Pugh A while the other half were Child B/C. The regimen used consisted of interferon-alpha-2b 3 MU/daily and ribavirin 800 mg/day. A virologic response was achieved in 30% of patients. Pretreatment viral load was significantly lower in responders than in non-responders (3 ¥ 105 versus 6.5 ¥ 105 IU/ml). Of the nine patients who were HCV RNA-negative at the time of transplantation, six remain free of infection after a median follow-up of 46 weeks. Although side effects and dose reductions were frequent, particularly due to cytopenias, no patients died while on therapy. However, treatment had to be discontinued in six patients due to thrombocytopenia (n = 4) and sepsis (n = 2). An association was found between viral load at 4 weeks of initiation of therapy and virologic response, an association that could be used in the clinical practice to guide discontinuations of therapy in those with minimal chances of obtaining viral clearance. The positive and negative predictive values of an early decrease in viral load (≥2 logs) were 82 and 100%, respectively. In summary, prevention of recurrence may be achieved in 20% of treated patients on the waiting list. The recommended dose is still unclear. Results were similar in terms of virological response in the three studies regardless of the regime used. Indeed, in two studies, therapy was started with low doses of interferon ± ribavirin, and doses were increased slowly as tolerated. In contrast, complete doses of both drugs were used in the third.
Post-Transplantation Antiviral Therapy Post-transplantation therapy may be started early following surgery or, alternatively, at a later stage, when disease is established. The major goal of pre-emptive post-transplantation therapy is to prevent reinfection of the graft, and, in so doing, to reduce the rate and/or severity of recurrent disease. In contrast, the major goal of treatment of established disease is to eradicate viremia, and, in so doing, to improve histology. Post-Transplantation Pre-emptive Therapy. Interferon alone or in combination with ribavirin has been attempted within the first 2
weeks after liver transplantation.83 The rate of sustained virological response achieved with interferon monotherapy ranges from 0 to 17%. Slightly better results may be achieved with interferon in combination with ribavirin (11–33%). However, side effects are common, leading to dose modifications or drug discontinuation in 28–50% and 47–85% of those treated with interferon or interferon–ribavirin, respectively. PEG-IFNs have also been attempted in this setting. In a recent multicenter US study, 54 transplant recipients were randomized within the first 3 weeks to receive 48 weeks of PEG-IFN-a2a or no treatment. Therapy was tolerated reasonably well, with only 8% of the patients discontinuing PEG-IFN due to side effects. However, efficacy was dismal, with a sustained virological response of only 8% in the treated group as opposed to 0% in the untreated arm.84 The applicability of this approach is rather limited since a substantial proportion of patients (40–60%) will not meet minimal criteria in the first 2–4 weeks post-transplantation to receive therapy with interferon ± ribavirin.
Treatment of Patients with Recurrent Hepatitis C Monotherapy has resulted in extremely disappointing results, with biochemical, virological, and/or histological response being rarely achieved with interferon or ribavirin as single agents.52,79 Although combination therapy probably yields the greatest potential benefit, results are still far from satisfactory (Table 52-7). Studies on therapy of recurrent hepatitis C are scarce, non-randomized, and generally based on small sample sizes.52,79,85,86 Sustained virological response achieved in studies of standard interferon with ribavirin ranges from 9% to 33%. Both dose adjustments (up to 78%) and drug discontinuations (30–50%), mainly due to ribavirin toxicity, are frequent. Although less common, severe adverse effects, particularly hepatic decompensation, may occur in treated patients (5% in previous studies). Treatment has also resulted in frequent hospital admissions, blood transfusions, as well as use of growth factors. The optimal duration of therapy is still unknown. Six versus 12 months of combination therapy were compared in 57 transplant patients.85 A sustained virological response was achieved in six of 27 patients treated for 6 months (22%) and in five of 30 patients treated for 12 months (17%) (P = 0.4). Response was better in those infected with genotype non-1 than in genotype 1, after both 6 and 12 months of therapy (43% versus 15% and 43% versus 9%, respectively), although the numbers of patients studied were too small to draw definitive conclusions regarding treatment duration.
Table 52-7. Combination Therapy with Interferon-Alpha or Pegylated Interferon and Ribavirin in Patients with Established Recurrent Hepatitis C Treatment regimen IFN + ribavirin Pegylated IFN Pegylated IFN + ribavirin
End-treatment virological response (%)
Sustained virological response (%)a
Histological improvement
Treatment reduction (%)
D/C (%)
35 35 37
22 19 25–30
70% 60% 50%
65 75 75
45 15 45
IFN, interferon; D/C, discontinuation. a Prognostic factors: sustained virological response significantly higher in hepatitis C virus genotype 2–3-infected patients, in those with low levels of viremia, and in those without advanced liver damage.
988
Chapter 52 MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
In terms of dosage, some authors have suggested that starting therapy with low doses with subsequent increases as tolerated may yield better results in terms of patient tolerance and compliance than starting with full doses, although such an approach would be predicted to compromise an already low virological response. Regardless of the protocol used, tolerance remains poor in this setting, with 80% of the patients requiring dose reductions of one or both drugs. Many patients have renal insufficiency secondary to immunosuppressive agents, and because of impaired renal clearance of ribavirin, drug-associated hemolysis can be profound.86 Thus if ribavirin is initiated as part of combination therapy, the dose should be 600–800 mg, depending on renal function and presence of anemia. Given that anemia is the major side effect of therapy posttransplantation, erythropoietin would be predicted to improve the tolerability of ribavirin. In contrast to non-transplant patients, the natural history of recurrent hepatitis C may not necessarily improve after successful viral eradication.75,76 The cause of persistent hepatitis and absence of fibrosis regression in many transplant patients who achieve a sustained viral clearance remains speculative, and includes infections with cryptic viruses, persistence of HCV RNA in liver tissue,88 and chronic allograft rejection. Indeed, although problems with interferon-induced rejection are less evident with combination therapy than with monotherapy,89,90 there are preliminary data suggesting an increased risk of rejection following a successful course of antiviral therapy, possibly as a consequence of low levels of immunosuppression. It is suggested that antiviral therapy improves hepatocyte microsomal function which leads to decreased immunosuppression levels. Persistence of elevated liver enzymes despite viral clearance warrants routine follow-up liver biopsy in order to rule out posttreatment rejection. In summary, response to standard interferon and ribavirin is generally lower following liver transplantation than in immunecompetent patients. Reasons for this low response rate likely include all of the following: (1) high levels of viremia; (2) high prevalence of HCV genotype 1; (3) low tolerability of interferon and ribavirin leading to frequent dose reductions; (4) previous non-response to interferon; and (5) lower responses to HCV therapy in patients with impaired immune function. Indeed, preliminary data from kinetics studies have shown that viral decay in the first 24 hours after the injection of interferon is significantly lower in transplant patients compared to those who are immune-competent. In addition, the efficacy of therapy appears to be reduced in recipients with advanced recurrent hepatitis and reversal of fibrosing cholestatic hepatitis is highly unusual.91 With combination PEG-IFN–ribavirin, the rate of sustained eradication has slightly increased to 26–45%.88,92,93 Response is generally associated with improvement in histology, particularly necroinflammation and, to a lesser extent, fibrosis. The main predictors of response include genotype and the completion of the treatment course. Tolerance remains an issue, with up to 90% requiring dose reductions and 24–49% drug discontinuations. PEGIFN may yield a higher risk of inducing acute cellular rejection than standard interferon. Pegylation properties, including extended half-life and increased serum concentrations of interferon, could make it more likely to increase HLA expression, and therefore risk of rejection. Rejection associated with PEG-IFN therapy can
lead to serious consequences such as graft loss from resistant rejection.89 There are still many aspects that need to be addressed, such as the optimal dose and duration of therapy, whether ribavirin maintenance is required following interferon discontinuation, the potential benefit of using growth factors, and whether fibrosis progression may be slowed down by continuing interferon therapy in those who do not achieve a sustained virological response. In summary, each of the strategies has advantages and disadvantages, but all remain highly unsatisfactory (Table 52-6). Prophylactic therapy while the patient is awaiting transplantation is the best theoretical approach. It is however limited by the low applicability and low tolerance in patients with advanced liver disease. Preemptive post-transplantation therapy is another attractive approach from a theoretical point of view. Although effective in some patients and apparently not associated with a substantial increased rate of rejection, tolerance is problematic and benefits are low. In addition, with these two strategies treatment is offered to all patients while only a proportion will develop serious complications from recurrent infection. With the available drugs, treatment of the established disease is probably the most attractive option.94 Although still limited by a low efficacy, tolerance appears to be better than when these drugs are given pre-emptively. Treatment should be offered preferentially to patients who develop histological progressive liver disease. Protocol liver biopsies perhaps at yearly intervals will identify early histologic changes that herald a progressive disease. A major barrier to improved response is the frequent occurrence of untoward effects requiring dose reduction or even cessation of therapy. In conclusion, pretransplantation antiviral therapy with PEG-IFN and ribavirin may be tried in patients with Child A cirrhosis, especially in those infected with genotypes 2 and 3. In the early posttransplantation period, pre-emptive treatment before development of disease cannot be advocated because PEG-IFN plus ribavirin is very poorly tolerated. In patients with established disease, treatment should be initiated, if no contraindications are present, once portal fibrosis and/or moderate necroinflammation is detected. Protocol liver biopsies at intervals may be necessary to identify such patients. Optimal dose and duration of therapy are unknown. While we wait for studies to define the best regimen, most clinicians follow the same guidelines that are used in immune-competent patients. In transplant patients, though, therapy may need to be indefinite in those with the most severe forms of recurrent hepatitis.91
ALTERNATIVE APPROACHES Since the efficacy of antiviral therapy is limited in HCV-infected recipients, selection of patients who will be at low risk for severe recurrence and optimal management of long-term immune suppression will likely be important in improving long-term outcomes. Some authors have developed predictive models of severe outcome based on simple variables, including age of the donor and type of immunosuppression.61 Validation of these models is however required before they can be generalized. Given the deleterious effect of intense immune suppression on the progression of recurrent HCV disease, an adequate management of immunosuppression appears mandatory in these patients. As a general rule, the optimal strategy should be to achieve a balance
989
Section VIII. Liver Transplantation
between prevention of acute and chronic rejection while minimizing the adverse effects of immunosuppression on recurrent hepatitis C. Based on the available information, the following recommendations were made at the recent consensus conference:53,66,67 (1) induction immunosuppression should be performed with two drugs or reduced doses of one or more agents when using triple-drug therapy; and (2) when rejection is diagnosed histologically, the first approach should be to increase the dose(s) of the agents used for maintenance immunosuppression while avoiding bolus corticosteroids or T-cell-depleting agents. Additionally, when doubts exist between rejection and hepatitis C because of overlapping histological findings, serial biopsies should be performed to clarify the clinical picture, rather than therapeutic trials of steroids. Features more suggestive of HCV infection include lymphoid aggregates, fatty changes, and sinusoidal dilatation, while those more suggestive of rejection include endotheliitis, bile duct necrosis, and a mixed portal inflammatory infiltrate (eosinophils and neutrophils as well as mononuclear cells).
insufficiency, and as a result, most patients with recurrent allograft failure due to hepatitis C will only receive an organ at a point when they are unlikely to survive retransplantation. Thus, while retransplantation may be a reasonable option in low-risk patients, it is unlikely to be a feasible option unless a live donor is available (see below). Whether retransplantation is justified in patients with several variables associated with poor outcome needs carefully consideration. If retransplantation is performed, therapeutic strategies, including different immunosuppressive protocols and use of antivirals, should be implemented despite the frequent coexistence of comorbidities and development of side effects. Additional important aspects to be considered and that were recommended at the consensus conference53 are: (1) the presence of cholestatic HCV disease should preclude retransplantation other than in exceptional circumstances; (2) additional transplantation beyond the second graft for recurrent hepatitis should be discouraged; (3) there are minimal data concerning the role of retransplantation in living donor recipients with recurrent HCV disease and no recommendations could be made.
RETRANSPLANTATION Retransplantation is the last option for patients with failing grafts due to recurrent disease. The results of retransplantation are inferior to those reported for first transplants. As predicted from natural history studies, the prevalence of HCV infection in patients undergoing retransplantation has progressively increased in most transplant centers (from 6.5% in 1990 to 38.4% in 1995), reaching a plateau thereafter.95 It has thus become imperative to determine whether all patients with graft failure due to recurrent HCV disease are candidates for further transplantation, or whether there is a subset in whom the outcomes would be so poor that retransplantation should not be undertaken.96 Early reports have suggested poor outcomes and with increasing shortage of organ donors, retransplantation is likely not the most beneficial use of a limited resource. In addition, the severity of recurrent HCV disease in the first graft may predict the severity of recurrence in the second graft.101 While we wait for a consensus, it has become apparent that this procedure is becoming less common at many centers. The fear with retransplantation, particularly in those with early severe recurrence, is related to four major aspects: 1. early reports suggesting a worse outcome following retransplantation in HCV-infected recipients than in those uninfected; 2. uncertainty regarding the natural history of recurrent hepatitis C in the second graft; 3. frequent comorbidities in these patients who generally have an advanced age by the time they require retransplantation; 4. increased organ shortage. Most series have shown that the outcome is generally poor, significantly worse than that obtained with retransplantation for other causes of late graft loss.95,96–100 Most cases of death occur in the first 6 months and are due to sepsis. However, it has also been shown that the outcome may be improved if performed before significant renal impairment and hepatic failure develop,98–101 and with the use of younger donors.98 Unfortunately, under the current MELD organ allocation system, patients have no realistic hope of receiving an organ until they have developed significant coagulopathy and renal
990
CONCLUSION Viral hepatitis is the leading indication for liver transplantation in the majority of transplant centers. Post-transplantation outcome in these patients largely depends on the prevention of allograft reinfection. In contrast to hepatitis B, where excellent results have been achieved following the implementation of effective interventions to prevent and to treat disease, recurrent hepatitis C is an increasing problem facing liver transplant hepatologists and surgeons. Currently no effective prophylactic therapy is available for hepatitis C so that recurrent hepatitis C occurs invariably. Progression to severe allograft fibrosis is often rapid. Current antivirals, including PEGIFNs, carry with them substantial toxicities that compromise efficacy. Hence, it is not surprising that, although some improvements have been made in the treatment of recurrent hepatitis C, a substantial proportion of HCV-infected patients develop recurrent allograft end-stage liver disease, leading to decreased graft survival, increased need for retransplantation, and ultimately, decreased patient survival. Only therapies that are not yet available are likely to change this picture.
REFERENCES 1. Keeffe EB. Liver transplantation: current status and novel approaches to liver replacement. Gastroenterology 2001; 120:749–762. 2. Roche B, Samuel D. Liver transplantation for hepatitis B virusrelated liver disease: indications, prevention or recurrence and results. J Hepatol 2003; 39:S181–S189. 3. Berenguer M, Lopez-Labrador FX, Wright TL. Hepatitis C and liver transplantation. J Hepatol 2001; 35:666–678. 4. Steinmuller T, Seehofer D, Rayes N, et al. Increasing applicability of liver transplantation for patients with hepatitis Brelated liver disease. Hepatology 2002; 35:1528–1535. 5. Roche B, Feray C, Gigou M, et al. HBV DNA persistence 10 years after liver transplantation despite successful anti-HBs passive immunoprophylaxis. Hepatology 2003; 38:86–95. 6. Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet 2003; 362:1907–1917.
Chapter 52 MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
7. Majno P. Treatment of hepatocellular carcinoma at the dawn of the third millennium: liver transplantation and its alternatives. Acta Gastroenterol 2004; 67:206–222. 8. Prieto M, Gómez MD, Berenguer M, et al. De novo hepatitis B after liver transplantation from hepatitis B core antibody-positive donors in an area with high prevalence of anti-HBc positivity in the donor population. Liver Transpl 2001; 7:51–58. 9. Muñoz SJ. Use of hepatitis B core antibody-positive donors for liver transplantation. Liver Transpl 2002; 8:S82–S87. 10. Villamil FG. Prophylaxis with anti-HBs immune globulins and nucleoside analogues after liver transplantation for HBV infection. J Hepatol 2003; 39:466–474. 11. Yu AS, Keeffe EB. Nucleosides analogues and other antivirals for treatment of hepatitis B in the peritransplant period. Clin Liver Dis 2003; 7:551–572. 12. Fontana RJ. Management of patients with decompensated HBV cirrhosis. Semin Liver Dis 2003; 23:89–100. 13. Villeneuve J, Condreay LD, Willems B, et al. Lamivudine treatment for decompensated cirrhosis resulting from chronic hepatitis B. Hepatology 2000; 31:207–210. 14. Yao FY, Bass NM. Lamivudine treatment in patients with severely decompensated cirrhosis due to replicating hepatitis B infection. J Hepatol 2000; 33:301–307. 15. Kapoor D, Gutpan RC, Wakil S, et al. Beneficial effects of lamivudine in hepatitis B virus related decompensated cirrhosis. J Hepatol 2000; 3:308–312. 16. Yao FY, Terrault NA, Freise C, et al. Lamivudine treatment is beneficial in patients with severely decompensated cirrhosis and actively replicating hepatitis B infection awaiting liver transplantation: a comparative study using a matched, untreated cohort. Hepatology 2001; 34:411–416. 17. Perrillo RP, Wright T, Rakela J, et al. A muticenter United States–Canadian trial to assess lamivudine monotherapy before and after liver transplantation for chronic hepatitis B. Hepatology 2001; 33:424–432. 18. Fontana RJ, Keeffe EB, Carey W, et al. Effect of lamivudine treatment on survival of 309 North American patients awaiting liver transplantation for chronic hepatitis B. Liver Transpl 2002; 8:433–439. 19. Andreone P, Biselli M, Gramenezi A, et al. Efficacy of lamivudine therapy for advanced liver disease in patients with precore mutant hepatitis B virus infection awaiting liver transplantation. Transplantation 2002; 74:1119–1124. 20. Fontana RJ, Hann H-WL, Perrillo RP, et al. Determinants of early mortality in patients with decompensated chronic hepatitis B treated with antiviral therapy. Gastroenterology 2002; 123:719–727. 21. Hann H-WL, Fontana RJ, Wright TL, et al. A United States compassionate use study of lamivudine treatment in nontransplantation candidates with decompensated hepatitis B virus-related cirrhosis. Liver Transpl 2003; 9:49–56. 22. Seehofer D, Rayes N, Steinmüller T, et al. Occurrence and clinical outcome of lamivudine resistant hepatitis B after liver transplantation. Liver Transpl 2001; 7:976–982. 23. Saab S, Kim M, Wright TL, et al. Successful orthotopic liver transplantation for lamivudine-associated YMDD mutant hepatitis B virus. Gastroenterology 2000; 119:1382–1384. 24. Rosenau J, Bahr MJ, Tillman HL, et al. Lamivudine and lowdose hepatitis B immune globulin for prophylaxis of hepatitis B reinfection after liver transplantation: possible role of mutations in the YMDD motif prior to transplantation as a risk factor for reinfection. J Hepatol 2001; 34:895–902. 25. Seehofer D, Rayes N, Naumann U, et al. Preoperative antiviral treatment and postoperative prophylaxis in HBV DNA positive patients undergoing liver transplantation. Transplantation 2001; 72:1381–1385. 26. Schiff ER, Lai CL, Hadziyannis S, et al. Adefovir dipivoxil therapy for lamivudine-resistant hepatitis in pre- and
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
post-liver transplantation patients. Hepatology 2003; 38:1419–1427. Angus P, Vaughan R, Xiong S, et al. Resistance to adefovir dipivoxil therapy associated with the selection of a novel mutation in the HBV polymerase. Gastroenterology 2003; 125:292–297. Mutimer D, Pillay D, Dragon E, et al. High pre-treatment serum hepatitis B virus titre predicts failure of lamivudine prophylaxis and graft re-infection after liver transplantation. J Hepatol 1999; 30:715–721. Naoumov NV, Lopes AR, Burra P, et al. Randomized trial of lamivudine versus hepatitis B immunoglobulin for long-term prophylaxis of hepatitis B recurrence after liver transplantation. J Hepatol 2001; 34:888–894. Markowitz JS, Martin P, Conrad AJ, et al. Prophylaxis against hepatitis B recurrence following liver transplantation using combination lamivudine and hepatitis B immune globulin. Hepatology 1998; 28:585–589. Yao FY, Osorio RW, Roberts JP, et al. Intramuscular hepatitis B immune globulin combined with lamivudine for prophylaxis against hepatitis B recurrence after liver transplantation. Liver Transpl Surg 1999; 5:491–496. Yoshida EM, Erb SR, Partovi N, et al. Liver transplantation for chronic hepatitis B infection with the use of combination lamivudine and low-dose hepatitis B immune globulin. Liver Transpl Surg 1999; 5:520–525. Angus PW, McCaughan GW, Gane EJ, et al. Combination lowdose hepatitis B immune globulin and lamivudine therapy provides effective prophylaxis against posttransplantation hepatitis B. Liver Transpl 2000; 6:429–433. Marzano A, Salizzoni M, Debernardi-Venon W, et al. Prevention of hepatitis B virus recurrence in cirrhotic patients treated with lamivudine before surgery and combined with passive immunoprophylaxis after liver transplantation. J Hepatol 2001: 34; 903-910. Han SH, Ofman J, Holt C, et al. An efficacy and costeffectiveness analysis of combination hepatitis B immune globulin and lamivudine to prevent recurrent hepatitis B after orthotopic liver transplantation compared with hepatitis B immune globulin monotherapy. Liver Transpl 2001; 6:741–748. McCaughan GW, Spencer J, Koorey D, et al. Lamivudine therapy in patients undergoing liver transplantation for hepatitis B virus precore mutant-associated infection: high resistance rates in treatment of recurrence but universal prevention if used as prophylaxis with very low-dose hepatitis B immune globulin. Liver Transpl Surg 1999; 6:512–519. Sanchez-Fueyo A, Rimola A, Grande L, et al. Hepatitis B immunoglobulin discontinuation followed by hepatitis B virus vaccination: a new strategy in the prophylaxis of hepatitis B virus recurrence after liver transplantation. Hepatology 2000; 31:496–501. Sanchez-Fueyo A, Martinez-Bauer E, Rimola A. Hepatitis B vaccination after liver transplantation. Hepatology 2002; 36:257. Angelico M, Di Paolo D, O Trinito M. Failure of a reinforced triple course of hepatitis B vaccination in patients transplanted for HBV-related cirrhosis. Hepatology 2002; 35:176–181. Bienzle U, Gunther M, Neuhaus R, et al. Immunization with an adjuvant hepatitis B vaccine after liver transplantation for hepatitis B virus-related disease. Hepatology 2003; 38:811–819. Dodson SF, de Dodson SF, de Vera ME, et al. Lamivudine after hepatitis B immune globulin is effective in preventing hepatitis B recurrence after liver transplantation. Liver Transpl 2000; 6:434–439. Buti M, Prieto M, Casafont F, et al. A randomized study comparing lamivudine monotherapy after a short course of hepatitis B immune globulin (HBIg) and lamivudine with longterm lamivudine plus HBIg in the prevention of hepatitis B virus
991
Section VIII. Liver Transplantation
43. 44.
45.
46.
47.
48.
49.
50.
51. 52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
992
recurrence after liver transplantation. J Hepatol 2003;38:811–817. Berenguer M, Wright TL. Treatment of recurrence of hepatitis B in transplant patients. J Hepatol 2003; 39: S190–S193. Perrillo R, Schiff E, Yoshida E, et al. Adefovir dipivoxil for the treatment of lamivudine-resistant hepatitis B mutants. Hepatology 2000; 32:129–134. Bock CT, Tillmann HL, Torresi J, et al. Selection of hepatitis B virus polymerase mutants with enhanced replication by lamivudine treatment after liver transplantation. Gastroenterology 2002; 122:264–273. Roche B, Samuel D, Feray C, et al. Retransplantation of the liver for recurrent hepatitis B virus infection: the Paul Brousse experience. Liver Transpl Surg 1999; 5:166–174. Davis GL, Albright JE, Cook SF, Rosenberg DM. Projecting future complications of chronic hepatitis C in the United States. Liver Transpl 2003; 9:331–338. Berenguer M, Prieto M, San Juan F, et al. Contribution of donor age to the recent decrease in patient survival among HCV-infected liver transplant recipients. Hepatology 2002; 36:202–210. Forman LM, Lewis JD, Berlin JA, et al. The association between hepatitis C infection and survival after orthotopic liver transplantation. Gastroenterology 2002; 122:889–896. Gane E. The natural history and outcome of liver transplantation in hepatitis C virus-infected recipients. Liver Transpl 2003; 9 (Suppl 3):S28–S34. Berenguer M. What determines the natural history of recurrent hepatitis C? J Hepatol 2005; 42:448–479. Berenguer M, Wright LT. Treatment strategies for hepatitis C: intervention prior to liver transplant, pre-emptively or after established disease. Clin Liver Dis 2003; 7:631–650. Wiesner RH, Sorrell M, Villamil F. Report of the First International Liver Transplant Society consensus conference on liver transplantation and hepatitis C. Liver Transpl 2003; 9 (Suppl 3):S1–S9. Feray C, Caccamo L, Alexander GJM, et al. European collaborative study on factors influencing the outcome after liver transplantation for hepatitis C. Gastroenterology 1999; 117:619–622. McCaughan GW, Zekry A. Mechanisms of HCV reinfection and allograft damage after liver transplantation. J Hepatol 2004; 40:368–374. Berenguer M, Ferrell L, Watson J, et al. HCV-related fibrosis progression following liver transplantation: increase in recent years. J Hepatol 2000; 32:673–684. Berenguer M, Aguilera V, Prieto M, et al. Delayed onset of severe hepatitis C-related liver damage following liver transplantation: a matter of concern? Liver Transpl 2003; 9:1152–1158. Berenguer M, Ferrell L, Watson J, et al. HCV-related fibrosis progression following liver transplantation: increase in recent years. J Hepatol 2000; 3:673–684. Neumann UP, Berg T, Bahra M, et al. Long-term outcome of liver transplants for chronic hepatitis C: a 10-year follow-up. Transplantation 2004; 77:226–231. Prieto M, Berenguer M, Rayón M, et al. High incidence of allograft cirrhosis in hepatitis C virus genotype 1b infection following transplantation: relationship with rejection episodes. Hepatology 1999; 29:250–256. Berenguer M, Crippin J, Gish R, et al. A model to predict severe HCV-related disease following liver transplantation. Hepatology 2003; 38:34–41. Charlton M, Seaberg E, Wiesner R, et al. Predictors of patient and graft survival following liver transplantation for hepatitis C. Hepatology 1998; 28:823–830. Burak KW, Kremers WK, Batts KP, et al. Impact of cytomegalovirus infection, year of transplantation, and donor age on outcomes after liver transplantation for hepatitis C. Liver Transpl 2002; 8:362–369.
64. Berenguer M, Prieto M, Rayón JM, et al. Natural history of clinically compensated HCV-related graft cirrhosis following liver transplantation. Hepatology 2000; 32:852–858. 65. Charlton M. Liver biopsy, viral kinetics, and the impact of viremia on severity of hepatitis C virus recurrence. Liver Transpl 2003; 9 (Suppl 3):S58–S62. 66. McCaughan GW, Zekry A. Impact of immunosuppression on immunopathogenesis of liver damage in hepatitis C virusinfected recipients following liver transplantation. Liver Transpl 2003; 9 (Suppl 3):S21–S27. 67. Lake JR. The role of immunosuppression in recurrence of hepatitis C. Liver Transpl 2003; 9:S63–S66. 68. Garcia-Retortillo M, Forns X, Feliu A, et al. Hepatitis C virus kinetics during and immediately after liver transplantation. Hepatology 2002; 35:680–687. 69. Lopez-Labrador FX, Berenguer M, Sempere A, et al. Genetic variability of hepatitis C virus NS3 protein in human leukocyte antigen-A2 liver transplant recipients with recurrent hepatitis C. Liver Transpl 2004; 10:217–227. 70. Wali M, Harrison RF, Gow PJ, Mutimer D. Advancing donor liver age and rapid fibrosis progression following transplantation for hepatitis C. Gut 2002; 51:248–252. 71. Machicao VI, Bonatti H, Krishna M, et al. Donor age affects fibrosis progression and graft survival after liver transplantation for hepatitis C. Transplantation 2004; 77:84–92. 72. Berenguer M, Rayón M, Prieto M, et al. Are post-transplantation protocol liver biopsies useful in the long-term? Liver Transpl 2001; 7:790–796. 73. Sebagh M, Rifai K, Feray C, et al. All liver recipients benefit from the protocol 10-year liver biopsies. Hepatology 2003; 37:1293–1301. 74. Sugawara Y, Makuuchi M. Should living donor liver transplantation be offered to patients with hepatitis C virus cirrhosis? Second forum in liver transplantation. J Hepatol 2005; 42. 75. Bizollon T, Admed SNS, Radenne S, et al. Long-term histologic improvement and clearance of intrahepatic hepatitis C virus RNA following sustained response to interferon-ribavirin combination therapy in liver transplant patients with hepatitis C recurrence. Gut 2003; 52:283–287. 76. Abdelmalek MF, Firpi RJ, Soldevila-Pico C, et al. Sustained viral response to interferon and ribavirin in liver transplant recipients with recurrent hepatitis C. Liver Transpl 2004; 10:199–207. 77. Feray C, Gigou M, Samuel D, et al. Incidence of hepatitis C in patients receiving different preparations of hepatitis B immunoglobulins after liver transplantation. Ann Intern Med 1998; 128:810–816. 78. Wright TL. Treatment of patients with hepatitis C and cirrhosis. Hepatology 2002; 36 (Suppl 1):S185–S194. 79. Garcia-Retortillo M, Forns X. Prevention and treatment of hepatitis C virus recurrence after liver transplantation. J Hepatol 2004; 41:2–10. 80. Crippin JS, Sheiner P, Terrault NA, et al. A pilot study of the tolerability and efficacy of antiviral therapy in patients awaiting liver transplantation for hepatitis C. Liver Transpl 2002; 8: 350–355. 81. Everson GT. Treatment of patients with hepatitis C virus on the waiting list. Liver Transpl 2003; 9 (11):S90–S94. 82. Forns X, Garcia-Retortillo M, Serrano T, et al. Antiviral therapy of patients with decompensated cirrhosis to prevent recurrence of hepatitis C after liver transplantation. J Hepatol 2003; 39: 389–396. 83. Terrault N. Prophylactic and preemptive therapies for hepatitis C-virus infected patients undergoing liver transplantation. Liver Transpl 2003; 9 (Suppl 3):S95–S100. 84. Chalasani N, Manzarbeitia C, Teperman L, et al. Peginterferon alfa 2a for recurrence of hepatitis C after liver transplantation: two randomised controlled trials. Hepatology 2005; 41:289–298.
Chapter 52 MANAGEMENT OF RECURRENT VIRAL HEPATITIS B AND C
85. Lavezzo B, Franchello A, Smedile A, et al. Treatment of recurrent hepatitis C in liver transplants: efficacy of a six versus twelve month course of interferon alfa 2b with ribavirin. J Hepatol 2002; 37:247–252. 86. Samuel D, Bizollon T, Feray C, et al. Interferon-alpha 2b plus ribavirin in patients with chronic hepatitis C after liver transplantation: a randomized study. Gastroenterology 2003; 124:642–650. 87. Jain AB, Eghtesad B, Venkataramanan R, et al. Ribavirin dose modification based on renal function is necessary to reduce hemolysis in liver transplant patients with hepatitis C virus infection. Liver Transpl 2002; 8:1007–1013. 88. Neff GW, O’Brien CB, Cirocco R, et al. Prediction of sustained virological response in liver transplant recipients with recurrent hepatitis C virus following combination pegylated interferon alfa2b and ribavirin therapy using tissue hepatitis C virus reverse transcriptase polymerase chain reaction testing. Liver Transpl 2004; 10:595–598. 89. Saab S, Kalmaz D, Gajjar NA, et al. Outcomes of acute rejection after interferon therapy in liver transplant recipients. Liver Transpl 2004; 10:859–867. 90. Todd-Stravitz R, Shiffman ML, Sanyal AJ, et al. Effects of interferon treatment on liver histology and allograft rejection in patients with recurrent hepatitis C following liver transplantation. Liver Transpl 2004; 10:850–858. 91. Gopal DV, Rosen HR. Duration of antiviral therapy for cholestatic HCV recurrence may need to be indefinite. Liver Transpl 2003; 9:348–353. 92. Dumortier J, Scoaxec JY, Chevallier P, Boillot O. Treatment of recurrent hepatitis C after liver transplantation: a pilot study of
93.
94.
95.
96.
97.
98. 99.
100.
101.
peginterferon alfa 2b and ribavirin combination. J Hepatol 2004; 40:669–674. Rodríguez-Luna H, Khatib A, Sharma P, et al. Treatment of recurrent hepatitis C infection after liver transplantation with combination of pegylated interferon alfa 2b and ribavirin: an open label series. Transplantation 2004; 77:190–194. Saab S, Ly D, Han SB, et al. Is it cost-effective to treat recurrent hepatitis C infection in orthotopic liver transplantation patients? Liver Transpl 2002; 8:449–457. Watt KD, Lyden ER, McCashland TM. Poor survival after liver retransplantation: is hepatitis C to blame? Liver Transpl 2003; 9:1019–1024. Biggins SW, Beldecos A, Rabkin JM, Rosen HR. Retransplantation for hepatic allograft failure: prognostic modeling and ethical considerations. Liver Transpl 2002; 8:313–322. Berenguer M, Prieto M, Palau A, et al. Severe recurrent hepatitis C following liver retransplantation for HCV-related graft failure. Liver Transpl 2003; 9:228–235. Roayaie S, Schiano TD, Thung SN, et al. Results of retransplantation for recurrent hepatitis C. Hepatology 2003; 38:1428–1436. Facciuto M, Heidt D, Guarrera J, et al. Retransplantation for late liver graft failure: predictors of mortality. Liver Transpl 2000; 6:174–179. Ghobrial RM. Retransplantation for recurrent hepatitis C in the model for end-stage liver disease era: how should we or shouldn’t we? Liver Transpl 2003; 9:1025–1027. Rosen H, Prieto M, Casanovas-Taltavull T, et al. Validation and refinement of survival models for liver retransplantation. Hepatology 2003; 38:460–469.
993
Section VIII: Liver Transplantation
MANAGEMENT OF RECURRENT NON-VIRAL CONDITIONS
53
James Neuberger Abbreviations AIH autoimmune hepatitis AMA antimitochondrial antibodies ERC endoscopic retrograde cholangiography
HLA MRC NAFLD
human leukocyte antigen magnetic resonance cholangiography non-alcoholic fatty liver disease
INTRODUCTION Recognition that the allograft may be affected by the same disease process that resulted in the failure of the native liver is of both clinical and academic importance. The recipient needs to be counseled about the possibility of recurrence and the possible impact on graft function and survival; the clinician needs to be aware of the potential of recurrence, to interpret the clinical, laboratory, radiological, and histological findings appropriately and, where appropriate, alter management. Finally, understanding which conditions recur in the allograft and those factors that are associated with recurrence may shed light on the pathogenesis of the disease. In this chapter, recurrence of non-viral and non-malignant diseases will be discussed, together with diagnosis and management.
DIAGNOSIS OF RECURRENT DISEASE IN THE ALLOGRAFT The criteria for the diagnosis of disease in the native liver may not be applicable in the allograft. The difference between the liver and host human leukocyte antigen (HLA) and other antigens and the effects of immunosuppression may modify the pattern of recurrent disease. Furthermore, the clinical, serological, and histological features of recurrent disease may be mimicked by other causes of graft damage. Thus the diagnosis of recurrent disease is often not straightforward. It is important that stringent criteria for the diagnosis of recurrent disease are agreed and followed (Table 53-1).
AUTOIMMUNE DISEASES Primary Biliary Cirrhosis (PBC) Diagnosis The criteria for diagnosis of recurrent PBC are shown in Table 532. Thus, the diagnosis is made primarily on the basis of histology. As with PBC in the native liver, the diagnostic histologic features of recurrent PBC may be found in the presence of normal liver tests. It is important to recognize that antimitochondrial antibodies (AMA), which, in the native liver, may antedate the clinical, biochemical, and histological features of PBC, will persist after liver
PBC PSC UDCA
primary biliary cirrhosis primary sclerosing cholangitis ursodeoxycholic acid
transplantation. The aberrant distribution of the dihydrolipoamide acetyl transferase (the antigen recognized by the AMA) in the biliary epithelial cells that is characteristic of PBC in the native liver may be seen in the graft as early as 7 days after transplantation. Titers of AMA may show a transient fall and then return to or exceed levels seen pretransplantation. Similarly, serum immunoglobulins may be elevated, especially the serum immunoglobulin M, and again this does not correlate with recurrence.1
Incidence The reported incidence of recurrent PBC varies in the literature; this variation is in part dependent on the criteria used to define PBC and whether protocol biopsies are used in the center. Histological features of recurrent PBC may be found in the presence of normal liver tests. Our own data (Figure 53-1) show that PBC recurs in 17% of patients2 at an overall median time to detection of recurrence of about 3 years.3 The prevalence of recurrent PBC is between 8 and 20% at 5 years and 20–30% at 10 years.2,4,5 Our recent studies suggest that immunosuppression with ciclosporin is associated with a longer time to recurrence compared with tacrolimus (median time to diagnosis of recurrence was 62 months for those on tacrolimus and 123 months on ciclosporin).6
Treatment There is no definitive treatment for those with recurrent PBC. However, most centers offer treatment with ursodeoxycholic acid (UDCA) 10–15 mg/kg per day as this bile acid is recommended for treatment in those with PBC in the native liver. There is no evidence whether this alters the natural history of recurrent disease. UDCA is well tolerated and may affect absorption of other medication; this has not been a significant problem in practice. Given the observation that recurrence is more rapid in those on tacrolimus compared to those on ciclosporin medication, it may be that switching between calcineurin inhibitors may be of benefit. Likewise, both azathioprine and mycophenolate have been suggested to be of benefit in the native liver in slowing progression, so modification of the immunosuppressive regime may be of benefit.
Outcome In the short to medium term, recurrent disease does not affect patient or graft survival. In our own series of 400 patients, only two developed graft failure over a 10-year follow-up.
995
Section VIII. Liver Transplantation
Table 53-1. Some causes of graft damage that may complicate the diagnosis of recurrent disease in the allograft
0.8 Recurrence of PBC
Infection Drug toxicity Graft damage Immune-mediated Acute rejection Chronic rejection De novo autoimmune disease Non-immune-mediated Ischemia Reperfusion Infection Viral Bacterial Mycobacterial
1.0
0.6 0.4 0.2 0.0 –0.2 –20
0
20
40
60
80
100
120
Months Table 53-2. Criteria for the diagnosis of recurrent primary biliary cirrhosis (PBC) Transplantation for PBC Characteristic histologic features of PBC Mononuclear inflammatory infiltrates Lymphoid aggregates Epithelioid granulomas Bile duct damage Persistence of antimitochondrial antibodies Elevated immunoglobulins Exclusion of other causes of graft damage Note: normal liver tests do not exclude the diagnosis of recurrent PBC
Autoimmune Hepatitis (AIH) Diagnosis The criteria for the diagnosis of recurrent AIH are shown in Table 53-3.7 It should be noted that all criteria should be met to establish the diagnosis. None of the features of recurrent AIH is specific to the diagnosis and all may be found in other conditions. Few studies have been done to look at the target antigens in recurrent AIH but one study suggested that the immune response was directed to a graft rather than host antigen,8 raising the possibility that this is, in fact, rejection rather than a true autoimmune recurrence. Application of the scoring systems developed to define AIH in the native liver to the transplant situation is not validated and therefore this system should not be used in the diagnosis of recurrent AIH.
Incidence The incidence of recurrent AIH is difficult to determine as few series have used stringent criteria to make the diagnosis.9–12 Our own series of 93 patients suggests that 13 have developed recurrent disease and of these, three have developed graft failure as a consequence.13
Treatment The approach to treatment is generally to increase immunosuppression. The first approach is to add corticosteroids (such as prednisolone 20 mg/day). If this does not result in resolution of the serological and histological features, then, for those on ciclosporinbased treatment, switching to tacrolimus should be considered.14
996
Figure 53-1. Risk of survival free of recurrence of primary biliary cirrhosis confirmed histologically in patient receiving ciclosporin and tacrolimus (data from Birmingham, UK).
Table 53-3. Criteria for the diagnosis of recurrent autoimmune hepatitis (after Manns18) Liver transplant for autoimmune hepatitis Autoantibodies in significant titer (<1 : 40) Sustained rise in serum aminotransferase activity (more than twice normal) Elevated serum immunoglobulins Compatible liver histology Corticosteroid dependency Exclusion of other causes of graft dysfunction (such as rejection and hepatitis C virus infection)
Table 53-4. Criteria for the diagnosis of recurrent primary sclerosing cholangitis (PSC) Transplant for PSC Multiple non-anastomotic biliary strictures Exclusion of other causes of sclerosing cholangitis (see Box 53-5) “Cholestatic” liver tests Characteristic liver histology may be seen but absence of characteristic features (such as pericentral fibrosis) does not exclude the diagnosis
Addition of mycophenolate 2 g/day may be of benefit. The roles of sirolimus and leflunomide is not established. Retransplantation should be considered for those with graft failure.
Outcome The response to treatment in this group of patients is, according to the literature, variable. About one-half seem to respond to additional immunosuppression.10,11,15 Part of the variation may be due to the differing criteria used to give the diagnosis of recurrent disease.
Primary Sclerosing Cholangitis (PSC) Diagnosis The criteria for the diagnosis of recurrent PSC are listed in Table 53-4. There are many causes of non-anastomotic strictures in the
Chapter 53 MANAGEMENT OF RECURRENT NON-VIRAL CONDITIONS
Survival Functions
Table 53-5. Causes of non-anastomotic biliary strictures 1.1
allograft (Table 53-5) and therefore distinction between recurrent primary sclerosing cholangitis and secondary sclerosing cholangitis may be difficult, if not impossible, to make. Most patients grafted for PSC will have a Roux en Y loop so there will be an increased risk of biliary infection. The diagnosis of recurrent PSC is made by showing multiple intra- and extrahepatic biliary strictures and exclusion of other causes of non-anastomotic strictures. These strictures can be shown by imaging the biliary tree with either magnetic resonance cholangiography (MRC) or percutaneous or endoscopic retrograde cholangiography (ERC). While MRC has the advantage that it is safer, the specificity and sensitivity have not been fully determined. ERC may not be technically possible as most patients grafted for PSC have a Roux loop. Liver histology is of limited benefit since the characteristic diagnostic features of PSC (fibrous cholangitis and fibro-obliterative disease) are seen in less than half the cases of PSC.
Incidence The probability of developing recurrent PSC is shown in Figure 532. Our own studies (which have not been confirmed) suggest that PSC is less likely to recur in those who have had a colectomy either before or during the transplant procedure, whereas colectomy posttransplant does not protect against recurrence.16 Although colectomy is well tolerated, these observations do not indicate the benefit of prophylactic colectomy to prevent recurrent PSC.
Treatment There is no specific treatment for non-anastomotic biliary strictures. Most centers offer these patients UDCA 20–25 mg/kg per day since this treatment has been suggested to be of benefit in those with PSC in the native liver. For those with associated ulcerative colitis, use of UDCA has the additional benefit of reducing the risk of developing colonic adenoma and carcinoma. Immunosuppression appears to have little effect on PSC recurrence.17 For those with cholestatic symptoms, consideration should be given to treatment of the symptoms of itching and associated complications such as osteopenia, fat-soluble vitamin maladsorption, and nutrition.
Outcome Unlike recurrent PBC, the prognosis of recurrent PSC is less good, with graft failure developing in 3%.16–18
Implications of Recurrent Autoimmune Diseases The observation that autoimmune disease may recur in the allograft is intriguing. The simplistic definition of an autoimmune disease is
Initial immunity 1.0 Disease Free Survival
Biliary infection Ischemia Hepatic artery thrombosis ABO-incompatible graft Reperfusion injury Graft-versus-host disease
Tac Tac-censor
0.9
Cya 0.8
Cya-censor
0.7 0.6 0.5 0 –100
0
100
200
300
Disease free survival (months) Figure 53-2. Rate of recurrence of primary sclerosing cholangitis (data from Birmingham, UK).
where the host mounts an immune-mediated tissue-damaging attack against host antigens; treatment with immunosuppressive agents will usually control such immune-mediated damage. Following transplantation, the host is immunosuppressed yet organ damage may still develop. Host HLA genotype is a major risk factor for developing AIH, but there is no clear consensus whether donor or recipient HLA predisposes to AIH recurrence. Recurrent AIH occurs in a small proportion of recipients and may be similar to de novo AIH that is seen in those individuals, especially children. As with recurrent AIH, treatment with increased immunosuppression or with corticosteroids may not be effective. Whether recurrent AIH is really autoimmune is not clear as there is no good evidence to show that the target antigens are self rather than host antigens; it may be that this is just another form of graft rejection. Where there is an immune-mediated response to graft antigens, both acute and chronic rejection may be associated with organ non-specific autoantibodies. With PBC, the implications of recurrent disease, with respect to disease pathophysiology, are uncertain. The report (not confirmed) that the aberrant distribution of the target of the AMA (PDC-E2) is seen within 1 week in the allograft of those grafted for PBC suggests that some extrahepatic factor triggers PBC: whether this is infectious (bacterial, viral, or other) remains controversial. However, aberrant distribution of E2 does not necessarily trigger overt liver damage (at least in the relatively short to medium term). That several reports suggest that the recurrence is greater in those receiving tacrolimus rather than ciclosporin raises other pointers to the pathogenesis of PBC: again, implications of this observation are not clear. However, the situation is complicated by several observations that report the features of recurrent PBC becoming overt only after reduction in calcineurin inhibitors. The implications of recurrent PSC are also uncertain. The observations of the effects of colectomy peri- or pretransplantation need either confirming or refuting. It has been suggested that
997
Section VIII. Liver Transplantation
this reflects the circulation of some gut-specific lymphocytes to the liver.
METABOLIC AND OTHER MISCELLANEOUS CONDITIONS Hemochromatosis The diagnosis of the disease (as opposed to the genotype) of genetic hemochromatosis is made on the basis of the appropriate genotype and evidence of iron overload in the absence of other causes of iron overload. Studies of iron metabolism in the liver allograft recipient are few and difficult to interpret: following liver transplantation, there may be additional iron input (associated blood transfusion) and redistribution of iron stores. Furthermore, estimation of iron stores is made difficult as serum ferritin, an acute-phase protein, may not accurately reflect total body iron stored. Current evidence suggests that the main defect in genetic hemochromatosis results in excess absorption of iron by the enterocyte and that the liver is secondary in the pathogenesis of the disease (Chapter 67).19,20 It is likely that the transplant will not correct the defect and the recipient will remain at risk of iron overload and its associated complications. A recent study of 22 patients grafted for genetic hemochromatosis21 evaluated 11 patients for evidence of iron indices and concluded that reaccumulation of iron in the graft is rare and there was no evidence of iron excess as reflected by both serum ferritin and transferrin saturation. However, median follow-up was only 4 years and the maximum follow-up was 11 years; thus it would seem prudent to monitor these patients with serum ferritin and transferring saturation in the longer term for evidence of iron toxicity, and consider venesection if there is evidence of progressive iron overload.
Budd–Chiari Syndrome Budd–Chiari syndrome is a rare indication for liver transplantation. There are many factors that may predispose to the thrombosis of the hepatic veins; in some cases, where the major predisposing factor lies within the liver, such as protein C or protein S deficiency, the liver transplant will effectively cure the patient. However, in some cases, such as where there is an underlying myeloproliferative disorder, these risk factors will persist and the recipient will remain at risk of thrombosis of the hepatic veins or other vessels. Thus, the King’s College Hospital group (London, UK) reported that Budd–Chiari syndrome recurred in two of 19 patients grafted for Budd–Chiari syndrome, at 4 months and 7 years posttransplant,22 and the Dallas group (Texas, USA) reported recurrent disease in one of 17 patients 124 months post-transplant.23 A variety of approaches have been adopted to reduce the risk of recurrence: some centers will use prophylactic anticoagulation with warfarin and others antiplatelet therapy (such as aspirin or clopidogrel); others have used hydroxyurea. However, recurrent Budd–Chiari may develop despite full anticoagulation.
Sarcoidosis Sarcoidosis is a systemic disease of uncertain etiology. Significant liver involvement is relatively uncommon but occasionally hepatic sarcoidosis is an indication for transplantation. Occasional cases of
998
recurrent sarcoidosis have been described24 but recurrence is not inevitable; use of corticosteroids post-transplantation may modify the pattern of recurrence.
Erythropoietic Protoporphyria Erythropoietic protoporphyria is an inherited disorder of heme synthesis which, in a small proportion of patients, is associated with progressive liver failure. As the marrow is a site of excess protoporphyrin production, it is not surprising that liver transplantation may not be curative. There are several cases of graft damage, with cholestasis, fibrosis, associated with features such as photosensitivity and elevated levels of protoporphyrins in blood and feces.25 Use of cholestyramine may reduce the severity of clinical problems and plasmapheresis and intravenous heme albumin or hematin infusions may be of benefit in improving the biochemical and liver abnormalities.26,27
Amyloid Transthyretin amyloidosis is an accepted indication for liver transplantation. Since the major defect lies within the liver, transplantation is indicated in some cases, even though the liver is otherwise functionally normal. Survival after transplantation is good in the short term but longer-term follow-up is needed. Cardiac disease is the main cause of death.28 Generally, following liver transplantation there is usually some improvement in the neurological consequences and, for those transplanted in the early stages, there may be some improvement. In contrast, gastrointestinal disturbances are generally unchanged.29 Transplantation does not appear to affect oculoleptomeningeal amyloid deposition.30
Alveolar Echinococcosis Alveolar echinococcosis is occasionally an indication for palliative liver transplantation. Reinfection of the graft may occur, although the use of drugs such as benzimidazole may reduce the risk of both graft infection and extrahepatic disease.31
Giant-Cell Hepatitis Giant-cell hepatitis is a rare indication for liver transplantation in adults. The condition may recur in the allograft.32 Ribavirin may be an effective treatment in the allograft.33
Alcoholic Liver Disease A return to alcohol consumption after liver transplantation probably occurs in the majority of those transplanted for alcoholic liver disease. The duration of pretransplant abstinence does not reliably predict recurrence;34,35 in contrast, a family history of alcoholism is a positive predictor factor for relapse.36 However, liver damage, either as a direct result of alcohol toxicity or indirectly as a consequence of non-compliance, is rare, affecting less than 5% of grafts at 5 years.34,37 Recognition of return to alcohol consumption may be difficult. Questioning of the patient and their family and support group may be helpful. Serological testing, using either carbohydrate deficient transferrin or gamma-glutamyltransferase levels may be helpful but
Chapter 53 MANAGEMENT OF RECURRENT NON-VIRAL CONDITIONS
are not specific.38 Some will directly measure alcohol in breath, blood, or urine but these approaches will detect only recent intake. Treatment is supportive. The use of agents such as naltrexone or acamprosate is as yet uncertain.39
Non-Alcoholic Fatty Liver Disease (NAFLD) There are many causes of fatty liver; these are discussed in Chapter 55. Where the metabolic cause persists, such as in those with the metabolic syndrome, it is not surprising that fatty liver will affect the graft.40,41 Treatment is of the underlying condition. NAFLD may recur in up to 100% cases at 5 years.42,43 The diagnosis of NAFLD is suggested by the characteristic ultrasound pattern, showing an echo-bright liver and confirmed by liver histology. Other causes of fatty liver, such as recurrent or de novo hepatitis C virus infection or alcohol consumption, should be excluded. Liver tests are of little diagnostic help. The treatment is of the underlying cause: weight loss and control of diabetes, if present, is indicated. Whether medication, such as metformin or glitazones, is effective has not been demonstrated in the liver allograft recipient. In severe cases, Roux en Y gastric bypass has been undertaken with associated improvement in liver function.44
Cryptogenic Cirrhosis Inevitably, the cause of cryptogenic cirrhosis is unknown. Indeed, it is likely that there are several causes that will have a varying outcome after transplantation.41,45 Several studies have suggested that cryptogenic cirrhosis may recur in the allograft. Some of these may represent “burnt-out” NAFLD since some (up to 40%) of those grafted for so-called cryptogenic cirrhosis will develop fatty liver disease that may progress to fibrosis and cirrhosis.41,45,46
Diseases where Liver Transplantation is Contraindicated Because of Recurrence/Persistence There are many metabolic diseases where the liver is affected and the patient may develop end-stage disease, but liver replacement is not indicated because the metabolic defect lies wholly or partly outside the liver or the degree of extrahepatic disease is so marked that liver transplantation alone will not allow the patient to return to an acceptable quality of life. Conversely, there are some indications for transplantation where the metabolic defect will not be partly or wholly corrected by liver transplantation but disease recurrence can be either prevented or controlled following liver replacement. Diseases where transplantation is not indicated include sea-blue histiocytic syndrome, mitochondrial cytopathies with multisystemic involvement, and Alpers disease with valproate toxicity.
REFERENCES 1. Haydon GH, Neuberger J. Primary biliary cirrhosis: an infectious disease? Gut 2000; 47:586–588. 2. Garcia R, Garcia C, McMaster P, Neuberger JM. Transplantation for primary biliary cirrhosis: retrospective analysis of 400 patients in a single center. Hepatology 2001; 33:22–27.
3. Balan V, Banns BM, Porayko M, et al. Histological evidence for recurrence of primary biliary cirrhosis after liver transplantation. Hepatology 1993; 18:1392–1398. 4. Haagsma EB. Clinical relevance of recurrence of primary biliary cirrhosis after liver transplantation. Eur J Gastroenterol Hepatol 1999; 11:639–642. 5. Levitsky J, Hart J, Cohen SM, Te HS. The effect of immunosuppressive regimes on the recurrence of primary biliary cirrhosis after liver transplantation. Liver Transpl 2003; 9:733–736. 6. Neuberger J, Gunson B, Hubscher SG, Nightingale PG. Immunosuppression affects the rate of recurrent primary biliary cirrhosis after liver transplantation. Liver Transpl 2004; 10:488–491. 7. Manns MP, Bahr MJ. Recurrent autoimmune hepatitis after liver transplantation – when self becomes non-self. Hepatology 2000; 32:868–870. 8. Aguilera I, Wichman I, Sousa JM, et al. Antibodies against glutathione-S-transferase T1 (GSTT1) in patients with de novo autoimmune hepatitis following liver transplantation. Clin Exp Immunol 2001; 126:535–539. 9. Hubscher SG. Recurrent autoimmune hepatitis after liver transplantation: diagnostic criteria, risk factors and outcome. Liver Transpl 2001; 7:285–291. 10. Ayata G, Gordon FD, Lewis WD, et al. Liver transplantation for autoimmune hepatitis: a long-term pathologic study. Hepatology 2000; 32:185–192. 11. Ratziu V, Samuel D, Sebagh M, et al. Long-term follow-up after liver transplantation for autoimmune hepatitis: evidence of recurrence of primary disease. J Hepatol 1999; 30:131–141. 12. Bahar RJ, Yanni GS, Martin MG, et al. Orthotopic liver transplantation for autoimmune hepatitis and cryptogenic chronic hepatitis in children. Transplantation 2001; 72:829–833. 13. Milkiewicz P, Hubscher S, Skiba G, et al. Recurrence of autoimmune hepatitis after liver transplantation. Transplantation 1999; 68:253–256. 14. Hurtova M, Duclos-Vallee JC, Johanet C, et al. Successful tacrolimus therapy for a severe recurrence of type 1 autoimmune hepatitis in a liver allograft recipient. Liver Transpl 2001; 7:556–558. 15. Neuberger J. Transplantation for autoimmune hepatitis. Semin Liver Dis 2002; 22:379–386. 16. Vera A, Moledina S, Gunson BK, et al. Risk factors for recurrence of primary sclerosing cholangitis of liver allograft. Lancet 2002; 360:1943–1944. 17. Kugelmas M, Spiegelman P, Osgood MJ, et al. Different immunosuppressive regimens and recurrence of primary sclerosing cholangitis after liver transplantation. Liver Transpl 2003; 9:727–732. 18. Graziadei IW, Wiesner RH, Batts KP, et al. Recurrence of primary sclerosing cholangitis following liver transplantation. Hepatology 1999; 29:1050–1056. 19. Kowdley KV. Liver transplantation: an in vivo model for the pathophysiology of hemochromatosis? Hepatology 2004; 39:1495–1498. 20. Pietrangelo A. Hereditary hemochromatosis – a new look at an old disease. N Engl J Med 2004; 250:2383–2397. 21. Crawford DH, Fletcher LM, Hubscher SG, et al. Patient and graft survival after liver transplantation for hereditary hemochromatosis. Hepatology 2004; 39:1655–1662. 22. Srinivasan P, Rela M, Prachalias A, et al. Liver transplantation for Budd–Chiari syndrome. Transplantation 2002; 73:973–977. 23. Melear JM, Goldstein RM, Levy MF, et al. Hematologic aspects of liver transplantation for Budd–Chiari syndrome with special reference to myeloproliferative disorders. Transplantation 2002; 74:1090–1095.
999
Section VIII. Liver Transplantation
24. Hunt J, Gordon FD, Jenkins RL, et al. Sarcoidosis with selective involvement of a second liver allograft: report of a case and review of the literature. Mod Pathol 1999; 12:325–328. 25. Meerman L, Haagsma EB, Gouw AS, et al. Long-term follow-up after liver transplantation for erythropoietic protoporphyria. Eur J Gastroenterol Hepatol 1999; 11:431–438. 26. Do KD, Banner BF, Katz E, et al. Benefits of chronic plasmapheresis and intravenous heme-albumin in erythropoietic protoporphyria after orthotopic liver transplantation. Transplantation 2002; 73:469–472. 27. Dellon MS, Szczepiiorkowski ZM, Dzik WH, et al. Treatment of recurrent allograft dysfunction with intravenous hematin after liver transplantation for erythropoietic protoporphyria. Transplantation 2002; 73:911–915. 28. Herlenius G, Wilczek HE, Larsson M, et al. Ten years of international experience with liver transplantation for familial amyloidotic polyneuropathy: results from the Familial Amyloidotic Polyneuropathy World Transplant Registry. Transplantation 2004; 77:64–71. 29. Suhr OB. Impact of liver transplantation on familial amyloidotic polyneuropthay patients’ symptoms and complications. Amyloid 2003; 10 (Suppl 1):77–83. 30. Ando Y, Terazaki H, Namura M, et al. A different amyloid formation mechanism: de novo oculoleptomeningeal amyloid deposits after liver transplantation. Transplantation 2004; 77:345–349. 31. Bresson-Hadni S, Koch S, Beurton I, et al. Primary disease recurrence after liver transplantation for alveolar echinoccosis: long term evaluation in 15 patients. Hepatology 1999; 30:857–864. 32. Nair S, Baisden B, Boitnott J, et al. Recurrent, progressive giant cell hepatitis in two consecutive liver allografts in a middle-aged woman. J Clin Gastroenterol 2001; 32:454–456. 33. Hassoun Z, N’Guyen B, Cote J, et al. A case of giant cell hepatitis recurring after liver transplantation and treated with ribavirin. Can J Gastroenterol 2000; 14:729–731. 34. Neuberger J, Schulz KH, Day C, et al. Transplantation for alcoholic liver disease. J Hepatol 2002; 36:130–137.
1000
35. Pageaux GP, Bismuth M, Perney P, et al. Alcohol relapse after liver transplantation for alcoholic liver disease: does it matter? J Hepatol 2003; 38:629–634. 36. Jauhar S, Talwalkar JA, Schneekloth T, et al. Analysis of factors that predict relapse following liver transplantation. Liver Transpl 2004; 10:408–411. 37. Webb K, Neuberger J. Transplantation for alcoholic liver disease. Br Med J 2004; 329:63–64. 38. Berlakovich GA, Windhager T, Freundofer E, et al. Carbohydrate deficient transferrin for detection of alcohol relapse after orthotopic liver transplantation for alcoholic cirrhosis. Transplantation 1999; 67:1231–1235. 39. DiMartini A, Weinrieb R, Fireman M. Liver transplantation in patients with alcohol and other substance use disorders. Psychiatr Clin North Am 2002; 25:195–209. 40. Angelico F, Del Ben M, Francioso S, et al. Recurrence of insulin resistant metabolic syndrome following liver transplantation. Eur J Gastreoenterol Hepatol 2003; 15:99–102. 41. Burke A, Lucey MR. Non-alcoholic fatty liver disease, nonalcoholic steatohepatitis and orthotopic liver transplantation. Am J Transplant 2004; 4:686–693. 42. Ayata G, Gordon FD, Lewis WD, et al. Cryptogenic cirrhosis: clinicopathological findings at and after liver transplantation. Hum Pathol 2002; 33:1098–1104. 43. Contos MJ, Cales W, Sterling RK, et al. Development of nonalcoholic fatty liver disease after orthotopic liver transplantation for cryptogenic cirrhosis. Liver Transpl 2001; 7:363–373. 44. Duchini A, Brunson ME. Roux-en-Y gastric bypass for recurrent non-alcoholic steatohepatitis in liver transplant recipients with morbid obesity. Transplantation 2001; 72:156–159. 45. Berg T, Neuhaus R, Klein R, et al. Distinct enzyme profiles in patients with cryptogenic cirrhosis reflect heterogeneous causes with different outcomes after liver transplantation (OLT): a longterm documentation before and after liver transplantation. Transplantation 2002; 74:792–798. 46. Ong J, Yanossi ZM, Reddy V, et al. Cryptogenic cirrhosis and post transplantation non-alcoholic fatty liver disease. Liver Transpl 2001; 7:797–801.
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
54
THE LIVER IN PREGNANCY Rebecca W. Van Dyke Abbreviations ABC ATP-binding cassette AFLP acute fatty liver of pregnancy ALT alanine aminotransferase AST aspartate aminotransferase BSEP bile-salt export pump BSP sulfobromophthalein CoA coenzyme A CT computed tomography DIC disseminated intravascular coagulopathy
FNH GGTP HBV HCV HIV ICP LCHAD
focal nodular hyperplasia g-glutamyltransferase activity hepatitis B virus hepatitis C virus human immunodeficiency virus Intrahepatic cholestasis of pregnancy long-chain 3-hydroxyacyl-CoA dehydrogenase
INTRODUCTION
MRI MRP2 NTCP PFIC SAME TIPS
magnetic resonance imaging multidrug resistance-associated protein-2 Na+/taurocholate co-transporting polypeptide progressive familial intrahepatic cholestasis S-adenosyl-l-methionine transjugular intrahepatic portosystemic shunt
CHANGES IN LIVER FUNCTION Plasma Proteins
The pregnant state is associated with physiologic changes in hepatic function that may cause uncertainty about the presence or absence of liver disease. In addition, common disorders of the liver may present with unusual features during pregnancy. Finally, several types of liver disease are unique to pregnancy. The aspects of liver function and liver disease during pregnancy discussed in this chapter include: (1) normal changes in hepatic function during pregnancy; (2) unusual aspects of common liver disorders; (3) liver disorders related to pregnancy; and (4) liver disorders unique to pregnancy.
Serum albumin concentrations in pregnancy and during use of oral contraceptives are 10–60% below those in control women, reflecting the increase in plasma volume in pregnancy and a decrease in albumin synthesis.10–13 Similarly, plasma levels of antithrombin III fall during pregnancy or use of oral contraceptives, attributed to decreased hepatic synthesis (Table 54-1).12,14 Plasma concentrations of other serum proteins (e.g., ceruloplasmin, fibrinogen, thyroxinebinding globulin, transferrin) increase during pregnancy and with use of oral contraceptives, presumably due to increased hepatic synthesis.10,12,14,15 Prothrombin times remain normal during pregnancy.16
CHANGES IN LIVER ANATOMY AND FUNCTION DURING PREGNANCY
Plasma Lipids
LIVER ANATOMY AND HISTOLOGY Liver size and gross appearance do not change during normal pregnancy.1,2 Subtle changes in histologic appearance may be seen but are non-specific in nature. These changes include: (1) increased variability in hepatocyte size and shape; (2) granularity of hepatocyte cytoplasm; (3) more frequent cytoplasmic fat vacuoles in centrilobular hepatocytes; and (4) hypertrophied Kupffer cells.3,4 Hepatocytes in women during normal pregnancy also exhibit proliferation of the smooth and rough endoplasmic reticulum, enlarged, rodshaped and giant mitochondria with paracrystalline inclusions, and increased numbers of peroxisomes.5 Many of these changes are seen in women taking oral contraceptives.6
LIVER BLOOD FLOW Blood volume increases 40–50%,7,8 accompanied by an increase in cardiac stroke volume and output and a decrease in peripheral resistance.7 Hepatic blood flow is unaltered during pregnancy.9 Therefore, hepatic blood flow in late pregnancy amounts to a smaller fraction of cardiac output.
There is increased peripheral lipolysis of triglycerides, increased flux of free fatty acids through the liver, increased hepatic synthesis of triglycerides, and increased hepatic synthesis and secretion of verylow-density lipoprotein, high-density lipoprotein, and low-density lipoprotein during pregnancy.10,17–19 These changes are reflected in a progressive increase in plasma triglycerides (up to 300%), cholesterol (25–60%), and phospholipids during pregnancy.11,17,18 The increased hepatic fat in biopsies of some pregnant women may be due to these changes in triglyceride and lipoprotein metabolism.
Drug Metabolism 17-Ethynyl estrogens decrease cytochrome P450 activity, probably via suicide inactivation of the enzyme.20 Pharmacokinetic studies suggest impaired metabolism of promazine, pethidine, and cortisol,21,22 an increased clearance of phenytoin, carbamezepine, metoprolol, and ciclosporin23–26 in pregnant women, and depressed antipyrine metabolism in women receiving oral contraceptives.27 Estrogens and pregnancy have been shown to impair hepatic activity of glucuronosyltransferase activity while progestational agents induce hepatic mixed-function oxidase activity in animals28 and oral contraceptives increase morphine clearance in humans.29 Thus, preg-
1003
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Table 54-1. Liver Tests During Pregnancy
Serum bile acids
Sulfobromophthalein (BSP)
Progressive increase to 2–3-fold non-pregnant values Progressive increase in plasma retention 45 min after administration and decrease in plasma clearance.
Markers of hepatocellular damage Aminotransferases No change (AST/ALT) Gamma glutamyl Serum activity normal transpeptidase or low Markers of liver function Serum albumin Decrease (by 10–60%)
Pathophysiology Release of placental and bone isoenzymes Release of placental enzyme Mild impairment of bilirubin transport Impaired biliary secretion Decrease in biliary transport maximum (to 27–70% of nonpregnant values)
Impaired release
40 30 20 10 0
1000 800 600 400 200 0 Non pregnant
Hemodilution; decreased synthesis
Prothrombin time
No change
Markers of cirrhosis Spider angiomata Palmar erythema
Appear in up to 67% Appear in up to 67%
Vascular effect of estrogens
Other Anti-thrombin III
Decrease
Decreased hepatic synthesis
nancy and oral contraceptives appear to alter drug metabolism, but the effects on any one drug are not readily predictable. The clinical significance of these changes in humans remains unclear.
Bile Formation Organic anion transport, including bilirubin30 and sulfobromophthalein (BSP),2,31,32 is impaired during pregnancy or use of estrogens or oral contraceptives. These changes are likely due primarily to estrogen/pregnancy-induced decreases in the canalicular organic anion-transporting pump multidrug resistance-associated protein-2 (MRP2: ABC C2).33–35 Concentrations of bile salts in blood are within the normal range in most pregnant women, but levels of glycocholate, taurocholate, and chenodeoxycholate may rise progressively until term and exceed levels measured early in pregnancy by two- to threefold.36–38 Pregnancy or estrogen-induced decreases in bile-salt transport (and bile formation) are likely due to decreases in both sinusoidal (Na+/taurocholate co-transporting polypeptide (NTCP)) and canalicular (bile-salt export pump (BSEP), ATP-binding cassette (ABC) B11) bile-salt transporters.33–35,39–41
1004
50 Alkaline Phosphatase Units
Change during pregnancy
Markers of cholestasis Alkaline Progressive increase in phosphatase serum activity to 1.5–2.5¥ non-pregnant values Leucine Progressive increase aminopeptidase Serum bilirubin None to slight increase
Leucine Aminopeptidase Units
Test
First
Second Third Trimester
In labor
Post partum
Figure 54-1. Serum activities of two enzymes in normal non-pregnant women and in various stages of pregnancy. The broken lines indicate the highest value obtained for each enzyme in non-pregnant women. (Modified from Walker FB, Hoblit DL, Cunningham FG, et al. Gamma glutamyl transpeptidase in normal pregnancy. Obstet Gynecol 1974; 43:745.)
LIVER TESTS During normal pregnancy, alkaline phosphatase activity rises progressively, particularly during the last 4 months (Figure 54-1, Table 54-1). At term, 42–77% of women exhibit alkaline phosphatase activities above the upper limit of normal,11,13,31,42,43 although values rarely exceed twice the upper limit of normal.11,16,18,42 The origin of the elevated plasma alkaline phosphatase activity during pregnancy is the placenta and bone, not the liver.44,45 Plasma leucine aminopeptidase activity also rises during pregnancy46 (Figure 54-1) due to release of a placental enzyme. 5-Nucleotidase is reported either to increase during pregnancy or not to change.46,47 Plasma levels of g-glutamyltransferase activity (GGTP) may fall slightly during a normal pregnancy.13,46,48 Levels of GGTP in women who have viral hepatitis and who are also either late in pregnancy or taking oral contraceptives are inappropriately low.48 Serum aminotransferase activities (aspartate aminotransferase (AST)/alanine aminotransferase (ALT)) remain within the normal range during pregnancy,13,16,18,31 making them useful tests for identifying hepatocellular injury in these individuals. Due to hemodilution and lower serum albumin concentrations, serum bilirubin levels tend to be lower than normal during normal pregnancy.13
SKIN MANIFESTATIONS OF LIVER DISEASE Several cutaneous vascular changes, usually associated with chronic liver disease, often appear during pregnancy. Vascular spiders may
Chapter 54 THE LIVER IN PREGNANCY
appear as early as the second month of pregnancy.49,50 At term, vascular spiders can be observed on the face, neck, anterior chest, lower arms, and dorsum of the hand in up to 67% of European-American women and in up to 14% of African-American women.49,50 In contrast, spiders are seen in only 12% of non-pregnant EuropeanAmerican women. The spiders disappear in 75% of affected women by 7 weeks postpartum, although a few women, particularly those with frequent pregnancies, continue to exhibit vascular spiders indefinitely. Palmar erythema is observed during pregnancy in 63% of European-American women and 39% of African-American women,50 beginning in the second month of pregnancy and increasing in incidence to a peak at term. By 7 weeks postpartum, less than 10% of women continue to exhibit palmar erythema.
COINCIDENT OCCURRENCE OF LIVER DISEASE AND PREGNANCY This section comprises a brief review of the manifestations of a group of fairly common liver diseases in the pregnant woman. The
Table 54-2. Acute Viral Hepatitis in Pregnancya Feature
Characteristics
Incidence Etiology
0.026–1.2% of gestations 4 series: epidemics of NANB hepatitis 1 series: epidemic of hepatitis A 5 series: mix of viruses, 11–78% hepatitis B 3 series: mix of viruses in India, predominantly hepatitis E ~30% (range 0–70%b) 1.2% in the USA, Australia, Israel ~30% in India, Middle East ~25% (range 8–54%) ~8% in the USA, Australia, Israel ~50% in India, Middle East
Fulminant hepatic failure Maternal mortality Premature birth Fetal/neonatal death
NANB, non-A, non-B. a Data derived from 1346 patients in series published from 1962 to 1983 (representative references:53–60) b The high rates of fulminant hepatic failure derive from countries where hepatitis E is endemic, accounting for 25–75% of viral hepatitis in pregnancy.
reader is referred to other chapters for complete discussions of these disorders.
ACUTE VIRAL HEPATITIS In 17 reviews of jaundice during pregnancy, hepatitis, presumably viral in etiology, accounted for 40% of 654 cases of clinical jaundice.16,42,51 The true incidence of clinical and subclinical viral hepatitis in pregnancy is not known, but in 1970–1974 at Parkland Hospital in Dallas, an incidence of 1 in 700 deliveries was noted.52 The data in Table 54-253–60 indicate the important features of the disease for mother and fetus when women have sufficient symptoms to seek medical care at large medical centers.
Maternal Features The clinical manifestations of acute viral hepatitis in pregnant women do not differ from those noted in non-pregnant women or in men. Similarly, laboratory data, when reported, are similar to values in non-pregnant individuals. In developed countries, maternal mortality from fulminant viral hepatitis and liver failure is low. Acute viral hepatitis, however, is associated with a greater, albeit still small, rate of fetal loss or premature birth even in well-nourished mothers. In contrast, in India and the Middle East, pregnant women with hepatitis E appear to manifest more severe disease, with 25–70% developing fulminant hepatic failure (Table 54-2).56,58,60–64 Mortality of both mothers and babies in hepatitis E-induced fulminant hepatic failure approaches 100% in many series.
Vertical Transmission Vertical transmission of hepatitis viruses from acutely infected mothers to infants has been reported when mothers are viremic at the time of birth57,65–68 (Table 54-3). Although vertical transmission of hepatitis A is extremely rare, exposed infants may be given immune serum globulin and the first dose of hepatitis A vaccine immediately after delivery.65,66,68 Infants born to mothers with acute hepatitis B should receive hepatitis B immune globulin and hepatitis B vaccine as recommended for infants of chronically infected mothers.69,70 There is no known prophylaxis against vertical transmission of C, E, or non-A, non-B, non-C viruses, although hepatitis E is commonly transmitted vertically and may cause severe disease in infected infants.63,64,71
Table 54-3. Vertical Transmission of Hepatitis Viruses Virus
Maternal status
Rate of transmission to infant
Prophylaxis for infant
Hepatitis A
Active infection within 2 weeks before/after delivery
Rare but possible
Hepatitis B
HBsAg-positive
10–30%
HBsAg-positive and high viral DNA load (>1.2 ¥ 109 gEq/ml) HCV RNA-positive HCV RNA-positive and HIV positive HBsAg-positive and HDV-positive Active infection at time of birth
> 85%
1. Immune serum globulin and hepatitis A vaccine after delivery 2. Hepatitis A vaccine repeated at 5–6 months of age 1. HBIG and vaccine at birth 2. Vaccine repeated at 1 and 6 months of life 1. HBIG and vaccine as above 2. Consider lamivudine for mother before delivery None None Same as hepatitis B ? Immune serum globulin
Hepatitis C Hepatitis D Hepatitis E
8.5% Up to 30% Similar to hepatitis B 50–100%
HBIG, hepatitis B immunoglobulin; HCV, hepatitis C virus; HVD, hepatitis D virus.
1005
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
ALCOHOL AND PREGNANCY The unique aspect of alcohol ingestion during pregnancy is fetal involvement and the fetal alcohol syndrome. This syndrome generally includes facial abnormalities, congenital malformations, growth retardation, and central nervous system dysfunction. Liver involvement has also been found in some affected infants,72–74 including hepatomegaly and elevated serum levels of transaminases and alkaline phosphatase. Liver histology is usually abnormal, with varying degrees of fatty infiltration, centrizonal hepatocyte degeneration, perivenular sclerosis, portal and perisinusoidal fibrosis, and proliferation of bile ducts. One affected child exhibited cirrhosis and esophageal varices by the age of 8.74 These reports suggest that maternal ethanol intake can cause chronic liver disease in offspring. Analysis of additional cases will be needed to define the spectrum of liver pathology.
CHRONIC LIVER DISEASE Impact of Chronic Liver Disease on Pregnancy Women with well-controlled mild chronic hepatitis and normal liver function appear to have normal fertility and to tolerate pregnancy well without adverse fetal or maternal outcomes.75 However, women with active liver disease, significant liver dysfunction, and/or cirrhosis exhibit decreased fertility76–78 and may experience liver deterioration during pregnancy and have higher rates of spontaneous abortion, premature birth, and perinatal death. Indeed, women with alcoholic liver disease often exhibit severe and irreversible gonadal failure, amenorrhea, and infertility, and rarely become pregnant. Infants born alive, however, are generally normal and do well, although mothers with clinically significant liver disease are more likely to die before their children reach adulthood. Contraceptive options for women with chronic liver disease include sterilization, barrier methods, and progestin-containing contraceptives.79 Pregnancy-related issues for specific liver diseases are outlined below.
Autoimmune Hepatitis Women with autoimmune hepatitis treated with immunosuppressive therapy are surviving for longer periods of time and on therapy many regain fertility and some become pregnant. In general, women with well-controlled autoimmune hepatitis receiving immunosuppressive therapy appear to tolerate pregnancy fairly well.76,80–84 Modest deteriorations in liver tests, particularly the serum bilirubin and alkaline phosphatase, may occur. These changes in tests usually return to the patient’s previous baseline values after delivery and most likely represent the imposed cholestatic effects of pregnancy. Reports of severe flares, liver failure, and even death in women who stopped immunosuppressive therapy during pregnancy or who were not on therapy during pregnancy indicate that successful therapy should not be stopped during pregnancy and that patients need continued monitoring.76,81,84 Whether these flares are due to pregnancy per se is not known, and remission during pregnancy has also been reported.81–83,85 Autoimmune hepatitis is associated with increased fetal morbidity and mortality: 15 spontaneous abortions and 9 perinatal deaths were reported in 128 pregnancies.76,80–85 The infants born alive were healthy and did well. With use of immunosuppressive therapy, including azathioprine, it appears that women with autoimmune hepatitis can conceive and deliver healthy
1006
children with relative safety.81,82 Postpartum, women should be monitored closely as disease flares may also occur in the first few months after delivery, likely related to immune reactivation.81,82,84,85 Finally, autoimmune hepatitis may present in the early postpartum period.86
Wilson’s Disease Chelation therapy has allowed patients with Wilson’s disease to survive in good health into and through the reproductive years. Many such patients become pregnant and bear children. Amenorrhea, infertility, and spontaneous abortions are common in symptomatic untreated women (due in part to high tissue copper levels as well as to the effects of liver dysfunction), but therapy restores fertility and allows a normal reproductive life.87–89 Some women with Wilson’s disease, although satisfactorily treated with chelators, have liver disease, including cirrhosis, that antedates initiation of therapy. These women have increased fetal and maternal morbidity and mortality. In both normal women and women with Wilson’s disease, concentrations of copper and ceruloplasmin in serum and urine increase during pregnancy or use of estrogens.15,87,90 In women with Wilson’s disease, ceruloplasmin and copper concentrations in sera may double by the third trimester of pregnancy. The former may increase into the low-normal range. Currently patients are treated lifelong with D-penicillamine, trientine (triethylene tetramine dihydrochloride), and/or zinc.87,89,91 Discontinuing therapy with D-penicillamine during pregnancy has been associated with symptomatic, and occasionally fatal, flare-ups of disease activity.89,92 Although D-penicillamine is potentially teratogenic, in 153 babies born to 111 mothers receiving the drug for Wilson’s disease, there were only 2 miscarriages, 3 premature births, 1 baby with a chromosomal defect, and 1 with cleft palate.89 Trientine and zinc appear to be similarly well tolerated during pregnancy.87,89,91 It is recommended that treatment with D-penicillamine or trientine (0.75–1 g/day during the first two trimesters and 0.5 g/day during the third trimester) or zinc be continued throughout pregnancy.87,89,91,93 Also, because of the antipyridoxine effects of D-penicillamine, oral supplementation with pyridoxine is recommended.
Cirrhosis Women with cirrhosis can and occasionally do become pregnant, although pregnancy in these women is uncommon.77,78 Reports of at least 156 pregnancies in 125 women with cirrhosis of varying etiology have been published.80,94–107 Evaluating the actual risk of hepatic complications during pregnancy is difficult, however, as only one study94 identified a potential control group of non-pregnant, cirrhotic women. Similarly, few authors have compared rates of obstetric complications in women with cirrhosis to rates in women without liver disease. During the course of pregnancy, liver tests (most commonly serum bilirubin and alkaline phosphatase activity) were reported to deteriorate in 30–40% of cirrhotic women,80,95–97 but in two-thirds of these cases, postpartum tests returned to baseline values. Much of this apparent deterioration may, in fact, reflect the cholestatic effect of pregnancy.
Chapter 54 THE LIVER IN PREGNANCY
Maternal morbidity and mortality are high during pregnancy (10.5% mortality in the 115 reported cases). Development of jaundice, ascites, hepatic encephalopathy, and postpartum hemorrhage are also common (Table 54-4). Maternal deaths are primarily due to gastrointestinal hemorrhage from varices, with liver failure accounting for many of the remaining deaths (Table 54-4). This degree of morbidity and mortality may not differ greatly from the natural history of cirrhosis in these women. Borhanmanesh and Haghighi94 noted, over a 40-month period, 2 deaths among 9 pregnant cirrhotic women and 3 deaths among 12 age-matched, non-pregnant, cirrhotic women. Bleeding from esophageal varices occurs in 18–32% of pregnant women with cirrhosis, but in up to 50% of women known to have portal hypertension.97,108 Patients with a past history of variceal bleeding may or may not bleed again during pregnancy97,99–101 and the risk is reduced after successful portosystemic shunting (Table 54-4). During or before pregnancy it is reasonable to screen for varices to estimate the risks of bleeding and, if varices are large, consider use of beta-blockers or of prophylactic banding. However, there are no data to indicate whether prophylactic treatment of varices with propranolol, sclerotherapy, or banding reduces bleeding and mortality during pregnancy. Variceal bleeding can be treated with octreotide infusion, sclerotherapy, banding, placement of a transjugular intrahepatic portosystemic shunt (TIPS) or by surgical portosystemic shunting.95,97,99–102,105,106,108 Balloon tamponade may also be used. Finally, although elective delivery by cesarean section has been recommended to avoid the strain of labor and risk of precipitating variceal hemorrhage, there is no evidence that vaginal deliveries precipitate hemorrhage and large intra-abdominal collateral veins may complicate surgical delivery.97,107 The effects of cirrhosis on the fetus are varied. First, before or during pregnancy, risks of maternal medications on the fetus should be considered. Spironolactone, which can cause genital malformations, should be stopped; propranolol, which might impair fetal growth, should be considered; and the need for other medications should be reassessed. Second, the rates of spontaneous abortion,
Table 54-4. Pregnancy and Cirrhosisa Features Variceal hemorrhage Maternal death Percentage of deaths from: Gastrointestinal hemorrhage Liver failure Other Complications Jaundice Ascites Hepatic encephalopathy Spontaneous abortions Premature births Prenatal deaths Postpartum hemorrhage a
Shunted (29 women)
Not shunted (90 women)
0 4%
18–32% 13%
0 100% 0
40–70% 15–25% 10–35%
NA NA NA 3% NA 17% 24%
28% 17% 4% 17% 23% 20% 9%
Data compiled from case reports and series published from 1968 to 1999 (representative references: 94,97,101,103). NA, not available.
premature birth, and perinatal death are all greater than expected in women with cirrhosis (Table 54-4). Infants born alive, however, are generally normal and do well. Third, fetal distress and perinatal mortality may be due, in part, to maternal hepatic decompensation and its attendant metabolic abnormalities. For example, severe maternal hyperbilirubinemia (16 and 33 mg/dl) has been reported109,110 to result in severely jaundiced infants due to maternal-to-infant placental transfer of unconjugated bilirubin, fetal distress in utero, and many postnatal complications, including kernicterus.109 It would seem prudent to monitor the fetuses of cirrhotic women and to consider early delivery when fetal distress and/or severe maternal hyperbilirubinemia is detected. Non-cirrhotic portal hypertension, associated with normal liver function, is present in 0.1% of pregnant women in India111 and does not appear to alter fertility to any significant degree. These women are reported to be at high risk (35%) of variceal bleeds during pregnancy. Although these episodes can usually be treated successfully, fetal loss during bleeding is reported to be as high as 40%.111
Viral Hepatitis B Chronic hepatitis B infection per se does not appear to alter fertility, conception, or pregnancy beyond the effects of liver dysfunction or cirrhosis itself.112 Vertical transmission of hepatitis B virus (HBV) from chronically infected mothers to offspring occurs during pregnancy or at delivery and prevention is an important clinical issue, as > 90% of infected infants become chronically infected.57,113–115 The risk of vertical transmission is related to the HBV viral load and replication rate, being over 85% in infants of women who exhibit HBeAg or detectable HBV DNA by qualitative assays.57,113,114 Therefore the American College of Obstetrics and Gynecology and the Centers for Disease Control recommend universal screening for HBsAg of all pregnant women during the third trimester.69,70 Infants of HBsAg-positive women should be treated immediately after birth (within 24 hours) with a single intramuscular injection of hepatitis B immunoglobulin and one injection of hepatitis B vaccine (Table 54-3).69,70,112,116,117 Further doses of vaccine should be given at 1 and 6 months of age. This immunoprophylaxis is highly effective in preventing > 80–90% of vertically transmitted HBV infection and provides future protection against horizontal transmission of both HBV and hepatitis D.112,115–118 However, for women with very high HBV DNA levels (>1.2 ¥ 109 gEq/ml), vaccine and HBIG prophylaxis fail in 28% of cases. Addition of maternal lamivudine treatment in the last month of pregnancy reduces this failure rate by half.119
Viral Hepatitis C Most women with chronic hepatitis C have mild disease and normal liver function and experience uncomplicated pregnancies. However, during pregnancy many women exhibit normalization of serum transaminase levels (AST, ALT), often associated with an increase in the hepatitis C virus (HCV) viral load, that reverses after delivery (Figure 54-2).120,121 These changes may be related to subtle immunologic shifts during pregnancy. The long-term consequences, if any, are not known, but a preliminary report122 suggests that after pregnancy HCV-infected women may experience worsening of necroinflammatory and fibrosis changes on liver biopsy. Further, an epidemiologic study of over 16 000 pregnant women identified a
1007
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
100
Serum HCV RNA Serum ALT
150 10
100 50
Normal
0 Prepregnant
1st
2nd
3rd
After delivery
Serum HCV RNA (105 viral particles/ml)
Serum ALT (IU/mI)
200
1
Figure 54-2. Reciprocal changes in serum alanine aminotransferase (ALT) and in hepatitis C virus (HCV) RNA levels before, during, and after pregnancy in women with chronic hepatitis C infection. Data represent the median values of 26 patients. (Modified from Gervais A, Bacq Y, Bernuau J, et al. Decrease in serum ALT and increase in serum HCV RNA during pregnancy in women with chronic hepatitis C. J Hepatol 2000; 32: 293–299, with permission from EASL.)
greater rate of cholestasis of pregnancy in HCV-positive women (15.9%) compared to HCV-negative women (0.8%), although the mechanism for this association is unclear.123 Hepatitis C can be transmitted to babies at or around the time of birth from infected or viremic mothers.124 Vertical transmission of hepatitis C does not appear to be related to the method of delivery or to breast-feeding.120,125,126–128 In 1800 pregnancies from 11 studies (most from Italy, where HCV infection is prevalent), 8.5% of babies were infected as determined by HCV RNA testing.120,125–127,129–136 The transmission rate appears to be up to two- to threefold higher, up to 30%, if mothers are also infected with human immunodeficiency virus (HIV),120,125,126,128,131 although antiretroviral therapy may decrease this risk.120,125,126,128 No prophylaxis is available (Table 54-3). Infants may exhibit HCV RNA early after delivery or not until 3–6 months of life.120,126,137 Up to 20% of infants will spontaneously eliminate the virus138 and maternal anti-HCV antibody may persist in uninfected infants for up to 18 months.120,126 Thus, efficient detection of true vertical transmission is best achieved by testing babies for HCV RNA and ALT at 3 and 6 months of age. Negative infants should be tested for HCV antibody once more at 18 months of age and positive values confirmed by HCV RNA testing.137,138 Outcome of children infected at birth is not well understood. One group of 62 asymptomatic children, 93% of whom exhibited abnormal ALT values, were followed for a mean of 5 years (range 2–11 years). By the end of follow-up, 81% were still viremic.138 High ALT levels and genotype 3 were more common in those infants who spontaneously cleared the virus. The only other data involve a 20year follow-up of children infected with HCV through blood transfusions for cardiac surgery. Only 1 patient developed cirrhosis in that time period (2.7%), and all other patients had normal liver tests.139 There are few data on the efficacy, side effects, or long-term benefits of treating children with hepatitis C,140 although treatment has been successful in some children.141
1008
As HCV-positive women cannot be reliably identified by history or examination, broad-based prenatal screening for HCV RNA has been proposed to counsel infected women and provide for long-term follow-up and eventual treatment of these women and any infected offspring. The safety and efficacy of treating pregnant HCV-positive women with interferon-based regimens are not known. A few case reports suggest treatment is safe;142 however, interferon is an abortifacient in some animals.
Primary Biliary Cirrhosis and Primary Sclerosing Cholangitis Cholestasis and pruritus in these disorders may be exacerbated or spontaneously improve during pregnancy and may respond to ursodeoxycholic acid (UDCA) therapy.143–146 Pregnancy outcome is more dependent on liver function and portal hypertension than the disease per se.
Liver Masses A wide variety of liver masses may be identified coincidentally during pregnancy, including liver cysts, intrahepatic pregnancy, hepatic hemangioma, liver cell adenoma, and hepatocellular carcinoma.147–150 Simple liver cysts are benign, are not affected by estrogens or pregnancy, and require no treatment. Intrahepatic pregnancy149 is extremely rare and may require surgical intervention. Other liver masses may be estrogen-responsive, may be affected by pregnancy, and are discussed later in this chapter.
Dubin–Johnson Syndrome Pregnancy or use of oral contraceptives in women with the Dubin–Johnson syndrome causes a reversible 2–2.5-fold increase in plasma concentrations of bilirubin.151,152 Plasma concentrations of bile acids remain normal.151 Affected women may be deeply jaundiced during pregnancy, but pruritus and other signs of generalized cholestasis are not seen. This transient exacerbation of the Dubin–Johnson syndrome is related, presumably, to the cholestatic effects of estrogens superimposed upon a liver with markedly impaired capacity for canalicular excretion of conjugated bilirubin.
PREGNANCY AFTER LIVER TRANSPLANTATION Although women with cirrhosis and severe liver dysfunction are usually amenorrheic and infertile,77,78 premenopausal women usually regain menstrual function and fertility after successful liver transplantation,77,78,153 most by 7 months after surgery. Pregnancy has been reported as early as 3 weeks post-transplantation,78 therefore contraceptive methods should be discussed soon after transplantation.77–79 Immunosuppressive drugs must be continued throughout pregnancy and blood levels monitored as changes in drug metabolism, especially of ciclosporin, may occur.78 Although azathioprine is teratogenic in animals and immunosuppressive drugs other than corticosteroids have not been adequately tested for safety in pregnancy,77,154 to date there is no evidence of increased fetal malformations in offspring of mothers with liver or kidney trans-
Chapter 54 THE LIVER IN PREGNANCY
plants.78,153–156 Many of these drugs are probably present in breast milk and the effects of breast-feeding are unknown. Pregnancy after liver transplantation is high-risk, with ~18% spontaneous abortions, 2% stillbirths, 36% premature births, 31% low-birth-weights, and 25% neonatal complications in 136 pregnancies in 130 women.154 Only 70% of pregnancies resulted in a live birth; however, these babies all did well. Mothers experience a variety of medical problems, including hypertension (40%), pre-eclampsia (25%), infections (30%, including cytomegalovirus infections that may adversely affect the fetus), and acute rejection (10%).154 Complications of pregnancy may be increased in women with pre-existing decreased renal function. Maternal mortality is low and related to recurrent liver disease and renal failure rather than to pregnancy per se. Liver transplantation has been performed during pregnancy, usually but not always resulting in fetal loss.
LIVER DISORDERS PROBABLY RELATED TO PREGNANCY BILIARY TRACT DISEASE Gallstone Formation Women develop cholesterol gallstones and clinical symptoms related to gallstones more frequently than do men.157,158 The increased incidence of gallstone formation begins at puberty, is related to the number of pregnancies, and tapers off after menopause, suggesting that sex hormones may be important etiologic factors.159 Compared with men, women exhibit increased saturation of bile with cholesterol and have a smaller pool of bile acids.157 Use of oral contraceptives14,157,160–162 or pregnancy163,164 increases the concentration of cholesterol and its total output in hepatic and gallbladder bile. The pool of bile acids is also increased but more of the bile acid pool is sequestered in the gallbladder and intestine due to decreased motility.164 As a result, there is little change or even a decrease in bile acids secreted into bile, and the enterohepatic cycling of bile acids is decreased in pregnancy, as is the proportion of chenodeoxycholic acid relative to cholic acid. All of these changes predispose to precipitation of cholesterol. Further, development of gallstones is promoted by pregnancy and oral contraceptive-induced decreases in gallbladder contractility.162–167 During pregnancy, biliary sludge develops in about one-third of women and by the time of delivery 10–12% of women exhibit gallstones on ultrasonographic examination.168–171 During pregnancy, biliary colic occurs in approximately one-third of those with existing stones, but not in those with sludge or a normal gallbladder.168,169 Biliary pain in most women responds to conservative medical management. Postpartum, biliary sludge disappears in virtually all women, but only about one-third of small stones disappear. Epidemiologic studies from pregnant women, women taking oral contraceptives, postmenopausal women receiving estrogens, men treated with diethylstilbestrol, and young women undergoing cholecystectomy all suggest that estrogen exposure may accelerate the development of symptoms in patients with pre-existing gallstones.172–175 Thus pregnancy may predispose not only to formation of gallstones but to presentation with clinical symptoms.
Biliary Tract Disease During Pregnancy Acute cholecystitis is second to appendicitis as the most common cause of non-obstetric surgery during pregnancy, accounting for 1–8 cases per 10 000 pregnancies.176 Furthermore, common duct stones are a common cause of jaundice during pregnancy.16 Diagnosis of biliary tract disease, with modern ultrasound, is straightforward,176–178 and, if necessary, magnetic resonance imaging may be safe.179 99Technetium-labeled hydroxyiminodiacetic acid (99TcHIDA) and other nuclear medicine scans are probably best avoided during pregnancy. Therapy for symptomatic disease during pregnancy is often conservative. About 55–85% of pregnant women with biliary colic, acute cholecystitis, or gallstone pancreatitis respond to general medical management and surgery may be postponed until after delivery.176,180,181 Patients with recurrent or worsening symptoms or common bile duct obstruction may require treatment during pregnancy. This can be accomplished with relatively little maternal or fetal mortality, even with open cholecystectomy.176,180,181 Laparoscopic cholecystectomy may be even safer during pregnancy as over 180 such operations have been performed with no maternal mortality, 1.7% spontaneous abortions, and 3.9% premature deliveries.182,183 Endoscopic management of gallstone pancreatitis or biliary obstruction during pregnancy also appears to be safe and effective in the few cases reported.184
HERPES SIMPLEX VIRUS HEPATITIS Although herpes simplex virus hepatitis (types I or II) is rare in previously healthy adults, about half the cases reported have occurred in association with pregnancy, and the mortality rate is about 40%.185–189 Patients generally present with a 4–14-day history of fever, systemic viral-type symptoms, and abdominal or right upper quadrant pain. Hepatitis is characterized by very high aminotransferase levels (> 1000 units), an increased prothrombin time, and a low bilirubin level, typically less than 3 mg/dl. Liver biopsy may be diagnostic, showing areas of focal or confluent hemorrhagic and coagulative necrosis, relatively little inflammatory infiltrate, and “ground-glass” nuclear inclusions or Cowdry type A inclusions at the periphery of areas of necrosis that are positive on immunohistochemical stain (Figure 54-3).187 Liver, vaginal, cervical, or throat cultures are often positive. Since therapy with agents such as aciclovir has been successful in salvaging both mothers and infants,185,186,189 aggressive evaluation of pregnant women who exhibit fever, a viral syndrome, and elevated transaminase values with a modestly elevated or low serum bilirubin should be initiated followed by immediate institution of antiviral therapy. Although vertical transmission of herpes simplex to infants, either in utero or at the time of delivery, is not inevitable, infants born to these women should be closely observed and treated as appropriate.
ESTROGEN-RESPONSIVE HEPATIC NEOPLASMS AND PREGNANCY The liver is an estrogen-responsive organ. Estrogens, either endogenous or exogenous, are thought to be involved in several hepatic vascular and neoplastic processes. These include hepatic sinusoidal dilatation, focal nodular hyperplasia, hepatocellular adenoma, and, possibly, some cases of hepatocellular carcinoma.147,190,191 Circum-
1009
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
A
B
Figure 54-3. Herpes simplex hepatitis during pregnancy. (A) Large area of confluent hepatocellular necrosis (double-headed arrow); (B) infected hepatocytes with intranuclear eosinophilic Cowdry type A inclusions (arrows). Hematoxylin and eosin. (Courtesy of Dr. H. Appelman.)
stantial epidemiological evidence links some of these disorders to use of oral contraceptives; the association between pregnancy and these abnormalities is based on case reports and by analogy to the effects of oral contraceptive use. With the widespread availability of high-quality ultrasonography and magnetic resonance imaging, hepatic mass lesions may be safely identified and monitored throughout pregnancy.179 Hepatic sinusoidal dilatation with associated hepatomegaly and abdominal pain has been reported in a few women receiving oral contraceptives. The lesion has been noted in livers that contained oral contraceptive-associated adenomas.190,192–194 Improvement follows discontinuing oral contraceptives. The prognosis is benign.192–194 Focal nodular hyperplasia (FNH) is a benign lesion, consisting of normal liver elements disposed around a central stellate scar that is often found incidentally. The lesion occurs almost exclusively in women.147,195 An association with oral contraceptive use is suggested by some,195–197 but not all, studies.198 FNH has not been reported to rupture during pregnancy or delivery, thus surgery could be considered only in those women with proven estrogen-responsive lesions who desire to bear children. Indeed, 22 women with FNH of
1010
4–13 cm tolerated 25 pregnancies without complications or an increase in tumor size.198,199 Hemangiomas also are very common and rarely grow during pregnancy or exhibit a clear estrogen-sensitivity.147 Rupture of even large hemangiomas is very rare, with only 2 cases occurring during pregnancy. Pregnancy and vaginal delivery are well tolerated, even in women with very large hemangiomas and prophylactic treatment does not seem to be indicated. Hepatocellular adenomas are benign hepatocellular tumors linked causally to estrogens and use of oral contraceptives.147,195,196,200,201 It is not known whether estrogens initiate the adenomas, but they appear to promote growth202 and the development of clinical symptoms, such as a mass lesion, abdominal pain, acute hemorrhage, necrosis, or rupture.147 Estrogens and/or pregnancy have also been suggested to promote progress to hepatocellular carcinoma. Conversely, many adenomas regress after removal of estrogens.203 Pregnancy has not been associated with an increased incidence of adenomas, but pregnancy is associated with growth of adenomas, development of symptoms (nausea, vomiting, right upper quadrant pain), and a risk that large adenomas (>6.5 cm) may rupture.147,199,200 Women with adenomas of 5 cm or greater in size should be monitored with ultrasound during pregnancy and resection considered for rapidly enlarging adenomas. In non-pregnant women, surgical excision should be considered for large, symptomatic tumors, for those that do not shrink after stopping oral contraceptives, and for tumors >5 cm in women who desire to bear children.147,204 Resection should be performed immediately for ruptured adenomas in both pregnant and non-pregnant women. Indeed, surgical resection of large adenomas has been successfully carried out during pregnancy. Hepatocellular carcinoma is extremely rare in pregnancy, but may be aggressive, perhaps related to the high levels of estrogen and subtle immunosuppression of pregnancy.147,150 Management options are few. Overall evaluation and management of this, like other liver masses detected during pregnancy, must be individualized with consideration of risks to both the mother and the fetus.
BUDD–CHIARI SYNDROME Budd–Chiari syndrome associated with use of oral contraceptives was noted as early as 1966, and the association has been well documented.205–208 Development of hepatic vein thrombosis is attributed to an oral contraceptive-induced increase in clotting factors plus a generalized propensity to venous thrombosis. Budd–Chiari syndrome associated with pregnancy is much less common. Thirtyone cases have been reported.209–212 The predisposing factors for hepatic vein occlusion are thought to be the estrogen-related increases in clotting factors and decreases in the activity of plasma antithrombin III likely associated with an underlying inherited thrombophilia.211,213 In some women, hepatic vein occlusion is associated with widespread venous thrombosis and may represent local propagation of clot originating in the iliac veins and inferior vena cava. Another syndrome that clinically resembles the Budd–Chiari syndrome, hepatic veno-occlusive disease, has been reported in 3 women postpartum214 and in 1 woman receiving an oral progestational agent for contraceptive purposes.215 Clinical symptoms of the Budd–Chiari syndrome frequently begin postpartum or immediately after an abortion rather than during the pregnancy itself. Management is the same as in the non-pregnant patient, although the fetus
Chapter 54 THE LIVER IN PREGNANCY
may be at high risk of intrauterine death if maternal liver function is poor.
Table 54-5. Intrahepatic Cholestasis of Pregnancy:a Clinical Features Incidence in pregnancy
LIVER DISORDERS UNIQUE TO PREGNANCY Four unique syndromes of liver dysfunction have been identified during pregnancy: (1) hepatic involvement in hyperemesis gravidarum; (2) intrahepatic cholestasis of pregnancy; (3) acute fatty liver of pregnancy (AFLP); and (4) pre-eclampsia/eclampsia-related liver disease.
HEPATIC INVOLVEMENT IN HYPEREMESIS GRAVIDARUM Hyperemesis gravidarum is not a liver disease, but liver dysfunction occurs in severe cases. For example, among women affected severely enough to require hospitalization for dehydration and weight loss, liver dysfunction and jaundice were noted in 13–33% and 10–13% of cases, respectively.16,216,217 Liver dysfunction usually presents in the first trimester, within 1–3 weeks after the onset of severe vomiting. Jaundice, dark urine, and occasionally pruritus are the major hepatic manifestations.16,42,216–218 Mild hyperbilirubinemia is the most frequently noted laboratory abnormality (mean value 1–7 mg/dl). Moderate increases in serum transaminase activities (2–3 times normal) occur in slightly more than half the patients and, rarely, values up to 800 IU/l have been noted.216 Alkaline phosphatase activities are elevated in a minority of patients. Autopsy specimens from 19 women who died of hyperemesis exhibited excess pigment in centrilobular areas and some fat, but no necrosis.219 Cholestasis has been seen in liver biopsies from some affected women,218 but most biopsies are normal. The etiology of hepatic dysfunction is unknown, but may be related to dehydration and malnutrition as similar liver findings are seen in patients with kwashiorkor and prolonged fasting. Hepatic dysfunction in hyperemesis gravidarum is a relatively benign process with little clinical consequence. Women who have died of hyperemesis in the past did so from starvation and dehydration, not from liver failure. If vomiting is controlled the hepatic dysfunction rapidly resolves, usually within a few days, although it may recur in subsequent pregnancies.16,42,216,218
INTRAHEPATIC CHOLESTASIS OF PREGNANCY Intrahepatic cholestasis of pregnancy (ICP) is a relatively benign cholestatic disorder that generally commences late in pregnancy, disappears abruptly after delivery, and frequently recurs with subsequent pregnancies. The main clinical manifestations are pruritus and jaundice. The term pruritus gravidarum is frequently applied to women with pruritus and biochemical cholestasis, whereas the terms cholestatic jaundice of pregnancy or cholestasis of pregnancy are often applied to those women who also develop clinically apparent jaundice.
Incidence ICP is identified in less than 2% of pregnancies in the USA and Europe (Table 54-5).220–241 The disorder appears to be more frequent
Onset Onset of pruritus Onset of jaundice Recurrence in subsequent pregnancies Signs and symptoms Pruritus Jaundice Nausea, vomiting Abdominal pain Skin excoriations
USA and Europe: 0.5–1.7% India: 0.8%–1.4% Scandinavia: 0.5–3% Chile: 4.7–6.5% 70% in the third trimester 30% before the third trimester Average: 28–30 weeks; range: 7–40 weeks 1–4 weeks after pruritus 21–70% 100% 10–25% 5–75% 9–24% Common
a
Data derived from over 1000 patients in case reports and series published from 1967 to 2003 (representative references: 219–240).
in Scandinavia and in Chile, being reported in 1–6% of all pregnancies there.220–231 It accounts for 20–50% of all causes of jaundice in pregnancy in series reported from Scandinavia;16,42,220 however, the overall incidence appears to exhibit seasonal fluctuations and to have decreased in the past two decades.225,232,233,240,241 In an Italian population, ICP was diagnosed more commonly (16%) in women who also had hepatitis C than in those who did not (0.8%), although the basis for this association is not known.123
Etiology The etiology and pathogenesis of ICP remain poorly defined but ICP appears to be the same disorder as oral contraceptive-induced cholestasis. Indeed, after the clinical introduction of oral contraceptives (containing high doses of estrogen), a number of women developed cholestasis which resembled ICP.242–244 Further, as many as 50% of women who experience ICP also develop pruritus and cholestasis when given oral contraceptives, and vice versa.245,246 It is likely that both genetic and hormonal factors are important in the development of ICP and/or contraceptive-induced cholestasis. The best hypothesis is that these disorders reflect a genetic sensitivity to the cholestatic effects of estrogens, although ICP has also been associated with altered plasma levels of selenium, zinc, and copper, as well as changes in biliary secretion of sulfated progesterones.247,248 Estrogens, especially ethinylestradiol and its 17b-glucuronide metabolite, reproducibly cause mild cholestasis in both humans and animals, as does pregnancy in animals, likely through inhibition of the hepatocyte basolateral bile-salt transporter NTCP as well as the canalicular BSEP (ABC B11).33–35,39–41,249,250 Estrogens and pregnancy also markedly impair the canalicular bilirubin transporter MRP2 (ABC C2),34,35,250 further promoting development of jaundice. Thus pregnancy and/or use of oral contraceptives can be considered to be states of mild and usually asymptomatic cholestasis. However clinical cholestasis (ICP) develops in the minority of these women due to underlying, presumably inheritable factors. Indeed, ICP frequently affects female relatives of index cases,222,225 including up to three generations of women in some families.229,251,252 ICP and famil-
1011
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
ial benign recurrent intrahepatic cholestasis have been observed in the same family.253 ICP and/or estrogen-related cholestasis are identified in 10–15% of both mothers and sisters of women who develop cholestasis while receiving oral contraceptives.254,255 Further, ethinylestradiol administration impairs biliary excretion of the organic anion BSP (a substrate for the canalicular bilirubin transporter MRP2) in both men and women, and the effect is much more marked both in women with a history of ICP and in women and men with a family history of ICP.256 Thus, it is likely that ICP results from a combination of high estrogen levels in women with mild mutations in one or more genes involved in bile acid transport and/or bile formation. Mutations in the canalicular phospholipid transporter MDR3 (ABC B4) have been described in several families in which heterozygously affected women have developed ICP.257–259 Homozygous children in these families have developed progressive familial intrahepatic cholestasis (PFIC) type 3. In these families, cholestasis is characteristically accompanied by high levels of GGTP. However, ICP must be genetically heterogeneous with involvement of other genes and/or other ABC B4 mutations as only 6/389 other women with ICP have been found to carry known ABC B4 mutations:260–266 sequencing of the genes for MDR3 (ABC B4) and BSEP (ABC B11) in 21 patients with ICP and in 40 control woman identified a further 10 new ICP-related mutations in MDR3 in 9 of the 21 patients.267 In this study,267 no ICP-related mutations in BSEP were identified, although analysis of single nucleotide polymorphisms in 57 women with ICP suggested associations with the gene for the canalicular bile salt transporter BSEP (ABC B11).268 At this time genetic testing is not generally available or clinically useful. However research laboratories may be able to test selected women with ICP who also have high GGTP levels and/or a family history of ICP, PFIC, or other cholestatic liver diseases.
Clinical Features The clinical and laboratory features of ICP233,240,269 are summarized in Tables 54-5 and 54-6,37,220–228,230–235,237–240,261,270–272 and Figure 54-
4. In addition, lipoprotein X may be identified in plasma, and gallbladder size and residual volume are often increased.273 Serum transaminase activities in a few patients with ICP are high enough to overlap with those typical of hepatocellular disorders such as acute viral hepatitis. Serologic tests for hepatitis viruses A and B, and the clinical course of the disease, particularly after delivery, may be helpful in the differential diagnosis. Liver biopsy is generally unnecessary for diagnosis. Liver failure and hepatic encephalopathy are not reported in ICP and their appearance indicates another etiology for the liver disease. The pruritus can be disabling, and in exceptional cases, can be so severe as to mandate termination of pregnancy. Fat malabsorption and vitamin K deficiency223 may develop in severe cases and may be responsible for some instances of maternal postpartum hemorrhage.269
Pathology The histopathology of ICP is that of intrahepatic cholestasis (Figure 54-5). Typical findings include centrilobular cholestasis, canaliculi containing bile plugs, and bile pigment in hepatocytes.16,274 Cholestasis may be patchy and subtle. Inflammation and hepatocellular necrosis are usually absent. Portal tracts and interlobular bile ducts are normal. Electron microscopic examination shows dilated bile canaliculi with loss of microvilli and occasional abnormal mitochondria.274,275 Histologic changes typically disappear after delivery and resolution of clinical symptoms.16,274
Natural History and Prognosis The cholestasis of ICP generally progresses until the time of delivery or termination of pregnancy.233,240,269 The severity of cholestasis and of laboratory abnormalities can be quite variable however, both during one pregnancy and between different pregnancies, with recurrence rates of 21–70%.224,233,237,240 For example, serum transaminase activities and even bilirubin concentrations may fluctuate and even temporarily normalize as pregnancy progresses.224,233,239,276 Pru-
Table 54-6. Intrahepatic Cholestasis of Pregnancy:a Laboratory Findings Test
Patients (n)
% of Women with abnormal values
Average value reported
Range of values reported
Normal valuesb
0.4–8.4
£ 1.1 mg/dl
nl–750
£ 60 KU/l
nl–1127 nl–1734 nl–430 nl–148 10–33 —
£ 40 IU/l £ 40 IU/l £ 6.5 mmol/l £ 5 mmol/l £ 12% —
2–31 1–10
£ 7 g/24 h £ 7 g/24 h
Bilirubin
991
12–27
Alkaline phosphatase
753
AST ALT Serum bile salts Serum cholic acid BSP retention at 45 minutes Prothrombin time Fecal fat: Patients with jaundice Patients with pruritus
837 749 576 91 77 177
73 (at least 2.4-fold upper limit of normal) 64 62 96–100 79 97–100 14c
1.0 (all patients) 2.9 (jaundiced patients) 146 ± 66 (up to 12.5-fold upper limit of normal) 115 160 35 24 23 —
— —
14.0 4.0
12 11
AST, aspartate aminotransferase; ALT, alanine aminotransferase; BSP, bromosulfophthalein; nl, normal. a Data derived from 1052 patients in case reports and series published from 1967 to 2002 (representative references: 37,219–227,229–234,236–239,260,268–270). b For non-pregnant individuals. c All corrected with vitamin K.
1012
Chapter 54 THE LIVER IN PREGNANCY
17
CA
ALAT 110
15 90 13 70 Enzyme activity (U/l)
11
Bile acid concentration (mmol/l)
9 7 5
50 30 10 90
ASAT
3 70 1 50
CDCA 7
30
5
10
3
16
20
30 32 34 36 38 40 2 4 35–60
Weeks of pregnancy
1 B
Days after delivery
DCA 3 1 16
20
30 32 34 36 38 40 2 4 35–60
Weeks of pregnancy A
Days after delivery
Figure 54-5. Intrahepatic cholestasis of pregnancy showing canalicular bile plugs (arrows) with well-preserved hepatocytes containing yellow pigment. Hematoxylin and eosin. (Courtesy of Dr. H. Appelman.)
Figure 54-4. Serum bile acid concentrations (A) and liver tests (B), during pregnancy in control women (shaded areas; mean ± 2 SD) and in 8 women who developed intrahepatic cholestasis of pregnancy (line; mean values). CA, cholic acid; CDCA, chenodeoxycholic acid; DCA, deoxycholic acid; ALAT, alanine aminotransferase; ASAT, aspartate aminotransferase. (Modified from Heikkinen J. Serum bile acids in the early diagnosis of intrahepatic cholestasis of pregnancy. Obstet Gynecol Scand 1975; 54:437.)
ritus, however, rarely improves before delivery. After delivery, pruritus quickly disappears, usually within 24–48 hours. Biochemical abnormalities and histologic findings resolve over the following weeks to months.31 Rarely, symptoms may persist for several weeks postpartum and respond to a short course of prednisone.277 Cholestasis may recur during treatment with oral contraceptives, although this appears to be less common with low-dose estrogen contraceptives.233 During long-term (up to 15-year) follow-up, prognosis for women who have had typical ICP is excellent, aside from a higher incidence (1.4–2.3-fold) of cholelithiasis and gallbladder disease.42,221,278 Chronic cholestatic liver disease, however, has developed in rare familial cases, probably due to specific and uncommon gene mutations.279 The prognosis for the fetus is not as benign. Problems include increases in premature labor (4–20%), intrauterine growth retardation (8–10%), and neonatal death (0.6–2.5%).37,220–222,225,226,230,231,234,237,238,241,268,272,280–284 Fetal monitoring has documented high rates of premature labor and delivery,237 fetal distress during labor (19–60%), and meconium staining
1013
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
(15–45%).228,231,237,238,241,280,283,284 The true incidence of fetal death due to ICP is not known; however, when several large series are combined, there is a trend towards increased fetal complications and death related to ICP (1.75% mortality in 679 infants born to women with ICP compared to 1.01% mortality in 710 infants born to unaffected women), especially in older series without aggressive obstetric management.31,37,221,222,237 The mechanism(s) of premature labor, fetal death, and meconium staining are not known but these events are attributed to elevated bile acids leading to increased uterine and fetal colonic muscle contractions, meconium passage, changes in fetal cardiac and vascular function, umbilical vein constriction, and acute fetal anoxia.229,238,269,285,286 These findings have prompted experienced physicians to recommend aggressive obstetric management.283,287,288 It is recommended that all women with ICP be closely followed from 34 weeks or the onset of jaundice, possibly including weekly fetal non-stress tests. Aggressive monitoring is thought to be especially important if onset of ICP is before 32 weeks, if there is a twin pregnancy, if jaundice occurs, or if there is a past history of a fetal death. Babies should be delivered promptly if any signs of fetal distress or meconium passage are found but even in the absence of fetal distress, labor should be induced, once fetal lung maturity has been documented, at 37 weeks for mild ICP and at 36 weeks for women who are jaundiced.222,223,226,231,237,240,241,269,283,287,288 Some have advocated even more aggressive care, including hospitalization and fetal non-stress tests at least once a day.231,238,284,289 Although the management plan outlined here appears to be associated with fewer sudden fetal deaths, fetuses still die in utero within hours or days of normal stress tests.237,238,283,289
Therapy In the past women were treated symptomatically for pruritus with cholestyramine,269,290 phenobarbital,290 and/or hypnotics. Success was variable and these agents did not improve fetal outcome. Fetal hemorrhage has been reported due to vitamin K deficiency from ICP and cholestyramine usage269 and thus vitamin K supplementation should be given near term to all women with jaundice and/or prolonged cholestasis. UDCA, a hydrophilic bile acid that improves other cholestatic liver diseases, possibly by stimulating biliary excretion of other, potentially toxic, bile acids or sulfated progesterones, is the most promising treatment for ICP.240,291 UDCA appears to be safe as no adverse outcomes have been reported in over 180 pregnancies.239,270–272,276,282,292–296a During both open-label and placebocontrolled treatment trials (especially when doses of > 1 gram (or 15 mg/kg) per day were used), most women receiving UDCA exhibit substantial improvement in pruritus, liver tests, and serum bile acid levels239,270,271,276,292–300 and better placental bile acid transport.271 Pregnancies were allowed to go further preceding delivery as well.276,292,294,296,296a,298 It is not clear whether UDCA actually improves fetal outcome when pregnancies are also aggressively managed, although anecdotal reports suggest such might be the case.282,293 Other less well studied therapeutic options include S-adenosyl-Lmethionine (SAME), which methylates phospholipids and improves experimental cholestasis301,302 and dexamethasone.303 SAME, which must be given parenterally and is not widely available, has been
1014
tested in four trials239,295,304,305 with either no295,304 or slight239,305 benefits for the mother. Overall, UDCA would appear to be the treatment of first choice.
ACUTE FATTY LIVER OF PREGNANCY AFLP is a rare and potentially fatal idiopathic disorder that appears during the third trimester of pregnancy.306,307 Its principal and distinctive histologic feature is infiltration of hepatocytes with microvesicular fat, histology strikingly similar to that in Reye’s syndrome, Jamaican vomiting sickness, valproic acid hepatotoxicity, tetracycline hepatotoxicity, and medium- and long-chain acyl coenzyme A (CoA) dehydrogenase deficiency.308–312 This group of disorders collectively have been termed the hepatic microvesicular steatoses, and they also share many clinical and laboratory features, suggesting a common abnormality in lipid metabolism and, perhaps, related etiology.
Incidence The frequency of AFLP appears to be 1 in 900 to 1 in 6660 pregnancies,235,306 including mild cases identified and treated early. AFLP accounts for 16–70% of cases of severe liver disease during pregnancy and, consequently, a significant number of maternal and fetal deaths.235,313,314 AFLP is more common in primagravidas, in twin pregnancies, and in those with male fetus (Table 54-7),235,306,311,313–324 although it can occur with any pregnancy.
Etiology AFLP, like other microvesicular steatoses, may be due to a combination of increased flux of triglycerides and fatty acids from adipose tissue to liver and defects in mitochondrial b-oxidation of fatty acids.307,310,317,325–327 During late pregnancy, increased plasma levels of triglycerides and fatty acids as well as mildly impaired boxidation have been documented in experimental animals and humans, possibly due to the effects of estrogens, progesterones, and fetal metabolic demands.307,312,325,328,329 In women who develop AFLP it is thought that additional factors further impair boxidation, leading to steatosis, decreased ATP generation, lipid peroxidation, elevated free fatty acids to toxic levels, inhibition of gluconeogenesis and, ultimately, liver failure.307,309,325,327 Potential triggering factors may include drugs, like aspirin or non-steroidal anti-inflammatory drugs that are known to impair b-oxida-
Table 54-7. Acute Fatty Liver of Pregnancy: Clinical Features and Presentationa Feature
Frequency
Incidence Age Primagravidas Twin gestations Male fetus Recurrences Onset Pre-eclampsia
1/875–1/6660 deliveries 28 years (16–40 years) 46% 10% 71% 3/22 35.5 weeks (26–40 weeks) 28%
a
Data derived from 202 patients in case reports and series published from 1955 to 2002 (representative references: 234,304,309,311–322).
Chapter 54 THE LIVER IN PREGNANCY
tion,309,312,330 inflammatory cytokines,325 or pre-eclampsia, which is also associated with increased fatty acid fluxes.328 Although most cases of AFLP appear to be sporadic, some cases, perhaps up to 20%,307,325,326,331,332 are likely due to a specific mutation in the gene for long-chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD), which catalyzes an essential step in mitochondrial boxidation of fatty acids.307,310,325,327,333,334 Indeed, heterozygosity for the common LCHAD mutations occurs in European-derived populations at rates of 1/150–1/680.323,325 Children homozygous for LCHAD mutations usually present early in life with rapidly fatal hypoketotic hypoglycemia, fatty liver, a Reye’s-like syndrome and skeletal or cardiac myopathies. The mothers of these children appear to be at high risk of AFLP: during 82 pregnancies in which 59 mothers heterozygous for LCHAD mutations carried homozygous affected fetuses, AFLP occurred in 43% and pre-eclampsia occurred in 22%.323,326,333–337 These same women had 74 pregnancies with normal or heterozygous fetuses and AFLP (2.7%) and preeclampsia (1.4%) were uncommon. Therefore, in women heterozygous for LCHAD deficiency, AFLP likely develops because of the combination of maternal genetic and pregnancy-related defects in liver mitochondrial b-oxidation of fatty acids exacerbated by placental defects in fatty acid metabolism,338,339 by unknown toxic factors generated by a homozygous affected fetus and, possibly, by concurrent pre-eclampsia.325,328 AFLP may be expected to recur in such women in future pregnancies if the fetus is homozygous for LCHAD deficiency. Prenatal testing for families with identified LCHAD mutations has been accomplished by research laboratories.340,341 Most infants born to women with AFLP, however, do not exhibit evidence of clinically significant defects in fatty acid metabolism. Thus, some combination of more subtle maternal and/or fetal genetic307,326,331,332,342 and environmental factors affecting maternal fatty acid oxidation may account for sporadic AFLP in the remaining 80% of cases. Based on these findings it is recommended that, in all cases of AFLP, both baby and mother be tested for at least the most common
(Glu474Gln) LCHAD mutation, that babies be screened for abnormal organic acids, acyl CoAs and acyl carnitines, and that babies be kept on a high-carbohydrate diet with frequent feedings until metabolic abnormalities have been excluded.307,325,331,332,334,340–342 Testing for the common LCHAD mutation is available on a fee-for-service basis through several laboratories that can be located through www.genetests.org. Determination of other mutations would likely require analysis by research laboratories.
Clinical Features The clinical and laboratory features of AFLP are summarized in Tables 54-7 to 54-10. Illness usually begins in the third trimester, around 35 weeks of gestation, although onset may be as early as 26 weeks or as late as the immediate postpartum period. The earliest manifestations are non-specific and include nausea, fatigue, malaise, vomiting, and abdominal distress (right upper quadrant or epigastric
Table 54-8. Acute Fatty Liver of Pregnancy: Signs and Symptomsa Feature
No. of patients for which data are available
Frequency (%)
198 202
77 54
195 202 199 88
56 86 44 32
132 158 77
32 84 13
Nausea, vomiting Abdominal pain; epigastric distress Encephalopathy Jaundice Hypoglycemia Ascites Extrahepatic manifestations Gastrointestinal hemorrhage Renal impairment Pancreatitis a
Data derived from 202 patients in case reports and series published from 1955 to 2002 (representative references: 304,309,311–321).
Table 54-9. Acute Fatty Liver of Pregnancy: Laboratory Abnormalitiesa Test
No. of patients
Frequency of abnormal results (%)
Abnormal values Average
Range
b
RBC smear White blood count Platelet count Prothrombin time Antithrombin III level Bilirubin Alkaline phosphatase
38 142 167 182 30 202 141
50 93 73 92 100 95 90
AST (nl £ 40) ALT (nl £ 40) Serum uric acid Serum creatinine
100 108 76 144
99 95 89 87
— — 22 s 11% of normal 12.2 mg% 4.4-fold upper limit of normal 218 IU/l 366 IU/l — 3.0 mg/dl
12 000–46 000 5000–121 000 nl–78 s 1.8–36 mg% nl to 10-fold upper limit nl–1300 nl–3670 up to 18.5 mg% up to 6.6 mg/dl
RBC, red blood cell; AST, aspartate aminotransferase; ALT, alanine aminotransferase; nl, normal. a Data derived from 202 patients in case reports and series published from 1955 to 2002 (representative references: 304,309,311–322). b Nucleated RBCs seen frequently; RBC fragments in those with disseminated intravascular coagulation.
1015
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Table 54-10. Acute Fatty Liver of Pregnancy: Outcome Maternal mortality All reportsa Reports from 1991 to 2002 in which delivery was initiated promptlyb
26% 5%
(59/224 cases) (5/101 cases)
Infant mortality All reportsa Reports from 1991 to 2002 in which delivery was initiated promptlyb
32% 9%
(80/249 cases) (10/108 cases)
a
Data derived from 224 patients in case reports and series published from 1955 to 2002 (representative references: 304,312,313,318–321,331). b Data derived from 101 patients in case reports and series published from 1991 to 2002 (representative references: 304,312,318–321,324,331).
pain). Fever, headache, diarrhea, back pain suggestive of pancreatitis, and myalgias are reported in some patients. Clinical signs of liver dysfunction and even frank liver failure, such as jaundice, hepatic encephalopathy, or bleeding, may ensue 1–2 weeks later. Mild cases may exhibit few, if any, signs or symptoms. Physical findings are often minimal, especially early in the disease. Right upper quadrant tenderness may be the only abnormality found. The liver is generally small and not palpable. As the disease progresses, jaundice, changes in mental status, edema, and ascites may appear. Signs and symptoms of pre-eclampsia and its hepatic complications are seen in 28% or more of patients.306,314,320,334,343 This is a much greater incidence than in normal pregnancies and may reflect roles for abnormalities in triglyceride and fatty acid metabolism in both AFLP and pre-eclampsia.325,328 Acute fatty liver of pregnancy is associated with a number of abnormal laboratory findings (Table 54-9). The blood smear frequently contains nucleated red blood cells and, in those patients with disseminated intravascular coagulopathy (DIC), fragmented erythrocytes, and Burr cells. Leukocytosis is frequent, even in women with no evidence of infection. Coagulation disorders are common, occasionally progressing to frank DIC. The abnormalities in clotting factors probably represent impaired hepatic synthesis as well as accelerated consumption. Indeed, a profound decrease in plasma antithrombin III activity has been found in all tested patients, even before the onset of AFLP.324 Liver test abnormalities include hyperbilirubinemia, increases in alkaline phosphatase activity, and modest elevations of serum transaminase activities (usually <400 IU/l). A few patients with otherwise typical AFLP, however, do exhibit markedly elevated serum transaminase values. The generally modest abnormalities of serum transaminases can be misleading as they do not reflect accurately the severe degree of liver dysfunction in these patients. For example, many patients with only modestly elevated AST/ALT may have evidence of fulminant hepatic failure, such as hypoglycemia, impaired synthesis of clotting factors, and hepatic encephalopathy. Modest renal dysfunction with elevation of blood urea nitrogen, creatinine, and uric acid values occurs in most patients with AFLP. Frank renal failure may occur in severe cases and require interim dialysis. The cause of renal failure is not known, although renal steatosis could be responsible.344 The characteristic increase in
1016
plasma uric acid often occurs prior to the onset of renal insufficiency, suggesting an early defect in renal tubular function. Imaging studies of the liver, using ultrasound or computed tomography (CT) or magnetic resonance imaging (MRI), may suggest the presence of a fatty liver; however, sensitivity and specificity of these findings are poor, and a normal study does not rule out the diagnosis of AFLP.306,320,345
Pathology The histologic appearance of AFLP is well described.311,315,318 At the light microscopic level, lobular architecture is intact, but the lobules are swollen. Infiltration of hepatocytes with microvesicular fat is the key feature of AFLP (Figure 54-6), although diffuse cytoplasmic ballooning or large fat vacuoles may be seen.311 Fat deposits are predominantly centrilobular (zone 3) with a more variable involvement of zones 1 and 2. The fat is readily appreciated when fat stains are used on frozen sections of tissue. The lesion is frequently missed when fixed tissue is examined as the fat is removed during the fixation process. In addition to fatty metamorphosis, centrilobular cholestasis, as indicated by bile canalicular plugs and bile-stained hepatocytes, is frequently seen. Although widespread hepatic necrosis is rarely seen, subtle signs of hepatocyte necrosis and hepatocytolysis are evident. Other clues suggesting that substantial loss of hepatocytes has in fact occurred include the small size of livers at autopsy, close apposition of portal tracts and central veins, prominent lipid- and lipofuscin-laden Kupffer cells, occasional acidophilic bodies, and numerous mitotic figures.311 Confluent centrizonal coagulative necrosis is seen in occasional patients who experience preterminal shock and is most likely a superimposed lesion of hepatic ischemia.311 Inflammation is commonly subtle or absent, although up to 25% of biopsies may exhibit substantial infiltration by lymphocytes and plasma cells in the lobule and/or portal tracts and, in these patients, the histological picture can resemble that of acute viral hepatitis, particularly if fat stains are not performed on frozen tissue. At the electron microscopic level, the swollen hepatocytes contain numerous small cytoplasmic lipid droplets.311,318,319 Mitochondria are large, pleomorphic, and ameboid-shaped and contain lamellar crystalline inclusions. Bile canaliculi are dilated with loss of microvilli. Precipitated material is seen in canalicular lumens as well as in intracellular pericanalicular vesicles. Other organs may also be involved and fatty metamorphosis of pancreatic acinar cells and renal tubular epithelia is welldescribed.344
Natural History and Prognosis All the complications of fulminant hepatic failure can occur in AFLP, including cerebral edema, gastrointestinal hemorrhage, renal failure, and infection. Once severe liver dysfunction has become evident, spontaneous labor or fetal death in utero commonly ensues, and delivery is often complicated by postpartum hemorrhage. After delivery, liver function rapidly improves, although some women experience transient worsening during the first few days postpartum. Signs, symptoms, liver function, and liver histology resolve over several weeks after delivery and there is no evidence of chronic liver disease.
Chapter 54 THE LIVER IN PREGNANCY
detection of milder cases that resolve quickly after delivery. Maternal morbidity, however, remains high, with average hospital stays of 11–16 days. Fetal mortality is related to rapid maternal decompensation, premature delivery, and maternal DIC, with fibrin deposition in the placenta leading to placental infarcts, placental insufficiency, and fetal asphyxia.346 AFLP seems to be a sporadic disease except in families with genetic mutations in fatty acid oxidation. In the past, at least 22 women without known genetic mutations who survived the disorder have become pregnant again, four of them twice. AFLP recurred in only three cases, and these may have had undiagnosed genetic mutations.313,315,316,320,322,347 A
B
Therapy There is no specific therapy for AFLP. Based on the observations that the disease only occurs in pregnancy, usually progresses relentlessly and rapidly during pregnancy, only improves after delivery, and has a potentially high fatality rate, prompt termination of pregnancy as soon as the diagnosis is suspected or made seems appropriate.306,320 Since AFLP usually presents late in pregnancy and affected infants often die in utero with little warning, immediate delivery would be expected to improve fetal survival as well. Mothers with liver failure should be transferred to centers experienced in the care of acute liver failure. Liver transplantation has been successfully employed in at least 2 patients with fulminant hepatic failure due to AFLP. The decision to proceed with transplantation should only be made after careful consideration, as AFLP usually resolves rapidly after delivery.
LIVER DISEASE RELATED TO PREGNANCYINDUCED HYPERTENSION Pre-eclampsia is a hypertensive disorder of pregnancy characterized by the trilogy of hypertension, proteinuria, and edema, while progression to seizures defines eclampsia. Although pre-eclampsia and eclampsia are not primarily liver diseases, any organ system can be involved in the disease process and a substantial fraction of affected women manifest some degree of liver damage, ranging from mild hepatocellular necrosis to hepatic rupture.
Incidence of Pre-eclampsia C
Figure 54-6. Acute fatty liver of pregnancy. (A) Fat accumulation is greater in pericentral hepatocytes (arrow) compared to periportal hepatocytes (arrowhead); hematoxylin and eosin. (B) The small (microvesicular) fat droplets surround but do not displace hepatocyte nuclei (arrows); hematoxylin and eosin. (C) Fat is readily appreciated with oil-red O stain on frozen liver tissue. (Courtesy of Dr. H Appelman.)
Overall outcome is summarized in Table 54-10. Both maternal and fetal mortality were high in the past, but have improved in recent series in which prompt delivery was undertaken along with improved medical treatment for fulminant liver failure and for premature infants. Improved outcome probably also reflects earlier
Pre-eclampsia develops in 5–11% of all pregnancies,348–350 more frequently in younger primagravidas, and at an extraordinarily high rate in women with the lupus anticoagulant. The risk is also two- to sixfold higher in women with other procoagulant disorders such as the factor V Leiden and factor II G20210A mutations.351,352 The disorder can be mild or can rapidly progress to multiorgan damage.
Pathophysiology The pathophysiology underlying pregnancy-induced hypertension, pre-eclampsia, and eclampsia is not completely understood; however, defective placental invasion of the maternal circulation and poor placental perfusion are thought to lead to maternal endothelial activation, increased vasoconstriction, increased peripheral resistance, hypertension, and activation of platelets and fibrinogen.349,350,353–356 Platelet aggregation and development of fibrin
1017
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
thrombi lead to thrombocytopenia, microangiopathic hemolytic anemia, and ischemic injury of a variety of organs, including liver. Abnormal lipid metabolism may also play a role in pre-eclampsia as plasma free fatty acids and triglycerides are increased in women with the disease.325,328,355 However documented genetic mutations in fatty acid metabolism have been identified in only a few women with pre-eclampsia-related liver disease.326,331,357
Liver Involvement Although most women with pre-eclampsia exhibit no clinical or laboratory evidence of liver disease,31 the liver is involved in at least 10–20% of women with pre-eclampsia.16,350,358–361 Indeed, preeclampsia-related liver disease occurs in up to 1.6% of all deliveries235,360–363 and accounts for about 5% of cases of jaundice in pregnancy. Liver involvement is more common – up to 70% – in severe pre-eclampsia/eclampsia,361,364 accounting for 20% of maternal mortality in pre-eclampsia.349,364,365 Two distinct syndromes of liver disease in pre-eclampsia have been described. One is a moderate to severe disorder of patchy hepatocyte necrosis associated with thrombocytopenia, termed the HELLP syndrome. The second is acute liver hemorrhage and rupture, a rare but often fatal event.
Hepatic Pathology Although liver biopsies in most women with pre-eclampsia are normal,3,4,366 17% or more of women with pre-eclampsia and 70% of women dying with eclampsia exhibit histologic liver injury.3,4,366–368 The characteristic hepatic lesions of pre-eclampsia consist of three features (Figure 54-7). First is diffuse deposition of fibrin along the sinusoids and, occasionally, in portal tract capillaries, hepatic arterioles, and portal veins.366,369 Second is ischemic necrosis of hepatocytes, usually focal but occasionally confluent.3,4,364,366,367,369 Third is periportal and portal tract hemorrhage.3,4,367,369 In cases with severe liver necrosis, diffuse patchy areas of hemorrhage and necrosis are visible grossly both within the parenchyma of the liver and in subcapsular locations. Large hematomas may form beneath the capsule (Figure 54-8) and may rupture, resulting in exsanguination.
HELLP Syndrome This syndrome was defined in 1982 and named for the characteristic laboratory features of hemolysis, elevated liver enzymes, and low platelet count370 and occurs in about 10% of all women with pre-eclampsia.359,371,372 Patients with the HELLP syndrome probably represent the middle of the spectrum of liver involvement in pre-eclampsia. Accepted criteria for diagnosis usually include abnormal red blood cell morphology (or increased lactate dehydrogenase), increased AST (or ALT) and a platelet count <100 000 or <150 000/mm3.370,372 Although liver damage is considered part of the HELLP syndrome, AST levels may reflect injury to organs other than the liver and a few patients who fit these criteria actually may have little or no histologic liver injury.3,4,366 Clinical features are summarized in Table 54-11.235,359–363,370,373–375 Approximately 30% of cases present postpartum.360,361,375 The relationship between HELLP and clinically evident pre-eclampsia is variable as hypertension is mild in 25% of patients with HELLP and is absent in 10% of cases, although the other clinical features and characteristic liver histologic changes are present. Some women exhibit considerable hepatic tenderness while large-volume transudative ascites occurs in 8–10% of cases.360,376
A
B
Figure 54-7. Liver in pre-eclampsia. (A) Periportal patchy hemorrhage and necrosis (arrow); (B) sinusoidal deposition of fibrin (arrow). Hematoxylin and eosin. (Courtesy of Dr. H. Appelman.)
1018
Figure 54-8. Abdominal computed tomography scan showing a subcapsular hematoma (arrow) in the right lobe in a woman with the HELLP syndrome (hemolysis, elevated liver enzymes, and low platelet count).
Chapter 54 THE LIVER IN PREGNANCY
Characteristic laboratory abnormalities are summarized in Table 54-12.359–361,370,373,374,376 Evidence of DIC is found in 8–21% of patients.360,375,377 The degree to which overall liver function is impaired may be difficult to assess. For example, hyperbilirubinemia, present in up to 42% of reported cases, is likely due to a combination of hemolysis and liver dysfunction, whereas elevated values for the prothrombin time probably reflect a combination of decreased synthesis and increased consumption. At least mild renal dysfunction is virtually universal. Maternal morbidity from HELLP is high, especially in severe cases identified by a platelet count of < 50 000/mm3.348,356,361,362 Approx-
Table 54-11. HELLP Syndrome: Clinical Featuresa Feature
Average value
Age Gestational duration Postpartum onset Primiparas Frequency: in all pregnant women in women with pre-eclampsia Hypertensionb Proteinuriab Nausea, vomiting Epigastric/right upper quadrant painb Headacheb Edemab
24 years (range 13–40 years) 32.3 weeks (range 23–40 years) 28% 53% 0.1–1.6% 10% 89% 80% 31% 52% 48% 54%
a
Data compiled from 1695 patients in case reports and series published from 1954 to 2000 (representative references: 234,357–361,368,371–373). Not all data are available for all patients. Signs and symptoms that are also commonly seen in women with pre-eclampsia who do not manifest the characteristic laboratory abnormalities of the HELLP syndrome (hemolysis, elevated liver enzymes, and low platelet count). b
Table 54-12. HELLP Syndrome: Laboratory Abnormalitiesa Test
Percentage with abnormal values
Average value
Range of values
Abnormal RBC morphology Lactic dehydrogenase Thrombocytopenia Fibrinogen Fibrin degradation products Prothrombin time Partial thromboplastin time Serum bilirubin AST (nl £ 40) ALT (nl £ 40) BUN, Cr
89.5b
—
—
98% 95% 8–16% 8–42%
— — Decreased Increased
nl–5000 12 000–100 000 — —
13% 13%
Increased Increased
— —
5–42% 89% 67% 50–100%
— 293 IU/l 173 IU/l Increased
nl–26 mg/dl nl–6200 IU/l nl–702 IU/l usually mild
HELLP, hemolysis, elevated liver enzymes, and low platelet count; RBC, red blood cells; AST, aspartate aminotransferase; ALT, alanine aminotransferase; BUN, blood urea nitrogen; nl, normal. a Data compiled from 1512 patients in case reports and series published from 1954 to 2000 (representative references: 357–359,368,371,372,374). Not all data are available on all patients. b RBC fragments, schistocytes, spherocytes; signs of microangiopathic hemolytic anemia.
imately 30–50% of such women develop cardiopulmonary, renal, hepatic, central nervous system, and/or bleeding complications. Maternal mortality ranges from 0 to 3.5% in experienced centers.235,353,355,360–363,375,378 Since maternal (and fetal) outcomes are worse when pre-eclampsia is complicated by HELLP,361 some have recommended laboratory screening of all women with preeclampsia for the laboratory features of HELLP,379 although it is not known if early identification alters treatment decisions or outcome. The HELLP syndrome rapidly resolves after delivery, although laboratory abnormalities may not peak until 24–48 hours postpartum.380 Long-term outcome is excellent as the liver lesions heal completely. Pre-eclampsia (and HELLP) may develop in subsequent pregnancies: rates of 3–27% have been reported.381,382 Pre-eclampsia and HELLP are far more dangerous to the fetus than to the mother, with high rates of intrauterine growth retardation and premature delivery as well as fetal/neonatal death rates of 3–23%.235,354,355,361–363,375,378 Fetal deaths are likely due to placental insufficiency and hypoxia.371 Therapy is directed primarily at urgent delivery as progressive damage to the liver and other organs as well as fetal death can take place quickly, and the disease resolves rapidly after delivery. Babies with adequate lung maturity are usually delivered within 24 hours of diagnosis, whereas less mature babies are often delivered after a few days of steroid treatment.354–356 High-dose steroid therapy appears to improve pre-eclampsia and HELLP in severely affected women before delivery or in women whose disease worsens postpartum.355,356,383,384 Plasmapheresis may also offer some benefits.
Overlap Between AFLP and Pre-eclampsia/HELLP The very rare AFLP and the more common pre-eclampsia/HELLP syndrome appear to coexist much more frequently than would be expected by chance alone: 28% or more of women with AFLP also manifest pre-eclampsia/HELLP (Table 54-7, Figure 549).306,314,320,334,343 Conversely, liver biopsies from 51 women with pre-eclampsia all showed microvesicular steatosis characteristic of AFLP.368,385 A common factor may be the abnormal lipid metabolism characteristic of both disorders.325,328,355 Indeed, women heterozygous for LCHAD deficiency have up to a 30% incidence of pre-eclampsia/HELLP whether they carry normal or affected fetuses.323,333,334 However, additional genetic/fetal/environmental factors may determine whether a pregnant woman develops one or both of these disorders or neither.
Hepatic Hemorrhage and Rupture Hepatic rupture in pregnancy was first described in 1844.386 More than 300 cases have been reported.364,386–395 Hepatic hemorrhage and/or rupture is probably due to confluent hepatic necrosis from pre-eclampsia and it is reported to occur in between 1/15 000 and 1/45 000 deliveries and in 1–2% of women with preeclampsia.360,361,364,386–388 The highest figures are derived from a study in which abdominal CT scans were used to identify contained hepatic hematomas in women with few symptoms other than right upper quadrant abdominal pain.387 Hemorrhage occurs late in pregnancy or up to 48 hours postpartum. Clinical or histologic features of pre-eclampsia are found
1019
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Microvesicular steatosis Modest necrosis Modest AST/ALT elevation Markedly impaired liver function disproportionate to AST/ALT elevation
AFLP
Pre-eclampsia: HELLP syndrome
Fibrin deposition, hemorrhage Ischemic injury with extensive necrosis Striking AST/ALT elevation Liver dysfunction proportionate to AST/ALT elevation Figure 54-9. AFLP and pre-eclampsia/HELLP syndrome (hemolysis, elevated liver enzymes, and low platelet count) overlap in some patients, but the pathophysiology and mechanism for liver dysfunction differ.
in 80–90%386 and many patients would satisfy the criteria for the HELLP syndrome. Hepatic hemorrhage and rupture are heralded by the sudden onset of right upper quadrant pain, which may, rarely, recur over several days. Hepatic tenderness, diffuse abdominal pain, peritoneal findings, chest pain, and/or shoulder pain are noted in some patients. Shock ensues in over half the patients within a few hours of the onset of pain or up to 48 hours later in women with contained hematomas. Anemia or a rapid drop in hematocrit is noted in many patients. These signs and symptoms are not diagnostic and are similar to those seen in a variety of intra-abdominal catastrophes, including rupture of a hepatic adenoma, abruptio placentae, perforated viscus, or intestinal infarction. Diagnosis is based on clinical suspicion and visualization of the liver, either radiologically or at laparotomy. Abdominal CT scans (Figure 54-8) may be the most sensitive and specific way to detect hepatic hemorrhage and/or rupture.386,387,389 Liver/spleen scans, ultrasound and angiograms may also be useful. Paracentesis can confirm intra-abdominal hemorrhage but does not clarify the site of bleeding and will be negative in patients with contained hematomas. MRI scans display chronic hematomas well, but are less accurate than CT scans in detecting acute hemorrhage. Hemorrhage into a contained hematoma may resolve with medical support.386,387,390 Women with contained hepatic hematomas may do well without surgery if they receive adequate hemodynamic support and are followed closely.386,387,390 Exploratory laparotomy should be undertaken, however, if there is any suspicion of free rupture into the peritoneum. By contrast, survival of an untreated liver rupture has not been reported. Treatment of hepatic rupture requires rapid diagno-
1020
sis and intensive therapy by a multidisciplinary team. The most important factors in ensuring maternal survival are identification of the disorder before irreversible shock has occurred, rigorous hemodynamic support with large-volume transfusions of blood and platelets, and control of bleeding by surgical or angiographic means.386,391 At surgery, hematomas are drained and hemostasis attempted with packing, sutures, and/or lobectomy.391–393 Outcome may be better with packing and drainage than with lobectomy.391–393 Surgical or angiographic hepatic artery interruption has been successful as well.386,388,390,392,393 Liver transplantation was successful in salvaging one woman with uncontrollable hemorrhage.394 Overall maternal and fetal mortality after hepatic rupture is high – approximately 50% and 60–70%, respectively. Mortality is lower in those with contained hematomas and in those in whom the diagnosis of liver rupture is made at or shortly after delivery.388 In those who survive, hematomas slowly resolve387,395 and subsequent successful pregnancy has been reported. Recurrence in a subsequent pregnancy has been reported once.386
DIFFERENTIAL DIAGNOSIS OF SEVERE LIVER DISEASE IN PREGNANCY The principal liver diseases that cause severe liver disease during pregnancy are acute viral hepatitis, AFLP, pre-eclampsia-related liver disease, and tetracycline-induced fatty liver. ICP, by contrast, rarely, if ever, causes severe liver disease. The clinical presentations of all four diseases can be similar, but laboratory tests are helpful in distinguishing them. Diagnosis of AFLP is based on compatible clinical and laboratory abnormalities in a women in the third trimester of pregnancy in conjunction with a liver biopsy (including fat stains on frozen tissue) showing microvesicular fat. In contrast to AFLP, acute viral hepatitis may present at any time during pregnancy. A history of exposure to an individual with hepatitis is often obtained, serologic tests for hepatitis viruses A or B may be positive, coagulopathy or DIC is uncommon, and serum transaminase values tend to be much higher than is seen in AFLP. Pre-eclampsia-related liver disease is associated with hypertension and proteinuria, whereas pre-eclampsia is often absent in AFLP. Pre-eclampsia-related liver disease exhibits a greater degree of hematologic abnormalities and of hepatocellular necrosis than does AFLP. Indeed, in severe cases, liver infarction and hemorrhage can be seen. However, some women exhibit both AFLP and pre-eclampsia-related liver disease (Figure 54-9). In the past, tetracycline, particularly when administered in high doses and/or to patients with impaired renal function, caused a syndrome that was virtually identical to AFLP. It is important to distinguish AFLP, pre-eclampsia, and possibly ICP from other causes of liver disease in pregnancy as treatment for these disorders, early delivery, and supportive care are critical to the outcome of both mother and child.
REFERENCES 1. Sheehan HL. Jaundice in pregnancy. Am J Obstet Gynecol 1961; 181:427. 2. Combes B, Shibata H, Adams R, et al. Alterations in sulfobromophthalein sodium-removal mechanisms from blood during normal pregnancy. J Clin Invest 1963; 42:1431.
Chapter 54 THE LIVER IN PREGNANCY
3. Ingerslev M, Teilum G. Biopsy studies on the liver in pregnancy I, II, III. Acta Obstet Gynecol 1945/46; 25:339, 352, 361. 4. Anita FP, Bharadwaj TP, Watsa MC, et al. Liver in normal pregnancy, pre-eclampsia, and eclampsia. Lancet 1958; 2:776. 5. Perez V, Gorodisch S, Casavilla F, et al. Ultrastructure of human liver at the end of normal pregnancy. Am J Obstet Gynecol 1971; 110:428. 6. Martinez-Manautou J, Aznar-Ramos R, Bautista-O’Farrill J, et al. The ultrastructure of liver cells in women under steroid therapy. II. Contraceptive therapy. Acta Endocrinol 1970; 65:7. 7. Rovinsky JJ, Jaffin H. Cardiovascular hemodynamics in pregnancy. Am J Obstet Gynecol 1966; 95:787. 8. Lund CJ, Donovan JC. Blood volume during pregnancy. Am J Obstet Gynecol 1967; 98:393. 9. Laakso L, Ruotsalainen P, Punnonen R, et al. Hepatic blood flow during late pregnancy. Acta Obstet Gynecol Scand 1971; 50:75. 10. Song CS, Rifkind AB, Gillette PN, et al. Hormones and the liver. The effects of estrogens, progestins, and pregnancy on hepatic function. Am J Obstet Gynecol 1969; 105:813. 11. McNair RD, Jaynes RV. Alterations in liver function during normal pregnancy. Am J Obstet Gynecol 1960; 80:500. 12. Ramcharan S, Sponzilli EE, Wingerd JC. Serum protein fractions. Effects of oral contraceptives and pregnancy. Obstet Gynecol 1976; 48:211. 13. Bacq Y, Zarka O, Brechot J-F, et al. Liver function tests in normal pregnancy: a prospective study of 103 pregnant women and 103 matched controls. Hepatology 1996; 23:1030. 14. Shaaban MM, Hammad WA, Fathalla MF, et al. Effects of oral contraception on liver function tests and serum proteins in women with past viral hepatitis. Contraception 1982; 26:65. 15. Burrows S, Pekala B. Serum copper and ceruloplasmin in pregnancy. Am J Obstet Gynecol 1971; 109:907. 16. Haemmerli P. Jaundice during pregnancy with special emphasis on recurrent jaundice during pregnancy and its differential diagnosis. Acta Med Scand 1966; 179 (Suppl 444):1. 17. Svanborg A, Vikrot O. Plasma lipid fractions, including individual phospholipids, at various stages of pregnancy. Acta Med Scand 1965; 178:615. 18. Knopp RH, Warth MR, Carrol CJ. Lipid metabolism in pregnancy. I. Changes in lipoprotein triglyceride and cholesterol in normal pregnancy and the effects of diabetes mellitus. J Reprod Med 1973; 10:95. 19. Novikoff PM. Intracellular organelles and lipoprotein metabolism in normal and fatty livers. In Arias IM, Popper H, Schachter D, et al, eds. The liver: biology and pathobiology. New York: Raven Press, 1982:143. 20. Schmid SE, Au WYW, Hill DE, et al. Cytochrome P-450dependent oxidation of the 17a-ethinyl group of synthetic steroids. D-homoannulation or enzyme inactivation. Drug Met Disp 1983; 11:531. 21. Crawford JS, Rudofsky S. Some alterations in the pattern of drug metabolism associated with pregnancy, oral contraceptives, and the newly-born. Br J Anaesth 1966; 38:446. 22. Vore M, Bauer J, Pascucci T. The effect of pregnancy on the metabolism of [14C] phenytoin in the isolated perfused rat liver. J Pharmacol Exp Ther 1978; 206:439. 23. Landon MJ, Kirkley M. Metabolism of diphenylhydantoin (phenytoin) during pregnancy. Br J Obstet Gynaecol 1979; 86:125. 24. Dam M, Christiansen J, Munck O, et al. Antiepileptic drugs: metabolism in pregnancy. Clin Pharmacokinet 1979; 4:53. 25. Hogstedt S, Lindberg B, Peng DR, et al. Pregnancy-induced increase in metoprolol metabolism. Clin Pharmacol Ther 1985; 37:688. 26. Casele HL, Laifer SA. Pregnancy after liver transplantation. Semin Perinatol 1998; 22:149.
27. Homeida M, Halliwell M, Branch RA. Effects of an oral contraceptive on hepatic size and antipyrine metabolism in premenopausal women. Clin Pharm Ther 1978; 24:228. 28. Brock WJ, Vore M. Hepatic morphine and estrone glucuronyltransferase activity and morphine biliary excretion in the isolated perfused rat liver. Drug Met Disp 1982; 10:336. 29. Watson KJR, Ghabrial H, Mashford ML, et al. The oral contraceptive pill increases morphine clearance but does not increase hepatic blood flow. Gastroenterology 1986; 90:1779A. 30. Schreiber AJ, Simon FR. Estrogen-induced cholestasis: clues to pathogenesis and treatment. Hepatology 1983; 3:607. 31. Ylostalo P. Liver function in hepatosis of pregnancy and preeclampsia with special reference to modified bromsulphthalein tests. Acta Obstet Gynecol Scand 1970; 49 (Suppl 4):4. 32. Kleiner GJ, Kresh L, Arias IM. Studies of hepatic excretory function. II. The effect of norethynodrel and mestranol on bromsulfalein sodium metabolism in women of childbearing age. N Engl J Med 1965; 273:420. 33. Cao J, Huang L, Liu Y, et al. Differential regulation of hepatic bile salt and organic anion transporters in pregnant and postpartum rats and the role of prolactin. Hepatology 2001; 33:140. 34. Cao J, Stieger B, Meier PJ, et al. Expression of rat hepatic multidrug resistance-associated proteins and organic anion transporters in pregnancy. Am J Physiol Gastrointest Liver Physiol 2002; 283:G757. 35. Geier A, Dietrich CG, Gerloff T, et al. Regulation of basolateral organic anion transporters in ethinylestradiolinduced cholestasis in the rat. Biochim Biophys Acta 2003; 1609:87. 36. Jarnfelt-Samsioe A, Eriksson B, Waldenstrom J, et al. Serum bile acids, gamma-glutamyltransferase and routine liver function tests in emetic and nonemetic pregnancies. Gynecol Obstet Invest 1986; 21:169. 37. Heikkinen J. Serum bile acids in the early diagnosis of intrahepatic cholestasis of pregnancy. Obstet Gynecol Scand 1975; 54:437. 38. Lunzer M, Barnes P, Byth K, et al. Serum bile acid concentrations during pregnancy and their relationship to obstetric cholestasis. Gastroenterology 1986; 91:825. 39. Lee J, Boyer JL. Molecular alterations in hepatocyte transport mechanisms in acquired cholestatic liver disorders. Semin Liver Dis 2000; 20:373. 40. Arrese M, Trauner M, Ananthanarayanan M, et al. Downregulation of the Na+/taurocholate cotransporting polypeptide during pregnancy in the rat. J Hepatol 2003; 38:148. 41. Trauner M, Boyer JL. Bile salt transporters: molecular characterization, function, and regulation. Physiol Rev 2003; 83:633. 42. Thorling L. Jaundice in pregnancy. A clinical study. Acta Med Scand 1955; 151 (Suppl 302):1. 43. Kessler WB, Andros GJ. Hepatic function during pregnancy and puerperium. Obstet Gynecol 1964; 23:372. 44. Sammour MB, Ramadan MA, Khalil FK, et al. Serum and placental lactic dehydrogenase and alkaline phosphatase isoenzymes in normal pregnancy and in pre-eclampsia. Acta Obstet Gynecol Scand 1975; 54:393. 45. Sussman HH, Bowman M, Lewis JL. Placental alkaline phosphatase in maternal serum during normal and abnormal pregnancy. Nature 1968; 218:359. 46. Walker FB, Hoblit DL, Cunningham FG, et al. Gamma glutamyl transpeptidase in normal pregnancy. Obstet Gynecol 1974; 43:745. 47. Seitanidis B, Moss DW. Serum alkaline phosphatase and 5¢nucleotidase levels during normal pregnancy. Clin Chim Acta 1969; 25:183. 48. Combes B, Shore GM, Cunningham FG, et al. Serum gglutamyl transpeptidase activity and in viral hepatitis:
1021
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
49. 50. 51. 52. 53.
54.
55. 56.
57.
58. 59.
60.
61. 62.
63.
64.
65. 66. 67. 68.
69.
70.
71. 72. 73.
1022
suppresion in pregnancy and by birth control pills. Gastroenterology 1977; 72:271. Bean WB. The cutaneous arterial spider: a survey. Medicine 1945; 24:243. Bean WB, Cogswell R, Dexter M. Vascular changes of the skin in pregnancy. Surg Obstet Gynecol 1949; 88:739. Friedlander P, Osler M. Icterus and pregnancy. Am J Obstet Gynecol 1967; 97:894. Hieber JP, Dalton D, Shorey J, et al. Hepatitis and pregnancy. J Pediatr 1977; 91:545. Rao AVN, Devi CS, Savithri P, et al. Infectious hepatitis in pregnancy and puerperium – a study of 60 cases. Indian J Med Sci 1969; 23:471. Borhanmanesh F, Haghighi P, Hekmat K, et al. Viral hepatitis during pregnancy. Severity and effect on gestation. Gastroenterology 1973; 64:304. Singh DS, Balasubramaniam M, Krishnaswami S, et al. Viral hepatitis during pregnancy. J Indian Med Assoc 1979; 73:90. Khuroo MS, Teli MR, Skidmore S. Incidence and severity of viral hepatitis in pregnancy. Am J Med 1981; 70:252. Tong MJ, Thursby M, Rakela J, et al. Studies on the maternal–infant transmission of the viruses which cause acute hepatitis. Gastroenterology 1981; 80:999. Jaiswal SP, Jain AK, Naik G, et al. Viral hepatitis during pregnancy. Int J Gynecol Obstet 2001; 72:103. Nayak NC, Panda SK, Datta R, et al. Aetiology and outcome of acute viral hepatitis in pregnancy. J Gastroenterol Hepatol 1989; 4:345. Khuroo MS, Samili S. Aetiology, clinical course and outcome of sporadic acute viral hepatitis in pregnancy. J Viral Hep 2003; 10:61. Harrison TJ. Hepatitis E virus – an update. Liver 1999; 19:71. Hamid SS, Jafri SMW, Khan H, et al. Fulminant hepatic failure in pregnant women: acute fatty liver or acute viral hepatitis? J Hepatol 1996; 25:20. Singh S, Mohanty A, Joshi YK, et al. Mother-to-child transmission of hepatitis E virus infection. Indian J Pediatr 2003; 70:37. Kumar RM, Uduman S, Rana S, et al. Sero-prevalence and mother-to-infant transmission of hepatitis E virus among pregnant women in the United Arab Emirates. Eur J Obstet Gynecol Reprod Biol 2001; 100:9. Michielsen, PP, Van Damme P. Viral hepatitis and pregnancy. Acta Gastroenterol Belg 1999; 62:21. Duff P. Hepatitis in pregnancy. Semin Perinatol 1998; 22:277. Dinsmoor MJ. Hepatitis in the obstetric patient. Infect Dis Clin North Am 1997; 11:78. Urganci N, Arapoglu M, Akyildiz B, et al. Neonatal cholestasis resulting from vertical transmission of hepatitis A infection. Pediatr Infec Dis J 2003; 22:381. Advisory Committee on Immunization Practices, CDC. Hepatitis B virus: a comprehensive strategy for eliminating transmission in the United States through universal childhood vaccination: recommendations of the Immunization Practices Advisory Committee (ACIP). Morbid Mortal Week Rep 40(RR-13):1, 1991 (www.cdc.gov/mmwr/preview/mmwrhtml/00033405.htm). ACOG Committee Opinion. Committee on Obstetrics. Maternal and fetal medicine: guidelines for hepatitis B virus screening and vaccination during pregnancy. Int J Gynecol Obstet 1993; 40:172. Khuroo MS, Saleem K, Shahid J. Vertical transmission of hepatitis E virus. Lancet 1995; 345:1025. Lefkowitch JH, Rushton AR, Feng-Chen K-C. Hepatic fibrosis in fetal alcohol syndrome. Gastroenterology 1983; 85:951. Moller J, Brandt NJ, Tygstrup I. Hepatic dysfunction in patient with fetal alcohol syndrome. Lancet 1979; i:605.
74. Habbick BF, Casey R, Zaleski WA, et al. Liver abnormalities in three patients with fetal alcohol syndrome. Lancet 1979; i:580. 75. Infeld DS, Borkowf HI, Varma RR. Chronic-persistent hepatitis and pregnancy. Gastroenterology 1979; 77:524. 76. Steven MM, Buckley JD, Mackay IR. Pregnancy in chronic active hepatitis. Q J Med 1979; 48:519. 77. Armenti VT, Herrine SK, Moritz MJ. Reproductive function after liver transplantation. Clin Liver Dis 1997; 1:471. 78. Laifer SA, Guido RS. Reproductive function and outcome of pregnancy after liver transplantation in women. Mayo Clin Proc 1995; 70:388. 79. Connolly TJ, Zuckerman AL. Contraception in the patient with liver disease. Semin Perinatol 1998; 22:178. 80. Whelton MJ, Sherlock S. Pregnancy in patients with hepatic cirrhosis. Lancet 1968; ii:995. 81. Heneghan MA, Norris SM, O’Grady JG, et al. Management and outcome of pregnancy in autoimmune hepatitis. Gut 2001; 48:97. 82. Schramm C, Kanzler S, Weidenbach H, et al. Autoimmune hepatitis and pregnancy: course of disease and outcome of pregnancies. Hepatology 2003; 38:486A. 83. Colle I, Hautekeete M. Remission of autoimmune hepatitis during pregnancy: a report of two cases. Liver 1999; 19:55. 84. Levine AB. Autoimmune hepatitis in pregnancy. Obstet Gynecol 2000; 95:1033. 85. Buchel E, Van Steenbergen W, Nevens F, et al. Improvement of autoimmune hepatitis during pregnancy followed by flare-up after delivery. Am J Gastroenterol 2002; 97:3160. 86. Samuel D, Riordan S, Strasser S, et al. Severe autoimmune hepatitis first presenting in the early post partum period. Clin Gastro Hepatol 2004; 2:622. 87. Walshe JM. The management of pregnancy in Wilson’s disease treated with trientine. Q J Med 1986; 58:81. 88. Mustafa MS, Shamina AH. Five successful deliveries following 9 consecutive spontaneous abortions in a patient with Wilson disease. Aust NZ J Obstet Gynaecol 1998; 38:312. 89. Sternlieb I. Wilson’s disease and pregnancy. Hepatology 2000; 31:531. 90. German JL III, Bearn AG. Effect of estrogens on copper metabolism in Wilson’s disease. J Clin Invest 1961; 40:445. 91. Brewer GJ, Johnson VD, Dick RD, et al. Treatment of Wilson’s disease with zinc. XVII: treatment during pregnancy. Hepatology 2000; 31:364. 92. Nunns D, Hawthorne B, Goulding P, et al. Wilson’s disease in pregnancy. Eur J Obstet Gynecol Reprod Biol 1995; 62:141. 93. Walshe JM. Pregnancy in Wilson’s disease. Q J Med 1977; 46:73. 94. Borhanmanesh F, Haghighi P. Pregnancy in patients with cirrhosis of the liver. Obstet Gynecol 1970; 36:315. 95. Schreyer P, Caspi E, El-Hindi JM, et al. Cirrhosis – pregnancy and delivery: a review. Obstet Gynecol Surv 1982; 37:304. 96. Cheng Y-S. Pregnancy in liver cirrhosis and/or portal hypertension. Am J Obstet Gynecol 1977; 128:812. 97. Britton RC. Pregnancy and esphageal varices. Am J Surg 1982; 143:421. 98. Niven P, Williams D, Zeegen R. Pregnancy following the surgical treatment of portal hypertension. Am J Obstet Gynecol 1971; 110:1100. 99. Salam AA, Warren WD. Distal splenorenal shut for the surgical treatment of variceal bleeding during pregnancy. Arch Surg 1972; 105:643. 100. Chattopadhyay TK, Kapoor VK, Iyer KS, et al. Successful splenorenal shunt performed during pregnancy. Jpn J Surg 1984; 14:405. 101. Brown HJ. Splenorenal shunt during pregnancy. Am Surg 1971; 37:441.
Chapter 54 THE LIVER IN PREGNANCY
102. Krol-Van Straaten J, De Maat CEM. Successful pregnancies in cirrhosis of the liver before and after portacaval anastomosis. Neth J Med 1984; 27:14. 103. Cerqui AJ, Haran M, Brodribb R. Implications of liver cirrhosis in pregnancy. Aust NZ J Obstet Gynaecol 1998; 38:93. 104. Aggarwal N, Sawnhey H, Suril V, et al. Pregnancy and cirrhosis of the liver. Aust NZ J Obstet Gynaecol 1999; 39:503. 105. Starkel P, Horsmans Y, Geubel A. Endoscopic band ligation: a safe technique to control bleeding esophageal varices in pregnancy. Gastrointest Endosc 1998; 48:212. 106. Zeeman GG, Moise KJ. Prophylactic banding of severe esophageal varices associated with liver cirrhosis in pregnancy. Obstet Gynecol 1999; 94:842. 107. Russell MA, Craigo SD. Cirrhosis and portal hypertension in pregnancy. Semin Perinatol 1998; 22:156. 108. Pajor A, Lehoczky D. Pregnancy and extrahepatic portal hypertension. Gynecol Obstet Invest 1990; 30:193. 109. Waffarn F, Carlisle S, Pena I, et al. Fetal exposure to maternal hyperbilirubinemia. Am J Dis Child 1982; 136:416. 110. Cotton DB, Brock BJ, Schifin BS. Cirrhosis and fetal hyperbilirubinemia. Obstet Gynecol 1981; 57 (Suppl):25S. 111. Aggarwal N, Sawhney H, Vasishta K, et al. Non-cirrhotic portal hypertension in pregnancy. Int J Gynecol Obstet 2001; 72:1. 112. Wong S-J, Chan LY-S, Yu V, et al. Hepatitis B carrier and perinatal outcome in singleton pregnancy. Am J Perinatol 1999; 16:485. 113. Zanetti AR, Ferroni P, Magliano EM, et al. Perinatal transmission of the hepatitis B virus and of the HBV-associated delta agent from mothers to offspring in Northern Italy. J Med Virol 1982; 9:139. 114. Stevens CE, Neurath RA, Beasley RP, et al. HBeAg and antiHBe detection by radioimmunoassay: correlation with vertical transmission of hepatitis B virus in Taiwan. J Med Virol 1979; 3:237. 115. Beasley RP, Hwang L-Y, Lee G C-Y, et al. Prevention of perinatally transmitted hepatitis B virus infections with hepatitis B immune globulin and hepatitis B vaccine. Lancet 1983; 2:1099. 116. Huang K-Y, Lin S-R. Nationwide vaccination: a success story in Taiwan. Vaccine 2000; 18:S35. 117. Euler GL, Copeland JR, Rangel MC, et al. Antibody response to postexposure prophylaxis in infants born to hepatitis B surface antigen-positive women. Pediatr Infect Dis 2003; J 22:123. 118. Roome A, Rak M, Hadler J. Follow-up of infants of hepatitis Binfected women after hepatitis B vaccination, Connecticut, 1994 to 1997. Pediatr Infect Dis J 2000; 19:573. 119. van Zonneveld M, van Nunen AB, Niesters HGM, et al. Lamivudine treatment during pregnancy to prevent perinatal transmission of hepatitis B virus infectiion. J Viral Hep 2003; 10:294. 120. Conte D, Fraquelli M, Prati D, et al. Prevalence and clinical course of chronic hepatitis C virus (HCV) infection and rate of HCV vertical transmission in a cohort of 15 250 pregnant women. Hepatology 2000; 31:751. 121. Gervais A, Bacq Y, Bernuau J, et al. Decrease in serum ALT and increase in serum HCV RNA during pregnancy in women with chronic hepatitis C. J Hepatol 2000; 32:293. 122. Fontaine H, Nalpas B, Gillet A. Worsening of HCV-related hepatitis after pregnancy. Hepatology 2000; 32:278A. 123. Locatelli A, Roncaglia N, Arreghini A, et al. Hepatitis C virus infection is associated with a higher incidence of cholestasis of pregnancy. Br J Obstet Gynaecol 1999; 106:498. 124. Newell M-L, Pembrey L. Mother-to-child transmission of hepatitis C virus infection. Drugs Today 2002; 38:321. 125. Eriksen NL. Perinatal consequences of hepatitis C. Clin Obstet Gynecol 1999; 42:121.
126. Zanetti AR, Tanzi E, Newell ML. Mother-to-infant transmission of hepatitis C virus. J Hepatol 1999; 31 (Suppl. 1):96. 127. Resti M, Azzari C, Galli L, et al. Maternal drug use is a preeminent risk factor for mother-to-child hepatitis C virus transmission: results from a multicenter study of 1372 mother–infant pairs. J Infect Dis 2002; 185:567. 128. Dal Molin G, D’Agaro P, Ansaldi F, et al. Mother-to-infant transmission of hepatitis C virus: rate of infection and assessment of viral load and IgM anti-HCV as risk factors. J Med Virol 2002; 67:137. 129. Caudai C, Battiata M, Riccardi MP, et al. Vertical transmission of the hepatitis C virus to infants of anti-human immunodeficiency virus-negative mothers: molecular evolution of hypervariable region 1 in prenatal and perinatal or postnatal infections. J Clin Microbiol 2003; 41:3955. 130. Steininger C, Kundi M, Jatzko G, et al. Increased risk of mother-to-infant transmission of hepatitis C virus by intrapartum infantile exposure to maternal blood. J Infect Dis 2003; 187:345. 131. Ferrero S, Lungaro P, Bruzzone BM, et al. Prospective study of mother-to-infant transmission of hepatitis C virus: a 10-year survey (1990–2000). Acta Obstet Gynecol Scand 2003; 82:229. 132. Reinus JF, Leikin EL, Alter HJ, et al. Failure to detect vertical transmission of hepatitis C virus. Ann Intern Med 1992; 117:881. 133. Matsubara T, Sumazaki R, Takita H. Mother-to-infant transmission of hepatitis C virus: a prospective study. Eur J Pediatr 1995; 154:973. 134. Pipan C, Amici S, Astori G, et al. Vertical transmission of hepatitis C virus in low-risk pregnant women. Eur J Clin Microbiol Infect Dis 1996; 15:116. 135. Resti M, Azzari, C, Mannelli F, et al. Mother to child transmission of hepatitis C virus: prospective study of risk factors and timing of infection in children born to women seronegative for HIV-1. Br Med J 1998; 317:437. 136. Pappalardo BL. Influence of maternal human immunodeficiency virus (HIV) co-infection on vertical transmission of hepatitis C virus (HCV): a meta-analysis. Int J Epidemiol 2003; 32:727. 137. Resti M, Bortolotti F, Vajro P, et al. Guidelines for the screening and follow-up of infants born to anti-HCV positive mothers. Digest Liver 2003; Dis 35:453. 138. Resti M, Jara P, Hierro L, et al. Clinical features and progression of perinatally acquired hepatitis C virus infection. J Med Virol 2003; 70:373. 139. Vogt M, Lang T, Klingler C, et al. Prevalence and clinical outcome of hepatitis C infection in children who underwent cardiac surgery before the implementation of blood-donor screening. N Engl J Med 1999; 341:866. 140. Jonas MM. Children with hepatitis C. Hepatology 2002; 36:S173. 141. Wirth S, Lang T, Gehring S, et al. Recombinant alfa-interferon plus ribavirin therapy in children and adolescents with chronic hepatitis C. Hepatology 2002; 36:1280. 142. Ozasian E, Yilmaz R, Simsek H, et al. Interferon therapy for acute hepatitis C during pregnancy. Ann Pharmacother 2002; 36:1715. 143. Olsson R, Loof L, Wallerstedt S. Pregnancy in patients with primary biliary cirrhosis – a case for dissuasion? Liver 1993; 13:316. 144. Nir A, Sorokin Y, Abramovici H, et al. Pregnancy and primary biliary cirrhosis. Int J Gynaecol Obstet 1989; 28:279. 145. Chazouilleres O, Poupon R, Bonnand AM, et al. Pregnancy and ursodeoxycholic acid (UDCA) treatment induce remission of primary biliary cirrhosis (PBC). Hepatology 1998; 28:545A. 146. Janczewska I, Olsson R, Hultcrantz R, et al. Pregnancy in patients with primary sclerosing cholangitis. Liver 1996; 16:326.
1023
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
147. Cobey FC, Salem RR. A review of liver masses in pregnancy and a proposed algorithm for their diagnosis and management. Am J Surg 2004; 187:181. 148. Athanassiou AM, Craigo SD. Liver masses in pregnancy. Semin Perinatol 1998; 22:166. 149. Delabrousse E, Site O, Le Mouel A, et al. Intrahepatic pregnancy: sonography and CT findings. Am J Roentgenol 1999; 173:1377. 150. Lau WY, Leung WT, Ho S, et al. Hepatocellular carcinoma during pregnancy and its comparison with other pregnancyassociated malignancies. Cancer 1995; 75:2669. 151. Cohen L, Lewis C, Arias IM. Pregnancy, oral contraceptives, and chronic familial jaundice with predominantly conjugated hyperbilirubinemia (Dubin–Johnson syndrome). Gastroenterology 1972; 62:1182. 152. Di Zoglio JD, Cardillo E. The Dubin–Johnson syndrome and pregnancy. Obstet Gynecol 1973; 42:560. 153. Cundy TF, O’Grady JG, Williams R. Recovery of menstruation and pregnancy after liver transplantation. Gut 1990; 31:337. 154. Armenti VT, Herrine SK, Radomski JS, et al. Pregnancy after liver transplantation. Liver Transpl 2000; 6:671. 155. Wu A, Nashan B, Messner U, et al. Outcome of 22 successful pregnancies after liver transplantation. Clin Transplant 1998; 12:454. 156. Jain A, Venkataramanan R, Fung JJ, et al. Pregnancy after liver transplantation under tacrolimus. Transplantation 1997; 64:559. 157. Bennion LJ, Grundy SM. Risk factors for the development of cholelithiasis in man. N Engl J Med 1978; 299:1221. 158. Thistle JL. Gallstones in women. Med Clin North Am 1974; 58:811. 159. Glasnovic JC, Mege RM, Mannovic I, et al. Cholelithiasis in young women. Gastroenterology 1993; 104:A76. 160. Pertsemlidis D, Panveliwalla D, Ahrens EH. Effects of clofibrate and of an estrogen–progestin combination on fasting biliary lipids and cholic acid kinetics in man. Gastroenterology 1974; 66:565. 161. Bennion JL, Mott DM, Howard BV. Oral contraceptives raise the cholesterol saturation of bile by increasing biliary cholesterol secretion. Metabolism 1980; 29:18. 162. Kern F, Everson GT, DeMark B. Biliary lipids, bile acids, and gallbladder function in the human female: effects of contraceptive steroids. J Lab Clin Med 1982; 99:798. 163. Kern F, Everson GT, DeMark B, et al. Biliary lipids, bile acids, and gallbladder function in the human female. Effects of pregnancy and the ovulatory cycle. J Clin Invest 1981; 68:1229. 164. Everson GT. Pregnancy and gallstones. Hepatology 1993; 17:159. 165. Braverman DZ, Johnson ML, Kern F Jr. Effects of pregnancy and contraceptive steroids on gallbladder function. N Engl J Med 1980; 302:362. 166. Everson RB, Byar DP, Bischoff AJ. Estrogen predisposes to cholecystectomy but not to stones. Gastroenterology 1982; 82:4. 167. Braverman DZ, Herbet D, Goldstein R, et al. Postpartum restoration of pregnancy-induced cholecystoparesis and prolonged intestinal transit time. J Clin Gastroenterol 1988; 10:642. 168. Maringhini A, Ciambra M, Baccelliere P, et al. Biliary sludge and gallstones in pregnancy. Ann Intern Med 1993; 119:116. 169. Valdivieso V, Covarrubias C, Siegel F, et al. Pregnancy and cholelithiasis. Hepatology 1993; 17:1. 170. Van Bodegraven AA, Bohmer CJM, Manoliu RA, et al. Gallbladder contents and fasting gallbladder volumes during and after pregnancy. Scand J Gastroenterol 1998; 33:993. 171. Ko CW, Sekijima JH, Lee SP. Biliary sludge. Ann Intern Med 1999; 130:301.
1024
172. Wingrave SJ, Kay CR. Oral contraceptives and gallbladder disease. Royal College of General Practitioners oral contraception study. Lancet 1982; ii:957. 173. Boston Collaborative Drug Surveillance Program. Surgically confirmed gallbladder disease, venous thromboembolism, and breast tumors in relation to postmenopausal estrogen therapy. N Engl J Med 1974; 290:15. 174. Everson GT, McKinley C, Lawson M. Gallbladder function in the human female: effect of the ovulatory cycle, pregnancy, and contraceptive steroids. Gastroenterology 1982; 82:711. 175. Glenn F, McSherry CK. Gallstones and pregnancy among 300 young women treated by cholecystectomy. Surg Gynecol Obstet 1968; 127:1067. 176. Landers D, Carmona R, Crombleholme W, et al. Acute cholecystitis in pregnancy. Obstet Gynecol 1987; 69:131. 177. Bartoli E, Calonaci N, Nenci R. Ultrasonography of the gallbladder in pregnancy. Gastrointest Radiol 1984; 9:35. 178. Mintz MC, Grumbach K, Arger PH, et al. Sonographic evaluation of bile duct size during pregnancy. Am J Radiol 1985; 145:575. 179. Zaretsky MV, Twickler DM. Magnetic resonance imaging in obstetrics. Clin Obstet Gynecol 2003; 46:868. 180. Ghumman E, Barry M, Grace PA. Management of gallstones in pregnancy. Br J Surg 1997; 84:1646. 181. Glasgow RE, Visser BC, Harris HW, et al. Changing management of gallstone disease during pregnancy. Surg Endosc 1998; 12:241. 182. Sungler P, Heinerman PM, Steiner H, et al. Laparoscopic cholecystectomy and interventional endoscopy for gallstone complications during pregnancy. Surg Endosc 2000; 14:267. 183. Cosenza CA, Saffari B, Jabbour N, et al. Surgical management of biliary gallstone disease during pregnancy. Am J Surg 1999; 178:545. 184. Tham TCK, Vandervoort J, Wong RCK, et al. Safety of ERCP during pregnancy. Am J Gastroenterol 2003; 98:308. 185. Kang AH, Graves CR. Herpes simplex hepatitis in pregnancy: a case report and review of the literature. Obstet Gynecol Surv 1999; 54:463. 186. Glorioso DV, Molloy PJ, Van Thiel DH, et al. Successful empiric treatment of HSV hepatitis in pregnancy. Dig Dis Sci 1996; 41:1273. 187. Jacques SM, Qureshi F. Herpes simplex virus hepatitis in pregnancy. Hum Pathol 1992; 23:183. 188. Goyert GL, Bottoms SF, Sokol RJ. Anicteric presentation of fatal herpetic hepatitis in pregnancy. Obstet Gynecol 1985; 65:585. 189. Klein NA, Mabie WC, Shaver DC, et al. Herpes simplex virus hepatitis in pregnancy. Gastroenterology 1991; 100:239. 190. Ishak KG. Hepatic lesions caused by anabolic and contraceptive steroids. Semin Liver Dis 1981; 1:116. 191. Christopherson WM, Mays ET, Barrows GHA. Liver oncogenesis and steroids. In Ariel IM, ed. Progress in clinical cancer, vol. III. New York: Grune and Stratton, 1978:153. 192. Winkler K, Poulsen H. Liver disease with perioportal sinusoidal dilatation. A possible complication to contraceptive steroids. Scand J Gastroenterol 1975; 10:699. 193. Spellberg MA, Mirro J, Chowdhury L. Hepatic sinusoidal dilatation related to oral contraceptives. Am J Gastroenterol 1979; 72:218. 194. Thung SN, Gerber MA. Precursor stage of hepatocellular neoplasm following long exposure to orally administered contraceptives. Hum Pathol 1981; 12:472. 195. Kerlin P, Davis GL, McGill DB, et al. Hepatic adenoma and focal nodular hyperplasia: clinical, pathologic, and radiologic features. Gastroenterology 1983; 84:994. 196. Klatskin G. Hepatic tumors. Possible relationship to use of oral contraceptives. Gastroenterology 1977; 73:386.
Chapter 54 THE LIVER IN PREGNANCY
197. Scott LD, Katz AR, Duke JH, et al. Oral contraceptives, pregnancy, and focal nodular hyperplasia of the liver. JAMA 1984; 251:1461. 198. Mathieu D, Kobeiter H, Maison P, et al. Oral contraceptive use and focal nodular hyperplasia of the liver. Gastroenterology 2000; 118:560. 199. Weimann A, Mossinger M, Fronhoff K, et al. Pregnancy in women with observed focal nodular hyperplasia of the liver. Lancet 1998; 351:1251. 200. Rooks JB, Ory HW, Ishak KG, et al. Epidemiology of hepatocellular adenoma. JAMA 1979; 242:644. 201. Edmondson HA, Henderson B, Benton B. Liver-cell adenomas associated with use of oral contraceptives. N Engl J Med 1976; 294:470. 202. Wanless IR, Medline A. Role of estrogens as promoters of hepatic neoplasia. Lab Invest 1982; 46:313. 203. Buhler H, Pirovino M, Akovbiantz A, et al. Regression of liver cell adenoma. Gastroenterology 1982; 82:775. 204. Trotter JF, Everson GT. Benign focal lesions of the liver. Clin Liver Dis 2001; 5:17. 205. Powell-Jackson PR, Melia W, Canalese J, et al. Budd–Chiari syndrome: clinical patterns and therapy. Q J Med 1982;51:79. 206. Lewis JH, Tice HL, Zimmerman HJ. Budd–Chiari syndrome associated with oral contraceptive steroids. Review of treatment of 47 cases. Dig Dis Sci 1983; 28:673. 207. Singh V, Sinha SK, Nain CK, et al. Budd–Chiari syndrome: our experience of 71 patients. J Gastroent Hepatol 2000; 15:550. 208. Valla D-C. The diagnosis and management of the Budd–Chiari syndrome: consensus and controversies. Hepatology 2003; 38:793. 209. Khuroo MS, Datta DV. Budd–Chiari syndrome following pregnancy. Report of 16 cases, with roentgenologic, hemodynamic and histologic studies of the hepatic outflow tract. Am J Med 1980; 68:113. 210. Artigas JMG, Estabanez JS, Faure MR. Pregnancy and the Budd–Chiari syndrome. Dig Dis Sci 1982; 27:89. 211. Fickert P, Ramschak H, Kenner L, et al. Acute Budd–Chiari syndrome with fulminant hepatic failure in a pregnant woman with factor V Leiden mutation. Gastroenterology 1996; 111:510A. 212. Grant WJ, McCashland T, Botha JF, et al. Acute Budd–Chiari syndrome during pregnancy: surgical treatment and orthotopic liver transplantation with successful completion of the pregnancy. Liver Transpl 2003; 9:976. 213. Walker ID. Thrombophilia in pregnancy. J Clin Pathol 2000; 53:573. 214. Hodkinson HJ, McKibbin JK, Tim JO, et al. Postpartum venoocclusive disease treated with ascitic fluid reinfusion. S Afr Med J 1978; 54:366. 215. Girardin M-FS-M, Zafrani ES, Prigent A, et al. Unilobular small hepatic vein obstruction: possible role of progestogen given as oral contraceptive. Gastroenterology 1983; 84:630. 216. Abell TL, Riely CA. Hyperemesis gravidarum. Gastroenterol Clin North Am 1992; 21:835. 217. Morali GA, Braverman DZ. Abnormal liver enzymes and ketonuria in hyperemesis gravidarum. J Clin Gastroenterol 1990; 12:303. 218. Larrey D, Rueff B, Feldman G, et al. Recurrent jaundice caused by recurrent hyperemesis gravidarum. Gut 1984; 25:1414. 219. Sheehan HL. The pathology of hyperemesis and vomiting of late pregnancy. J Obstet Gynaecol Br Emp 1939; 46:685. 220. Haemmerli UP, Wyss HI. Recurrent intrahepatic cholestasis of pregnancy. Report of six cases, and review of the literature. Medicine 1967; 46:299. 221. Johnson P, Samsioe G, Gustafson A. Studies in cholestasis of pregnancy. I. Clinical aspects and liver function tests. Acta Obstet Gynecol Scand 1975; 54:77.
222. Ylostalo P, Ylikorkala O. Hepatosis of pregnancy. A clinical study of 107 patients. Ann Chir Gynaecol Fenniae 1975; 64:128. 223. Reyes H, Radrigan ME, Gonzalez MC, et al. Steatorrhea in patients with intrahepatic cholestasis of pregnancy. Gastroenterology 1987; 93:584. 224. Reyes H, Gonzalez MC, Ribalta J, et al. Colestasia intrahepatica de la embarazada: variabilidad clinica y bioquimica. Rev Med Chile 1982; 110:631. 225. Berg B, Helm G, Petersohn L, et al. Cholestasis of pregnancy. Acta Obstet Gynecol Scand 1986; 65:107. 226. Reyes H. The enigma of intrahepatic cholestasis of pregnancy: lessions from Chile. Hepatology 1982; 2:87. 227. Reyes H, Gonzalez MC, Ribalta J, et al. Prevalence of intrahepatic cholestasis of pregnancy in Chile. Ann Intern Med 1978; 88:487. 228. Reyes H, Taboada G, Ribalta J. Prevalence of intrahepatic cholestasis of pregnancy in La Paz, Bolivia. J Chron Dis 1979; 32:499. 229. Glantz A, Marschall H-U, Mattsson L-A. Intrahepatic cholestasis of pregnancy: Relationships between bile acid levels and fetal complication rates. Hepatology 2004; 40:467. 230. Paternoster DM, Fabris F, Palu G, et al. Intra-hepatic cholestasis of pregnancy in hepatitis C virus infection. Acta Obstet Gynecol Scand 2002; 81:99. 231. Roncaglia N, Arreghini A, Locatelli A, et al. Obstetric cholestasis: outcome with active management. Eur J Obstet Gynecol Reprod Biol 2002; 100:167. 232. Glasinovic JC, Marinovic I, Vela P. The changing scene of cholestasis of pregnancy in Chile. An epidemiological study. Hepatology 1986; 6:1161A. 233. Bacq Y, Sapey T, Brechot MC, et al. Intrahepatic cholestasis of pregnancy: a French prospective study. Hepatology 1997; 26:358. 234. Shanmugam S, Thappa DM, Habeebullah S. Pruritus gravidarum: a clinical and laboratory study. J Dermatol 1998; 25:582. 235. Ch’ng CL, Morgan M, Hainsworth I, et al. Prospective study of liver dysfunction in pregnancy in Southwest Wales. Gut 2002; 51:876. 236. Abedin P, Weaver JB, Egginton E. Intrahepatic cholestasis of pregnancy: prevalence and ethnic distribution. Ethn Health 1999; 4:35. 237. Rioseco AJ, Ivankovic MB, Manzur A, et al. Intrahepatic cholestasis of pregnancy: a retrospective case-control study of perinatal outcome. Am J Obstet Gynecol 1994; 170:890. 238. Gaudet R, Merviel P, Berkane N, et al. Fetal impact of cholestasis of pregnancy: experience at Tenon Hospital and literature review. Fetal Diagn Ther 2000; 15:191. 239. Nicastri PL, Diaferia A, Tartagni M, et al. A randomised placebo-controlled trial of ursodeoxycholic acid and Sadenosylmethionine in the treatment of intrahepatic cholestasis of pregnancy. Br J Obstet Gynaecol 1998; 105:1205. 240. Reyes H. Intrahepatic cholestasis. A puzzling disorder of pregnancy. J Gastroenterol Hepatol 1997; 12:211. 241. Heinonen S, Kirkinen P. Pregnancy outcome with intrahepatic cholestasis. Obstet Gynecol 1999; 94:189. 242. Cullberg G, Lundstrom R, Stenram U. Jaundice during treatment with an oral contraceptive, lyndiol. Br Med J 1965; 1:695. 243. Eisalo A, Jarvinen PA, Luukkainen T. Hepatic impairment during the intake of contraceptive pills: clinical trial with postmenopausal women. Br Med J 1964; 2:426. 244. Thulin KE, Nermark J. Seven cases of jaundice in women taking an oral contraceptive, anovlar. Br Med J 1966; 1:684. 245. Kreek MJ, Weser E, Sleisenger MH, et al. Idiopathic cholestasis of pregnancy. N Engl J Med 1967; 277:1391.
1025
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
246. Kreek MJ, Sleisenger MH, Jeffries GH. Recurrent cholestatic jaundice of pregnancy with demonstrated estrogen sensitivity. Am J Med 1967; 43:795. 247. Reyes H, Sjovall J. Bile acids and progesterone metabolites in intrahepatic cholestasis of pregnancy. Ann Med 2000; 32:94. 248. Reyes H, Baez ME, Gonzalez MC, et al. Selenium, zinc and copper plasma levels in intrahepatic cholestasis of pregnancy, in normal pregnancies and in healthy individuals, in Chile. J Hepatol 2000; 32:542. 249. Stieger B, Fattinger K, Madon J, et al. Drug- and estrogeninduced cholestasis through inhibition of the hepatocellular bile salt export pump (BSEP) of rat liver. Gastroenterology 2000; 118:422. 250. Micheline D, Emmanuel J, Serge E. Effect of ursodeoxycholic acid on the expression of the hepatocellular bile acid transporters (NTCP and BSEP) in rats with estrogen-induced cholestasis. J Pediatr Gastroenterol Nutr 2002; 35:185. 251. Reyes H, Ribalta J, Gonzalez-Ceron M. Idiopathic cholestasis of pregnancy in a large kindred. Gut 1976; 17:709. 252. Holzbach RT, Sivak DA, Braun WE. Familial recurrent intrahepatic cholestasis of pregnancy: a genetic study providing evidence for transmission of a sex-limited, dominant trait. Gastroenterology 1983; 85:175. 253. De Pagter AGF, van Berge Henegouwen GP, ten Bokkel Huinink JA, et al. Familial benign recurrent intrahepatic cholestasis. Gastroenterology 1976; 71:202. 254. Dalen E, Westerholm B. Occurrence of hepatic impariment in women jaundiced by oral contraceptives and in their mothers and sisters. Acta Med Scand 1974; 195:459. 255. Eloranta M-J, Heinonen S, Mononen T, et al. Risk of obstetric cholestasis in sisters of index patients. Clin Genet 2001; 60:42. 256. Reyes H, Ribalta J, Gonzalez C, et al. Sulfobromophthalein clearance tests before and after ethinyl estradiol administration, in women and men with familial history of intrahepatic cholestasis of pregnancy. Gastroenterology 1981; 81:226. 257. Jacquemin E, Cresteil D, Manouvrier S, et al. Heterozygous non-sense mutation of the MDR3 gene in familial intrahepatic cholestasis of pregnancy. Lancet 1999; 353:210. 258. Strautnieks S, Lopes A, Underhill J, et al. Critical residues in the multidrug resistance 3 protein gene associated with adult onset cholangiopathy and intrahepatic cholestasis of pregnancy. Hepatology 2002; 36:415A. 259. Jacquemin E, De Vree JML, Cresteil D, et al. The wide spectrum of multidrug resistance 3 deficiency: from neonatal cholestasis to cirrhosis of adulthood. Gastroenterology 2001; 120:1448. 260. Dixon PH, Weerasekera N, Linton KJ, et al. Heterozygous MDR3 missense mutation associated with intrahepatic cholestasis of pregnancy: evidence for a defect in protein trafficking. Hum Mol Genet 2000; 9:1209. 261. Simon E, Marschall H-U, Glantz A, et al. Mutations of hepatocanalicular ABC transporters in intrahepatic cholestasis of pregnancy (ICP). Hepatology 2002; 36:335A. 262. Lucena J-F, Herrero JI, Quiroga J, et al. A multidrug resistance 3 gene mutation causing cholelithiasis, cholestasis of pregnancy, and adulthood biliary cirrhosis. Gastroenterology 2003; 124:1037. 263. Gendrot C, Bacq Y, Brechot M-C, et al. A second heterozygous MDR3 nonsense mutation associated with intrahepatic cholestasis of pregnancy. J Med Genet 2003; 40:e32, (http://www.jmedgenet.com/cgi/content/full/40/3/e32). 264. Mullenbach R, Linton KJ, Wiltshire S, et al. ABCB4 gene sequence variation in women with intrahepatic cholestasis of pregnancy. J Med Genet 2003; 40:e70, (http://www.jmedgenet.com/cgi/content/full/40/5/e70). 265. Savander M, Ropponen A, Avela K, et al. Genetic evidence of heterogeneity in intrahepatic cholestasis of pregnancy. Gut 2003; 52:1025.
1026
266. Eloranta M-L, Heiskanen JTM, Hiltunen MJ, et al. Multidrug resistance 3 gene mutation 1712delT and estrogen receptor a gene polymorphisms in Finnish women with obstetric cholestasis. Eur J Obstet Gynecol Reprod Biol 2002; 105:132. 267. Pauli-Magnus C, Lang T, Meier Y, et al. Sequence analysis of bile salt export pump (ABCB11) and multidrug resistance pglycoprotein 3 (ABCB4, MDR3) in patients with intrahepatic cholestasis of pregnancy. Pharmacogenetics 2004; 14:91. 268. Eloranta J-L, Hakli T, Hiltunen M, et al. Association of single nucleotide polymorphisms of the bile salt export pump gene with intrahepatic cholestasis of pregnancy. Scand J Gastroenterol 2003; 38:648. 269. Davidson KM. Intrahepatic cholestasis of pregnancy. Semin Perinatol 1998; 22:104. 270. Brites D, Rodrigues CMP, Oliveira N, et al. Correction of maternal serum bile acid profile during ursodeoxycholic acid therapy in cholestasis of pregnancy. J Hepatol 1998; 28:91. 271. Serrano MA, Brites D, Larena MG, et al. Beneficial effect of ursodeoxycholic acid on alterations induced by cholestasis of pregnancy in bile acid transport across the human placenta. J Hepatol 1998; 28:829. 272. Laifer SA, Stiller RJ, Siddiqui DS, et al. Ursodeoxycholic acid for the treatment of intrahepatic cholestasis of pregnancy. J Maternal-Fetal Med 2001; 10:131. 273. Kirkinen P, Ylostalo P, Heikkinen J, et al. Gallbladder function and maternal bile acids in intrahepatic cholestasis of pregnancy. Eur J Obstet Gynec Reprod Biol 1984; 18:29. 274. Adlercreutz H, Svanborg A, Anberg A. Recurrent jaundice in pregnancy. I. A clinical and ultrastructual study. Am J Med 1967; 42:335. 275. Van Haelst U, Bergstein N. Electron microscopic study of the liver in so-called idiopathic jaundice of late pregnancy. Path Eur 1970; 5:198. 276. Zapata R, Sandoval L, Palma J, et al. Ursodeoxycholic acid in the treatment of intrahepatic cholestasis of pregnancy. A 10year experience on its efficacy, safety and the perinatal outcome. Gastroenterology 2000; 118:A1008. 277. Olsson R, Tysk C, Aldenborg F, et al. Prolonged postpartum course of intrahepatic cholestasis of pregnancy. Gastroenterology 1993; 105:267. 278. Furhoff A-K, Hellstrom K. Jaundice in pregnancy. A follow-up study of the series of women originally reported by L. Thorling. II. Present health of the women. Acta Med Scand 1974; 196:181. 279. Leevy CB, Koneru B, Klein KM. Recurrent familial prolonged intrahepatic cholestasis of pregnancy associated with chronic liver disease. Gastroenterology 1997; 113:966. 280. Laatikainen T, Ikonen E. Fetal prognosis in obstetric hepatosis. Ann Chir Gynaecol Fenniae 1975; 64:155. 281. Reid R, Ivey KJ, Rencoret RH, et al. Fetal complications of obstetric cholestasis. Br Med J 1976; 1:870. 282. Davies MH, da Silva RCMA, Elias E, et al. Fetal mortality associated with cholestasis of pregnancy and the potential benefit of therapy with ursodeoxycholic acid. Gut 1995; 37:580. 283. Alsulyman OM, Ouzounian JG, Ames-Castro M, et al. Intrahepatic cholestasis of pregnancy: perinatal outcome associated with expectant management. Am J Obstet Gynecol 1996; 175:957. 284. Savonius H, Riikonen S, Gylling H, et al. Pregnancy outcome with intrahepatic cholestasis. Acta Obstet Gynecol Scand 2000; 79:323. 285. Germain AM, Kato S, Carvajal JA, et al. Bile acids increase response and expression of human myometrial oxytocin receptor. Am J Obstet Gynecol 2003; 189:577. 286. Howard PJ, Murphy GM. Bile acid stress in the mother and baby unit. Eur J Gastroenterol Hepatol 2003; 15:317.
Chapter 54 THE LIVER IN PREGNANCY
287. Germain AM, Carvajal JA, Glasinovic JC, et al. Intrahepatic cholestasis of pregnancy: an intriguing pregnancy-specific disorder. J Soc Gynecol Invest 2002; 9:10. 288. Lammert F, Marschall H-U, Glantz A, et al. Intrahepatic cholestasis of pregnancy: molecular pathogenesis, diagnosis and management. J Hepatol 2000; 33:1012. 289. Matos A, Bernardes J, Ayres-de-Campos D, et al. Antepartum fetal cerebral hemorrhage not predicted by current surveillance methods in cholestasis of pregnancy. Obstet Gynecol 1997; 89:803. 290. Heikkinen J, Maentausta O, Ylostalo P, et al. Serum bile acid levels in intrahepatic cholestasis of pregnancy during treatment with phenobarbital or cholestryamine. Eur J Obstet Gynec Reprod Biol 1982; 14:153. 291. Trauner M, Graziadei IW. Mechanisms of action and therapeutic applications of ursodeoxycholic acid in chronic liver diseases. Aliment Pharmacol Ther 1999; 13:979. 292. Palma J, Reyes H, Ribalta J, et al. Effects of ursodeoxycholic acid in patients with intrahepatic cholestasis of pregnancy. Hepatology 1992; 15:1043. 293. Palma J, Reyes H, Ribalta J, et al. Ursodeoxycholic acid in the treatment of cholestasis of pregnancy: a randomized, double-blind study controlled with placebo. J Hepatol 1997; 27:1022. 294. Diaferia A, Nicastri PL, Tartagni M, et al. Ursodeoxycholic acid therapy in pregnant women with cholestasis. Int J Gynaecol Obstet 1996; 52:133. 295. Floreani A, Paternoster D, Melis A, et al. S-adenosylmethionine versus ursodeoxycholic acid in the treatment of intrahepatic cholestasis of pregnancy: preliminary results of a controlled trial. Eur J Obstet Gynecol Reprod Biol 1996; 67:109. 296. Mazzella G, Nicola R, Francesco A, et al. Ursodeoxycholic acid administration in patients with cholestasis of pregnancy: effects on primary bile acids in babies and mothers. Hepatology 2001; 33:504. 296a. Kondrackiene J, Beuers U, Kupcinskas L. Efficacy and safety of ursodeoxycholic acid versus cholestyramine in intrahepatic cholestasis of pregnancy. Gastroenterology 2005; 129:894. 297. Meng L-J, Reyes H, Palma J, et al. Effects of ursodeoxycholic acid on conjugated bile acids and progesterone metabolites in serum and urine of patients with intrahepatic cholestasis of pregnancy. J Hepatol 1997; 27:1029. 298. Jenkins JK, Boothby LA. Treatment of itching associated with intrahepatic cholestasis of pregnancy. Ann Pharmacother 2002; 36:1462. 299. den Dulk M, Valentijn RM, Welten CAHC, et al. Intrahepatic cholestasis of pregnancy. Netherlands J Med 2002; 60:366. 300. Mela M, Mancuso A, Burroughs AK. Review article: pruritus in cholestatic and other liver diseases. Aliment Pharmacol Ther 2003; 17:857. 301. Di Padova C, Tritapepe R, Cammareri G, et al. S-adenosyl-Lmethionine antagonizes ethynylestradiol-induced bile cholesterol supersaturation in humans without modifying the estrogen plasma kinetics. Gastroenterology 1982; 82:233. 302. Stramentinoli G, di Padova C, Gualano M, et al. Ethynylestradiol-induced impairment of bile secretion in the rat: protective effects of S-adenosyl-L-methionine and its implication in estrogen metabolism. Gastroenterology 1981; 80:154. 303. Hirvioja M-L, Tuimala R, Vuori J. The treatment of intrahepatic cholestasis of pregnancy by dexamethasone. Br J Obstet Gynaecol 1992; 99:109. 304. Ribalta J, Reyes H, Gonzalez M, et al. S-Adenosyl-L-methionine in the treatment of patients with intrahepatic cholestasis of pregnancy: a randomized, double-blind, placebo-controlled study with negative results. Hepatology 1991; 13:1084. 305. Frezza M, Centini G, Cammareri G, et al. Sadenosylmethionine for the treatment of intrahepatic
306.
307. 308.
309. 310. 311. 312.
313.
314.
315. 316.
317.
318.
319.
320.
321.
322.
323.
324.
325.
326.
cholestasis of pregnancy. Results of a controlled clinical trial. Hepato-gastroenterol 1990; 37 (Suppl II):122. Castro MA, Fassett MJ, Reynolds TB, et al. Reversible peripartum liver failure: a new perspective on the diagnosis, treatment, and cause of acute fatty liver of pregnancy, based on 28 consecutive cases. Am J Obstet Gynecol 1999; 181:389. Treem WR. Mitochondrial fatty acid oxidation and acute fatty liver of pregnancy. Semin Gastrointest Dis 2002; 13:55. Bourgeois C, Olson L, Comer D, et al. Encephalopathy and fatty degeneration of the viscera: a clinicopathologic analysis of 40 cases. Am J Clin Pathol 1971; 56:558. Passayre D, Mansouri A, Haouzi D, et al. Hepatotoxicity due to mitochondrial dysfunction. Cell Biol Toxicol 1999; 15:367. Rinaldo P, Matern D, Bennett MJ. Fatty acid oxidation disorders. Annu Rev Physiol 2002; 64:477. Rolfes DB, Ishak KG. Acute fatty liver of pregnancy: a clinicopathologic study of 35 cases. Hepatology 1985; 5:1149. Fromenty B, Berson A, Pessayre D. Microvesicular steatosis and steatohepatitis: role of mitochondrial dysfunction and lipid peroxidation. J Hepatol 1997; 26 (Suppl. 1):13. Davies MH, Wilkinson SP, Hanid MA, et al. Acute liver disease with encephalopathy and renal failure in late pregnancy and early puerperium: a study of fourteen patients. Br J Obstet Gynaecol 1980; 87:1005. Pereira SP, O’Donohue J, Wendon J, et al. Maternal and perinatal outcome in severe pregnancy-related liver disease. Hepatology 1997; 26:1258. Pockros PJ, Peters RL, Reynolds TB. Idiopathic fatty liver of pregnancy: findings in ten cases. Medicine 1984; 63:1. Reyes H, Sandoval L, Wainstein A, et al. Acute fatty liver of pregnancy: a clinical study of 12 episodes in 11 patients. Gut 1994; 35:101. Eisele JW, Barker EA, Smuckler EA. Lipid content in the liver of fatty metamorphosis of pregnancy. Am J Pathol 1975; 81:545. Weber FL Jr, Snodgrass PJ, Powell DE, et al. Abnormalities of hepatic mitochondrial urea-cycle enzyme activities and hepatic ultrastructure in acute fatty liver of pregnancy. J Lab Clin Med 1979; 94:27. Burroughs AK, Seong NH, Dojcinov DM, et al. Idiopathic acute fatty liver of pregnancy in 12 patients. Q J Med 1982; 51:481. Usta IM, Barton JR, Amon EA, et al. Acute fatty liver of pregnancy: an experience in the diagnosis and management of fourteen cases. Am J Obstet Gynecol 1994; 171:1342. Schoeman MN, Batey RG, Wilcken B. Recurrent acute fatty liver of pregnancy associated with a fatty-acid oxidation defect in the offspring. Gastroenterology 1991; 100:544. Visconti M, Manes G, Giannattasio F, et al. Recurrence of acute fatty liver of pregnancy. J Clin Gastroenterol 1995; 21:243. Tyni T, Ekholm E, Pihko H. Pregnancy complications are frequent in long-chain 3-hydroxyacyl-coenzyme A dehydrogenase deficiency. Am J Obstet Gynecol 178:603, 1998. Castro MA, Goodwin TM, Shaw KJ, et al. Disseminated intravascular coagulation and antithrombin III depression in acute fatty liver of pregnancy. Am J Obstet Gynecol 1996; 174:211. Strauss AW, Bennett MJ, Rinaldo P, et al. Inherited long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency and a fetal–maternal interaction cause maternal liver disease and other pregnancy complications. Semin Perinat 1999; 23:100. Yang Z, Zhao Y, Bennett MJ, et al. Fetal genotypes and pregnancy outcomes in 35 families with mitochondrial trifunctional protein mutations. Am J Obstet Gynecol 2002; 187:715.
1027
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
327. Rakheja D, Bennett MJ, Rogers BB. Long-chain L-3hydroxyacyl-coenzyme A dehydrogenase deficiency: a molecular and biochemical review. Lab Invest 2002; 82:815. 328. Sattar N, Gaw A, Packard CJ, et al. Potential pathogenic roles of aberrant lipoprotein and fatty acid metabolism in preeclampsia. Br J Obstet Gynaecol 1996; 103:614. 329. Grimbert S, Fisch C, Deschamps D, et al. Effects of female sex hormones on mitochondria: possible role in acute fatty liver of pregnancy. Am J Physiol 1995; 268:G107. 330. Baldwin GS. Do NSAIDs contribute to acute fatty liver of pregnancy? Med Hypoth 2000; 54:846. 331. Yang Z, Yamada J, Zhao Y, et al. Prospective screening for pediatric mitochondrial trifunctional protein defects in pregnancies complicated by liver disease. JAMA 2002; 288:2163. 332. Preece MA, Green A. Pregnancy and inherited metabolic disorders: maternal and fetal complications. Ann Clin Biochem 2002; 39:444. 333. Ibdah JA, Bennett MJ, Rinaldo P, et al. A fetal fatty-acid oxidation disorder as a cause of liver disease in pregnant women. N Engl J Med 1999; 340:1723. 334. Treem WR, Shoup ME, Hale DE, et al. Acute fatty liver of pregnancy, hemolysis, elevated liver enzymes, and low platelets syndrome, and long chain 3-hydroxyacyl-coenzyme A dehydrogenase deficiency. Am J Gastroenterol 1996; 91:2293. 335. Wilcken B, Leung K-C, Hammond J, et al. Pregnancy and fetal long-chain 3-hydroxyacyl coenzyme A dehydrogenase deficiency. Lancet 1993; 341:407. 336. Treem WR, Rinaldo P, Hale DE, et al. Acute fatty liver of pregnancy and long-chain 3-hydroxyacyl-coenzyme A dehydrogenase deficiency. Hepatology 1994; 19:339. 337. Isaacs JD, Sims HF, Powell CK, et al. Maternal acute fatty liver of pregnancy associated with fetal trifunctional protein deficiency: molecular characterization of a novel maternal mutant allele. Pediatr Res 1996; 40:393. 338. Rakheja D, Bennett MJ, Foster BM, et al. Evidence for fatty acid oxidation in human placenta, and the relationship of fatty acid oxidation enzyme activities with gestational age. Placenta 2002; 23:447. 339. Shekhawat P, Bennett MJ, Sadovsky Y, et al. Human placenta metabolizes fatty acids: implications for fetal fatty acid oxidation disorders and maternal liver diseases. Am J Physiol Endocrinol Metab 2003; 284:E1098. 340. Hintz SR, Matern D, Strauss A, et al. Early neonatal diagnosis of long-chain 3-hydroxyacyl coenzyme A dehydrogenase and mitochondrial trifunctional protein deficiencies. Molec Genet Metab 2002; 75:120. 341. Ibdah JA, Zhao Y, Viola J, et al. Molecular prenatal diagnosis in families with fetal mitochondrial trifunctional protein mutations. J Pediatr 2001; 138:396. 342. Tein I. Metabolic disease in the fetus predisposes to maternal hepatic complications of pregnancy. Pediatr Res 2000; 47:6. 343. Riely CA, Latham PS, Romero R, et al. Acute fatty liver of pregnancy: a reassessment based on observations in nine patients. Ann Intern Med 1987; 106:703. 344. Slater DN, Hague WM. Renal morphological changes in idiopathic acute fatty liver of pregnancy. Histopathology 1984; 8:567. 345. Farine D, Newhouse J, Owen J, et al. Magnetic resonance imaging and computed tomography scan for the diagnosis of acute fatty liver of pregnancy. Am J Perinatol 1990; 7:316. 346. Moise KJ Jr, Shah DM. Acute fatty liver of pregnancy: etiology of fetal distress and fetal wastage. Obstet Gynecol 1987; 69:482. 347. Barton JR, Sibai BM, Mabie WC, et al. Recurrent acute fatty liver of pregnancy. Am J Obstet Gynecol 1990; 163:534.
1028
348. Jenkins SM, Head BB, Hauth JC. Severe preeclampsia at <25 weeks of gestation: maternal and neonatal outcomes. Am J Obstet Gynecol 2002; 186:790. 349. Pridjian G, Puschett JB. Preeclampsia. Part 1: clinical and pathophysiologic considerations. Obstet Gynecol Surv 2002; 57:598. 350. Alexander BT, Bennett WA, Khalil RA, et al. Preeclampsia: linking placental ischemia with cardiovascular-renal dysfunction. News Physiol Sci 2001; 16:282. 351. Benedetto C, Marozio L, Salton L, et al. Factor V Leiden and factor II G20210A in preeclampsia and HELLP syndrome. Acta Obstet Gynecol Scand 2002; 81:1095. 352. Bloomenthal D, von Dadelszen P, Liston R, et al. The effect of factor V Leiden carriage on maternal and fetal health. CMAJ 2002; 167:48. 353. VanWijk MJ, Kublickiene K, Boer K, et al. Vascular function in preeclampsia. Cardiovasc Res 2000; 47:38. 354. Saphier CJ, Repke JT. Hemolysis, elevated liver enzymes, and low platelets (HELLP) syndrome: a review of diagnosis and management. Semin Perinatol 1998; 22:118. 355. Ellison J, Sattar N, Greer I. HELLP syndrome: mechanisms and management. Hosp Med 1999; 60:243. 356. Magann EF, Martin JN. Twelve steps to optimal management of HELLP syndrome. Clin Obstet Gynecol 1999; 42:532. 357. den Boer ME , Ijlst L, Wijburg FA, et al. Heterozygosity for the common LCHAD mutation (1528G > C) is not a major cause of HELLP syndrome and the prevalence of the mutation in the Dutch population is low. Pediatr Res 2000; 48:151. 358. Beller FK, Dame WR, Ebert C. Pregnancy induced hypertension complicated by thrombocytopenia, haemolysis and elevated liver enzymes (HELLP) syndrome. Renal biopsies and outcome. Aust NZ J Obstet Gynecol 1985; 25:83. 359. Sibai BM, Taslimi MM, El-Nazer A, et al. Maternal–perinatal outcome associated with the syndrome of hemolysis, elevated liver enzymes, and low platelets in severe preeclampsiaeclampsia. Am J Obstet Gynecol 1986; 155:501. 360. Sibai BM, Ramadan MK, Usta I, et al. Maternal morbidity and mortality in 442 pregnancies with hemolysis, elevated liver enzymes, and low platelets (HELLP syndrome). Am J Obstet Gynecol 1993; 169:1000. 361. Martin JN, Rinehart BK, May WL, et al. The spectrum of severe preeclampsia: comparative analysis by HELLP (hemolysis, elevated liver enzyme levels, and low platelet count) syndrome classification. Am J Obstet Gynecol 1999; 180:1373. 362. Ertan AK, Wagner S, Hendrik HJ, et al. Clinical and biophysical aspects of HELLP-syndrome. J Perinat Med 2002; 30:483. 363. Abraham KA, Connolly G, Farrell J, et al. The HELLP syndrome, a prospective study. Renal Failure 2001; 23:705. 364. Villegas H, Azuela JC, Maqueo M. Spontaneous rupture of liver in toxemia of pregnancy. Int J Gynecol Obstet 1970; 8:836. 365. Hibbard LT. Maternal mortality due to acute toxemia. Obstet Gynecol 1973; 42:263. 366. Arias F, Mancilla-Jimenez R. Hepatic fibrinogen depositis in pre-eclampsia. N Engl J Med 1976; 295:578. 367. Sheehan HL, Lynch JB. Pathology of toxemia of pregnancy. Edinburgh: Churchill Livingstone, 1973:328. 368. Dani R, Mendes GS, Medeiros JdL, et al. Study of the liver changes occurring in preeclampsia and their possible pathogenetic connection with acute fatty liver of pregnancy. Am J Gastroenterol 1996; 91:292. 369. Barton JR, Riely CA, Adamee TA. Hepatic histopathologic condition does not correlate with laboratory abnormalities in HELLP syndrome (hemolysis, elevated liver enzymes, and low platelet count). Am J Obstet Gynecol 1992; 167:1538.
Chapter 54 THE LIVER IN PREGNANCY
370. Weinstein L. Syndrome of hemolysis, elevated liver enzymes, and low platelet count: a severe consequence of hypertension in pregnancy. Am J Obstet Gynecol 1982; 142:159. 371. Reubinoff BE, Schenker JG. HELLP syndrome – a syndrome of hemolysis, elevated liver enzymes and low platelet count – complicating preeclampsia-eclampsia. Int J Gynecol Obstet 1991; 36:95. 372. Sibai BM. The HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets). Am J Obstet Gynecol 1990; 162:311. 373. Clark SL, Phelan JR, Allen SH, et al. Antepartum reversal of hematologic abnormalities associated with the HELLP syndrome. J Reprod Med 1986; 31:70. 374. Weinstein L. Preeclampsia/eclampsia with hemolysis, elevated liver enzymes, and thrombocytopenia. Obstet Gynecol 1985; 66:657. 375. Haddad B, Barton JR, Livingston JC, et al. Risk factors for adverse maternal outcomes among women with HELLP (hemolysis, elevated liver enzymes, and low platelet count) syndrome. Am J Obstet Gynecol 2000; 183:444. 376. Woods JB, Blake PG, Perry KG Jr, et al. Ascites: a portent of cardiopulmonary complications in the preeclamptic patient with the syndrome of hemolysis, elevated liver enzymes, and low platelets. Obstet Gynecol 1992; 80:87. 377. Freund G, Arvan DA. Clinical biochemistry of preeclampsia and related liver diseases of pregnancy. Clinica Chimica Acta 1990; 191:123. 378. Faridi A, Heyl W, Rath W. Preliminary results of the International HELLP-Multicenter-Study. Int J Gynaecol Obstet 2000; 69:279. 379. Rath W, Loos W, Kuhn W, et al. The importance of early laboratory screening methods for maternal and fetal outcome in cases of HELLP syndrome. Eur J Obstet Gynecol Reprod Biol 1990; 36:43. 380. Martin JN Jr, Blake PG, Perry KG Jr, et al. The natural history of HELLP syndrome. Am J Obstet Gynecol 1991; 164:1500. 381. Sullivan CA, Magann EF, Perry KG, et al. The recurrence risk of the syndrome of hemolysis, elevated liver enzymes, and low platelets (HELLP) in subsequent gestations. Am J Obstet Gynecol 1994; 171:940. 382. Sibai BM, Ramadan MK, Chari RS, et al. Pregnancies complicated by HELLP syndrome (hemolysis, elevated liver
383.
384.
385.
386. 387.
388.
389.
390.
391.
392.
393.
394.
395.
enzymes, and low platelets): subsequent pregnancy outcome and long-term prognosis. Am J Obstet Gynecol 1995; 172:125. Martin JN, Thigpen BD, Rose CH, et al. Maternal benefit of high-dose intravenous corticosteroid therapy for HELLP syndrome. Am J Obstet Gynecol 2003; 189:830. Isler CM, Magann EF, Rinehart BK, et al. Dexamethasone compared with betamethasone for glucocorticoid treatment of postpartum HELLP syndrome. Int J Gynecol Obstet 2003; 80:291. Minakami H, Oka N, Sato T, et al. Preeclampsia: a microvesicular fat disease of the liver? Am J Obstet Gynecol 1988; 159:1043. Ralston SJ, Schwaitzberg SD. Liver hematoma and rupture in pregnancy. Semin Perinatol 1998; 22:141. Manas KJ, Welsh JD, Rankin RA, et al. Hepatic hemorrhage without rupture in preeclampsia. N Engl J Med 1985; 312:424. Stain SC, Woodburn DA, Stephens AL. Spontaneous hepatic hemorrhage associated with pregnancy treatment by hepatic arterial interruption. Ann Surg 1996; 224:73. Dammann HG, Hagemann J, Runge M, et al. In vivo diagnosis of massive hepatic infarction by computed tomography. Dig Dis Sci 198; 27:73. Sheikh RA, Yasmeen S, Pauly MP, et al. Spontaneous intrahepatic hemorrhage and hepatic rupture in the HELLP syndrome. J Clin Gastroenterol 1999; 28:323. Marsh FA, Kaufmann SJ, Bhabra K. Surviving hepatic rupture in pregnancy – a literature review with an illustrative case report. J Obstet Gynaecol 2003; 23:109. Smith LG Jr, Moise KJ Jr, Dildy GA III, et al. Spontaneous rupture of liver during pregnancy. Obstet Gynecol 1991; 77:171. Rinehart BK, Terrone DA, Magann EF, et al. Preeclampsiaassociated hepatic hemorrhage and rupture: mode of management related to maternal and perinatal outcome. Obstet Gynecol Surv 1999; 54:196. Hunter SK, Martin M, Benda JA, et al. Liver transplant after massive spontaneous hepatic rupture in pregnancy complicated by preeclampsia. Obstet Gynecol 1995; 85:819. Lavery JP, Berg J. Subcapsular hematoma of the liver during pregnancy. S Med J 1989; 82:1568.
1029
Section IX: Liver Affected by Other Organs (Non-Hepatic Conditions)
55
NASH Brent A. Neuschwander-Tetri Abbreviations ALT alanine aminotransferase AMPK AMP-activated protein kinase apoE apolipoprotein E ASH alcoholic steatohepatitis ATP adenosine triphosphate BMI body mass index CAH chronic active hepatitis CT computed tomography DPP diabetes prevention program GGT g-glutamyltranspeptidase HCC hepatocellular carcinoma HCV hepatitis C virus HMG-CoA hydroxymethylglutaryl-coenzyme A
HOMA-IR homeostasis model assessment method for insulin resistance IRS-1 and insulin receptor substrates 1 and 2 IRS-2 LDL low-density lipoprotein MCPA methylenecyclopropylalanine MTTP microsomal triglyceride transfer protein MR magnetic resonance NADH nicotinamide-adenine dinucleotide NAFLD non-alcoholic fatty liver disease NAFL non-alcoholic fatty liver NASH non-alcoholic steatohepatitis NCEP national cholesterol education program
INTRODUCTION Non-alcoholic steatohepatitis (NASH) is a subset of non-alcoholic fatty liver disease (NAFLD). It is a disorder currently characterized by a constellation of histological abnormalities identified on liver biopsy that are similar to those seen in alcoholic liver disease but in patients who consume little or no alcohol. NASH is a worrisome form of liver disease because its prevalence is probably increasing in parallel with the dramatic increases in obesity and sedentary lifestyle that have become well documented over the past two decades. Of greatest concern, NASH also appears to be responsible for increased rates of cirrhosis, hepatocellular cancer, liver transplantation, and death.
HISTORICAL ASPECTS Jurgen Ludwig, a Mayo Clinic pathologist, popularized the term “non-alcoholic steatohepatitis” in a paper published in 1980 that describes a constellation of histological abnormalities in patients who denied alcohol abuse yet had histological findings on liver biopsy strongly suggesting alcoholic liver disease.1 The main message of this seminal paper was to challenge the assumption held by many at the time that such patients were covertly consuming excessive amounts of alcohol. This paper led to a wider recognition of NASH and sparked exponential growth in the field over the ensuing decades. By current standards, this early study was relatively small, with only 20 patients, some of whom may have had other causes of liver disease such as hepatitis C (testing was not possible at the time), hepatitis B (1 out of 9 tested had detectable HBsAg), and primary biliary cirrhosis (1 out of 5 tested had a positive antimitochondrial antibody titer). Nevertheless, the current level of awareness of NASH and research into its etiology and treatment can be attributed to research prompted by this case series.
PCOS polycystic ovary syndrome PI-3 kinase phosphatidyl inositol-3 kinase PPARa peroxisome proliferator-activated receptor-a QUICKI quantitative insulin sensitivity check index SIClamp glucose clamp-derived index of insulin sensitivity SVR sustained virologic response rate TNF-a tumor necrosis factor-a TZDs thiazolidinediones UDCA ursodeoxycholic acid ULN upper limit of normal VLDL very-low-density lipoprotein
What we now recognize as NASH did not completely escape the attention of clinicians and pathologists before Ludwig’s paper was published. Earlier papers and abstracts described similar patients but did not deliver quite the same message. For example, in 1952, Zelman described liver biopsy findings in 19 obese men that included steatosis and varying degrees of inflammation and fibrosis.2 Leevy described 270 patients with fatty liver, 64 of whom did not consume alcohol,3 and others also described similar experiences. These early studies set the stage for the Ludwig series that convinced many in the field that NASH was a potentially serious disorder that was truly unrelated to alcohol consumption and could no longer be ignored.
NOMENCLATURE NAFL, NAFLD, AND NASH The abnormalities in the liver described in patients historically associated with obesity or diabetes have been given various names, including fatty liver hepatitis, idiopathic steatohepatitis, steatofibrosis, steatonecrosis, pseudoalcoholic liver disease, and alcohol-like liver disease.4 After the term NASH gained widespread use, a need to name the full spectrum of liver disease associated with fat accumulation in the absence of alcohol abuse and other identifiable causes became apparent, whether or not necroinflammatory changes and fibrosis were present. Thus, the umbrella terms non-alcoholic fatty liver (NAFL) and NAFLD, both of which include NASH as a subset, were adopted.5,6 Which of these two terms, NAFL or NAFLD, is more accurate may be a philosophical question rather than a medical issue.7 The question of whether the presence of fat in the liver constitutes a disease or an inconsequential abnormality may be impossible to answer with empirical data. Undoubtedly, some patients may have a small amount of fat in the liver that is of no consequence. On the
1031
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
other hand, substantial fat in the liver, even without steatohepatitis, accelerates the progression of concomitant liver diseases such as chronic hepatitis C,8 increases the risk of surgery, and renders a liver unacceptable as a donor organ for transplant because of its intolerance to ischemia and reperfusion. The presence of significant amounts of fat with low-grade necroinflammatory changes that fall short of the current definition of NASH may deserve to be identified as a disease, or NAFLD, because of these associated adverse sequelae. NASH is currently defined as a constellation of findings that includes steatosis and necroinflammatory injury (Figure 55-1, Table 55-1).4 Essential for the diagnosis is the pathologist’s overall impression that the findings meet the criteria for steatohepatitis along with the clinician’s assessment that the patient is drinking little or no alcohol. Excluding other causes of liver disease was once thought to be important, but NASH is now recognized to occur in the pres-
ence of other forms of liver disease.9 This is probably even true in the case of alcoholic liver disease. The patient who is obese and has features of insulin resistance, but also consumes hepatotoxic quantities of alcohol, may have features of both forms of liver injury, yet current diagnostic criteria cannot make this distinction. Epidemiological data suggest that both insulin resistance and alcohol may contribute to the liver disease in such patients since obese alcoholics tend to have more hepatic steatosis than either non-obese alcoholics or obese non-alcoholics.10 Because NASH is a histological diagnosis and not a pathophysiological process per se, one could argue that it is best described as a syndrome rather than a disease. An apt hepatological precedent is the term chronic active hepatitis, a term once applied to chronic inflammatory changes in the liver that were most commonly localized around the portal tract. This term, and its abbreviation (CAH), served a valuable purpose in identifying patients who had similar findings so that they could be studied and specific underlying pathophysiological mechanisms understood. However, by 1995, the term was dropped in favor of specific diseases such as chronic hepatitis B and C, diseases that were subsequently characterized with respect to their own underlying mechanisms of injury, spectrum of histological abnormalities, natural history, and treatment. Looking towards the future, the understanding and nomenclature of NASH may evolve similarly to the point where specific underlying abnormalities are identified (e.g., insulin resistance; see below) and the full histological spectra of liver abnormalities associated with these abnormalities are described.
SIMILAR DISORDERS Wilson’s disease can cause triglyceride accumulation of necroinflammatory changes characteristic of NASH. It can mimic NASH
A
Table 55-1. Diagnostic Criteria for Non-Alcoholic Steatohepatitis (NASH)4 Histopathological criteria
Comment
Hepatocellular steatosis
> 5% of hepatocytes must contain macrovesicular fat Findings must be consistent with a diagnosis of steatohepatitis Highly suggestive but not required feature Ropy condensation of cytoskeletal elements often seen in conjunction with ballooning Mix of mononuclear and polymorphonuclear cells (presence of neutrophils not required for diagnosis) Represent apoptotic hepatocytes Pericellular “chicken-wire” fibrosis around hepatocytes in zone 3 (near central vein) is characteristic of NASH and alcoholic liver disease but is not a required diagnostic feature
Necroinflammatory changes Hepatocellular ballooning Mallory’s hyaline
Inflammatory cells
Acidophil bodies Fibrosis
B
Figure 55-1. Histopathological features of non-alcoholic steatohepatitis (NASH). (A) Moderate steatosis and necroinflammatory changes are evident in a predominantly zone-3 distribution (around the central vein) (hematoxylin and eosin stain, 40¥). (B) Pericellular (“chicken-wire”) fibrosis is seen as delicate strands of collagen by the blue staining around hepatocytes in zone 3 of the hepatic acinus (trichrome stain, 40¥). (Courtesy of E.M. Brunt.)
1032
Clinical criteria Absence of alcohol abuse
Corroborated consumption <20 grams daily confirms “non-alcoholic”; alcohol consumption between 20 and 60 grams daily is of uncertain relevance
Chapter 55 NASH
at the light microscopy level yet it is typically excluded as a cause of NASH because the underlying disorder is better understood. As such, Wilson’s disease may be an example of a transition to identifying steatohepatitis in the setting of specific underlying abnormalities rather than identifying steatohepatitis in the absence of alcohol abuse as a disease. The phospholipidoses are liver disorders characterized by the accumulation of amphipathic molecules within lysosomal membranes leading to marked redundancy of the membranes into “whorl” patterns that can be visualized by electron microscopy. Although many drugs can cause phospholipidosis, the disorder only develops with chronic use of the implicated drug. Amiodarone is one of the most commonly used drugs that can cause phospholipidosis in addition to a NASH-like pattern of injury in some patients. While the phospholipidosis is pathogenetically distinct from NAFLD, the appearance by light microscopy can be similar to NASH.
EPIDEMIOLOGY RISK FACTORS FOR NAFLD When any new disease process is first recognized, one of the first steps to a better understanding is to describe the common features shared by patients with the disease. When the pathogenesis of the disease is understood, then the full spectrum of abnormalities associated with the underlying etiologic abnormality can be fully appreciated. Thus it happens that the “classic” findings of a disease process may ultimately be recognized to occur in but a minority of patients with the disease. Early series of patients fitting the criteria for NASH were described as being obese, diabetic, and female. In retrospect, the impact of referral bias on the characterization of NASH patients is now appreciated and the current view has expanded to include patients who are lean, non-diabetic, and male.11 The most common underlying risk factor of the development of NASH is the presence of insulin resistance.12–20 With this understanding, many of the previously identified risk factors for NASH are recognized as causes or other consequences of insulin resistance (Table 55-2). Obesity, especially centripetal obesity, and sedentary lifestyle are two of the largest risk factors for insulin resistance.21,22 Genetic predisposition also plays a major role, explaining why even lean people can develop type 2 diabetes, a late complication of insulin resistance, with aging. The full spectrum of liver abnormalities and their prevalence in patients with insulin resistance have not been defined and remain an important research area. One serious challenge to acquiring these data is the difficulty in defining what is meant by the term insulin resistance. If insulin resistance is defined as sufficiently impaired insulin signaling to cause a phenotypic change such as fatty liver, then the logic applied to defining each becomes slightly circular. Nonetheless, some have suggested that the presence of fatty liver is an early and sensitive indicator of insulin resistance.23 Female gender is no longer considered a risk factor for NASH. Early series that identified female gender as a risk factor may have suffered from referral bias and the confounding effect of obesity. When these factors are taken into account, NASH is found to occur equally in both genders.
Table 55-2. Conditions Associated with Non-Alcoholic Fatty Liver Disease (NAFLD) Insulin resistance Obesity Sedentary lifestyle Type 2 diabetes Hypertriglyceridemia Hypertension Drugs Tamoxifen Corticosteroids Amiodarone Estrogens Calcium-channel blockers (case reports, true association uncertain) Toxins Extensive exposure to volatile hydrocarbons278 Dietary abnormalities Carbohydrate excess (e.g., dietary, total parenteral nutrition) Protein deficiency Rapid weight loss Vitamin B12 deficiency Choline deficiency (?) Altered small-bowel anatomy Obesity surgery with blind loop of small bowel Small-bowel diverticula Short gut Metabolic diseases (resulting in NASH-like histology) Hypobetalipoproteinemia Abetalipoproteinemia Wilson’s disease Lipodystrophies Andersen’s disease Weber–Christian syndrome Mauriac syndrome Infections Chronic hepatitis C (usually genotype 3) AIDS Bacillus cereus infection? NASH, non-alcoholic steatohepatitis; AIDS, acquired immunodeficiency syndrome.
Hyperlipidemia, typically hypertriglyceridemia, is associated with NAFLD. In most circumstances, this association is explained by the hypertriglyceridemia associated with insulin resistance. Other factors, such as specific drugs that may have a more direct etiological role in the development of NAFLD, are discussed further below in the discussion of pathogenesis.
PREVALENCE OF NAFLD IN ADULTS Firm data describing the true prevalence of NAFLD and NASH in adults are not available. Case series are compromised by factors such as variable definitions of NAFLD and NASH, referral bias, case selection bias, and problems with excluding other diseases such as chronic hepatitis C in earlier studies. Postmortem series offer insight into the true prevalence but few such studies have been reported. In a series of 351 patients who died as inpatients and who underwent autopsy, steatohepatitis was identified in 2.7% of lean patients and 18.5% of markedly obese patients. Type 2 diabetes, rapid weight loss before death, and preterminal intravenous glucose were identified as risk factors in this select group.12 An early series found fatty liver in 24% of individuals who died in traffic accidents, although
1033
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
this may also have been a group at particular risk for having had recent alcohol ingestion.24 The best estimates based on currently available data indicate that about 20% of adults in the USA have NAFLD and about 2–3% of all adults have NASH (Table 55-3).22,25 These surprisingly high figures have been viewed with an element of skepticism because such a disease burden continues to be underrecognized by most care providers. Multiple factors contribute to the underdiagnosis of NAFLD, highlighting the importance of continued education of the medical community (Table 55-4).
PREVALENCE OF NAFLD IN CHILDREN The full spectrum of NAFLD is found in children.26 Whereas adults with NAFLD may have a normal body weight, children are almost invariably overweight or obese and typically have elevated fasting serum triglycerides. The prevalence of NAFLD in children is unknown, but appears be growing in parallel with the rise in obesity. One study in Japan of 810 children aged 4–12 years old demonstrated the presence of sonographically detectable NAFLD in 2.6% and its presence correlated with obesity.27
PROGRESSION OF DISEASE It is tempting and perhaps deceptively intuitive to think that, in some people, simple fatty liver progresses to steatohepatitis and then to fibrosis and cirrhosis. However, an equally plausible alternate hypothesis is that individuals prone to develop necroinflammatory injury do so as the fat accumulates (Figure 55-2). In fact, the limited long-term follow-up studies support the latter paradigm more than the former. The study reported by Matteoni et al. found that most people with just fat and no steatohepatitis continued to
Table 55-4. Factors Contributing to the Underestimation of NonAlcoholic Fatty Liver Disease (NAFLD) and Non-Alcoholic Steatohepatitis (NASH) by Health Care Providers
• • • • • • •
Table 55-3. Prevalence of Non-Alcoholic Fatty Liver Disease (NAFLD) and Non-Alcoholic Steatohepatitis (NASH) in Adults22
•
NAFLD (including NASH) NASH NASH with fibrosis
•
Two thirds of adults are overweight or obese; 70% are sedentary
20% 3–4% 1% (?)
•
Few or no symptoms associated with NAFLD and NASH, even with aggressive disease25 Unrecognized presence of cirrhosis caused by NASH10 Failure to realize that NAFLD and NASH are not rare diseases in patients who are not female, obese, and diabetic Normal or near-normal aminotransferases in some patients despite aggressive disease150 Rising upper reference range of aminotransferase values unrelated to test characteristics153 Absence of reliable laboratory tests and non-invasive diagnostic techniques Discounting mild aminotransferase elevations as trivial without further investigation Discounting incidentally discovered NAFLD evident on imaging studies as trivial without further investigation Lack of effective therapy to spur a more aggressive diagnostic approach Likely increasing prevalence in parallel with the increase in obesity222
Obesity Sedentary lifestyle Genetics Drugs Other
Insulin resistance
20% of adults have NAFLD; 3–4% have NASH
Type 2 diabetes Hypertension Hyperlipidemia Vascular disease etc.
Steatosis with necroinflammatory Arterial changes (NASH) vasoconstriction
?
Steatosis alone
Progressive fibrosis in some patients
NASH and cryptogenic cirrhosis account for about 12% of liver transplants
Cirrhosis ("cryptogenic" in many)
HCC
1034
Liver transplant
Death
Figure 55-2. Progression of non-alcoholic fatty liver disease (NAFLD). The most common underlying risk factor for NAFLD is the presence of insulin resistance. Obesity, sedentary lifestyle, and genetic factors are major causes of insulin resistance, whereas diabetes, hypertension, hyperlipidemia, and vascular disease are other common consequences of insulin resistance besides NAFLD. A fraction of patients with NAFLD develop necroinflammatory changes and meet current criteria for non-alcoholic steatohepatitis (NASH). It is primarily these patients who are at risk for developing progressive fibrosis, cirrhosis, and hepatocellular carcinoma (HCC). How many patients with NASH progress to fibrosis and then to cirrhosis is uncertain. The prevalence of cirrhosis caused by NASH may be underestimated because it can be unrecognized yet still complicate other comorbidities such as diabetes, vascular disease, and complications of obesity.
Chapter 55 NASH
have just fat over time and typically did not progress to NASH and its sequelae.5 Knowing how many individuals with NASH progress to cirrhosis has proven difficult because the population cannot be screened economically and non-invasively.28 Serum aminotransferases lack sufficient sensitivity and specificity to serve as a surrogate for defining the true prevalence of NASH. Based on studies described above, 10–15% of people with NAFLD will have NASH and a subset – perhaps one-third – of those will be at risk of developing cirrhosis.29 The factors that determine whether a patient with NAFLD also develops necroinflammatory changes and fibrosis are not known. Possibilities include genetics, dietary composition, and concomitant forms of other liver disease (e.g., chronic hepatitis C). There may be important racial and ethnic predispositions, but these remain poorly characterized at this time.30,31 The clinical parameters that may predict the presence of progressive disease are discussed further below (Tables 55-5 and 55-6).
LIVER TRANSPLANTATION, HEPATOCELLULAR CARCINOMA, AND DEATH FROM LIVER FAILURE If NASH only caused biopsy abnormalities without clinically significant sequelae, the causes and consequences of fatty liver disease would be of only minor interest. However, with increasing recognition of this disorder has emerged a parallel recognition of its role in causing liver failure and cancer. Historically, about 2% of liver transplants were performed for a known diagnosis of NASH. The true fraction of the 4000–5000 liver transplants performed annually in the USA due to NASH is likely severalfold greater if cryptogenic
cirrhosis is included. About 10% of liver transplants are performed for cryptogenic cirrhosis and epidemiological data (discussed below) have implicated prior NASH as the most likely causative factor in most.32,33 It is unknown how many patients thought to be dying of end-stage diabetes or other terminal complications of insulin resistance also have occult cirrhosis that contributes to their death.
PATHOGENESIS OF NAFLD AND NASH ACCUMULATION OF TRIGLYCERIDE IN HEPATOCYTES By definition, accumulation of excess hepatocellular triglyceride is necessary for the development of NAFLD. While this is obvious in the case of NAFLD without concomitant steatohepatitis, we can only infer that the presence of fat is necessary for the development of the ballooning, inflammation, fibrosis and other necroinflammatory lesions of NASH because these cytological features of liver injury are not seen in the absence of fat in the non-alcoholic. The major mechanisms of triglyceride accumulation in hepatocytes were well described by Lombardi 40 years ago (Figure 55-3). Fat is delivered to hepatocytes in the form of free fatty acids bound to albumin or it is synthesized within hepatocytes de novo. Free fatty acids delivered to the liver originate primarily from adipose tissue through the action of hormone-sensitive lipase and other lipolytic enzymes. A small fraction of circulating free fatty acids derives from the short-chain fatty acids absorbed directly from the
Table 55-5. Predictors of Non-Alcoholic Steatohepatitis (NASH) versus Simple Steatosis Klain et al.279
Wanless and Lentz12
Silverman et al.280
Dixon et al.15 “HAIR”
Cohort
Bariatric surgery
Autopsy of hospitalized patients
Bariatric surgery; mean BMI 49 kg/m2
Bariatric surgery; BMI > 35 kg/m2
Age
Yes
Gender
Female
No
Male
BMI
Yes
Yes
No
No
Centripetal obesity (WHR)
Yes
Type 2 diabetes mellitus
Yes
Yes
Yes
HbA1c
Yes
Insulin resistance
Yes
ALT elevation
Yes
Yes
AST/ALT ratio
No
GGT
Yes
Hypertension
Yes
Triglycerides
Yes
Ferritin
Yes Yes
BMI, body mass index; WHR, waist-to-hip ratio; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, g-glutamyltransferase. Independent risk factors identified by multivariate analysis are noted in bold type.
1035
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Table 55-6. Predictors of Fibrosis in Patients with Non-Alcoholic Steatohepatitis (NASH)
Cohort
Wanless and Lentz12
Silverman et al.281
Marceau et al.130
Angulo et al.177
Ratziu et al.282 “BAAT”
GarcíaMonzón et al.150
Dixon et al.15
Crespo et al.283
Chitturi et al.109
Autopsy of hospitalized patients
Bariatric surgery; mean BMI 49 kg/m2
Bariatric surgery; mean BMI 47 kg/m2
Biopsied to evaluate liver disease
Abnormal LFTs and BMI >25 kg/m2
BMI >40 kg/m2
BMI >35 kg/m2
Bariatric surgery; mean BMI 47 kg/m2
Abnormal LFTs
Yes
>45
≥50
Yes
Yes
Yes
Yes
Yes
≥30 kg/m
Age Gender BMI
Male Yes
Centripetal obesity Type 2 diabetes mellitus
Female 2
≥28 kg/m
Male
Harrison et al.284
Female ≥28 kg/m2
2
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes >5.2%
HbA1c Insulin resistance
Yes
C-peptide
Yes
ALT elevation
Yes
AST/ALT ratio
>1
2 ¥ ULN
Yes >0.8
Hypertension
Yes
Triglycerides
>1.7 mmol/l
Ferritin
No
No
BMI, Body mass index; LFT, liver function test; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ULN, upper limit of normal. Independent risk factors identified by multivariate analysis noted in bold type.
small bowel during the absorptive phase of digestion. The liver also synthesizes free fatty acids, a metabolic step that is the major fate of excess circulating glucose not needed as an immediate energy source. The caveat is that a low-fat diet may do little to decrease the amount of fat delivered directly to the liver, yet a diet characterized by excessive carbohydrates may rapidly increase the accumulation of fat in the liver through de novo synthesis. A minor source of triglyceride accumulation in the liver is the hepatocellular endocytosis of lipoprotein remnants, specifically lowdensity lipoprotein (LDL) and chylomicron remnants. These lipoproteins originate from very-low-density lipoprotein (VLDL) and chylomicrons secreted into the circulation by the liver and gut respectively, and the remnants containing a small amount of triglyceride are taken up by the liver once most of their triglyceride content has been removed by muscle and adipose tissue. Free fatty acids in the liver have two major fates (Figure 55-3). They can be shuttled into mitochondria by the carnitine cycle where they serve as a source of energy (adenosine triphosphate: ATP) or they can be esterified into triglyceride, packaged as VLDL, and secreted by exocytosis into the circulation. A relatively minor pathway in terms of its overall consumption of free fatty acids is their oxidation in peroxisomes. Peroxisomes are small hepatocellular organelles that metabolize very-long-chain fatty acids and dicarboxylic acids by oxidation.
1036
Microvesicular Steatosis Hepatocellular triglyceride accumulation in NAFLD is typically macrovesicular or a mix of microvesicular and macrovesicular steatosis. Microvesicular steatosis is defined as lipid droplets in hepatocytes less than 1 μm in diameter, or roughly less than the diameter of hepatocyte nuclei. Pure microvesicular steatosis can sometimes be difficult to identify without specific fat stains such as the oil-red O. It is often associated with life-threatening acute liver dysfunction and is almost uniformly associated with disorders of mitochondrial dysfunction, such as drug toxicity,34 acute fatty liver of pregnancy, Reye’s syndrome, or toxins such as methylenecyclopropylalanine (MCPA, hypoglycin A) in unripe ackee fruit (Blighia sapida). Eating unripe ackee is responsible for Jamaican vomiting sickness and death from liver failure. Why triglyceride accumulates as small droplets in some disorders and large droplets that displace cell contents peripherally in other disorders such as NAFLD is not known. The rate of fat accumulation may be one explanation. In most cases of microvesicular steatosis, the fat accumulates over a relatively short period, whereas macrovesicular steatosis is commonly associated with more prolonged metabolic disturbances. Inconsistent with this explanation is the development of macrovesicular steatosis very early in the course of alcohol ingestion, as shown in an early study by Rubin and Lieber.35 In general, disorders of purely microvesicular fat accumu-
Chapter 55 NASH
Table 55-7. Multiple Components of Very-Low-Density Lipoprotein (VLDL)
VLDL
Triglyceride Peroxisomal oxidation
Mitochondrial -oxidation Free fatty acids
Lipoprotein remnants Free fatty acids bound to albumin
Excess glucose
Lipolysis in adipose tissue
Figure 55-3. Uptake and disposal of fat by the liver. Two major sources contribute to fat in the liver: delivery from peripheral adipose tissue as free fatty acids bound to albumin in the circulation or de novo synthesis within the liver from excess carbohydrate. Uptake of lipoprotein remnants delivers triglyceride to the liver but probably accounts for a relatively minor contribution. Free fatty acids in the liver have two major fates: mitochondrial b-oxidation or esterification to triglyceride and export as very-low-density lipoprotein (VLDL). Peroxisomal oxidation has a quantitatively minor role in the disposition of fat, although it probably plays an important metabolic role. Fat accumulates as triglyceride to present as non-alcoholic fatty liver disease when the delivery or de novo synthesis exceeds oxidation or secretion.
lation are not classified as NAFLD or NASH, even with associated necroinflammatory changes.
VLDL Synthesis and Secretion (Figure 55-3) In the postprandial state, ingested triglycerides are hydrolyzed to free fatty acids in the lumen of the small bowel (and a small amount is hydrolyzed in the mouth by lingual lipase), and the released free fatty acids are absorbed by enterocytes. Only short-chain fatty acids are then delivered from enterocytes directly into the blood, whereas most are esterified back into triglycerides within enterocytes and secreted into the lymphatics as chylomicrons. Adipose tissue and muscle remove free fatty acids from chylomicrons and store them as triglyceride or use them immediately as a source of energy. Also in the postprandial state, the liver converts unneeded carbohydrates to free fatty acids that are esterified and secreted into the blood as VLDL. The synthesis and secretion of VLDL require multiple intact metabolic pathways. Seemingly disparate dietary deficiencies and metabolic abnormalities comprising any one of these pathways can cause the accumulation of fat in the liver, i.e., NAFLD (Table 55-7). Few data are available on rates of VLDL production in NASH, although one study suggested synthesis is impaired.36 Impaired secretion of triglyceride from hepatocytes is likely the most common metabolic defect contributing to the development of NAFLD and thus warrants particular attention.37 VLDL synthesis requires the protein apoB100, a 550-kDa protein translated from a liver-specific splice form of the APOB transcription product. An
Essential for VLDL formation
Function
Functional apolipoprotein B100 gene Minor apolipoproteins Amino acid availability
Key component285,286
Protein synthesis Functional microsomal triglyceride transfer protein Choline/betaine/SAMe Essential fatty acids Vesicle trafficking/intact cytoskeleton
Key component of VLDL in circulation38 Necessary for lipoprotein synthesis (e.g., NAFLD associated with kwashiorkor) Necessary for lipoprotein synthesis Necessary for apoB100 lipidation38,40,132
Methyl donors necessary to form phosphatidyl choline (lecithin)255 Arachidonic acid necessary to form phosphatidyl choline (lecithin)255 Necessary for secretion from hepatocytes
NAFLD, non-alcoholic fatty liver disease; SAMe, S-adenosylmethionine.
alternate splice form of the same gene is used by enterocytes to form chylomicrons. Apolipoprotein E (apoE) is also an important component of circulating VLDL and polymorphisms of apoE such as apoE3-Leiden are associated with hepatic steatosis in mice.38 Whether apoE is incorporated in VLDL in hepatocytes or in the circulation has not been firmly established. In the liver, newly formed apoB co-translationally inserts into the endoplasmic reticulum where it undergoes essential disulfide bond formation and lipidation by triglyceride. Two proteins, protein disulfide isomerase and microsomal triglyceride transfer protein (MTTP) are needed for lipidation.39 Impaired MTTP expression is a cause of abetalipoproteinemia, a condition associated with NASH and cirrhosis. Less severe defects cause hypobetalipoproteinemia and have been linked to an increased risk of hepatic steatosis in patients with diabetes40 and increased liver injury after exposure to endotoxin.41 Whereas abetalipoproteinemia presents in infancy with severe fat malabsorption and early death, hypobetalipoproteinemia is found in 1/500 to 1/1000 people and is associated with NAFLD in adults. A clinical clue to the presence of this rare abnormality is an unexpectedly low serum cholesterol level (below 150 mg/dl).42,43
Mitochondrial b-Oxidation
Fat that is not secreted from hepatocytes as VLDL undergoes b-oxidation, either in mitochondria or the peroxisomes. Mitochondrial boxidation converts the energy stored in fat to the metabolically usable forms of nicotinamide-adenine dinucleotide (NADH) and ATP. It is also the source of the ketone bodies acetoacetate, acetone, and D-3-hydroxybutyrate. These are essential metabolic fuel sources for peripheral tissues, especially neurons, muscle, and brain when glucose is in short supply because of the inability of these tissues to use free fatty acids as an energy source. A number of drugs, toxins, and genetic abnormalities of mitochondrial function have been described, often in association with NALFD.34,44 Alcohol causes significant disruption of mitochondrial function and thus may share this mechanism of injury with NAFLD. Morphological abnormalities in mitochondria, characterized as crystalline inclusions, have also been observed in patients with NASH,
1037
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Peroxisomal Oxidation Peroxisomal b-oxidation of fatty acids may have a significant role in the pathogenesis of NASH. Peroxisomes metabolize long-chain free fatty acids (22 carbons and greater) to shorter fatty acids that can be further metabolized by mitochondrial b-oxidation.48 About 10% of short- and medium-chain free fatty acids are also metabolized by peroxisomes. The enzymes of the peroxisomal compartment are induced by the fibrates and a high-fat diet.49 An obligate product of peroxisomal b-oxidation is the production of hydrogen peroxide. Although peroxisomes are well endowed with the enzyme catalase to facilitate the destruction of this reactive oxygen molecule, peroxisomal metabolism of fatty acids remains a potential source of oxidant stress. While this might suggest that increased activity of the peroxisomal pathway could participate in the development of NASH, the data are conflicting. The peroxisomal knockout mouse develops morphological features of steatohepatitis50–52 but upregulation of peroxisomal b-oxidation with a peroxisome proliferator-activated receptor-a (PPARa) ligand prevents fat accumulation and NASH in the methionine- and choline-deficient diet mouse model53 and a mouse model of lipoatrophy.54
Insulin Resistance, Hyperinsulinemia, and Hepatocellular Triglyceride Accumulation In the fed state, circulating insulin levels rise and signal muscle and adipose tissue to take up glucose from the blood for metabolism or storage (as glycogen). Insulin also signals adipose tissue to stop the production of free fatty acids from triglyceride and signals the liver to stop the de novo synthesis of glucose needed for survival during fasting. Down-regulating both of these metabolic steps in the fed state is logical since circulating free fatty acids are not needed by muscle and the liver as a source of energy and blood levels of glucose are sufficient such that the liver does not need to make glucose to keep the central nervous system functioning. When the responses of muscle, adipose tissue, and the liver to insulin are impaired, blood glucose levels rise and insulin secretion by the pancreas increases to compensate and keep glucose levels within a narrow range. Such an abnormal metabolic state is called insulin resistance with compensatory hyperinsulinemia. When the compensatory hyperinsulinemia becomes insufficient to facilitate normal glucose disposal, glucose levels rise and diabetes results. At some point, after years to decades of insulin overproduction, the pancreatic beta-cell population begins to diminish by attrition in response to the excessive demand, and diabetes progresses to the point where even fasting blood glucose levels become elevated. Most patients with NAFLD and NASH are now recognized as having underlying insulin resistance (Figure 55-4).14,17,18,55–57 The role of insulin resistance and its compensatory hyperinsulinemia in causing fatty liver is demonstrated by epidemiological studies, anatomical observations, and well-described biochemical pathways.23,57–60 If fat accumulation in the liver is a common metabolic
1038
6 P= 0.007 5 Insulin (ng/ml)
although their functional significance is unknown.16,45,46 Although assessment of mitochondrial respiratory chain components shows fairly global impairment in NASH patients,47 the overall role of mitochondrial dysfunction in the pathogenesis of NASH has not been determined.
4 3 2 1 NASH
Control
0 0
30
60
90
120
Minutes Figure 55-4. Fasting and postprandial hyperinsulinemia in non-diabetic patients with non-alcoholic steatohepatitis (NASH). Eighteen subjects with NASH and 18 controls matched for body mass index, age, and gender were given a 500-calorie meal; fasting insulin levels were elevated in the NASH group and the homeostasis model assessment method for insulin resistance (HOMAIR) was 6.5 ± 5.0 in the NASH group compared to 3.2 ± 3.0 in controls (P < 0.001). (Reproduced from Chalasani N, Deeg MA, Persohn S, Crabb DW. Metabolic and anthropometric evaluation of insulin resistance in nondiabetic patients with nonalcoholic steatohepatitis. Am J Gastro 2003; 98:1849–1855, with permission.)
consequence of hyperinsulinemia, then it is not surprising that emerging evidence suggests that NAFLD may be a very early indicator of insulin resistance.23 Insulin resistance is also likely the major underlying disorder in children with NAFLD.61 As is the case in adults, obesity is the primary cause of insulin resistance in children,62,63 especially centripetal adiposity.64 Epidemiological evidence has also shown that alanine aminotransferase (ALT) elevations, a surrogate for NAFLD, correlate strongly with insulin resistance in children.65 The complexities of obesity in children are underscored by the suggestion that intrauterine imprinting occurs, in which birth weight and being underweight in the first 2 years predispose to later weight gain and diabetes.66,67
MECHANISMS OF INSULIN RESISTANCE Postreceptor Signal Transduction and Insulin Resistance Understanding the common mechanisms of insulin resistance is essential to identify potentially effective treatments for most patients with NAFLD. Abnormalities of the insulin receptor have been associated with severe insulin but are quite rare. Most patients with insulin resistance have acquired defects of postreceptor intracellular signaling. The insulin receptor is a tyrosine kinase that autophosphorylates itself in response to insulin binding. The activated receptor then phosphorylates other intracellular signaling mediators, the most prevalent of which are the insulin receptor substrates 1 and 2 (IRS-1 and IRS-2, Figure 55-5). The net outcome is increased glucose disposal, decreased gluconeogenesis, increased lipogenesis, decreased lipolysis, and stimulation of cell growth.
Chapter 55 NASH
Insulin binding to insulin receptor
TNF␣
IRS-1
Free fatty acids
PI3 kinase
IRS-2
Normal effects of insulin: Increased glucose disposal Increased lipogenesis Decreased gluconeogenesis Decreased lipolysis Trophic effects
Figure 55-5. Intracellular signaling by insulin. Tumor necrosis factor-a (TNF-a) and high levels of intracellular free fatty acids both impair signal transduction after insulin binds its receptor, especially in target tissues such as muscle and adipose tissue. The result is a loss of the downstream effects of insulin with impaired glucose disposal and a failure to down-regulate hepatic gluconeogenesis and peripheral lipolysis. These effects of TNF-a and free fatty acids may be mediated by increased activation of the intracellular kinase JNK1.
Other
Impaired activation of phosphatidyl inositol-3 kinase (PI-3 kinase) by IRS-1 has been identified as a common site for impaired insulin signaling. Tumor necrosis factor-a (TNF-a) and increased free fatty acids both impair this pathway.68–70 In response, glucose disposal is impaired and serum glucose levels rise, which signals the pancreas to increase insulin output. A state of relative hyperinsulinemia thus follows. If the compensatory increases in insulin levels are sufficient to bring glucose levels under control, patients simply have insulin resistance without diabetes. When the elevated insulin levels are inadequate to achieve normal glucose disposal, then impaired glucose tolerance and diabetes develop. The relative roles played by the liver, muscle, and adipose tissue in modulating insulin sensitivity and predisposing to NAFLD have not been clarified. Data obtained primarily in animal studies have also shown that signaling defects in the liver play an important role in the development of insulin resistance.71–74 For example, liverspecific knockout of the insulin receptor causes severe insulin resistance in mice, manifested by glucose intolerance and hyperinsulinemia.75 On the other hand, knockout of adipose tissue glucose transporter GLUT4 also causes insulin resistance.76
TNF-a and Free Fatty Acids as Common Mediators of Insulin Resistance The most common mediators of impaired insulin signaling are elevated intracellular levels of free fatty acids and increased circulating levels of the cytokine TNF-a (Figure 55-5). A key role of TNF-a in mediating insulin resistance has been demonstrated in TNF-a knockout mice which fail to develop insulin resistance when fed to the point of obesity70 or in mice treated with antibodies to TNFa.77 Excessive intracellular free fatty acids stimulate the serinethreonine kinase JNK-1, which, through serine phosphorylation of key members of the insulin signal transduction pathway such as IRS1, prevents normal insulin signaling.78 TNF-a probably also impairs insulin signaling by the same mechanism of JNK1 activation, placing JNK1 activation in a central role in the development of insulin resistance. Further evidence is provided by the JNK1 knockout mouse that also fails to develop obesity-induced insulin resistance.78 Therefore, modulating JNK1 activity is being examined as a treatment option for diabetes and may be a reasonable focus for future studies in NASH.
Liver Disease as a Cause of Insulin Resistance The preponderance of data indicates that the liver is the target of too much insulin and suffers the consequences of hyperinsulinemia. However, the presence of liver disease can also contribute to the underlying state of insulin resistance. For example, cirrhosis from any cause is a risk factor for the development of insulin resistance.79,80 Additionally, studies in rats with NAFLD caused by tetracycline develop insulin resistance. Nonetheless, the observations that insulin resistance and hyperinsulinemia appear well before the development of significant liver dysfunction in patients with NASH suggest that the insulin resistance comes first and NAFLD follows as a consequence in most patients.
Iron and Insulin Resistance The relationship between hepatic iron stores and insulin resistance is complex and poorly understood. Current data indicate that the relationship may be bidirectional: excess iron stores can contribute to insulin resistance81,82 and yet insulin resistance appears to increase iron accumulation.83 The mechanisms underlying this mutually synergistic effect are the subject of ongoing investigation. Perhaps both components of this positive feedback mechanism may not exist commonly in the same patients or much more iron accumulation associated with insulin resistance would be expected.84–86
Lipodystrophies and Insulin Resistance Adipose tissue in the right place and in the right amount may be desirable because adipocytes can serve as a depot for fat, thus sparing other tissues from inappropriate fat deposition. Disorders of peripheral adipose tissue development and survival, the lipodystrophies, are defined by partial or complete inability to form adipose tissue. NAFLD and cirrhosis are known sequelae of these disorders, with the degree of hepatic steatosis being proportional to the extent of adipose tissue loss.87 For example, congenital generalized lipodystrophy is a rare disorder characterized by nearly absent peripheral fat, severe hepatic steatosis, and a significant risk of cirrhosis.88 Mutations associated with partial lipodystrophies include abnormalities of gene encoding PPAR-g, PPARG,89 and the nuclear envelope protein lamin A, LMNA.90 Lamin A may participate in regulating
1039
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Increased hepatocellular triglyceride content
Mitochondrial dysfunction
Substrate for lipid peroxidation
Cellular ATP depletion
Increased gut-derived endotoxin
Increased circulating free fatty acids
Cellular membrane dysfunction
Figure 55-6. Mechanisms of hepatocellular injury in non-alchoholic fatty liver disease (NAFLD). Several mechanisms of hepatocellular injury have been proposed to explain the necroinflammatory changes characteristic of non-alcoholic steatohepatitis (NASH). The mechanisms are not mutually exclusive and probably each contributes to the development of NASH in varying degrees.
Oxidant stress
Hepatocellular necrosis, apoptosis, release of inflammatory cytokines
sterol regulatory element-binding proteins 1 and 2 (SREBP-1 and -2). These transcription factors modulate insulin signaling and lipid synthesis in hepatocytes. In animal studies, overexpression of SREBP-1 is associated with severe hepatic steatosis91 whereas loss of SREBP-1 makes mice resistant to the development of fatty liver.92
Hepatocellular Injury in the Triglyceride-Loaded Hepatocyte Much more is known about how triglyceride accumulates in hepatocytes than why fat-laden hepatocytes are prone to injury. Oxidant stress in the setting of ample substrate for lipid peroxidation has been proposed as one link between the accumulation of fat and subsequent injury.93 Alternative explanations of cellular injury in NASH include toxicity of free fatty acids and impaired mitochondrial ATP production (Figure 55-6). These mechanisms are not mutually exclusive and a combination of these or yet to be discovered additional mechanisms of injury is probably important. One challenging idea proposed in the 1970s is that the underlying abnormality in patients with NASH causes both the injury and the fat to accumulate such that the presence of fat is simply a marker of the disease process.94
Oxidant Stress and Lipid Peroxidation Oxidant stress is defined as the production of reactive oxygen species in excess of what can be handled by endogenous antioxidant mechanisms leading to altered cellular physiology. Day and James proposed a “two-hit” hypothesis for the pathogenesis of NASH that states that the accumulation of fat is the primary abnormality and secondary oxidant stress leads to the inflammation, injury, and fibrogenesis that characterize NASH.95 The evidence that oxidant stress plays a role in the pathogenesis of NASH is largely derived from animal studies. For example, ferrets with diet-induced steatosis have higher liver markers of oxidant stress than control animals without steatosis 96 An animal model that recapitulates many of the abnormalities of NASH, the methionine- and choline-deficient diet, also causes increased oxidant stress in the liver.97 Rats fed this diet develop steatosis in 2 weeks, necroinflammatory changes in 5 weeks,
1040
and fibrosis in 12–17 weeks.98 Additionally, they had lower levels of the endogenous antioxidant glutathione, suggesting demands on defense mechanisms. Other studies in animals have also found upregulated antioxidant pathways, suggesting that the liver is responding to oxidant stress. Despite these findings in animal models, limited data in patients with NASH have confirmed the role of oxidant stress in NASH. While some studies confirm increased markers of oxidant stress,99 few have shown that antioxidants prevent NASH (see Treatment, below).
Possible Sources of Oxidant Stress in NAFLD Multiple sources of oxidant stress in the fatty liver have been identified and include cytochrome P450,100 peroxisomal b-oxidation, mitochondrial electron leak, and recruited inflammatory cells. The presence of excess fat in the liver provides ample substrate for lipid peroxidation and reactive lipid peroxidation products such as 4hydroxynonenal can further amplify the oxidant stress that led to their initial formation.93 Electron transport initiated by cytochrome P450 enzymes can cause oxidant stress101 and these enzymes are up-regulated in patients with NASH102,103 and in animal models of steatohepatitis.104 Definitive proof of the role of one key P450 isoenzyme, CYP2E1, was sought by feeding CYP2E1 knockout mice the methionine and choline-deficient diet. Interestingly, mice with this model of steatohepatitis diet still developed steatohepatitis, indicating that 2E1 is not necessary for liver injury in this model. One explanation is that another isoenzyme, CYP4A, was up-regulated and in vitro inhibition with an antibody to this enzyme decreased oxidant stress.97,100 Increased flux of free fatty acids through the liver may be the major stimulus for the induction of P450 activity in NASH patients. The reversibility of this process has been shown by the downregulation of CYP2E1 associated with weight loss following bariatric surgery.105 Finally, the peroxisomal b-oxidation pathway has been examined as a source of oxidant stress. If this metabolic pathway is responsible for injury, then its down-regulation should be beneficial. Paradoxically, the pathway is up-regulated by PPARa ligands
Chapter 55 NASH
(fibrates) and these agents may be beneficial in NASH (see Treatment, below). Moreover, the peroxisomal knockout mouse is not protected against the development of steatohepatitis,50,51 suggesting that PPARa-mediated disposal of fat may be an important fate of fat.53
Iron as a Contributor to Oxidant Stress and Insulin Resistance Excessive hepatic iron accumulation plays an important role in causing oxidative stress and chronic liver injury in patients with genetic hemochromatosis. Lesser degrees of iron accumulation are commonly present in other forms of liver disease, especially alcoholic liver disease and NAFLD. Whether this iron contributes to the liver disease and its likelihood to progress to cirrhosis and hepatocellular carcinoma (HCC) is less certain.106,107 If excess iron is important in the pathogenesis of NASH, then abnormalities of the HFE gene associated with the iron overload of classical hemochromatosis might be overrepresented in patients with NASH. One study in Australia of 51 subjects with NASH found that 31% had an abnormal HFE genotype compared to a normal prevalence of 13% in the general population.108 In this study, HFE mutations were associated with increased iron staining on biopsy, increased transferrin saturation, and increased fibrosis. In contrast, another study conducted in Australia found that HFE mutations and stainable iron were did not correlate with fibrosis.109 Detecting excess hepatic iron in patients with NAFLD generally requires a liver biopsy with an iron stain or even iron quantitation. Serum ferritin is unreliable because it can be elevated by the lowgrade necroinflammatory injury in the liver. In patients with NASH, serum ferritin did not correlate with either biopsy-detectable iron or abnormal HFE genotypes.108 Some studies have suggested that ferritin correlates better with insulin resistance and the presence of NAFLD rather than true iron overload.110 In summary, the two-hit theory of the pathogenesis of NASH is appealing but finding evidence that confirms a causal role of oxidant stress in the pathogenesis of liver injury in humans with NASH has been difficult. Markers of oxidant stress are increased, in both the liver and the serum. However, treatment trials of various antioxidants have generally been disappointing. This could be explained by a lack of efficacy of currently used antioxidants or that oxidant stress is not critical to the pathogenesis of NASH. Hopefully, continued trials with antioxidants and further work into the pathogenesis of NASH will provide clarity and guidance for therapy.
Free Fatty Acid Toxicity The increased flux of free fatty acids through the liver in states of increased peripheral lipolysis not only promotes insulin resistance,12 but could play a direct role in hepatocellular injury. One of the difficulties in examining the role of free fatty acids in hepatocellular injury is the lack of reliable methods to measure their intracellular levels. However, if free fatty acids are agents of cellular injury in the pathogenesis of NAFLD, abundant and overlapping protective mechanisms against this toxicity would be expected and could provide indirect evidence of their toxicity. Indeed, such mechanisms exist. Hepatocytes in particular are well endowed with mechanisms to bind, transform, catabolize, and export excess free fatty acids
through the combined actions of fatty acid-binding proteins, triglyceride synthesis, secretion as VLDL, mitochondrial b-oxidation, and enzymatic removal of lipid peroxidation products. One example is up-regulation of the nuclear receptor PPARa, which plays a central role in sensing excess free fatty acids and up-regulating the genetic program of fatty acid disposal.
Impaired Hepatocellular ATP Production Mitochondrial dysfunction in the liver may occur in NAFLD and contribute to the hepatocellular injury that leads to NASH.16,45,111–113 Hepatocellular mitochondria oxidize free fatty acids as an energy source and use this energy to generate ATP. Without normal mitochondrial function, hepatocytes become depleted of ATP and normal cellular functions become crippled, leading to cellular injury and death. Cells can become similarly impaired by oxygen deprivation and thus metabolic mitochondrial dysfunction shares mechanistic features with hypoxic injury. One provocative finding is that dietary constituents can increase demands on mitochondrial function. Fructose requires ATP to initiate its catabolism and an oral fructose load was found to cause more depletion of hepatic ATP in patients with NAFLD than normal.114,115 Whether the increased oral fructose consumption associated with the common use of high-fructose corn syrup in soft drinks and other foods contributes to liver injury in NAFLD is unknown. Some data also suggest that muscle mitochondrial dysfunction plays a significant role in the development of insulin resistance. A study of healthy but insulin-resistant children of diabetics demonstrated that their muscle is insulin-resistant and exhibits impaired ATP production.116 Although the functional significance of this is uncertain, insulin resistance increases with aging, in conjunction with a decrease in muscle mitochondrial function. Thus, progressive deterioration of muscle mitochondrial function probably occurs with aging and is associated with progressively worsening insulin resistance independent of obesity.117
Inflammation and Cytokines NASH, by definition, is an inflammatory condition. Understanding the roles of intrahepatic and extrahepatic cytokines and their respective receptors in mediating the inflammatory process is in its infancy.63 TNF-a is a potent proinflammatory cytokine that has been examined for a possible role in NAFLD and NASH. Several studies have measured circulating levels of TNF-a and found that they are increased in patients with NASH.118–120 Since visceral or mesenteric adipose tissue is a major source of TNF-a production and the liver is immediately downstream in its venous drainage, the liver may be a primary target of this cytokine. Because not all patients with NAFLD develop significant necroinflammatory changes in response to the hepatocellular fat accumulation, there may be genetic polymorphisms in proinflammatory cytokines and their receptors that predispose to unbridled inflammation in a subset of patients.121 Evidence also suggests that a TNFa receptor polymorphism is overexpressed in patients with NASH and a similar observation has been made in patients with alcoholic hepatitis. On the other hand, blocking the effect of TNF-a does not prevent NAFL in a diabetic mouse model, suggesting that TNF-a may not be central to the pathogenesis of NAFL.122
1041
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Endotoxin
AMIODARONE
One stimulus for an inflammatory state in NASH may be the abnormal production and absorption of bacterial endotoxin from the gut.93,119 This phenomenon has been the leading explanation for the severe and rapidly progressive fatty liver disease that developed in patients who underwent jejunoileal bypass several decades ago. Following this procedure, the surgically created blind loop served as a reservoir for bacterial overgrowth and the production of endotoxin. Even without surgically altered small-bowel anatomy, abnormalities such as small-bowel diverticula can be a source of bacterial overgrowth and may contribute to the development of NASH in selected patients. Animal models have provided additional evidence that gutderived endotoxin may play a role in fatty liver disease. The obese leptin-deficient ob/ob mouse develops fatty liver and the extent of liver injury, as measured by serum ALT levels and hepatic NFkB activation is modulated by gut-derived endotoxin.77,123
Amiodarone may be one of the most common causes of NASH induced by currently available drugs. Amiodarone use is associated with multiple histologic abnormalities; it causes phospholipidosis, or the accumulation of whorls of lysosomal membrane, that requires electron microscopy to identify and it can also cause the histological abnormalities typically associated with NASH. Specifically, macrovesicular and microvesicular steatosis, hepatocellular ballooning, Mallory’s hyaline, fibrosis, acidophil bodies and glycogen nuclei can be features of amiodarone hepatotoxicity. These abnormalities occur in 1–3% of patients treated with the drug and have been rarely associated with severe progressive liver disease.133
HEPATIC FIBROGENESIS IN NASH An imbalance of extracellular matrix production in the liver in response to chronic injury is probably the cause of cirrhosis in NASH as it is in all other forms of chronic liver disease. Why some individuals develop fibrosis and some do not has not been established, but roughly a third of patients with any form of chronic liver disease appear to develop fibrosis and are therefore at risk for progressing to cirrhosis. Genetic predisposition related to a variety of polymorphisms may play a key role.124,125 Stellate cells are the primary source of hepatic extracellular matrix production and these cells have been shown to be in an activated state in NASH.126,127 Especially relevant to NASH is the finding that insulin may stimulate activation of hepatic stellate cells to a more fibrogenic phenotype.128 A direct role for hyperinsulinemia that accompanies insulin resistance may also explain the increased risk of fibrosis in patients with diabetes.15,129,130
DRUGS THAT CAUSE NAFLD AND NASH Compared to insulin resistance, drugs are a relatively uncommon cause of NAFLD and NASH. A number of drugs have been associated with NASH (Table 55-2), many of which are identified in case reports. Such single cases or small case series of drug-induced NASH cannot establish with certainty that NASH was not a pre-existing disorder before the drug or if the drug in question exacerbated underlying insulin resistance, leading to the development of NASH by mechanisms common to most patients.131 Several mechanisms are important in drug-induced NAFLD and NASH. Drugs may increase the burden of free fatty acids delivered to the liver (e.g., corticosteroid-induced peripheral lipolysis). Perhaps more commonly, some drugs impair the removal of fat from the liver by impairing mitochondrial b-oxidation34 or compromising the normal formation and secretion of VLDL.132 Two drugs, amiodarone and tamoxifen, are particularly relevant to current clinical practice because of the difficult decisions required for optimal patient management if drug-induced NASH develops.
1042
TAMOXIFEN Tamoxifen has been clearly implicated as a cause of NAFLD and even NASH in several studies.134–136 In some patients there may be irresolvable issues as to whether tamoxifen caused NASH or the NASH was a pre-existing condition. Nonetheless, in a case series of 66 patients undergoing serial computed tomography (CT) imaging for tumor surveillance, 24 developed NAFLD and 11 of those developed liver enzyme elevations while being treated with tamoxifen.137 Onset of NAFLD occurred 1–44 months after beginning tamoxifen and resolution occurred 1–14 months after stopping tamoxifen. Some studies have shown that being overweight predisposes to developing NAFLD on tamoxifen, suggesting that, mechanistically, tamoxifen may cause NASH by exacerbating underlying metabolic abnormalities.138–140 The mechanisms of tamoxifen-induced NAFLD and NASH have been debated. While commonly used as an estrogen antagonist, tamoxifen also exerts estrogen agonist effects in a tissue-dependent fashion. On the other hand, estrogens may be important for normal lipid metabolism141 and the antagonism of estrogen effects or the absence of estrogens may also contribute to the development of NAFLD.140,142,143 The management of tamoxifen-induced NASH is challenging (see Special issues, below). Comparatively few data are available for alternative agents such as raloxifene and toremifene, but early reports have suggested the development of NASH while on toremifene.138,144 Initial reports of a benefit of the PPARa ligand bezafibrate in patients with tamoxifen-induced NASH have suggested that this approach needs further study.138,140
CLINICAL FEATURES OF NAFLD AND NASH (Table 55-8) DIAGNOSIS The presence of NAFLD is often first suspected based on the results of imaging studies such as the finding of a diffusely echogenic liver on ultrasound or a relatively hypodense liver on CT. A diagnosis of NAFLD can be confirmed by liver biopsy, although the current practice of most clinicians is not to pursue the diagnostic impression further if liver enzymes are normal. The potential pitfalls of this clinical algorithm are discussed further below. NASH, the subset of NAFLD characterized by significant necroinflammatory changes, can only be diagnosed by biopsy because there are currently no symp-
Chapter 55 NASH
toms, signs, or useful non-invasive markers of the features of NASH that set it apart from the broader inclusive diagnosis of NAFLD.
SYMPTOMS NASH is most commonly asymptomatic.25 This is in contrast to patients with alcoholic liver disease, who tend to be more sympto-
matic with similar degree of fatty liver.145,146 Many patients with NASH note fatigue and poor exercise tolerance. Case series have also highlighted right upper quadrant abdominal pain as an associated symptom. Few data exist on the prevalence and nature of this pain. It may be more common than is currently perceived and the symptoms can be difficult to discern from symptoms of cholelithiasis in some patients.
Table 55-8. Clinical Features of Non-Alcoholic Steatohepatitis (NASH)
ALCOHOL HISTORY
Symptoms and physical findings • Fatigue (correlates poorly with histological stage) • Vague aching right upper quadrant pain (usually mild but may be mistaken for gallstone disease) • Hepatomegaly • Increased waist circumference indicates central adiposity220 • Acanthosis nigricans (especially in children206) • Lipodystrophy Laboratory • Aminotransferase elevations (rarely more than 10-fold above upper reference range; aminotransferases can be in the normal range with NASH or cirrhosis266) • ALT typically greater than AST (AST > ALT suggests occult alcohol abuse or significant fibrosis • Elevated insulin ¥ glucose product (basis of the HOMA-IR and QUICKI) • Hypertriglyceridemia • Positive antinuclear antibody in about one-third of patients287 • Abnormal iron indices Imaging • No imaging modality can reliably identify fibrosis and stage the disease167 • Ultrasound demonstrates a “bright” liver but is insensitive (cannot detect steatosis less than 25–30%) and non-specific (increased echogenicity of fibrosis or cirrhosis can be mistaken for steatosis) • CT allows accurate quantification of fat but at increased cost • MR imaging and spectroscopy allow measurement of fat content and possibly ATP levels in fatty liver114
Excluding significant alcohol use in a patient with steatohepatitis is essential for establishing a diagnosis of NASH. How much alcohol constitutes enough to contribute to liver disease is debated. A large population study in Italy, the Dionysos study, suggested that alcohol consumption less than 30 grams daily is not associated with adverse sequelae.10 This seems quite generous to many, and an upper limit of alcohol consumption for most studies of NAFLD is set at 20 grams daily to exclude fully the possibility that alcohol could be playing a role. Certainly, when consumption reaches an amount over 60 grams daily, then it most likely plays a role in the development of hepatic steatosis. For an amount between 20 and 60 grams daily, the role of alcohol is uncertain. Quantifying alcohol consumption can be at best a rough approximation in the clinical setting. For the sake of clinical studies, daily alcohol consumption is typically described as grams of ethanol consumed daily. This of course begs the question of exactly how much beer, wine, or liquor delivers a specified amount of alcohol. A commonly used conversion is that 10 grams of alcohol is roughly the alcohol content of one beer, one glass of wine, or one standard drink containing distilled spirits. In reality, these drinks typically contain anywhere from 10 to 20 grams of ethanol, depending on the type of beer or wine, what constitutes a “glass of wine” and the volume of distilled spirits actually used to prepare a drink. Shown in Table 55-9 is a compilation of the alcohol content of various common beverages to assist in the accurate estimation of daily alcohol consumption.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; HOMA-IR, homeostasis model assessment method for insulin resistance; QUICKI, quantitative insulin sensitivity check index; CT, computed tomography; MR, magnetic resonance; ATP, adenosine triphosphate.
Table 55-9. Alcohol Content of Beverages Commonly Consumed in the USA Beverage
Nominal alcohol content (vol/vol)a Typical
Domestic beers Domestic light beers Malt liquors, ales, stoutsb Wine “Fortified” winec Distilled spirits
Range
Alcohol content (g/dl)
Volume (ml) for 10 g alcohol
Typical
Range
Typical
Range
“Standard drink” volume
Grams alcohol per standard drink Typical
Range
4.66% (e.g., Budweiser) 3.43% (e.g., Bud Light)
4–7% (3.2–5.5% wt/vol) 3.1–4.3% (2.5–3.4% wt/vol)
3.7 2.7
3.2–5.5 2.5–3.4
270 370
180–320 300–410
12 oz (355 ml) 12 oz (355 ml)
13 10
11–20 8.7–12
5.89% (e.g., Colt 45)
5–12% (4–9% wt/vol)
4.7
4.0–9.5
215
110–260
12 oz (355 ml)
16.5
14–32
14% 20% 40%
12–15% 18–20% 40–75% (80–151 proof )
11.1 15.8 32
9.5–11.9 14.2–15.8 32–60
90 60 32
85–105 60–70 17–32
4 oz (115 ml) 4 oz (115 ml) 1 oz (30 ml)
13 18 9.5
11–14 16–18 9.5–18
a
By federal law, the alcohol content of beer in the USA is measured in wt% (g/dl). All other alcoholic beverages are measured in vol% (ml/dl). The conversion is g/dl ¥ 1.27 = ml/dl, or ml/dl ¥ 0.79 = g/dl. Since 1935, federal law has prohibited mention of alcohol content on beer labels (to avoid using it to market a product) unless otherwise required by state law. This law was successfully challenged in a 1992 federal court case. b In California, a product labeled “beer” must have an alcohol content of less than 4% wt/vol (5% vol/vol). Products with higher alcohol content are identified by other names, such as malt liquor, ale, or stout. c Fortified wines are produced by adding distilled alcohol after fermentation is complete to increase the alcohol content.
1043
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
EXAMINATION FINDINGS There are no specific physical examination findings that reliably indicate the presence of NAFLD or NASH. Because insulin resistance is an underlying abnormality in most patients with NAFLD, centripetal obesity, hypertension, and possibly acanthosis nigricans may be seen in association with NAFLD. Hepatomegaly is present in up to 75% of patients,25 but can be difficult to identify on physical exam in patients with marked visceral adiposity. An atypical distribution of body fat may suggest a lipodystrophy, but subtle forms of lipodystrophy can be difficult to distinguish from the spectrum fat distribution observed in the normal population.147
LABORATORY TESTS Clues to the presence or absence of NAFLD can be obtained from laboratory testing, although the presence of NASH in the setting of NAFLD cannot be established without a liver biopsy. Laboratory testing also serves to exclude the presence of other diseases or identify diseases such as hepatitis C infection that may coexist with NAFLD.
Aminotransferases Elevated serum levels of the aminotransferases ALT and aspartate aminotransferase (AST) are commonly used as a screen for unsuspected liver diseases such as NAFLD. Once other causes of liver disease have been excluded, NASH is the most common cause of elevated aminotransferases.148 On the other hand, the prevalence of elevated serum ALT and AST levels in patients with NAFLD or NASH is unknown.
Disease Despite “Normal” Aminotransferases Data obtained from morbidly obese subjects undergoing bariatric surgery have demonstrated that the entire spectrum of NAFLD from minor amounts of fat to aggressive NASH and cirrhosis can be found, even with normal aminotransferases.22,149,150 Although aminotransferases lack specificity and sensitivity for the detection of NAFLD,22 better tests short of liver biopsy are not available.
Defining Normal Reference Ranges for Serum Aminotransferases A contributing factor to poor sensitivity of aminotransferases in detecting NAFLD or other liver diseases such as hepatitis C is how the upper limit of the reference range (commonly called the upper limit of normal, or ULN) is defined.151 By convention, normal ranges are defined by the results of any test in a disease-free population. Since imaging and liver biopsy are needed to exclude NAFLD and NASH, clinical laboratories cannot exclude these in the reference population used to define normal. Surrogate markers such as body mass index (BMI) and other clinical markers of insulin resistance that correlate with an elevated ALT in population studies are not used to exclude patients from the reference population by clinical laboratories.22,152 This has resulted in a wide range of upper reference range values used by different clinical laboratories, each with its own reference population. Although some variability is introduced by the use of different analyzers in clinical laboratories, the major contributor to the variability is variability of reference population chosen (B. Neuschwan-
1044
der-Tetri, unpublished data). One study based on a carefully screened population suggested that the upper reference range for serum ALT activity for men and women should be 30 and 19 U/l respectively.153 This compares to an upper reference range among laboratories that varies from 30 to 75 U/l. One disadvantage to lowering the upper reference range to increase the sensitivity of the ALT is the loss of specificity. This could lead to unnecessary testing until a new paradigm for the clinical management of mildly elevated levels (e.g., 35 U/l) emerges.154
Usefulness of ALT:AST Ratio One of the ways in which measuring the ALT and AST can contribute to the diagnosis of NAFLD is to calculate the ratio of one to the other. An AST greater than the ALT is highly suggestive of alcoholic liver disease whereas the ALT is typically greater than the AST in the non-cirrhotic NAFLD patient.155 However, with the development of significant fibrosis or cirrhosis in patients with NASH, the AST can exceed the ALT. One series of 70 patients with NASH found AST:ALT ratios of 0.7, 0.9, and 1.4 with no fibrosis, mild fibrosis, or cirrhosis respectively.156
Cholestatic Enzymes NAFLD is primarily a hepatocellular disease with only occasional biliary tract abnormalities identified on biopsy. Nonetheless, g-glutamyltranspeptidase (GGT) can be elevated in NAFLD in proportion to the degree of fat accumulation149 and elevated serum alkaline phosphatase levels have been reported in obese patients.157 A pilot study of an insulin-sensitizing agent (see below) that caused improvement in aminotransferases also found corresponding decreases in both the serum alkaline phosphatase and GGT.158
Other Laboratory Tests The role of other laboratory tests in the diagnosis of NAFLD is primarily to exclude other causes of liver disease or to identify concomitant diseases. Antibodies associated with autoimmune hepatitis are sometimes found in low to moderate titers.159 High titers provide an additional impetus to perform a liver biopsy to exclude fully this treatable disease. Identifying insulin resistance by laboratory testing can suggest a diagnosis of NAFLD (see below). Serum lipid levels can provide important information regarding other correlates of the metabolic syndrome that require appropriate management. Identifying the relatively rare patient with hypobetalipoproteinemia as a cause of NAFLD can be facilitated by an astute observation of an unexpectedly low serum cholesterol. Wilson’s disease must be fully excluded as an underlying disorder in children and adolescents with NAFLD by measuring the serum ceruloplasmin and other testing as needed (Chapter 66).
LIVER BIOPSY The diagnosis of NASH can only be established with certainty by performing a liver biopsy (see review of histological findings in Chapter 13).4,160 An unresolved issue requiring further studies before definitive recommendations can be established is the appropriate selection of patients who would benefit from a liver biopsy. The decision to perform a liver biopsy in a patient suspected of having NAFLD should take into account the inherent risks associ-
Chapter 55 NASH
ated with a biopsy balanced against the benefits.161,162 An uninterested patient, substantial comorbidities such as advanced vascular disease, and a variety of other clinical circumstances lessen the enthusiasm for performing a liver biopsy. Despite the known reasons not to perform a liver biopsy, studies have suggested that significant benefit can be gained from the information obtained. For example, at least a third of patients suspected of having NAFLD on clinical grounds were found to have another cause of liver enzyme elevations when a biopsy was performed.163 Historically, biopsies have changed management in a fraction of patients, but the need to consider treatment will likely increase as new therapeutic alternatives for NAFLD emerge.155,163,164 Establishing the diagnosis of NASH and distinguishing it from simple steatosis, or non-NASH NAFLD, relies on finding steatosis, characteristic inflammation, and evidence of cellular injury (Table 55-1).4 The extent of necroinflammatory changes and fibrosis (if any) needed to establish the diagnosis of NASH continues to be refined. An occasional finding is a component of microvesicular steatosis along with macrovesicular steatosis. Histologically, the hepatocellular triglyceride accumulation in NAFLD is commonly a mixture of microvesicular and macrovesicular fat droplets. Microvesicular fat is identified as droplets smaller than the nucleus that do not displace the nucleus whereas macrovesicular fat displaces cell contents, including the nucleus to the cell periphery. Whether the differences between the two patterns of fat accumulation are related to different pathophysiological processes or the rate at which fat accumulates is unknown.
IMAGING STUDIES Imaging studies often provide the first evidence that a patient has otherwise unsuspected NAFLD. Unfortunately, a rational and costeffective approach of what to do next has not been developed. A reasonable approach includes estimating insulin sensitivity, excluding other causes of liver disease, and using this information to determine whether a liver biopsy should be pursued (Figure 55-7). The characteristics of commonly used imaging studies in NAFLD are discussed below. Although each has its strengths and weaknesses, imaging variably identifies steatosis and cirrhosis. No imaging study can assess the necroinflammatory changes or fibrosis that distinguish NASH from less worrisome forms of NAFLD.165–167
Ultrasound Ultrasound examination of the liver may be the least costly imaging method, but it lacks both specificity and sensitivity. Fat in the liver confers a “bright” appearance on ultrasound, but significant fibrosis without fat can have a similar appearance.165 Methods have been proposed to increase the sensitivity of ultrasound using quantitative techniques, but these have not been widely accepted or established in clinical practice. Despite these shortcomings, ultrasound remains a commonly used method of identifying the NAFLD as a cause of unexplained liver enzyme elevations.
Computed Tomography CT imaging of the liver is more sensitive than ultrasound for detecting NAFLD, but at a greater cost.165 Like ultrasound, CT cannot
Incidental fatty liver on imaging (ultrasound, CT, MRI)
ALT > 40 U/L
Evaluate for other causes of liver disease •Alcohol •Viral •Drug •Autoimmune •Other metabolic (e.g., Wilson's disease, alpha-1 antitrypsin deficiency) Assess insulin sensitivity: •Non-diabetic: measure fasting insulin and glucose •Type 2 diabetic: assume insulin resistance
If NAFLD or NASH suspected:
Consider biopsy
Consider empiric therapy Lifestyle modification (exercise, dietary modification)
Figure 55-7. A simple diagnostic algorithm for non-alchoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH). Most patients come to medical attention because of unexplained serum alanine aminotransferase (ALT) elevations or evidence of NAFLD on imaging studies. One approach is to evaluate other possible causes of enzyme elevations or fatty liver and then assess insulin sensitivity. If NAFLD or NASH is suspected based on the results of this testing, then a decision whether to perform a liver biopsy or simply recommend empiric therapy needs to be made. Factors favoring a biopsy include risks for advanced disease (higher age, diabetes, obesity, hypertension), level of patient concern, and level of confidence that other diseases have been excluded, especially treatable diseases such as autoimmune hepatitis and hemochromatosis. The degree of ALT elevation undoubtedly influences the decision of most clinicians, although this probably lacks a rational basis because advanced NASH can be found with normal liver enzymes.
identify necroinflammatory changes or early fibrosis that signify the presence of NASH. Radiographic tissue density is estimated using Hounsfield units. The fatty liver has a lower density than normal and the presence of fat can be calculated by comparing the liver density to spleen or paraspinal muscle density (Figure 55-8). Common formulas include either subtracting the liver density from the spleen density or calculating the liver-to-spleen ratio. A difference between liver and spleen of > 10 Hounsfield units indicates liver fat,168 as does a liver-to-spleen ratio of less than 1.
Magnetic Resonance (MR) MR imaging and MR spectroscopy are the most sensitive means of detecting NAFLD, with the tradeoff of also being the most costly.165 In fact, the specificity and sensitivity of MR techniques may exceed the accuracy of the subjective interpretation of a limited sample of tissue obtained by liver biopsy. A number of MR techniques have been proposed for optimizing the detection of fat169 and spectroscopy techniques have also proved useful in detecting disruption of ATP production in experimental studies.115
1045
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Figure 55-8. Low-density liver on abdominal computed tomographic imaging of a patient with nonalcoholic steatohepatitis (NASH). Whereas the liver density should be roughly equivalent to the spleen, in this typical example, the liver has a mean density of 26 Hounsfield units (HU) compared to a spleen density of 48 HU, resulting in a liver-to-spleen density ratio of 0.54. By definition, water has a density of 0 HU and air a density of –1000 HU.
Nuclear Medicine Improvements in ultrasound, CT, and MR have led to general abandonment of nuclear medicine studies as a means of detecting hepatic abnormalities in general. One very sensitive means of estimating liver fat content is the use of 133Xe washout, a technique reported in the past but no longer in clinical use.
Focal Fat and Focal Sparing The common use of abdominal imaging to evaluate suspected liver abnormalities or other intra-abdominal processes has led to increasing discovery of incidental focal lesions in the liver caused by focal sparing in an otherwise fatty liver or focal fat within an otherwise normal liver. Focal fat may be caused by aberrant venous drainage of insulin-enriched pancreatic blood into a specific region of the liver such as the posterior aspect of segment IV near the porta hepatis (Figure 55-9).60 Focal sparing may be related to variations in the blood supply as well. Focal sparing adjacent to the gallbladder is common and can be explained by a localized blood supply originating from the gallbladder. In 290 patients with hepatic steatosis, 78% of those with a gallbladder had focal sparing in this area.170 In rare circumstances (about 1% of patients), focal fat or focal sparing can occur adjacent to the falciform ligament and may be difficult to distinguish from tumor.171 Focal sparing in this area may be related to non-insulin-enriched blood draining directly into the liver from a gastric vein that fails to join the portal vein before it enters the liver.60
CLINICAL ESTIMATION OF INSULIN RESISTANCE The ability to estimate insulin resistance and thus predict the presence of complications associated with this disorder has become
1046
necessary in the management of patients with NAFLD. The presence of insulin resistance suggests that therapeutic options aimed at improving insulin sensitivity may be beneficial, whereas the occasional patient with NAFLD but normal insulin sensitivity should be further evaluated for uncommon or unrecognized causes such as other metabolic abnormalities or covert alcohol abuse. Unfortunately, how best to define and measure insulin resistance is a source of continued debate. Perhaps it is no coincidence that the difficulties encountered in defining and measuring insulin resistance are no different from those with defining NAFLD and NASH,6 namely the difficulty in drawing a distinction between what is normal and what constitutes disease. This becomes impossible when patients present within a continuum from normal to overtly lifethreatening disease yet measures are sought that dichotomize between the presence and absence of a disorder. Some of the commonly used methods to assess insulin sensitivity and the practical issues related to reliably measuring serum insulin concentrations have been reviewed.172 Despite potential problems, biochemical assessment of insulin resistance commonly relies on measuring serum insulin levels, and two measures, the homeostasis model assessment method for insulin resistance (HOMA-IR) and the quantitative insulin sensitivity check index (QUICKI), use the fasting insulin level multiplied by the fasting glucose level.173 Although HOMA-IR has been widely used, Katz et al. suggest that the QUICKI is a reasonable estimate of the glucose clamp-derived index of insulin sensitivity (SIClamp) because the QUICKI is linearly related to the SIClamp.173–175 This may be due to the fact that both the QUICKI and the SIClamp calculations use log-transformed values. Nonetheless, the QUICKI and the HOMA-IR are both based on the product of the fasting insulin multiplied by the fasting glucose (Figure 55-10), a point missed by many authors when they calcu-
Chapter 55 NASH Figure 55-9. Focal fat caused by aberrant pancreatic venous drainage. (A) Focal fat in this patient in posterior segment IV (arrow) was attributed to pancreatic venous drainage into a focal area of the liver. This image was obtained during the portal venous phase after injection of contrast into the superior mesenteric artery, showing that the area of focal fat was not perfused by portal blood flow. (B) Selective injection of the superior pancreaticoduodenal artery resulted in venous filling of the region of focal fat, indicating that the insulin-enriched pancreatic blood flow was directed to the region with focal fat. (Reproduced with permission from Fukukura Y, Fujiyoshi F, Inoue H, et al. Focal fatty infiltration in the posterior aspect of hepatic segment IV: relationship to pancreaticoduodenal venous drainage. Am J Gastro 2000; 95:3590–3595.)
A
B
late that the two are found to be highly correlated in a given study. Because they are derived from the same number (glucose ¥ insulin), the HOMA-IR and QUICKI are potentially interchangeable in their utility.173 Because the QUICKI is log-transformed, the arithmetic mean of HOMA-IR values from a study cohort may not be equivalent to the arithmetic mean of the corresponding QUICKI values. The relevance of this difference in evaluating study data has not been determined. Because insulin sensitivity varies over a range and there are intermediate values associated with disease in some people and not in
others, defining a cutoff between normal and abnormal has proven difficult. Additionally, the absence of standardization of insulin measurements may prevent direct comparison of results among different study sites. A lower limit of normal QUICKI in the range of 0.357–0.382 is often reported with values less than this indicating insulin resistance (typically in the range of 0.25–0.35 for NASH patients). An upper limit of normal HOMA-IR is in the range of 1.0–1.5 with higher values signifying insulin resistance. By calculation, the QUICKI values of 0.25–0.35 typical of NASH patients equate to HOMA-IR values of 25–1.8 respectively. A potentially
1047
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
useful and expedient clinical pearl is that based on these normal limits, values for the glucose ¥ insulin product exceeding 700 when one multiplies insulin (mU/l) ¥ glucose (mg/dl) may correspond to insulin resistance (equivalent of QUICKI < 0.351 and HOMA-IR > 1.73). Why this simple product is not used in place of more complicated formulas is not justified in the literature. Although a simple mathematical relationship exists between the two indices of insulin sensitivity, the HOMA-IR has a greater spread between values in patients with insulin resistance and little spread between values of patients with normal or near-normal insulin sensitivity. The opposite is true for the QUICKI, in which patients with normal insulin sensitivity have little variation in their values. The clinical relevance of this significant difference is yet to be established, although a treatment that causes a modest improvement in insulin sensitivity in patients with severe insulin resistance will cause a large change in the HOMA-IR and a small change in the QUICKI. Obtaining postprandial insulin levels after an oral glucose challenge may prove to be an even better means of assessing insulin resistance,158,176 but this too will need further study.
HOMA-IR =
glucose[mM] x insulin[mU/L] 22.5
(or) HOMA-IR =
glucose[mg/dL] x insulin[mU/L] 405 1
QUICKI =
log(glucose[mg/dL] x insulin[mU/L]) Figure 55-10. Clinical estimation of insulin sensitivity. Two easily used measures of insulin sensitivity are the homeostasis model assessment of insulin resistance (HOMA-IR) and the quantitative insulin sensitivity check index (QUICKI). Both indices are derived from the fasting insulin level multiplied by a simultaneously obtained fasting glucose level and are therefore mathematically directly related to each other. The QUICKI is log-transformed, which makes it linearly related to the SIClamp, a more accurate measure of insulin sensitivity based on the insulin clamp technique.
PREDICTORS OF NASH AND FIBROSIS Appropriate management of patients with suspected NASH requires the ability to predict which patients with elevated liver enzymes or NAFLD detected by imaging are at risk for progressive disease and thus warrant more aggressive evaluation and treatment (Tables 55-5 and 55-6). One study of severely obese patients undergoing bariatric surgery (BMI > 35 kg/m2) found that hypertension, ALT elevation, and insulin resistance predicted the presence of NASH in patients with NAFLD.15 In fact, three-quarters of these severely obese patients with both hypertension and diabetes had NASH whereas 7% with neither condition had NASH (Figure 5511). Diabetes and hypertension were also predictive of advanced fibrosis. All patients with advanced fibrosis in this study had at least diabetes or hypertension. Another study established the importance of age; significant fibrosis (stage 3 or 4) was present in only 4% of NASH patients under the age of 45 yet it was present in 40% of those 45 years and older.177 The available data can be summarized by observing that the greatest risk for significant fibrosis on a liver biopsy is the presence of obesity and diabetes in a patient over the age of 45 years with an AST:ALT ratio > 1. Many studies evaluating risk factors for the presence of fibrosis identify the degree of steatosis and necroinflammatory changes on the liver biopsy as predictive.150 While this association is informative with respect to the pathogenesis of fibrosis, it offers no clinical utility. The association of one biopsy finding to another obviously does not help in deciding whom to biopsy. Moreover, the association between necroinflammatory changes and fibrosis on a given biopsy has not been shown to predict who will develop fibrosis with time. The ability to predict who will develop fibrosis based on findings in a non-fibrotic biopsy would be a major contribution to clinical management.
DIFFERENTIAL DIAGNOSIS Because the diagnosis of NASH is based on histological findings, alcoholic hepatitis must be excluded from the outset. Other causes of steatohepatitis that must also be excluded with appropriate clinical and laboratory evaluation are Wilson’s disease and drug-induced steatohepatitis. Whether these entities share pathophysiologic
Figure 55-11. Type 2 diabetes and hypertension (HTN) as major risk factors for non-alcoholic steatohepatitis (NASH) in obese patients. The presence of both diabetes and HTN confers the greatest risk for having NASH in severely obese patients undergoing bariatric surgery, whereas the risk of having NASH is low in the absence of both risk factors. (Reproduced from Dixon JB, Bhathal PS, O’Brien PE. Nonalcoholic fatty liver disease: predictors of nonalcoholic steatohepatitis and liver fibrosis in the severely obese. Gastroenterology 2001; 121:91–100, with permission.)
Percentage of patients with NASH
80 70 60 50 40 30 20 10 0 Neither diabetes nor HTN (n=57)
1048
Hypertension alone (n=29)
Diabetes alone (n=8)
Both diabetes and HTN (n=11)
Chapter 55 NASH
mechanisms with NASH is yet to be defined and therefore whether they cause NASH or must be excluded to establish a diagnosis of NASH is a source of ongoing debate.
CONCOMITANT DISEASES NASH was once thought of as a diagnosis of exclusion, implying that the diagnosis could not be reached unless all other causes of liver disease were excluded. As the histologic criteria have been refined and underlying pathophysiologic mechanisms better understood, the coexistence of NASH with other causes of liver disease is increasingly recognized. For example, 5.5% of biopsies with steatohepatitis were found to have concomitant other diseases in one series.9 In this series, chronic hepatitis C was the most common; conversely, NASH was found in about 4% of biopsies obtained for hepatitis C. Another series identified alpha1-antitrypsin defects as the most common concomitant disease process in NAFLD, occurring in about 8% of biopsies.178
NAFLD and Chronic Hepatitis C Infection NAFLD and hepatitis C are two common forms of liver disease and overlap between the two is not surprising.9,178,179 However, the presence of NAFLD has been found to be more common in patients infected with hepatitis C than the general population in both adults and children. For example, a series of 148 hepatitis C virus (HCV) patients in Australia identified NAFLD in 61% compared to the noninfected adult prevalence of about 20%.180 Studies in humans and animals have shed some light on the interaction between NAFLD and HCV infection to show that obesity and insulin resistance are the major factors predisposing to NAFLD in HCV patients (Figure 55-12).181,182 HCV infection does contribute to insulin resistance and could promote the development of NAFLD by this mechanism.183,184 Although obesity and insulin resistance are the underlying causes of NAFLD in most patients with HCV, HCV genotype 3 infection can directly cause NAFLD.181,185–187 Patients with genotype 3 infection often have lower circulating lipids than patients with other genotypes, indicat-
HCV infection
+
+
Insulin resistance
NAFLD/NASH
(genotype 3) +
+
Fibrosis
Cirrhosis/HCC
Figure 55-12. The interaction between non-alcoholic fatty liver disease (NAFLD) and hepatitis C virus (HCV) infection. The primary contributor to NAFLD in most patients with HCV infection is insulin resistance caused by obesity, sedentary lifestyle and probably genetic factors. However, HCV infection contributes to the progression of NAFLD by increasing the degree of insulin resistance, accelerating fibrogenesis, and contributing to the risk of developing hepatocellular carcinoma (HCC). In the case of genotype 3 infection, HCV directly promotes the accumulation of fat in the liver.
ing impaired hepatic secretion of triglyceride as VLDL.8,186,188,189 Moreover, serum lipids typically increase190 and the NAFLD in these patients improves with successful treatment of the HCV infection.191 Significantly impaired secretion of fat from the liver is rare with other genotypes. Studies in animals and cultured cells have explored the basis for the interaction between viral proteins and hepatocellular fat trafficking.192 HCV core and NS-5 proteins can impair triglyceride export as VLDL from hepatocytes. Viral proteins associate with nascent lipid droplets, apolipoprotein A, or microsomal triglyceride transfer protein. This interaction and increased oxidant stress have been proposed as mechanisms of virally impaired hepatocellular triglyceride handling.193–196 Paradoxically, in vitro studies reported to date have used only genotype 1 virus, yet the clinical association between HCV infection and NAFLD is strongest in genotype 3 infected patients. The coexistence of the two disorders is important because the presence of NAFLD accelerates the progression of liver disease associated with hepatitis C. Liver fibrosis in HCV patients correlated with the presence of NAFLD independent of age and obesity in one study,197 Another study from Italy found that NAFLD doubled the rate of fibrosis progression from approximately one Scheuer stage every 8 years in HCV without NAFLD to one stage every 4 years in HCV with coexisting NAFLD.198 Similar findings of accelerated progression in studies from around the world have been reported in patients with NAFLD or its surrogate, obesity.8,178,187,199,200 Just as it increases the rate of fibrosis, the presence of NAFLD may also increase the risk of developing HCC in HCV patients.201 The presence of NAFLD also appears to reduce significantly the likelihood of obtaining a sustained virologic response to treatment with interferon and ribavirin, even with appropriate weight-based drug doses. Analysis of 1428 patients treated with interferon and ribavirin in a study that excluded NASH demonstrated that patients without steatosis achieved a sustained virologic response rate (SVR) of 66% whereas those with simple steatosis had a lower average SVR of 50%.8 A negative impact on response to treatment has been reported in other studies as well.178,187 Obesity, a surrogate marker for insulin resistance and NAFLD, has also been associated with a lower response rate despite weight-based dosing.202,203 The optimal management of patients with both NAFLD and HCV infection is not known. Some studies suggest that insulin resistance can directly promote fibrogenesis,204 possibly through the expression of TNF-a and cytochrome P450 2E1.205 This observation may provide further evidence of the need to assess insulin sensitivity in patients with HCV infection and direct initial management towards improving insulin sensitivity. The one notable exception is genotype 3 patients who may have NAFLD as a direct outcome of the viral infection and the NAFLD improves with treatment of the infection.
DISTINGUISHING ALCOHOLIC STEATOHEPATITIS FROM NASH Excluding alcohol as a cause of steatohepatitis can be challenging in a subset of patients. Contributing to this challenge is the likely, but unproved, possibility that some patients may have both excessive alcohol consumption and insulin resistance as underlying causes of
1049
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
steatohepatitis. Identifying such patients will require a better understanding of the pathogenesis of both diseases and the development of specific disease markers. In addition, some authors have concluded that NASH and ASH are indistinguishable and, indeed, the landmark paper by Ludwig that popularized the term NASH did so to establish that the previously described “alcohol-like” liver disease was a separate entity but with remarkably similar histologic findings to alcoholic liver disease. The published data and growing experience of clinicians suggest that a distinction can often be established from clinical and pathological findings.145 Specifically, the features below can be used to help distinguish alcoholic steatohepatitis (ASH) from NASH (Table 55-10).
Clinical and Historical Findings Compared to ASH, NASH is more commonly asymptomatic. Vague right upper quadrant abdominal pain is increasingly recognized in patients with NASH, but significant symptoms such as anorexia, fever, jaundice, and weight loss are characteristics of patients with severe alcoholic liver disease.145 These symptoms tend to be found in patients hospitalized with severe ASH while the patient with covert alcohol abuse and less severe ASH may present with minimal symptoms similar to patients with NASH.146 NASH tends to be
Table 55-10. Features Suggestive of Either Non-Alcoholic Steatohepatitis (NASH) or Alcoholic Steatohepatitis (ASH) Features associated with NASH Clinical/historical Alcohol <20 g/day (corroborated history) Type 2 diabetes Obesity/overweight Hypertension Polycystic ovary syndrome/ dysmenorrhea
Features associated with ASH Alcohol use/abuse
Physical Exam Findings While obesity (BMI ≥ 30 kg/m2) is not a prerequisite for the development of NASH, it is definitely a risk factor for NASH whereas it is not recognized as a risk for ASH. Hepatomegaly is a feature of both, although it can be more difficult to appreciate in the obese patient with NASH. Acanthosis nigricans (regionally hyperpigmented skin) is a feature of insulin resistance and may be more common in NASH (especially in children206), whereas it is not a feature of ASH.
Laboratory Abnormalities The aminotransferase ratio is one of the most helpful objective measures in distinguishing NASH from unsuspected ASH in the covert drinker. NASH is typically characterized by a serum ALT that exceeds the AST by up to two- to threefold, whereas the inverse is true in ASH, when the AST typically exceeds the ALT by a ratio that can be up to two- to threefold.145 In both diseases, AST or ALT values over 300 U/l are unusual. Hyperbilirubinemia is not a feature of NASH that has not progressed to cirrhosis whereas it is relatively common in patients with severe ASH. Similarly, peripheral leukocytosis can be a feature of ASH but not NASH.
Liver Biopsy Features
Pathogenesis Fat accumulation is primary Injury is secondary to fat (oxidant stress?)
Mitochondrial dysfunction? Acetaldehyde toxicity?
On a given liver biopsy, the distinction between ASH and NASH can be impossible without sufficient clinical information. Despite the many similarities, certain features can suggest one diagnosis over the other.207 Mallory bodies (Mallory’s hyaline) tend to be more numerous and well-formed in ASH whereas they can be indistinct or even absent in NASH.145,146 Conversely, the steatosis is typically more severe in NASH and it can be relatively mild in some cases of ASH. The presence of steatosis is probably a necessary prerequisite for the development of NASH whereas it may be a relative epiphenomenon in ASH. The presence of glycogenated nuclei (clear inclusions nearly filling the nuclei) favors NASH over ASH,208 although these are a non-specific finding and can be found in diabetes, Wilson’s disease, and a number of other disorders. Sometimes found in ASH but not NASH is evidence of severe perivenous (zone 3) injury, including sclerosing hyaline necrosis and the presence of central vein injury with the appearance of a veno-occlusive-like lesion.209 The presence of these abnormalities strongly favors alcoholism as the underlying disorder. Ballooned hepatocytes and the characteristic zone 3 “chickenwire” fibrosis surrounding hepatocytes in the region of the central vein are features of both diseases.
Prognosis Cirrhosis is major adverse outcome, risk is about 1%
Liver failure, risk up to 50%
ASSOCIATED CONDITIONS
Physical exam Obesity Acanthosis nigricans Laboratory ALT > AST (unless cirrhotic) Normal bilirubin
Liver biopsy Poorly formed or undetectable Mallory bodies Steatosis always present (except in cirrhosis) Glycogenated nuclei
Anorexia Leukocytosis, fever
Jaundice
AST > ALT Elevated bilirubin Hypoalbuminemia Peripheral leukocytosis Well-formed, numerous Mallory bodies Steatosis may be less prominent Sclerosing hyaline necrosis and central vein injury
ALT, alanine aminotransferase; AST, aspartate aminotransferase.
1050
more associated with features of the metabolic syndrome, such as diabetes, hypertension, obesity, dysmenorrhea, and possibly polycystic ovary syndrome (PCOS).145 Patients with severe ASH have a mortality that approaches 50% related to liver failure per se whereas patients with NASH are primarily at risk for the gradual development of cirrhosis and its complications.
For the first two decades after NASH was described, case series identified associated conditions such as obesity, diabetes, hyperten-
Chapter 55 NASH
sion, and hyperlipidemia in an effort to understand better the causes and identify treatments for NASH. With recent advances in the understanding of NASH, especially with respect to the role of insulin resistance, the importance of identifying associated conditions has evolved. As the importance of using associated conditions to identify pathophysiologic mechanisms has diminished, the role of identifying associated conditions has changed to identifying patients at risk for having NASH. The associated conditions identified in earlier studies relate mostly to insulin-resistance: obesity as an underlying cause of insulin resistance and diabetes, hypertension, and hyperlipidemia as common manifestations of insulin resistance. Additionally, acanthosis nigricans, the dermatologic abnormality associated with insulin resistance, may be associated with NASH, especially in children. Anecdotal reports suggest that PCOS may also be overrepresented in patients with NASH. PCOS is the ovarian response to hyperinsulinemia and is characterized by ovarian testosterone production, leading to menstrual irregularity, hirsutism, and acne.210 Interestingly, ovarian cysts are no longer required for the diagnosis.
DISEASE COMPLICATIONS The complication of NAFLD that is the primary cause of concern is its risk of causing progressive fibrosis, cirrhosis, and the complications of cirrhosis, including HCC and death. Our current understanding of how often this occurs and the clinical risk factors associated with progression to cirrhosis are discussed in the review of natural history. NAFLD is not recognized to be a direct cause of extrahepatic complications. In general, the associated conditions described above are thought to arise from the underlying metabolic abnormalities that lead to fatty liver and are not a direct consequence of fatty liver.
TREATMENT No effective therapies or preventive measures have been established by rigorous clinical studies for NASH. A number of agents and interventions have been examined in small trials and case series, often with encouraging findings initially, but few randomized controlled trials have been conducted in patients with NASH (Table 55-11).211 One exception was a well-designed and appropriately powered trial of ursodeoxycholic acid (UDCA) that followed an initial promising pilot study.212 Unfortunately, this trial determined that UDCA was no better than placebo (see below).213 As more complex studies are designed and executed, a therapeutic approach to NASH may emerge that combines a multidisciplinary approach to obesity, insulin resistance, sedentary lifestyle, dietary imbalances, and genetic variations. Current understanding of the pathogenesis of NASH raises the possibility that addressing insulin resistance may be an effective preventive measure, as well as a valuable treatment option.
IMPROVING INSULIN SENSITIVITY Exercise and Lifestyle Modifications Experimental data from animals and humans convincingly demonstrate that exercise reverses insulin resistance,214,215 the most
Table 55-11. Treatment Options for Non-Alcoholic Steatohepatitis (NASH) Target
Possible intervention(s)
Comments
Obesity
Caloric restriction Exercise Drugs
Insulin resistance
Exercise
Several case series suggest a benefit in NASH Necessary for sustained weight loss Insufficient data with antiobesity drugs Most effective means of improving insulin sensitivity known;288 effect on NASH to be established Most effective in conjunction with exercise for improving insulin sensitivity Pilot studies suggest a benefit of metformin and thiazolidinediones Needs confirmatory data
Weight loss
Drugs Abnormal hepatic fat metabolism Oxidant stress
Betaine
Vitamin E
Hepatocellular injury
Other antioxidants Cytoprotective agents
Fibrogenesis
Antifibrotics
Beneficial in one small pilot study in children No controlled trials Rationale for ursodeoxycholic acid and silymarin; ursodeoxycholic acid proven to be ineffective Effective antifibrotics for liver disease yet to be found
common cause of NAFLD. Perhaps the best rationale for promoting a non-sedentary lifestyle as a treatment for NAFLD is provided by the results of the Diabetes Prevention Program (DPP) trial.216 Subjects at risk for developing diabetes who adhered to a modified lifestyle that included increased regular physical activity experienced better improvement in their insulin sensitivity than similar subjects treated with the insulin-sensitizing agent metformin. Additionally, an observational study of over 50 000 nurses also found that less television-watching and more physical activity prevented the onset of diabetes, a disease that represents the final stages of insulin resistance.217 Despite the preponderance of evidence that exercise reduces mortality from cardiovascular causes, data supporting a similar salutary effect in patients with NASH are lacking.211 However, several small trials have shown that exercise may be an effective means of treating fatty liver.218 One 12-week trial of diet and exercise resulted in improved liver enzymes but histopathology was not evaluated.120 How much exercise is needed to normalize insulin sensitivity continues to be debated. Any activity is better than none and this is the approach taken by the National Cholesterol Education Program (NCEP) in recommending a minimum of 30 minutes daily of brisk walking or other moderate activities for the 70% of adults who are sedentary or minimally active. Although such a modest step represents a large accomplishment for many patients, more intense exercise provides a further benefit for those physically able and willing to pursue such lifestyle changes.219
1051
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Weight Reduction The difficulties in achieving and maintaining weight loss are abundantly evident to health care providers and the public alike,220,221 yet the adverse health outcomes of obesity may soon outpace smoking as a preventable cause of illness and death.222 Current treatment approaches to obesity include exercise,223 dietary modifications,224 surgery,225 and drugs.226 Weight loss to treat NAFLD and NASH has been the subject of a number of small trials and case series.221 These trials each have their weaknesses and no firm conclusions can be drawn. In trials that included liver biopsies, a concern has been raised that inflammation or fibrosis could worsen. However, the histological criteria used to assess NASH were those used to evaluate chronic viral hepatitis. The changes associated with improvement of NASH may be different from the resolution of viral hepatitis. For example, a trial of rosiglitazone found that, as lobular inflammation diminished, periportal inflammation increased.227 If a histologic grading scheme is used that assesses only periportal inflammation, then such changes could be interpreted as worsening rather than what may turn out to be a step towards resolution of NASH. A current trend in the dietary management of obesity is to favor low-carbohydrate diets, eschewing the carbohydrate-based food pyramid. Although low-carbohydrate diets may be beneficial in the short term, long-term benefits have not been found in the absence of sustained lifestyle modifications that include increased physical activity.223,228,229
Bariatric Surgery Frustration by patients and their physicians with the failure of standard treatment modalities has led to increasing reliance on obesity surgery which, at a price in terms of risks and side effects, can improve insulin sensitivity and its complications.225,230,231 The current enthusiasm for surgery is fueled in part by the relative safety of current procedures compared to the jejunoileal bypass operation performed three decades ago. The older procedure left a blind loop of small bowel that may have caused or exacerbated a NASH-like form of aggressive liver disease. The procedures as they are currently performed have not been associated with progressive liver disease and death from liver failure. In fact, preliminary data suggest that the NASH associated with severe obesity may improve following the Roux en Y gastric bypass.149,232–235 A multicenter study group sponsored by the National Institutes of Health is currently gathering data that should provide definitive information about the safety of current bariatric surgery and whether it is as effective in treating NASH as it is in treating the other manifestations of insulin resistance.
Insulin-Sensitizing Agents Using pharmacological agents to improve insulin sensitivity can provide additional evidence for the role of insulin resistance in NAFLD and these agents may be an important adjunct to therapy in certain patients such as those unable to increase physical activity or lose weight. Two classes of drugs are currently available to improve insulin sensitivity: the thiazolidinediones (TZDs) and the biguanides.
Thiazolidinediones The TZD PPAR-g nuclear receptor ligands were developed as therapeutic agents for the treatment of type 2 diabetes by improving
1052
insulin sensitivity.236 Experimentally, these agents lead to diminished fat accumulation in the liver and muscle of diabetic animals and humans,237 suggesting that they could be beneficial in patients with NASH. Three clinically available agents, troglitazone, rosiglitazone, and pioglitazone, have been evaluated in NASH but only in small pilot studies. In a series of 7 subjects treated for 3–6 months, troglitazone was associated with improved liver enzymes,46 although it has been withdrawn from clinical use because of its idiosyncratic hepatotoxicity. Similar toxicity has not been observed with pioglitazone or rosiglitazone, possibly because troglitazone is metabolized by cytochrome P450 CYP3A4 and this is not a major pathway for the other two. Nonetheless, occasional adverse reactions have been observed with both pioglitazone and rosiglitazone which justify caution with their use. Weight gain, typically caused by peripheral fat accumulation rather than increased visceral adiposity, is a common side effect of the TZDs and can be disheartening to patients with NASH. Preliminary studies with pioglitazone and rosiglitazone indicate that both agents improved liver enzymes and the histological features of NASH.227,238,239 However, 1 patient treated with rosiglitazone experienced a precipitous rise in aminotransferases that correlated with the concomitant use of corticosteroids.158 This observation underscores the need to examine the utility of these agents within the structure of clinical trials. Whether these agents improve NASH by improving insulin sensitivity or through a recently recognized anti-inflammatory mechanism is unknown. Studies of human240 and mouse241 monocytes and macrophages have demonstrated down-regulation of cytokine production in response to PPAR-g ligands. The clinical utility of PPARg ligands as anti-inflammatory agents was further supported by their beneficial effect in patients with ulcerative colitis. The TZDs may also be beneficial because of a direct antifibrotic effect. Experimental evidence has demonstrated the ability of PPAR-g ligands to inhibit collagen production by hepatic stellate cells, the cell type responsible for abnormal fibrogenesis in the liver.242,243
Metformin Metformin, the only clinically available biguanide, is increasingly used to normalize hepatic gluconeogenesis and improve insulin sensitivity in patients with type 2 diabetes.244 It activates hepatic AMPactivated protein kinase (AMPK), shifting metabolism of free fatty acids away from esterification into triglyceride and towards mitochondrial b-oxidation.245 It also decreases hepatocellular SREBP-1 expression which down-regulates the hepatic synthesis of fatty acids. AMPK may also be important in the beneficial effect of metformin of increasing skeletal muscle glucose uptake.246 How these beneficial metabolic changes in diabetes might be helpful in NASH is uncertain. One pilot study of 20 patients with NASH treated with 1.5 gram metformin daily for 4 months found improved liver enzymes, insulin sensitivity, and liver volume, but the effect on histopathology was not determined.247 Metformin improved histopathologic findings of NAFLD in the leptin-deficient mouse,248 yet it did not reduce liver fat content in a study of diabetics.237 Like the TZDs, metformin has been established as a good first-line therapy for patients with diabetes, but both need further evaluation in clinical trials as treatments for NASH.
Chapter 55 NASH
PPARa LIGANDS
Betaine
Unlike the PPAR-g nuclear receptor ligands which promote adipocyte differentiation, PPARa ligands promote the disposal of fat through metabolic oxidation in the liver and other tissues.236 Mouse models of NAFLD have shown that PPARa ligands can prevent fat accumulation53,54 and that genetically modified mice lacking PPARa are predisposed to NAFLD.53 The clinically available PPARa ligands, the fibrates, have been examined in small studies of patients with NASH. Clofibrate was found to have no effect on liver enzymes or histopathology in a pilot study of 16 patients.212 A 4-week treatment of 23 patients with gemfibrozil demonstrated improved liver enzymes but the effects on histopathology were not ascertained.249 Limited experience with bezafibrate, a PPARa ligand not available in the USA, suggests that it might reduce the NASH associated with tamoxifen use.138,140
In a pilot study of patients with NASH, administration of betaine improved both the biochemical and histological abnormalities of NASH.254 Betaine, pronounced either bee-tane or beta-een, was first isolated from extracts of the common sugar beet (Beta vulgaris) and synthesized well over a century ago. Chemically, betaine is simply trimethylglycine. Endogenous betaine, once considered an inert metabolic byproduct, is derived from oxidation of choline or from dietary intake. Studies in the 1940s established that betaine can prevent lipid accumulation in the liver, and as such it was identified as a lipotropic agent. Betaine can substitute for choline and function as an effective dietary methyl donor when dietary methionine and choline are inadequate to meet metabolic needs. A major function of choline is to serve as a substrate for the production of phosphatidyl choline (lecithin), a component of cell membranes and an obligate component of VLDL. Because VLDL synthesis and export are the only mechanism by which fat can be exported from hepatocytes, insufficient phosphatidyl choline can prevent the synthesis of functional VLDL in the liver.255 Betaine also functions as a methyl donor for the conversion of homocysteine to methionine and thus alleviates defects in the methylation pathway caused by deficiencies of the folate cycle and vitamin B12. Homocysteine accumulation can impair insulin signaling,256 suggesting that betaine-mediated metabolism of homocysteine could improve insulin sensitivity, although this has not been examined experimentally.
STATINS AND OTHER HYPOLIPIDEMIC AGENTS Because NAFLD is associated with hyperlipidemia, one therapeutic approach has been to treat the hyperlipidemia. This approach has met with little success and the published experience fails to record the results of negative pilot studies, an unfortunate but common phenomenon.250 The hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors, or statins, were developed to impair cholesterol synthesis, reduce circulating cholesterol levels, and reduce the risk of vascular disease and its complications. Based on the original mechanism of action, there is little theoretical rationale for using the statins as therapeutic agents to treat NAFLD. However, newly recognized but important anti-inflammatory effects of these drugs have rekindled interest in their use in liver disease. The hypolipidemic agent probucol has antioxidant properties that could be beneficial in NASH. An open-label pilot study with probucol in 17 patients led to improved liver enzymes but liver biopsies were not examined.251
OTHER DRUGS Vitamin E If oxidant stress is a requisite pathophysiological process in the development of NASH, then antioxidants should prove to be effective therapies. Unfortunately, this has not been the case. Admittedly, the absence of data demonstrating a benefit of antioxidants could be due to the lack of available and effective pharmacological antioxidants or alternatively because of insufficiently powered studies to identify a benefit. On the other hand, the detection of oxidant stress by various means in patients with NASH may be documenting an epiphenomenon unrelated to the pathophysiology of the disease. One pilot study did show that vitamin E at daily doses ranging from 400 to 1200 IU improved liver enzymes in children and may also have improved histopathological changes in some.252 However a small pilot study in adults showed that vitamin E supplementation did not offer significant benefits beyond those achieved with exercise.120 One study in adults suggested that a combination of vitamin E and vitamin C was beneficial, yet the control group demonstrated a similar benefit.253
Ursodeoxycholic Acid UDCA is a minor component of human bile and its pharmacologic supplementation replaces more toxic bile acids within the bile acid pool. Because of its potential to reduce injury in the liver, UDCA has been investigated as a therapy for NASH. A widely quoted but uncontrolled pilot study of 24 patients treated for 12 months showed improvements in the serum ALT level and a small improvement in the degree of steatosis.212 A subsequent double-blind placebo-controlled study for 2 years in 126 adults treated with UDCA 13–15 mg/kg found that treatment was not different from placebo (Figure 55-13).213 This well-conducted clinical trial not only demonstrated the inefficacy of UDCA in NASH, it also highlights the need to include placebo-controlled treatment groups in studies of NASH because subtle lifestyle changes that may accompany the diagnosis and subsequent management could lead to improvement in the control group. Additional negative data with UDCA include a trial in 31 children with NASH that showed weight loss was effective in normalizing enzymes and sonographically detected NAFLD whereas UDCA was not.257
LIVER TRANSPLANTATION When liver disease caused by NAFLD progresses to decompensated cirrhosis, liver transplantation can be an option. About 2% of transplants are performed for a known diagnosis of NASH and a much larger fraction are performed for cryptogenic cirrhosis, a condition that is mostly caused by NASH, as described below. One of the biggest challenges often facing these patients and the transplant team coordinating their care is the presence of significant obesity.
1053
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
60 UDCA Percentage of subjects
Placebo 40
20
0 –2
–1
0
1
2
Steatosis change Figure 55-13. Ursodeoxycholic acid (UDCA) as a treatment for non-alcoholic steatohepatitis (NASH). UDCA, 13–15 mg/kg, given for 24 months did not cause significant improvement in hepatic steatosis (shown above), necroinflammatory changes, or serum alanine aminotransferase (ALT) levels compared to controls. Although significant improvement was found in each of these parameters when compared to baseline values, the improvement was the same in UDCA and placebo-treated groups. (Data from Lindor KD, Kowdley KV, Heathcote EJ, et al. Ursodeoxycholic acid for treatment of nonalcoholic steatohepatitis: results of a randomized trial. Hepatology 2004; 39:770–778.)
Also, because NASH develops in the setting of insulin resistance, the underlying metabolic abnormality most likely remains after transplantation. Recurrent NASH after transplant occurs in patients with prior diagnoses of both NASH and cryptogenic cirrhosis.258
SPECIAL ISSUES USE OF HMG-COA REDUCTASE INHIBITORS (“STATINS”) IN PATIENTS WITH NASH As a class, the HMG-CoA reductase inhibitors are known to cause occasional increases in the serum aminotransferase levels and the manufacturers of these drugs warn against their use in patients with known liver disease. This warning appears to be based on hypothetical concerns, as there are no data to suggest that patients with pre-existing chronic liver disease are any more susceptible to increased liver enzymes or serious liver injury than patients with normal enzymes. In fact, a survey of one large university practice found that patients with elevated liver enzymes before starting a statin may have experienced overall improvement in their liver enzymes.259 Therefore, a reasonable approach is to use this class of drugs as they are intended to treat hypercholesterolemia with appropriate routine monitoring of liver enzymes as recommended in patients with normal liver enzymes.
TAMOXIFEN USE The use of the antiestrogen agent tamoxifen after resection of estrogen receptor-positive breast cancer provides a clear survival benefit. Reports of women found to have NAFLD, NASH, and cirrhosis while taking tamoxifen have raised the question of whether tamoxifen use is causally implicated or whether the liver disease may have
1054
pre-existed but was unrecognized before treatment. Early reports established the presence of significant NAFLD in patients treated with tamoxifen,135–137,260–262 and subsequent reports showed that, at least in some patients, NAFLD developed while being treated with tamoxifen and thus provide evidence for causality.134,139,140,263 The approach to a patient with NASH while taking tamoxifen is challenging. Some studies have shown that overweight and obesity are risk factors for the development of NAFLD after starting tamoxifen.138–140 This is important because it suggests that improving body weight might diminish the risk of developing NAFLD or possibly lead to improvement after it has developed. Alternative antiestrogen agents might also be considered. However, one study found that the use of an alternative antiestrogen agent, toremifene, was also associated with the development of NASH, suggesting that this approach may be of little benefit.144 Additional studies are needed of other antiestrogen agents and also the fibrates as concomitant treatment to prevent NASH in this setting.138,140 The best advice is to approach each situation on a case-by-case basis, weighing the potential risks and benefits of discontinuing tamoxifen. Available data suggest that overweight patients should be counseled on the benefits of weight reduction.
HEPATITIS A VACCINATION Severe hepatitis A is rare, but pre-existing liver disease may increase the likelihood of developing acute liver failure leading to death or liver transplant after hepatitis A infection. For this reason, the Centers for Disease Control and the Advisory Committee on Immunization Practices recommend that all patients with chronic liver disease should be vaccinated against hepatitis A. The vaccine, given as two injections 6 months apart, is highly effective and well tolerated. Most patients with NAFLD have not been vaccinated for hepatitis A and it is reasonable to consider recommending the vaccine to patients when a diagnosis of NAFLD or any other form of chronic liver disease is established.
HOW MUCH ALCOHOL IS ALLOWABLE? No uniformly accepted recommendation for the allowable amount of alcohol ingestion exists for patients with chronic liver disease, with the exception of abstinence for patients with alcoholic liver disease. Most clinicians recommend allowable alcohol consumption that ranges from complete abstinence to one drink (10–20 grams ethanol; see Table 55-9) weekly. The Italian Dionysos population study did not find adverse consequences of consumption < 30 grams daily, but few clinicians are comfortable making recommendations to patients allowing this much regular alcohol consumption. On the other hand, there are no data to indicate that the occasional (less frequent than weekly) single drink poses a danger to the liver. Decisions must be made based on the patient’s desires, expectations, and severity of underlying liver disease.
PROGNOSIS AND NATURAL HISTORY Accurately describing the natural history of NAFLD is difficult because the existing data are compromised by variable definitions of
Chapter 55 NASH
NAFLD and NASH, the possible unrecognized coexistence of HCV in earlier studies, the prolonged duration of study needed to identify progression, and probably a failure to recognize cirrhosis as a contributor to death from the known complications of cardiovascular disease.
PROGRESSION TO CIRRHOSIS The prognosis of NAFLD is directly related to its likelihood to progress to cirrhosis. Early studies of small cohorts followed for relatively short durations suggested that NASH was a relatively benign disease. Despite more recent data indicating that NASH may have become the most common form of chronic liver disease leading to cirrhosis, HCC, and death,25,264 a nihilistic diagnostic and therapeutic approach to these patients persists. The absence of hard figures to describe the prevalence of cirrhosis caused by NASH undoubtedly contributes to a non-aggressive diagnostic approach. However, as awareness increases, more patients may be recognized as having significant silent liver disease while being treated for the complications of insulin resistance.
A
CRYPTOGENIC CIRRHOSIS If NASH now has a prevalence of 2–3% of the adult population and a significant proportion of these patients will have progressive fibrosis and be at risk for cirrhosis, then a key question is why more patients succumbing from NASH cirrhosis have not been identified in death or transplant statistics. One factor may be that cirrhosis has been a silent and unrecognized contributor to death from obesity, cardiovascular disease, and diabetes.265 The presence of such silent cirrhosis has been well documented in morbidly obese patients undergoing bariatric surgery.266 The second factor contributing to a lack of accounting of NASH as a cause of cirrhosis is the curious loss of the typical histological features of NASH after cirrhosis develops. When this happens, a liver biopsy may reveal relatively bland cirrhosis and, because of the absence of markers for other liver disease, the cirrhosis is classified as cryptogenic (Figure 55-14).32,33 The recognition that “burned-out” NASH may explain many patients with cryptogenic cirrhosis is based on the following two observations: (1) diabetes and obesity are overrepresented in patients with cryptogenic cirrhosis;267 and (2) NASH often occurs after liver transplantation for cryptogenic cirrhosis as it does in patients with known NASH.268–270 About 12% of liver transplants are currently performed for a diagnosis of cryptogenic cirrhosis. If most of these patients had NASH as a cause of cirrhosis, then these patients plus the 2% transplanted for known NASH add up to 12–14% of liver transplants being performed for cirrhosis caused by NASH. This percentage is unlikely to represent the total fraction of the population with cirrhosis caused by NASH. Because end-stage liver disease caused by NASH is often associated with complicated diabetes or massive obesity, many of these patients will not be candidates for liver transplant and succumb to the complications of their liver disease or other manifestations of insulin resistance.
HEPATOCELLULAR CARCINOMA Once considered primarily a complication of chronic hepatitis B and C and a few less common forms of liver disease, hepatocellular car-
B
Figure 55-14. Evolution of non-alcoholic steatohepatitis (NASH) to cirrhosis. (A) A liver biopsy performed to evaluate elevated liver enzymes showed necroinflammatory changes, hepatocellular ballooning, and some hepatocytes containing Mallory’s hyaline, findings diagnostic of NASH (top panel, 40¥). (B) Twelve years later, a second biopsy showed none of the necroinflammatory changes of NASH but showed regenerative nodules among dense bands of fibrosis consistent with cirrhosis. Without the prior biopsy, this case would fulfill criteria for “cryptogenic” cirrhosis, and demonstrates the loss of active lesions of steatohepatitis with progression to cirrhosis. (Courtesy of E.M. Brunt.)
cinoma (HCC, Figure 55-15) is now recognized to be a complication of cirrhosis from any cause. The development of HCC in patients with cirrhosis caused by NASH occurs with the same frequency as in patients with chronic hepatitis C.271–275 Additionally, evidence suggests that the surrogates of NAFLD, obesity and diabetes,272,274 are associated with an increased risk of HCC and synergistically contribute to the risk of HCC in patients with coexisting viral hepatitis.276 Because of the comparatively high prevalence of HCV and NAFLD compared to other forms of chronic liver disease, these two diseases are now the most common underlying causes of HCC.273 As reviewed above, most patients diagnosed with cryptogenic cirrhosis had prior NASH as the cause of their liver disease. Given this fact, HCC should be associated with cryptogenic cirrhosis just as it is with NASH cirrhosis. Indeed, one survey of patients with HCC identified it as a complication of cryptogenic cirrhosis and, in other studies, patients with cryptogenic cirrhosis had risk factors for prior NASH as a cause of their liver disease.277
1055
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Figure 55-15. Hepatocellular carcinoma developing in a patient with non-alcoholic steatohepatitis (NASH) cirrhosis. An obese non-alcoholic 58-yearold male presented with ascites and a liver biopsy demonstrated cirrhosis with steatohepatitis. One year later, a surveillance computed tomography scan of the liver demonstrated a 6-cm mass in the right lobe. The patient subsequently underwent transarterial chemoembolization; the lipiodolenhanced image used to direct therapy is shown. (Courtesy of A.M. Di Bisceglie.)
Table 55-12. Important Areas for Future Research
• • • • •
Identify the pathogenesis of hepatocellular injury in patients with NAFLD that leads to the development of NASH Identify non-invasive predictors of NASH in patients with NAFLD or elevated liver enzymes Identify non-invasive predictors of fibrosis (e.g., demographics, comorbidities, blood tests, genomic polymorphism analysis, imaging) Define the spectrum and prevalence of liver disease associated with insulin resistance Determine the role of exercise, weight loss, dietary changes, antioxidants, insulin-sensitizing agents and antifibrotic agents in the prevention and treatment of NASH
NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis.
CONCLUSIONS More than two decades of active research into the causes and consequences of NAFLD have advanced the field significantly. NAFLD is now recognized as a remarkably common disorder, affecting 1 in 5 adults. NASH, which was once thought to be an uncommon disorder affecting obese diabetic women, is now recognized as a relatively common cause of liver-related morbidity. It is associated with insulin resistance, affects both genders, and is a cause of significant liver disease before diabetes develops. Much work is still needed to identify the causes of NASH at the molecular level, establish better diagnostic tools, and identify effec-
1056
tive therapeutic strategies. Major areas needing active investigation that will build on the current foundation are listed in Table 55-12.20
REFERENCES 1. Ludwig J, Viggiano TR, McGill DB, Ott BJ. Nonalcoholic steatohepatitis. Mayo Clinic Proc 1980; 55:434–438. 2. Zelman S. The liver in obesity. Arch Intern Med 1952; 90:141–156. 3. Leevy CM. Fatty liver: a study of 270 patients with biopsy proven fatty liver and a review of the literature. Medicine 1962; 41:249–276. 4. Brunt EM. Nonalcoholic steatohepatitis: definition and pathology. Semin Liver Dis 2001; 21:3–16. 5. Matteoni CA, Younossi ZM, Gramlich T, et al. Nonalcoholic fatty liver disease: a spectrum of clinical and pathological severity. Gastroenterology 1999; 116:1413–1419. 6. Sanyal A. AGA technical review on nonalcoholic fatty liver disease. Gastroenterology 2002; 1213:1705–1725. 7. Clark JM, Diehl AM. Defining nonalcoholic fatty liver disease: implications for epidemiologic studies. Gastroenterology 2003; 124:248–250. 8. Poynard T, Ratziu V, McHutchison J, et al. Effect of treatment with peginterferon or interferon alfa-2b and ribavirin on steatosis in patients infected with hepatitis C. Hepatology 2003; 38:75–85. 9. Brunt EM, Ramrahkiani S, Cordes BG, et al. Concurrence of histologic features of steatohepatitis with other forms of chronic liver disease. Mod Pathol 2003; 16:49–56. 10. Bellentani S, Tiribelli C. The spectrum of liver disease in the general population: lesson from the Dionysos study. J Hepatol 2001; 35:531–537.
Chapter 55 NASH
11. Bacon BR, Farahvash MJ, Janney CG, Neuschwander-Tetri BA. Nonalcoholic steatohepatitis: an expanded clinical entity. Gastroenterology 1994; 107:1103–1109. 12. Wanless IR, Lentz JS. Fatty liver hepatitis (steatohepatitis) and obesity: an autopsy study with analysis of risk factors. Hepatology 1990; 12:1106–1110. 13. Luyckx FH, Lefebvre PJ, Scheen AJ. Non-alcoholic steatohepatitis: association with obesity and insulin resistance, and influence of weight loss. Diabetes Metab 2000; 26:98–106. 14. Willner IR, Waters B, Patil SR, et al. Ninety patients with nonalcoholic steatohepatitis: insulin resistance, familial tendency, and severity of disease. Am J Gastro 2001; 96:2957–2961. 15. Dixon JB, Bhathal PS, O’Brien PE. Nonalcoholic fatty liver disease: predictors of nonalcoholic steatohepatitis and liver fibrosis in the severely obese. Gastroenterology 2001; 121:91–100. 16. Sanyal AJ, Campbell-Sargent C, Mirshani F, et al. Nonalcoholic steatohepatitis: association of insulin resistance and mitochondrial abnormalities. Gastroenterology 2001; 120:1183–1192. 17. Chitturi S, Abeygunasekera S, Farrell GC, et al. NASH and insulin resistance: insulin hypersecretion and specific association with the insulin resistance syndrome. Hepatology 2002; 35:373–379. 18. Pagano G, Pacini G, Musso G, et al. Nonalcoholic steatohepatitis, insulin resistance, and metabolic syndrome: further evidence for an etiologic association. Hepatology 2002; 35:367–372. 19. Chalasani N, Deeg MA, Persohn S, Crabb DW. Metabolic and anthropometric evaluation of insulin resistance in nondiabetic patients with nonalcoholic steatohepatitis. Am J Gastro 2003; 98:1849–1855. 20. Neuschwander-Tetri BA, Caldwell SH. Nonalcoholic steatohepatitis: summary of an AASLD single topic conference. Hepatology 2003; 37:1202–1219. 21. Kral JG, Buckley MC, Kissileff HR, Schaffner F. Metabolic correlates of eating behavior in severe obesity. Int J Obesity 2001; 25:258–264. 22. Ruhl CE, Everhart JE. Determinants of the association of overweight with elevated serum alanine aminotransferase activity in the United States. Gastroenterology 2003; 124:71–79. 23. Malnick SD, Beergabel M, Knobler H. Non-alcoholic fatty liver: a common manifestation of a metabolic disorder. Q J Med 2003; 96:699–709. 24. Hilden M, Christoffersen P, Juhl E, Dalgaard JB. Liver histology in a ‘normal’ population: examination of 503 consecutive fatal traffic casualties. Scand J Gastro 1977; 12:593–597. 25. Falck-Ytter Y, Younossi ZM, Marchesini G, McCullough AJ. Clinical features and natural history of nonalcoholic steatosis syndromes. Semin Liver Dis 2001; 21:17–26. 26. Lavine JE, Schwimmer JB. Nonalcoholic fatty liver disease in the pediatric population. In Sanyal AJ, ed. Clinical liver disease, vol. 8. Philadelphia: Elsevier, 2004:549–558. 27. Tominaga K, Kurata JH, Chen YK, et al. Prevalence of fatty liver in Japanese children and relationship to obesity. An epidemiological ultrasonographic survey. Dig Dis Sci 1995; 40:2002–2009. 28. Ruhl CE, Everhart JE. Epidemiology of nonalcoholic fatty liver. In Sanyal A, ed. Clinical liver disease, vol. 8. Philadelphia: Elsevier, 2004:501–519. 29. Harrison SA, Torgerson S, Hayashi PH. The natural history of nonalcoholic fatty liver disease: a clinical histopathological study. Am J Gastro 2003; 98:2042–2047. 30. Caldwell SH, Harris DM, Patrie JT, Hespenheide EE. Is NASH underdiagnosed among African Americans? Am J Gastro 2002; 97:1496–1500.
31. Browning JD, Kumar KS, Saboorian MH, Thiele DL. Ethnic differences in the prevalence of cryptogenic cirrhosis. Am J Gastro 2004; 99:292–298. 32. Clark JM, Diehl AM. Nonalcoholic fatty liver disease: an underrecognized cause of cryptogenic cirrhosis. JAMA 2003; 289:3000–3004. 33. Caldwell S, Crespo DM. The spectrum expanded: cryptogenic cirrhosis and the natural history of non-alcoholic fatty liver disease. J Hepatol 2004; 40:578–584. 34. Fromenty B, Pessayre D. Inhibition of mitochondrial betaoxidation as a mechanism of hepatotoxicity. Pharmacol Ther 1995; 67:101–154. 35. Rubin E, Lieber CS. Alcohol-induced hepatic injury in nonalcoholic volunteers. N Engl J Med 1968; 278:869–876. 36. Charlton M, Sreekumar R, Rasmussen D, et al. Apolipoprotein synthesis in nonalcoholic steatohepatitis. Hepatology 2002; 35:898–904. 37. Fisher EA, Ginsberg HN. Complexity in the secretory pathway: the assembly and secretion of apolipoprotein B-containing lipoproteins. J Biol Chem 2002; 277:17377–17380. 38. Mensenkamp AR, van Luyn MJA, Havinga R, et al. The transport of triglycerides through the secretory pathway of hepatocytes is impaired in apolipoprotein E deficient mice. J Hepatol 2004; 40:599–699. 39. Leung GK, Veniant MM, Kim SK, et al. A deficiency of microsomal triglyceride transfer protein reduces apolipoprotein B secretion. J Biol Chem 2000; 275:7515–7520. 40. Bernard S, Touzet S, Personne I, et al. Association between microsomal triglyceride transfer protein gene polymorphism and the biological features of liver steatosis in patients with type II diabetes. Diabetologia 2000; 43:995–999. 41. Bjorkegren J, Beigneux A, Bergo MO, et al. Blocking the secretion of hepatic very low density lipoproteins renders the liver more susceptible to toxin-induced injury. J Biol Chem 2002; 277:5476–5483. 42. Wishingrad M, Paaso B, Garcia G. Fatty liver due to heterozygous hypobetalipoproteinemia. Am J Gastro 1994; 89:1106–1107. 43. Tarugi P, Lonardo A, Ballarini G, et al. Fatty liver in heterozygous hypobetalipoproteinemia caused by a novel truncated form of apolipoprotein B. Gastroenterology 1996; 111:1125–1133. 44. Berson A, De Beco V, Lettéron P, et al. Steatohepatitis-inducing drugs cause mitochondrial dysfunction and lipid peroxidation in rat hepatocytes. Gastroenterology 1998; 114:764–774. 45. Caldwell SH, Swerdlow RH, Khan EM, et al. Mitochondrial abnormalities in non-alcoholic steatohepatitis. J Hepatol 1999; 31:430–434. 46. Caldwell SH, Hespenheide EE, Redick JA, et al. A pilot study of a thiazolidinedione, troglitazone, in nonalcoholic steatohepatitis. Am J Gastro 2001; 96:519–525. 47. Pérez-Carreras M, Del Hoyo P, Martín MA, et al. Defective hepatic mitochondrial respiratory chain in patients with nonalcoholic steatohepatitis. Hepatology 2003; 38:999–1007. 48. Rao MS, Reddy JK. Peroxisomal beta-oxidation and steatohepatitis. Semin Liver Dis 2001; 21:43–55. 49. Reddy JK. Nonalcoholic steatosis and steatohepatitis. III. Peroxisomal b-oxidation, PPARa, and steatohepatitis. Am J Physiol 2001; 281:G1333–G1339. 50. Fan CY, Pan J, Usuda N, et al. Steatohepatitis, spontaneous peroxisome proliferation and liver tumors in mice lacking peroxisomal fatty acyl-CoA oxidase. Implications for peroxisome proliferator-activated receptor alpha natural ligand metabolism. J Biol Chem 1998; 273:15639–15645. 51. Bass NM. Three for the price of one knockout – a mouse model of a congenital peroxisomal disorder, steatohepatitis, and hepatocarcinogenesis. Hepatology 1999; 29:606–608.
1057
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
52. Kim BH, Won YS, Kim EY, et al. Phenotype of peroxisome proliferator-activated receptor-alpha(PPARalpha)deficient mice on mixed background fed high fat diet. J Vet Sci 2003; 4:239–244. 53. Ip E, Farrell GC, Robertson G, et al. Central role of PPARadependent hepatic lipid turnover in dietary steatohepatitis in mice. Hepatology 2003; 38:123–132. 54. Chou CJ, Haluzik M, Gregory C, et al. WY14,643, a peroxisome proliferator-activated receptor a (PPARa) agonist, improves hepatic and muscle steatosis and reverses insulin resistance in lipoatrophic A-ZIP/F-1 mice. J Biol Chem 2002; 277:24484–24489. 55. Neuschwander-Tetri BA. A resistance movement in NASH. Am J Gastro 2001; 96:2813–2814. 56. Marchesini G, Bugianesi E, Forlani G, et al. Nonalcoholic fatty liver, steatohepatitis, and the metabolic syndrome. Hepatology 2003; 37:917–923. 57. Kelley DE, McKolanis TM, Hegazi RA, et al. Fatty liver in type 2 diabetes mellitus: relation to regional adiposity, fatty acids, and insulin resistance. Am J Physiol 285:E906–E916, 2003. 58. Marchesini G. Corrections. Hepatology 2003; 38:536. 59. Browning JD, Horton JD. Molecular mediators of hepatic steatosis and liver injury. J Clin Invest 2004; 114:147–152. 60. Fukukura Y, Fujiyoshi F, Inoue H, et al. Focal fatty infiltration in the posterior aspect of hepatic segment IV: relationship to pancreaticoduodenal venous drainage. Am J Gastro 2000; 95:3590–3595. 61. Schwimmer JB, Deutsch R, Rauch JB, et al. Obesity, insulin resistance, and other clinicopathological correlates of pediatric nonalcoholic fatty liver disease. J Pediatr 2003; 143:500–505. 62. Sinha R, Fisch G, Teague B, et al. Prevalence of impaired glucose tolerance among children and adolescents with marked obesity. N Engl J Med 2002; 346:802–810. 63. Weiss R, Dziura J, Burgert TS, et al. Obesity and the metabolic syndrome in children and adolescents. N Engl J Med 2004; 350:2362–2374. 64. Weiss R, Dufour S, Taksali SE, et al. Prediabetes in obese youth: a syndrome of impaired glucose tolerance, severe insulin resistance, and altered myocellular and abdominal fat partitioning. Lancet 2003; 362:951–957. 65. Kawasaki T, Hashimoto N, Kikuchi T, et al. The relationship between fatty liver and hyperinsulinemia in obese Japanese children. J Pediatr Gastroenterol Nutr 1997; 24:317–321. 66. Bhargava SK, Sachdev HS, Fall CHD, et al. Relation of serial changes in childhood body-mass index to impaired glucose tolerance in young adulthood. N Engl J Med 2004; 350:865–875. 67. Dietz WH. Overweight in childhood and adolescence. N Engl J Med 2004; 350:855–857. 68. Kim JK, Fillmore JJ, Chen Y, et al. Tissue-specific overexpression of lipoprotein lipase causes tissue-specific insulin resistance. Proc Natl Acad Sci 2001; 98:7522–7527. 69. Hotamisligil GS, Arner P, Caro JF, et al. Increased adipose tissue expression of tumor necrosis factor-a in human obesity and insulin resistance. J Clin Invest 1995; 95:2409–2415. 70. Uysal KT, Wiesbrock SM, Marino MW, Hotamisligil GS: Protection from obesity-induced insulin resistance in mice lacking TNF-a function. Nature 1997; 389:610–614. 71. Shimomura I, Matsuda M, Hammer RE, et al. Decreased IRS-2 and increased SREBP-1c lead to mixed insulin resistance and sensitivity in livers of lipodystrophic and ob/ob mice. Molecular Cell 2000; 6:77–86. 72. Kruszynska YT, Olefsky JM, Frias JP. Effect of obesity on susceptibility to fatty acid-induced peripheral tissue insulin resistance. Metab Clin Exp 2003; 52:233–238. 73. Kim SP, Ellmerer M, Van Citters GW, Bergman RN. Primacy of hepatic insulin resistance in the development of the metabolic
1058
74.
75.
76.
77.
78. 79.
80.
81.
82.
83.
84.
85. 86.
87.
88.
89.
90.
91.
92.
93. 94.
syndrome induced by an isocaloric moderate-fat diet in the dog. Diabetes 2003; 52:2453–2460. Rajala MW, Obici S, Scherer PE, Rossetti L. Adipose-derived resistin and gut-derived resistin-like molecule-b selectively impair insulin action on glucose production. J Clin Invest 2003; 111:225–230. Michael MD, Kulkarni RN, Postic C, et al. Loss of insulin signaling in hepatocytes leads to severe insulin resistance and progressive hepatic dysfunction. Mol Cell 2000; 6:87–97. Abel ED, Peroni O, Kim JK, et al. Adipose-selective targeting of the GLUT4 gene impairs insulin action in muscle and liver. Nature 2001; 409:729–733. Li Z, Yang S, Lin H, et al. Probiotics and antibodies to TNF inhibit inflammatory activity and improve nonalcoholic fatty liver disease. Hepatology 2003; 37:343–350. Hirosumi J, Tuncman G, Chang L, et al. A central role of JNK in obesity and insulin resistance. Nature 2002; 420:333–336. Marchesini G, Pacini G, Bianchi G, et al. Glucose disposal, bcell secretion, and hepatic insulin extraction in cirrhosis. A minimal model assessment. Gastroenterology 1990; 99:1715–1722. Kruszynska YT, Harry DS, Bergman RN, McIntyre N. Insulin sensitivity, insulin secretion and glucose effectiveness in diabetic and non-diabetic cirrhotic patients. Diabetologia 1993; 36:121–128. Guillygomarc’h A, Mendler MH, Moirand R, et al. Venesection therapy of insulin resistance-associated hepatic iron overload. J Hepatol 2001; 35:344–349. Facchini FS, Hua NW, Stoohs RA. Effect of iron depletion in carbohydrate-intolerant patients with clinical evidence of nonalcoholic fatty liver disease. Gastroenterology 2002; 122:931–939. Vigano M, Vergani A, Trombini P, et al. Insulin resistance influence iron metabolism and hepatic steatosis in type II diabetes. Gastroenterology 2000; 118:986–987. Mendler M-H, Turlin B, Moirand R, et al. Insulin resistanceassociated hepatic iron overload. Gastroenterology 1999; 117:1155–1163. Ferrannini E. Insulin resistance, iron, and the liver. Lancet 2000; 355:2181–2182. Fernández-Real JM, López-Bermajo A, Ricart W. Cross-talk between iron metabolism and diabetes. Diabetes 2002; 51:2348–2354. Garg A, Misra A. Hepatic steatosis, insulin resistance, and adipose tissue disorders. J Clin Endo Metab 2002; 87:3019–3022. Agarwal AK, Arioglu E, De Almeida S, et al. AGPAT2 is mutated in congenital generalized lipodystrophy linked to chromosome 9q34. Nat Genet 2002; 31:21–23. Agarwal AK, Garg A. A novel heterozygous mutation in peroxisome proliferator-activated receptor-g gene in a patient with familial partial lipodystrophy. J Clin Endo Metab 2002; 87:408–411. Shackleton S, Lloyd DJ, Jackson SN, et al. LMNA, encoding lamin A/C, is mutated in partial lipodystrophy. Nat Genet 2000; 24:153–156. Shimano H, Horton JD, Hammer RE, et al. Overproduction of cholesterol and fatty acids causes massive liver enlargement in transgenic mice expressing truncated SREBP-1a. J Clin Invest 1996; 98:1575–1584. Yahagi N, Shimano H, Hasty AH, et al. Absence of sterol regulatory element-binding protein-1 (SREBP-1) ameliorates fatty livers but not obesity or insulin resistance in Lep(ob)/Lep(ob) mice. J Biol Chem 2002; 277:19353–19357. Day CP, James OF. Hepatic steatosis: innocent bystander or guilty party? Hepatology 1998; 27:1463–1466. Thaler H. Relation of steatosis to cirrhosis. Clin Gastroenterol 1975; 4:273–280.
Chapter 55 NASH
95. Day CP, James OF. Steatohepatitis: a tale of two “hits”? Gastroenterology 1998; 114:842–845. 96. Shavila J, King LJ, Parke DV. Spontaneous development of fatty liver in ferrets in a toxicology study. Toxicology 1996; 112:105–116. 97. Leclercq IA, Farrell GC, Field J, et al. CYP2E1 and CYP4A as microsomal catalysts of lipid peroxides in murine nonalcoholic steatohepatitis. J Clin Invest 2000; 105:1067–1075. 98. George J, Pera N, Phung N, et al. Lipid peroxidation, stellate cell activation and hepatic fibrogenesis in a rat model of chronic steatohepatitis. J Hepatol 2003; 39:756–764. 99. Hasegawa T, Yoneda M, Nakamura K, et al. Plasma transforming growth factor-b1 level and efficacy of a-tocopherol in patients with non-alcoholic steatohepatitis: a pilot study. Aliment Pharmacol Ther 2001; 15:1667–1672. 100. Robertson G, Leclercq I, Farrell GC. Nonalcoholic steatosis and steatohepatitis. II. Cytochrome P-450 enzymes and oxidative stress. Am J Physiol 2001; 281:G1135–1139. 101. Caro AA, Cederbaum AI. Oxidative stress, toxicology, and pharmacology of CYP2E1. Annu Rev Pharmacol Toxicol 2004; 44:27–42. 102. Niemelä O, Parkkila S, Juvonen RO, et al. Cytochromes P450 2A6, 2E1, and 3A and production of protein-aldehyde adducts in the liver of patients with alcoholic and non-alcoholic liver diseases. J Hepatol 2000; 33:893–901. 103. Chalasani N, Gorski JC, Asghar MS, et al. Hepatic cytochrome P450 2E1 activity in nondiabetic patients with nonalcoholic steatohepatitis. Hepatology 2003; 37:544–550. 104. Weltman MD, Farrell GC, Liddle C. Increased hepatocyte CYP2E1 expression in a rat nutritional model of hepatic steatosis with inflammation. Gastroenterology 1996; 111:1645–1653. 105. Emery MG, Fisher JM, Chien JY, et al. CYP2E1 activity before and after weight loss in morbidly obese subjects with nonalcoholic fatty liver disease. Hepatology 2003; 38:428–435. 106. Bugianesi E, Manzini P, D’Antico S, et al. Relative contribution of iron burden, HFE mutations, and insulin resistance to fibrosis in nonalcoholic fatty liver. Hepatology 2004; 39:179–187. 107. Bhattacharya R, Kowdley KV. Iron and HFE mutations in nonalcoholic steatohepatitis: innocent bystanders or accessories to the crime? Gastroenterology 2003; 125:615–616; discussion 617. 108. George DK, Goldwurm S, MacDonald GA, et al. Increased hepatic iron in nonalcoholic steatohepatitis is associated with increased fibrosis. Gastroenterology 1998; 114:311–318. 109. Chitturi S, Weltman M, Farrell GC, et al. HFE mutations, hepatic iron, and fibrosis: ethnic-specific association of NASH with C282Y but not with fibrotic severity. Hepatology 2002; 36:142–149. 110. Hsiao TJ, Chen JC, Wang JD. Insulin resistance and ferritin as major determinants of nonalcoholic fatty liver disease in apparently healthy obese patients. Int J Obesity Rel Metab Disord: J Int Assoc Study Obesity 2004; 28:167–172. 111. Pessayre D, Berson A, Fromenty B, Mansouri A. Mitochondria in steatohepatitis. Semin Liver Dis 2001; 21:57–69. 112. Miele L, Grieco A, Armuzzi A, et al. Hepatic mitochondrial beta-oxidation in patients with nonalcoholic steatohepatitis assessed by 13C-octanoate breath test. Am J Gastro 2003; 98:2335–2336. 113. Caldwell SH, Chang CY, Nakamoto RK, Krugner-Higby L. Mitochondrial abnormalities in fatty liver disease. In Sanyal A, ed. Clinics in liver disease, vol. 8. Philadelphia: Elsevier, 2004:595–617. 114. Cortez-Pinto H, Chatham J, Chacko VP, et al. Alterations in liver ATP homeostasis in human nonalcoholic steatohepatitis: a pilot study. JAMA 1999; 282:1659–1664.
115. Nair S, Chacko VP, Arnold C, Diehl AM. Hepatic ATP reserve and efficiency of replenishing: comparison between obese and nonobese normal individuals. Am J Gastro 2003; 98:466–470. 116. Petersen KF, Dufour S, Befroy D, et al. Impaired mitochondrial activity in the insulin-resistant offspring of patients with type 2 diabetes. N Engl J Med 2004; 350:664–671. 117. Petersen KF, Befroy D, Dufour S, et al. Mitochondrial dysfunction in the elderly: possible role in insulin resistance. Science 2003; 300:1140–1142. 118. Crespo J, Cayón A, Fernández-Gil P, et al. Gene expression of tumor necrosis factor a and TNF-receptors, p55 and p75, in nonalcoholic steatohepatitis patients. Hepatology 2001; 34:1158–1163. 119. Wigg AJ, Roberts-Thomson IC, Dymock RB, et al. The role of small intestinal bacterial overgrowth, intestinal permeability, endotoxaemia, and tumour necrosis factor a in the pathogenesis of non-alcoholic steatohepatitis. Gut 2001; 48:206–211. 120. Kugelmas M, Hill DB, Vivian B, et al. Cytokines and NASH: a pilot study of the effects of lifestyle modification and vitamin E. Hepatology 2003; 38:413–419. 121. Tilg H, Diehl AM. Cytokines in alcoholic and nonalcoholic steatohepatitis. N Engl J Med 2000; 343:1467–1476. 122. Memon RA, Grunfeld C, Feingold KR. TNF-alpha is not the cause of fatty liver disease in obese diabetic mice. Nat Med 2001; 7:2–3. 123. Yang S, Lin H, Diehl AM. Fatty liver vulnerability to endotoxininduced damage despite NF-kB induction and inhibited caspase 3 activation. Am J Physiol 2001; 281:G382–G392. 124. Bataller R, North KE, Brenner DA. Genetic polymorphisms and the progression of liver fibrosis: a critical appraisal. Hepatology 2003; 37:493–503. 125. Dixon JB, Bhathal PS, Jonsson JR, et al. Pro-fibrotic polymorphisms predictive of advanced liver fibrosis in the severely obese. J Hepatol 2003; 39:967–971. 126. Washington K, Wright K, Shyr Y, et al. Hepatic stellate cell activation in nonalcoholic steatohepatitis and fatty liver. Hum Pathol 2000; 31:822–828. 127. Cortez-Pinto H, Baptista A, Camilo ME, de Moura MC. Hepatic stellate cell activation occurs in nonalcoholic steatohepatitis. Hepato-Gastroenterol 2001; 48:87–90. 128. Paradis V, Perlemuter G, Bonvoust F, et al. High glucose and hyperinsulinemia stimulate connective tissue growth factor expression: a potential mechanism involved in progression to fibrosis in nonalcoholic steatohepatitis. Hepatology 2001; 34:738–744. 129. Latry P, Bioulac-Sage P, Echinard E, et al. Perisinusoidal fibrosis and basement membrane-like material in the livers of diabetic patients. Hum Pathol 1987; 18:775–780. 130. Marceau P, Biron S, Hould F-S, et al. Liver pathology and the metabolic syndrome X in severe obesity. J Clin Endo Metab 1999; 84:1513–1517. 131. Farrell GC. Drugs and steatohepatitis. Semin Liver Dis 2002; 22:185–194. 132. Lettéron P, Sutton A, Mansouri A, et al. Inhibition of microsomal triglyceride transfer protein: another mechanism for drug-induced steatosis in mice. Hepatology 2003; 38:133–140. 133. Lewis JH, Ranard RC, Caruso A, et al. Amiodarone hepatotoxicity: prevalence and clinicopathologic correlations among 104 patients. Hepatology 1989; 9:679–685. 134. Murata Y, Ogawa Y, Saibara T, et al. Unrecognized hepatic steatosis and non-alcoholic steatohepatitis in adjuvant tamoxifen for breast cancer patients. Oncol Rep 2000; 7:1299–1304. 135. Pinto HC, Baptista A, Camilo ME, et al. Tamoxifen-associated steatohepatitis – report of three cases. J Hepatol 1995; 23:95–97. 136. Cai Q, Bensen M, Greene R, Kirchner J. Tamoxifen-induced transient multifocal hepatic fatty infiltration. Am J Gastro 2000; 95:277–279.
1059
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
137. Ogawa Y, Murata Y, Nishioka A, et al. Tamoxifen-induced fatty liver in patients with breast cancer. Lancet 1998; 351:725. 138. Saibara T, Onishi S, Ogawa Y, et al. Bezafibrate for tamoxifeninduced non-alcoholic steatohepatitis. Lancet 1999; 353:1802. 139. Nguyen MC, Stewart RB, Banerji MA, et al. Relationships between tamoxifen use, liver fat and body fat distribution in women with breast cancer. Int J Obesity 2001; 25:296–298. 140. Nemoto Y, Saibara T, Ogawa Y, et al. Tamoxifen-induced nonalcoholic steatohepatitis in breast cancer patients treated with adjuvent tamoxifen. Intern Med 2002; 41:345–350. 141. Kanel KT, Wolmark N, Thompson PD. Delayed severe hypertriglyceridemia from tamoxifen (letter). N Engl J Med 1997; 337:281–282. 142. Jones ME, Thorburn AW, Britt KL, et al. Aromatase-deficient (ArKO) mice have a phenotype of increased adiposity. Proc Natl Acad Sci 2000; 97:12735–12740. 143. Maffei L, Murata Y, Rochira V, et al. Dysmetabolic syndrome in a man with a novel mutation of the aromatase gene: effects of testosterone, alendronate, and estradiol treatment. J Clin Endo Metab 2004; 89:61–70. 144. Hamada N, Ogawa Y, Saibara T, et al. Toremifene-induced fatty liver and NASH in breast cancer patients with breastconservation treatment. Int J Oncol 2000; 17:1119–1123. 145. Diehl AM, Goodman Z, Ishak KG. Alcohol-like liver disease in nonalcoholics. A clinical and histologic comparison with alcoholinduced liver injury. Gastroenterology 1988; 95:1056–1062. 146. Pinto HC, Baptista A, Camilo ME, et al. Nonalcoholic steatohepatitis. Clinicopathological comparison with alcoholic hepatitis in ambulatory and hospitalized patients. Dig Dis Sci 1996; 41:172–179. 147. Garg A. Acquired and inherited lipodystrophies. N Engl J Med 2004; 350:1220–1234. 148. Berasain C, Betés M, Panizo A, et al. Pathological and virological findings in patients with persistent hypertransaminasaemia of unknown aetiology. Gut 2000; 47:429–435. 149. Luyckx FH, Desaive C, Thiry A, et al. Liver abnormalities in severely obese subjects: effect of drastic weight loss after gastroplasty. Int J Obesity Rel Metab Disord 1998; 22:222–226. 150. García-Monzón C, Martín-Pérez E, Iacono OL, et al. Characterization of pathogenic and prognostic factors of nonalcoholic steatohepatitis associated with obesity. J Hepatol 2000; 33:716–724. 151. Ratziu V, Imbert-Bismut F, Messous D, Poynard T. The elusiveness of “normal” ALT in fatty liver. Hepatology 2004; 39:1172–1173. 152. Clark JM, Brancati FL, Diehl AM. The prevalence and etiology of elevated aminotransferase levels in the United States. Am J Gastro 2003; 98:960–967. 153. Prati D, Taioli E, Zanella A, et al. Updated definitions of healthy ranges for serum alanine aminotransferase levels. Ann Intern Med 2002; 137:1–9. 154. Kaplan MM. Alanine aminotransferase levels: what’s normal? Ann Intern Med 2002; 137:49–51. 155. Van Ness MM, Diehl AM. Is liver biopsy useful in the evaluation of patients with chronically elevated liver enzymes? Ann Intern Med 1989; 111:473–478. 156. Sorbi D, Boynton J, Lindor KD. The ratio of aspartate aminotransferase to alanine aminotransferase: potential value in differentiating nonalcoholic steatohepatitis from alcoholic liver disease. Am J Gastro 1999; 94:1018–1022. 157. Andersen T, Christoffersen P, Gluud C. The liver in consecutive patients with morbid obesity: a clinical, morphological, and biochemical study. Int J Obesity Rel Metab Disord 1984; 8:107–115. 158. Neuschwander-Tetri BA, Brunt EM, Wehmeier KR, et al. Interim results of a pilot study demonstrating the early effects of the PPAR-g ligand rosiglitazone on insulin sensitivity,
1060
159.
160. 161.
162.
163.
164.
165. 166. 167.
168.
169.
170.
171.
172. 173.
174.
175.
176.
177. 178.
179.
aminotransferases, hepatic steatosis and body weight in patients with non-alcoholic steatohepatitis. J Hepatol 2003; 38:434–440. Loria P, Lonardo A, Leonardi F, et al. Non-organ-specific autoantibodies in nonalcoholic fatty liver disease: prevalence and correlates. Dig Dis Sci 2003; 48:2173–2181. Bianchi L. Liver biopsy in elevated liver functions tests? An old question revisited. J Hepatol 2001; 35:290–294. Mathiesen UL, Franzén LE, Frydén A, et al. The clinical significance of slightly to moderately increased liver transaminase values in asymptomatic patients. Scand J Gastro 1999; 34:85–91. Pratt DS, Kaplan MM. Evaluation of abnormal liver-enzyme results in asymptomatic patients. N Engl J Med 2000; 342:1266–1271. Skelly MM, James PD, Ryder SD. Findings on liver biopsy to investigate abnormal liver function tests in the absence of diagnostic serology. J Hepatol 2001; 35:195–199. Sorbi D, McGill DB, Thistle JL, et al. An assessment of the role of liver biopsies in asymptomatic patients with chronic liver test abnormalities. Am J Gastro 2000; 95:3206–3210. Siegelman ES, Rosen MA. Imaging of hepatic steatosis. Semin Liver Dis 2001; 21:71–80. Mortele KJ, Ros PR. Imaging of diffuse liver disease. Semin Liver Dis 2001; 21:195–212. Saadeh S, Younossi ZM, Remer EM, et al. The utility of radiological imaging in nonalcoholic fatty liver disease. Gastroenterology 2002; 123:745–750. Jacobs JE, Birnbaum BA, Shapiro MA, et al. Diagnostic criteria for fatty infiltration of the liver on contrast-enhanced helical CT. Am J Roentgenol 1998; 171:659–664. Fishbein MH, Stevens WR. Rapid MRI using a modified Dixon technique: a non-invasive and effective method for detection and monitoring of fatty metamorphosis of the liver. Pediatr Radiol 2001; 31:806–809. Aubin B, Denys A, Lafortune M, et al. Focal sparing of liver parenchyma in steatosis: role of the gallbladder and its vessels. J Ultrasound Med 1995; 14:77–80. Soyer P, Devine N, Somveille E, et al. Hepatic pseudolesion around the falciform ligament: prevalence on CT examination. Abdom Imag 1996; 21:324–328. Wallace TM, Matthews DR. The assessment of insulin resistance in man. Diabet Med 2002; 19:527–534. Mather KJ, Hunt AE, Steinberg HO, et al. Repeatability characteristics of simple indices of insulin resistance: implications for research applications. J Clin Endo Metab 2001; 86:5457–5464. Katz A, Nambi SS, Mather K, et al. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endo Metab 2000; 85:2402–2410. Abbasi F, Reaven GM. Evaluation of the quantitative insulin sensitivity check index as an estimate of insulin sensitivity in humans. Metab Clin Exp 2002; 51:235–237. Ikai E, Ishizaki M, Suzuki Y, et al. Association between hepatic steatosis, insulin resistance and hyperinsulinaemia as related to hypertension in alcohol consumers and obese people. J Hum Hypertens 1995; 9:101–105. Angulo P, Keach JC, Batts KP, Lindor KD. Independent predictors of liver fibrosis in patients with nonalcoholic steatohepatitis. Hepatology 1999; 30:1356–1362. Sanyal AJ, Contos MJ, Sterling RK, et al. Nonalcoholic fatty liver disease in patients with hepatitis C is associated with features of the metabolic syndrome. Am J Gastro 2003; 98:2064–2071. Lonardo A, Adinolfi LE, Loria P, et al. Steatosis and hepatitis C virus: mechanisms and significance for hepatic and extrahepatic disease. Gastroenterology 2004; 126:586–597.
Chapter 55 NASH
180. Hourigan LF, Macdonald GA, Purdie D, et al. Fibrosis in chronic hepatitis C correlates significantly with body mass index and steatosis. Hepatology 1999; 29:1215–1219. 181. Monto A, Alonzo J, Watson JJ, et al. Steatosis in chronic hepatitis C: relative contributions of obesity, diabetes mellitus, and alcohol. Hepatology 2002; 36:729–736. 182. Colloredo G, Sonzogni A, Rubbia-Brandt L, Negro F. Hepatitis C virus genotype 1 associated with massive steatosis of the liver and hypo-b-lipoproteinemia (letter). J Hepatol 2004; 40:562–563. 183. Aytug S, Reich D, Sapiro LE, et al. Impaired IRS-1/PI3-kinase signaling in patients with HCV: a mechanism for increased prevalence of type 2 diabetes. Hepatology 2003; 38:1384– 1392. 184. Shintani Y, Fujie H, Miyoshi H, et al. Hepatitis C virus infection and diabetes: direct involvement of the virus in the development of insulin resistance. Gastroenterology 2004; 126:840–848. 185. Mihm S, Fayyazi A, Hartmann H, Ramadori G. Analysis of histopathological manifestations of chronic hepatitis C virus infection with respect to virus genotype. Hepatology 1997; 25:735–739. 186. Hofer H, Bankl HC, Wrba F, et al. Hepatocellular fat accumulation and low serum cholesterol in patients infected with HCV-3a. Am J Gastro 2002; 97:2880–2885. 187. Patton HM, Patel K, Behling C, et al. The impact of steatosis on disease progression and early and sustained treatment response in chronic hepatitis C patients. J Hepatol 2004; 40:484–490. 188. Petit JM, Benichou M, Duvillard L, et al. Hepatitis C virusassociated hypobetalipoproteinemia is correlated with plasma viral load, steatosis, and liver fibrosis. Am J Gastro 2003; 98:1150–1154. 189. Negro F. Hepatitis C virus and liver steatosis: when fat is not beautiful. J Hepatol 2004; 40:533–535. 190. Serfaty L, Andreani T, Giral P, et al. Hepatitis C virus induced hypobetalipoproteinemia: a possible mechanism for steatosis in chronic hepatitis C. J Hepatol 2001; 34:428–434. 191. Kumar D, Farrell GC, Fung C, George J. Hepatitis C virus genotype 3 is cytopathic to hepatocytes: reversal of hepatic steatosis after sustained therapeutic response. Hepatology 2002; 36:1266–1272. 192. Lai MM. Hepatitis C virus proteins: direct link to hepatic oxidative stress, steatosis, carcinogenesis and more. Gastroenterology 2002; 122:568–571. 193. Lerat H, Honda M, Beard MR, et al. Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus. Gastroenterology 2002; 122:352–365. 194. Perlemuter G, Sabile A, Lettéron P, et al. Hepatitis C virus core protein inhibits microsomal triglyceride transfer protein activity and very low density lipoprotein secretion: a model of viralrelated steatosis. FASEB J 2002; 16:185–194. 195. Shi ST, Polyak SJ, Tu H, et al. Hepatitis C virus NS5A colocalizes with the core protein on lipid droplets and interacts with apolipoproteins. Virology 2002; 292:198–210. 196. Okuda M, Li K, Beard MR, et al. Mitochondrial injury, oxidative stress, and antioxidant gene expression are induced by hepatitis C virus core protein. Gastroenterology 2002; 122:366–375. 197. Clouston AD, Jonsson JR, Purdie DM, et al. Steatosis and chronic hepatitis C: analysis of fibrosis and stellate cell activation. J Hepatol 2001; 34:314–320. 198. Adinolfi LE, Gambardella M, Andreana A, et al. Steatosis accelerates the progression of liver damage of chronic hepatitis C patients and correlates with specific HCV genotype and visceral obesity. Hepatology 2001; 33:1358–1364. 199. Castéra L, Hézode C, Roudot-Thoraval F, et al. Worsening of steatosis is an independent factor of fibrosis progression in untreated patients with chronic hepatitis C and paired liver biopsies. Gut 2003; 52:288–292.
200. Hu K-Q, Kyulo NL, Esrailian E, et al. Overweight and obesity, hepatic steatosis, and progression of chronic hepatitis C: a retrospective study on a large cohort of patients in the United States. J Hepatol 2004; 40:147–154. 201. Ohata K, Hamasaki K, Toriyama K, et al. Hepatic steatosis is a risk factor for hepatocellular carcinoma in patients with chronic hepatitis C virus infection. Cancer 2003; 97:3036–3043. 202. Bressler BL, Guindi M, Tomlinson G, Heathcote EJ. High body mass index is an independent risk factor for nonresponse to antiviral treatment in chronic hepatitis C. Hepatology 2003; 38:639–644. 203. McCullough A. Obesity and its nurturing effect on hepatitis C. Hepatology 2003; 38:557–559. 204. Hickman IJ, Powell EE, Prins JB, et al. In overweight patients with chronic hepatitis C, circulating insulin is associated with hepatic fibrosis: implications for therapy. J Hepatol 2003; 39. 205. Gochee PA, Jonsson JR, Clouston AD, et al. Steatosis in chronic hepatitis C: association with increased messenger RNA expression of collagen I, tumor necrosis factor-alpha and cytochrome P450 2E1. J Gastroenterol Hepatol 2003; 18:386–392. 206. Rashid M, Roberts EA. Nonalcoholic steatohepatitis in children. J Pediatr Gastroenterol Nutr 2000; 30:48–53. 207. Brunt EM. Alcoholic and nonalcoholic steatohepatitis. Clin Liver Dis 2002; 6:399–420. 208. Itoh S, Yougel T, Kawagoe K. Comparison between nonalcoholic steatohepatitis and alcoholic hepatitis. Am J Gastro 1987; 82:650–654. 209. Goodman ZD, Ishak KG. Occlusive venous lesions in alcoholic liver disease: a study of 200 cases. Gastroenterology 1982; 83:786–796. 210. Legro RS. Polycystic ovary syndrome: the new millennium. Mol Cell Endocrinol 2001; 184:87–93. 211. Angulo P, Lindor KD. Treatment of nonalcoholic fatty liver: present and emerging therapies. Semin Liver Dis 2001; 21:81–88. 212. Laurin J, Lindor KD, Crippin JS, et al. Ursodeoxycholic acid or clofibrate in the treatment of non-alcohol-induced steatohepatitis: a pilot study. Hepatology 1996; 23:1464–1467. 213. Lindor KD, Kowdley KV, Heathcote EJ et al. Ursodeoxycholic acid for treatment of nonalcoholic steatohepatitis: results of a randomized trial. Hepatology 2004; 39:770–778. 214. Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001; 344:1343–1350. 215. Center for Disease Control and Prevention Primary Prevention Working Group. Primary prevention of type 2 diabetes mellitus by lifestyle intervention: implications for health policy. Ann Intern Med 2004; 140:951–957. 216. Diabetes Prevention Program Research Group. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002; 346:393–403. 217. Hu FB, Li TY, Colditz GA, et al. Television watching and other sedentary behaviours in relation to risk of obesity and type 2 diabetes mellitus in women. JAMA 2003; 289:1785–1791. 218. Gauthier MS, Couturier K, Latour JG, Lavoie JM. Concurrent exercise prevents high-fat-diet-induced macrovesicular hepatic steatosis. J Appl Physiol 2003; 94:2127–2134. 219. Slentz CA, Duscha BD, Johnson JL, et al. Effects of the amount of exercise on body weight, body composition, and measures of central obesity. Arch Intern Med 2004; 164:31–39. 220. Klein S, Wadden T, Sugerman HJ. AGA technical review on obesity. Gastroenterology 2002; 123:882–932. 221. Wang RT, Koretz RL, Yee HF, Jr. Is weight reduction an effective therapy for nonalcoholic fatty liver? A systematic review. Am J Med 2003; 115:554–559.
1061
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
222. Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000. JAMA 2004; 291:1238–1245. 223. Bonow RO, Eckel RH. Diet, obesity, and cardiovascular risk. N Engl J Med 2003; 348:2057–2058. 224. Klein S. Outcome success in obesity. Obesity Res 2001; 9:354S–358S. 225. Mun EC, Blackburn GL, Matthews JB. Current status of medical and surgical therapy for obesity. Gastroenterology 2001; 120:669–681. 226. Yanovski SZ, Yonovski JA. Obesity. N Engl J Med 2002; 346:591–602. 227. Neuschwander-Tetri BA, Brunt EM, Wehmeier KR, et al. Improved nonalcoholic steatohepatitis after 48 weeks of treatment with the PPAR-g ligand rosiglitazone. Hepatology 2003; 38:1008–1017. 228. Foster GD, Wyatt HR, Hill JO, et al. A randomized trial of a low-carbohydrate diet for obesity. N Engl J Med 2003; 348:2082–2090. 229. Samaha FF, Iqbal N, Seshadri P, et al. A low-carbohydrate as compared with a low-fat diet in severe obesity. N Engl J Med 2003; 348:2074–2081. 230. Polyzogopoulou EV, Kalfarentzos F, Vagenakis AG, Alexandrides TK. Restoration of euglycemia and normal acute insulin response to glucose in obese subjects with type 2 diabetes following bariatric surgery. Diabetes 2003; 52:1098–1103. 231. Steinbrook R. Surgery for severe obesity. N Engl J Med 2004; 350:1075–1079. 232. Ranløv I, Hardt F. Regression of liver steatosis following gastroplasty or gastric bypass for morbid obesity. Digestion 1990; 47:208–214. 233. Marceau P, Hould FS, Lebel S, et al. Malabsorptive obesity surgery. Surg Clin North Am 2001; 81:1113–1127. 234. Duchini A, Brunson ME. Roux-en-Y gastric bypass for recurrent nonalcoholic steatohepatitis in liver transplant recipients with morbid obesity. Transplantation 2001; 72:156–171. 235. Kral JG, Thung SN, Biron S, et al. Effects of surgical treatment of the metabolic syndrome on liver fibrosis and cirrhosis. Surgery 2004; 135:48–58. 236. Yki-Järvinen H. Thiazolidinediones. N Engl J Med 2004; 351:1106–1118. 237. Tiikkainen M, Häkkinen A-M, Korsheninnikova E, et al. Effects of rosiglitazone and metformin on liver fat content, hepatic insulin resistance, insulin clearance and gene expression in adipose tissue in patients with type 2 diabetes. Diabetes 2004; 53:2169–2176. 238. Shadid S, Jensen MD. Effect of pioglitazone on biochemical indices of non-alchoholic fatty liver disease in upper body obesity. Clin Gastroenterol Hepatol 2003; 1:384–387. 239. Promrat K, Lutchman G, Uwaifo GI, et al. A pilot study of pioglitazone treatment for nonalcoholic steatohepatitis. Hepatology 2004; 39:188–196. 240. Ricote M, Li AC, Willson TM, et al. The peroxisome proliferator-activated receptor-g is a negative regulator of macrophage activation. Nature 1998; 391:79–82. 241. Jiang C, Ting AT, Seed B. PPAR-g agonists inhibit production of monocyte inflammatory cytokines. Nature 1998; 391:82–86. 242. Miyahara T, Schrum L, Rippe R, et al. Peroxisome proliferatoractivated receptors and hepatic stellate cell activation. J Biol Chem 2000; 46:35715–35722. 243. Galli A, Crabb DW, Ceni E, et al. Antidiabetic thiazolidinediones inhibit collagen synthesis and hepatic stellate cell activation in vivo and in vitro. Gastroenterology 2002; 122:1924–1940. 244. Kirpichnikov D, McFarlane SI, Sowers JR. Metformin: an update. Ann Intern Med 2002; 137:25–33. 245. Fryer LGD, Parbu-Patel A, Carling D. The anti-diabetic drugs rosiglitazone and metformin stimulate AMP-activated protein
1062
246.
247. 248.
249.
250.
251.
252. 253.
254.
255.
256.
257.
258.
259.
260. 261. 262. 263.
264.
265.
266.
267.
268.
kinase through distinct signaling pathways. J Biol Chem 2002; 277:25226–25232. Zhou G, Myers R, Li Y, et al. Role of AMP-activated protein kinase in mechanism of metformin action. J Clin Invest 2001; 108:1167–1174. Marchesini G, Brizi M, Bianchi G, et al. Metformin in nonalcoholic steatohepatitis (letter). Lancet 2001; 358:893–894. Lin HZ, Yang SQ, Chuckaree C, et al. Metformin reverses fatty liver disease in obese, leptin-deficient mice. Nat Med 2000; 6:998–1003. Basaranoglu M, Acbay O, Sonsuz A. A controlled trial of gemfibrozil in the treatment of patients with nonalcoholic steatohepatitis. J Hepatol 1999; 31:384. Clark JM, Brancati FL. Negative trials in nonalcoholic steatohepatitis: why they happen and what they teach us. Hepatology 2004; 39:602–603. Merat S, Malekzadeh R, Sohrabi MR, et al. Probucol in the treatment of non-alcoholic steatohepatitis: a double-blind randomized controlled study. J Hepatol 2003; 38:414–418. Lavine JE. Vitamin E treatment of nonalcoholic steatohepatitis in children: a pilot study. J Pediatr 2000; 136:734–738. Harrison SA, Torgerson S, Hayashi P, et al. Vitamin E and vitamin C treatment improves fibrosis in patients with nonalcoholic steatohepatitis. Am J Gastro 2003; 98:2485–2490. Abdelmalek MF, Angulo P, Jorgensen RA, et al. Betaine, a promising new agent for patients with nonalcoholic steatohepatitis: results of a pilot study. Am J Gastro 2001; 96:2711–2717. Yao Z, Vance DE. The active synthesis of phosphatidylcholine is required for very low density lipoprotein secretion from rat hepatocytes. J Biol Chem 1988; 263:2998–3004. Najib S, Sánchez-Margalet V. Homocysteine thiolactone inhibits insulin signaling, and glutathione has a protective effect. J Mol Endocrinol 2001; 27:85–91. Vajro P, Franzese A, Valerio G, et al. Lack of efficacy of ursodeoxycholic acid for the treatment of liver abnormalities in obese children. J Pediatr 2000; 136:739–743. Yu AS, Keeffe EB. NAFLD and NASH: important diseases before and after liver transplantation. Hepatology 2001; 34:842–843. Chalasani N, Aljadhey H, Kesterson J, et al. Patients with elevated liver enzymes are not at higher risk for statin hepatotoxicity. Gastroenterology 2004; 126:1287–1292. Pratt DS, Knox TA, Erban J. Tamoxifen-induced steatohepatitis. Ann Intern Med 1995; 123:236. Van Hoof M, Rahier J, Horsmans Y. Tamoxifen-induced steatohepatitis. Ann Intern Med 1996; 124:855–856. Oien KA, Moffat D, Curry GW, et al. Cirrhosis with steatohepatitis after adjuvant tamoxifen. Lancet 1999; 353:36–37. Chu CH, Lin SC, Shih SC, et al. Fatty metamorphosis of the liver in patients with breast cancer: possible associated factors. World J Gastroenterol 2003; 9:1618–1620. Hui JM, Kench J, Chitturi S, et al. Long-term outcomes of cirrhosis in nonalcoholic steatohepatitis compared with hepatitis C. Hepatology 2003; 38:420–427. Ioannou GN, Weiss NS, Kowdley KV, Dominitz JA. Is obesity a risk factor for cirrhosis-related death or hospitalization? A population-based cohort study. Gastroenterology 2003; 125:1053–1059. Sorrentino P, Tarantino G, Conca P, et al. Silent non-alcoholic fatty liver disease – a clinical-histological study. J Hepatol 2004; 41:751–757. Poonawala A, Nair SP, Thuluvath PJ. Prevalence of obesity and diabetes in patients with cryptogenic cirrhosis: a case-control study. Hepatology 2000; 32:689–692. Caldwell SH, Hespenheide EE. Obesity and cryptogenic cirrhosis. In: Leuschner U, James O, Dancygier H, eds.
Chapter 55 NASH
269.
270.
271.
272.
273.
274.
275.
276.
277.
278.
Steatohepatitis (ASH and NASH). Falk Symposium 121. Norwell, MA: Kluwer Academic, 2001:151. Ong J, Younossi ZM, Reddy V, et al. Cryptogenic cirrhosis and posttransplantation nonalcoholic fatty liver disease. Liver Transpl 2001; 7:797–801. Contos MJ, Cales W, Sterling RK, et al. Development of nonalcoholic fatty liver disease after orthotopic liver transplantation for cryptogenic cirrhosis. Liver Transpl 2001; 7:363–373. Ratziu V, Bonyhay L, Di Martino V, et al. Survival, liver failure, and hepatocellular carcinoma in obesity-related cryptogenic cirrhosis. Hepatology 2002; 35:1485–1493. Nair S, Mason A, Eason J, et al. Is obesity an independent risk factor for hepatocellular carcinoma in cirrhosis? Hepatology 2002; 36:150–155. Marrero JA, Fontana RJ, Su GI, et al. NAFLD may be a common underlying liver disease in patients with hepatocellular carcinoma in the United States. Hepatology 2002; 36:1349–1354. El-Serag HB, Tran T, Everhart JE. Diabetes increases the risk of chronic liver disease and hepatocellular carcinoma. Gastroenterology 2004; 126:460–468. Anagnostopoulos GK, Arvanitidis D, Tsiakos S, et al. Is hepatocellular carcinoma part of the natural history of nonalcoholic steatohepatitis? J Clin Gastroenterol 2003; 37:88–89. Yuan J-M, Govindarajan S, Arakawa K, Yu MC. Synergism of alcohol, diabetes, and viral hepatitis on the risk of hepatocellular carcinoma in blacks and whites in the US. Cancer 2004; 101:1009–1017. Bugianesi E, Leone N, Vanni E, et al. Expanding the natural history of nonalcoholic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology 2002; 123:134–140. Cotrim HP, Andrade ZA, Parana R, et al. Nonalcoholic steatohepatitis: a toxic liver disease in industrial workers. Liver 1999; 19:299–304.
279. Klain J, Fraser D, Goldstein J, et al. Liver histology abnormalities in the morbidly obese. Hepatology 1989; 10:873–876. 280. Silverman JF, O’Brien KF, Long S, et al. Liver pathology in morbidly obese patients with and without diabetes. Am J Gastro 1990; 85:1349–1355. 281. Silverman EM, Sapala JA, Appelman HD. Regression of hepatic steatosis in morbidly obese persons after gastric bypass. Am J Clin Pathol 1995; 104:23–31. 282. Ratziu V, Giral P, Charlotte F, et al. Liver fibrosis in overweight patients. Gastroenterology 2000; 118:1117–1123. 283. Crespo J, Fernandez-Gil P, Hernandez-Guerra M, et al. Are there predictive factors of severe liver fibrosis in morbidly obese patients with non-alcoholic steatohepatitis? Obesity Surg 2001; 11:254–257. 284. Harrison SA, Oliver DA, Torgerson S, et al. NASH: clinical assessment of 501 patients from two separate academic medical centers with validation of a clinical scoring system for advanced hepatic fibrosis (abstract). Hepatology 2003; 34 (suppl 1):511A. 285. Tarugi P, Lonardo A, Ballarini G, et al. A study of fatty liver disease and plasma lipoproteins in a kindred with familial hypobetalipoproteinemia due to a novel truncated form of apolipoprotein B (APO B-54.5). J Hepatol 2000; 33:361–370. 286. Tarugi P, Lonardo A, Gabelli C, et al. Phenotypic expression of familial hypobetalipoproteinemia in three kindreds with mutations of apolipoprotein B gene. J Lipid Res 2001; 42:1552–1561. 287. Tajiri K, Takenawa H, Yamaoka K, et al. Nonalcoholic steatohepatitis masquerading as autoimmune hepatitis. J Clin Gastroenterol 1997; 25:538–540. 288. Ross R, Dagnone D, Jones PJH, et al. Reduction in obesity and related comorbid conditions after diet-induced weight loss or exercise-induced weight loss in men. Ann Intern Med 2000; 133:92–103.
1063
Section IX: Liver Affected by Other Organs (Non-Hepatic Conditions)
56
THE LIVER IN SYSTEMIC ILLNESS Naga Chalasani and Oscar W. Cummings Abbreviations ALT alanine aminotransferase APS antiphospholipid syndrome AST aspartate aminotransferase CDC centers for disease control CPAP continuous positive airway pressure CREST syndrome (calcinosis, Raynaud’s phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia
FAP FMF HBV HCV HPS IL-6 LDH
familial amyloidotic polyneuropathy familial mediterranean fever hepatitis B virus hepatitis C virus hemophagocytic syndrome interleukin-6 lactate dehydrogenase
INTRODUCTION When patients with abnormal liver tests are encountered, it is often assumed that the liver is the primary culprit in the disease process. However, numerous systemic illnesses and diseases of other organs can produce signs and symptoms that are indistinguishable from primary liver diseases. The hepatic manifestations in these disorders may range from mild enzyme abnormalities to significant liver injury and liver failure. In this chapter, we will review liver involvement in selected systemic disorders such as heart disease, pulmonary disease, amyloidosis, connective tissue disorders, Reye’s syndrome, jejunoileal bypass, and hematological disorders. Other systemic disorders with hepatic involvement, such as sarcoidosis, cystic fibrosis, and sepsis, are covered elsewhere in this textbook.
HEART DISEASE Liver involvement occurs in patients with both acute and chronic heart disease and its spectrum ranges from asymptomatic increases in liver biochemistries to fulminant liver failure. The liver receives a significant portion of the cardiac output and therefore any condition that decreases cardiac output will lead to a fall in hepatic perfusion. The liver is able to compensate for changes in hepatic blood flow via vasoactive mechanisms and by increasing oxygen extraction during periods of hepatic hypoperfusion.1 However, when critical levels of left or right heart failure are reached, hepatic injury may occur. In right-sided heart failure, it has been suggested that this damage is caused by elevation in right atrial pressure, leading to elevation in hepatic venous pressure that causes distention of hepatic sinusoids and liver cell hypoxia. In left-sided heart failure, decreased cardiac output results in diminished hepatic perfusion, which leads to hepatic hypoxia. In general terms, liver involvement in right heart failure is referred to as congestive hepatopathy, whereas liver injury resulting from left heart failure is known as ischemic hepatitis. However, these two phenomena often coexist and may be indistin-
NRSSS OSA OTC PBC SARS SCoV SLE
national reye syndrome surveillance system obstructive sleep apnea ornithine transcarbamylase primary biliary cirrhosis severe acute respiratory syndrome SARS corona virus systemic lupus erythematosus
guishable in clinical practice. The final common pathway for liver damage appears to be centrilobular (zone III) hepatocellular necrosis. This portion of the liver lobule is the most vulnerable to hypoxic injury due to the organization of the liver acinus. Highly oxygenated blood enters the hepatic lobule via branches of the hepatic artery and portal vein in the periportal region. As it passes through the hepatic sinusoids toward the terminal hepatic venule, oxygen is extracted and hepatocytes in the centrilobular area are perfused by blood that is the least well oxygenated.1
LIVER IN RIGHT HEART FAILURE (CONGESTIVE HEPATOPATHY) Liver abnormalities in patients with right heart failure are common. Right heart failure can be isolated (due to cor pulmonale or primary pulmonary hypertension) or, more likely, a consequence of left ventricular failure. In a large study of 175 patients with both acute and chronic right heart failure,2 hepatomegaly was present by physical examination in over 90% and splenomegaly in 20% of these patients.2 Other findings of right heart failure, such as peripheral edema, pleural effusion, and ascites, were also frequently present (Table 56-1). Ascites is more prominent in patients with chronic right heart failure than in acute right heart failure.2 Characteristic changes in histology are seen in the liver of patients with congestive heart failure. On gross inspection, the congested liver appears enlarged and purple with blunt edges.3 The classically described “nutmeg” appearance is caused by alternative areas of pale, more normal-appearing parenchyma contrasting with congested, hemorrhagic areas that correspond to the centrilobular regions of the liver (Figure 56-1). Microscopically, central veins and sinusoids in the centrilobular region become dilated and engorged with erythrocytes. Inflammation is noticeably absent (Figure 56-2). Adjacent hepatocytes may become compressed and atrophied. With long-standing hepatic congestion, fibrosis and cirrhosis may develop (cardiac cirrhosis) (Figure 56-3). Hepatic congestion due to the right heart failure results in numerous biochemical abnormalities (Table 56-2). In chronic congestive
1065
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Table 56-1. Symptoms and Signs of Congested Livers in 175 Patents with Right-Sided Heart Failure
Table 56-2. Liver Tests of 175 Patients with Right-Sided Heart Failure
Symptom/sign
Liver tests
Hepatomegaly Peripheral edema Pleural effusion Splenomegaly Ascites
Acute heart failure (%)
Chronic heart failure (%)
99 77 25 20 7
95 71 17 22 20
Adapted from Richman SM, Delman AJ, Grob D. Alterations in indices of liver function in congestive heart failure with particular reference to serum enzymes. Am J Med 1961; 30:211.
Acute heart failure
Bilirubin BSP retention Alkaline phosphatase Aspartate aminotransferase Alanine aminotransferase Globulins Prothrombin time Albumin Cholesterol
Chronic heart failure
n
Abnormal(%)
n
Abnormal (%)
86 71 80 67 53 100 68 100 87
37 87 10 48 15 60 84 32 49
57 55 55 37 29 67 43 67 60
21 71 9 5 3 37 74 27 42
BSP, Bromosulfophthalein. Adapted from Richman SM, Delman AJ, Grob D. Alterations in indices of liver function in congestive heart failure with particular reference to serum enzymes. Am J Med 1961; 30:211.
Figure 56-1. Nutmeg liver. In chronic passive congestion of the liver, red cells pool and distend the sinuses around the central vein. These regions develop a darker red-violet color, in contrast to the surrounding tan liver parenchyma. This color stippling is reminiscent of the cut surface of a nutmeg.
Figure 56-2. Liver congestion. The sinuses around the central vein are distended by normal red cells. As the severity of this lesion increases, the adjacent hepatocytes become atrophic. (Hematoxylin and eosin.)
1066
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS Figure 56-3. Cardiac cirrhosis. Dense fibrous bands emanate from the central veins (arrows) to surround a nodule of regenerating hepatocytes. (Trichrome.)
heart failure, hyperbilirubinema occurs in over 20% of patients.2 The elevation is generally mild, less than 3 mg/dl, and composed predominantly of unconjugated bilirubin. Serum aminotransferase levels are usually normal or minimally elevated in compensated, chronic congestive heart failure, but may become elevated during exacerbations of heart failure. Prothrombin time is prolonged in the majority of patients with hepatic congestion from right heart failure. Patients anticoagulated with warfarin sodium (warfarin) for dilated cardiomyopathy may have decreased warfarin requirements during exacerbations of congestive heart failure and this effect, if not appreciated, could result in dangerously prolonged prothrombin times. Hepatic biochemical abnormalities generally improve with improvement in cardiac function. Several signs and symptoms of congestive heart failure (e.g., ascites, pedal edema, mild hyperbilirubinemia) are also seen in patients with decompensated hepatic cirrhosis and distinguishing these two entities may be difficult. In some patients, it is particularly difficult to distinguish cardiac ascites from cirrhotic ascites clinically. In such cases, characterization of ascitic fluid or measurement of hepatic venous pressure gradient may be of assistance. In a prospective study of 13 patients with cardiac ascites, the serum ascites albumin concentration gradient was 1.1 g/dl and the total protein was 2.5 g/dl.3 Additionally, cardiac ascites had significantly more ascitic fluid red cell counts and higher levels of lactate dehydrogenase.3
LEFT HEART FAILURE AND ISCHEMIC HEPATITIS Ischemic hepatitis can be defined as hepatocellular necrosis associated with a decrease in hepatic perfusion.4–6 It is relatively infrequent, with a reported incidence of 0.16–1.5% of hospitalized patients. It can affect any age group, although it is most frequently reported in the older population. This undoubtedly reflects the increased risk of underlying cardiovascular disease in older people. When seen in children, it is often associated with congeni-
tal heart disease or overwhelming sepsis.7 The term ischemic hepatitis is a misnomer because ischemic liver injury is characterized by centrilobular necrosis in the absence of inflammation. The diagnosis of ischemic hepatitis should be considered in any patient with elevations of liver enzymes (aspartate aminotransferase (AST), alanine aminotransferase (ALT), and lactate dehydrogenase (LDH)) in the setting of documented or suspected systemic hypotension.
Causes and Pathogenesis Liver is a highly vascular organ, receiving approximately 25% of cardiac output. Seventy percent of the hepatic blood flow is derived from the portal system. The other 30% is delivered by the hepatic artery and liver arterial perfusion expresses a linear relation between blood pressure and blood flow.8 The liver can maintain normal oxygen uptake by increasing oxygen extraction, with as much as 95% of the oxygen from the blood being extracted in a single pass through the liver.6 This remarkable compensatory mechanism most likely accounts for the low incidence of liver damage in shock (resistance to ischemia). Nevertheless, these compensatory mechanisms are overwhelmed in some patients with severely diminished hepatic perfusion leading to ischemic liver injury. Cardiogenic shock from any cause (e.g., myocardial infarction, tamponade) is the most commonly reported risk factor for the development of ischemic hepatitis. Transiently decreased cardiac output seen in patients with arrhythmia or valvular heart disease may also result in hepatic injury even in the absence of bona fide hypotension (? relative hypotension).9 Episodes of hypotension resulting in ischemic hepatitis may be very brief and sometimes there may not be any documented episodes of hypotension. Although diminished hepatic perfusion from systemic hypotension (either absolute or relative) is essential, a recent study suggested that systemic hypotension or shock alone is insufficient to cause ischemic hepatitis.10 In this study, 31 patients with ischemic hepatitis were compared to a control group consisting of 31 previously
1067
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
healthy subjects with major non-hepatic trauma and marked hypotension (mean systolic pressure of 54 ± 22 mmHg lasting for a mean of 19 ± 14 minutes). All 31 patients with ischemic hepatitis had organic heart disease and more than 90% had demonstrable right heart failure. None of the subjects in the control group developed ischemic hepatitis despite profound hypotension. These findings suggest that right heart failure with passive hepatic congestion renders the liver susceptible to ischemic liver injury during transient systemic hypotension.10 Henrion et al. recently published a paper that described clinical and hemodynamic features of 142 episodes of ischemic hepatitis. When ischemic hepatitis occurred in patients with congestive heart failure or acute heart failure, the hepatic hypoxia resulted primarily from the decreased hepatic blood flow and venous congestion. However, when ischemic hepatitis occurred in patients with toxic or septic shock, oxygen delivery to liver was decreased but oxygen needs were increased, leading to hepatic hypoxia.11 Non-cardiogenic causes of ischemic hepatitis include hypovolemic shock from hemorrhage or dehydration, heat stroke, and septic shock.12–14 Rare episodes of ischemic hepatitis have been reported in patients who ingest vasoactive medication (ergotamine overdose) and after protracted seizures in children.15,16
Clinical Syndrome The clinical picture is usually dominated by the cardiovascular, septic, or hemorrhagic illness that precipitated the hepatic hypoperfusion. A distinctive biochemical pattern is characteristic of
this disorder.17,18 Serum aminotransferase levels rise rapidly after an ischemic episode and peak within 1–3 days (Figure 56-4). With treatment of the underlying illness, serum aminotransferases usually return to near-normal within 7–10 days of the initial insult. Persistent elevation of serum aminotransferase levels beyond this period implies a poor prognosis due to continued hepatic hypoperfusion. Serum ALT and AST activity are strikingly elevated and may exceed 200 times the upper limits of normal (Figure 56-4). Less marked elevation (<500 U/l) have also been reported in biopsyproven ischemic hepatitis. Serum LDH activity is also markedly elevated in patients with ischemic hepatitis. When fractionated, serum LDH activity is mostly of hepatic origin.17 The level of LDH may rise to 30 times the upper limits of normal and parallels the pattern of aminotransferase activity, with a brisk rise and rapid resolution. Of note, the serum LDH is usually only slightly elevated in patients with acute viral hepatitis. Marked elevations of alkaline phosphatase or serum bilirubin are unusual in ischemic hepatitis and cholestasis has not been demonstrated on liver biopsies of those patients. Mild elevations of serum bilirubin may be seen, but this rarely exceeds four times the upper limit of normal.17 In one series of patients with ischemic hepatitis, additional biochemical features were noted that might be helpful in diagnosis.7 All patients in this series had transient abnormalities of serum creatinine and blood urea nitrogen. These changes were sometimes marked, consistent with acute renal failure, but resolved over 7–10 days. The authors speculated that the same hypotensive insult to the liver had similar adverse effects on the kidney. Both hyper-
6000 Patient 1 5000
Patient 2 Patient 3
Aspartate Aminotransferase U/L
4000
Patient 4 Patient 5
3000
Patient 6 Patient 7
2000
Patient 8 Patient 9
1000
750
500
250
0 Pre insult 1
2
3
4
5
6
7
8
Days
1068
9
10 11 12 13 14 15 16 17
Figure 56-4. Ischemic hepatitis. Serial alanine aminotransferase (ALT) levels in patients with ischemic hepatitis. (Adapted from Gitlin N, Serio KM. Ischemic hepatitis: widening horizons. Am J Gastroenterol 1992; 87:831, with permission.)
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS
glycemia and hypoglycemia can be seen.5,17,18 In one series, two-thirds of patients with ischemic hepatitis had new-onset hyperglycemia that occurred within 48 hours of their illness.18 In another report, hypoglycemia was seen in 33% of the patients with ischemic hepatitis.5 Histologic studies of ischemic hepatitis are limited. Patients in whom a diagnosis of ischemic hepatitis is being considered are usually critically ill, often with multiorgan failure, making a liver biopsy impractical. Nevertheless, the hallmark of ischemic hepatitis is centrilobular necrosis in the absence of an inflammatory infiltrate.19 Gibson and Dudley studied 17 patients with these characteristic histologic findings and concluded that a diagnosis of ischemic hepatitis could be made without a liver biopsy in the appropriate clinical setting (patients who had a potential cause for a fall in cardiac output) and a rapid rise in levels of serum aminotransferases and LDH.17
Differential Diagnosis Few primary liver diseases will give such marked elevations of serum aminotransferases followed by rapid resolution. The diagnosis of ischemic hepatitis can be made readily in a patient in the intensive care unit with a rapid and striking increase of serum aminotransferase and LDH activities who has recently suffered a documented, acute hypotensive episode that required pressor support. Acute viral hepatitis may mimic the clinical picture seen with ischemic hepatitis. Viral hepatitis will usually be accompanied by a symptomatic prodrome or a history of exposure to an infectious agent and serologies will help to exclude a viral etiology. Serum LDH activity is only mildly elevated in viral hepatitis. In patients who have only modest elevations in serum aminotransferases, chronic hepatitis B and hepatitis C should be excluded by history and appropriate serologic tests. Examination of previous liver enzyme results prior to the current illness may be invaluable in distinguishing the acute illness from chronic viral hepatitis. Special care should be taken to identify all medications taken by the patient prior to admission as well as those started since hospitalization. Drug hepatoxicity can be associated with striking elevations of serum aminotransferase and LDH activities. Acetaminophen toxicity should always be considered in patients with marked elevations of serum aminotransferase levels and renal insufficiency. Amiodarone is an antiarrhythmic agent that is increasingly being used in the management of unstable tachyarrhythmias. Its toxicity should also be considered in the differential diagnosis as several cases of acute hepatitis and fulminant liver failure have been reported with intravenous loading doses of amiodarone.19–21 Other causes of marked, acute elevations of serum AST activity include rhabdomyolysis, acute myocardial infarction, acute cholangitis, hepatic trauma, and hepatic infarction, all of which should be distinguishable from ischemic hepatitis by history and additional laboratory investigation.
Treatment and Prognosis The treatment of ischemic hepatitis is directed at the underlying illness. Therapy to improve cardiac output with inotropes and pressors will improve hepatic perfusion and result in resolution of the ischemic hepatitis. Similarly, volume resuscitation for patients with
hemorrhagic shock and appropriate treatment for septic shock will indirectly improve cardiac output and hepatic perfusion. Special consideration should be given to prescribing medications in patients with circulatory failure and ischemic hepatitis. The clearance of certain medications (e.g., lidocaine, calcium-channel blockers) is dependent on hepatic blood flow and their clearance can be greatly diminished in this setting. Indeed, an association has been reported between the use of calcium-channel blockers and antiarrhythmic agents in patients with ischemic hepatitis and increased mortality, although it was not clear from that study if the worse prognosis was merely related to the presence of more severe cardiac disease.22 Opiates and analgesics that are often prescribed for these critically ill patients also have the potential to accumulate and cause neurologic and respiratory depression. It has been suggested that the risk of acetaminophen toxicity is increased in patients with underlying cardiomyopathy, even at low doses in the absence of alcohol ingestion.23 The prognosis of ischemic hepatitis is largely dependent upon the prognosis of the underlying illness. Short-term mortality has been reported to be as high as 50% in patients with ischemic hepatitis. One-month and 1-year survival rates for patients with ischemic hepatitis in the series by Henrion et al. were 47% and 28%, respectively.11 However, despite the massive hepatic necrosis that may occur, deaths due to hepatic failure are extremely rare; the cause of death is usually a result of poor cardiac reserve.4 The aminotransferase levels do not seem to have any prognostic significance. When stratified according to peak AST activity, the survival for patients with AST levels lower than 2000 IU/l was 43%, compared to 41% for patients with a peak AST activity above this value. The pattern of AST activity, however, does seem to have some prognostic value. In patients with ischemic hepatitis who died, the level of serum AST did not drop appreciably from peak values while in those who survived, AST levels rapidly returned towards normal, suggesting improving cardiac function.4
LIVER IN CONSTRICTIVE PERICARDITIS Constrictive pericarditis may masquerade as a primary liver disease.24–26 Pulsatile hepatomegaly, splenomegaly, and ascites are often present in patients with constrictive pericarditis.27 Other important physical findings include elevated jugular venous pressure, pulsus paradoxus, and pericardial knock. It has been noted that ascites is more prominent than pedal edema.28 Histological features are usually non-specific: diffuse centrilobular congestion, necrosis, and fibrosis are the most common, but there may be mild abnormalities such as patchy fibrosis without congestion. Occasionally, there can be prominent sinusoidal dilation and hemorrhagic necrosis with hepatic vein thrombosis, leading to a misdiagnosis of Budd–Chiari syndrome.29,30 Tuberculosis remains an important cause of constrictive pericarditis, but other etiologies, such as cardiac surgery, connective tissue disorders, subclinical viral pericarditis, and malignancy, are becoming increasingly frequent. Pericardial calcification on chest radiography and low voltage on electrocardiography, when present, are very suggestive of constrictive pericarditis. A recent report indicates that pericardiectomy will improve liver function (assessed by indocyanine green clearance) in most patients with liver injury caused by constrictive pericarditis.31
1069
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
LIVER INJURY DUE TO CHRONIC RESPIRATORY FAILURE Although most cases of ischemic hepatitis are related to altered hepatic perfusion (e.g., congestive heart failure), it is now clear that severe arterial hypoxemia caused by severe respiratory disease can also lead to hepatic injury independent of hepatic perfusion abnormalities (hypoxic hepatitis).32–35 A recent study investigated the details of hemodynamic and oxygen transportation abnormalities in 19 consecutive episodes of hypoxic hepatitis caused by acute exacerbation of chronic respiratory failure without left heart failure.11 These patients had marked arterial hypoxemia (mean PaO2 34 mmHg and PaCO2 64 mmHg) with elevated central venous pressure but an elevated cardiac index and low systemic vascular resistance. Oxygen delivery was significantly decreased in these patients, despite reasonable hepatic blood flow, as measured by the low-dose galactose clearance test.11 The authors suggested that hypoxic hepatitis induced by acute exacerbation of chronic respiratory failure is due to severe arterial hypoxemia in the background of elevated central venous pressure and passive hepatic congestion.11,33 Most patients with hypoxic hepatitis caused by respiratory diseases have accompanying cardiac dysfunction; however, there are reports of severe obstructive sleep apnea (OSA) leading to hypoxic hepatitis in the absence of any cardiac dysfunction.33 A recent study suggests that chronic intermittent hypoxemia may cause chronic low-grade hepatic injury in patients with OSA.36 Chin et al. investigated the relationship between liver enzyme abnormalities and intermittent hypoxemia in 40 obese men with OSA. In this study, the prevalence of unexplained elevations in transaminases in obese patients with OSA was 35%. More interestingly, the authors systematically evaluated the effect of treatment with nasal continuous positive airway pressure (CPAP) on the levels of aminotransferases, insulin, and glucose in 40 obese subjects with OSA. There was a significant decrease in the levels of transaminases after overnight, 1 month, and 6 months of nasal CPAP treatment, with no significant changes in the degree of insulin resistance (Table 56-3).
JEJUNOILEAL BYPASS Jejunoileal bypass surgery was introduced in the 1960s as a surgical treatment for morbid obesity.37 Initial enthusiasm for this radical weight-loss procedure was tempered by the recognition of a myriad of associated complications, including electrolyte abnormalities, kidney problems, gallstones, severe malnutrition, and liver disease.37–41 The hepatic effects of jejunoileal bypass include fatty infiltration, cirrhosis, and progressive liver failure. This procedure is rarely, if ever, performed today, so new diagnoses of liver disease related to jejunoileal bypass are unusual. However, patients who have had surgery decades ago may present with end-stage liver disease, on occasion being referred to centers for consideration of liver transplantation. Rarely, non-alcoholic steatohepatitis has been reported following gastric partitioning surgery.42
MECHANISM OF HEPATIC INJURY Numerous theories have been proposed to explain the liver injury associated with jejunoileal bypass. The bypass procedure results in
1070
Table 56-3. Changes in aspartate aminotransferase (AST), Alanine Aminotransferase (ALT), and Insulin Resistance Following Nasal Continuous Positive Airway Pressure (CPAP) Treatment in Obese Patients with Obstructive Sleep Apnea (n = 40) Nasal CPAP treatment
AST before and after treatment (IU/l)
ALT before and after treatment (IU/l)
HOMA before and after treatment
Overnight CPAP
39 ± 28 versus 34 ± 24 (P < 0.001)
61 ± 53 versus 61 ± 48 (P = 0.9)
CPAP for 1 month
37 versus 25a (P < 0.001)
60 versus 40a (P = 0.001)
CPAP for 6 months
37 versus 25a (P < 0.001)
60 versus 40a (P = 0.001)
3.4 ± 2.2 versus 3.2 ± 1.6 (P = 0.3) versus 3.2 ± 1.6 (P = 0.3) 3.4 ± 2.2 versus 2.8 ± 1.4 (P = 0.2) N/A
HOMA, homeostasis model assessment method (a measure of insulin resistance). a Standard deviations for AST and ALT values for 1- and 6-month CPAP were not available in the manuscript. Adapted from Chin K, Nakamura T, Takahasi K, et al. Effects of obstructive sleep apnea on serum aminotransferase levels in obese patients. Am J Med 2003;114:370.
malabsorption of a number of nutrients, which accounts for some of its weight-loss effects and many of the associated complications. Early researchers noted the similarities of fatty infiltration seen in patients with protein malnutrition and in patients after jejunoileal bypass.43 Serum levels of essential amino acids were decreased when measured during the period of rapid weight loss and there was a concurrent increase in fatty infiltration of the liver. In a control group of patients who had stable weights after jejunoileal bypass, serum amino acid levels were higher and there was less hepatic steatosis. Nutritional supplements were reported to decrease the extent of hepatic injury in some patients, although others experienced continued deterioration. Animal models and studies in patients after jejunoileal bypass implicated bacterial overgrowth as a possible etiologic factor in the development of hepatic steatosis. In a dog model of jejunoileal bypass, vibramycin treatment prevented the appearance of fatty liver and death from liver failure.44 In rats, metronidazole therapy also seemed to ameliorate the postoperative changes to the liver, but to a lesser degree than did protein supplementation.45 Genetically obese ob/ob mice with hepatic steatosis were shown to have agerelated increases in the production of endogenous ethanol by the intestinal microflora.46 Drenick and associates47 treated patients with metronidazole administered at random time intervals after jejunoileal bypass. In untreated patients, hepatic lipid content as measured by morphologic assessment was elevated and remained abnormal over a 12-month period in most patients. In patients treated with metronidazole, hepatic fat content that was initially elevated following the bypass surgery decreased once the drug was started. In another group, in whom metronidazole was administered intermittently, levels of hepatic fat increased and then decreased in concert with antibiotic therapy. Moreover, in this study there was no correlation between hepatic steatosis and protein malnutrition. The mechanism of hepatic injury, although perhaps related to bacterial overgrowth, remains unknown.
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS
PATHOLOGY OF LIVER DISEASE The hepatic histology found in patients who have undergone jejunoileal bypass often resembles the changes seen in alcoholic liver disease. The most common findings include macrovesicular steatosis, Mallory’s hyaline, sinusoidal fibrosis, and infiltrates of polymorphonuclear leukocytes.48,49 The histologic changes of hepatic steatosis are often at their worst in the first year after surgery and then may improve in some patients, while others show continued progression to cirrhosis and liver failure. Vyberg and co-workers50 performed serial liver biopsies on 34 morbidly obese patients undergoing jejunoileal bypass. Postoperatively, 44% of patients had no or minimal histologic changes, while the remainder had varying degrees of steatosis, steatohepatitis, and perivenular fibrosis. Five to 9 months postoperatively, liver biopsies revealed progression of the hepatic injury in almost all patients. In the group of patients with minimal preoperative changes, 85% had developed moderate to marked steatosis. Those patients with more severe changes preoperatively showed increased steatohepatitis and fibrosis; 18% developed bridging fibrosis and 9% had changes of cirrhosis. Mallory’s hyaline was seen in almost one-third of patients. Nasrallah and colleagues51 attempted to identify pre- and postoperative variables that may predict histologic liver injury after jejunoileal bypass. Preoperatively, 59% of morbidly obese patients had normal liver histology. There was no correlation between preoperative histology, serum biochemical parameters, use of small amounts of alcohol, or the amount of weight loss and the postoperative histologic changes.
CLINICAL MANIFESTATIONS Liver injury after jejunoileal bypass may range from asymptomatic elevations of serum hepatic enzymes to cirrhosis with liver failure. It has been estimated that up to 2% of patients died of liver failure following this procedure.38,52 Liver injury may become clinically apparent within months of surgery or progress insidiously for more than 10 years before presenting with signs and symptoms of cirrhosis. Requarth and co-workers53 reported the long-term morbidity following jejunoileal bypass in 453 patients. During the follow-up, 24 patients developed acute liver failure (7%) and the actuarial 15-year probability of cirrhosis was 8.1%. In the early stages of hepatic injury associated with jejunoileal bypass, liver enzyme abnormalities do not correlate with histologic findings and are therefore of limited value in identifying patients at risk for liver failure.51,52,54 Mild elevations of serum aminotransferases and alkaline phosphatase levels may occur, but many patients with significant histologic changes have no biochemical abnormalities. Serial liver biopsies may be helpful in following the progression of liver disease. Once overt liver failure develops, elevation of serum bilirubin, a further fall in serum albumin, and prolongation of prothrombin time will be evident. In these patients, a trial of parenteral vitamin K administration is worthwhile to exclude malabsorption as a contributing cause to the coagulopathy. Patients who are discovered to have progressive liver injury may benefit from reversal of the bypass operation.41,55 In one series, 9 patients with cirrhosis had reversal of the bypass to re-establish continuity of the small intestine. Seven patients survived for at least 3 years after the surgery. Follow-up liver biopsies showed histologic
improvement in 4 patients with decreased steatosis and inflammation, while 3 had no obvious changes on liver histology. Two patients died of liver failure shortly after the reversal procedure; both of these patients had preoperative ascites, indicating that once decompensation has occurred the prognosis is very poor. For patients who have already developed decompensated cirrhosis, liver transplantation may be the only therapeutic option available to improve quality of life and prolong survival.56–59 In general, it is believed that jejunoileal bypass should be reversed either during or immediately after the liver transplantation.57,58 One should avoid the takedown of jejunoileal bypass in patients with advanced liver disease prior to transplantation because such surgical interventions are poorly tolerated. If the jejunoileal bypass is not reversed, there is a substantial risk of recurrent steatosis and progressive liver damage in the allografts.59 In such patients, liver biopsies should be performed on a regular basis during the post-transplant period to detect progressive liver damage. Morbid obesity is common in the post-transplant period when jejunoileal bypass is reversed, but allograft abnormalities are uncommon despite such weight gain.57–59
CONNECTIVE TISSUE DISEASES Patients with connective tissue disorders such as systemic lupus erythematosus (SLE), rheumatoid arthritis, Sjögren’s syndrome, and scleroderma may have clinical and biochemical evidence of associated liver disease. The severity of liver involvement can range from asymptomatic elevations of serum aminotransferases to cirrhosis and liver failure. Unusual liver lesions such as nodular regenerative hyperplasia have been reported with increased frequency in patients with connective tissue disorders. Over the last decade, there have been numerous reports associating chronic hepatitis C infection with Sjögren’s syndrome. An excellent review was recently published by Abraham and colleagues that provided a detailed review of the hepatic manifestations of autoimmune rheumatic diseases.60
SYSTEMIC LUPUS ERYTHEMATOSUS SLE is an autoimmune disorder involving skin, kidneys, cardiovascular system, and central nervous system. Strict criteria reflecting multiorgan involvement have been established by the American College of Rheumatology in order to diagnose SLE uniformly and distinguish it from other connective tissue disorders; patients are required to have four of 11 criteria before a diagnosis of SLE is established.61 These criteria need not be present at one time, but may develop sequentially over many years. Liver test abnormalities are not a part of these diagnostic criteria and the liver is generally not a major target for end-organ damage in patients with SLE. Nevertheless, many patients with SLE may have clinically significant liver disease. Runyon and associates62 reviewed over 200 patients who met the criteria for SLE. Twenty-one percent of patients had abnormal liver enzymes at some point during their illness and these elevations could not usually be attributed to other non-hepatic etiologies or comorbid conditions. In 20% of patients, the first liver enzyme elevations were noted during an exacerbation of SLE. Elevations of aminotransferase and alkaline phosphatase levels were usually mild – less
1071
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
than fourfold the upper limits of normal. However, more severe liver disease occurred, approximately 25% of the patients with abnormal liver enzymes were jaundiced, and 3 patients died of liver failure. Liver biopsy specimens were available from 33 patients and revealed a variety of hepatic lesions, including non-specific portal inflammation, chronic active hepatitis, and established cirrhosis. The most common finding was hepatic steatosis that was seen in over onethird of patients. In an autopsy study published in 1992, Matsumoto and associates63 studied liver specimens and clinical data from 52 patients with SLE. None of these patients died as a result of liver failure. Onethird of these patients had abnormalities of at least two different liver enzymes. The most common finding was hepatic congestion, although the authors felt that this lesion was most likely a result of the terminal event. As in the previous study, hepatic steatosis was very common and was found in 73% of patients. Twelve percent of patients had chronic hepatitis. Of interest, 21% of patients had histological evidence of arteritis, a finding previously considered rare in patients with SLE. Hepatic infarction occurred in 1 patient. Nodular regenerative hyperplasia, a lesion characterized by diffuse nodularity in the absence of associated fibrosis, was seen in 3 patients. Hemophagocytic syndrome (HPS) can be seen in patients with SLE.64,65 HPS is a clinicopathological entity characterized by systemic proliferation of benign hemophagocytic cells of monocyte–macrophage–histiocyte lineage. It is characterized by fever, cytopenia, liver dysfunction, and lymphadenopathy. This complication developed in nearly 10% of the patients with SLE seen at Yokohama City University Hospital over a 10-year period.64 The interpretation of any studies of liver disease in patients with SLE is complicated by comorbid conditions and the potential for drug toxicities that may mimic chronic liver disease. All of the above-mentioned studies were performed prior to the availability of hepatitis C testing. Thus, hepatitis C virus (HCV) cannot be excluded as a cause for some of the liver enzyme elevations or the abnormal hepatic histology. This is particularly important since substantial number of patients with SLE received blood transfusions prior to 1991. In addition, most patients in these studies received varying doses of prednisone and this could explain the frequent finding of hepatic steatosis in these patients. Hepatic injury from salicylates may also be a factor in producing some of the liver dysfunction associated with SLE and other connective tissue disorders.66 The salicylate-induced hepatotoxicity appears to be dose-dependent. Hepatic injury is not evident at doses less than 2.5 g/day or with a blood salicylate level <25 mg/dl. Furthermore, there is a correlation between blood salicylate levels and the serum ALT activity.67 Features of a hypersensitivity reaction, such as fever or rash, are usually absent. The elevation in the liver enzymes reflects hepatocellular injury, with elevations of serum aminotransferases, sometimes exceeding 1000 U/l. Discontinuation of aspirin results in prompt improvement of liver enzymes with no chronic sequelae. Ascites has been reported in several patients with SLE in the absence of liver disease or other secondary conditions.68 The pathogenesis of ascites in these patients is due to serositis with associated weeping of lymphatics. Ascites may be present even when SLE is relatively quiescent and other symptoms or signs of the disease are
1072
inactive. Increased immunosuppression with prednisone may be beneficial in reducing SLE-related ascites, once other etiologies have been carefully excluded.
RHEUMATOID ARTHRITIS Elevations of alkaline phosphatase are the most frequently reported liver test abnormalities associated with rheumatoid arthritis, and may be seen in as many as 50% of patients, while serum aminotransferase levels are usually normal.69 The source of alkaline phosphatase, bone versus liver, is somewhat controversial. In studies of alkaline phosphatase fractions, almost one-third of patients with rheumatoid arthritis had elevated levels of hepatobiliary alkaline phosphatase, suggesting liver involvement. However, other serum enzymes, such as 5¢-nucleotidase and g-glutamyltranspeptidase, that are often used a supplementary assays to support the hepatic origin of alkaline phosphatase, are frequently normal in patients with rheumatoid arthritis.69,70 Furthermore, these enzymes have been shown to have higher levels in the joint space than in serum, implying that they may originate from inflamed joints. Other investigators have demonstrated that the degree of total alkaline phosphatase elevation varies with the number of joints involved.71 Finally, the pattern of alkaline phosphatase fractions may vary over time.72 Thus, abnormal serum liver-associated enzymes in patients with rheumatoid arthritis must be interpreted with caution. A number of liver diseases have been reported in patients with adult and juvenile rheumatoid arthritis (Table 56-4). In 1997, Ruderman and associates73 published an autopsy study of hepatic histology in 182 patients with rheumatoid arthritis. The most common finding was hepatic congestion, although it is likely that this lesion was a result of the terminal event. As in SLE, hepatic steatosis was common and was found in 23% of the patients. Eleven percent of patients had fibrosis, 2.7% had established cirrhosis, and 5% had evidence of amyloidosis. Other disorders that are described in association with rheumatoid arthritis, such as primary biliary cirrhosis (PBC) and nodular regenerative hyperplasia, were not seen in this series. Nodular regenerative hyperplasia has been reported in association with rheumatoid arthritis, often with Felty’s syndrome and active joint disease (Figure 56-5).74 Features of portal hypertension such as varices and ascites are common in this latter group of patients.63 The pathogenesis of nodular regenerative hyperplasia is not known, although some authors have suggested it is related to drug toxicity or underlying portal venous thromboses.63,74 The latter hypothesis is intriguing given the frequency of antiphospholipid syndrome (APS) in patients with connective tissue diseases, the same
Table 56-4. Liver Disease Associated with Adult and Juvenile Rheumatoid Arthritis Adult rheumatoid arthritis
Juvenile rheumatoid arthritis
Hepatic steatosis Primary biliary cirrhosis Autoimmune hepatitis Nodular regenerative hyperplasia Amyloidosis Salicylate or methotrexate hepatotoxicity
Acute hepatitis Chronic non-specific hepatitis Massive liver enlargement Drug toxicity
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS Figure 56-5. Nodular regenerative hyperplasia. Thin bands of atrophic hepatocytes (arrows) outline a central focus of hepatocyte regeneration, producing a nodular appearance throughout the liver. The reticulin in the atrophic areas is condensed but there is no fibrosis. (Hematoxylin and eosin.)
population with an increased incidence of nodular regenerative hyperplasia. As with SLE, drug toxicities, particularly aspirin and other salicylates, may play a role in the liver abnormalities associated with rheumatoid arthritis. Gold therapy may cause intrahepatic cholestasis with features of a hypersensitivity reaction, including a skin rash and eosinophilia.75–77 Prolonged, high-dose gold salt therapy has also been reported in 1 patient to produce a dose-related form of toxicity characterized by jaundice and severe hepatocellular necrosis. Liver biopsy revealed submassive hepatic necrosis with lobular and portal inflammation. Brown-black pigment, identified as gold granules, was seen in macrophages, and electron microscopy demonstrated gold particles in lysosomes. It was believed that hepatic injury occurred once the concentration of gold exceeded the lysosomal storage capacity.78 Perhaps the most controversial issue in drug therapy and hepatotoxicity involves the use of methotrexate for the treatment of rheumatoid arthritis. This topic is reviewed elsewhere (Chapter 26).
ANTIPHOSPHOLIPID SYNDROME APS is characterized by the presence of antiphospholipid antibodies (anticardiolipin antibodies and/or lupus anticoagulant) in association with venous/arterial thromboses, recurrent fetal loss, and thrombocytopenia. Although APS can be a primary disorder, it is frequently seen in patients with SLE and other connective tissue disorders. The most commonly described hepatic manifestation of APS is Budd–Chiari syndrome.79,80 Pelletier and colleagues81 reported that, of their 22 patients with APS, 4 had Budd–Chiari syndrome, with no other cause of hepatic vein thrombosis. Several other liver disorders have been reported in patients with APS and these are summarized in Table 56-5.79–83
SJÖGREN’S SYNDROME Sjögren’s syndrome is an autoimmune disorder that mainly affects exocrine glands and usually presents as a persistent dryness of the
Table 56-5. Liver Complications in Antiphospholipid Syndrome Budd–Chiari syndrome Hepatic veno-occlusive disease Nodular regenerative hyperplasia Transient elevation of hepatic enzymes due to multiple fibrin thrombi Infarction of liver Autoimmune hepatitis HELLP syndrome HELLP, hemolysis, elevated liver enzymes, and low platelet count.
mouth and eyes due to functional impairment of the salivary and lacrimal glands. PBC is common in both primary and secondary forms of Sjögren’s syndrome. In a study of 300 patients with primary Sjögren’s syndrome, 7% of the patients had elevated antimitochondrial antibody titers.84 Of these, 60% had elevations of alkaline phosphatase. Liver biopsies revealed changes of early PBC, even in many patients with normal liver enzymes.84 Over the last decade, there have been numerous investigations exploring the relationship between HCV and Sjögren’s syndrome. In 1992, Haddad and colleagues85 postulated a causal relationship between Sjögren’s syndrome and HCV. The reported prevalence of HCV in patients with Sjögren’s syndrome depends on the methods of detection, population studied, and the criteria for diagnosing primary Sjögren’s syndrome. In European patients, the prevalence ranges between 14 and 19% by third-generation enzyme-linked immunosorbent assay and 5–19% by the second-generation immunoblot (RIBA-2) method, whereas HCV prevalence by RIBA2 was only 0–1% in American patients.86 HCV prevalence is much lower when the polymerase chain reaction method is used to detect HCV or when more objective criteria are applied to define primary Sjögren’s syndrome.87 Based on the current evidence, HCV seems to be a rare cause of primary Sjögren’s syndrome except in patients with liver involvement or cryoglobulinemia.86 Furthermore, HCV is
1073
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
a sialotropic virus and morphological evidence of sialadenitis is found in a significant proportion of patients with chronic HCV infection.86–90 In HCV-related sialadenitis, the lymphocytic infiltrate is predominantly pericapillary (unlike periductal, in primary Sjögren’s syndrome) and clinical symptoms of dryness are infrequent.
SCLERODERMA Reynolds and colleagues91 described 6 patients with typical PBC and varying features of scleroderma and CREST syndrome (calcinosis, Raynaud’s phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia). Since then, the association between these two autoimmune disorders has been well established. The close immunologic association between PBC and scleroderma is supported by the finding of positive antimitochondrial antibodies in more than one-quarter of patients with scleroderma and anticentromere antibody in one-quarter of patients with PBC. In a recent study of 40 patients with systemic and localized scleroderma, liver biopsy-confirmed PBC was seen in 5 patients (12.5%).92 Sometimes, patients with PBC develop symptomatology consistent with scleroderma. In a review of 558 patients with PBC, 4% were found to have symptoms of CREST syndrome.93 Clinical manifestations of CREST often antedated the development of PBC by as much as 28 years.
REYE’S SYNDROME Reye’s syndrome is an acute illness characterized by encephalopathy and fatty infiltration of the liver. It was first reported in 1963 by Reye and colleagues, who described 21 Australian children who developed loss of consciousness, vomiting, fever, and hypoglycemia that occurred shortly after a viral prodrome.94 Over 80% of patients in this series died. At autopsy, extensive fatty infiltration was noted in the liver and kidney, and to a lesser extent in the myocardium and pancreas.
700
EPIDEMIOLOGY The Centers for Disease Control (CDC) define a case of Reye’s syndrome as one in which there is: 1. acute, non-inflammatory encephalopathy, manifested clinically by alterations in the level of consciousness and documented, when such results are available, by the measurement of 8 or fewer leukocytes per cubic millimeter of cerebrospinal fluid or by the presence of cerebral edema without perivascular or meningeal inflammation in the histological section of the brain; 2. hepatopathy documented by liver biopsy or autopsy or a threefold or greater rise of AST, ALT, or serum ammonia; 3. no other more reasonable explanation for the cerebral or hepatic abnormalities.95,96 Reye’s syndrome predominantly affects children in the first decade of life, but up to 20% of patients may be older than 15 years and, thus, this condition may be seen by adult gastroenterologists and hepatologists.96–98 A study demonstrated decreasing incidence of Reye’s syndrome based on the epidemiological characteristics of 1207 cases reported to the National Reye Syndrome Surveillance System (NRSSS) from December 1, 1980 through November 30, 1997.96 A peak of 555 cases was reported in surveillance year 1980. From 1987 through 1993, no more than 36 cases were reported each year, and from 1994 through 1997, no more than 2 cases were reported each year (Figure 56-6).87 A similar decline in the incidence of Reye’s syndrome has been noted in the UK and elsewhere (Figure 56-7).98 Since the original description of Reye’s syndrome, epidemiologic studies have suggested an association between Reye’s syndrome and the use of aspirin during influenza or varicella viral infections.96,98–101 Before 1990, the incidence of Reye’s syndrome was higher in years with epidemics of influenza B than in years with epidemics of influenza A (H3N2 or H1N1), but this association was not found subsequently.96 Patients who developed Reye’s syndrome were much more likely to have received salicylates during a prodromal
Reports of possible relation between Reye’s syndrome & aspirin use
600
No. of Cases
500 400 300 200
Surgeon general’s advisory Labeling of aspirin-containing medications
100
19 74 19 77 19 78 19 79 19 80 19 81 19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97
0
1074
Figure 56-6. Number of reported cases of Reye’s syndrome in relation to the public announcement of the epidemiologic association of Reye’s syndrome with aspirin ingestion and the labeling of aspirin-containing medications. (Reproduced from Belay ED, Bresee JS, Holman RC et al. Reye’s syndrome in the United States from 1981 through 1997. N Engl J Med 340:1377, 1999, with permission.)
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS
illness than were those who did not develop the syndrome.100,101 In a previous study from the CDC, the odds ratio of developing Reye’s syndrome in association with a viral illness treated with aspirin was 16:1.101 Children chronically treated with salicylates for illnesses such as juvenile rheumatoid arthritis and Kawasaki’s disease are also at increased risk for Reye’s syndrome.96,102 Based upon these observations, in 1986, aspirin and aspirin-containing medications were labeled with an advisory not to administer to children with flu-like illnesses. The declining incidence of Reye’s syndrome in the USA and UK is temporally associated with public health warnings about the risk of Reye’s syndrome after the use of aspirin in children with varicella and influenza-like illnesses.96,98 In countries where aspirin
300 275
276
Aspirin warning in the UK began in June 1986
250 225 200 175 150 125
104
100 75
52
50 17
25 0 Aug 1981– July 1986
Aug 1986– July 1991
Aug 1991– July 1996
Aug 1996– July 1999
Figure 56-7. Number of cases of Reye’s syndrome reported to the British Paediatric Surveillance Unit. (Adapted from Hall SM, Lynn R. Reye’s syndrome. N Engl J Med 1999; 341:845, with permission.)
is inadvertently used in children with viral illnesses, Reye’s syndrome continues to be a significant problem.103 Erroneous diagnosis of Reye’s syndrome was not uncommon when presumptive criteria were used to establish the diagnosis. Several inherited metabolic disorders present with signs and symptoms that are indistinguishable from Reye’s syndrome (see below). When the charts of 49 patients originally diagnosed with Reye’s syndrome were blindly reviewed, only one case was felt to be truly due to Reye’s syndrome.104 In UK, 12.7% patients that were originally reported with Reye’s syndrome between 1981 and 1998 were subsequently diagnosed with an inherited metabolic disorder.98 Because Reye’s syndrome is now very rare, it has been suggested that any infant or children suspected of having this disorder should undergo investigation to exclude the presence of inborn metabolic disorders that can mimic Reye’s syndrome.96
HISTOPATHOLOGY AND PATHOGENESIS Liver specimens in patients with Reye’s syndrome exhibit several characteristic histologic abnormalities (Figure 56-8). Microvesicular steatosis and decreased or absent glycogen stores are common and appear to correlate with severity of illness and survival.105 There is minimal hepatocellular necrosis associated with these lesions. Hepatic glycogen content in biopsies taken within 24–48 hours of the onset of encephalopathy was significantly lower in patients who died. These lesions also appear to evolve over time so that in serial biopsy specimens glycogen stains and microvesicular steatosis may initially worsen and then improve within 1–2 weeks of the onset of encephalopathy. In survivors, liver biopsies performed 1–13 months after clinical improvement showed essentially normal histology, indicating that the hepatic lesions of Reye’s syndrome are completely reversible. Electron microscopic studies of the liver in Reye’s syndrome have also demonstrated characteristic findings that are consistent with the presumed pathogenesis of the disease. Ultrastructural study has shown that mitochondria within the hepatocytes of children with
Figure 56-8. Liver in Reye’s syndrome. In Reye’s syndrome the hepatic architecture is preserved and the portal tracts are unremarkable. The hepatocyte cytoplasm is finely vacuolated due to accumulated lipid. Note that the nuclei remain in the center of the cell, a feature of microvesicular steatosis. (Hematoxylin and eosin.)
1075
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Reye’s syndrome are swollen and irregularly shaped with expansion of the matrix space.106 These changes also regress in parallel with clinical improvement. An excellent review of metabolic abnormalities seen in Reye’s syndrome and other inherited disorders was published in 1994.107 Hyperammonemia may result from a decrease in the activity of enzymes involved in the urea cycle, carbamyl phosphate synthetase and ornithine transcarbamylase (OTC), due to damage to hepatic mitochondria. Both of these enzymes are normally found predominantly within the mitochondria; however, in patients with Reye’s syndrome, OTC activity is shifted to the cytosol, presumably due to damage to the mitochondrial membrane.108 The decreased levels of enzyme activity are transient and correlate with the clinical status of the patient.109 Increase in nitrogen load due to muscle breakdown has also been reported in patients with Reye’s syndrome and contributes to the profound hyperammonemia present in this disorder.107 The metabolism of fatty acids is also impaired in patients with Reye’s syndrome. Numerous intermediate metabolites are abnormally elevated in the liver, plasma, or urine of these patients and high levels of these compounds are believed to inhibit further mitochondrial enzymes involved in ureagenesis, gluconeogenesis, and fatty acid oxidation.107 Salicylates may inhibit mitochondrial enzymes involved in fatty acid oxidation, which may explain the association between these drugs and the development of Reye’s syndrome.107 Furthermore, in mice, infection with influenza virus potentiated the inhibitory effects of salicylates on mitochondrial fatty acid oxidation.110 Some of the deleterious effects of salicylates on mitochondrial structure and function may be cytokine-mediated; pretreatment with a-interferon ameliorated salicylate-induced damage to isolated rat liver mitochondria.111
CLINICAL FEATURES, DIFFERENTIAL DIAGNOSIS, AND TREATMENT Reye’s syndrome is predominantly seen in young children and adolescents but has been reported to occur infrequently in adults.112–114 Belay et al. found that 8% of the cases in the USA involved patients who were 15 years of age or older.96 Other illnesses, such as Jamaican vomiting sickness and drug toxicity from valproic acid or fialuridine, may present with similar changes of microvesicular steatosis. Several inborn metabolic disorders often present with signs and symptoms that are indistinguishable from Reye’s syndrome.115–117 The most common metabolic disorder mimicking Reye’s syndrome is medium-chain acyl-coenzyme A dehydrogenase deficiency.117 The majority of these inherited disorders are specific enzymatic defects that usually become evident before the age of 3 years. These disorders are characterized by recurrent episodes, the presence of similar disorder in the siblings, frequent hypoglycemia, cardiac enlargement, and muscle weakness. The most consistent distinguishing features of Reye’s syndrome on electron microscopy are ultrastructural changes in liver tissue, specifically the proliferation of smooth endoplasmic reticulum and peroxisomes, and the presence of enlarged and pleomorphic mitochondria with loss of dense granules.116–118 In enzymatic defects with Reye’s manifestations, in contrast, the mitochondria are normal in size and appearance. The illness begins with a viral prodrome that may begin to resolve. Within several days, patients present with intractable vomiting and
1076
mental status changes, ranging from irritability to mild encephalopathy with confusion and disorientation that may progress to deep coma. Serum aminotransferase levels are elevated as much as 50 times the upper limits of normal. Prothrombin time is usually prolonged and serum ammonia is elevated. Serum bilirubin levels are almost invariably normal or just slightly elevated so that jaundice is conspicuously absent in Reye’s syndrome.112 Treatment of Reye’s syndrome is generally supportive, with careful attention to hypoglycemia and electrolyte disturbances. High-concentration glucose solutions are required to maintain adequate serum glucose levels. With neurologic deterioration, consideration should be given to intracranial pressure monitoring in order to guide the effects of various interventions, such as hyperventilation and mannitol infusions, that may be used to decrease intracranial pressure. In the report from the CDC that described the outcomes of 1207 cases of Reye’s syndrome reported to NRSSS, the overall case fatality rate was 31.3%.96 The level of consciousness at the time of admission was a strong predictor of death; the mortality rate increased from 17.8% in stage 0 (alert and wakeful) to 89.6% in stage 5 patients (unarousable, areflexia, and fixed pupils) (Table 56-6). Additionally, age 5 years and serum ammonia concentration 45 mg/dl were independent predictors of mortality. Residual neurological deficits were reported in 9.8% of patients; the deficits were mild in 6.9% and severe in 2.8%.96 Patients with serum ammonia > 45 mg/dl had significantly higher risk of neurological complications (relative risk, 4.1; 95% confidence interval, 1.2–14.0).96
STAUFFER’S SYNDROME Stauffer’s syndrome is a paraneoplastic syndrome of liver test abnormalities in patients with renal cell carcinoma, first described in 1961 by Stauffer.119 In the early literature, its incidence in patients with renal cell cancer ranged from 4 to 40%.120–122 Its true incidence is difficult to ascertain from these early reports due to their inability adequately to exclude hepatic metastases as a cause for liver test abnormalities. In a recent study that reviewed the records of 365 patients with renal cell carcinoma, 21% had paraneoplastic elevation of alkaline phosphatase. Hepatic and bone metastases were excluded by computed tomography scans and bone scans.123 Recent evidence suggests that interleukin-6 (IL-6) is responsible for causing various paraneoplastic syndromes seen in renal cell
Table 56-6. Outcome of Reye’s Syndrome According to the Level of Consciousness Stage
Total
Outcome Complete recovery
0–1 2–3 4–5
505 473 128
391 (77%) 289 (61%) 15 (12%)
CNS sequelae Mild
Severe
16 (3%) 32 (6.7%) 5 (4%)
4 (0.8%) 10 (2.1%) 8 (6.2%)
Death
94 (19%) 142 (30%) 100 (78%)
CNS, central nervous system. Adapted from Belay ED, Bresee JS, Holman RC, et al. Reye’s syndrome in the United States from 1981 through 1997. N Engl J Med 1999; 340:1377, with permission.
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS
carcinoma. The following evidence supports IL-6 as the cytokine involved: 1. when given systemically, IL-6 induces findings similar to paraneoplastic syndromes associated with renal cell carcinoma (fever, elevated alkaline phosphatase);124,125 2. IL-6 is expressed by most of the renal cancer cell lines and a strong correlation exists between serum IL-6 and paraneoplastic elevation of alkaline phosphatase;126,127 3. administration of anti-IL-6 monoclonal antibodies normalized the alkaline phosphatase levels in patients with Stauffer’s syndrome.128 Strickland and Schenker reviewed 29 published cases of nephrogenic hepatic dysfunction in whom sufficient data were available to apply strict criteria for the diagnosis.129 Fever and weight loss were the most common symptoms. Hepatomegaly was present in twothirds of patients. Elevation of alkaline phosphatase was the most common biochemical abnormality, occurring in 90% of reported cases, while abnormalities of serum aminotransferases and serum bilirubin were much less common. Histologic examination of the liver in patients with nephrogenic hepatic dysfunction has shown only non-specific changes. Steatosis, mild focal hepatocyte necrosis, portal lymphocytic infiltration, and Kupffer cell hyperplasia have all been reported.120,129,130 The importance of recognizing this syndrome lies in the decision concerning resection of the tumor and differentiating these benign liver enzyme abnormalities from metastatic disease. Abnormalities of hepatic function usually resolve within 1–2 months after the primary tumor has been resected and, indeed, this is an important feature in the diagnosis of this syndrome.119–122,129,130 With recurrence of renal cell carcinoma, clinical and biochemical characteristics of nephrogenic hepatic dysfunction may again become evident.130
AMYLOIDOSIS Hepatic involvement is common in patients with systemic amyloidosis. It is disorder of abnormal protein deposition in various organs.
On histological examination, all amyloid proteins show extracellular, amorphous, eosinophilic hyaline-like material. This material stains with Congo red dye and shows apple-green birefringence when examined under polarized light. The precursor proteins, although vastly different chemically, share the conformational property of forming beta-pleated sheets as they precipitate. It is the beta-pleated sheet arrangement that leads to the characteristic histochemical findings.
PRIMARY AND SECONDARY AMYLOIDOSIS Amyloid is currently classified by placing an A in front of the abbreviation for the precursor protein (Table 56-7). There are at least 16 different variants. Many of these occur only focally in aged or tumorous organs and do not directly involve the liver. Primary or AL amyloidosis is related to the deposition of immunoglobulin light- or heavy-chain protein.131,132 This is probably the most common form of systemic amyloidosis in this country. Immunoglobulin is a normal component of the humoral immune system and can normally be found in the serum in a 2:1 ratio, kappa to lambda. In lymphoplasmocytic disorders such as multiple myeloma or Waldenströms’s macroglobulinemia, there is an overproduction of portions of the immunoglobulin molecule, generally the light chain. The light-chain protein itself or its breakdown products then precipitate out of the serum to form amyloid. Most patients with amyloidosis have precipitated lambda light chains, suggesting that this form is more likely to produce beta-pleated sheet arrangement. AA or secondary amyloid is associated with chronic infections such as osteomyelitis or tuberculosis. Cytokines associated with the inflammation, such as IL-1, IL-6, and tumor necrosis factor, stimulate the liver to produce serum amyloid A.133,134 This protein is an injury-specific component of high-density lipoprotein. The normal human AA protein exists as three isoforms and coded by three genes on the short arm of chromosome 11 (11p). This protein is continuously produced during chronic infections.135 People vary greatly in their ability to clear this protein. While most individuals cleave
Table 56-7. Different Types of Amyloidosis Variant
Precursor protein
Sites involved
Disease
AL ATTR
Immunoglobulin light chain Transthyretin
Systemic Systemic
Myeloma Hereditary
AA Ab2M AApoAI
(Apo)serum AA b2-microglobulin Apoprotein AI
Systemic Systemic Nerves, kidney, liver
Chronic infection Chronic hemodialysis Hereditary
ALys AGel AFib ACys Ab APrPsc ACal AIAPP AANF APro
Lysozyme variants Gelsolin Fibrinogen a-chain Cystatin C Ab-protein precursor Prion protein (Pro)calcitonin Islet amyloid protein Atrial natriuretic factor Prolactin
Kidney Cornea, nerves, skin Kidney Cerebral blood vessels Brain Nervous system Thyroid Islet of Langerhans Heart Aging pituitary
Hereditary Hereditary Hereditary Hereditary Alzheimer’s disease Spongiform encephalopathy Medullary carcinoma
Mutation Chromosome 18; Val30Met most common AA gene on 11p; FMF 16p Chromosome 11; Gly26Arg most common
Chromosome 21 PRNP gene Chromosome 20
Prolactinoma
1077
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
amyloid A into small peptides, some patients can only cleave it into larger sizes consistent with the amyloid subunit. Familial Mediterranean fever (FMF) is an autosomal recessive disorder associated with recurrent episodes of fever, arthritis, serositis, and skin rash. It is predominantly seen in non-Ashkenazi Jewish, and Turkish people. These patients develop renal disease and ultimately renal failure, which can be prevented with colchicine administration. Recently a genetic defect has been located on the short arm of chromosome 16 in patients with FMF.136,137 The defect occurs in a gene coding for a protein known as pyrin/marenostrin. It is unclear how this gene product interacts with the serum A protein to produce amyloidosis. Mutations in the transthyretin gene are associated with familial amyloidotic polyneuropathy (FAP).138 This is the most common type of heritable amyloidosis. The transthyretin gene product is a normal serum protein largely produced by the liver and it carries serum thyroxin and retinal-binding protein. The protein is a tetramer of identical subunits that is encoded by a single gene on chromosome 18. At least 78 different amino acid substitutions occurring at 51 different sites in the transthyretin gene have been reported.138 Most of these mutations are amyloidogenic. Although inherited as an autosomal dominant disorder, symptoms generally do not arise until the third to fourth decade of life. The disorder has been seen in Portugal, Sweden, Japan, and the USA. The US kindreds usually exhibit the defects associated with their country of origin. The clinical onset can be quite variable among these ethnic groups, even in those patients with the same mutation. The patients usually present with a lower-limb nephropathy, which progresses to the upper limbs. Autonomic nervous system involvement is also usually extensive, often resulting in diarrhea. Changes in the joints, skin, and cornea are also present. Rarely, other organs can be involved. Patients undergoing chronic hemodialysis develop amyloid secondary to precipitation of the b2-microglobulin protein.139 b2Microglobulin is a normal serum protein. The hemodialysis itself is thought to produce localized excess concentrations of this precursor, resulting in the deposition of amyloid. Generally, the musculoskeletal system and the kidneys are involved. Although amyloid is relatively homogeneous in its appearance, it is clear that its etiology can be quite variable. The mechanism of amyloid formation is likewise complex and probably various mechanisms are at work, depending on the type of amyloid. For instance, in the FAP disorders it is thought that the inheritable mutation causes beta-plated sheet-type fibrils to form upon normal proteolysis. This contrasts with secondary amyloidosis where not only overexpression of the AA protein but perhaps abnormal proteolysis results in the deposition of fibrils.
CLINICAL FEATURES, DIAGNOSIS, AND TREATMENT Amyloidosis can present as renal dysfunction/renal failure, heart failure/cardiomyopathy, peripheral neuropathy, skin rash, and blood dyscrasias.140 The most common finding related to the liver is hepatomegaly.141 Several other forms of presentation of hepatic amyloidosis are summarized in Table 56-8. Jaundice and cholestasis can be one of the initial manifestations of hepatic amyloidosis.142–148
1078
Table 56-8. Manifestations of Hepatic Amyloidosis Hepatosplenomegaly Splenomegaly Ascites Prolongation of prothrombin time Due to acquired factor X deficiency Cholestasis Jaundice Acute liver failure Spontaneous rupture
When present, it is an ominous finding and suggests a short survival time. Amyloid can also present catastrophically as liver failure or hepatic rupture.149 Ascites due to sinusoidal occlusion can also be a presenting symptom.150 Generally the transaminases are minimally elevated, although they can be quite high in the rare cases of fulminant failure. Jaundice is also infrequent except for the rare cholestatic cases. Imaging studies may show an infiltrative process but are generally not helpful in gauging the extent of disease. Demonstration of amyloid deposition is generally required to make the diagnosis. Fat pad aspirations are the easiest and most common approach to diagnose amyloidosis. If that is negative, rectal or skin biopsies are obtained, and finally biopsies of affected organs as needed. Amyloid deposition in the liver occurs in three basic patterns. The most common pattern is protein deposition in the spaces of Disse (Figure 56-9). As the material accumulates, it encroaches on the hepatocytes, causing extensive atrophy. Amyloid increases the mass of the liver, producing hepatomegaly, but only rarely compromises sinusoidal blood flow, leading to portal hypertension. Another pattern is deposition of amyloid only in the walls of hepatic arterioles and the third pattern is globular clusters of amyloid, again in the space of Disse. The latter pattern appears to be most uncommon. The Congo red stain is the mainstay of diagnosis.138,151 While commonly used as a diagnostic test, the Congo red stain can be difficult to perform correctly. The characteristic staining is highly dependent on the pH, salt concentration, and thickness of the material stained. Any variation in the above parameters can greatly affect the quality of staining. The Congo red staining is relatively specific for amyloid when properly performed; however, under some conditions, one can also see staining of fibrin, elastin, and collagen. However, these are usually not birefringent under polarized light. The staining properties with Congo red are related to the betapleated sheet arrangement of the proteins. Generally, all types of amyloid stain with the Congo red dye. However, in some patients with AL amyloid, it may be very difficult to obtain a satisfactory staining reaction. In these cases, it may be useful to attempt other stains, such as Thioflavin T and S, crystal violet, and methyl violet. Electron microscopy has been a gold standard for detection of amyloidosis. Ultrastructural examination shows non-branching fibrillar structures of indefinite length with a width of approximately 9.5 nm. Therapy is generally directed at treating the underlying condition.152,153 In the case of primary amyloid, therapy directed against the underlying lymphoplasmocytic neoplasm can sometimes improve the survival.154,155 Liver transplantation has been performed
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS Figure 56-9. Liver in amyloidosis. Amyloid is usually deposited in the space of Disse (arrows). As the deposits increase, the hepatocytes become sunken ribbons. (Hematoxylin and eosin.)
in this disorder and short-term survivors have been noted.156 Stem cell transplantation therapy has also been employed, with mixed results.157,158 However, generally the disease progresses and fatal complications from cardiac or renal failure eventually ensue. In the cases of amyloidosis due to chronic infections, eradication of underlying infection should be attempted. Colchicine is the treatment of choice in patients with FMF. Orthotopic liver transplantation has been very successful in patients with FAP.159–166 Most centers recommend transplantation in patients with FAP as soon as symptoms occur, as liver transplantation has been shown to diminish the disease progression. In general, there is little improvement in the autonomic dysfunction that is already established.165,166 Patients with advanced disease, including both upper and lower motor neuron symptoms, generally do poorly.165,166
LIVER IN HEMATOLOGIC DISEASES SICKLE-CELL ANEMIA Patients with sickle-cell anemia may have a variety of hepatic abnormalities involving both the hepatic parenchyma and the biliary tract. These abnormalities may be present during the asymptomatic phase of sickle-cell disease as well as during the episodes of sickle crisis. The incidence of hepatic involvement is very difficult to estimate due to the confounding effects of chronic hemolysis that may elevate serum bilirubin and aminotransferase activity. Of 100 consecutive patients seen as outpatients or inpatients at a university hospital, 24% had one or more chronic abnormalities of liver tests.167 The vast majority of the abnormalities were due to concurrent illnesses ranging from chronic viral hepatitis to congestive heart failure.155 Similarly, in another series, serum alkaline phosphatase was abnormal in 64% of patients with sickle-cell disease, although it appears that the alkaline phosphatase was mostly of bone origin.168 Thus, the true incidence of hepatic involvement is greatly dependent upon the criteria used to define liver disease in this population.
HEPATIC CRISIS Hepatic crisis in sickle-cell disease is characterized by right upper quadrant abdominal pain, jaundice, hepatomegaly, and fever. This constellation of findings is also suggestive of acute cholecystitis or cholangitis and differentiating these entities may be difficult. It has been estimated that 7–10% of hospitalizations for sickle-cell anemia were complicated by hepatic crises.169,170 The most marked serum biochemical abnormality during hepatic crisis is elevation in serum bilirubin. Total bilirubin is usually less than 15 mg/dl, although extreme levels of hyperbilirubinemia, greater than 50 mg/dl, may occur. This marked hyperbilirubinemia may not necessarily be associated with a worse prognosis.171–173 A large component of this bilirubin is the direct fraction, which may be as much as 50% of the total bilirubin in many cases. Serum aminotransferases are abnormal, with levels usually less than 10 times the upper limits of normal. Elevation of serum LDH activity, disproportionate to the increase of serum aminotransferases, is also common and reflects the ongoing hemolysis associated with sickle crisis.168 Evidence of extrahepatic sickle crisis, such as joint and flank pain, is usually, but not invariably, present. With general supportive care, clinical improvement is seen within several days, although hyperbilirubinemia may persist for several weeks. Deaths related to fulminant hepatic failure in the absence of other identifiable etiologies have been reported.168 Recent data suggest that acute hepatopathy due to sickle-cell anemia should be considered a contraindication to percutaneous liver biopsy.174 Out of 14 patients with sickle-cell disease who had percutaneous liver biopsy, 5 (36%) had serious hemorrhage and 4 of them died due to bleeding complications. All of those who had bleeding complications had acute sickling crisis at the time of their liver biopsy.
BILIARY TRACT DISEASE Pigment gallstones are frequently found in patients with sickle-cell anemia: estimates of the prevalence range between 40 and 80%.168,175–177 The incidence varies directly with the age of the
1079
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
patient.168 Choledocholithiasis was found in 18% of 65 patients undergoing cholecystectomy.168 Hepatic crisis will resolve quickly with supportive care while cholecystitis and cholangitis will ultimately require endoscopic and/or surgical intervention. Therefore, establishing the correct diagnosis is especially important. There is a suggestion that the operative complications of cholecystectomy in patients with sickle-cell anemia are greater than in the general population.165 Schubert reviewed published reports on 97 patients who underwent cholecystectomy and found 15% of patients had complications that were deemed serious, including pneumonia, seizures, and sickle crisis.168,178 Thus, cholecystectomy should be reserved for those patients with documented gallstones whose symptoms are clearly of biliary tract origin or for those in whom hepatic crisis and biliary tract disease cannot be adequately differentiated.168,178 Careful preoperative management with special attention to transfusion requirements and volume status is important.
erythrocyte sickling, and erythrophagocytosis was found in almost all patients (Figure 56-10).180 In a study of pregnant women given prophylactic red cell transfusions, erythrocyte sickling and erythrophagocytosis were present in all patients in the absence of hepatic crisis.184 Thus, these changes appear to be characteristic of sickle-cell anemia and do not correlate with aminotransferase levels or activity of liver disease. The presence of other histologic lesions may also be detected in the livers of patients with sickle-cell anemia and further confuse the clinical picture. Massive iron deposition is frequently identified by routine iron stains in the majority of patients.179,180,184 Also, of 19 patients with sickle-cell anemia on whom biopsies were performed because of abnormal liver tests, 9 (47%) had changes consistent with acute or chronic viral hepatitis.180 Cirrhosis has been reported to occur in 15–20% of patients with sickle-cell anemia and may be due to hypoxic injury from sickling and intrasinusoidal sludging of erythrocytes, chronic viral hepatitis, or massive hemosiderosis.168,180,186
VIRAL HEPATITIS As expected from the large numbers of transfusions required by many patients with sickle-cell anemia, viral hepatitis has been reported to occur, although from published reports, the incidence is difficult to determine. In patients with liver test abnormalities, the incidence of hepatitis B virus (HBV) or histologic changes of chronic hepatitis have ranged from 5 to 47%.167,175,179,180 The reported prevalence of hepatitis C in patients with sickle-cell anemia is 10–35%.181–183 Not surprisingly, the risk of hepatitis C was directly related to the number of transfusions received. The impact of hepatitis C infection on the natural history of patients with sickle-cell anemia is currently unknown.
HEPATIC HISTOLOGY Several studies have evaluated the histopathologic changes in the livers of patients with sickle-cell disease.171,180,184,185 These studies included biopsies from patients during hepatic crisis as well as biopsies during quiescent periods. Hepatic sinusoidal distention,
HODGKIN’S LYMPHOMA Hepatic involvement in Hodgkin’s lymphoma has been reported to occur in 5% of patients at the time of diagnosis, 30% during the course of the disease, and up to 50% at the time of autopsy.187 The histology of liver involvement in malignant lymphomas has been reviewed.174 The extent of tissue sampling correlates with the ability to stage hepatic involvement in Hodgkin’s disease accurately; percutaneous liver biopsy has the lowest yield while laparoscopy and laparotomy have similar yields. A diagnosis of hepatic involvement in Hodgkin’s disease requires the finding of the Reed–Sternberg cell (Figure 56-11). Non-specific inflammatory infiltrates are seen in 50% of liver biopsies in patients with Hodgkin’s disease and, alone, do not constitute grounds for diagnosing hepatic involvement. Noncaseating epithelioid granulomas may be seen in 25% of patients with Hodgkin’s disease.188,189 Granulomas may be seen in the portal tract and the hepatic lobules and do not necessarily indicate hepatic involvement by Hodgkin’s disease. Figure 56-10. Liver in sickle-cell anemia. Numerous sickled red blood cells distend the sinuses of the liver (arrow). (Hematoxylin and eosin.)
1080
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS Figure 56-11. Liver in Hodgkin’s lymphoma. Classic Reed–Sternberg cell (arrow) is seen in a polymorphous background of lymphocytes, plasma cell, and eosinophils which greatly expands a portal tract. (Hematoxylin and eosin.) The inset shows the Reed–Sternberg cells marked with anti-CD-30. (Modified immunoperoxidase method.)
Elevation of serum alkaline phosphatase levels is the most frequently abnormal liver test found in patients with Hodgkin’s disease. In a review of 111 inpatients with Hodgkin’s disease, 41% had abnormal serum alkaline phosphatase levels.190 Patients with more advanced stages of Hodgkin’s disease were more likely to have elevations of this enzyme; 14% of patients with stage I or II disease and 65% and 81% of patients with stage III or stage IV disease, respectively, had abnormal serum alkaline phosphatase levels. The elevations were generally mild – 1.5–2 times the upper limits of normal – although more marked increases were noted in patients with stage IV disease. The liver was felt to be the source of the alkaline phosphatase in the majority of patients, although several patients, adolescents with potential for bone growth, had elevation in the bone fraction.190 Abnormalities in alkaline phosphatase were seen in several patients in the absence of hepatic involvement, more commonly in patients with fever as a systemic manifestation of Hodgkin’s disease. Thus, in some patients abnormal serum alkaline phosphatase could represent the equivalent of an acute-phase response. Jaundice occurs infrequently in Hodgkin’s lymphoma, except in the late stages of the illness. The most frequent cause of jaundice is intrahepatic infiltration by the tumor, which was seen in 45% of jaundiced patients at the time of autopsy.187,191,192 Extrahepatic biliary tract obstruction occurs much less frequently and accounts for only 5–10% of jaundiced patients.191–193 A small number of patients with Hodgkin’s lymphoma have been described who have evidence of severe intrahepatic cholestasis with dramatic elevations in serum bilirubin and alkaline phosphatase levels in the absence of tumor infiltration or bile duct obstruction.173,179 The etiology of this syndrome is not known but one report suggested that it could be related to vanishing bile duct syndrome.194,195 An association between primary sclerosing cholangitis and Hodgkin’s lymphoma has also been suggested.196 Acute liver failure with encephalopathy, jaundice, and coagulopathy has also been reported in patients with Hodgkin’s and non-Hodgkin’s lymphoma either due to direct hepatic involvement,197,198 or as a paraneoplastic syndrome.199
NON-HODGKIN’S LYMPHOMA Hepatic involvement in non-Hodgkin’s lymphomas occurs very frequently, with estimates ranging between 24 and 43%.200,201 The hepatic infiltrate usually involves the portal triads and has a nodular appearance (Figure 56-12). Epithelioid granulomas may also be seen in the liver of these patients. Immunophenotyping using monoclonal antibodies may be performed on snap-frozen liver biopsy tissues in order to characterize the infiltrates.202 Rarely, primary hepatic lymphoma in the absence of systemic lymphoma has been reported.203,204 The clinical manifestations of hepatic infiltration with nonHodgkin’s lymphoma are similar to that seen with Hodgkin’s disease. Patients may remain asymptomatic despite extensive hepatic infiltration. Mild to moderate elevations in serum alkaline phosphatase and moderate to marked elevations of LDH activities may be present. In contrast to Hodgkin’s disease, non-Hodgkin’s lymphomas are more likely to produce jaundice as a result of extrahepatic obstruction, usually at the porta hepatis, rather than by direct hepatic infiltration.187 Reactivation of chronic hepatitis B in patients receiving cytotoxic therapy is well documented.205,206 Recent data suggest that lamivudine can prevent such reactivation of hepatitis B in patients receiving chemotherapy.207 While some reports have suggested that HCV may have a role in the development of non-Hodgkin’s lymphoma,208,209 other reports have failed to confirm such an association.210,211 Patients with lymphoma with severe liver dysfunction (due to either lymphoma involvement or coexisting liver problem) pose a significant therapeutic problem due to their inability to tolerate conventional chemotherapeutic agents. Ghobrial and colleagues summarized their experience with 41 such patients seen at the Mayo Clinic over a 5-year period. The authors found that mechlorethamine, high-dose corticosteroids, and rituximab are safe and effective in this patient population.212
MULTIPLE MYELOMA Liver may be directly involved by plasma cell infiltrate in up to 30% patients with multiple myeloma. The patients generally present with
1081
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Figure 56-12. Hepatic B-cell lymphoma. Sheets of small lymphocytes surround a regenerative liver nodule. The portal tract also appears to be involved. The B-cell nature of the infiltrate can be confirmed by immunohistochemistry or flow cytometry. (Hematoxylin and eosin.)
A
B
Figure 56-13. (A) Hepatic mastocytosis. (Hematoxylin and eosin.) It may involve the portal tracts, lobule, or both. The lesion is a stellate region of fibrosis occupied by lymphocytes, eosinophils, and cells with larger pale nuclei – the mast cell. (B) Hepatic mastocytosis (Leder). In this section of the liver the numerous mast cells (red) of systemic mastocytosis stand out against the background counterstain (chloroacetate esterase (Leder) stain).
hepatomegaly or ascites.213 It is important to distinguish myeloma liver involvement with myeloma from autoimmune hepatitis. In myeloma, the plasma cell infiltrate predominantly involves the sinusoids, with relative sparing of the portal tracts. This is distinct from autoimmune hepatitis, where the infiltrate is predominantly in the portal area. Another pattern of liver injury in myeloma patients is nodular regenerative hyperplasia.214 Waldenström’s macroglobulinemia is a neoplastic disorder of Blymphocytes and plasma cells that has been alluded to previously.215 The neoplastic cells secrete immunoglobulin heavy chain, which has rarely been associated with the development of amyloidosis. The lymphoplasmacytic tumor can directly involve the liver. Clinical liver disease in these patients is often mild, with minimal elevation of transaminases and alkaline phosphatase. Their symptoms are usually referred to the extrahepatic problems associated with the disease. Histologically, the infiltrate shows expanded portal tracts
1082
with lymphocytes and plasma cells, and larger atypical cells with occasional mitotic figures.
MASTOCYTOSIS Mastocytosis is commonly seen in children as a skin rash, urticaria pigmentosa. However, the disease can become systemic and involve the liver, especially in adult patients.216 When seen in adults, it often presents with fever, hepatosplenomegaly, steatorrhea or diarrhea, and weight loss. The liver chemistries are usually minimally abnormal; radiologic studies are often not contributory. A liver biopsy demonstrates the characteristic lesions that are similar to those seen in other organs, including spleen and lymph nodes. This lesion is characterized by irregular areas of fibrosis containing numerous eosinophils and a background of lymphocytes and other mononuclear cells (Figure 56-13).217 These mononuclear cells are mast cells that may be recognized by a number of methods. Because the
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS
REFERENCES
Figure 56-14. Lobular portion of liver from a patient with severe acute respiratory syndrome (SARS) showing marked regenerative activity characterized by numerous mitotic figures (arrows). (Reprinted from Chau TN, Lee KC, Yao H, et al. SARS-associated viral hepatitis caused by a novel coronavirus: report of three cases. Hepatology 2004; 39;302, with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons.)
cells are often degranulated, toluidine blue staining may not be helpful. Chloroacetate esterase staining, however, is usually diagnostic; the cells also exhibit c-kit (CD-117). These fibrotic clusters may be located either in the portal tracts or in the parenchyma. When the lesions abut the central veins, they may be responsible for the rare cases of portal hypertension associated with the disease.218,219
LIVER MANIFESTATIONS IN SARS Severe acute respiratory syndrome (SARS) is a serious infectious disease that spreads via airborne droplets and can result in severe acute pulmonary inflammation and epithelial damage. Studies have confirmed that a novel corona virus called SARS corona virus (SCoV) is the etiologic agent. Liver test abnormalities are seen in up to 50–60% of patients who were hospitalized with SARS, and the most common liver test abnormality is mild elevation in serum transaminases.220–223 Transaminase levels peak during the second week of illness and their levels improve with successful recovery.223 This phenomenon of mildly elevated transaminases is likely a nonspecific response to severe systemic illness with hypoxemia and possibly related to multiple medications. Occasionally, transaminase levels may be significantly elevated with peak ALT values ranging between 500 and 1000 IU/l.224 Chau and colleagues described the liver histology in 3 patients with SARS who had markedly elevated transaminases with no evidence of organ failure.224 All three biopsies had hepatocyte apoptosis and in 2 patients there were marked accumulation of cells in mitosis (Figure 56-14). Liver biopsies also exhibited balloon degeneration of the hepatocytes and mild to moderate lymphocytic infiltration. Reverse transcriptase polymerase chain reaction was positive for SCoV in liver tissue from all 3 patients, suggesting that this pattern of liver injury may indeed represent a form of SARS-associated viral hepatitis.225
1. Lautt WW, Greenway CV. Conceptual review of the hepatic vascular bed. Hepatology 1987; 7:952. 2. Richman SM, Delman AJ, Grob D. Alterations in indices of liver function in congestive heart failure with particular reference to serum enzymes. Am J Med 1961; 30:211. 3. Runyon BA. Cardiac ascites: a characterization. J Clin Gastroenterol 1988; 10:410. 4. Hickman PE, Potter JM. Mortality associated with ischemic hepatitis. Aust NZ J Med 1990; 20:32. 5. Fuchs S, Bogomolski-Yahalom V, Paltiel O, et al. Ischemic hepatitis. Clinical and laboratory observations of 34 patients. J Clin Gastroenterol 1998; 26:183. 6. Bacon BR, Joshi SN, Granger DN. Ischemia, congestive heart failure, Budd–Chiari syndrome, and veno-occlusive disease. In Kaplowitz N, ed. Liver and biliary diseases, 2nd edn. Philadelphia: Williams & Wilkins, 1996:421. 7. Garland JS, Werlin SL, Rice TB. Ischemic hepatitis in children, diagnosis and clinical course. Crit Care 1988; Med 16:1209. 8. Naschitz JE, Slobodin G, Lewis RJ, et al. Heart diseases affecting the liver and liver diseases affecting the heart. Am Heart J 2000; 140:111. 9. Naschitz JE, Yeshurun D, Shahar J. Cardiogenic hepatorenal syndrome. Angiology 1990; 41:893. 10. Seeto R, Fenn B, Rockey DC. Ischemic hepatitis: clinical presentation and pathogenesis. Am J Med 2000; 109: 109. 11. Henrion J, Schapira M, Luwaert R, et al. Hypoxic hepatitis. Clinical and hemodynamic study in 142 consecutive cases. Medicine 2003; 82:392. 12. Kew M, Bersohn I, Seftel H, et al. Liver damage in heat stroke. Am J Med 1970; 49:192. 13. Hassanein T, Razack A, Gavaler J, et al. Heat stroke: its clinical and pathological presentation, with particular attention to the liver. Am J Gastroenterol 1992; 87:1382. 14. Rubel LR, Ishak KG. The liver in fatal exertional heat stroke. Liver 1983; 3:249. 15. Deviere J, Reuse C, Askenasi R. Ischemic pancreatitis and hepatitis secondary to ergotamine poisoning. J Clin Gastroenterol 1987; 9:350. 16. Ussery XT, Henar EL, Black DD, et al. Acute liver injury after protracted seizures in children. J Pediatr Gastroenterol Nutr 1989; 9:421. 17. Gibson PR, Dudley FJ. Ischemic hepatitis: clinical features, diagnosis, and prognosis. Aust NZ J Med 1984; 14:822. 18. Gitlin N, Serio KM. Ischemic hepatitis: widening horizons. Am J Gastroenterol 1992; 87:831. 19. Bynum TE, Boitnott JK, Maddrey WC. Ischemic hepatitis. Dig Dis Sci 1979; 24:129. 20. Rhodes A, Eastwood JB, Smith SA. Early acute hepatitis with parenteral amiodarone: a toxic effect of the vehicle? Gut 1993; 34:565. 21. Tosetti C, Ongari M, Evangelisti A, et al. Acute hepatotoxicity from amiodarone. Minerva Medica 1995; 86:387. 22. Potter JM, Hickman PE. Cardiodepressant drugs and the high mortality rate associated with ischemic hepatitis. Crit Care Med 1992; 20:474. 23. Bonkovsky HL, Kane RE, Jones DP, et al. Acute hepatic and renal toxicity from low doses of acetaminophen in the absence of alcohol abuse or malnutrition: evidence for increased susceptibility to drug toxicity due to cardiopulmonary and renal insufficiency. Hepatology 1994; 19:1141. 24. Lowe MD, Harcombe AA, Grace AA, et al. Restrictiveconstrictive heart failure masquerading as liver disease. Br Med J 1999; 318:585.
1083
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
25. Van der Merwe S, Dens J, Daenen W, et al. Pericardial disease is often not recognized as a cause of chronic severe ascites. J Hepatol 2000; 32;164. 26. Arora A, Seth S, Acharya SK, et al. Hepatic coma as a presenting feature of constrictive pericarditis. Am J Gastroenterol 1993; 88:430. 27. Manga P, Vythilingum S, Mitha AS. Pulsatile hepatomegaly in constrictive pericarditis. Br Heart J 1984; 52:465. 28. Braunwald E. Disorders of cardiovascular disease. In: Fauci AS, Braunwald E, Isselbacher KJ, et al., eds. Harrison’s principles of internal medicine, 15th edn. New York: McGraw Hill; 2005:1301–1495. 29. Solano FX, Young E, Talamo TS, et al. Constrictive pericarditis mimicking Budd–Chiari syndrome. Am J Med 1986; 80:113. 30. Arora A, Tandon N, Sharma MP, et al. Constrictive pericarditis masquerading as Budd–Chiari syndrome. J Clin Gastroenterol 1991; 13:178. 31. Tagami H, Utoh J, Kunitomo R, et al. Hepatic surgery after surgical repair of constrictive pericarditis. Jpn J Thorac Surg 2000; 53:567. 32. Whelan G, Pierce AK, Schenker S, et al. Hepatic function in patients with hypoxaemia due to chronic pulmonary disease. Aust Ann Med 1969; 18:243. 33. Mathurin P, Durand F, Ganne N, et al. Ischemic hepatitis due to obstructive sleep apnea. Gastroenterology 1995; 109:1682. 34. Henrion J, Colin L, Schapira M, et al. Hypoxic hepatitis caused by severe hypoxemia from obstructive sleep apnea. J Clin Gastroenterol 1997; 24:245. 35. Henrion J, Minette P, Colin L, et al. Hypoxic hepatitis caused by acute exacerbation of chronic respiratory failure: a casecontrolled, hemodynamic study of 17 consecutive cases. Hepatology 1999; 29:427. 36. Chin K, Nakamura T, Takahashi K, et al. Effects of obstructive sleep apnea on serum aminotransferase levels in obese patients. Am J Med 2003; 114:370. 37. Scott HW, Law DH, Sandstead HH, et al. Jejunoileal shunt in surgical treatment of morbid obesity. Ann Surg 1970; 171:770. 38. Bo-Linn GW. Obesity, anorexia nervosa, bulimia, and other eating disorders. In: Slesinger MH, Fordtran JS, eds. Gastrointestinal disease, 5th edn. Philadelphia: WB Saunders; 1993:2109–2136. 39. Jewell WR, Hermreck AS, Hardin CA. Complications of jejunoileal bypass for morbid obesity. Arch Surg 1975; 110:1039. 40. Kirkpatrick JR. Jejunoileal bypass: a legacy of late complications. Arch Surg 1987; 122:610. 41. Dean P, Joshi S, Kaminski D. Long-term outcome of reversal of small intestinal bypass operations. Am J Surg 1990; 159:118. 42. Hamilton DL, Vest TK, Brown BS, et al. Liver injury with alcoholic like hyaline after gastroplasty for morbid obesity. Gastroenterology 1983; 85:722. 43. Moxley RT, Pozefsky T, Lockwood DH. Protein nutrition and liver disease after jejunoileal bypass for morbid obesity. N Engl J Med 1974; 290:921. 44. Hollenbeck JI, O’Leary JP, Maher JW, Woodward ER. An etiologic basis for fatty liver after jejunoileal bypass. J Surg Res 1975; 18:83. 45. Lewin MR, Aranjo JG, Sarmiento JL, et al. Bypass induced liver disease: an experimental study of post-operative protein supplementation and metronidazole therapy in an animal model. Br J Exp Pathol 1987; 68:15. 46. Cope K, Risby T, Diehl AM. Increased gastrointestinal ethanol production in obese mice: implications for fatty liver disease pathogenesis. Gastroenterology 2000; 119:1340. 47. Drenick EJ, Fisler J, Johnson D. Hepatic steatosis after intestinal bypass – prevention and reversal by metronidazole, irrespective of protein calorie malnutrition. Gastroenterology 1982; 82:535.
1084
48. Peters RL, Gay T, Reynolds TB. Post-jejunoileal bypass hepatic disease. Am J Clin Pathol 1975; 63:318. 49. Snover DC. Biopsy diagnosis of liver disease. Baltimore: Williams & Wilkins, 1992:183. 50. Vyberg M, Ravn V, Anderson B. Pattern of progression of liver injury following jejunoileal bypass for morbid obesity. Liver 1987; 7:271. 51. Nasrallah SM, Wills CE, Galambos JT. Liver injury following jejunoileal bypass: are there markers? Ann Surg 1980; 192:726. 52. Weismann RE, Johnson RE. Fatal hepatic failure after jejunoileal bypass. Am J Surg 1977; 134:253. 53. Requarth JA, Burchard KW, Colacchio TA, et al. Long-term morbidity following jejunoileal bypass. The continuing potential need for surgical reversal. Arch Surg 1995; 130:318. 54. Brown RG, O’Leary JP, Woodward ER. Hepatic effects of jejunoileal bypass for morbid obesity. Am J Surg 1974; 127:53. 55. Soyer MT, Ceballos R, Aldrete JS. Reversibility of severe hepatic damage caused by jejunoileal bypass after re-establishment of normal intestinal continuity. Surgery 1976; 79:601. 56. Burke GW, Cirocco R, Hensley G, et al. Liver transplantation for cirrhosis following jejuno-ileal bypass – regional cytokine differences associated with pathological changes in the transplant liver. Transplantation 1992; 54:374. 57. Markowtiz JS, Seu P, Goss JA, et al. Liver transplantation for decompensated cirrhosis after jejunoileal bypass: a strategy for management. Transplantation 1998; 65:570. 58. Lowell JA, Shenoy S, Ghalib R, et al. Liver transplantation after jejunoileal bypass for morbid obesity. J Am Coll Surg 1997; 185:123. 59. D’Souza-Gburek SM, Batts KP, Nikias GA, et al. Liver transplantation for jejunoileal bypass-associated cirrhosis: allograft histology in the setting of an intact bypassed limb. Liver Trans Surg 1997; 3:23. 60. Abraham S, Begum S, Isenberg D. Hepatic manifestations of autoimmune rheumatic diseases. Ann Rheum Dis 2004; 63;123. 61. Schur PH. Clinical features of SLE. In Kelley WN, Harris ED, Ruddy S, Sledge CB, eds. Textbook of rheumatology. Philadelphia: WB Saunders; 1993:1017. 62. Runyon BA, LaBrecque DR, Anuras S. The spectrum of liver disease in systemic lupus erythematosus. Am J Med 1980; 69:187. 63. Matsumoto T, Yoshimine T, Shimouchi K, et al. The liver in systemic lupus erthematosus: pathologic analysis of 52 cases and review of Japanese autopsy registry data. Hum Pathol 1992; 23:1151. 64. Tsuji T, Ohno S, Ishigatsubo Y. Liver manifestations in systemic lupus erythematosus: high incidence of hemophagocytic syndrome. J Rheum 2002; 29:1576. 65. Maeshima E, Kobayashi T, Mune M, Yukawa S. Systemic lupus erythematosus with haemophagocytosis and severe liver disorder. Ann Rheum Dis 2002; 61:753. 66. Seaman WE, Ishak KG, Plotz PH. Aspirin induced hepatotoxicity in patients with systemic lupus erythematosus. Ann Intern Med 1974; 80:1. 67. Farrell GC. Drug-induced liver disease. London: Churchill Livingstone; 1994:372–376. 68. Schousboe JT, Koch AE, Chang RW. Chronic lupus peritonitis with ascites: review of the literature with a case report. Semin Arthritis Rheum 1988; 18:121. 69. Thompson PE, Houghton BJ, Clifford C, et al. The source and significance of raised serum enzymes in rheumatoid arthritis. Q J Med 1990; 280:869. 70. Rau R, Pfenninger K, Boni A. Proceedings: liver function tests and liver biopsies in patients with rheumatoid arthritis. Ann Rheum Dis 1975; 34:198. 71. Cimmino MA, Buffrini L, Barisone G, et al. Alkaline phosphatase activity in the serum of patients with rheumatoid arthritis. Z Rheumatol 1990; 49:143.
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS
72. Cimmino MA, Accardo S. Changes in the isoenzyme pattern of alkaline phosphatase in patients with rheumatoid arthritis. Clin Chem 1990; 36:1376. 73. Ruderman EM, Crawford JM, Maier A, et al. Histologic liver abnormalities in an autopsy series of patients rheumatoid arthritis. Br J Rheum 1997; 36:210. 74. Wanless IR, Godwin TA, Allen F, Feder A. Nodular regenerative hyperplasia of the liver in hematologic disorders: a possible response to obliterative portal venopathy. Medicine 1980; 59:367. 75. Favreau M, Tannenbasum H, Lough J. Hepatic toxicity associated with gold therapy. Ann Intern Med 1977; 87:717. 76. Hanissian AS, Rothschild BM, Kaplan S. Gold: hepatotoxic and cholestatic reactions. Clin Rheumatol 1985; 4:183. 77. Farrell GC. Drug-induced liver disease. London: Churchill Livingstone, 1994:388. 78. Fleischner GM, Morecki R, Hanaichi T, et al. Light and electron microscopical studies of a case of gold salt-induced hepatotoxicity. Hepatology 1991; 14:422. 79. Tsutsumi A, Koike T. Hepatic manifestations of the antiphospholipid syndrome. Intern Med 2000; 39:6. 80. Asherson RA, Khamashta MA, Hughes GR. The hepatic complications of the antiphospholipid antibodies. Clin Exp Rheumatol 1991; 9:341. 81. Pelletier S, Landi B, Piette C, et al. Antiphospholipid syndrome as the second cause of non-tumorous Budd–Chiari syndrome. J Hepatol 1994; 21:76. 82. Haram K, Trovik J, Sandset PM, Hordnes K. Severe syndrome of hemolysis, elevated liver enzymes and low platelets (HELLP) in the 18th week of pregnancy associated with antiphospholipid-antibody syndrome. Acta Obst Gynecol Scand 2003; 82:679. 83. Roberts G, Gordon MM, Porter D, et al. Acute renal failure complicating HELLP syndrome, SLE and antiphospholipidantibody syndrome: successful outcome using plasma exchange therapy. Lupus 2003; 12:251. 84. Skopouli FN, Barbatis C, Moutsopoulos HM. Liver involvement in primary Sjögren’s syndrome. Br J Rheumatol 1994; 33:745. 85. Haddad J, Deny P, Munz-Gotheil C, et al. Lymphocytic sialadenitis of Sjögren’s syndrome associated with chronic hepatitis C virus liver disease. Lancet 1992; 339:321. 86. Ramos-Casals M, Garcia-Carrasco M, Cervera R, et al. Sjögren’s syndrome and hepatitis C virus. Clin Rheumatol 1999; 18:93. 87. Poet JL, Tonelli-Serabian Y, Garnier PP. Chronic hepatitis C and Sjögren’s syndrome. J Rheumatol 1994; 21:1376. 88. Pirisi M, Scott C, Fabris C, et al. Mild sialoadenitis: a common finding in patients with hepatitis C virus infection. Scand J Gastroenterol 1994; 29:940. 89. Pawlotsky JM, Roudot-Thoraval F, Simmonds P, et al. Extrahepatic immunologic manifestations in chronic hepatitis C virus serotypes. Ann Intern Med 1995; 122:169. 90. Loustaud-Ratti V, Richi A, Liozan E, et al. Prevalence and characteristics of Sjogren’s syndrome or Sicca syndrome in chronic hepatitis C virus infection: a prospective study. J Rheumatol 2001; 28:2245. 91. Reynolds TB, Denison EK, Frankl HD, et al. Primary biliary cirrhosis with scleroderma, Raynaud’s phenomenon, and telangiectasia. Am J Med 1970; 50:302. 92. Goring HD, Panzer M, Lakotta W, et al. Coincidence of scleroderma and primary biliary cirrhosis: results of a systematic study of a dermatologic patient sample. Hautarzt 1998; 49:361. 93. Powell FC, Schroter AL, Dickinson ER. Primary biliary cirrhosis and the CREST syndrome: a report of 22 cases. Q J Med 1987; 62:75. 94. Reye RDK, Morgan G, Baral J. Encephalopathy and fatty degeneration of the viscera: a disease entity in childhood. Lancet 1963; 2:749.
95. Hurwitz ES, Nelson DB, Davis C, et al. National surveillance for Reye’s syndrome: a five year review. Pediatrics 1982; 70:895. 96. Belay ED, Bresee JS, Holman RC, et al. Reye’s syndrome in the United States from 1981 through 1997. N Engl J Med 1999; 340:1377. 97. MMWR 1987; 36:689. 98. Hall SM, Lynn R. Reye’s syndrome. N Engl J Med 1999; 341:845. 99. Starko KM, Ray CG, Dominguez LB, et al. Reye’s syndrome and salicylate use. Pediatrics 1980; 6:859. 100. Walkmann RJ, Hall WN, McGee H, Van Amburg G. Aspirin as a risk factor in Reye’s syndrome. JAMA 1982; 247:3089. 101. Hurwitz ES, Barrett MJ, Bregman D, et al. Public health service study of Reye’s syndrome and medications: report of a pilot phase. N Engl J Med 1985; 313:849. 102. Remington PL, Shabino CL, McGee H, et al. Reye syndrome and juvenile rheumatoid arthritis in Michigan. Am J Dis Child 1985; 139:870. 103. Ghosh D, Dhadwal D, Aggarwal A, et al. Investigation of an epidemic of Reye’s syndrome in northern region of India. Ind Pediatr 1999; 36:1097. 104. Gauthier M, Guay J, LaCroix J, Lortie A. Reye’s syndrome: a reappraisal of diagnosis in 49 presumptive cases. Am J Dis Child 1989; 143:1181. 105. Bove KE, McAdams AJ, Partin JC, et al. The hepatic lesion in Reye’s syndrome. Gastroenterology 1975; 69:685. 106. Partin JC, Schubert WK, Partin JS. Mitochondrial ultrastructure in Reye’s syndrome (encephalopathy and fatty degeneration of the viscera). N Engl J Med 1971; 285:1339. 107. Teem WR. Inherited and acquired syndromes of hyperammonemia and encephalopathy in children. Semin Liver Dis 1994; 14:236. 108. Teem WR. Inherited and acquired syndromes of hyperammonemia and encephalopathy in children. Semin Liver Dis 1994; 14:236. 109. Woodfin BM, Davis LE. Displacement of hepatic carbamoyl transferase activity from mitochondria to cytosol in Reye’s syndrome. Biochem Med Metab Biol 1991; 46:255. 110. Trauner DA, Horvath E, David LE. Inhibition of fatty acid boxidation by influenza B virus and salicylic acid in mice: implications for Reye’s syndrome. Neurology 1988; 38:239. 111. Tomoda T, Taked K, Jurashige T, et al. Experimental study on Reye’s syndrome: inhibitory effect of interferon alfa on acetylsalicylate-induced injury to rat liver mitochondria. Metabolism 1992; 41:887. 112. Meythaler JM, Varma RR. Reye’s syndrome in adults. Arch Intern Med 1987; 147:61. 113. Peters J, Wiener GJ, Gilliam J, et al. Reye’s syndrome in adults: a case report and review of the literature. Arch Intern Med 1986; 146:2401. 114. Kirkpatrick DB, Ottoson C, Bateman LL. Reye’s syndrome in an adult patient. West J Med 1986; 144:223. 115. Chang PF, Huang SF, Hwu WL, et al. Metabolic disorders mimicking Reye’s syndrome. J Formosan Med Assoc 2000; 99:295. 116. Green A, Hall SM. Investigations of metabolic disorders resembling Reye’s syndrome. Arch Dis Child 1995; 145:964. 117. Clayton PT, Doig M, Ghafari S, et al. Screening for medium chain acyl-CoA dehydrogenase deficiency using electrospray ionization tandem mass spectrometry. Arch Dis Child 1998; 79:109. 118. Partin JS, Daugherty CC, McAdams AJ, et al. A comparison of liver ultrastructure in salicylate intoxication and Reye’s syndrome. Hepatology 1984; 4:687. 119. Stauffer MH. Nephrogenic hepatosplenomegaly (abstract). Gastroenterology 1961; 40:694.
1085
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
120. Utz DC, Warren MM, Gregg JA, Ludwig J. Reversible hepatic dysfunction associated with hypernephroma. Mayo Clin Proc 1970; 45:161. 121. Ramos CV, Taylor HB. Hepatic dysfunction associated with renal cell carcinoma. Cancer 1972; 29:1287. 122. Walsh PN, Kissane JM. Nonmetastatic hypernephroma with reversible hepatic dysfunction. Arch Intern Med 1968; 122:214. 123. Chuang YC, Lin ATL, Chen KK, et al. Paraneoplastic elevation of serum alkaline phosphatase in renal cell carcinoma: incidence and implication on prognosis. J Urol 1997; 158:1684. 124. Weber J, Gunn H, Yang J, et al. A phase I trial of intravenous interleukin-6 in patients with advanced cancer. J Immunother Emph Tumor Immunol 1994; 15:292. 125. Nieken J, Mulder NH, Buter J, et al. Recombinant human interleukin-6 induces a rapid and reversible anemia in cancer patients. Blood 1995; 86:900. 126. Walther MM, Johnson B, Culley D, et al. Serum interluekin-6 levels in metastatic renal cell carcinoma before treatment with interleukin-6 correlates with paraneoplastic syndromes but not patient survival. J Urol 1998; 159:718. 127. Tsukamoto T, Kumamoto Y, Miyao N, et al. Interleukin-6 in renal carcinoma. J Urol 1992; 148:1778. 128. Blay JY, Rossi JF, Wijdenes J, et al. Role of interleukin-6 in the paraneoplastic inflammatory syndrome associated with renal-cell carcinoma. Int J Cancer 1997; 72:424. 129. Strickland RC, Schenker S. The nephrogenic hepatic dysfunction syndrome: a review. Dig Dis 1977; 22:49. 130. Jacobi GH, Philipp T. Stauffer’s syndrome: diagnostic help in hypernephroma. Clin Nephrol 1975; 4:113. 131. Gertz MA, Lacy MQ, Dispenzieri A. Amyloidosis. Hematol Oncol Clin North Am 1999; 13:1211. 132. Gillmore JD, Lovat LB, Hawkins PN. Amyloidosis and the liver. J Hepatol 1999; 30:117. 133. Rienhoff HY Jr, Huang JH, Li XX, Liao WS. Molecular and cellular biology of serum amyloid A. Mol Biol Med 1990; 7: 282. 134. Pepys MB, Baltz ML. Acute phase proteins with special reference to C-reactive protein and related proteins (pentaxins) and serum amyloid A protein. Adv Immunol 1983; 34:141. 135. Urieli-Shoval S, Linke RP, Matzner Y. Expression and function of serum amyloid A, a major acute-phase protein, in normal and disease states. Curr Opin Hematol 2000; 764–769. 136. Aksentijevich I, Torosyan Y, Samules J, et al. Mutation and haplotype studies of familial Mediterranean fever reveal new ancestral relationships and evidence for a high carrier frequency with reduced penetrance in the Ashkenazi Jewish population. Am J Hum Genet 1999; 64:939. 137. Mulley JC. The genetic basis for periodic fever. Am J Hum Genet 1999; 64:939. 138. Murakami T, Uchino M, Ando M. Genetic abnormalities and pathogenesis of familial amyloidotic polyneuropathy. Pathol Int 1995; 45:1. 139. Drueke TB. Beta2-microglobulin and amyloidosis. Nephrol Dial Transpl 2000; 15:17. 140. Gertz MA, Kyle RA. Hepatic amyloidosis: clinical appraisal in 77 patients. Hepatology 1997; 25:118. 141. Gillmore JD, Lovat LB, Hawkins PN. Amyloidosis and the liver. J Hepatol 1999; 30:117. 142. Faa G, Van Eyken P, De Vos R, et al. Light chain deposition disease of the liver associated with AL-type amyloidosis and severe cholestasis. J Hepatol 1991; 12:75. 143. Girelli CM, Lodi G, Rocca F. Kappa light chain deposition disease of the liver. Eur J Gastro Hepatol 1998; 10:429. 144. Hoffman MS, Stein BE, Davidian MM, et al. Hepatic amyloidosis presenting as severe intrahepatic cholestasis: a case report and review of the literature. Am J Gastroenterol 1988; 83:783.
1086
145. Iwai M, Ishii Y, Mori T, et al. Cholestatic jaundice in two patients with primary amyloidosis: ultrastructural findings of the liver. J Clin Gastroenterol 1999; 28:162. 146. Mohr A, Miehlke S, Klauck S, et al. Hepatomegaly and cholestasis as primary clinical manifestations of an AL-kappa amyloidosis. Eur J Gastro Hepatol 1999; 11:921. 147. Peters RA, Koukoulis G, Gimson A, et al. Primary amyloidosis and severe intrahepatic cholestatic jaundice. Gut 1994; 35:1322. 148. Qureshi WA. Intrahepatic cholestatic syndromes: pathogenesis, clinical features and management. Dig Dis 1999; 17:49. 149. Bujanda L, Beguiristain A, Alberdi F, et al. Spontaneous rupture of the liver in amyloidosis. Am J Gastroenterol 1997; 92:1385. 150. Bion E, Brenard R, Pariente EA, et al. Sinusoidal portal hypertension in hepatic amyloidosis. Gut 1991; 32:227. 151. Elghetany MT, Saleem A. Methods for staining amyloid in tissues: a review. Stain Technol 1988; 63:201. 152. Gertz MA, Kyle RA. Amyloidosis: prognosis and treatment. Semin Arthr Rheum 1994; 24:124. 153. Tan SY, Pepys MB, Hawkins PN. Treatment of amyloidosis. Am J Kidney Dis 1995; 26:267. 154. Comenzo RL, Vosburgh E, Falk RH, et al. Dose-intensive melphalan with blood stem-cell support for the treatment of AL (amyloid light-chain) amyloidosis: survival and responses in 25 patients. Blood 1998; 91:3662. 155. Dhodapkar MV, Jagannath S, Vesole D, et al. Treatment of ALamyloidosis with dexamethasone plus alpha interferon. Leuk Lymph 1997; 27:351. 156. Nowak G, Westermark P, Wernerson A, et al. Liver transplantation as rescue treatment in a patient with primary AL kappa amyloidosis. Transpl Int 2000; 13:92. 157. Comenzo RL. Hematopoietic cell transplantation for primary systemic amyloidosis: what have we learned? Leuk Lymph 2000; 37:245. 158. Patriarca F, Geromin A, Fanin R, et al. Improvement of amyloidrelated symptoms after autologous stem cell transplantation in a patient with hepatomegaly, macroglossia and purpura. Bone Marrow Transplant 1999; 24:433. 159. Ericzon BG, Holmgren G, Lundgren E, et al. New structural information and update on liver transplantation in transthyretinassociated amyloidosis. Report from the 4th International Symposium on Familial Amyloidotic Polyneuropathy and Other Transthyretin Related Disorders and the 3rd International Workshop on Liver Transplantation in Familial Amyloid Polyneuropathy, Umea Sweden, June 1999. Amyloid 2000; 7:145. 160. Fournier B, Giostra E, Mentha G, et al. Three cases of liver transplantation for type I familial amyloid polyneuropathy. Transplant Proc 1997; 29:2416. 161. Garcia-Herola A, Prieto M, Pascual S, et al. Progression of cardiomyopathy and neuropathy after liver transplantation in a patient with familial amyloidotic polyneuropathy caused by tyrosine-77 transthyretin variant. Liver Transpl Surg 1999; 5:246. 162. Hemming AW, Cattral MS, Chari RS, et al. Domino liver transplantation for familial amyloid polyneuropathy. Liver Transpl Surg 1998; 4:236. 163. Monteiro E, Perdigoto R, Furtado AL. Liver transplantation for familial amyloid polyneuropathy. Hepato-Gastroenterol 1998; 45:1375. 164. Shaz BH, Gordon F, Lewis WD, et al. Orthotopic liver transplantation for familial amyloidotic polyneuropathy: a pathological study. Human Pathol 2000; 31:40. 165. Suhr OB, Herlenius G, Friman S, et al. Liver transplantation for hereditary transthyretin amyloidosis. Liver Transplant 2000; 6:263. 166. Tashima K, Ando Y, Terazaki H, et al. Outcome of liver transplantation for transthyretin amyloidosis: follow-up of
Chapter 56 THE LIVER IN SYSTEMIC ILLNESS
167. 168. 169. 170. 171. 172. 173.
174.
175.
176. 177.
178. 179.
180.
181. 182.
183. 184.
185.
186.
187. 188. 189. 190.
191. 192. 193. 194.
Japanese familial amyloidotic polyneuropathy patients. J Neuro Sci 1999; 17:119. Johnson CS, Omata M. Tong MJ, et al. Liver involvement in sickle cell disease. Medicine 1985; 64:349. Schubert T. Hepatobiliary system in sickle cell disease. Gastroenterology 1986; 90:2013. Diggs LW. Sickle cell crisis. Am J Clin Pathol 1965; 44:1. Sheehy TW. Sickle cell hepatopathy. South Med J 1977; 70:533. Rosenblate HJ. Eisenstein R, Holmes AW. The liver in sickle-cell anemia. Arch Pathol 1970; 90:235. Sheehy TW. Sickle cell hepatopathy. South Med J 1977; 70:533. Buchanan GR, Blader BE. Benign course of extreme hyperbilirubinemia in sickle cell disease: analysis of six cases. J Pediatr 1977; 91:21. Zakaria N, Knisely A, Portmann B, et al. Acute sickle cell hepatopathy represents a potential contraindication for percutaneous liver biopsy. Blood 2003; 101:101. Ballas SK, Lewis CN, Noone AM, et al. Clinical, hematological, and biochemical features of HbSC disease. Am J Hematol 1982; 13:37. Phillips JC, Gerald BE. The incidence of cholethiasis in sickle cell disease. AJR 1971; 113:27. Sarnaik S, Slovis TL, Corbett DP, et al. Incidence of cholelithiasis in sickle-cell anemia using the ultrasonic gray-scale technique. J Pediatr 1980; 96:1005. Flye MW, Silver D. Biliary tract disorders and sickle cell disease. Surgery 1972; 72:361. Comer GM, Ozick LA, Sachdev RK, et al. Transfusion related chronic liver disease in sickle-cell anemia. Am J Gastroenterol 1991; 86:1232. Omata M, Johnson CS, Tong M, Tatter D. Pathological spectrum of liver disease in sickle cell disease. Dig Dis Sci 1986; 31:247. De Vault KR, Friedman LS, Westerberg S, et al. Hepatitis C in sickle-cell anemia. Am J Gastroenterol 1994; 18:206. Hasan MF, March F, Posner G, et al. Chronic hepatitis C in patients with sickle cell disease. Am J Gastroenterol 1996; 91:1204. Hassan M, Hasan S, Giday S, et al. Hepatitis C virus in sickle cell disease. J Nat Med Assoc 2003; 95:939. Yeomans E, Lowe TW, Eigenbrodt EH, Cunningham FG. Liver histopathologic findings in women given prophylactic transfusion during pregnancy. Am J Obstet Gynecol 1990; 163:958. Mills LR, Mwakyusa D, Milner PF. Histopathologic features of liver biopsy specimens in sickle cell disease. Arch Pathol Lab Med 1988; 112:290. Bauer TW, Moore GW, Hutchins GM. The liver in sickle cell disease: a clinicopathologic study of 70 patients. Am J Med 1980; 69:833. Birrer MJ, Young RC. Differential diagnosis of jaundice in lymphoma patients. Semin Liver Dis 1987; 7:269. Jaffe ES. Malignant lymphomas: pathology of hepatic involvement. Semin Liver Dis 1987; 7:257. Kadin ME, Donaldson SS, Dorfman RF. Isolated granulomas in Hodgkin’s disease. N Engl J Med 1970; 283:859. Aisenberg AC, Kaplan MM, Rieder SV, Goldman JM. Serum alkaline phosphatase at the onset of Hodgkin’s disease. Cancer 1970; 26:318. Bouroncle BA, Old JW, Vazques AG. Pathogenesis of jaundice in Hodgkin’s disease. Arch Intern Med 1962; 110:872. Levitan R, Diamond HD, Craver LF. Jaundice in Hodgkin’s disease. Am J Med 1961; 30:99. Perera DR, Greene ML, Fenster F. Cholestasis associated with extrabiliary Hodgkin’s disease. Gastroenterology 1974; 67:680. Hubscher SG, Lumley MA, Elias E. Vanishing bile duct syndrome: a possible mechanism for intrahepatic cholestasis in Hodgkin’s lymphoma. Hepatology 1993; 17:70.
195. Crosbie OM, Crown JP, Nolan NP, et al. Resolution of paraneoplastic bile duct paucity following successful treatment of Hodgkin’s disease. Hepatology 1997; 26:5. 196. Man KM, Drejet A, Keeffe EB, et al. Primary sclerosing cholangitis and Hodgkin’s disease. Hepatology 1993; 18:1127. 197. Woolf GM, Petrovic LM, Rojter SE, et al. Acute liver failure due to lymphoma: a diagnostic concern when considering liver transplantation. Dig Dis Sci 1994; 39:1351. 198. Rowbotham D, Wendon J, Williams R. Acute liver failure secondary to hepatic infiltration: a single center experience of 18 cases. Gut 1998; 42:576. 199. Dourakis SP, Tzemanakis E, Deutsch M, et al. Fulminant hepatic failure as a presenting paraneoplastic manifestation of Hodgkin’s disease. Eur J Gastro Hepatol 1999; 11:1055. 200. Memeo L, Pecorello I, Ciardi A, et al. Primary non-Hodgkin’s lymphoma of the liver. Acta Oncol 1999; 38:655. 201. Lei KI. Primary non-Hodgkin’s lymphoma of the liver. Leuk Lymph 1998; 29:293. 202. Verdi CJ. Grogan TM, Protell R, et al. Liver biopsy immunotyping to characterize lymphoid malignancies. Hepatology 1986; 6:6. 203. Ryoo JW, Manaligod JR, Walker MJ. Primary lymphoma of the liver. J Clin Gastroenterol 1986; 8:308. 204. Osborne BM, Butler JJ, Guarda LA. Primary lymphoma of the liver. Cancer 1985; 56:2902. 205. Pinto PC, Hu E, Bernstein-Singer M, et al. Acute hepatic injury after withdrawal of immunosuppressive chemotherapy in patients with hepatitis B. Cancer 1990; 65:878. 206. Markovic S, Drozina G, Vovk M, et al. Reactivation of hepatitis B but not hepatitis C in patients with malignant lymphoma and immunosuppressive therapy. A prospective study of 305 patients. Hepato-Gastroenterol 1999; 46:2925. 207. Al-Taie OH, Mork H, Gassel AM, et al. Prevention of hepatitis B flare-up during chemotherapy using lamivudine: case report and review of the literature. Ann Hematol 1999; 78:247. 208. Ferri C, Caracciolo F, Zignego AL, et al. Hepatitis C virus infection in patients with non-Hodgkin’s lymphoma. Br J Hematol 1994; 88:392. 209. Ferri C, La Civita L, Monti M, et al. Chronic hepatitis C and Bcell non-Hodgkin’s lymphoma. Q J Med 1996; 89:117. 210. Collier JD, Zanke B, Moore M, et al. No association between hepatitis C and B-cell lymphoma. Hepatology 1999; 29:1259. 211. Pioltelli P, Gargantini L, Cssi E, et al. Hepatitis C virus in nonHodgkin’s lymphoma. A reappraisal after a prospective casecontrol study of 300 patients. Am J Hematol 2000; 64:95. 212. Ghobrial IM, Wolf RC, Pereira DL, et al. Therapeutic options in patients with lymphoma and severe liver dysfunction. Mayo Clinic Proc 2004; 79:169. 213. Karp SJ, Shareef D. Ascites as a presenting feature multiple myeloma. J R Soc Med 1897; 80:182. 214. Kitazono M, Saito Y, Kinoshita M, et al. Nodular regenerative hyperplasia of the liver in a patient with multiple myeloma and systemic amyloidosis. Acta Pathol Jpn 1985; 35:961. 215. Dimopoulos MA, Galani E, Matsouka C. Waldenström’s macroglobulinemia. Hematol Oncol Clin North Am 1999; 13:1351. 216. Pauls JD, Brems J, Pockros PJ, et al. Mastocytosis: diverse presentations and outcomes. Arch Intern Med 1999; 15:9401. 217. Horny H-P, Ruck P, Krober S, et al. Systemic mast cell disease (mastocytosis). General aspects and histopathological diagnosis. Histo Histopathol 1997; 12:1081. 218. Ghandur-Mnaymneh L, Gould E. Systemic mastocytosis with portal hypertension; autopsy findings and ultrastructural study of the liver. Arch Pathol Lab Med 1985; 10:976. 219. Kyriakou D, Kouroumalis E, Konsolas J, et al. Systemic mastocytosis: a rare cause of noncirrhotic portal hypertension simulating autoimmune cholangitis – report of four cases. Am J Gastroenterol 1998; 93:106.
1087
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
220. Humar A, McGilvray I, Phillips MJ. Severe acute respiratory syndrome and the liver. Hepatology 2004; 39:291. 221. Wu KL, Lu SN, Changchien CS, et al. Sequential changes of serum aminotransferase levels in patients with severe acute respiratory syndrome. Am J Trop Med Hyg 2004; 71:125. 222. Cui HJ, Tong XL, Li P, et al. Serum hepatic enzyme manifestations in patients with severe acute respiratory
1088
syndrome: retrospective analysis. W J Gastroenterol 2004; 10:1652. 223. Wong WM, Ho JC, Ooi GC, et al. Temporal patterns of hepatic dysfunction and disease severity in patients with SARS. JAMA 2003; 290:2663. 224. Chau TN, Lee KC, Yao H, et al. SARS-associated viral hepatitis caused by a novel coronavirus: report of three cases. Hepatology 2004; 39:302.
Section IX: Liver Affected by Other Organs (Non-Hepatic Conditions)
THE LIVER AND PARENTERAL NUTRITION
57
Kathleen M. Campbell and William F. Balistreri Abbreviations BSEP bile-salt export pump DEHT diethylhexylphthalate GGT g-glutamyltransferase NTCP sodium-taurocholate cotransporting polypeptide
PN PNAC PXR TNF-a
parenteral nutrition PN-associated cholestasis pregnane X receptor tumor necrosis factor-a
INTRODUCTION The use of parenteral nutrition (PN) as a substitute for or an adjuvant to enteral feeding has become standard practice in the care of preterm infants, individuals with intestinal failure or short-bowel syndrome, and children and adults during periods of stress and/or increased metabolic demands. However, the use of PN, especially in the absence of any enteral feeding (total PN or TPN), is associated with a number of complications involving the liver and biliary tract. The pattern and nature of these complications vary based on the age of the patient. They range from asymptomatic abnormalities in serum biochemistries to biliary sludge and cholelithiasis, steatohepatitis, or severe cholestasis, with possible progression to cirrhosis and end-stage liver disease. In all age groups the risk of hepatobiliary complications increases with prolonged dependence on PN, the presence of short-bowel syndrome, and the absence of significant enteral nutrition.
EPIDEMIOLOGY The nature of hepatobiliary complications associated with PN varies according to age (Figure 57-1). In neonates and young children, PNassociated cholestasis (PNAC) is the most common complication, and is marked biochemically by an elevation in serum alkaline phosphatase, g-glutamyltransferase (GGT), and conjugated bilirubin levels. The incidence of PNAC in this population is 40–60%.1 The risk of PNAC is inversely related to gestational age and volume of enteral feeding, and correlates directly with duration of PN. In addition, the risk of PNAC is greater in infants with intrauterine growth retardation as compared to those appropriate for gestational age, and increases with early systemic infection.2,3 Thus, it is no surprise that low-birth-weight, premature infants with congenital or acquired intestinal dysfunction are at highest risk. Beale et al. found that PNAC developed in 23% of 62 premature infants weighing <2000 grams at birth. The incidence was 50% in infants <1000 grams at birth, but only 7% in infants >1500 grams at birth. The incidence rose with increasing length of therapy.4 The overall incidence of PNAC in neonates has decreased over time, as described in a study
total PN or TPN SAME S-adenosylmethionine UDCA ursodeoxycholic acid
of 273 infants who received PN for at least 2 weeks. Between 1971 and 1983 the incidence of PNAC was 57%, compared with 31% between 1983 and 1987, and 25% in infants treated between 1992 and 1996.5 Factors associated with severe PNAC in that population included infection and intestinal stasis. The authors speculated that the decline in the incidence of PNAC over time was the result of better management of total caloric intake, improved TPN composition, and decreased catheter-related infections.5 Andorsky et al. examined variables that correlated with cholestasis in a group of infants with short-bowel syndrome exposed to PN for at least 90 days. They found that shorter bowel length, longer time with a diverting ostomy, greater number of Gram-positive infections, and fewer days of feeding were associated with a higher peak conjugated bilirubin level.6 In older children and adults, steatosis, steatohepatitis, and non-specific hepatic inflammation are more frequent responses to prolonged PN exposure. Non-specific elevation of serum aspartate aminotransferase, alanine aminotransferase, and alkaline phosphatase levels occur in 20–70% of adult patients receiving PN for more than 2 weeks.7–9 These abnormalities are transient in most patients; however, up to 15% of patients develop persistent biochemical abnormalities and are at risk for progression to chronic liver disease.10 The predominant histologic finding in adult patients with abnormal liver biochemistries is fatty infiltration, with or without inflammation.9 Cholestasis is reportedly less frequent in this age group. However, Cavicchi et al. found that chronic cholestasis developed in 65% of adult patients receiving long-term PN; with the prevalence of chronic cholestasis increasing from 55% after 2 years to 72% after 6 years of PN treatment. Chronic cholestasis was associated with an underlying independent risk for liver disease, a smallbowel length of <50 cm, and parenteral lipid intake of more than 1 g/kg per day.11 Biliary tract complications, ranging from gallbladder distention to biliary sludge to gallstone formation and acalculous or calculous cholelithiasis, are equally common in both children and adults receiving long-term PN. Serial ultrasound examinations of adult patients with no pre-existing biliary tract disease documented the time course of PN-associated biliary tract complications.12 In this study, 6% of patients had biliary sludge at 3 weeks, 50% had biliary sludge between 4 and 6 weeks, and 100% of patients had biliary
1089
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Table 57-1. Proposed Etiology of Parenteral Nutrition-Associated Hepatobiliary Complications
NEONATES
Cholestasis
ADULTS
Cholelithiasis/ Cholecystitis
Steatosis/ Steatohepatitis
Figure 57-1. Range of manifestations of parenteral nutrition (PN)-associated hepatobiliary complications. The predominant clinical and histologic presentation of PN-associated hepatobiliary injury varies with age. While cholestasis is the most common presentation in neonates, steatosis/steatohepatitis is a more common presentation in adolescents and adults. Biliary tract complications are equally common in all ages.
sludge when receiving PN for more than 6 weeks. Gallstones formed in 42% of patients with sludge, and 50% of patients with stones required cholecystectomy with removal of mixed bilirubin– cholesterol gallstones.12 Not surprisingly, the risk of biliary tract complications increases with increased duration of PN, with lack of a functional ileum, and with little to no enteral intake.13–15
PATHOGENESIS The pathophysiology of PN-associated hepatobiliary complications is presumably multifactorial (Table 57-1), depending on a number of factors, including the age of the patient, the presence of underlying gastrointestinal dysfunction, associated inflammatory and infectious conditions, and the intrinsic properties of the PN infusate.
1090
Cholestasis • Immaturity of hepatobiliary function and transport • Altered biliary constituents (increased lithocholate) • Sepsis • Bacterial overgrowth and translocation • Chronic inflammatory state • Interruption of the enterohepatic circulation • Toxicity of contaminants, including phytosterols • Oxidant/antioxidant imbalance Steatosis/steatohepatitis Provision of excess calories (dextrose or lipid) Essential fatty acid deficiency Bacterial overgrowth and translocation Choline deficiency Taurine deficiency Carnitine deficiency Sepsis Oxidant/antioxidant imbalance Increased portal vein insulin/glucagon ratio
• • • • • • • • •
Biliary tract complications Prolonged lack of enteral stimulation Reduced bile flow Interruption of enterohepatic circulation
• • •
nuclear receptors, including the farnesoid X receptor (FXR), which activates the transcription of the ileal sodium-dependent bile acid transporter and BSEP.20 Low levels of RNA expression, protein expression, and functional activity of FXR and other nuclear receptors in the prenatal and neonatal liver reflect the ontogeny of the hepatocanalicular transport system. The coordinate expression of these two classes of proteins leads to a low-flow state with regard to bile acid metabolism in the newborn. As bile acids are the driving force for bile formation, the result is physiologic cholestasis, which is particularly striking in the premature neonate and predisposes these infants to more profound PN-associated cholestasis.
IMMATURITY OF HEPATOBILIARY FUNCTION AND TRANSPORT
ROLE OF TOXIC BILE ACIDS
One of the primary factors leading to PNAC in neonates is agerelated immaturity of hepatobiliary function and transport. It has long been recognized that normal, term infants display a developmental delay in bile acid metabolism leading to decreased hepatic uptake of bile acids, decreased hepatic excretion, and increased peripheral serum concentrations (“physiologic cholestasis”).16 This condition is due in part to an inefficient enterohepatic circulation in neonates caused by poor ileal reabsorption of bile acids. In a similar manner, infants with necrotizing enterocolitis, and adults with inflammatory bowel disease or bowel resection, frequently lack a functional ileum, further disrupting the enterohepatic circulation. Neonates also have a reduced capacity for hepatic bile acid uptake and excretion. Studies in rats have shown that the expression of the ileal sodium-dependent bile acid transporter, the basolateral sodium-taurocholate cotransporting polypeptide (NTCP), and the canalicular bile-salt export pump (BSEP) are developmentally regulated, with low levels of expression prenatally and in the immediate postnatal period.17–19 The expression of these proteins is, in part, regulated at the transcriptional level by the activity of a number of
The generation of toxic bile acids, particularly the monohydroxy bile acid lithocholate, has been implicated as a marker and a mediator of PNAC in children and adults. A secondary bile acid, lithocholate is formed by bacterial dehydroxylation of primary bile acids. In both humans and animal models, increased concentrations of serum and biliary lithocholate have been associated with the use of PN, even in the absence of other biochemical markers of hepatobiliary dysfunction.21–23 This increase in lithocholate levels is in part due to the inefficient enterohepatic circulation associated with lack of enteral stimulation. The delivery of primary bile acids to the large intestine, where resident bacteria effect the conversion to lithocholate, is thus facilitated, and this secondary bile acid then efficiently re-enters the enterohepatic circulation. In healthy hepatocytes, lithocholate is detoxified by sulfation and/or hydroxylation. Expression of the enzymes responsible for these detoxifying reactions is controlled by another nuclear receptor, the pregnane X receptor (PXR), the absence of which leads to profound lithocholate toxicity in experimental animals.24–26 The expression of PXR is developmentally regulated, as with the nuclear receptors mentioned above, and may be
Chapter 57 THE LIVER AND PARENTERAL NUTRITION
especially low in the preterm neonate.20 In addition, expression of the cytochrome P450 3A system, responsible for the hydroxylation of lithocholate and other xenobiotics, is also regulated by PXR expression, and is low in the neonatal liver.27,28 Therefore, although toxic bile acids are likely to play a role in PNAC at all ages, the neonatal liver may be least equipped to attenuate their damage.
GASTROINTESTINAL DYSFUNCTION The primary indication for the use of long-term PN in all age groups is temporary or permanent gastrointestinal dysfunction due to a variety of causes. In infants, diseases such as necrotizing enterocolitis, volvulus, and gastroschisis account for many cases of intestinal dysfunction or failure necessitating PN. Motility disorders and inflammatory bowel disease are more common indications in older children and adults. Limited tolerance to enteral feedings is common in all of these disorders. The lack of nutrient-induced enteral stimulation worsens underlying gastrointestinal dysfunction and promotes the development of PN-associated hepatobiliary complications through a variety of mechanisms. In humans and animals, the fasted state is associated with decreased levels of gut hormones and peptides that impact secretion, motility, and mucosal growth.29–31 Serum levels of cholecystokinin, gastrin, enteroglucagon, motilin, glucagon-like peptide-2, and peptide YY are lower in fasted patients. In addition, bowel resection or mucosal disease may alter the secretion of these hormones, particularly peptide YY, which is produced in the distal small bowel and colon.32,33 The lack of these hormones leads to intestinal atrophy and impaired motility. This in turn promotes intestinal stasis and bacterial translocation, which increase the risk of sepsis. Gallbladder stasis also occurs, which increases the risk of biliary sludge and gallstone formation.34 Lack of enteric nutrients also leads to decreased mass of gut-associated lymphoid tissue and decreased mucosal immunoglobulin A secretion, impairing intestinal immunity. In the absence of sufficient mucosal barriers, bacterial pathogens display increased adherence to the intestinal mucosal surface, leading to further epithelial injury. The importance of gastrointestinal function in the development of PN-associated hepatobiliary complications is reflected in the protective effect of enteral feeding. Even minimal enteral feeds stimulate the secretion of many gut hormones and peptides, and the volume of enteral feeds correlates inversely with the risk of all PN-associated hepatobiliary complications.35
BACTERIAL OVERGROWTH In the absence of enteral stimulation, intestinal motility and immunoglobulin A secretion decrease, and gut permeability increases, thereby promoting intestinal stasis, bacterial overgrowth, and translocation. Translocation in turn leads to portal endotoxemia and sepsis. Overgrowth of intestinal bacteria also results in increased conversion of primary bile acids to toxic lithocholate in the large intestine, which further promotes hepatic injury.
INFECTION AND ACTIVATION OF INFLAMMATORY PATHWAYS Sepsis has been associated with the development of both cholestasis and steatosis in the setting of PN.3,36,37 Evidence supports that these effects are mediated through endotoxin and other proinflammatory molecules. Elevation of serum alkaline phosphatase and
GGT levels correlates with elevated erythrocyte sedimentation rate, serum neopterin, tumor necrosis factor-a (TNF-a) and soluble interleukin-2 receptor levels in adults with inflammatory bowel disease receiving home PN.36 Furthermore, antibody against TNF-a has proven effective at reducing hepatic steatosis in a rat model of PN-associated liver disease, and was associated with marked biochemical and histological improvement in PNAC when used to treat an enterocutaneous fistula in a patient with Crohn’s disease.38,39 Endotoxin also plays a key role in the initiation and perpetuation of PN-associated liver dysfunction. Infusion of endotoxin in rats receiving PN results in marked accumulation of lipid in the liver, even in the absence of significant changes in serum biochemistries. Endotoxin has long been implicated as an etiologic factor in multiple syndromes of intrahepatic cholestasis.40,41 Endotoxin acts within the hepatic microenvironment by stimulating Toll-like receptors on Kupffer cells, causing a release of proinflammatory cytokines. In addition, there is growing evidence that endotoxin directly stimulates hepatic stellate cells to release chemokines and cellular adhesion molecules, promoting ongoing inflammation and fibrosis.42 This proinflammatory storm results in decreased levels of NTCP, BSEP, and Mrp2, the multispecific organic anion transporter located on the hepatocyte canalicular membrane, a major contributor to bilesalt-independent bile flow.41 This combination of events results in cholestasis, as well as promoting intrahepatic inflammation and fibrosis.
OXIDANT/ANTIOXIDANT IMBALANCE Many of the mechanisms contributing to PN-associated hepatic complications act through the final common pathway of increased oxidant damage, either by increasing hepatic reactive oxygen species or decreasing natural antioxidants. One source of excess damaging oxidants may be the photo-oxidation of multivitamin solutions contained in PN formulations, leading to the production of peroxides.43 Hepatic glutathione, the most important intracellular antioxidant in human tissue, is depleted during PN administration in animal models, in association with the development of biochemical evidence of liver injury and oxidant damage.44 In neonates this may be due in part to immaturity of the hepatic transsulfuration pathway (Figure 57-2), usually a source of the protective antioxidants glutathione and S-adenosylmethionine (SAME), as well as developmental deficiencies in antioxidant compounds, including vitamin E, selenium, and taurine. Additional oxidative stress from any source (such as hypoxia or hypoperfusion) exacerbates the situation by leading to lipid peroxidation and cell membrane damage.
DEFICIENCIES ASSOCIATED WITH PN Prior to the advent of intravenous lipid preparations, essential fatty acid deficiency was a common cause of steatosis in patients receiving PN with minimal enteral feeding. In part, this was due to deficiency of phosphatidylcholine, resulting in defective incorporation of triglycerides into lipoproteins for extrusion from the hepatocyte. This deficiency may be exacerbated in circumstances that promote hepatic triglyceride synthesis. Even with current intravenous lipid preparations, plasma free choline deficiency has been identified in both children and adults receiving long-term PN with minimal enteral intake.45–47 In this setting, low plasma free choline levels are associated with hepatic steatosis with or without biochemical abnor-
1091
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Dietary protein/amino acids
Methionine ATP S-Adenosylmethionine
S-Adenosylhomocysteine ATP Homocysteine Serine Cystathione
Cysteine Glutamate + ATP ␥-Glutamylcysteine Glycine + ATP Glutathione Figure 57-2. Glutathione synthesis and the hepatic trans-sulfuration pathway. De novo synthesis of the intracellular antioxidant glutathione requires the precursors glutamate, cysteine, and glycine. Methionine is converted to cysteine via the hepatic trans-sulfuration pathway. Deficiency of necessary precursors, or reduced activity of the trans-sulfuration pathway, may lead to reduced intrahepatic glutathione stores and increased risk of oxidant injury.
malities.45 Although intravenous lipid provides some choline in the form of phosphatidylcholine, free choline is minimal. Defects or immaturity in the hepatic transsulfuration pathway may also contribute to low plasma choline levels by decreasing de novo choline synthesis. As it is a methyl donor in the metabolism of homocysteine, a key precursor to glutathione, choline deficiency may predispose to oxidant-mediated injury. Deficiencies of vitamin C, vitamin E, and/or selenium may contribute to steatosis in the setting of PN via loss of protective antioxidant activity in the liver. Deficiency of carnitine, essential for appropriate fat metabolism, is also associated with an increased incidence of steatosis.45,46 The degree to which altered concentrations of these compounds are a consequence, rather than a cause, of liver injury, is difficult to elucidate. For example, cholestasis itself alters intrahepatic carnitine handling, leading to secondary changes in fat metabolism and resulting in increased intracellular lipid storage.
INHERENT TOXICITIES OF PN CONSTITUENTS OR FORMULATION Caloric deficiency, caloric overload, and imbalance of calories from various constituents can lead to hepatic dysfunction. Excess calories
1092
via PN leads to hepatic steatosis and steatohepatitis in adults. The relative role of excess dextrose calories and excess lipid calories in hepatic dysfunction is controversial. Lipid emulsions have been implicated as a contributing factor based on the appearance of cholestasis in association with changes in lipid delivery, and the improvement in cholestasis noted with decreased lipid administration.48 Intravenous lipids have been characterized as proinflammatory; they interfere with cholesterol metabolism, promote lipid peroxidation, decrease clearance of bacterial endotoxin, and adversely impact macrophage function.48 In addition, administration of intravenous lipids increases hepatocyte apoptosis.49 These consequences appear to be dose-dependent, with a proportional effect starting at 1 g/kg per day.50 In addition to the quantity of lipid, the type of lipid in PN formulations may contribute to hepatic dysfunction. At least one study has suggested that a lipid formulation containing 50% medium-chain triglycerides and 50% long-chain triglycerides is superior to traditional soy-based lipid formulations at preventing cholestasis.51 On the other hand, excess or unbalanced dextrose calories are associated with both cholestasis and steatosis in animal models. Hyperglycemia itself induces decreased bile flow in isolated perfused rat livers and in diabetic rats via a decrease in bile acid independent bile flow.52,53 In the absence of lipid calories, excess dextrose is easily converted to triglycerides in the liver, leading to steatosis.54,55 Both of these processes may be promoted by the delivery of dextrose to the liver via peripheral veins rather than the portal vein. This altered delivery system is associated with changes in the relative concentrations of insulin and glucagon in the portal vein, with an increase in the insulin/glucagon ratio. As insulin promotes fatty acid biosynthesis and glucagon inhibits fatty acid synthesis and promotes hepatic release of fatty acids, this altered balance in the portal vein leads to steatosis.56 Addition of intravenous lipid decreases the insulin/glucagon ratio and prevents steatosis.56 The quantity and quality of amino acids present in PN have both been implicated in PNAC. In both animal models and isolated perfused rat livers, intravenous infusion of amino acids solutions produces cholestasis.57 This cholestatic reaction may be more profound in neonates, who have difficulty handling certain amino acids due to immature hepatic mitochondrial function. Particular amino acids have been implicated as either protective or harmful. For example, taurine has been shown to have a beneficial effect by stimulating increased bile flow and bile acid secretion, glutamine reduces bacterial translocation and enhances hepatic antioxidant capacity via the cytochrome P450 system and enhanced glutathione stores, while methionine infusion reproduces the liver injury of PNAC.58–61 Other micronutrients that have been implicated in PN-associated hepatobiliary disease include manganese and copper, excess levels of which have been associated with cholestasis in animal models and in humans.62 Again, whether these micronutrients represent a cause or an effect of liver dysfunction remains unclear. Non-nutrient components incorporated into or contaminating PN formulations are likely to play a role in PN-associated hepatic injury. A classic example is found in the case of polysorbates, previously used as emulsifying agents. These compounds were associated with a severe and fatal syndrome of cholestasis, thrombocytopenia, ascites, and renal failure in a group of low-birth-weight infants who received polysorbates as part of an intravenous vitamin E solution.63
Chapter 57 THE LIVER AND PARENTERAL NUTRITION
Tween 80, a surfactant, has been associated with cholestasis in animal models, and diethylhexylphthalate (DEHT), a peroxisome proliferator used as a softener in polyvinyl infusion circuits, inhibits phase 1 and phase 2 enzymes in the liver; these enzymes are necessary for detoxifying noxious substances.64 More recently, phytosterols, which are present in some commercial lipid formulations, have been implicated in PN-associated hepatic injury. Phytosterols cannot be efficiently metabolized by the liver and lead to cholestasis via a number of potential mechanisms. By displacing cholesterol in plasma membranes they may alter membrane fluidity and interfere directly with the function of hepatocanalicular transporters essential for bile formation.65,66 New evidence supports the additional role of phytosterols as direct antagonists to FXR, the bile acidsensitive nuclear receptor responsible for increasing expression of BSEP as a hepatoprotective mechanism in cholestatic conditions.67
CLINICAL FEATURES DIAGNOSIS The diagnosis of PN-associated liver disease is based on the development of characteristic clinical symptoms in the setting of exposure to PN, and the exclusion of other causes of liver disease. PN-associated cholestasis is operationally defined as a serum conjugated bilirubin >2.0 mg/dl and >50% of the total bilirubin in a patient receiving PN for at least 14 consecutive days.68 Most cases of PNAC present within 2–10 weeks of PN initiation.69 Specific criteria for the development of other forms of PN-associated liver disease are more difficult to define. Steatosis, in particular, may or may not be accompanied by biochemical evidence of liver disease. In any patient with biochemical or histologic evidence of hepatobiliary complications of PN, it is essential to rule out other causes of liver disease. This is particularly applicable to neonates, in whom other genetic, structural, and metabolic causes of cholestasis may present in a similar time frame.
HISTOLOGIC FEATURES The histologic features of PN-associated hepatobiliary disorders are protean and are seldom diagnostic. In Cavicchi’s study of 90 patients, primarily adults, receiving long-term home PN, 57 patients underwent liver biopsy. Histologic cholestasis was identified in 76% of patients with biochemical cholestasis and in 27% of patients without biochemical cholestasis. Other histologic findings included portal inflammation (87–90%), macrosteatosis (63–100%), ductular proliferation (59–72%), and hepatocyte necrosis (18–43%). Microsteatosis was identified in 63% of patients and severe steatosis was associated with more marked fibrosis or cirrhosis.11 In neonates many of the histologic findings are similar, although cholestasis, rather than steatosis, predominates. Zambrano’s review of 24 cases of PN-associated hepatobiliary disease identified periportal inflammation as the most prevalent finding (83%), followed by cholestasis (79%), bile duct proliferation (79%), and steatosis (29%). Fibrosis was identified in 71% of specimens, with a progression in the severity of fibrosis with increased duration of PN. Giantcell transformation (29%) and extramedullary hematopoiesis (79%) were additional findings that are specific to the neonatal liver.70 These histologic findings may look strikingly similar to other causes of neonatal cholestasis, including biliary atresia.
DIFFERENTIAL DIAGNOSIS Many individuals exposed to PN have multiple other risk factors for liver disease. Ensuring that there are no underlying or concurrent explanations for hepatobiliary dysfunction is the first step. The differential diagnosis includes medication use, hypoxic insults, viral hepatitis, heart disease and other causes of primary cholestasis, particularly important in neonates (cystic fibrosis, biliary atresia, syndromes of progressive familial intrahepatic cholestasis, a1antitrypsin deficiency, metabolic liver disease).
TREATMENT There is no specific treatment for PN-associated hepatobiliary disease. The definitive treatment is discontinuation of PN. In situations in which this is not possible, a variety of strategies and compounds have been employed to prevent or treat the consequences of PN (Table 57-2). In general, these therapies target one or more of the pathophysiologic mechanisms mentioned previously.
CHOLERETICS A variety of strategies have been employed in an effort to increase bile flow and protect the hepatocyte from bile acid-induced injury. Ursodeoxycholic acid (UDCA), a hydrophilic bile salt advocated in many forms of cholestatic liver disease, may be effective in the prevention and/or treatment of PN-associated hepatobiliary disease. In addition to acting as a choleretic agent, UDCA has been shown to abrogate hydrophobic bile acid-induced apoptosis in both in vitro and in vivo models.71 Furthermore, UDCA may be effective at
Table 57-2. Proposed Therapies for Parenteral Nutrition-Associated Hepatobiliary Complications Therapy
Mechanism of action
Ursodeoxycholic acid, tauro-ursodeoxycholic acid Cholecystokinin Rapid amino acid infusion b-carotene S-adenosylmethionine Homocysteine Antibiotics
Promotes bile flow, protects hepatocyte from toxic bile acids
Anti-inflammatory agents (anti-TNF Ab) Glutamine Glucagon Glucagon-like peptide 2 Cyclic parenteral nutrition infusion Enteral feeding
Promotes bile flow Increases gallbladder-emptying and decreases biliary sludge formation Antioxidant Antioxidant Antioxidant Prevent sepsis, decrease bacterial translocation Reduce circulating TNF Promotes intestinal mucosal health, decreases bacterial translocation Reduces insulin/glucagon ratio in portal vein Promotes intestinal mucosal health and integrity Promotes more physiologic pattern of insulin release, may protect hepatocyte from oxidant injury Promotes intestinal mucosal health, promotes bile flow, promotes gastrointestinal hormone release, may help replete hepatic glutathione stores
TNF, tumor necrosis factor.
1093
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
inhibiting apoptosis triggered by other insults via prevention of the mitochondrial membrane permeability transition and subsequent inhibition of reactive oxygen species production.72 Supplementation with enteral UDCA decreases biochemical markers of PNassociated hepatic injury and the incidence of hepatic steatosis in a rabbit model.73 Studies in humans have reported variable results of the use of UDCA and tauro-UDCA in PNAC. While most studies in children and adults report a beneficial effect on serum biochemistries and/or a shorter duration of cholestasis, at least one study in neonates found no benefit to the use of tauro-UDCA.74–78 These conflicting results might be explained by differences in subject age, duration of PN exposure, variable absorption of UDCA, and presence of additional risk factors (such as intestinal resection and infection). The use of cholecystokinin has been advocated as prevention or treatment for the development of PN-associated gallbladder disease and cholestasis. While daily intravenous dosing is associated with increased bile flow and decreased serum conjugated bilirubin levels, the long-term effects on gallstone formation and progressive cholestasis are not clear.79–81 Administration of additional intravenous amino acids daily over a 5-minute period to adult patients receiving PN may increase gallbladder-emptying and prevent the formation of biliary sludge.82 This rapid administration of intravenous protein simulates the postprandial state and may effect gallbladder emptying through enhanced cholecystokinin release.83 Additional compounds proposed as choleretics include chenodeoxycholate, which prevents calcium bilirubinate gallstones in prairie dogs on PN, and cisapride, which may act by increasing intestinal motility in addition to increasing gallbladder contractility and thereby promoting bile flow.84 Neither agent is currently available.
ANTIOXIDANTS As oxidant-mediated injury is likely a final common pathway in PNassociated hepatobiliary disease, a wide variety of antioxidant compounds have shown promise in ameliorating or preventing hepatic injury. In an in vitro model of cholestatic liver disease, b-carotene blocked hydrophobic bile acid-induced generation of reactive oxygen species, inhibited both apoptosis and necrosis of isolated rat hepatocytes, and may hold promise in PNAC.85 Addition of SAME to PN preparations has been shown to increase bile flow, decrease hepatic steatosis, and result in improvement in biochemical markers of liver damage in a rat model, and may have similar beneficial effects in humans.86,87 SAME acts as a precursor for glutathione and is a methyl donor in most transmethylation reactions, preventing glutathione depletion and mitochondrial dysfunction in several models of hepatic injury. On the other hand, addition of homocysteine, another glutathione precursor, to a standard PN formulation administered to rats increased bile flow without improving histology or changing markers of lipid and protein oxidation.88
ANTIMICROBIALS/ANTI-INFLAMMATORY AGENTS Addition of metronidazole to enteral UDCA supplementation in a rabbit model of PNAC resulted in an additional decrease in bilirubin levels in treated animals.73 Antibodies to TNF decrease total hepatic fat and triglyceride levels, and decrease TNF production by
1094
peritoneal macrophages in animal models of PN-associated steatosis.38 In addition, both probiotics and anti-TNF antibodies improve liver histology and decrease hepatic steatosis in the ob/ob mouse model of non-alcoholic fatty liver disease.89 Although the initiating factors for hepatic injury in obesity-related fatty liver disease may be different from those in PN-associated steatosis, the role of bacterial overgrowth and proinflammatory cytokines may represent a common pathway to liver injury. Polymyxin B, which exhibits specific anti-lipopolysaccharide activity in addition to antimicrobial activity, has also been shown to decrease total hepatic fat and triglyceride concentrations in the livers of rats receiving PN, but has yet to be tested in humans.90 Although choline has no intrinsic antimicrobial activity, addition of this compound to standard PN formulations led to decreased biochemical markers of liver injury and decreased radiographic evidence of steatosis in a small group of adults on chronic PN.45,46 Addition of glutamine to PN has been found to improve systemic and hepatic responses to Gram-negative bacterial challenge, and protect hepatic energy stores during acute stress from hemorrhagic shock.91,92 As the primary metabolic fuel of enterocytes, glutamine protects against intestinal ischemia–reperfusion injury and decreases the incidence of intestinal bacterial translocation.93,94
GASTROINTESTINAL HORMONES Addition of glucagon to PN infusions decreases portal and peripheral insulin/glucagon ratios, decreases total hepatic lipid content, and attenuates periportal fatty infiltration in adult rats receiving high-dextrose PN.95 Infusion of glucagon-like petide-2 in animal models maintained on PN rapidly increases portal blood flow, decreases intestinal crypt and villus apoptosis, stimulates glucose and protein uptake with a subsequent increase in local anabolic metabolism, and results in histologically healthy-appearing small intestinal mucosa.96–98
CYCLIC INFUSION OF PN Cyclic infusion of PN is advocated as hepatoprotective based on observations that early use of discontinuous infusions prevented further deterioration of biochemical markers in patients with PNassociated cholestasis.99 Continuous infusion has the potential to lead to constant release of insulin, stimulating persistent fatty acid synthesis. In addition to potentiating a more physiologic pattern of insulin secretion, cyclical PN may protect the hepatocyte from secondary injury due to hypoxia or infection by preserving mitochondrial respiratory chain function.100
ENTERAL STIMULATION Perhaps the most important, and most effective, method for preventing and treating PN-associated hepatobiliary complications is early reinstitution of enteral feeds. Enteral feedings are protective via a number of mechanisms. Reintroduction of enteral feeds following a period of fasting rapidly reverses the mucosal atrophy and intestinal dysfunction associated with lack of enteral stimulation. Enteral feeds stimulate bile flow, and feeds enriched with the glutathione precursors cysteine, glycine, and glutamine may increase hepatic glutathione stores and offer protection from endotoxinmediated hepatocellular damage.101
Chapter 57 THE LIVER AND PARENTERAL NUTRITION
TRANSPLANTATION In patients with end-stage liver disease secondary to PN, transplantation may be the only therapeutic option available. Progressive hepatic dysfunction is an indication for intestinal transplantation in patients with no realistic hope for sufficient gut adaptation. In patients believed to be capable of achieving freedom from PN if not for the presence of end-stage liver disease, isolated liver transplantation or sequential intestinal–liver transplantation may be an option; however, in patients with decompensated end-stage liver disease and no hope of freedom from PN, combined intestinal–liver or multiorgan transplant is the only hope for long-term survival.
PROGNOSIS AND NATURAL HISTORY Spontaneous resolution of hepatobiliary abnormalities over time after discontinuation of PN is the norm in most adult and pediatric patients. Persistence and continued progression of liver disease are less common. Prognosis is dependent on the degree and type of hepatobiliary abnormalities, as well as the underlying disease process necessitating PN. In Messing’s follow-up of 124 adult patients receiving PN for nonmalignant short-bowel syndrome, 6% of patients with permanent intestinal failure died from liver failure. No patients with reversible intestinal failure died from PN-associated liver disease.102 In Chan’s survey of adult patients receiving home TPN for more than 1 year, the median survival without cirrhosis on home TPN was 20 ± 5.1 years, although 15% progressed to advanced liver disease.103 Those who progressed to end-stage liver disease had 100% mortality at an average of 10.8 ± 7.1 months after initial elevation of serum bilirubin. The presence of an underlying inflammatory condition seemed to be a risk factor for advanced liver disease in this population.103 In Cavicchi’s study of 90 patients on long-term PN, 41.5% developed complicated PN-associated liver disease. The incidence of complicated liver disease increased with duration of PN and was associated with parenteral lipid intake of greater than 1 g/kg per day and the presence of chronic cholestasis. In this series, 22% of patients died from PN-associated liver disease.11 In addition to finding a decreased incidence of PNAC over time in infants, Kubota et al. also identified a decrease in mortality related to PNAC from 13% for patients treated from 1971 to 1983, to 3% in the group treated from 1992 to 1996.5 In another study on the incidence and outcome of PNAC in infants with neonatal intestinal resection and prolonged PN dependence, 25% of those who developed cholestasis progressed to liver failure. All of these patients either died or were listed for combined intestinal–liver transplant.3
CONCLUSIONS PN may be associated with a number of hepatobiliary complications, ranging from asymptomatic elevation of serum biochemistries to cholestasis or steatohepatitis with progression to cirrhosis. The risk of hepatobiliary complications clearly increases with increased time of exposure to PN and decreased enteral stimulation. The etiology of these complications is complex and is dependent on a number of
factors related to the patient and to the PN formulation. The neonate is particularly susceptible, due, in part, to immaturity of hepatic excretory function. Although the only definitive therapy for PN-associated hepatobiliary complications is discontinuation of PN, this is seldom possible. Advances in PN formulations, advances in the management of gastrointestinal diseases necessitating PN, and the development of intestinal transplantation as a viable option for patients with intestinal failure have combined to improve the outcome of PN-dependent patients. Ongoing research addressing the epidemiology, pathophysiology, and treatment of PN-mediated hepatic injury promises further strides in the understanding and management of this disorder.
REFERENCES 1. Kelly DA. Liver complications of pediatric parenteral nutrition – epidemiology. Nutrition 1998; 14:153–157. 2. Baserga MC, Sola A. Intrauterine growth restriction impacts tolerance to total parenteral nutrition in extremely low birth weight infants. J Perinatol 2004; 24:476–481. 3. Sondheimer JM, Asturias E, Cadnapaphornchai M. Infection and cholestasis in neonates with intestinal resection and long-term parenteral nutrition. J Pediatr Gastroenterol Nutr 1998; 27:131–137. 4. Beale EF, Nelson RM, Bucciarelli RL, et al. Intrahepatic cholestasis associated with parenteral nutrition in premature infants. Pediatrics 1979; 64:342–347. 5. Kubota A, Yonekura T, Hoki M, et al. Total parenteral nutritionassociated intrahepatic cholestasis in infants: 25 years’ experience. J Pediatr Surg 2000; 35:1049–1051. 6. Andorsky DJ, Lund DP, Lillehei CW, et al. Nutritional and other postoperative management of neonates with short bowel syndrome correlates with clinical outcomes. J Pediatr 2001; 139:27–33. 7. Luman W, Shaffer JL. Prevalence, outcome and associated factors of deranged liver function tests in patients on home parenteral nutrition. Clin Nutr 2002; 21:337–343. 8. Lindor KD, Fleming CR, Abrams A, Hirschkorn MA. Liver function values in adults receiving total parenteral nutrition. JAMA 1979; 241:2398–2400. 9. Baker AL, Rosenberg IH. Hepatic complications of total parenteral nutrition. Am J Med 1987; 82:489–497. 10. Bowyer BA, Fleming CR, Ludwig J, et al. Does long-term home parenteral nutrition in adult patients cause chronic liver disease? JPEN J Parenter Enteral Nutr 1985; 9:11–17. 11. Cavicchi M, Beau P, Crenn P, et al. Prevalence of liver disease and contributing factors in patients receiving home parenteral nutrition for permanent intestinal failure. Ann Intern Med 2000; 132:525–532. 12. Messing B, Bories C, Kunstlinger F, Bernier JJ. Does total parenteral nutrition induce gallbladder sludge formation and lithiasis? Gastroenterology 1983 ;84:1012–1019. 13. Manji N, Bistrian BR, Mascioli EA et al. Gallstone disease in patients with severe short bowel syndrome dependent on parenteral nutrition. JPEN J Parenter Enteral Nutr 1989; 13:461–464. 14. Roslyn JJ, Pitt HA, Mann LL, et al. Gallbladder disease in patients on long-term parenteral nutrition. Gastroenterology 1983; 84:148–154. 15. Colomb V, Goulet O, Rambaud C, et al. Long-term parenteral nutrition in children: liver and gallbladder disease. Transplant Proc 1992; 24:1054–1055. 16. Suchy FJ, Balistreri WF, Heubi JE, et al. Physiologic cholestasis: elevation of the primary serum bile acid
1095
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
1096
concentrations in normal infants. Gastroenterology 1981; 80:1037–1041. Hardikar W, Ananthanarayanan M, Suchy FJ. Differential ontogenic regulation of basolateral and canalicular bile acid transport proteins in rat liver. J Biol Chem 1995; 270:20841–20846. Tomer G, Ananthanarayanan M, Weymann A, et al. Differential developmental regulation of rat liver canalicular membrane transporters Bsep and Mrp2. Pediatr Res 2003; 53:288–294. Shneider BL, Dawson PA, Christie DM, et al. Cloning and molecular characterization of the ontogeny of a rat ileal sodium-dependent bile acid transporter. J Clin Invest 1995; 95:745–754. Balasubramaniyan N, Shahid M, Suchy FJ, Ananthanarayanan M. Multiple mechanisms of ontogenic regulation of nuclear receptors during rat liver development. Am J Physiol Gastrointest Liver Physiol 2004; 288(2):G251–260. Fouin-Fortunet H, Le Quernec L, Erlinger S, et al. Hepatic alterations during total parenteral nutrition in patients with inflammatory bowel disease: a possible consequence of lithocholate toxicity. Gastroenterology 1982; 82:932–937. Das JB, Uzoaru IL, Ansari GG. Biliary lithocholate and cholestasis during and after total parenteral nutrition: an experimental study. Proc Soc Exp Biol Med 1995; 210:253–259. Balistreri WF, Suchy FJ, Farrell MK, Heubi JE. Pathologic versus physiologic cholestasis: elevated serum concentration of a secondary bile acid in the presence of hepatobiliary disease. J Pediatr 1981; 98:399–402. Sonoda J, Xie W, Rosenfeld JM, et al. Regulation of a xenobiotic sulfonation cascade by nuclear pregnane X receptor (PXR). Proc Natl Acad Sci USA 2002; 99:13801–13806. Staudinger JL, Goodwin B, Jones SA, et al. The nuclear receptor PXR is a lithocholic acid sensor that protects against liver toxicity. Proc Natl Acad Sci USA 2001; 98:3369–3374. Xie W, Radominska-Pandya A, Shi Y, et al. An essential role for nuclear receptors SXR/PXR in detoxification of cholestatic bile acids. Proc Natl Acad Sci USA 2001; 98:3375–3380. Lacroix D, Sonnier M, Moncion A, et al. Expression of CYP3A in the human liver – evidence that the shift between CYP3A7 and CYP3A4 occurs immediately after birth. Eur J Biochem 1997; 247:625–634. Kearns GL, Abdel-Rahman SM, Alander SW, et al. Developmental pharmacology – drug disposition, action, and therapy in infants and children. N Engl J Med 2003; 349:1157–1167. Beer SF, Bircham PM, Bloom SR, et al. The effect of a 72-h fast on plasma levels of pituitary, adrenal, thyroid, pancreatic and gastrointestinal hormones in healthy men and women. J Endocrinol 1989; 120:337–350. Lucas A, Bloom SR, Aynsley-Green A. Metabolic and endocrine consequences of depriving preterm infants of enteral nutrition. Acta Paediatr Scand 1983; 72:245–249. Aynsley-Green A, Lucas A, Lawson GR, Bloom SR. Gut hormones and regulatory peptides in relation to enteral feeding, gastroenteritis, and necrotizing enterocolitis in infancy. J Pediatr 1990; 117:S24–S32. Nightingale JM, Kamm MA, van der Sijp JR, et al. Gastrointestinal hormones in short bowel syndrome. Peptide YY may be the ‘colonic brake’ to gastric emptying. Gut 1996; 39:267–272. Sharman-Koendjbiharie M, Piena-Spoel M, Hopman WP, et al. Gastrointestinal hormone secretion after surgery in neonates with congenital intestinal anomalies during starvation and introduction of enteral nutrition. J Pediatr Surg 2003; 38:1602–1606. Kaji T, Takamatsu H, Kajiya H. Motility of the gastrointestinal tract and gallbladder during long-term total parenteral nutrition in dogs. JPEN J Parenter Enteral Nutr 2002; 26:198–204.
35. Lucas A, Bloom SR, Aynsley-Green A. Gut hormones and ‘minimal enteral feeding’. Acta Paediatr Scand 1986; 75:719–723. 36. Reimund JM, Duclos B, Arondel Y, Baumann R. Persistent inflammation and immune activation contribute to cholestasis in patients receiving home parenteral nutrition. Nutrition 2001; 17:300–304. 37. Wolf A, Pohlandt F. Bacterial infection: the main cause of acute cholestasis in newborn infants receiving short-term parenteral nutrition. J Pediatr Gastroenterol Nutr 1989; 8:297–303. 38. Pappo I, Bercovier H, Berry E, et al. Antitumor necrosis factor antibodies reduce hepatic steatosis during total parenteral nutrition and bowel rest in the rat. JPEN J Parenter Enteral Nutr 1995; 19:80–82. 39. Forrest EH, Oien KA, Dickson S, et al. Improvement in cholestasis associated with total parenteral nutrition after treatment with an antibody against tumour necrosis factor alpha. Liver 2002; 22:317–320. 40. Dickerson RN, Karwoski CB. Endotoxin-mediated hepatic lipid accumulation during parenteral nutrition in rats. J Am Coll Nutr 2002; 21:351–356. 41. Trauner M, Fickert P, Stauber RE. Inflammation-induced cholestasis. J Gastroenterol Hepatol 1999; 14:946–959. 42. Paik YH, Schwabe RF, Bataller R et al. Toll-like receptor 4 mediates inflammatory signaling by bacterial lipopolysaccharide in human hepatic stellate cells. Hepatology 2003; 37:1043–1055. 43. Chessex P, Lavoie JC, Rouleau T, et al. Photooxidation of parenteral multivitamins induces hepatic steatosis in a neonatal guinea pig model of intravenous nutrition. Pediatr Res 2002; 52:958–963. 44. Sokol RJ, Taylor SF, Devereaux MW, et al. Hepatic oxidant injury and glutathione depletion during total parenteral nutrition in weanling rats. Am J Physiol 1996; 270: G691–G700. 45. Buchman AL, Dubin MD, Moukarzel AA, et al. Choline deficiency: a cause of hepatic steatosis during parenteral nutrition that can be reversed with intravenous choline supplementation. Hepatology 1995; 22:1399–1403. 46. Buchman AL, Ament ME, Sohel M, et al. Choline deficiency causes reversible hepatic abnormalities in patients receiving parenteral nutrition: proof of a human choline requirement: a placebo-controlled trial. JPEN J Parenter Enteral Nutr 2001; 25:260–268. 47. Burt ME, Hanin I, Brennan MF. Choline deficiency associated with total parenteral nutrition. Lancet 1980; 2:638–639. 48. Colomb V, Jobert-Giraud A, Lacaille F, et al. Role of lipid emulsions in cholestasis associated with long-term parenteral nutrition in children. JPEN J Parenter Enteral Nutr 2000; 24:345–350. 49. Tazuke Y, Drongowski RA, Btaiche I, et al. Effects of lipid administration on liver apoptotic signals in a mouse model of total parenteral nutrition (TPN). Pediatr Surg Int 2004; 20:224–228. 50. Cavicchi M, Crenn P, Beau P, et al. Severe liver complications associated with long-term parenteral nutrition are dependent on lipid parenteral input. Transplant Proc 1998; 30:2547. 51. Rubin M, Moser A, Vaserberg N, et al. Structured triacylglycerol emulsion, containing both medium- and long-chain fatty acids, in long-term home parenteral nutrition: a double-blind randomized cross-over study. Nutrition 2000; 16:95–100. 52. Marin JJ, Bravo P, Perez Barriocanal F, et al. Hyperglycemiainduced cholestasis in the isolated perfused rat liver. Hepatology 1991; 14:184–191. 53. Garcia-Marin JJ, Villanueva GR, Esteller A. Diabetes-induced cholestasis in the rat: possible role of hyperglycemia and hypoinsulinemia. Hepatology 1988; 8:332–340.
Chapter 57 THE LIVER AND PARENTERAL NUTRITION
54. Reif S, Tano M, Oliverio R, et al. Total parenteral nutritioninduced steatosis: reversal by parenteral lipid infusion. JPEN J Parenter Enteral Nutr 1991; 15:102–104. 55. Keim NL, Mares-Perlman JA. Development of hepatic steatosis and essential fatty acid deficiency in rats with hypercaloric, fatfree parenteral nutrition. J Nutr 1984; 114:1807–1815. 56. Nussbaum MS, Li S, Bower RH, et al. Addition of lipid to total parenteral nutrition prevents hepatic steatosis in rats by lowering the portal venous insulin/glucagon ratio. JPEN J Parenter Enteral Nutr 1992; 16:106–109. 57. Shattuck KE, Grinnell CD, Rassin DK. Amino acid infusions induce reversible, dose-related decreases in bile flow in the isolated rat liver. JPEN J Parenter Enteral Nutr 1993; 17:171–176. 58. Guertin F, Roy CC, Lepage G, et al. Effect of taurine on total parenteral nutrition-associated cholestasis. JPEN J Parenter Enteral Nutr 1991; 15:247–251. 59. Cooke RJ, Whitington PF, Kelts D. Effect of taurine supplementation on hepatic function during short-term parenteral nutrition in the premature infant. J Pediatr Gastroenterol Nutr 1984; 3:234–238. 60. Moss RL, Haynes AL, Pastuszyn A, Glew RH. Methionine infusion reproduces liver injury of parenteral nutrition cholestasis. Pediatr Res 1999; 45:664–668. 61. Schimpl G, Pesendorfer P, Steinwender G, et al. Allopurinol and glutamine attenuate bacterial translocation in chronic portal hypertensive and common bile duct ligated growing rats. Gut 1996; 39:48–53. 62. Fok TF, Chui KK, Cheung R, et al. Manganese intake and cholestatic jaundice in neonates receiving parenteral nutrition: a randomized controlled study. Acta Paediatr 2001; 90:1009–1015. 63. Balistreri WF, Farrell MK, Bove KE. Lessons from the E-Ferol tragedy. Pediatrics 1986; 78:503–506. 64. Loff S, Kabs F, Subotic U, et al. Kinetics of diethylhexylphthalate extraction from polyvinylchloride-infusion lines. JPEN J Parenter Enteral Nutr 2002; 26:305–309. 65. Iyer KR, Spitz L, Clayton P. BAPS prize lecture: New insight into mechanisms of parenteral nutrition-associated cholestasis: role of plant sterols. British Association of Paediatric Surgeons. J Pediatr Surg 1998; 33:1–6. 66. Clayton PT, Whitfield P, Iyer K. The role of phytosterols in the pathogenesis of liver complications of pediatric parenteral nutrition. Nutrition 1998; 14:158–164. 67. Carter BA, Prendergast DR, von Furstenberg R, Karpen SJ. Soy lipid-derived stigmasterol suppresses bile acid-activated bile salt export pump (BSEP, ABCB11) gene expression – role in total parenteral nutrition-associated cholestasis (TPNAC). Hepatology 2004; 40 (Suppl. 1):200A. 68. Krawinkel MB. Parenteral nutrition-associated cholestasis – what do we know, what can we do? Eur J Pediatr Surg 2004; 14:230–234. 69. Forchielli ML, Walker WA. Nutritional factors contributing to the development of cholestasis during total parenteral nutrition. Adv Pediatr 2003; 50:245–267. 70. Zambrano E, El-Hennawy M, Ehrenkranz RA, et al. Total parenteral nutrition induced liver pathology: an autopsy series of 24 newborn cases. Pediatr Dev Pathol 2004; 7:425–432. 71. Rodrigues CM, Fan G, Wong PY, et al. Ursodeoxycholic acid may inhibit deoxycholic acid-induced apoptosis by modulating mitochondrial transmembrane potential and reactive oxygen species production. Mol Med 1998; 4:165–178. 72. Rodrigues CM, Fan G, Ma X, et al. A novel role for ursodeoxycholic acid in inhibiting apoptosis by modulating mitochondrial membrane perturbation. J Clin Invest 1998; 101:2790–2799. 73. Gunsar C, Melek M, Karaca I, et al. The biochemical and histopathological effects of ursodeoxycholic acid and
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
metronidazole on total parenteral nutrition-associated hepatic dysfunction: an experimental study. Hepatogastroenterology 2002; 49:497–500. Heubi JE, Wiechmann DA, Creutzinger V, et al. Tauroursodeoxycholic acid (TUDCA) in the prevention of total parenteral nutrition-associated liver disease. J Pediatr 2002; 141:237–242. Beau P, Labat-Labourdette J, Ingrand P, Beauchant M. Is ursodeoxycholic acid an effective therapy for total parenteral nutrition-related liver disease? J Hepatol 1994; 20:240–244. Spagnuolo MI, Iorio R, Vegnente A, Guarino A. Ursodeoxycholic acid for treatment of cholestasis in children on long-term total parenteral nutrition: a pilot study. Gastroenterology 1996; 111:716–719. Levine A, Maayan A, Shamir R, et al. Parenteral nutritionassociated cholestasis in preterm neonates: evaluation of ursodeoxycholic acid treatment. J Pediatr Endocrinol Metab 1999; 12:549–553. Chen CY, Tsao PN, Chen HL, et al. Ursodeoxycholic acid (UDCA) therapy in very-low-birth-weight infants with parenteral nutrition-associated cholestasis. J Pediatr 2004; 145:317–321. Prescott WA Jr, Btaiche IF. Sincalide in patients with parenteral nutrition-associated gallbladder disease. Ann Pharmacother 2004; 38:1942–1945. Teitelbaum DH, Han-Markey T, Drongowski RA, et al. Use of cholecystokinin to prevent the development of parenteral nutrition-associated cholestasis. JPEN J Parenter Enteral Nutr 1997; 21:100–103. Dawes LG, Muldoon JP, Greiner MA, Bertolotti M. Cholecystokinin increases bile acid synthesis with total parenteral nutrition but does not prevent stone formation. J Surg Res 1997; 67:84–89. Wu ZS, Yu L, Lin YJ, et al. Rapid intravenous administration of amino acids prevents biliary sludge induced by total parenteral nutrition in humans. J Hepatobiliary Pancreat Surg 2000; 7:504–509. Nealon WH, Upp JR Jr, Alexander RW, et al. Intravenous amino acids stimulate human gallbladder emptying and hormone release. Am J Physiol 1990; 259:G173–G178. Broughton G 2nd, Fitzgibbons RJ Jr, Geiss RW, et al. IV chenodeoxycholate prevents calcium bilirubinate gallstones during total parenteral nutrition in the prairie dog. JPEN J Parenter Enteral Nutr 1996; 20:187–193. Gumpricht E, Dahl R, Devereaux MW, Sokol RJ. Beta-carotene prevents bile acid-induced cytotoxicity in the rat hepatocyte: evidence for an antioxidant and anti-apoptotic role of betacarotene in vitro. Pediatr Res 2004; 55:814–821. Li N, Zhang HH, Wang SH, et al. S-adenosylmethionine in treatment of cholestasis after total parenteral nutrition: laboratory investigation and clinical application. Hepatobiliary Pancreat Dis Int 2002; 1:96–100. Belli DC, Fournier LA, Lepage G, et al. S-adenosylmethionine prevents total parenteral nutrition-induced cholestasis in the rat. J Hepatol 1994; 21:18–23. Belli DC, Albrecht R, La Scala GC, et al. Homocysteine prevents total parenteral nutrition (TPN)-induced cholestasis without changes in hepatic oxidative stress in the rat. J Pediatr Gastroenterol Nutr 2003; 36:200–205. Li Z, Yang S, Lin H, et al. Probiotics and antibodies to TNF inhibit inflammatory activity and improve nonalcoholic fatty liver disease. Hepatology 2003; 37:343–350. Pappo I, Bercovier H, Berry EM, et al. Polymyxin B reduces total parenteral nutrition-associated hepatic steatosis by its antibacterial activity and by blocking deleterious effects of lipopolysaccharide. JPEN J Parenter Enteral Nutr 1992; 16:529–532.
1097
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
91. Dhar A, Kujath S, Van Way CW 3rd. Glutamine administration during total parenteral nutrition protects liver adenosine nucleotides during and after subsequent hemorrhagic shock. JPEN J Parenter Enteral Nutr 2003; 27:246–251. 92. Lin MT, Saito H, Furukawa S, et al. Alanyl-glutamine enriched total parenteral nutrition improves local, systemic, and remote organ responses to intraperitoneal bacterial challenge. JPEN J Parenter Enteral Nutr 2001; 25:346–351. 93. Wu GH, Wang H, Zhang YW, et al. Glutamine supplemented parenteral nutrition prevents intestinal ischemia-reperfusion injury in rats. World J Gastroenterol 2004; 10:2592–2594. 94. Ding LA, Li JS. Effects of glutamine on intestinal permeability and bacterial translocation in TPN-rats with endotoxemia. World J Gastroenterol 2003; 9:1327–1332. 95. Li SJ, Nussbaum MS, McFadden DW, et al. Addition of glucagon to total parenteral nutrition (TPN) prevents hepatic steatosis in rats. Surgery 1988; 104:350–357. 96. Chance WT, Sheriff S, Foley-Nelson T, et al. Maintaining gut integrity during parenteral nutrition of tumor-bearing rats: effects of glucagon-like peptide 2. Nutr Cancer 2000; 37:215–222. 97. Guan X, Stoll B, Lu X, et al. GLP-2-mediated up-regulation of intestinal blood flow and glucose uptake is nitric oxide-
1098
98.
99.
100.
101.
102.
103.
dependent in TPN-fed piglets 1. Gastroenterology 2003; 125:136–147. Burrin DG, Stoll B, Guan X, et al. Glucagon-like peptide 2 dose-dependently activates intestinal cell survival and proliferation in neonatal piglets. Endocrinology 2005; 146:22–32. Hwang TL, Lue MC, Chen LL. Early use of cyclic TPN prevents further deterioration of liver functions for the TPN patients with impaired liver function. Hepatogastroenterology 2000; 47:1347–1350. Morikawa N, Suematsu M, Kyokane T, et al. Discontinuous total parenteral nutrition prevents postischemic mitochondrial dysfunction in rat liver. Hepatology 1998; 28:1289–1299. Dzakovic A, Kaviani A, Eshach-Adiv O, et al. Trophic enteral nutrition increases hepatic glutathione and protects against peroxidative damage after exposure to endotoxin. J Pediatr Surg 2003; 38:844–847. Messing B, Crenn P, Beau P, et al. Long-term survival and parenteral nutrition-dependency of adult patients with nonmalignant short bowel. Transplant Proc 1998; 30:2548. Chan S, McCowen KC, Bistrian BR, et al. Incidence, prognosis, and etiology of end-stage liver disease in patients receiving home total parenteral nutrition. Surgery 1999; 126:28–34.
Section IX: Liver Affected by Other Organs (Non-Hepatic Conditions)
58
PREOPERATIVE AND POSTOPERATIVE HEPATIC DYSFUNCTIONS Enrique J. Martinez and Thomas D. Boyer Abbreviations ALT alanine aminotransferase AST aspartate aminotransferase
LDH MELD
lactate dehydrogenase model of end-stage liver disease
INTRODUCTION It is estimated that up to 10% of cirrhotic patients will undergo surgery in the final 2 years of their lives.1,2 This leads to the need for a careful review of the pre- and postoperative liver dysfunction of which physicians will need to be aware as these cases present. In this chapter, we will review many aspects of what is known with regard to how the normal and diseased liver reacts to the operative procedures. The liver is the major site for the metabolism of many drugs and is responsible for the synthesis of most serum proteins and the removal of endogenous toxins. Therefore postoperative liver dysfunction may slow the recovery of a patient who has undergone surgery. Because of the liver’s role in drug metabolism, it is susceptible to injury by a variety of xenobiotics. Alterations in hepatic blood flow may also affect liver function, especially in the patient with underlying chronic liver disease. Therefore it is not unexpected that abnormalities of liver tests are noted frequently in patients after surgery.3 However, clinical jaundice is rare (<1%) in patients with normal livers, and its development should prompt a thorough evaluation of its cause. Before reviewing the causes of postoperative liver dysfunction, there will be a brief discussion of the evaluation of the patient with abnormal liver tests found during preoperative testing. If the surgery is critical to the patient, the presence of preoperative liver dysfunction should have no effect on the decision to operate. However, if the surgery is elective, the finding of abnormal liver tests preoperatively should prompt an evaluation as to their cause, and an estimate of liver function and reserve should be made.
PREOPERATIVE LIVER DYSFUNCTION The frequency of unsuspected liver disease is approximately 1 in 700 of otherwise healthy surgical candidates, making this a common clinical problem.4,5 Of most concern to surgeons and anesthesiologists are elevations of the serum aspartate aminotransferase (AST), alanine aminotransferase (ALT), bilirubin, and disorders of coagulation. Elevations of alkaline phosphatase or g-glutamyltranspeptidase are of little clinical significance and should not prompt an evalua-
rFVIIa
recombinant factor VIIa
tion unless associated with other clinical findings (Chapter 14). The significance of a low serum albumin alone is difficult to interpret because of the multiple causes of a fall in the concentration of this protein (Chapter 14). It is useful to combine patients with preoperative liver dysfunction into three groups: (1) asymptomatic with normal physical examination; (2) symptomatic; and (3) physical and biochemical findings of chronic liver disease. Asymptomatic patients with a normal physical examination and abnormal liver tests (increased AST/ALT) are encountered frequently, raising concerns that they have underlying liver disease that may increase the risk of surgery. The abnormal liver tests most likely reflect a subacute or chronic form of liver injury resulting from viral hepatitis, drugs, and/or alcohol, or fatty liver. We lack information as to whether this group of patients is at any increased risk from surgery, but it is the authors’ opinion that if the serum bilirubin, albumin, and clotting tests are normal and the increase in aminotransferases is mild (twofold to fivefold), there is little, if any, increase in surgical risk. Deciding to perform surgery and ensuring its successful completion in this type of patient should not be an end to the evaluation of the liver disease. After surgery the liver tests need to be monitored; if persistently abnormal, a thorough evaluation should be performed. Greater increases in aminotransferases are more problematic only because they suggest more significant hepatic injury. Anesthetic agents may adversely affect hepatic function because of decreases in splanchnic blood flow and thereby oxygen delivery to the liver.6,7 Early studies suggested that this may be a real clinical concern when an increase in operative mortality was observed in patients with acute hepatitis.8–10 However, in other studies no increase in mortality has been found when surgery was performed in patients with coincidental acute viral hepatitis.11–13 All of these studies suffer from being anecdotal and reporting on small numbers of patients. In addition many patients were jaundiced, suggesting the presence of significant liver injury. Despite these uncertainties, if the surgery is elective, a delay in the operation is the most conservative approach. Liver tests should he performed 2–3 weeks later, and if the abnormalities persist for several months, the patient should be evaluated completely. If it is decided to perform an elective operation when AST/ALT is elevated more than fivefold but a normal albumin, prothrombin time, and bilirubin are noted, the risk to the patient
1099
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
probably remains small. However, if the liver tests worsen postoperatively, defining whether the surgery or the initial illness is at fault will be difficult. Because there is no evidence that the presence of liver disease increases the risk of developing anesthetic-induced hepatitis14 (Chapter 26), there is no need to alter the choice of anesthetic agent used for the surgery. Isolated elevations in the serum bilirubin are usually due to Gilbert’s syndrome (Chapter 74); this syndrome does not increase the risk of surgery. If the elevations of bilirubin are seen in association with elevations of AST/ALT or alkaline phosphatase, the cause of the liver injury needs to be determined before any elective surgery is performed. Isolated prolongation of the prothrombin time is an unusual manifestation of liver disease but suggests the presence of cirrhosis or a severe acute or subacute injury. The symptomatic patient with elevated liver tests is of greater concern when elective surgery is planned. The presence of symptoms (nausea, vomiting) in association with elevated liver tests suggests that the patient is developing an acute illness that may worsen before it improves. Therefore if the patient is subjected to surgery and the liver tests worsen, it will be very difficult to determine whether the patient is suffering from a pre-existing condition or a complication of surgery (i.e., anesthetic-induced hepatitis). In addition, there is uncertainty as to whether or not acute viral hepatitis increases the risk of surgery;9–13 therefore, elective surgery should be postponed until the hepatitis has resolved. A special subset of the symptomatic patient is represented by patients with markedly elevated transaminases. This group is varied as to cause of these elevations: causes range from ischemic hepatitis or cardiac dysfunction, drugs, liver trauma, cancer metastases to the liver, to rhabdomyolysis. A recent review demonstrated that the overall mortality of patients with serum AST greater than 3000 IU/l was 55% and that ischemic hepatitis patients had a 75% mortality compared to 33% for all other causes. This group should clearly be excluded from consideration for any elective surgery.15 The presence of clinical (i.e., splenomegaly, spiders, palmar erythema, or ascites) and biochemical (i.e., low albumin or prolonged prothrombin time) evidence of chronic liver disease is of greatest concern when planning elective surgery. The risk of the surgery is determined by how well the liver is functioning16 and the presence or absence of symptoms. For example, patients with chronic hepatitis appear to have an increased surgical risk if they are symptomatic.16,17 Elective surgeries in patients with evidence of chronic liver disease should be delayed until the cause of the liver disease is determined and the severity of the injury is fully assessed. The most common cause of chronic liver disease in the western world is alcoholism. Alcoholic liver disease can manifest as fatty liver, alcoholic hepatitis, cirrhosis, or a combination of the above (Chapter 29). The risk of surgery in the patient with fatty liver appears to be small.16 Surgical studies would suggest that it is not uncommon to find unsuspected cirrhosis in obese patients at the time of bariatric surgery. In a recent report of 125 patients with cirrhosis detected at surgery, the authors were able to proceed with their planned bariatric surgery in 74% of patients; the result was no intraoperative deaths and only a 4% mortality rate.18 Alcoholic hepatitis can be present as an asymptomatic illness or may be associated with jaundice and liver failure. Elective surgery in a patient with decompensated liver disease resulting from any cause is ill advised.
1100
Even in patients with better preserved liver function, elective surgery in patients with acute symptomatic alcoholic hepatitis is associated with increased morbidity and mortality and should not he performed.16,19 The effect of asymptomatic alcoholic hepatitis on surgical mortality has been studied best in patients undergoing portosystemic shunt surgery.20–23 The presence of large amounts of alcoholic hyaline in a liver biopsy specimen indicated a high likelihood of mortality in some series.24,25 In another series the 1-year survival of patients after the insertion of a portosystemic shunt was 70–74% in the absence of alcoholic hyaline and only 10% in those with alcoholic hyaline in a liver biopsy. In contrast, other investigators have found no correlation between the presence of alcoholic hyaline and survival.21–23 Despite these uncertainties, performance of elective surgery in a patient with alcoholic hepatitis should be avoided. It is very difficult, however, to determine whether an alcoholic patient has fatty liver or a more serious lesion (i.e., alcoholic hepatitis) based on liver test results alone. Therefore the most conservative approach in a chronic alcoholic patient who requires elective surgery and who has abnormal liver tests is either to perform a liver biopsy to define the nature of the liver injury or a period of abstinence (2–3 months) to allow the acute injury to resolve preceding surgery. There is little question that the presence of cirrhosis increases the risk of surgery, especially if it is an intra-abdominal operation.26–52 Published mortality rates vary from 5 to 67% and morbidity rates from 7 to 39%.18 These differences in survival reflect the variability in the clinical state of the patients reported in the different studies. The risk of surgery is defined best by the clinical severity of the cirrhosis (i.e., Child’s classification).16,26,27,31 Patients with good hepatic function, no ascites, and a good nutritional state (Child’s A) do well with surgery, whereas those with jaundice, low serum albumin, ascites, and muscle-wasting have a high operative mortality and postoperative morbidity. The presence of umbilical hernias is reported to be as high as 42% if ascites is present versus 10% if no ascites is present.31 In cases of spontaneous rupture of these umbilical hernias, supportive care is associated with a 50% mortality versus a 14% mortality with repair, but repair may be associated with wound dehiscence or peritonitis in a patient with cirrhotic ascites.32–34 Postoperative decreases in ascites either with medical management or with a transjugular intrahepatic portosystemic shunt reduce the recurrence of umbilical hernias postrepair.34 Groin hernia repair in cirrhotics appears to he less affected by ascites, with reports of 8% recurrence versus 60% for umbilical hernias in similar patients with ascites.35 Elective surgery can be performed in patients with decompensated liver disease but the surgeon must have experience in the management of this type of patient to minimize the risks of surgery. A difficult problem that faces many surgeons is performing a cholecystectomy in a patient with cirrhosis. Mortality rates of 7.5–25.5% and morbidity rates of 4.8–25% have been reported.26,36–46 In addition, many patients require transfusions, especially if they have decompensated liver disease.36 A subtotal cholecystectomy used to be suggested for patients with cirrhosis and portal hypertension.37 Previously, open cholecystectomy had been recommended for patients with cirrhosis, but recent studies would suggest that in Child’s A–B patients, elective laparoscopic cholecystectomy is preferred because of less bleeding, fewer wound infections, and decreased lengths of stay compared with open
Chapter 58 PREOPERATIVE AND POSTOPERATIVE HEPATIC DYSFUNCTIONS
cholecystectomy.38–43 The complication rates after urgent cholecystectomy are significantly higher (36%) compared to elective laparoscopic cholecystectomy in cirrhotics (16%), suggesting that, if possible, stabilization of the patient with medical management followed by elective surgery would be preferred.43 Given the high rates of mortality and morbidity in patients with advanced cirrhosis undergoing cholecystectomy and the increased difficulty associated with liver transplantation in patients with a previous cholecystectomy, the authors believe that the presence of recurrent cholecystitis in a patient with advanced liver disease should be considered an indication for liver transplantation while conservative management is being performed. This may necessitate an appeal to be filed with the United Network for Organ Sharing for those patients in the USA to allow them to have a higher-priority model of end-stage liver disease (MELD) score. In the acutely ill patient percutaneous cholecystostomy tubes may be attempted but may be difficult to perform in patients with significant ascites. Complication rates for cirrhotics with abdominal surgeries other than just cholecystectomy are also quite high. The overall mortality rate for urgent abdominal surgery has been reported to be 45–57% versus 10–18% with elective surgeries. Important predictive factors associated with mortality in some studies include the presence of ascites (P = 0.006), encephalopathy (P = 0.002), and coagulopathy (P = 0.021).43–46 The mortality rates by Child’s classification likewise demonstrates a progressive increased risk as liver disease worsens and are reported at 10% for class A, 30–31% for class B, and 76–82% for class C cirrhotics. The causes of the deaths from gastrointestinal surgeries in a recent series by Mansour et al. were coagulopathy or sepsis in 81% of cases.44 A prior report classified 87% of deaths as being due to multisystem organ failure.26 These data suggest that, wherever possible, the surgery should be converted from an emergent procedure to an elective one, but this may not be possible depending on the situation. It remains to be seen if the new MELD scores will be more predictive of operative morality for these procedures than is the Child’s classification. Given that coagulopathy plays a major role in postoperative abdominal surgery complications seen in cirrhotics, it is interesting to speculate what the future role of recombinant factor VIIa (rFVIIa) may be for these patients. Data on the use of rFVIIa have been growing in the literature and there are scattered reports of use of this medication for limited abdominal procedures such as diagnostic laparoscopy and some other radiological and surgical procedures.47–51 Recently, use of rFVIIa has been shown to improve the international normalized ratio in 49.8–58.5% of patients, regardless of the MELD score.51 It is unclear whether this effect is transient in nature, and if surgery is performed what impact this would have on overall postoperative bleeding complications and patient survival. Cardiac surgery in more advanced cirrhotic patients likewise has a reported high complication rate.28–30 No mortality was noted in one series of Child’s A patients who underwent coronary artery bypass grafting, valvular replacement, or both, but survival was reported to be 80% in Child’s B-class patients for similar surgeries. Likely contributors to the increased mortality include liver perfusion injuries related to cardiopulmonary bypass, as well as bypass aggravation of the coagulopathy which may be already be present in these patients. New advances in the use of cardiac stents and valvuloplasty techniques may help these patients by avoiding a thoracotomy.
Table 58-1. Causes of Postoperative Liver Dysfunction Hepatitis-like Drugs Anesthetic needs Ischemia Cardiac Shock (non-cardiac) Iatrogenic (ligation hepatic artery) Viral hepatitis Cholestasis Benign postoperative cholestasis Sepsis drugs Antibiotics, antiemetics Bile duct injury Choledocholithiasis/pancreatitis Cholecystitis
POSTOPERATIVE LIVER DYSFUNCTION Table 58-1 lists the most common causes of liver dysfunction and jaundice in postoperative patients. In a review of surgical complications in cirrhotics, the 30-day mortality rate was reported to be 11.6%, with a 30.1% perioperative complication rate.39 The causes of this mortality rate are separated into two large groups but many patients have a mixed picture, which increases the difficulty in making a correct cause to outcome association.
ANESTHETIC-INDUCED LIVER INJURY Anesthetic-induced hepatitis is of most concern because of a high incidence of hepatic failure; it is discussed in detail in Chapters 21 and 26. Liver injury resulting from any of the currently used anesthetics is rare.52,53 By 1985 more than 500 cases of halothane hepatitis had been reported; however, the estimated frequency of hepatitis resulting from halothane is 1 in 10 000 operations.14,53 The putative toxin is a metabolic product of halothane; therefore, other halogenated anesthetic agents that are less extensively metabolized than halothane have an even lower incidence of hepatotoxicity.14,53 Published mortality rates vary from 10 to 80%, with rates of 10–30% being most representative.53 The development of symptoms (fever or jaundice) is seen 7–14 days after a single exposure and 5–7 days after multiple exposures to halothane. Fever is the most common symptom of halothaneinduced hepatitis and may be present in the absence of jaundice. Laboratory tests show a marked increase (greater than 10-fold above normal) of serum levels of aminotransferases. Patients with severe injury may have a rise in serum bilirubin and prolongation of the prothrombin time. Many patients, however, have only a rise in aminotransferase levels without clinical icterus and therefore the injury may be missed unless laboratory tests are obtained in patients with postoperative fever.53–55 Eosinophilia and renal insufficiency may also be present.53–55 The other halogenated anesthetics also cause a hepatitis-like injury with clinical features that are similar to those seen with halothane hepatitis.
1101
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
Separately, the use of anesthetics is known to reduce the hepatic arterial blood flow and oxygen uptake by hepatocytes.6,7,56–58 In healthy volunteers, this reduction averaged 35% and was seen in the first 30 minutes of induction but returns to baseline during the procedure. It is possible that this fall in oxygen delivery may affect more the cirrhotic liver.
ISCHEMIC HEPATITIS Surgery is commonly performed in patients with cardiac disease, and, if they develop congestive heart failure, they may develop ischemic hepatitis. Ischemic hepatitis is marked by a rapid rise in serum levels of AST, ALT, and lactate dehydrogenase (LDH). The levels are frequently greater than 10-fold above normal and in severe cases may be associated with jaundice and prolongation of the prothrombin time.59 In contrast to halothane hepatitis, elevation in liver tests can he seen any time after surgery and is not associated with fever or eosinophilia. In addition, the liver tests tend to return rapidly to normal (elevations last 3–11 days).59 Non-cardiogenic shock liver is seen in association with hypotensive episodes (resulting from bleeding or sepsis) and also is marked by a rapid rise and fall in serum levels of AST and ALT. When comparing all causes of massive elevations of AST (more than 3000 IU/l), ischemic hepatitis can be found to be associated with a mortality rate of 75% versus 33% for all other causes combined.60 Accidental or deliberate ligation of the hepatic artery or its branches may result in hepatic ischemia and necrosis associated with a rise in serum levels of AST and ALT.61 When accidental ligation of the hepatic artery occurs, it is usually during cholecystectomy, and should be suspected in a patient who develops a rise in serum levels of AST and ALT after that operation.
ACUTE VIRAL HEPATITIS The development of acute viral hepatitis in the postoperative period is rare if patients are shown to have normal liver tests preceding the surgery. If the patient does develop acute viral hepatitis, a gradual rise in the serum levels of AST and ALT will be observed with or without other systemic symptoms. Of note, the serum levels of LDH are only slightly increased relative to the degree of elevation of the AST and ALT; hence, measurement of the serum LDH is a useful test for separating ischemic and drug-induced hepatitis from viral hepatitis. The appropriate tests for the diagnosis of acute viral hepatitis are discussed in Chapters 30–34.
DRUG-INDUCED HEPATITIS There are a large number of drugs that may cause an acute hepatitis-like injury and these are discussed in detail in Chapter 26. Acetaminophen is a direct hepatotoxin and is commonly used in the postoperative patient. Toxicity from acetaminophen is generally observed when more than 7.5 grams is ingested as a single dose.62 However, it has become apparent that therapeutic doses of acetaminophen ingested by alcoholics may be associated with significant hepatic injury.63 In addition, the toxic metabolite of acetaminophen is formed by P450 2E1; this enzyme is susceptible to induction by a number of drugs, including alcohol.62 The co-administration of an
1102
inducing agent may increase the generation of the toxic metabolite of acetaminophen and cause toxicity with ingestion of therapeutic doses (3–4 g/day) of the drug. Therefore if a patient develops an increase in serum levels of AST/ALT postoperatively, the amount of acetaminophen taken by the patient should be determined. Most hepatitis-like drug reactions are idiosyncratic and can be due to a variety of agents, as discussed in Chapter 26. An important fact in the differential diagnosis of postoperative liver dysfunction is that most idiosyncratic drug reactions develop after at least 2 weeks of therapy. Therefore the development of abnormal liver tests within 2 weeks of surgery is unlikely to be due to drugs started after the operation. In addition, drugs taken for more than 12 months preceding the surgery are unlikely causes of postoperative liver dysfunction.
BENIGN POSTOPERATIVE CHOLESTASIS The development of jaundice postoperatively is observed in less than 1% of patients undergoing major surgery. Patients with pre-existing liver or cardiovascular disease and who suffer trauma have a significantly higher incidence of postoperative jaundice.3 The cause of the jaundice is multifactorial and includes anesthetic-induced reduction in liver function because of decreased hepatic blood flow, increased pigment load from hematomas and transfused blood, and impaired bile formation secondary to bacterial sepsis.3,64–68 The breakdown of 50 ml of blood yields 250 mg of bilirubin, which can be easily handled by the normal liver but leads to a rapid rise in serum bilirubin in patients with impaired liver function.3 A fall in liver blood flow is observed with almost all general anesthetics, which may lead to a decline in liver function, especially in patients with underlying liver disease.6,7,16 If the patient has suffered an episode of hypotension, it will affect hepatocyte function and predispose the patient to the development of cholestasis.3 Last, endotoxemia reduces bile flow, and intra-abdominal sepsis is frequently associated with abnormal liver tests and jaundice.64,67 Benign postoperative cholestasis develops most commonly within the first 10 days after surgery. Cholestasis is observed most frequently in patients with sepsis, after cardiovascular surgery, and after prolonged operations during which and after the patient received multiple transfusions.64–67 Although an increase in serum bilirubin is common and may reach levels of 40 mg/dl, it is not universal. Serum levels of alkaline phosphatase are frequently elevated, whereas AST and ALT levels are normal or only mildly elevated (less than fivefold). Serum albumin levels may be normal or slightly reduced and the prothrombin time is usually normal. Liver biopsy shows cholestasis and variable degrees of fat.3,66 The condition is referred to as benign because, if the patient recovers from the surgery and any associated complications, the cholestasis resolves. However, patients developing postoperative cholestasis have a significant mortality. Patients with serum bilirubin levels of greater than 6 mg/dl have a 46% mortality if they have suffered abdominal trauma and an 86% mortality if they have intraabdominal sepsis.68 These latter groups of patients frequently die with the syndrome termed multiple organ system failure, and worsening liver function is seen in association with renal failure and acute respiratory distress syndrome.42,69 The liver plays a passive role
Chapter 58 PREOPERATIVE AND POSTOPERATIVE HEPATIC DYSFUNCTIONS
Table 58-2. Differential Diagnosis of Postoperative Liver Dysfunction Disorder
Type of surgery
Fever
Onset postoperatively
ALT (U/l)
AP
Halothane hepatitis Viral hepatitis Benign postoperative jaundice Shock Bile duct injury
No relationship No relationship Major surgery with sepsis
Common Uncommon Common
2–15 days >3 weeks <7 days
>500 >500 slight ≠
slight ≠ slight ≠ ≠≠
No relationship (cardiac disease) Biliary tract and stomach
Uncommon Common
1–4 days Days–weeks
>500 200–300
Slight ≠(LDH) ≠≠
ALT, Alanine aminotransferase; AP, alkaline phosphatase; LDH, lactate dehydrogenase; ≠, increase; ≠≠, greater increase.
in multiple organ system failure in that acute liver failure (encephalopathy with a coagulopathy) is not the cause of death.
BILE DUCT OBSTRUCTION A common concern is the development of extrahepatic bile duct obstruction in the postoperative patient who becomes icteric. Coincidental choledocholithiasis after surgery is rare. A far more common occurrence is bile duct injury after biliary tract or gastric surgery. Bile duct injury after laparoscopic cholecystectomy is an increasingly common problem and frequently goes unrecognized during the cholecystectomy.69–71 The patient develops clinical jaundice with or without signs of cholangitis days to weeks after the initial surgery. Diagnosis is made by endoscopic retrograde cholangiopancreatography or transhepatic cholangiography (see Chapters 16 and 64). Postoperative pancreatitis may also cause bile duct obstruction because of edema of the head of the pancreas. The diagnosis is made by finding an elevated serum level of amylase and a computed tomography scan of the abdomen showing edema of the pancreas and bile duct dilation. The jaundice resolves as the patient recovers from the pancreatitis.72 In the postoperative jaundiced patient who has not undergone biliary or gastric surgery and who does not have evidence of pancreatitis, biliary tract disease is uncommon and other causes of jaundice should be considered initially. Acute cholecystitis (calculous or acalculous) may occur postoperatively and can be associated with abnormal liver tests and jaundice.73 The presence of right upper quadrant abdominal pain and fever suggests the diagnosis, with the ultrasound findings of pericholecystic fluid, thickening of the gallbladder wall, and perhaps stones supporting the clinical suspicion.74 Gangrene, perforation, and empyema of the gallbladder are common in the postoperative patient and associated with a high mortality.73 Abnormal liver tests are frequently observed in patients receiving total parenteral nutrition, which is discussed in detail in Chapter 57. Fatty liver with mild elevations of the serum aminotransferases and alkaline phosphatase is commonly observed.75 Less common, but of greater concern, especially in children, is the development of jaundice. The abnormal liver tests develop days to weeks after the institution of therapy.75 The liver biopsy findings are non-specific and the diagnosis is one of excluding the other causes of postoperative hepatic dysfunction. The cause of the disorder remains poorly understood.
EVALUATION OF THE PATIENT WITH POSTOPERATIVE LIVER DYSFUNCTION If the patient is within the first 2 weeks of surgery and has a hepatitis-like injury, anesthetic-related hepatitis or ischemic hepatitis is of major concern (Table 58-2). Injury by a direct hepatotoxin such as acetaminophen should also be considered. The development of cholestasis in the immediate postoperative period in a patient who has undergone biliary or gastric surgery suggests bile duct injury. If the patient has undergone major cardiac or abdominal surgery and is infected or has received multiple blood transfusions, benign postoperative cholestasis should be the initial diagnosis. If the abnormal liver tests develop more than 2 weeks after surgery, drug or total parenteral nutrition-induced liver injury should be considered, as should bile duct injury if gallbladder surgery had been performed. Postoperative cholecystitis is associated with abdominal pain and fever, which are unusual features of the other types of injury, and abdominal ultrasonography should be performed in this situation. Hepatitis C should be considered in the transfused patient who develops elevated AST/ALT levels more than 3 weeks after exposure to blood products but is very rare unless the donor was incubating the virus when the blood was donated. Antibody tests may be negative during the acute illness and identification of viral ribonucleic acid in the serum by polymerase chain reaction may be required (Chapter 32). Tests for acute hepatitis A and B are not usually necessary because they infrequently cause post-transfusion hepatitis.
REFERENCES 1. Jackson FC, Christophersen ER, Peternel WW. Preoperative management of patients with liver disease. Surg Clin North Am 1968; 48:907. 2. LaMont JT. Postoperative jaundice. Surg Clin North Am 1974; 54:637. 3. Probst A, Probst T, Zengerl G, et al. Prognosis and life expectancy in chronic liver disease. Dig Dis Sci 1995; 40:1905. 4. Schemel WH. Unexpected hepatic dysfunction found by multiple laboratory screening. Anesth Analg (Cleve) 1976; 55:810.
1103
Section IX. Liver Affected by Other Organs (Non-Hepatic Conditions)
5. Watanecyawech M, Kelly KA Jr. Hepatic diseases unsuspected before surgery. NY State J Med 1975; 75:1278. 6. Ngai SH. Effects of anesthetics on various organs. N Engl J Med 1980; 302:564. 7. Cooperman LH. Effects of anesthesia on the splanchnic circulation. Br J Anaesth 1972; 44:967. 8. Handle DD, Summerskill WHJ. Surgery in acute hepatitis. JAMA 1963; 184:257. 9. Powell-Jackson P, Greenway B, Williams R. Adverse effects of exploratory laparotomy in patients with unsuspected liver disease. Br J Surg 1982; 69:419. 10. Shaldon S, Sherlock S. Virus hepatitis with features of prolonged bile retention. Br Med J 1957; 2:734. 11. Hardy KJ, Hughes ESR. Laparotomy in viral hepatitis. Med J Aust 1968; 1:710. 12. Strauss AA, Strauss SF, Schwartz AH, et al. Decompression by drainage of the common bile duct in subacute and chronic jaundice: a report of 73 cases with hepatitis or concomitant biliary duct infection as cause. Am J Surg 1959; 97:137. 13. Bourke JB, Cannon P, Ritchie HD. Laparotomy for jaundice. Lancet 1967; ii:521. 14. Farrell GC. Postoperative hepatic dysfunction. In Zakim D, Boyer TD, eds. Hepatology: a textbook of liver disease, 2nd edn. Philadelphia: WB Saunders, 1990:869. 15. Johnson RD, O’Connor ML, Kerr RM. Extreme serum elevations of aspartate aminotransferase. Am J Gastro 1995; 90:1244. 16. Friedman LS, Maddrey WC. Surgery in the patient with liver disease. Med Clin North Am 1987; 71:453. 17. Hargrove MD. Chronic active hepatitis: possible adverse effect of exploratory laparotomy. Surgery 1971; 68:771. 18. Brolin RE, Bradley LJ, Taliwal RV. Unsuspected cirrhosis discovered during elective obesity operations. Arch Surg 1998; 133:84. 19. Greenwood SM, Leffler CT, Minkowitz S. The increased mortality rate of open liver biopsy in alcoholic hepatitis. Surg Gynecol Obstet 1972; 134:600. 20. Eckhauser F, Appelman H, O’Leary T, et al. Hepatic pathology as a determinant of prognosis after portal decompression. Am J Surg 1980; 139:105. 21. Kanel G, Kaplan M, Zawacki I, et al. Survival in patients with post necrotic cirrhosis and Laennec’s cirrhosis undergoing therapeutic portacaval shunt. Gastroenterology 1977; 73:679. 22. Bell RH, Miyai K, Orloff MJ. Outcomes in cirrhotic patients with acute alcoholic hepatitis after emergency portacaval shunt for bleeding esophageal varices. Am J Surg 1984; 147:78. 23. Reichle R, Fahmy W, Golsorkhi M. Prospective comparative clinical trial with distal splenorenal and mesocaval shunts. Am J Surg 1979; 137:13. 24. Mikkelsen W. Therapeutic portacaval shunt. Preliminary data on controlled trial. Arch Surg 1974; 108:302. 25. Pande N, Resnick R, Yee W, et al. Cirrhotic portal hypertension. Morbidity of continued alcoholism. Gastroenterology 1978; 74:64. 26. Garrison RN, Cryer HM, Howard DA, et al. Clarification of risk factors for abdominal operations in patients with hepatic cirrhosis. Ann Surg 1984; 199:648. 27. Klemperer JD, Ko W, Connolly M, et al. Cardiac operations in patients with cirrhosis. Ann Thorac Surg 1998; 65:85. 28. Morris JJ, Hellman CL, Gawey BJ, et al. Three patients requiring both coronary artery bypass surgery and orthotopic liver transplantation. J Cardiothorac Vasc Anesth 1995; 9:311. 29. Pollard RJ, Sidi A, Gibby GL. Aortic stenosis with end stage liver disease: prioritizing surgical and anesthetic therapies. J Clin Anesth 1998; 10:253. 30. Brown MW, Burk RE. Development of intractable ascites following upper abdominal surgery in patients with cirrhosis. Am J Med 1986; 80:879. 31. Chapman CB, Snell AM, Rowntree LG. Decompensated portal cirrhosis. Report on one hundred and twelve cases. Clinical
1104
32. 33. 34. 35. 36. 37.
38.
39.
40.
41. 42.
43.
44.
45. 46. 47.
48.
49.
50.
51.
52.
53.
54. 55. 56.
features of the ascitic stage of cirrhosis of the liver. JAMA 1981; 97:237. Leonetti JP, Aranha GV, Wilkinson WA, et al. Umbilical herniorrhaphy cirrhotic patients. Arch Surg 1894; 119:441. Yonemoto RH, Davidson CS. Herniorrhaphy in cirrhosis of the liver with ascites. N Engl J Med 1956; 255:733. Maniatis AG, Hunt CM. Therapy for spontaneous umbilical hernia rupture. Am J Gastro 1995; 90:310. Hurst RD. Management of groin hernias in patients with ascites. Ann Surg 1992; 216:696. Bloch RS, Allaben RD, Wait AJ. Cholecystectomy in patients with cirrhosis. Arch Surg 1985; 120:669. Bornman PC, Terblanche I. Subtotal cholecystectomy: for the difficult gallbladder in portal hypertension and cholecystitis. Surgery 1985; 98:1. Yerdel MA, Tsuge H, MiMura H, et al. Laparoscopic cholecystectomy cirrhotic patients: expanding indications. Surg Laparosc Endosc 1993; 3:180. Yerdel MA, Koksoy C, Aras N, et al. Laparoscopic versus open cholecytectomy in cirrhotic patients: a prospective study. Surg Laparosc Endosc 1997; 7:483. Sleeman D, Namias N, Levi D, et al. Laparoscopic cholecystectomy in cirrhotic patients. J Am Coll Sum 1998; 187:400. Gopalswamy N. Risks of intra-abdominal nonshunt surgery in cirrhotics. Dig Dis 1998; 16:225. D’Alburquerque LA, de Miranda MP, Genzini T, et al. Laparoscopic cholecystectomy in cirrhotic patients. Surg Laparosc Endosc 1995; 5:272. Friel CM. Laparoscopic cholecystectomy in patients with hepatic cirrhosis: a five-year experience. J Gastrointest Surg 1999; 3:286. Mansour A, Watson W, Shayani V, et al. Abdominal operations in patients with cirrhosis: still a major surgical challenge. Surgery 1997; 122:730. Doberneck RC, Sterlin WA, Allison DC. Morbidity and mortality after operation in nonbleeding cirrhotics. Am J Surg 1983; 146:306. Aranha GV, Greenlee HB. Intra-abdominal surgery in patients with advanced cirrhosis. Arch Surg 1986; 121:275. Bernstein DE. Recombinant factor VIIa corrects prothrombin time in cirrhotic patients: a preliminary study. Gastroenterology 1997; 113:1930. Berstein DE. Effectiveness of the recombinant factor VIIa in patients with the coagulopathy of advanced Child’s B and C cirrhosis. Semin Thromb Hemost 2000; 26:437. Jeffers L. Safety and efficacy of recombinant factor VIIa in patients with liver disease undergoing laparoscopic liver biopsy. Gastroenterology 2002; 123:118. Slappendel R, Huvers FC, Benraad B, et al. Use of recombinant factor VIIa to reduce postoperative bleeding after total hip arthroplasty in a patient with cirrhosis and thrombocytopenia. Anesthesiology 2002; 96:1525. Wehbi MA, Obideen K, Martinez EJ, et al. Recombinant VVI (RFVIIA) as a safe and effective therapeutic option for the correction of moderate to severe hepatic coagulopathy. Hepatology 2003; 34:667A. Ziser A, Plevak DJ, Weisner RH, et al. Morbidity and mortality in cirrhotic patients undergoing anesthesia and surgery. Anesthesiology 1999; 90:42. Farrell GC. Liver disease due to anaesthetic agents. In Farrell GC, ed: Drug-induced liver disease. Edinburgh: Churchill Livingstone, 1994:389. Touloukian J, Koplowitz N. Halothane-induced hepatic disease. Semin Liver Dis 1981; 1:134. Cousins MJ, Plummer JL, Hau PM. Risk factors for halothane hepatitis. Aust NZ J Surg 1989; 59:5. Friedman LS. The risk of surgery in patients with liver disease. Hepatology 1999; 29:1617.
Chapter 58 PREOPERATIVE AND POSTOPERATIVE HEPATIC DYSFUNCTIONS
57. Strunin L. Anesthetic management of patients with liver disease. In: Millward-Sadler GH, Wright R, Arthur MJP, eds. Wright’s liver and biliary disease. London: Saunders; 1992:1381. 58. Cowan RE, Jackson BT, Grainger SL, et al. Effects of anesthetics and abdominal surgery on the liver blood flow. Hepatology 1991; 14:1161. 59. Gibson PR, Dudley FJ. Ischemic hepatitis: clinical features, diagnosis and prognosis. Aust NZ I Med 1984; 14:822. 60. Johnson RD, O’Connor ML, Kerr RM. Extreme serum elevations of aspartate aminotransferase. Am J Gastro 1995; 90:1244. 61. Brittain RS, Marchioro TL, Hermann G, et al. Accidental hepatic artery ligation in humans. Am J Surg 1964; 107:822. 62. Farrell GC. Paracetamol-induced hepatotoxicity. In: Farrell GC, ed: Drug-induced liver disease. Edinburgh: Churchill Livingstone;1994:205. 63. Kumar S, Rex DK. Failure of physicians to recognize acetaminophen hepatotoxicity in chronic alcoholics. Ann Intern Med 1991; 151:1189. 64. Gottlieb JE, Menashe PI, Cruz E. Gastrointestinal complications in critically ill patients: the intensivists’ overview. Am J Gastroenterol 1986; 81:227. 65. LaMont JT, Isselhacher KJ. Postoperative jaundice. N Engl J Med 1974; 288:305. 66. Schmid M, Hefti ML, Gattiker R, et al. Benign postoperative intrahepatic cholestasis. N Engl J Med 1965; 272:545.
67. Kantrowitz PA, Jones WA, Greenberger NJ, et al. Severe postoperative hyperbilirubinemia simulating obstructive jaundice. N Engl J Med 1967; 276:591. 68. Boekhorst T, Urlus M, Doesburg W, et al. Etiologic factors of jaundice severely ill patients. A retrospective study in patients admitted to intensive care unit with severe trauma or with septic intraabdominal complications following surgery and without evidence of bile duct obstruction. J Hepatol 1988; 7:111. 69. Waxman K. Postoperative multiple organ failure. Crit Care Clin 1987; 3:429. 70. Moossa AR, Easter DW, Van Sonnenberg E, et al. Laparoscopic injuries to the bile duct. A cause for concern. Ann Surg 1992; 215:203. 71. Davidoff AM, Pappas TN, Murran EA, et al. Mechanisms of major biliary injury during laparoscopic cholecystectomy. Ann Surg 1992; 215:196. 72. Thompson JS, Bragg LE, Hodgson PE, et al. Postoperative pancreatitis. Surg Gynecol Obstet 1988; 167:377. 73. Frazee RC, Nagorney UM, Mucha P. Acute acalculous cholecystitis. Mayo Clin Proc 1989; 64:163. 74. Becker CD, Burckhardt B, Terrier F. Ultrasound in postoperative acalculous cholecystitis. Gastrointest Radiol 1986; 11:47. 75. Baker AL, Rosenberg IH. Hepatic complications of total parenteral nutrition. Am J Med 1987; 82:489.
1105
Section X: Tumors of the Liver
59
HEPATOCELLULAR CARCINOMA Jordi Bruix, Concepció Bru and Josep M. Llovet Abbreviations AAIR age adjusted incidence rate AFP a1-fetoprotein AgNOR nucleolar organizer region protein ALT alanine aminotransferase BCLC barcelona-clínic-Liver-cancer CT computed tomography HBV hepatitis B virus
HCC HCV HIV MELD MRI NASH PCNA
hepatocellular carcinoma hepatitis C virus human immunodeficiency virus model for end-stage liver disease magnetic resonance non-alcoholic steatohepatitis proliferating cell nuclear antigen
INTRODUCTION Some years ago it was common to consider hepatocellular carcinoma (HCC) as an infrequent cancer in the western world, as it was seen as a neoplasm mostly restricted to sub-Saharan Africa and Asia. At the same time, it was assumed that the diagnosis would always be made when the tumor was already advanced and thus, treatment would be unfeasible and the prognosis very grim. These concepts induced a nihilistic approach to the clinical management of patients diagnosed with HCC. However, in recent decades the situation has changed dramatically. It has been acknowledged that the incidence of HCC has increased in the western world1–3 and that this neoplasm is now the leading cause of death in patients with cirrhosis.4,5 At the same time, it has been shown that diagnosis can be achieved at an early stage when effective therapy is feasible and long-term survival is not anecdotal.6 The present chapter will summarize the most relevant aspects regarding the epidemiology, pathogenesis, diagnosis, and treatment of this neoplasm.
EPIDEMIOLOGY Recent estimations indicate that primary liver cancer is now the fifth most common cancer in the world and the third cause of cancerrelated mortality.1,2 More than half a million cases are diagnosed every year on a worldwide basis, but there are relevant geographic differences in incidence (Table 59-1).3 Eastern Asia and sub-Saharan Africa account for most cases and the annual incidence rates there largely exceed 15/100 000 habitants. Incidence rates are intermediate (between 5 and 15/100 000) in Mediterranean countries and southern Europe, while the rate is low (below 5/100 000) in northern Europe and the USA.3 However, slight changes have been detected in some specific areas. Incidence rates have decreased in Hong Kong, Shanghai, and Singapore,3 while effective vaccination against the hepatitis B virus (HBV) has significantly reduced the incidence in Taiwan.7 The same effect can be expected in other areas where this virus is the main oncogenic agent for HCC development and if vaccination is implemented. In contrast, the incidence has increased in Australia, the UK,8 Canada,9 and the USA;10,11 this probably reflects the spread of hepatitis C virus (HCV) infection. This has occurred later than in countries such as Japan12 where the
PEI PHT PS RCT TNM US TAE
percutaneous ethanol injection portal hypertension performance status randomized controlled trial tumor node metastasis ultrasonography transarterial embolization
incidence may have reached a plateau,13 or even initiated a decrease, as described in some registries in Italy. In addition to the heterogeneous geographical incidence, there are some differences of race and ethnicity. It is likely that these racerelated differences reflect different exposure to risk factors and time of acquisition, rather than genetic predisposition. Accordingly, the difference in HCC incidence according to race may vanish in populations of mixed ethnicity but with homogeneous risk profile. In all areas and cancer registries males have a higher prevalence than females. The magnitude of the ratio varies greatly, with most countries showing values between 2:1 and 4:1. The ratio exceeds 4:1 in some particular regions of France, Italy, and Switzerland and is lower than 2:1 in South America.1,3 The male predominance may be due to specific genetic and hormonal profiles together with a higher prevalence of risk factors such as viral infection, alcoholism, and smoking. The age at which HCC appears also varies according to gender, geographic area, and risk factor associated with cancer development. In most areas female age is higher than male.1,3 In high-incidence areas where HBV is the main etiologic agent, the peak age appears after 40 years, while in low-incidence areas such as the USA, the peak of age appears beyond 75 years.1,3 As previously noted, the prevalence of risk factors has a marked geographic heterogeneity and this undoubtedly accounts for a major part of the described epidemiological pattern. HBV is widespread in Asia (with the exception of Japan) and sub-Saharan Africa. HBV is mostly acquired at birth or during early childhood and, in most areas, viral oncogenicity overlaps with that of aflatoxin B-1, a powerful oncogenic agent that contaminates food stored in humid conditions in several countries of Asia. The main risk factors in Japan, Europe, and Asia are HCV infection and alcohol intake. The spread of HCV has occurred at different decades in these areas and this explains why the peak incidence of HCC in Japan and southern Europe has been registered earlier than in the USA.
RISK FACTORS FOR HCC One of the relevant characteristics of HCC is that cirrhosis underlies HCC in almost 80% of affected individuals.4,14 Thus, any agent leading to chronic liver damage and ultimately cirrhosis should be
1109
Section X. Tumors of the Liver
Table 59-1. Incidence of Hepatocellular Carcinoma According to Geographical Area and Distribution of Risk Factors Geographic area
Europe Western Europe Southern Northern North America Northern Southern Asia and Africa (Japan) Eastern Asia South-Eastern Asia Middle Africa World
AAIR Male/female
Risk factors HCV
HBV
Alcohol
Other
60–70%
10–15%
20%
10%
50–60%
20%
20%
10%
20% 70%
70% 10–20%
10% 10%
Aflatoxin 10%
5.8/1.6 9.8/3.4 2.6/1.3 4.1/1.6 4.8/3.6
35.4/12.6 18.3/5.7 24.2/12.9 14.9/5.5
AAIR, age adjusted incidence rate; HCV, hepatitis C virus; HBV, hepatitis B virus. Reproduced from Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet 2003; 362:1907–1917, with permission.
seen as a risk factor for HCC. Due to their high prevalence rate the major causes of cirrhosis and HCC are HBV, HCV, and alcohol, but less prevalent conditions such as hemocromatosis, primary biliary cirrhosis, non-alcoholic steatohepatitis (NASH), and Wilson’s disease have also been associated with HCC development.
HEPATITIS B VIRUS This infective agent affects 300 million people worldwide and it constitutes the most important oncogenic factor for HCC development. The evidence linking HBV with HCC is unquestionable.15 Large cohort studies have shown that the incidence in HBV carriers is significantly increased. Seminal studies in Taiwan showed a 100-fold increase in HBV carriers,16 but several aspects should temper this value. Active viral replication implies a higher risk and longstanding active infection resulting in cirrhosis is the major event that gives rise to highly increased risk.17–19 The incidence of HCC in inactive HBV carriers without liver cirrhosis is less than 0.3%.20,21 This has been clearly shown in studies in Europe20–22 and the USA,23,24 but Asian studies in HBV carriers who acquired the infection early in life indicate that in this specific population the incidence of HCC is higher even in the absence of cirrhosis.18,19 The role of specific genotypes or mutations is not well established.22,25,26 In addition to this epidemiologic evidence, it has been shown that HBV can be integrated into the host cellular genome and induce genetic damage (instability, deletion, and rearrangements).15 In addition, some of the HBV proteins disrupt cellular functions27 and may favor neoplastic transformation, induce proliferation, and impede apoptosis.28 Interestingly, occult HBV infection may become apparent if properly investigated by molecular techniques even in the absence of serological markers of HBV itself.29,30 Accordingly, the relevance of HBV may be even higher than suggested by studies based on serologic assessment of HBV infection. Finally, the implementation of vaccination against HBV has resulted in a significant decrease of HCC incidence7 and this is the final proof of the importance of this virus in the genesis of this cancer.
1110
HEPATITIS C VIRUS The discovery of this virus has enabled the proper classification of most patients with so-called non-A, non-B hepatitis. Immediately after a serological test became available it was clear that chronic HCV infection was a major risk factor for HCC.31 The prevalence of HCV in HCC cohorts varies according to the penetration of the agent in the population of each geographic area.3 A study including 12 000 men described a 20-fold increased risk of HCC in infected individuals;32 this figure is very close to the estimated risk obtained in a meta-analysis of 21 case control studies.11 The risk is clearly related to the degree of liver damage induced by the virus.33,34 There are some case reports of so-called healthy HCV carriers with HCC,35 but several cohort studies indicate that the incidence in patients with chronic hepatitis is low (below 1%) and that the risk increases sharply when cirrhosis is established.33,34 At that time, the annual incidence ranges between 2 and 8%.4 The process from acute infection to cirrhosis may take 20–30 years.36 Hence, those studies recruiting patients with chronic hepatitis will detect HCC development if follow-up has been long enough, while those studies including patients with established cirrhosis will register the risk early during follow-up. Patients infected with the human immunodeficiency virus (HIV) are now effectively treated with combined regimes and if coinfected with the HCV they present a faster evolution to cirrhosis and thus, are at risk for HCC development. In fact, liver disease and/or HCC are the major causes of death in these patients.37
ALCOHOL The estimation of risk of HCC in alcoholics was surely overestimated until the availability of a test for HCV infection, as a noteworthy proportion of alcoholic cirrhotics are infected by HCV.38 The risk for HCC appears when alcohol intake exceeds 60 g/day and beyond this cut-off it increases linearly.39 Thereby, alcohol consumption above 80 g/day for more than 10 years is associated with a fivefold increased risk. The risk of HCC development is further increased on development of cirrhosis and the annual incidence
Chapter 59 HEPATOCELLULAR CARCINOMA
increases beyond 2% in decompensated cirrhosis.40 Coinfection with HCV or HBV exerts a synergistic affect and increases HCC risk.
AFLATOXIN This is a powerful oncogenic agent that induces a specific damage in the p53 gene (G to T transversions at codon 249).41 It is produced by fungi that contaminate food stored in humid conditions, as frequently happens in some undeveloped areas of the world. Epidemiological studies have established a correlation between the magnitude of aflatoxin intake (that can be recognized by the presence of aflatoxin-albumin adducts in sera or by the detection of aflatoxin metabolites in urine) and the incidence of HCC. However, as previously mentioned, the areas of heavy aflatoxin contamination overlap with areas of high prevalence of HBV infection. A synergistic effect induces a 60-fold increase in risk, while aflatoxin alone increases the risk by four.42
TOBACCO The role of smoking in promoting HCC has been difficult to ascertain due to the frequent confounding role of other risk factors. However, recent data suggest that smoking increases the risk for HCC.55–57 Contrarily, coffee consumption reduces the risk.58
HORMONAL COMPOUNDS Oral contraceptives have been linked to the development of adenoma and in some series of HCC. While the relationship with the first entity is well established, the causality with HCC is less robust.59 Viral infection, especially HCV, may not be ruled out in old studies and the oncogenic risk in long-term users may be due to the combination of estrogens and androgens.59 Androgenic treatment for hematologic conditions or athletic improvement also increases the risk of adenomas and has been associated with HCC development,60 but strong data are not available.
IRON AND COPPER DEPOSITION Patients with hereditary hemochromatosis have a highly increased risk of HCC upon cirrhosis development.14,43,44 If cirrhosis is not present the risk is less but there are cases of HCC in patients in whom treatment prevented the progression of the disease.45 Iron deposition is responsible for oxidative damage and cancer development and the same should happen in patients with Wilson’s disease. However, a high incidence in patients with Wilson’s disease has not been documented. Interestingly, an experimental model of copper accumulation results in cancer development46 that is effectively prevented by avoiding oxidative stress.47
NON-ALCOHOLIC STEATOHEPATITIS, OBESITY, AND DIABETES Several epidemiological studies conducted in Europe and the USA have linked obesity and diabetes to a higher HCC risk.48–50 The risk is higher in males50 and the current epidemic of overweight and obesity may prompt an increase of HCC incidence linked to this entity. Data on the prevalence and incidence of HCC in NASH patients are still limited and the available information suggests that some patients with HCC in cryptogenic cirrhosis may have indeed suffered NASH,51–53 but the magnitude of HCC risk in NASH patients who have reached cirrhosis is currently unknown.
AUTOIMMUNE HEPATITIS The risk in this entity is low. Some of the reported patients with HCC had coincidental viral infection and the annual incidence is 0.1%.
a1-ANTITRYPSIN DEFICIENCY This condition may be associated with cirrhosis and HCC in the PIZZ phenotype associated with intrahepatocellular protein deposition. Again, the confounding role of viral infection is relevant, as most patients with HCC present evidence of either HBV or HCV.
PRIMARY BILIARY CIRRHOSIS The risk of HCC appears at advanced stages when cirrhosis is established and affects mostly males.54
CIRRHOSIS As previously described, cirrhosis development through any damaging agent constitutes a major event regarding the magnitude of risk for the development of HCC.4,14 However, the risk varies according to etiology61,62 and, more importantly, according to the degree of liver function impairment. There is a steady increase in incidence from early cirrhosis without portal hypertension (PHT) to decompensated cirrhosis fitting into Child–Pugh C class.63–65 Proper evaluation of unscreened compensated cirrhotics may show a 5% prevalence of HCC and the percentage increases to 15–20% when considering patients with variceal bleeding or bacterial peritonitis.4 Older age appears in most studies as a relevant predictor but it is possibly a surrogate marker of the duration of the disease. Male sex is also a marker of increased risk, reflecting either the simultaneous oncogenic effect of coincidental agents (smoking, alcohol intake) or a higher oncogenicity related to androgens. The activity of the liver disease is another relevant factor in HCC promotion. Sustained HBV replication and higher HCV-related inflammation as reflected by increased alanine aminotransferase (ALT) levels have been shown to correlate with a higher HCC incidence. More intense disease activity can also be depicted by proliferating cell nuclear antigen (PCNA)66 or nucleolar organizer region protein (AgNOR)67 staining and recent data indicate that subjects with increased proliferation bear an increased HCC risk. Other histology findings have been linked to HCC risk. The value of large-cell dysplasia is controversial as there are studies with discrepant results.63,68 Japanese authors have described irregular regeneration of hepatocytes as a powerful predictor,69 but this awaits confirmation. The parameter that has been consistently linked to increased risk is abnormal a1-fetoprotein (AFP) concentration.4,70 It is known that chronic hepatic inflammation and regeneration may induce transient increases of this tumor marker. This is well described in both chronic HBV and HCV infection even in the absence of cirrhosis.71,72 It could be argued that increased AFP levels reflect the presence of an HCC that cannot be recognized by current imaging techniques, but this would not modify the clinical relevance of the finding.
1111
Section X. Tumors of the Liver
PATHOGENESIS The relationship between any agent inducing chronic liver injury and HCC development indicates that active inflammation with oxidative damage should be a key step in the origin of the genetic damage leading to cancer. This would be present both in viral infection and in the metabolic derangement that occurs in metal accumulation, alcohol consumption, or NASH. The molecular events responsible for the development and progression of HCC are not elucidated.73,74 Malignant hepatocytes are the result of cumulative genetic changes occurring in mature hepatocytes or in stem cells. The step-by-step process is initiated through generation of foci of dysplastic hepatocytes without overt malignant phenotype. While low-grade dysplastic nodules harbor a low malignant potential, the likelihood of evolving to HCC is significant in high-grade dysplastic foci. Thirty percent of these nodules turn into HCC after a follow-up of 5 years.75,76 Dysplastic nodules and very early HCC retain their hepa-
tocytic differentiation and do not show an extensive net of newly formed vessels.77,78 At this stage the nodules usually measure <2 cm in diameter, but they slowly increase in size and exhibit an active neoangiogenic process resulting in an enhanced blood supply derived from the hepatic artery.79,80 At this time point, the tumor has reached an unequivocal malignant profile, its differentiation degree is progressively lost and the tumor displays the invasiveness and dissemination that characterize all neoplasms.79,80 Active research is trying to define the specific genetic changes that determine all these sequential events in cancer progression. Loss of heterozygosity at 1p (1p36-p34) and in M6F/IGF2R is frequently observed at early stages together with telomere shortening.81 Altered expression of genes such as SP70, glypican-3, or survivin, or some sets of genes, have been proposed as early HCC markers. More advanced tumor stages may exhibit allelic alterations in all chromosomes. The most frequently affected chromosomes are 1, 4, 8, 16, and 17 (Table 59-2), but none of them is abnormal in more than 60% of the cases.73 Down-regulation of p53 is observed in up to 40%
Table 59-2. Chromosomal Abnormalities in Human Hepatocellular Carcinoma and Genes Involved Chromosome
Numerical aberrations
1
Loss of heterozygosity
Genes involved
del 1p22 del 1p32-p36
1p 1p35–36
L-myc
3 4
-
5
-
t5;9
6 7 8 9
+ -
del 6 (q13-q-ter) tl7;7 (p13-p14) inv 8 (q10) t5;9 inv 98p12:q12)
10 11
-; +
12 13
-;+
15 16
-; + -
17
-
18
dup 11p15 del 11 (p13-p14) del 11p11 t11;22
4p 4q, 4q32, 4q11–23 5p 5q, 5q35-q-ter 6q26-q27 8q
10q 11p 11p13-p15.1
b-catenin AFP/Alb/a-FGF
APC (5q21) M6P/IGF-IIR c-myc (8q24) p16 INK4A p19 ARF (9p21) PTEN (10q23) WT-1 IGF-II Cyclin D (11q13)
13q 13q12 13q12-q31
Rb-1 (13q-14)
del 16q
16p 16q22–24
17p, 17p13, 17p13-pter
-; +
t17;7 (p13;p14) t17;18 (q25;q11) t17;X del 17 (p12) t17;18 (q25;q11)
Axin-1 (16p13.3) E-cadherin (16q22.1) SOCS-1 (16p13.13) p53 (17p13)
20 21 22
+ -; + -
t22;?
22q 22q11–12
X
-
t17:X
AFP, a-fetoprotein; a-FGF, APC, IGF, PTEN, SOCS-1, SMAD2 Reproduced with permission from Cancer Cell.82
1112
Structural aberrations
SMAD2 SMAD4 (18q21)
Chapter 59 HEPATOCELLULAR CARCINOMA
of cases and G to T mutation reflects genetic damage due to aflatoxin intake.41 The pathway leading to HCC development may differ according to the underlying etiologic agent. HBV-related HCC exhibits a more intense genetic instability with multiple chromosome changes.83 By contrast, non-HBV-related HCC would be less unstable genetically, and more frequently present mutations of the b-catenin gene with activation of the wnt signaling pathway.84 This would allow the development of large non-disseminated tumors. Array studies have allowed the study of subsets of tumors grouped according to etiology and/or stage.85–87 This has led to the identification of a large number of genes with differential expression, but the biological meaning of the findings is not established. Identified genes are linked to cell proliferation, inflammation, collagen deposition, angiogenesis, and apoptosis. Some studies have established a relationship between expression of some genes with tumor progression (p16, p27, skp2, SOCS-1, PEG10) or dissemination and metastases (nm23-H1, osteopontin, ARHC[RhoC], KA-1, MMP14). However, all these data need extensive validation before they can become well-established prognostic tools in clinical practice.88
PATHOLOGY The pathology characteristics of HCC are modified along its transition from early to advanced stage.79,80 Malignant cells resemble normal hepatocytes but while in the initial stages of development the tumors retain their differentiation degree, at advanced stages the differentiation of the cells is progressively lost. Traditionally, the differentiation degree is established following the Edmonson and Steiner criteria. These define grades from I to IV according to the degree of nuclear irregularity, hyperchromatism, and the nuclear/cytoplasmic ratio. While it is difficult to distinguish grade I from benign hepatic just by cell pattern, grade IV is characterized by highly abnormal cells. Since early small HCC are usually well differentiated, extensive expertise is required to establish a confident diagnosis by analyzing biopsy material. Malignant cells have well-defined cell membranes and exhibit a finely granular eosinophilic cytoplasm. This gives the cell a clear appearance and on ultrasound prompts a hyperechoic pattern. Cells may accumulate bile and at early stages a variable amount of fat deposition is frequently recognized.89 Mallory bodies and a1-antitrypsin globules can also be recognized in a minority of cases. Immunostaining can recognize the presence of cytokeratins 7, 8, 18, and 19, carcinoembryonic antigen, a-fetoprotein, and several others.79,80 However, most are not specific and thus, cannot be used to diagnose a tumor such as HCC confidently. The most usual growth pattern of malignant hepatocytes is described as trabecular, in which cells accumulate in thin (microtrabecular) or thick (macrotrabecular) layers separated by sinusoids that may still contain Kupffer and stellate cells (Figure 59-1). Reticulin fibers are scanty and the endothelium vessels show a capillary pattern. The trabecula may form dilated canaliculi and depict a pseudoglandular or acinar pattern. A minority of HCCs develop an intense stroma and this results in a scirrhous pattern. Less than 5% of HCC exhibit a mixed pattern combining cholangiocarcinoma and hepatocarcinoma features. This probably reflects the embryonic
Figure 59-1. Microscopic view of hepatocellular carcinoma. It presents a moderate degree of differentiation. Cells are irregular in shape and exhibit abnormal nuclei with prominent nucleoli.
Figure 59-2. Solitary hepatocellular carcinoma within a cirrhotic liver. The nodule measures less than 3 cm in diameter, is well delimited by a thin capsule, and no satellites are detected in its vicinity. Note that the tumor has a heterogeneous content with areas of distinct appearance.
origin of biliary cells and hepatocytes and does not imply that a specific evolutionary pattern leads to a distinct outcome. According to their gross macroscopic appearance, the tumors can be described as expansive, infiltrative, and diffuse. Expansive tumors compress the surrounding liver and frequently exhibit a reticulin pseudocapsule that surrounds the tumor partially or completely.79,80 Infiltrative tumors have no distinct margins and invade the surrounding liver without formation of pseudocapsule. Finally, diffuse tumors appear as a multinodular entity that resembles cirrhotic nodules. The most common appearance is a distinct nodule of varying size (Figure 59-2) that increases in size as it evolves and is associated with the growth of satellite foci in the vicinity of the main tumor while simultaneously invading the segmental portal pedicle.
1113
Section X. Tumors of the Liver
Figure 59-3. Small hepatocellular carcinoma treated by percutaneous ablation. Note that the main tumor is necrotic, but there are several additional tumor foci and a tumor thrombi inside a portal venule. This feature indicates advanced tumor stage with risk of hematogenous spread.
Thereafter, the tumor spreads and several HCC nodules are detected in the liver. The prevalence of portal vein invasion (Figure 59-3) increases as the disease evolves and finally, all or most of the liver volume is replaced by the tumor. Extrahepatic dissemination is not frequent at early stages but at late stages it is common to detect lymph node involvement in the hepatic hilum and metastatic nests in adrenal glands, lung, and bones. Vascular invasion is one of the most characteristic findings in this neoplasm and its existence has significant prognostic consequences. It reflects an advanced malignant phenotype and hematogenous spread (Figure 59-3) that may lead to recurrence after initially effective treatment. The prevalence of vascular invasion increases in parallel with the loss of differentiation and with tumor size.79 It can be recognized in 20% of HCC < 2 cm in diameter, while the percentage increases to 30–60% in HCC between 2 and 5 cm and to 60–90% in larger nodules. The risk of vascular invasion in small tumors is relevant and Japanese authors have distinguished two subtypes according to growth pattern, even if both are smaller than 2 cm and very similar at ultrasonography.90 The indistinct type has ill-defined margins (Figure 59-4) and has not yet evolved into a real mass; it does not show satellites and the rate of vascular invasion is 10%. The distinct nodular type is clearly identified as a small mass and careful analysis reveals satellites in 10% and vascular invasion in up to 25% of cases. Distinct nodules show arterial vascularization and thus represent a more advanced tumor stage, despite their small size. In contrast, the indistinct type would correspond to carcinoma in situ;91,92 currently this can only be confidently diagnosed after surgical ablation.
FIBROLAMELLAR CARCINOMA This is a specific variant of HCC that has a unique pattern. In contrast with regular HCC, it appears in a normal liver and it is more frequent in young women. It has no relationship with contraceptive usage and is not related to viral infection. It usually appears as a large non-encapsulated mass of firm consistency. It may contain large areas of fibrous septa and cells may form thick trabecula. Malignant hepatocytes are large and polyhedral or round in shape. Their cyto-
1114
Figure 59-4. Pathology slice of an explanted cirrhotic liver. It shows a small hepatocellular carcinoma (HCC) measuring 12 mm. It has ill-defined margins and is located in the vicinity of a major vessel. Some of these very small HCCs correspond to carcinoma in situ as they have not reached fully malignant phenotype. Vascular invasion is absent and there is no evidence of spread.
plasm is eosinophilic and shows marked granularity due to a larger number of mitochondria.80 The growth rate of fibrolamellar HCC is reported to be less than that of common HCC and it is claimed to disseminate less and always at a more advanced stage. This more favorable cancer biology may confer a less grim prognosis, but this is still statistically unproven.
CLINICAL MANIFESTATIONS The fact that most HCCs appear in the setting of cirrhosis defines the most common clinical symptoms of the patients at the time of diagnosis, as reported in large cohort studies. A large part of the findings will be undistinguishable from the clinical findings observed in patients with cirrhosis without HCC. A variable proportion of cases will present with jaundice, ascites, encephalopathy, or bleeding due to ruptured esophageal varices. Cancer-related symptoms such as abdominal pain or constitutional syndrome (weight loss, anorexia, and malaise) are a reflection of advanced tumor stage. A minor proportion of patients will be diagnosed because of acute hemoperitoneum due to ruptured HCC or because of bone metastases. The proportion of symptomatic patients depends on the type of population recruited. Cohorts of cirrhotics entered into surveillance will collect patients with early disease without cancer-related symptoms, and the main findings will be related to degree of failure of the underlying cirrhosis.65,93 By contrast, patients diagnosed in the community who have not previously been aware of liver disease will report cancer symptoms more frequently as these will be the trigger for medical evaluation and diagnosis. This is usually the case in young patients with normal liver in whom the diagnosis is made homogeneously at advanced stages once symptoms present. In some areas, such as South Africa, tumor growth is reported to be faster and, despite regular screening, the stage at diagnosis is always advanced and patients are highly symptomatic with large hepatic masses inducing pain associated with physical deterioration and death within a short time. The biochemical pattern at diagnosis is also dominated by the underlying liver disease. Large tumors may induce cholestasis with
Chapter 59 HEPATOCELLULAR CARCINOMA
increased alkaline phosphatase and g-glutamyltranspeptidase. Old series have reported a change in the ALT/AST aminotransferase ratio, but the clinical value of the finding is minimal. A minority of patients present with hypoglycemia or hypercholesterolemia. The latter affects up to 30% of black Africans with HCC and is the result of excessive cholesterol production by tumor cells lacking the proper regulatory mechanism. Several paraneoplastic manifestations have been described in patients with HCC. The most common are diarrhea and severe hypoglycemia. Diarrhea is the result of tumor release of vasoactive substances and is usually intermittent. However, in some cases it can be intense and may become the major symptom leading to diagnosis. Other manifestations include hypercalcemia, sexual changes, polymyositis, thrombophlebitis, and skin rashes.
DIAGNOSIS AND STAGING Years ago, most HCCs were diagnosed at an advanced stage when cancer-related symptoms (pain, constitutional syndrome, weight loss, malaise, and anorexia) were already present and a large mass could easily be recognized by physical examination. Diagnosis confirmation was obtained by the analysis of tissue samples obtained by biopsy of the mass without any need for image guidance or by the presence of increased AFP. Hepatic scintigraphy was the first imaging tool able to suspect a hepatic mass still not recognized by clinical symptoms or laboratory abnormalities, but the major advance for an easy and early diagnosis has been the development of ultrasonography (US), computed tomography (CT), and magnetic resonance (MRI). These imaging techniques have been instrumental in detecting HCC at an early subclinical stage during the
evaluation of patients with chronic liver disease and also in engaging surveillance plans based on regular US in the population at risk, namely patients with liver cirrhosis. No randomized controlled trial (RCT) has been performed to establish the benefits of regular screening, but expert opinion supports the incorporation of cirrhotic patients who would be treated if diagnosed with HCC in structured screening programs.4,94 These aim to diminish tumor-related mortality by detecting HCC at an early stage when application of effective treatment would be feasible.4,94 According to this end-point it is suggested to limit screening to Child–Pugh A and B patients. Child–Pugh C patients should be evaluated for liver transplantation. If this is not available, surveillance will not be cost-effective as early detection and treatment will never translate into improved survival. A panel of experts set up by the European Association for the Study of the Liver recommended screening based on abdominal ultrasound and serum AFP every 6 months.4 The interval was derived according to the available data on tumor volume doubling time and Figure 59-5 depicts the strategy proposed to confirm the diagnosis once suspicion had been raised. If the nodule is < 1 cm, close follow-up is recommended since reliable diagnosis is not feasible with current diagnostic techniques. In nodules between 1 and 2 cm, HCC diagnosis requires positive cytology or histology. However, biopsy will be negative for malignancy in up to 30–40% of cases, even if the nodules are indeed malignant. Accordingly, a negative biopsy report does not confidently exclude malignancy. Several studies have shown that the recognition of dysplasia within the nodules detected by US is a powerful predictor of malignant transformation during follow-up75,76 and the same applies if the nodules are shown to present arterial hypervascularization.95–97 Probably, development of an extensive net of newly formed vessels
Cirrhotic patients* US+AFP/6m
Liver nodule
≥ 1cm
< 2cm
> 2cm
FNAB
AFP ≥ 400 ng/mL CT-MRI-Angiography
No nodule
< 1cm
Increased AFP**
US/3m
Spiral CT
Figure 59-5. Screening strategy for early hepatocellular carcinoma detection in cirrhotic patients. (Reproduced from Bruix J, Sherman M, Llovet JM, et al. Clinical management of hepatocellular carcinoma: conclusions of the Barcelona-2000 EASL conference. J Hepatol 2001; 35:421–430, with permission.)
Normal AFP
No HCC
HCC***
Surveillance US+AFP/6m
* Available for curative treatments if diagnosed with HCC ** AFP levels to be defined *** Pathological confirmation or non-invasive criteria
1115
Section X. Tumors of the Liver
already indicates transition into overt malignancy that will be properly diagnosed if adequately evaluated. If the nodule is > 2 cm and the underlying liver is cirrhotic, the diagnosis of HCC can be established by non-invasive criteria: two coincident imaging techniques showing a focal lesion > 2 cm with characteristic arterial hypervascularization and washout of contrast during the venous phase (Figure 59-6), or one imaging technique with specific patterns associated with AFP > 400 ng/ml. With state of the art equipment and assessment by expert radiologists the non-invasive diagnostic criteria could be further refined. Thereby, two coincidental imaging techniques showing intense arterial uptake of contrast following by contrast washout in the venous/delayed phase would be able to set confident HCC diagnosis in nodules between 1 and 2 cm within a cirrhotic liver.98 One single technique with a specific vascular profile would suffice for nodules larger than 2 cm within cirrhosis. Finally, nodules smaller than 1 cm in diameter would still be almost impossible to be confidently diagnosed by biopsy or imaging techniques.98 The policy to follow on detection of an increased concentration of AFP is less conclusive. A minor increase in AFP may be seen in patients with chronic liver disease as a result of inflammation flares and thus, in some cases it will not be due to development of cancer.71,72 However, most investigations have shown that persistent elevation of AFP during follow-up is a strong marker of increased risk of HCC development.4 It is important to note that increased risk does not mean faster tumor progression and thus, in high-risk patients there is no basis for carrying out screening at shorter time periods. Furthermore, it is well known that up to 40–60% of patients with HCC present normal or minimally increased AFP concentrations that do not reach the diagnostic cut-off.71,72 This limitation in sensitivity and the higher accuracy of imaging techniques and guided biopsy for the diagnosis of HCC have reduced the clinical efficacy of this tumor marker that will be useful in only a minority of cases. In addition to the low diagnostic sensitivity, AFP determination also has problems in specificity. AFP increases may be observed during
inflammation flares in patients with chronic HBV and HCV infection. Accordingly, AFP should not be used as a tool for screening and diagnosis since imaging techniques have better sensitivity and specificity.70 Hence, even in the presence of increased AFP concentration, the detection of an atypical radiological pattern should raise doubts about HCC diagnosis and biopsy should be indicated.98 The use of AFP fractions such as lectin-bound AFP, desgammacarboxyprothrombin,99 or glypican100 has been proposed to surpass the efficacy of AFP but their use has not fully entered clinical practice. Proteomic techniques may help to identify new markers.101 After HCC detection and diagnosis, the key issue is to stage tumor extent properly. This should be based on state-of-the-art CT or MRI.6,96,102 Both CT and MRI are equally effective for the detection of tumor sites > 2 cm.102 However, MRI angiography is more effective for nodules below this cut-off, but there are still problems with the characterization of minute nodules < 1 cm that do not exhibit the specific profile in terms of mass recognition and vascular enhancement. Angiography has almost no role in diagnosis and staging, while lipiodol CT is not reliable.4 This oily contrast is mostly retained within HCC foci after its injection in the hepatic artery, while it is cleared from non-tumoral liver. However, it has been shown that it has both false-positive and false-negative results and hence, lacks adequate clinical usefulness. Critical aspects of HCC staging are the detection of additional tumor sites and of vascular invasion. Both events reflect advanced tumor stage and higher likelihood of dissemination prior to therapy leading to recurrence after initially effective therapies such as resection, transplantation, or percutaneous ablation. Extrahepatic spread is infrequent at early stages but should be ruled out by chest CT. Bone metastases are usually symptomatic and should be ruled out if needed by bone scintigraphy. Brain metastases are exceptional. In addition to the evaluation of tumor burden according to size, number of nodules, vascular invasion, and dissemination, it is relevant to evaluate patients in terms of liver function, general health, and the presence of associated morbidity. The degree of liver function impairment because of tumor stage and underlying cirrhosis may reduce life expectancy and at the same time limit the application of therapy. This is not the case for liver transplantation, but it is very relevant to establish the feasibility of surgical resection and of invasive procedures such as transarterial chemoembolization. Finally, assessment of cancer-related symptoms provides a rough estimation of cancer stage and, if present, may indicate poor shortterm survival.
PROGNOSTIC PREDICTION
Figure 59-6. Magnetic resonance imaging scan showing a hepatocellular carcinoma located in the right lobe. Note the additional tumor site located near to the surface of the liver. Both nodules show arterial contrast uptake and washout in the venous phase.
1116
Outcome prediction is relevant to offer adequate information to patients and relatives at the time of diagnosis and when treatment has been initiated. In most neoplasms the prognosis is defined by tumor stage at the time of diagnosis. However, in patients with HCC this is more complex. Cirrhosis underlies HCC in most patients and thus prognosis depends on the evolutionary stage in which the neoplasm is diagnosed, the degree of liver function impairment of the underlying cirrhosis, and the treatment received.6 Simultaneously, the impairment of liver function determines the feasibility of treatment as this may further impair liver function and even induce death.
Chapter 59 HEPATOCELLULAR CARCINOMA
According to these comments, any proposal aimed at stratifying patients into different prognostic groups or at linking staging with treatment indication has to consider this complex interaction. Systems that only consider one dimension will be unable to predict outcome accurately. This is indeed the case for the tumor node metastasis (TNM) classification,103 that just takes into account tumor burden and disregards liver function. The same applies to the Child–Pugh104 or the model for end-stage liver disease (MELD)105 systems that disregard tumor extent, or to scores that merely evaluate general health status and physical capacity such as the performance status (PS)106 or the Karnofsky index. The sole usefulness of all unidimensional systems is to identify patients with very advanced disease stage and reduced life expectancy. The classification proposed by Okuda et al.107 has been used for many years but has currently been replaced by more accurate proposals. Several multidimensional systems have been proposed in recent years in France,108 the USA,109 Spain,110,111 Italy,112,113 Austria,114 Germany,115 Hong Kong,116 and Japan,117,118 to grade patients according to life expectancy (Table 59-3). All consist of a combination of tumor parameters and liver function variables, and usually provide a stratification of patients into separate groups. However, only the Italian119,120 and the Barcelona-Clínic-LiverCancer (BCLC)121–123 proposals have been validated. In addition, the BCLC system is the only one that links staging with treatment indication (Figure 59-7). This system was developed some years ago as a result of several cohort investigations and RCTs assessing the relevant outcome predictors within different tumor stages and treatment options. It conducts a first division of the patients into the relevant evolutionary stages at which patients may be diagnosed.6 Stage 0 comprises small tumors usually < 2 cm that have not reached an invasive phenotype with increased vascularization and appearance of microscopic vascular invasion and satellites. Currently, confident diagnosis of these very early HCCs corresponding to so-called carcinoma in situ is not feasible prior to resection, but in the next few
years, these tumors should be a common result in well-conducted surveillance programs. Stage A comprises tumors diagnosed at an apparently early stage when curative treatment (resection, transplant, ablation) is feasible. Patients have a preserved liver function (Child–Pugh A or B) and present with solitary tumors or up to three nodules, each < 3 cm in size. As noted, these patients are candidates for surgery or percutaneous ablation. Survival at 5 years may range between 50 and 75%. Patients with large or multifocal disease that are asymptomatic belong to an intermediate stage (stage B). These patients are candidates for transarterial chemoembolization and, depending on baseline characteristics and treatment success, will achieve a 3-year survival around 50%. Patients who report cancerrelated symptoms (pain, constitutional syndrome) or present with vascular invasion or extrahepatic spread correspond to an advanced stage (stage C). There is no standard treatment for them and their survival at 3 years is less than 10%. Finally, patients with severe impairment of liver function (Child–Pugh C) or major physical deterioration (PS = 3) or corresponding to class 3 of the Okuda staging system correspond to stage D. Their median survival is less than 6 months. The stratification of patients into separate stages according to tumor stage, liver function, and presence of symptoms, with the development of specific prognostic tools for each, is practical from a clinical point of view and also when assessing the impact of treatment. While at intermediate and advanced stages the natural history of untreated patients is better known, as there are several studies defining this issue, this is not the case for patients diagnosed at an early stage. Some groups have reported small series of untreated patients with early-stage disease. However, in normal conditions these patients are treated with any of the available options and, if this is not the case, the same reason prompting no treatment (late diagnostic confirmation, delay in staging, refusal of treatment, associated conditions preventing therapy) may impair survival and make any analysis flawed. Hence, any attempt to define prognosis in this
Table 59-3. Prognostic and Staging Systems to Predict Outcome in Hepatocellular Carcinoma Patients Author (year)
108
n
Stuart (1996)
314
CLIP (1998)111
435
Chevret (1999)107
761
Llovet (1999)109
102
BCLC (1999)110
—
Villa (2000)112 CUPI (2002)115
96 926
JIS (2003)116 SliDe (2004)117
722 177
Prognostic variables Tumor stage
Liver function
Health status
Portal vein invasion AFP Tumor morphology AFP, portal vein invasion Portal vein invasion AFP
Albumin
—
Child–Pugh
—
Bilirubin Alkaline phosphatase —
Karnofsky Performance status
Child–Pugh
Performance status
Bilirubin Bilirubin, ascites alkaline phosphatase Child–Pugh Liver damage by LCSGJ PIVKA
— Symptoms
Portal vein invasion Metastases Tumor size, number Vascular invasion Extraheptic spread Estrogen receptor status TNM AFP TNM by LCSGJ TNM by LCSGJ
— —
AFP, a-fetoprotein; TNM, tumor node metastasis; LCSGJ, Liver Cancer Study Group of Japan; PIVKA, protein induced by vitamin K absence.
1117
Section X. Tumors of the Liver
HCC
Stage A–C Okuda 1-2, PST 0-2, Child–Pugh A–B
Stage 0 PST 0, Child–Pugh A
Very early stage (0) Single < 2 cm Carcinoma in situ
Early stage (A) Single or 3 nodules < 3 cm, PS 0
Stage D Okuda, PST >2, Child–Pugh C
Intermediate stage (B) Multinodular, PS 0
Advanced stage (C) Portal invasion, N1, M1, PS 1-2
Terminal stage (D)
Single Portal invasion, N1, M1 Portal pressure/bilirubin
Increased
Normal
Resection
3 nodules < 3 cm
Associated diseases
No
Liver transplantation (CLT/LDLT) Curative treatments 50%–75% at 5 years
No
Yes
Yes
PEI/RF
Chemoembolization
New Agents
Randomized controlled trials 40%–50% at 3 years vs 10% at 3 years
Symptomatic treatment
Figure 59-7. BCLC staging and treatment strategy. Patients are divided into separate stages according to tumor burden, liver function, and physical condition. This stratification allows a rough estimation of life expectancy and guide treatment indication. (Reproduced from reference 110 with permission).
stage will have to incorporate the treatment options to be applied and use the relevant predictors within them. As previously mentioned, the natural history of the so-called nonsurgical HCC is widely studied. Optimally, the analysis of these patients should exclude those cases diagnosed at an end-stage. Otherwise, the prognostic predictors will merely identify the markers of end-stage disease. Years ago we took advantage of the two prospective RCTs comparing transarterial embolization (TAE)124 and tamoxifen versus no treatment and joined the two control groups of these investigations into a cohort of 102 patients who presented with large multifocal HCC, but who did not report a heavily impaired physical status (PS > 2) or advanced liver disease (Child–Pugh C).110 Their 1-, 2-, and 3-year survival was 54, 40, and 28%, respectively and we identified the presence of cancer-related symptoms (PS 1 or 2, constitutional syndrome) and detection of vascular invasion or extrahepatic spread as independent predictors of survival. Accordingly, asymptomatic patients without vascular invasion or extrahepatic spread showed a 1-, 2-, and 3-year survival of 80, 65, and 50%, respectively, while these figures decreased to 29, 16, and 8% in those with at least one adverse characteristic (Figure 59-8).109,110 As exposed, the current prognostic estimation in patients with HCC is derived from rough assessment of tumor stage combined
1118
with liver function evaluation and registration of cancer-related symptoms. However, it is expected that current effort in translational research will facilitate basing the prediction on the identification of markers associated with the activation of the mechanisms that govern cancer progression and dissemination.
TREATMENT There are several options to be considered in patients with HCC, but only a few can achieve long-term cure. These include surgical resection, liver transplantation, and percutaneous ablation.6 Unfortunately, the success of these approaches is restricted to patients diagnosed at an early stage and currently, this involves fewer than 40% of patients evaluated in referral units. Most patients are diagnosed at a more advanced stage and the only option that has been shown to have a positive impact on survival is transarterial chemoembolization.125 However, its applicability is reduced to <20% of cases. Accordingly, almost half of patients will have no option for effective therapy and thus, they might be considered for entering research investigations to evaluate new therapeutic strategies or will receive best supportive care. Treatment indication requires a careful evaluation of the status of the patient in terms of tumor stage, degree of liver failure, and
Chapter 59 HEPATOCELLULAR CARCINOMA Figure 59-8. Survival of untreated patients with non-surgical hepatocellular carcinoma. Patients can be divided into two separate subgroups according to the presence of symptoms or an invasive pattern, as shown by vascular invasion or extrahepatic spread.
Natural history of non-surgical HCC Survival according to the prognostic factors 100 80%
80
p<0.00001
Probability (%)
65% 60 50% 40
29% 16%
20
8%
0 Patients at risk
0
12
24
36
48
(BCLC Stage B) 48
37
30
16
7
(BCLC Stage C) 54
17
9
2
–
general health. Figure 59-7 shows the staging and treatment algorithm that is followed in our group.6 As previously mentioned, the evaluation of tumor burden should define the size of the tumor, the presence of daughter nodules, and the existence of extrahepatic disease and/or malignant vein invasion, specially of the portal vein. This can be accurately established using a combination of US and CT scan or MRI. In general, the detection of vascular invasion or extrahepatic spread implies advanced disease stage and precludes any effective therapy. The degree of liver function impairment is the second critical aspect. Patients with non-cirrhotic disease have a normal liver function and thus are good candidates for liver resection or other treatments, with potential deleterious effects on liver function. However, the majority of patients present with underlying cirrhosis and careful evaluation of hepatic function is mandatory. Decompensated cirrhosis prevents almost all forms of therapy except liver transplantation. By contrast, in patients with preserved liver function, surgical resection is still a valuable option, but should be limited to individuals with normal bilirubin concentration and portal pressure below 10 mmHg as measured by hepatic vein catheterization.126 If these conditions are not met, the morbidity and mortality associated with surgery increase significantly126 and patients will be better served by percutaneous ablation or transarterial chemoembolization. The evaluation of the general condition of the patient is part of the clinical assessment in any entity and should precede any therapeutic decision. Concomitant morbidities increase the risk of treatment-related complications and death. More specific is the assessment of cancer-related symptoms, as estimated by their PS score. This quantifies physical status in cancer patients and has similar value to the more detailed Karnofsky index. The intense impairment of these scores (PS = 3–4 or Karnofsky index < 60) identifies patients with advanced or terminal disease that will not benefit from any anticancer therapy and should just receive symptomatic care to avoid unnecessary suffering.
60 Months
SURGICAL TREATMENT For years surgical resection was the sole treatment option with accepted efficacy. Patients who could not be resected because of advanced tumor stage or impaired liver function were grouped together into a broad category named non-surgical HCC that clearly included a heterogeneous population. The development of liver transplantation and the availability of techniques to achieve tumor ablation through percutaneous approach or laparoscopy have fueled the debate about the current role of liver resection and whether it should still be viewed as the first therapy option to be considered or whether this prominent position should be given to liver transplantation. The absence of RCTs comparing resection, transplantation, and ablation prevents any robust recommendation and the therapeutic algorithm to follow should be the review of cohort studies and the analysis of the resources available in each setting.
Surgical Resection There is general consensus in considering resection as the best option for HCC in a normal liver. These patients represent fewer than 5% of HCC patients in western countries, while the proportion may reach 40% in some Asian countries where HBV acquired during childhood is the main etiologic agent. Accordingly, the therapeutic controversy affects just those patients with underlying chronic liver disease if this is still compensated. If liver decompensation is present, there is no doubt that liver transplantation is the option offering better long-term outcome. To solve the controversy in patients with compensated liver disease, it is mandatory to review critically all the data available as the outcome clearly depends on the selection policy applied to consider patients as surgical candidates. Twenty years ago, when resection was the sole choice, the postoperative mortality could exceed 20% and the 3-year survival was usually less than 50%. More recently, the earlier diagnosis in asymptomatic phases, the more accurate tumor staging, the refined functional evaluation of the underlying liver function, and the
1119
Section X. Tumors of the Liver
Table 59-4. Best Reported Outcomes after Surgical Resection and Liver Transplantation Treatment
Surgical resection Llovet et al. (1999)109 global No PHT, normal bile Takayama et al. (2000)129 Arii et al. (2000)128 Stage 1, <2 cm 2–5 cm Stage 2, <2 cm 2–5 cm Wayne et al. (2002)131 Poon et al. (2002)130 Child–Pugh A £ 5 cm/3 £ 3 cm Yamamoto et al. (2004)132 Stage 0 £ 2 cm Stage A £ 2 cm Liver transplantation Mazzaferro et al. (1996)141 Bismuth et al. (1999)140 Jonas et al. (2001)142 Yao et al. (2001)143 Plessier et al. (2004)144
n
Survival (%) 1 year
5 years
77 35 74
85 91 100
51 74 62
1318 2722 502 1548 249
96 95 92 95 83
72 58 55 58 41
135
90
70
37 149
— —
85 60
48 45 120 64
84 82 90 87
741 74 71 73
PHT, portal hypertension.
improvement in surgical techniques have allowed a sharp increase in potential long-term outcome. Major hepatectomies are contraindicated in cirrhotic patients and, with this limit, it is the current standard to have a perioperative mortality of < 3% (some groups have reported 0% mortality127,128), a transfusion rate of < 10%, and a 5-year survival > 50% (Table 59-4). This may indeed exceed 70% at 5 years in patients with preserved liver function and small solitary tumors.126,129–133 While it is usual to restrict resection to patients with a single tumor, the limits in size are not well established. The likelihood of vascular invasion and dissemination increases with size,79,134 but, in some instances, tumors may grow slowly and reach a large size without evidence of invasion or satellite disease. In these cases the risk of recurrence postresection is not significantly increased compared with smaller tumors. Resection has to include intraoperative US to guide anatomical resection (segmentectomy or subsegmentectomy) and to detect additional nodules not discovered during preoperative staging. The most common system to evaluate the hepatic functional reserve is the Child–Pugh classification104 and usually patients in classes B and C are excluded. However, this is far from optimal, as even Child–Pugh A patients may perform poorly after resection and develop liver decompensation and early death. The Japanese groups rely on the indocyanine retention test of hepatic function, using the rate of hepatic clearance of the dye to guide their decisions regarding the appropriateness and extent of liver resection.127 Portal pressure and bilirubin have been established as the best parameters to select optimal surgical candidates in Europe. Clinically relevant PHT can be defined as the presence of either a hepatic vein pressure gradient of > 10 mmHg, esophageal varices, or splenomegaly with a platelet count of <100 000/mm3. Subjects without relevant PHT
1120
and normal bilirubin achieve 5-year survival rates of 70%, whereas this decreases to 50% in patients with PHT, and to 25% in those with PHT and a raised bilirubin.125 The major drawback of surgical resection is the high recurrence rate during follow-up. This may exceed 50% at 3 years and its appearance reduces long-term survival.126,131,135 Most of the recurrent sites will appear early during follow-up (within 2 years) and be secondary to tumor spread prior to resection. By contrast, late recurrence may be due to the emergence of metachronic HCC originated in a separate cellular clone.126,131,135 The distinction between disseminated sites or new tumor clones can be estimated by genetic analysis through genomic hybridization, DNA fingerprinting, pattern of HBV integration sites, or microarray technology. The most powerful predictors of postoperative recurrence due to dissemination are the presence of microvascular invasion, poor differentiation, and satellite lesions.126,131,135 Predictors of de novo tumors should be related to parameters reflecting ongoing oncogenic events such as increased inflammation activity reflected by abnormal transaminase levels.135 Currently there is no effective method to diminish recurrence rate.136 Preoperative chemoembolization or adjuvant chemotherapy has no efficacy and may even complicate the intervention. Preliminary studies indicating benefit from internal radiation136 and adoptive immunotherapy (using activated lymphocytes)130 require validation, as do the apparent successes of retinoids and interferon in preventing de novo tumors. Interferon has also been tested after percutaneous ablation138 but robust evidence showing a positive impact is still lacking. Treatment of recurrence offers disappointing results. In the majority of cases, HCC recurrence will be due to disseminated disease and appear as multifocal disease.139 Accordingly, patients will not become candidates for resection or salvage liver transplantation. Since liver transplantation in patients at a stage similar to that of resected patients presents a lower risk of recurrence, some groups have proposed offering enlistment for liver transplantation to patients in whom analysis of the resected HCC reveals a high likelihood of dissemination.140 Initial encouraging results of transplantation because of risk of recurrence await confirmation. If recurrence appears late during follow-up, the treatment strategy should be the same as for the initial HCC. Late recurrence is more frequently solitary and corresponds to early-stage de novo malignancy in the cirrhotic liver.
Liver Transplantation Patients with HCC constituted a frequent target to assess the potential of liver transplantation during the pioneering years.141 The lack of therapeutic alternatives and their preserved physical conditions allowed their recruitment for a highly invasive and risky experimental procedure. The experience accumulated was instrumental in disclosing that the procedure was feasible and at the same time it became evident that advanced tumor stage, as reflected by large tumor involvement, vascular invasion, or lymph node dissemination, was always associated with disease recurrence and death within a short time. Interestingly, those tumors detected incidentally after the explant of an end-stage cirrhotic liver were not associated with recurrence. Improvement in imaging techniques allowed the confident recognition of these early tumors and this prompted Bismuth
Chapter 59 HEPATOCELLULAR CARCINOMA
et al.141 and Mazzaferro et al.142 to propose the indication of liver transplantation to patients with solitary HCC < 5 cm or with up to three nodules, each measuring < 3 cm. These criteria have been widely known as the Milan criteria and currently constitute an established definition for selecting the optimal candidates for liver transplantation. Several groups have confirmed that the survival of patients selected according to this definition may exceed 70% at 5 years (Table 59-4).143–145 Thus, this is comparable to the outcome of the best candidates for surgical resection, but with a lower disease recurrence that is around 15%. Tumor recurrence affects preferentially liver, lymph nodes, lung, and bones. Its incidence is higher if the pathology analysis of the explant discloses vascular invasion (macro- or microscopic) or additional tumor nests.143,145 These adverse parameters are more frequent in tumors exceeding 5 cm and if they show areas of poor differentiation.79,134 These data argue against a potential expansion of the enlistment criteria as proposed by some groups.144,146 Therefore, restrictive selection provides optimal results, but perhaps excludes a number of patients who, despite having a slightly more advanced tumor stage, would still achieve an encouraging outcome exceeding 50% at 5 years. Poor differentiation in large tumors surely reflects a higher risk of dissemination, but assessment of this parameter prior to therapy147 lacks the required strength in decision-making. Large solitary tumors contain areas with different degrees of differentiation and thus, assessment of risk would be flawed. Due to the limited number of livers to be transplanted, the decision to expand enlistment criteria will probably have to wait until there are strong data defining the new limits. Probably, these data may be obtained within live donation programs. This technical advance has eliminated the requirement to get an optimal outcome with the limited number of livers and several groups have launched live donation programs with expanded criteria:148–150 long-term results are eagerly awaited. A large multicenter survey in Japan with limited follow-up indicates that the survival data in live donation are the same as in cadaveric programs.151 Therefore, patients who meet the Milan criteria achieve an 80% survival rate at 3 years, while this is reduced to 60% and associated with higher HCC recurrence if the tumor stage slightly exceeds this limit.151 These results represent a major reinforcement of the usefulness of live donation in patients with HCC, but there are several issues that should temper the enthusiasm. The applicability of live donation is still limited because of the absence of a valid donor or because of the refusal of patients to put a relative at risk. The risk of death for donors is around 0.5% and, in addition, the graft may suffer several complications related either to the surgical technique (20% morbidity) or viral reinfection.152 In addition, some studies have suggested that the course of HCV reinfection after live donation is more aggressive than that after cadaveric transplantation and this induces faster progression to cirrhosis and liver failure.153 This issue is currently very controversial156 and only a follow-up of large cohorts will establish whether this concern has to be taken into account. As just mentioned, the main problem for liver transplantation is the limited availability of organs, even if using marginal livers or developing split liver155 or domino transplantation.156 Active campaigns are needed to increase the donation rate within the community and there are huge differences between countries. Even in those with higher donation rates, such as Spain, the number of candidates
exceeds that of donors and this implies a steadily growing number of patients waiting for transplantation and the longer time that patients have to wait to obtain a liver. At the beginning of liver transplantation programs the waiting time was usually less than 60 days, but the success of the procedure has progressively increased this period, during which the tumor may progress and impede transplantation. Waiting times longer than 12 months imply a 25% probability of exclusion while waiting.126,157 As a result, in some countries where the waiting time is around 2 years, liver transplantation is not an option for HCC patients as most will never reach the operation. This lack of organs implies that the analysis of liver transplantation according to the intention-to-treat principle offers a less encouraging outcome that may be around 60% at 2 years for waiting times of 12 months. There are no RCTs comparing any treatment option versus no treatment on the waiting list and thus, there is no strong evidence showing that a given intervention is effective to prevent progression and exclusion. Surgical resection, percutaneous ablation, and transarterial chemoembolization have been applied by most groups with a long enough waiting time between enlistment and transplantation. The benefits of treatment aiming to prevent progression have to be balanced against the risk associated with therapy and the expected waiting time is the major parameter in deciding to apply treatment. As a whole, percutaneous ablation appears to be effective and with an acceptable safety profile, but surgical resection and chemoembolization imply a higher risk of liver failure if patients do not have enough liver function reserve.158,159 Systemic chemotherapy has no efficacy. In any case, in most programs time on the waiting list has appeared to be the main predictor of effective transplantation and thus, of long-term survival. According to this situation those patients with more severe disease or more active HCC would never reach transplantation and this has prompted several groups or governments to develop priority strategies to transplant the sickest. In the USA, United Network for Organ Sharing adopted the MELD system to offer priority to patients awaiting transplantation. This composite score, based on the bilirubin, prothrombin time, and creatinine, is used for non-cancer patients and a variable score was initially granted to HCC patients. Patients in stage I (single tumor < 2 cm) received 24 points while patients in stage II (single tumor 2–5 cm or < 3 cm) had 29 points. After initial testing it was recognized that HCC patients received an unfair priority and had a higher likelihood of transplantation in the short term. Thus, points were finally reduced to 20 for stage 2, with no priority given to stage 1.160 Continuous analysis of the performance of the system is being conducted and will no doubt be further modified to achieve a fair distribution of available livers across categories of disease. At the same time, care should be taken not to give priority to too advanced patients whose outcome will be unequivocally dismal, as this would be a misuse of the scarce number of livers.
PERCUTANEOUS TREATMENTS This is a therapeutic option that has grown rapidly during the last decade. Destruction or ablation of tumor cells can be achieved by injecting chemical substances (ethanol, acetic acid, boiling saline) or by inserting a probe that modifies local tumor temperature (radiofrequency, microwave, laser, and cryotherapy). All these alter-
1121
Section X. Tumors of the Liver
natives are usually done percutaneously with minimal invasiveness or during laparoscopy. Currently, percutaneous ablation is the best option for patients with early HCC who are ineligible for resection or transplantation.6 Treatment has to be repeated on separate days and the efficacy of ablation is assessed at 1 month by dynamic CT (Figure 59-9).4 The absence of contrast uptake in the tumor reflects tumor necrosis, while the recognition of contrast uptake constitutes treatment failure. Contrast-enhanced US may also inform about the success of therapy, but has lower efficacy for the detection of new tumor sites during follow-up. Recurrence rate after percutaneous ablation is similar to that seen after surgical resection and occurs within the vicinity of the treated nodule or in separate segments of the liver.138,161 In addition, some of the tumors initially classified as completely necrosed present intratumoral recurrence later on and this should be seen as treatment failure that has not been detected immediately after therapy. Percutaneous ethanol injection (PEI), usually performed under US guidance, is the best known and studied of the percutaneous therapies. It is highly effective for small HCCs and has a low rate of adverse effects. It achieves complete tumor necrosis in 90–100% of HCCs < 2cm in diameter. This is reduced to 70% in 2–3-cm tumors and 50% in 3–5-cm tumors The 5-year survival of Child–Pugh A candidates with a complete response is 50%.140,162–164 The presence of septa within large tumor nodules prevents the diffusion of ethanol into the whole tumor volume and hence, complete necrosis is rarely achieved in tumors >3 cm. The need to repeat ethanol injection and the suboptimal efficacy in large tumors has prompted the development of alternative techniques able to ablate larger nodules in a smaller number of procedures. Radiofrequency ablation is the most extensively used alternative to PEI. It can be applied through single or multiple cooled-tip electrodes percutaneously, laparoscopically, or intraoperatively. It achieves the same objective response as PEI in significantly fewer sessions and it is superior in tumors >2 cm.140,162–164 The main drawbacks of radiofrequency ablation are its higher cost and associated 10% morbidity.165 Superficial tumors should not be treated percutaneously by direct puncture because of the risk of tumor seeding. In addition, tumors in close proximity to the hilum or gallbladder may damage the biliary tree. Treatment of tumors in close proximity to the heart or major vessels should also be avoided.
PALLIATIVE THERAPY A large list of therapeutic interventions has been proposed for patients with advanced HCC. The lack of a standard therapy with unequivocal positive impact on survival has allowed the evaluation of different approaches, but unfortunately only a few of them have been properly tested in prospective RCTs comparing active therapy with best supportive care.125 Most options have been studied within phase II trials or within RCTs where two active interventions are compared. In only a few options is there suggestive evidence of therapeutic benefit and, when attempting a structured review and metaanalysis of all the available studies, the sole intervention that offers a positive effect on survival is chemoembolization.125 Strong data have also established the absence of efficacy of estrogen blockade with tamoxifen.125 Antiandrogenic treatment166 and octreotide administration167 have also shown no efficacy. Vitamin D derivates have no effect. Interferon administration has been tested in small
1122
Table 59-5. Systemic Agents used for Advanced Hepatocellular Carcinoma and Response Rate Systemic chemotherapy Doxorubicin as single agent Cisplatin Epirubicin Mitoxantrone Etoposide 5-fluorouracil, paclitaxel, iridotecan, gemcitabine Antiandrogen Interferon Octreotide Seocalcitol (vitamin D derivative)
No. of patients
Response rate
>1000 48 62 118 26
10–18% 10% 11% 16% 13% <10%
376 60 60 372
<10% <10% <5% <5%
studies, with negative results, and it is not well tolerated.168 Systemic chemotherapy has just marginal activity (<10% objective responses) (Table 59-5) and may be associated with severe toxicity. Ultimately, there is no impact on survival. As the majority of intraarterial chemotherapy trials have not included a no-treatment arm, no definitive conclusions can be drawn. Internal radiation with I132,169 or other isotopes,170 and radiation171 have been assessed in too few small studies to provide solid conclusions.
Arterial Embolization The majority of hepatic blood flow in a normal liver arises from the portal vein (70%), while the remainder arises from the hepatic artery. In contrast, the blood supply to HCC is predominantly arterial. In fact, the intense angiogenic activity in malignant tumors results in the typical hypervascular appearance of the tumors and offers the pathologic basis of both the characteristic diagnostic radiologic features of HCC and the therapeutic rationale supporting arterial obstruction. Hepatic artery blood flow results in extensive tumor necrosis and has been a therapeutic option for years. Extrahepatic disease should be ruled out and the absence of portal blood flow (secondary to portal vein obstruction with thrombosis or tumor, portosystemic anastomosis, or hepatofugal flow) constitutes the main contraindication. Patients with advanced disease (Child–Pugh C) should also be excluded from this treatment. Gelfoam prepared as 1-mm cubes is the most commonly used agent for arterial obstruction, but polyvinyl alcohol, beads, alcohol, starch microspheres, blood clots, and metallic coils have all been used.172 Hepatic artery obstruction is performed during an angiographic procedure and is known as TAE. The catheter is advanced into the hepatic artery and the end-point is to interrupt blood flow to the tumor in as selective a way as possible (Figure 59-10). This will ensure maximal ischemic action and avoid injuring the surrounding non-tumor liver. However, if treating multifocal HCC involving both right and left hepatic lobes, both hepatic arteries may be safely occluded if the liver function is well preserved. Injection of any agent has to be done carefully to avoid the backward flow of particles with embolization of arterial vessels outside the liver. The cystic artery must be preserved to avoid necrosis of the gallbladder. When TAE is combined with the prior injection into the hepatic artery of chemotherapy, most commonly doxorubicin, mitomycin, or cisplatin, the procedure is known as transarterial chemoembolization. It is usual to suspend the chemotherapeutic agent in
Chapter 59 HEPATOCELLULAR CARCINOMA Figure 59-9. Percutaneous treatment of a small solitary hepatocellular carcinoma (HCC) located in the right lobe. (A) HCC at baseline. Arterial enhancement is recognized. (B) Tumor after successful treatment. There is no contrast uptake in the treated area. This reflects complete tumor necrosis.
A
B
1123
Section X. Tumors of the Liver
Figure 59-10. (A) Hepatic angiography showing two large hepatocellular carcinoma nodules nourished by the left hepatic artery. (B) Selective catheterization with injection of gelfoam sponge achieves complete obstruction of the left hepatic artery with preserved flow through the right branch.
A
B
lipiodol, an oily contrast agent that is selectively retained within the tumor. This aims to enhance the exposure of the tumor cells to the chemotherapy. It is common practice to inject 25% of chemotherapy into the tumor-free lobe, aiming to act on any potentially undetected tumor cells.172 The side effects of intra-arterial injection of chemotherapy are the same as for systemic administration: nausea, vomiting, bone marrow suppression, alopecia, and renal impairment. In addition, the hepatic artery obstruction with induced necrosis of the tumor is followed in 50% of patients by the postembolization syndrome.172 This con-
1124
sists of fever, abdominal pain, and a moderate degree of ileus. It is usually self-limited and most patients can be discharged within 48–72 hours. Fasting is required for 24 hours prior to the procedure and intravenous hydration is mandatory. Prophylactic antibiotics are not routinely used. Postprocedure fever is usually a reflection of tumor necrosis, but if it persists, severe infectious complications, such as a hepatic abscess or cholecystitis, should be ruled out. There is large heterogeneity in the assessment of response to therapy and criteria employed. Objective responses registered by the identification of intratumoral necrosis and reduced tumor
Chapter 59 HEPATOCELLULAR CARCINOMA
burden on dynamic CT scan or MRI are seen in 15–55% of patients. Therapeutic efficacy is accompanied by delayed tumor progression and vascular invasion.173,174 Tumor necrosis is also reflected by a decrease in tumor marker concentration. However, the residual tumor typically recovers its blood supply and repeated treatment is necessary. The positive impact of embolization on survival was unclear until very recently. The available RCTs comparing chemoembolization with no treatment124,175–177 did not evidence a benefit until the recent publications of two studies from Hong Kong178 and Barcelona.174 The type of patients recruited and the type of agent used were different but both registered a benefit in survival and allowed a cumulative meta-analysis to become significant (Figure 59-11). Clearly, there is room for improvement in the type of agents used to obstruct the arterial blood flow, the chemotherapy to be injected, and the timing of treatment. However, currently, chemoembolization is the sole option that accomplishes all the evidence-based requests to be considered standard of care. Bland embolization has a relevant antitumoral effect, but evidence of a positive impact on survival is lacking.124
Future Agents The current therapies for HCC are directed towards the physical elimination of the tumor, while treatment based on selective biological targets governing the evolution of the tumor are not in place. Current advances in the knowledge of the molecular abnormalities that are responsible for tumor development and progression should dramatically change the range of available therapies. Gene therapy to correct an abnormal gene profile is the most awaited advance, but it is still far from maturity. In contrast, a large number of new agents have recently been launched and are currently under evaluation. These include tyrosine kinase inhibitors, antiangiogenic agents against vascular endothelial growth factor, inhibitors of epidermal
Author, Journal year
Patients
Lin, Gastroenterology 1988
63
GETCH, NEJM 1995
96
Bruix, Hepatology 1988
80
Pelletier, J Hepatol 1998
73
Lo, Hepatology 2002
79
Llovet, Lancet 2002
112
OVERALL
503
Heterogeneity: p=0.14
growth factor, and telomerase pathway inhibitors, among others. Some of these agents do not pretend to necrose the tumor directly, but rather prevent its progression and, in that way, delay or avoid cancer-related death. All those agents showing promising results in the early clinical development phase will have to undergo critical evaluation of their efficacy and benefits. For this aim, the sole method of establishing clinical benefits is to conduct large RCTs comparing active therapy with best supportive care. This is based on a scientific and ethical rationale. Supportive care should be considered the standard of care in patients with advanced HCC, while chemotherapy or other medication should be disregarded because of a lack of efficacy and potential induction of severe side effects that may reduce quality of life and even survival.
PREVENTION Several cancers have well-defined risk factors that, if avoided, could prompt a major reduction in cancer-related death. HCC is among these neoplasms, as in almost all cases there is an etiologic agent that has induced liver disease and ultimately, liver cancer. Viral infection is the leading cause of HCC and the dissemination of both HBV and HCV can be effectively prevented. Active vaccination against HBV reduces the carrier rate in the population and, in countries where universal vaccination has been implemented, the reduction in HCC incidence has become apparent within the next decade.7 Accordingly, HCC may be one of the leading cancers that could be prevented by vaccination. There is no effective vaccination for HCV and prevention of its transmission is based on an improvement in health standards. Control of blood transfusion and use of blood derivates has been a major advancement. Avoiding contact with blood is of critical importance; this includes not only the use of disposable needles, but also careful use of razors, piercing equipment,
Random effects model (DerSimonian & Lairo OR (95% CI) 0.01 0.1 0.5 1 2 10 100
Figure 59-11. Cumulative meta-analysis of the randomized controlled trials testing transarterial embolization versus no treatment for patients with non-surgical hepatocellular carcinoma. The sum of all results offers a significant difference favoring treatment. (Modified from Llovet JM, Bruix J. Systematic review of randomized trials for unresectable hepatocellular carcinoma: chemoembolization improves survival. Hepatology 2003; 37:429–442, with permission.)
p=0.017
Favors treatment
Favors control
1125
Section X. Tumors of the Liver
and any other instrument that may be spoiled by contaminated blood. Health campaigns in the population should reduce excessive alcohol intake and healthy lifestyle should control the obesity epidemic that is responsible for a large proportion of liver disease frequently classified as cryptogenic. If primary prevention has failed and the etiologic agent is already present or has induced liver damage, the tools for prevention and their efficacy are less than optimal. Antiviral treatment against HBV may abrogate viral replication and achieve an inactive HBV carrier state.179 In Europe and the USA, the risk of HCC in the absence of cirrhosis is low. By contrast, in Asia, even in the absence of activity, the risk may persist. If cirrhosis is already present, the reduction of HCC risk related to antiviral therapy is difficult to prove.180 Most drugs marketed in the last decade to treat HBV have shown the capacity to reduce viral replication and seroconvert patients, but their preventive effect is still not well established, despite promising preliminary results with lamivudine.181 The same applies to treatment with interferon. Analysis of the several cohort studies that have been conducted worldwide shows a significant effect of interferon in Asia; however this has not been detected in the west.180 The efficacy of antiviral treatment or prolonged interferon administration in patients with chronic liver disease due to HCV is controversial, but less contradictory. There is general agreement that effective treatment cures viral infection and, if this is the case, noncirrhotic chronic liver disease does not progress and may even reverse. When cirrhosis is already established, the impact of treatment is less clear. Viral infection may be eradicated but long-standing liver injury may have already induced the genetic damage needed for malignant transformation. Hence, HCC may emerge at the same rate as in non-cured cirrhotics. The two available RCTs show discrepant results and the suggestive data registered in several nonrandomized cohort studies may be the result of a selection bias, rather than an effect of treatment.182 Large RCTs are currently ongoing in HCV cirrhotics and the eagerly awaited results should clarify this issue. There are other health interventions that may be effective in prevention in addition to avoidance of viral infection and alcoholism. Iron deposition should be controlled in patients with hereditary hemochromatosis and thus prevent progression to cirrhosis and cancer. Aflatoxin contamination of food could be eliminated by proper storage conditions. Some drugs, such as Oltipraz, may help to metabolize aflatoxin and eliminate toxic metabolites,183 but the real impact of this strategy is unknown. The protective effects of soya consumption are not established. Selenium administration as an antioxidant is currently being tested, while garlic has no effect. Finally, a small study has suggested a beneficial effect of vitamin K administration,184 but the small size of this study prevents any robust conclusions being drawn.
FUTURE PROSPECTS Major advances have taken place over the last 20 years. These have prompted earlier diagnosis, better evaluation and staging of the disease, and a more structured decision process at the time of treatment indication. Over the next few years a major effort should be made to implement the effective prevention strategies outlined
1126
above. In addition, screening of the population at risk through blood sampling and non-invasive imaging techniques should further advance the time of diagnosis when the tumor has not acquired a fully overt malignant phenotype. This would imply detection and diagnosis at a premalignant stage or when the lesion could be defined as carcinoma in situ. The efficacy of therapy should be further improved by developing adjuvant agents and targeting biological mechanisms. In fact, identifying the markers of abnormal regulation of the relevant mechanisms should allow both tailored treatment and also a better prediction of the evolution of the disease and response to treatment. Fulfillment of all these expectations will come from the genomic and proteomic research currently conducted and HCC should benefit from the same gene and protein profiling that has changed the classification and management of malignant disease of breast, prostate, lung cancer, non-Hodgkin’s Blymphoma, and melanoma. All these predictions will take years to become reality. Their development will have to be done jointly by basic and clinical researchers who should share the ultimate goal of transforming this now major cancer killer into a preventable illness or, if this is not achieved, into a curable condition.
ACKNOWLEDGMENTS Josep M Llovet is supported by a grant from AGAUR (2003BEAI00138 and 2004BE00226, Generalitat de Catalunya, Spain), Instituto de Salud Carlos III (Fondo de Investigaciones Sanitarias 2002-2005, PI02/0596) and Programa “Ramon y Cajal” (IDIBAPS, Ministerio de Ciencia y Tecnología, Spain).
REFERENCES 1. Parkin DM, Whelan M, Ferlay J, et al. Cancer incidence in five continents. Lyon: IARC Scientific Publications; 2002. 2. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin 2005; 55:74–108. 3. Bosch FX, Ribes J, Diaz M, et al. Primary liver cancer: worldwide incidence and trends. Gastroenterology 2004; 127 (Suppl 1):S5–S16. 4. Bruix J, Sherman M, Llovet JM, et al. Clinical management of hepatocellular carcinoma: conclusions of the Barcelona-2000 EASL conference. J Hepatol 2001; 35:421–430. 5. Benvegnu L, Gios M, Boccato S, et al. Natural history of compensated viral cirrhosis: a prospective study on the incidence and hierarchy of major complications. Gut 2004; 53:744–749. 6. Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet 2003; 362:1907–1917. 7. Chang MH, Chen CJ, Lai MS, et al. Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. N Engl J Med 1997; 336:1855–1859. 8. Khan SA, Taylor-Robinson SD, Toledano MB, et al. Changing international trends in mortality rates for liver, biliary and pancreatic tumours. J Hepatol 2002; 37:806–813. 9. elSaadany S, Tepper M, Mao Y, et al. An epidemiologic study of hepatocellular carcinoma in Canada. Can J Public Health 2002; 93:443–446. 10. El Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999; 340:745–750. 11. El Serag HB. Hepatocellular carcinoma: recent trends in the United States. Gastroenterology 2004; 127 (Suppl 1):S27–S34.
Chapter 59 HEPATOCELLULAR CARCINOMA
12. Tanaka Y, Hanada K, Mizokami M, et al. A comparison of the molecular clock of hepatitis C virus in the United States and Japan predicts that hepatocellular carcinoma incidence in the United States will increase over the next two decades. Proc Natl Acad Sci USA 2002; 99:15584–15589. 13. Kiyosawa K, Umemura T, Ichijo T, et al. Hepatocellular carcinoma: recent trends in Japan. Gastroenterology 2004; 127 (Suppl 1):S17–S26. 14. Fattovich G, Stroffolini T, Zagni I, et al. Hepatocellular carcinoma in cirrhosis: incidence and risk factors. Gastroenterology 2004; 127 (Suppl 1):S35–S50. 15. Brechot C. Pathogenesis of hepatitis B virus-related hepatocellular carcinoma: old and new paradigms. Gastroenterology 2004; 127 (Suppl 1):S56–S61. 16. Beasley RP, Hwang LY, Lin CC, et al. Hepatocellular carcinoma and hepatitis B virus. A prospective study of 22 707 men in Taiwan. Lancet 1981; 2:1129–1133. 17. Evans AA, Chen G, Ross EA, et al. Eight-year follow-up of the 90 000-person Haimen City cohort: I. Hepatocellular carcinoma mortality, risk factors, and gender differences. Cancer Epidemiol Biomarkers Prev 2002; 11:369–376. 18. Yang HI, Lu SN, Liaw YF, et al. Hepatitis B e antigen and the risk of hepatocellular carcinoma. N Engl J Med 2002; 347:168–174. 19. Yuen MF, Wong DK, Sablon E, et al. HBsAg seroclearance in chronic hepatitis B in the Chinese: virological, histological, and clinical aspects. Hepatology 2004; 39:1694–1701. 20. Fattovich G. Natural history of hepatitis B. J Hepatol 2003; 39 (Suppl 1):S50–S58. 21. Manno M, Camma C, Schepis F, et al. Natural history of chronic HBV carriers in northern Italy: morbidity and mortality after 30 years. Gastroenterology 2004; 127:756–763. 22. Sanchez-Tapias JM, Costa J, Mas A, et al. Influence of hepatitis B virus genotype on the long-term outcome of chronic hepatitis B in western patients. Gastroenterology 2002; 123:1848–1856. 23. Sherman M, Peltekian KM, Lee C. Screening for hepatocellular carcinoma in chronic carriers of hepatitis B virus: incidence and prevalence of hepatocellular carcinoma in a North American urban population. Hepatology 1995; 22:432–438. 24. McMahon BJ, Holck P, Bulkow L, et al. Serologic and clinical outcomes of 1536 Alaska Natives chronically infected with hepatitis B virus. Ann Intern Med 2001; 135:759–768. 25. Kuang SY, Jackson PE, Wang JB, et al. Specific mutations of hepatitis B virus in plasma predict liver cancer development. Proc Natl Acad Sci USA 2004; 101:3575–3580. 26. Kao JH, Chen PJ, Lai MY, et al. Basal core promoter mutations of hepatitis B virus increase the risk of hepatocellular carcinoma in hepatitis B carriers. Gastroenterology 2003; 124:327–334. 27. Brechot C, Gozuacik D, Murakami Y, et al. Molecular bases for the development of hepatitis B virus (HBV)-related hepatocellular carcinoma (HCC). Semin Cancer Biol 2000; 10:211–231. 28. Ferber MJ, Montoya DP, Yu C, et al. Integrations of the hepatitis B virus (HBV) and human papillomavirus (HPV) into the human telomerase reverse transcriptase (hTERT) gene in liver and cervical cancers. Oncogene 2003; 22:3813–3820. 29. Brechot C, Thiers V, Kremsdorf D, et al. Persistent hepatitis B virus infection in subjects without hepatitis B surface antigen: clinically significant or purely “occult”? Hepatology 2001; 34:194–203. 30. Pollicino T, Squadrito G, Cerenzia G, et al. Hepatitis B virus maintains its pro-oncogenic properties in the case of occult HBV infection. Gastroenterology 2004; 126:102–110. 31. Bruix J, Barrera JM, Calvet X, et al. Prevalence of antibodies to hepatitis C virus in Spanish patients with hepatocellular carcinoma and hepatic cirrhosis [see comments]. Lancet 1989; 2:1004–1006.
32. Sun CA, Wu DM, Lin CC, et al. Incidence and cofactors of hepatitis C virus-related hepatocellular carcinoma: a prospective study of 12 008 men in Taiwan. Am J Epidemiol 2003; 157:674–682. 33. Yoshida H, Shiratori Y, Moriyama M, et al. Interferon therapy reduces the risk for hepatocellular carcinoma: national surveillance program of cirrhotic and noncirrhotic patients with chronic hepatitis C in Japan. IHIT Study Group. Inhibition of Hepatocarcinogenesis by Interferon Therapy. Ann Intern Med 1999; 131:174–181. 34. Liang TJ, Heller T. Pathogenesis of hepatitis C-associated hepatocellular carcinoma. Gastroenterology 2004; 127 (Suppl 1):S62–S71. 35. De Mitri MS, Poussin K, Baccarini P, et al. HCV-associated liver cancer without cirrhosis. Lancet 1995; 345:413–415. 36. Lauer GM, Walker BD. Hepatitis C virus infection. N Engl J Med 2001; 345:41–52. 37. Puoti M, Bruno R, Soriano V, et al. Hepatocellular carcinoma in HIV-infected patients: epidemiological features, clinical presentation and outcome. AIDS 2004; 18:2285–2293. 38. Yamanaka T, Shiraki K, Nakazaawa S, et al. Impact of hepatitis B and C virus infection on the clinical prognosis of alcoholic liver cirrhosis. Anticancer Res 2001; 21:2937–2940. 39. Bagnardi V, Blangiardo M, La Vecchia C, et al. Alcohol consumption and the risk of cancer: a meta-analysis. Alcohol Res Health 2001; 25:263–270. 40. Morgan TR, Mandayam S, Jamal MM. Alcohol and hepatocellular carcinoma. Gastroenterology 2004; 127 (Suppl 1):S87–S96. 41. Ozturk M. Genetic aspects of hepatocellular carcinogenesis. Semin Liver Dis 1999; 19:235–242. 42. Qian GS, Ross RK, Yu MC, et al. A follow-up study of urinary markers of aflatoxin exposure and liver cancer risk in Shanghai, People’s Republic of China. Cancer Epidemiol Biomarkers Prev 1994; 3:3–10. 43. Fracanzani AL, Conte D, Fraquelli M, et al. Increased cancer risk in a cohort of 230 patients with hereditary hemochromatosis in comparison to matched control patients with non-iron-related chronic liver disease. Hepatology 2001; 33:647–651. 44. Elmberg M, Hultcrantz R, Ekbom A, et al. Cancer risk in patients with hereditary hemochromatosis and in their firstdegree relatives. Gastroenterology 2003; 125:1733–1741. 45. Britto MR, Thomas LA, Balaratnam N, et al. Hepatocellular carcinoma arising in non-cirrhotic liver in genetic haemochromatosis. Scand J Gastroenterol 2000; 35:889–893. 46. Choudhury S, Zhang R, Frenkel K, et al. Evidence of alterations in base excision repair of oxidative DNA damage during spontaneous hepatocarcinogenesis in Long Evans Cinnamon rats. Cancer Res 2003; 63:7704–7707. 47. Chang B, Nishikawa M, Nishiguchi S, et al. L-carnitine inhibits hepatocarcinogenesis via protection of mitochondria. Int J Cancer 2005; 113:719–729. 48. Adami HO, Chow WH, Nyren O, et al. Excess risk of primary liver cancer in patients with diabetes mellitus. J Natl Cancer Inst 1996; 88:1472–1477. 49. El Serag HB, Richardson PA, Everhart JE. The role of diabetes in hepatocellular carcinoma: a case-control study among United States Veterans. Am J Gastroenterol 2001; 96:2462–2467. 50. Calle EE, Rodriguez C, Walker-Thurmond K, et al. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of US adults. N Engl J Med 2003; 348:1625–1638. 51. Nair S, Mason A, Eason J, et al. Is obesity an independent risk factor for hepatocellular carcinoma in cirrhosis? Hepatology 2002; 36:150–155. 52. Ratziu V, Bonyhay L, Di M V, et al. Survival, liver failure, and hepatocellular carcinoma in obesity-related cryptogenic cirrhosis. Hepatology 2002; 35:1485–1493.
1127
Section X. Tumors of the Liver
53. Bugianesi E, Leone N, Vanni E, et al. Expanding the natural history of nonalcoholic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology 2002; 123:134–140. 54. Caballeria L, Pares A, Castells A, et al. Hepatocellular carcinoma in primary biliary cirrhosis: similar incidence to that in hepatitis C virus-related cirrhosis. Am J Gastroenterol 2001; 96:1160–1163. 55. Wang LY, You SL, Lu SN, et al. Risk of hepatocellular carcinoma and habits of alcohol drinking, betel quid chewing and cigarette smoking: a cohort of 2416 HBsAg-seropositive and 9421 HBsAgseronegative male residents in Taiwan. Cancer Causes Control 2003; 14:241–250. 56. Kuper H, Tzonou A, Kaklamani E, et al. Tobacco smoking, alcohol consumption and their interaction in the causation of hepatocellular carcinoma. Int J Cancer 2000; 85:498–502. 57. Marrero JA, Fontana RJ, Fu S, et al. Alcohol, tobacco and obesity are synergistic risk factors for hepatocellular carcinoma. J Hepatol 2005; 42:218–224. 58. Inoue M, Yoshimi I, Sobue T, et al. Influence of coffee drinking on subsequent risk of hepatocellular carcinoma: a prospective study in Japan. J Natl Cancer Inst 2005; 97:293–300. 59. MILTS. Oral contraceptives and liver cancer. Results of the Multicentre International Liver Tumor Study. Contraception 1997; 56:275–284. 60. Yu MW, Yang YC, Yang SY, et al. Hormonal markers and hepatitis B virus-related hepatocellular carcinoma risk: a nested case-control study among men. J Natl Cancer Inst 2001; 93:1644–1651. 61. Fasani P, Sangiovanni A, De Fazio C, et al. High prevalence of multinodular hepatocellular carcinoma in patients with cirrhosis attributable to multiple risk factors. Hepatology 1999; 29:1704–1707. 62. Benvegnu L, Noventa F, Bernardinello E, et al. Evidence for an association between the aetiology of cirrhosis and pattern of hepatocellular carcinoma development. Gut 2001; 48: 110–115. 63. Degos F, Christidis C, Ganne-Carrie N, et al. Hepatitis C virus related cirrhosis: time to occurrence of hepatocellular carcinoma and death. Gut 2000; 47:131–136. 64. Velazquez RF, Rodriguez M, Navascues CA, et al. Prospective analysis of risk factors for hepatocellular carcinoma in patients with liver cirrhosis. Hepatology 2003; 37:520–527. 65. Sangiovanni A, Del Ninno E, Fasani P, et al. Increased survival of cirrhotic patients with a hepatocellular carcinoma detected during surveillance. Gastroenterology 2004; 126:1005–1014. 66. Donato MF, Arosio E, Del Ninno E, et al. High rates of hepatocellular carcinoma in cirrhotic patients with high liver cell proliferative activity. Hepatology 2001; 34:523–528. 67. Borzio M, Trere D, Borzio F, et al. Hepatocyte proliferation rate is a powerful parameter for predicting hepatocellular carcinoma development in liver cirrhosis. Mol Pathol 1998; 51:96–101. 68. Ganne-Carrie N, Chastang C, Chapel F, et al. Predictive score for the development of hepatocellular carcinoma and additional value of liver large cell dysplasia in western patients with cirrhosis. Hepatology 1996; 23:1112–1118. 69. Shibata M, Morizane T, Uchida T, et al. Irregular regeneration of hepatocytes and risk of hepatocellular carcinoma in chronic hepatitis and cirrhosis with hepatitis-C-virus infection. Lancet 1998; 351:1773–1777. 70. Sherman M. Alphafetoprotein: an obituary. J Hepatol 2001; 34:603–605. 71. Trevisani F, D’Intino PE, Morselli-Labate AM, et al. Serum alpha-fetoprotein for diagnosis of hepatocellular carcinoma in patients with chronic liver disease: influence of HBsAg and antiHCV status. J Hepatol 2001; 34:570–575. 72. Gupta S, Bent S, Kohlwes J. Test characteristics of alphafetoprotein for detecting hepatocellular carcinoma in patients
1128
73. 74. 75.
76.
77.
78.
79. 80.
81.
82. 83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
with hepatitis C. A systematic review and critical analysis. Ann Intern Med 2003; 139:46–50. Buendia MA. Genetics of hepatocellular carcinoma. Semin Cancer Biol 2000; 10:185–200. Thorgeirsson SS, Grisham JW. Molecular pathogenesis of human hepatocellular carcinoma. Nat Genet 2002; 31:339–346. Borzio M, Fargion S, Borzio F, et al. Impact of large regenerative, low grade and high grade dysplastic nodules in hepatocellular carcinoma development. J Hepatol 2003; 39:208–214. Seki S, Sakaguchi H, Kitada T, et al. Outcomes of dysplastic nodules in human cirrhotic liver: a clinicopathological study. Clin Cancer Res 2000; 6:3469–3473. Kojiro M, Nakashima O. Histopathologic evaluation of hepatocellular carcinoma with special reference to small early stage tumors. Semin Liver Dis 1999; 19:287–296. Kojiro M. Focus on dysplastic nodules and early hepatocellular carcinoma: an eastern point of view. Liver Transpl 2004; 10 (Suppl 1):S3–S8. Nakashima T, Kojiro M. Hepatocellular carcinoma. Tokyo: Springer Verlag; 1987. Ishak KG, Goodman ZD, Stocker JT. Tumors of the liver and intrahepatic bile ducts. Washington, DC: Armed Forces Institute of Pathology; 2001. Lechel A, Manns MP, Rudolph KL. Telomeres and telomerase: new targets for the treatment of liver cirrhosis and hepatocellular carcinoma. J Hepatol 2004; 41:491–497. Bruix J, Boix L, Sala M, et al. Focus on hepatocellular carcinoma. Cancer Cell 2004; 5(3): 215–219. Laurent-Puig P, Legoix P, Bluteau O, et al. Genetic alterations associated with hepatocellular carcinomas define distinct pathways of hepatocarcinogenesis. Gastroenterology 2001; 120:1763–1773. Calvisi DF, Factor VM, Ladu S, et al. Disruption of beta-catenin pathway or genomic instability define two distinct categories of liver cancer in transgenic mice. Gastroenterology 2004; 126:1374–1386. Ye QH, Qin LX, Forgues M, et al. Predicting hepatitis B viruspositive metastatic hepatocellular carcinomas using gene expression profiling and supervised machine learning. Nat Med 2003; 9:416–423. Smith MW, Yue ZN, Geiss GK, et al. Identification of novel tumor markers in hepatitis C virus-associated hepatocellular carcinoma. Cancer Res 2003; 63:859–864. Lee JS, Thorgeirsson SS. Genome-scale profiling of gene expression in hepatocellular carcinoma: classification, survival prediction, and identification of therapeutic targets. Gastroenterology 2004; 127 (Suppl 1):S51–S55. Llovet JM, Wurmbach E. Gene expression profiles in hepatocellular carcinoma: not yet there. J Hepatol 2004; 41:336–339. Kutami R, Nakashima Y, Nakashima O, et al. Pathomorphologic study on the mechanism of fatty change in small hepatocellular carcinoma of humans. J Hepatol 2000; 33:282–289. Nakashima Y, Nakashima O, Tanaka M, et al. Portal vein invasion and intrahepatic micrometastasis in small hepatocellular carcinoma by gross type. Hepatol Res 2003; 26:142–147. Takayama T, Makuuchi M, Hirohashi S, et al. Early hepatocellular carcinoma as an entity with a high rate of surgical cure. Hepatology 1998; 28:1241–1246. Sakamoto M, Hirohashi S. Natural history and prognosis of adenomatous hyperplasia and early hepatocellular carcinoma: multi-institutional analysis of 53 nodules followed up for more than 6 months and 141 patients with single early hepatocellular carcinoma treated by surgical resection or percutaneous ethanol injection. Jpn J Clin Oncol 1998; 28:604–608. Bolondi L, Sofia S, Siringo S, et al. Surveillance programme of cirrhotic patients for early diagnosis and treatment of
Chapter 59 HEPATOCELLULAR CARCINOMA
94. 95.
96.
97.
98. 99.
100.
101.
102. 103. 104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
hepatocellular carcinoma: a cost-effectiveness analysis. Gut 2001; 48:251–259. Bolondi L. Screening for hepatocellular carcinoma in cirrhosis. J Hepatol 2003; 39:1076–1084. Fracanzani AL, Burdick L, Borzio M, et al. Contrast-enhanced Doppler ultrasonography in the diagnosis of hepatocellular carcinoma and premalignant lesions in patients with cirrhosis. Hepatology 2001; 34:1109–1112. Burrel M, Llovet JM, Ayuso C, et al. MRI angiography is superior to helical CT for detection of HCC prior to liver transplantation: an explant correlation. Hepatology 2003; 38:1034–1042. Gaiani S, Celli N, Piscaglia F, et al. Usefulness of contrastenhanced perfusional sonography in the assessment of hepatocellular carcinoma hypervascular at spiral computed tomography. J Hepatol 2004; 41:421–426. Bruix J, Sherman M. Management of Hepatocellular Carcinoma. Hepathology 2005; 42:1208–1236. Marrero JA, Su GL, Wei W, et al. Des-gamma carboxyprothrombin can differentiate hepatocellular carcinoma from nonmalignant chronic liver disease in American patients. Hepatology 2003; 37:1114–1121. Capurro M, Wanless IR, Sherman M, et al. Glypican-3: a novel serum and histochemical marker for hepatocellular carcinoma. Gastroenterology 2003; 125:89–97. Paradis V, Degos F, Dargere D, et al. Identification of a new marker of hepatocellular carcinoma by serum protein profiling of patients with chronic liver diseases. Hepatology 2005; 41:40–47. Lencioni R, Cioni D, Crocetti L, et al. Magnetic resonance imaging of liver tumors. J Hepatol 2004; 40:162–171. Greene FL, Page DL, Fleming ID, et al. AJCC Cancer staging handbook. New York: Springer; 2002. Pugh RN, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973; 60:646–649. Forman LM, Lucey MR. Predicting the prognosis of chronic liver disease: an evolution from child to MELD. Mayo end-stage liver disease. Hepatology 2001; 33:473–475. Sorensen JB, Klee M, Palshof T, et al. Performance status assessment in cancer patients. An inter-observer variability study. Br J Cancer 1993; 67:773–775. Okuda K, Ohtsuki T, Obata H, et al. Natural history of hepatocellular carcinoma and prognosis in relation to treatment. Study of 850 patients. Cancer 1985; 56:918–928. Chevret S, Trinchet JC, Mathieu D, et al. A new prognostic classification for predicting survival in patients with hepatocellular carcinoma. Groupe d’Etude et de Traitement du Carcinome Hepatocellulaire. J Hepatol 1999; 31:133– 141. Stuart KE, Anand AJ, Jenkins RL. Hepatocellular carcinoma in the United States. Prognostic features, treatment outcome, and survival. Cancer 1996; 77:2217–2222. Llovet JM, Bustamante J, Castells A, et al. Natural history of untreated nonsurgical hepatocellular carcinoma: rationale for the design and evaluation of therapeutic trials. Hepatology 1999; 29:62–67. Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999; 19:329–338. The Cancer of the Liver Italian Program (CLIP) investigators. a new prognostic system for hepatocellular carcinoma: a retrospective study of 435 patients. Hepatology 1998; 28:751–755. Villa E, Moles A, Ferretti I, et al. Natural history of inoperable hepatocellular carcinoma: estrogen receptors’ status in the tumor is the strongest prognostic factor for survival. Hepatology 2000; 32:233–238.
114. Schoniger-Hekele M, Muller C, Kutilek M, et al. Hepatocellular carcinoma in central Europe: prognostic features and survival. Gut 2001; 48:103–109. 115. Herold C, Reck T, Fischler P, et al. Prognosis of a large cohort of patients with hepatocellular carcinoma in a single European centre. Liver 2002; 22:23–28. 116. Leung TW, Tang AM, Zee B, et al. Construction of the Chinese University Prognostic Index for hepatocellular carcinoma and comparison with the TNM staging system, the Okuda staging system, and the Cancer of the Liver Italian Program staging system: a study based on 926 patients. Cancer 2002; 94:1760–1769. 117. Kudo M, Chung H, Haji S, et al. Validation of a new prognostic staging system for hepatocellular carcinoma: the JIS score compared with the CLIP score. Hepatology 2004; 40:1396–1405. 118. Omagari K, Honda S, Kadokawa Y, et al. Preliminary analysis of a newly proposed prognostic scoring system (SLiDe score) for hepatocellular carcinoma. J Gastroenterol Hepatol 2004; 19:805–811. 119. Ueno S, Tanabe G, Sako K, et al. Discrimination value of the new western prognostic system (CLIP score) for hepatocellular carcinoma in 662 Japanese patients. Cancer of the Liver Italian Program. Hepatology 2001; 34:529–534. 120. Levy I, Sherman M. Staging of hepatocellular carcinoma: assessment of the CLIP, Okuda, and Child–Pugh staging systems in a cohort of 257 patients in Toronto. Gut 2002; 50:881–885. 121. Cillo U, Bassanello M, Vitale A, et al. The critical issue of hepatocellular carcinoma prognostic classification: which is the best tool available? J Hepatol 2004; 40:124–131. 121. Grieco A, Pompili M, Caminiti G, et al. Prognostic factors for survival in patients with early-intermediate hepatocellular carcinoma undergoing non-surgical therapy: comparison of Okuda, CLIP, and BCLC staging systems in a single Italian centre. Gut 2005; 54:411–418. 123. Marrero J, Fontana R, Barrat A. Prognosis of hepatocellular carcinoma: comparison of seven staging systems in an American cohort. Hepatology 2005; 41:707–716. 124. Bruix J, Llovet JM, Castells A, et al. Transarterial embolization versus symptomatic treatment in patients with advanced hepatocellular carcinoma: results of a randomized, controlled trial in a single institution. Hepatology 1998; 27:1578–1583. 125. Llovet JM, Bruix J. Systematic review of randomized trials for unresectable hepatocellular carcinoma: chemoembolization improves survival. Hepatology 2003; 37:429–442. 126. Llovet JM, Fuster J, Bruix J. Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology 1999; 30:1434–1440. 127. Torzilli G, Makuuchi M, Inoue K, et al. No-mortality liver resection for hepatocellular carcinoma in cirrhotic and noncirrhotic patients: is there a way? A prospective analysis of our approach. Arch Surg 1999; 134:984–992. 128. Fan ST, Lo CM, Liu CL, et al. Hepatectomy for hepatocellular carcinoma: toward zero hospital deaths. Ann Surg 1999; 229:322–330. 129. Arii S, Yamaoka Y, Futagawa S, et al. Results of surgical and nonsurgical treatment for small-sized hepatocellular carcinomas: a retrospective and nationwide survey in Japan. Hepatology 2000; 32:1224–1229. 130. Takayama T, Sekine T, Makuuchi M, et al. Adoptive immunotherapy to lower postsurgical recurrence rates of hepatocellular carcinoma: a randomised trial. Lancet 2000; 356:802–807. 131. Poon RT, Fan ST, Lo CM, et al. Long-term survival and pattern of recurrence after resection of small hepatocellular carcinoma in patients with preserved liver function: implications for a strategy of salvage transplantation. Ann Surg 2002; 235:373–382.
1129
Section X. Tumors of the Liver
132. Wayne JD, Lauwers GY, Ikai I, et al. Preoperative predictors of survival after resection of small hepatocellular carcinomas. Ann Surg 2002; 235:722–730. 133. Yamamoto M, Takasaki K, Otsubo T, et al. Favorable surgical outcomes in patients with early hepatocellular carcinoma. Ann Surg 2004; 239:395–399. 134. Esnaola NF, Mirza N, Lauwers GY, et al. Comparison of clinicopathologic characteristics and outcomes after resection in patients with hepatocellular carcinoma treated in the United States, France, and Japan. Ann Surg 2003; 38:711–719. 135. Imamura H, Matsuyama Y, Tanaka E, et al. Risk factors contributing to early and late; 2003; 38:200–207. 136. Schwartz JD, Schwartz M, Mandeli J, et al. Neoadjuvant and adjuvant therapy for resectable hepatocellular carcinoma: review of the randomised clinical trials. Lancet Oncol 2002; 3:593–603. 137. Lau WY, Leung TW, Ho SK, et al. Adjuvant intra-arterial iodine131-labelled lipiodol for resectable hepatocellular carcinoma: a prospective randomised trial. Lancet 1999; 353:797–801. 138. Shiratori Y, Shiina S, Teratani T, et al. Interferon therapy after tumor ablation improves prognosis in patients with hepatocellular carcinoma associated with hepatitis C virus. Ann Intern Med 2003; 138:299–306. 139. Minagawa M, Makuuchi M, Takayama T, et al. Selection criteria for repeat hepatectomy in patients with recurrent hepatocellular carcinoma. Ann Surg 2003; 238:703–710. 140. Sala M, Fuster J, Llovet JM, et al. High pathological risk of recurrence after surgical resection for hepatocellular carcinoma: an indication for salvage liver transplantation. Liver Transpl 2004; 10:1294–1300. 141. Bismuth H, Majno PE, Adam R. Liver transplantation for hepatocellular carcinoma. Semin Liver Dis 1999; 19:311– 322. 142. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334:693–699. 143. Jonas S, Bechstein WO, Steinmuller T, et al. Vascular invasion and histopathologic grading determine outcome after liver transplantation for hepatocellular carcinoma in cirrhosis. Hepatology 2001; 33:1080–1086. 144. Yao FY, Ferrell L, Bass NM, et al. Liver transplantation for hepatocellular carcinoma: expansion of the tumor size limits does not adversely impact survival. Hepatology 2001; 33:1394–1403. 145. Plessier A, Codes L, Consigny Y, et al. Underestimation of the influence of satellite nodules as a risk factor for posttransplantation recurrence in patients with small hepatocellular carcinoma. Liver Transpl 2004; 10 (Suppl 2):S86–S90. 146. Marsh JW, Dvorchick I. Liver organ allocation for hepatocellular carcinoma: are we sure? Liver Transpl 2003; 9:693–696. 147. Cillo U, Vitale A, Bassanello M, et al. Liver transplantation for the treatment of moderately or well-differentiated hepatocellular carcinoma. Ann Surg 2004; 239:150–159. 148. Steinmuller T, Pascher A, Sauer I, et al. Living-donation liver transplantation for hepatocellular carcinoma: time to drop the limitations? Transplant Proc 2002; 34:2263–2264. 149. Kawasaki S. Living-donor liver transplantation for hepatocellular carcinoma. Hepatogastroenterology 2002; 49:53–55. 150. Gondolesi GE, Roayaie S, Munoz L, et al. Adult living donor liver transplantation for patients with hepatocellular carcinoma: extending UNOS priority criteria. Ann Surg 2004; 239:142–149. 151. Todo S, Furukawa H. Living donor liver transplantation for adult patients with hepatocellular carcinoma: experience in Japan. Ann Surg 2004; 240:451–459. 152. Forman LM, Lewis JD, Berlin JA, et al. The association between hepatitis C infection and survival after orthotopic liver transplantation. Gastroenterology 2002; 122:889–896.
1130
153. Garcia-Retortillo M, Forns X, Llovet JM, et al. Hepatitis C recurrence is more severe after living donor compared to cadaveric liver transplantation. Hepatology 2004; 40:699–707. 154. Forman LM, Trotter JF, Emond J. Living donor liver transplantation and hepatitis C. Liver Transpl 2004; 10:347–348. 155. Renz JF, Yersiz H, Reichert PR, et al. Split-liver transplantation: a review. Am J Transplant 2003; 3:1323–1335. 156. Azoulay D, Samuel D, Castaing D, et al. Domino liver transplants for metabolic disorders: experience with familial amyloidotic polyneuropathy. J Am Coll Surg 1999; 189:584–593. 157. Yao FY, Bass NM, Nikolai B, et al. Liver transplantation for hepatocellular carcinoma: analysis of survival according to the intention-to-treat principle and dropout from the waiting list. Liver Transpl 2002; 8:873–883. 158. Llovet JM, Mas X, Aponte JJ, et al. Cost effectiveness of adjuvant therapy for hepatocellular carcinoma during the waiting list for liver transplantation. Gut 2002; 50:123–128. 159. Sala M, Varela M, Bruix J. Selection of candidates with HCC for transplantation in the MELD era. Liver Transpl 2004; 10 (Suppl 2):S4–S9. 160. Freeman RB, Wiesner RH, Edwards E, et al. Results of the first year of the new liver allocation plan. Liver Transpl 2004; 10:7–15. 161. Sala M, Llovet JM, Vilana R, et al. Initial response to percutaneous ablation predicts survival in patients with hepatocellular carcinoma. Hepatology 2004; 40:1352–1360. 162. Livraghi T, Goldberg SN, Lazzaroni S, et al. Small hepatocellular carcinoma: treatment with radio-frequency ablation versus ethanol injection. Radiology 1999; 210:655–661. 163. Lencioni RA, Allgaier HP, Cioni D, et al. Small hepatocellular carcinoma in cirrhosis: randomized comparison of radiofrequency thermal ablation versus percutaneous ethanol injection. Radiology 2003; 228:235–240. 164. Lin SM, Lin CJ, Lin CC, et al. Radiofrequency ablation improves prognosis compared with ethanol injection for hepatocellular carcinoma ≤ 4 cm. Gastroenterology 2004; 127:1714–1723. 165. Livraghi T, Solbiati L, Meloni MF, et al. Treatment of focal liver tumors with percutaneous radio-frequency ablation: complications encountered in a multicenter study. Radiology 2003; 226:441–451. 166. Groupe d’Etude et de Traitement du Carcinome Hepatocellulaire. Randomized trial of leuprorelin and flutamide in male patients with hepatocellular carcinoma treated with tamoxifen. Hepatology 2004; 40:1361–1369. 167. Yuen MF, Poon RT, Lai CL, et al. A randomized placebocontrolled study of long-acting octreotide for the treatment of advanced hepatocellular carcinoma. Hepatology 2002; 36:687–691. 168. Llovet JM, Sala M, Castells L, et al. Randomized controlled trial of interferon treatment for advanced hepatocellular carcinoma. Hepatology 2000; 31:1–5. 169. Raoul JL, Guyader D, Bretagne JF, et al. Prospective randomized trial of chemoembolization versus intra-arterial injection of 131Ilabeled-iodized oil in the treatment of hepatocellular carcinoma. Hepatology 1997; 26:1156–1161. 170. Carr BI. Hepatic arterial 90yttrium glass microspheres (Therasphere) for unresectable hepatocellular carcinoma: interim safety and survival data on 65 patients. Liver Transpl 2004; 10 (Suppl 1):S107–S110. 171. Park HC, Seong J, Han KH, et al. Dose–response relationship in local radiotherapy for hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2002; 54:150–155. 172. Bruix J, Sala M, Llovet JM. Chemoembolization for hepatocellular carcinoma. Gastroenterology 2004; 127 (Suppl 1):S179–S188.
Chapter 59 HEPATOCELLULAR CARCINOMA
173. GETCH. A comparison of lipiodol chemoembolization and conservative treatment for unresectable hepatocellular carcinoma. Groupe d’Etude et de Traitement du Carcinome Hepatocellulaire. N Engl J Med 1995; 332:1256–1261. 174. Llovet JM, Real MI, Montanya X, et al. Arterial embolization, chemoembolization versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomized controlled trial. Lancet 2002; 359:1734–1739. 175. Lin DY, Liaw YF, Lee TY, et al. Hepatic arterial embolization in patients with unresectable hepatocellular carcinoma – a randomized controlled trial. Gastroenterology 1988; 94:453–456. 176. Pelletier G, Ducreux M, Gay F, et al. Treatment of unresectable hepatocellular carcinoma with lipiodol chemoembolization: a multicenter randomized trial. J Hepatol 1998; 29:129–134. 177. GETCH. A comparision of lipiodol chemoembolization in European patients with unresectable hepatocellular carcinoma. N Engl J Med 1995; 332:1256–1261. 178. Lo CM, Ngan H, Tso WK, et al. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology 2002; 35:1164–1171.
179. EASL International Consensus Conference on Hepatitis B. 13–14 September, 2002: Geneva, Switzerland. Consensus statement (short version). J Hepatol 2003; 38:533–540. 180. Lok AS. Prevention of hepatitis B virus-related hepatocellular carcinoma. Gastroenterology 2004; 127 (Suppl 1):S303–S309. 181. Liaw YF, Sung JJ, Chow WC, et al. Lamivudine for patients with chronic hepatitis B and advanced liver disease. N Engl J Med 2004; 351:1521–1531. 182. Heathcote EJ. Prevention of hepatitis C virus-related hepatocellular carcinoma. Gastroenterology 2004; 127 (Suppl 1):S294–S302. 183. Kensler TW, He X, Otieno M, et al. Oltipraz chemoprevention trial in Qidong, People’s Republic of China: modulation of serum aflatoxin albumin adduct biomarkers. Cancer Epidemiol Biomarkers Prev 1998; 7:127–134. 184. Habu D, Shiomi S, Tamori A, et al. Role of vitamin K2 in the development of hepatocellular carcinoma in women with viral cirrhosis of the liver. JAMA 2004; 292:358–361.
1131
Section X: Tumors of the Liver
60
CHOLANGIOCARCINOMA Konstantinos N. Lazaridis and Gregory J. Gores Abbreviations CC cholangiocarcinoma CDK cell division kinase COX-2 cyclooxygenase-2 DAPI 4¢,6-diamidino-2-phenylindole DIA digitized image analysis EBRT external-beam radiation therapy EGF epidermal growth factor EGFR receptor ERCP endoscopic retrograde cholangiopancreatography EUS endoscopic ultrasound FISH fluorescence in situ hybridization FLIP flice-inhibitory protein
HAAH HCV HGF HIV hMLH1 hTERT IL-6 MAPK Mcl-1 MMP MRCP
human aspartyl b-hydroxylase hepatitis C virus hepatocyte growth factor human immunodeficiency virus human Mut L homologue 1 human telomerase reverse transcriptase interleukin-6 mitogen-activated protein kinase myeloid cell leukemia-1 matrix metalloproteinases magnetic resonance cholangiopancreatography
INTRODUCTION Cholangiocarcinoma (CC) is a malignancy of the bile ducts with apparently rising incidence and clinical significance. CC is caused by malignant transformation of the cholangiocyte – the epithelial cell that lines the bile ducts - and accounts for 10–15% of all hepatobiliary neoplasms.1 In clinical practice, CC is divided into two types based on its location along the biliary apparatus, namely, intra- and extrahepatic. At the outset, we would like to stress the distinction between intra- and extrahepatic CC. Although these two clinical conditions have comparable characteristics, each entity has discrete epidemiological as well as clinical features and probable independent etiopathogenetic origins.1 At present, we need to better understand the biology of CC. Gaining such knowledge will likely improve CC early detection, prognosis, and hopefully therapy, which is currently disappointingly limited.2
EPIDEMIOLOGY AND RISK FACTORS EPIDEMIOLOGY Over past decades the incidence of CC has increased.3 Approximately 5000 new cases of CC are diagnosed every year in the USA.4 Of those, about two-thirds involve the extrahepatic bile ducts and the remaining one-third affects the intrahepatic biliary tree. It should be noted that, in many registries, hilar CCs, which involves the right and left hepatic ducts and their confluence, are classified as intrahepatic CC. However, in this review, we consider hilar CCs to be extrahepatic given their risk factors, growth patterns, clinical presentation and biology. Evidence suggests that the epidemiology of intra- as opposed to extrahepatic CC is different. Possible misclassification between these two kinds of CC may have an effect on the available epidemiological parameters. In the USA, based on the Surveillance
MRI MRI NO PDT PSC PTC STAT TNF TNM WISP1v
magnetic resonance imaging computed tomography nitric oxide photodynamic therapy primary sclerosing cholangitis percutaneous transhepatic cholangiography signal transducer and activator of transcription tumor necrosis factor tumor node metastasis WNT1-inducible signaling pathway protein 1
Epidemiology and End Results registries, the age-adjusted incidence rates of intrahepatic CC increased from 0.32/100 000 in 1975–1979 to 0.85/100 000 in 1995–1999.3 In contrast, the incidence of extrahepatic CC decreased slightly from 1.08/100 000 in 1979 to 0.82/100 000 in 19983 (Figure 60-1). Because of the epidemiological disparity, we discuss the epidemiology of intra- versus extrahepatic CC independently. The incidence of intrahepatic CC varies across the world.5 For instance, it is highest in north-east Thailand (96/100 000 in men and 38/100 000 in women), probably due to the high prevalence of liverfluke infestations. In the past, the average age of diagnosis of intrahepatic CC was the mid-50s. However, recently there has been an observed shift in age towards the mid-60s. This observation may relate to: (1) the development of CC in the context of everincreasing chronic liver disease in the present aging population; and (2) improved follow-up and treatment of risk factors (i.e., primary sclerosing cholangitis (PSC), choledochal cysts) in younger individuals. In the USA, the male-to-female ratio for intrahepatic CC is about 1.5. Caucasians and African-Americans have a comparable age-adjusted incidence of intrahepatic CC. In contrast, the incidence among Asians is twice as high as that of Caucasians. However, Caucasians are the only ethnic group in which there is reported gradual increase in the age-adjusted incidence of intrahepatic CC. The mortality related to intrahepatic CC is also increasing worldwide.5 In fact, the percentage of intrahepatic CC increased mortality was greater than that observed for hepatocellular carcinoma. In the USA, the age-adjusted mortality rate for intrahepatic CC increased from 0.07/100 000 in 1973 to 0.69/100 000 in 1997.6 Overall, the 5-year survival of patients with intrahepatic CC remains disappointingly low and practically unchanged over the last two decades. This lack of progress occurs despite earlier detection, employment of aggressive surgical approaches (i.e., hepatectomy), and promising new therapies (i.e., photodynamic therapy [PDT], brachytherapy).
1133
Section X. Tumors of the Liver
Intrahepatic
Extrahepatic
Age-adjusted incidence rate per 100 000
1.20 1.00 0.80 0.60 0.40 0.20 0.00 1975–79
1980–84
1985–89
1990–94
1995–99
Time period Figure 60-1. Trends of intra- and extrahepatic cholangiocarcinoma in the USA. Note that the term intrahepatic cholangiocarcinoma includes hilar lesions. (Modified from Shaib, Y, El-Serag, HB. The epidemiology of cholangiocarcinoma. Semin Liver Dis 2004; 24:115–125.)
Table 60-1. Cholangiocarcinoma Risk Factors Age (over 65 years old) Primary sclerosing cholangitis (PSC) Liver fluke infestation Opisthorchis viverrini Clonorchis sinensis Caroli’s disease Choledochal cysts Bile duct adenoma and biliary papillomatosis Chronic intraductal stones (i.e., hepatolithiasis) Liver cirrhosis Hepatitis C virus Human immunodeficiency (HIV) virus Thorotrast Surgical biliary-enteric drainage procedures
The incidence of extrahepatic CC also differs across the globe. In the USA, the age-adjusted incidence of extrahepatic CC has been reported to be 1.2/100 000 for men and 0.8/100 000 for women. As mentioned above, the overall incidence of extrahepatic CC is slightly decreasing (Figure 60-1). Moreover, the age-adjusted mortality rates of extrahepatic CC are declining in western countries,5 with the exception of Italy and Japan. In the USA, age-adjusted mortality rates declined from 0.6/100 000 in 1979 to 0.3/100 000 in 1998.6 Indeed, there is evidence of minor improvements in 5year survival rates of extrahepatic CC from 11.7% in 1973–1977 to 15.1% in 1983–1987.7 Nevertheless, the small decline in extrahepatic CC in the USA is followed by a true increase of intrahepatic CC, which determines the overall rising incidence of CC.3
RISK FACTORS Most patients diagnosed with CC do not have or have not been exposed to known risk factor(s) associated with the disease. Nonetheless, Table 60-1 describes risk factors that have been associated with the development of CC. PSC is a definite risk factor for CC. The risk for developing CC in a patient with PSC is approximately 1.5% per year after diagno-
1134
sis of the cholestatic liver disease.8 Of interest, among the PSC patients who develop CC, about 30% will be found to have malignancy of the bile duct5 within 2 years of the diagnosis of PSC.9,10 Indeed, the risk of developing CC is not associated with the duration of PSC.9 Liver flukes, namely Opisthorchis viverrini and Clonorchis sinensis, are strongly associated with CC. These liver worms inhabit the bile ducts and sporadically the gallbladder. Individuals become infected with these parasites by eating undercooked fish. In patients with choledochal cysts (i.e., congenital cystic dilatation of bile ducts) the lifetime risk of developing CC is about 10–15%, and the median age of diagnosis is 34 years.11 Intrahepatic biliary stones (i.e., hepatolithiasis) are frequent in Asia but rare in western countries. Studies have reported the association of hepatolithiasis with peripherally located intrahepatic CC.12 Thorotrast is a colloidal suspension of 232ThO2, which mainly emits alpha-particles and was used as a contrast agent in radiology from the 1930s to 1950s. Exposure to Thorotrast has been linked to the development of CC. Thorotrast causes microsatellite instability and subsequently CC, probably via clonal expansion of cholangiocytes and inactivation of human Mut L homologue 1 (hMLH1) by hypermethylation.13 Recently, hepatitis C virus (HCV) and human immunodeficiency virus (HIV) have been added to the list of risk factors for development of CC.98 Common features among several of the risk factors of CC include chronic inflammation of the bile ducts and cholestasis. As discussed below both these events likely contribute to cholangiocyte malignant transformation of the cholangiocyte.
MOLECULAR PATHOGENESIS In the last decade, there has been considerable progress in understanding the pathogenesis of CC. In several instances, malignant transformation of cholangiocytes takes place amidst a milieu of chronic inflammation of the bile ducts and chronic cholestasis. This environment causes increased production of cytokines and reactive oxygen species, resulting in protracted cellular (i.e., cholangiocyte) stresses and accrual of irreversible DNA damage.1 Subsequently, cholangiocytes undergo malignant transformation, namely, attain molecular changes and develop cellular and subcellular characteristics that are otherwise lacking in normal conditions. The molecular alteration of cholangiocytes which leads to carcinogenesis is a multifaceted process of interrelated events. Nevertheless, Table 60-2 shows the proposed contributing pathways. The described molecular mechanisms of cholangiocarcinogenesis are mainly derived from experimental studies of intrahepatic CC because of the easy access to tissue. It remains uncertain, however, whether these pathogenetic pathways are pertinent to extrahepatic CC. For example, hilar CCs are highly desmoplastic and therefore obtaining adequate cells for studies is challenging. Under normal conditions, cholangiocytes retain tissue homeostasis despite exposure to proliferation signals. However, chronic biliary inflammation causes local interleukin-6 (IL-6) and hepatocyte growth factor (HGF) production, which are initially derived from periductal stromal cells (i.e., stellate cells). IL-6 is a powerful mitogen involved in cholangiocyte proliferation, as reported in both animal and human studies.14–17 IL-6 binds to its plasma membrane
Chapter 60 CHOLANGIOCARCINOMA
Table 60-2. Molecular Alterations of Cholangiocyte Malignant Transformation Contribution in carcinogenesis
Molecular mechanisms
References
Autologous proliferation signaling
IL-6, gp80/gp130 up-regulation HGF/c-met up-regulation EGF/c-erbB-2 COX-2 up-regulation K-ras mutations p53 mutations p21/WAF mutations Mdm-2 up-regulation p16 INK4 mutation FLIP up-regulation NO inhibition of caspases Bcl-2 up-regulation Bcl-XL up-regulation Mcl-1 up-regulation COX-2 up-regulation Telomerase expressed VEGF expressed E-cadherin decreased a-catetin and b-catetin decreased Matrix metalloproteinase (MMP) up-regulation Human aspartyl (asparaginyl) b-hydroxylase expression WISP1v expression
Sugawara et al.19 Yokomuro et al.,16 Lai et al.,21 Aishima et al22 Aishima et al.,22 Ito et al.,24 Collier et al.,27 Kiguchi et al26 Chariyalertsak et al.,28 Endo et al.,29 Yoon et al35 Kang et al.,31 Tannapfel et al.,32 Tada et al.33 Kang et al.31 Furubo et al.94 Furubo et al.94 Tannapfel et al.,32 Ahrendt et al.40 Que et al.42 Torok et al.43 Harnois et al.95 Okaro et al.96 Yoon et al.34 Nzeako et al.44 Itoi et al.47,48 Benckert et al.49 Ashida et al.97 Ashida et al.97 Terada et al.51
Loss of antigrowth signaling
Evasion of apoptosis
Unlimited replicative potential Angiogenesis Tissue invasiveness and metastasis
Lavaissiere et al.,52 Ince et al.,53 Maeda et al.54 Tanaka et al.55
IL-6, interleukin-6; HGF, hepatocyte growth factor; EGF, epidermal growth factor; COX-2, cyclooxygenase-2; WAF, wild-type p53 activated fragment 1; FLIP, flice-inhibitory protein; NO, nitric oxide; VEGF, vascular endothelial growth factor; WISP1v, WNT1-inducible signaling pathway protein 1. Modified from Berthiaume EP, Wands J. Molecular pathogenesis of cholangiocarcinoma. Semin Liver Dis 2004; 24:127–137.
receptor, forming the active heterodimer, gp80/gp130, which in turn stimulates cellular transcription through the mitogen-activated protein kinase (MAPK)/signal transducer and activator of transcription (STAT) pathway.18 Of interest, malignant, but not normal, cholangiocytes also produce high levels of IL-619 and overexpress the gp80/gp130 heterodimer. HGF also promotes cholangiocyte growth via its plasma membrane receptor, c-met.14,20 In addition, CC cells achieve an endogenous capacity to produce HGF and up-regulate its c-met receptor.16,21,22 Hence, through the IL-6 and HGF pathways malignant cholangiocytes maintain autologous proliferating mechanisms. Another mechanism that contributes to cholangiocarcinogenesis is the epidermal growth factor (EGF) and its receptor (EGFR).23,24 Interaction of EGF with EGFR leads to activation of the MAPK pathway.25 The c-erb-B2 protein, a homolog of the EGFR, is a tyrosine kinase which is activated in CC.22 Indeed, constitutive expression of c-erb-B2 in gallbladder epithelium leads to adenocarcinoma.26 Cyclooxygenase-2 (COX-2), an isoform that catalyzes the formation of prostaglandins from arachidonic acid, is induced by mitogens and cytokines27 and is involved in the pathogenesis of CC. COX-2 is overexpressed in malignant, but not normal, cholangiocytes.28,29 The complex and interrelated processes of carcinogenesis in bile ducts is indicated by the fact that IL-6, HGF, and EGF stimulate COX-2 expression in cholangiocytes.26,30 Nonetheless, the exact mechanism by which COX-2 causes CC is uncertain, but likely involves inhibition of apoptotic pathways. K-ras is an oncogene that plays an important role in mitogenic cellular signals. Mutations of this gene have been detected in 20–100%
of biopsy-proven CC.31,32 K-ras mutations have also been associated with hilar CC and periductal tumor extension.31,33 Further verification of these observations is needed because, if proven reliable, then detection of K-ras and p53 (see below) mutations may be incorporated in the clinical management of patients at high risk for developing CC (i.e., PSC). For example, PSC patients with specific mutations of K-ras and/or p53 may benefit from aggressive surveillance protocols for CC, development of chemopreventive treatments, and early orthotopic liver transplantation (OLT). In addition to biliary inflammation as a precipitant of cholangiocarcinogenesis, CC demonstrates an inherent tropism for bile. To this end, bile acids have been reported to transactivate the EGFR and to promote the expression of COX-2 in cholangiocytes.34,35 Critical pathways that inhibit cell proliferation are usually lost in the development of cholangiocarcinogenesis. For example, loss of heterozygocity for the tumor suppressor gene p53 is frequent in CC.36 The p53 gene directs the cellular machinery of cell cycle and apoptosis. Regarding the cell cycle, p53 regulates the p21/WAF1 (wild-type p53 activated fragment 1) protein which binds to the cell division kinase (CDK) 4–cyclin D complex (Figure 60-2). Thus, p53 causes negative feedback of the CDK4–cyclin D complex, therefore averting phosphorylation of Rb and, as a result, release of the E2F transcription factor (Figure 60-2). Subsequently, the E2F molecule controls the transcription of multiple cellular proteins important in the S-phase of cell cycle.37,38 Concerning apoptosis, p53 can induce apoptosis by promoting Bax insertion in the mitochondrial membrane and stimulating mitochondrial depolarization and subsequently apoptosis (Figure 60-2). Moreover, inactivation of the
1135
Section X. Tumors of the Liver
cyc D
CDK4 p21
p16 cyc
CDK4
mdm-2
Rb E2F
p53
Rb
P E2F
Bax Bax
Bcl-2
Bcl-2 Apoptosis
Cell cycle
Figure 60-2. p53 regulates both cholangiocyte cell cycle and apoptosis. p53 controls the cell division kinase (CDK)4–cyclin D complex via the p21/WAF1 protein which binds to the latter. As a result, p53 causes negative feedback of the CDK4–cyclin D complex, therefore averting phosphorylation of Rb and leading to release of the E2F transcription factor. Subsequently, E2F regulates the transcription of multiple proteins in the S-phase of the cell cycle. p53 can also induce apoptosis by promoting Bax insertion into the mitochondrial membrane, stimulating mitochondrial depolarization and resulting in apoptosis. The p16 protein also inhibits the CDK4–cyclin D complex. (Modified from Berthiaume EP, Wands J. Molecular pathogenesis of cholangiocarcinoma. Semin Liver Dis 2004; 24:127–137.)
p14/mdm/p53 pathway and the p16 (a tumor suppressor gene) via a variety of molecular mechanisms has been described in CC and in PSC-associated CC.39,40 As shown in Table 60-2, p53, p21/WAF1, and p16 mutations are involved in loss of critical cell signaling which may permit the development of CC. Apoptosis (i.e., programmed cell death) is an important cellular mechanism that controls tissue homeostasis. Disarrangement of apoptosis may lead to aberrant cell proliferation and subsequently to carcinogenesis. Ligand activation of the Fas/TRAIL (tumor necrosis factor [TNF]-related apoptosis-inducing ligand)/TNF-receptor family or release of cytochrome c by mitochondria causes activation of caspases, resulting in DNA fragmentation and cell destruction.41 As mentioned above, both chronic biliary inflammation and chronic cholestasis cause persistent cholangiocyte stress and DNA damage. Apoptosis therefore serves as a scavenger for cells that develop malignant transformation. Thus, loss of the protective function of apoptosis in cholangiocytes may result in development of CC. Cholangiocytes express the Fas receptor on plasma membrane30 and respond to FAS ligand stimulation with apoptosis. Nevertheless, in a CC cell line there was diminished responsiveness of the Fas/FAS ligand due to alteration of the flice-inhibitory protein (FLIP).30 FLIP inhibits activation of procaspase 8, causing diminished signaling of the Fas/FAS ligand pathway.42 Moreover, cholangiocytes under the effect of nitric oxide (NO) display inhibition of both caspase 3 and 9, likely via nitrosylation, and become relatively resistant to apoptosis.43 Overall, it is the balance of pro- and antiapoptotic signals in the cholangiocyte that guides the depolarization of the mitochondrial membrane, releasing cytochrome c into the cytosol and then the activation of caspases. In addition, COX-2 induced by inflammation is antiapoptotic. Up-regulation of COX-2 inhibits Fas/FAS ligand-induced apoptosis
1136
by increasing expression of the inhibitory protein myeloid cell leukemia-1 (Mcl-1).44 Of interest, bile acids which are elevated in CC have a positive effect on the expression of Mcl-1 protein via inhibition of proteosome degradation.34 Besides bile acids, inflammatory mediators up-regulate Mcl-1. As a result, potential malignant cholangiocytes are averted from apoptosis. CCs are able to grow perpetually. In contrast, normal cholangiocytes, like other types of cells, are undergoing a defined number of cellular divisions prior to undergoing senescence. Key to this process is progressive telomere shortening. Telomeres are long stretches of repeat sequences present at the end of chromosomes that are involved in DNA synthesis. After multiple cell cycles, telomere shortening causes chromosomal instability and renders cells unable to divide.45 Preservation of telomere shortening via overexpression of the human telomerase reverse transcriptase (hTERT) allows cancers, including CC, to sustain chromosomal replication and therefore to maintain eternal proliferation.46 In fact, detectable hTERT activity and increased expression of hTERT mRNA have been reported in intrahepatic CC.47,48 Angiogenesis is promoted by many malignancies to ensure the adequate blood supply of oxygen and nutrients to constantly dividing tumor cells. CCs have a high vascular supply and intrahepatic CCs express vascular endothelial growth factor (VEGF).49 Moreover, in a CC cell line, increased expression of VEGF was dependent on transforming growth factor-b stimulation.49 A feature of neoplasia is invasiveness of the surrounding tissue and metastasis. E-cadherin, a cell surface protein, is involved in cellular adhesion. Mutations or deletions of the gene cause diminished cell adhesiveness, facilitating tissue invasion and metastasis. Intrahepatic CCs have reduced expression of E-cadherin associated with advanced tumor histologic stage.50 Matrix metalloproteinases (MMP) have been reported to be up-regulated in CC, an event which was associated with clinical invasiveness.51 Moreover, the human aspartyl b-hydroxylase (HAAH), a protein involved in tumor invasion, is expressed in hepatocellular carcinoma and CC.52 HAAH involves post-transcriptional hydroxylation of b-carbons on specific residues present in EGF-like domains of proteins that participate in cell migration and motility.52 Expression of HAAH in transfected cell lines was associated with anchorage-independent growth and tumor development in nude mice.53 CC cell lines have been reported to overexpress HAAH.54 Another protein associated with connective tissue growth factor family, termed WISP1v (WNT1inducible signaling pathway protein 1), is also overexpressed in CC.55 Overexpression of WISP1v was linked with lymphatic and perineural CC invasion.55
PATHOLOGY AND CLASSIFICATION Microscopically, both intra- and extrahepatic CCs are adenocarcinomas. The commonest histological appearance is a well- to moderately differentiated tubular adenocarcinoma within a prominent, dense, desmoplastic stroma. In addition to tubular adenocarcinoma, other histologic variants of CC include papillary adenocarcinoma, signet-ring carcinoma, squamous cell or mucoepidermoid carcinoma, and a lymphoepithelioma-like form.
Chapter 60 CHOLANGIOCARCINOMA Figure 60-3. The term cholangiocarcinoma (CC) refers to tumors involving the entire (i.e., intra- and extrahepatic) biliary tree. Intrahepatic CC denotes malignancies affecting the bile ducts inside the liver. Extrahepatic CCs are divided into hilar or Klatskin tumor, middle, and distal tumors. Common hepatic duct Hilar CC
Gallbladder Common bile duct
Middle CC
Extra-hepatic CC
Distal CC
Type I
Type IIIb
Type II
Type IIIa
Figure 60-4. Type I cholangiocarcinoma (CC) affects the common hepatic duct; type II CC involves the common hepatic duct and the confluence of the right and left hepatic ducts; type IIIa and IIIb CC includes the common hepatic duct and either the right or left hepatic duct, respectively; and type IV CC involves the biliary confluence and extends to both right and left hepatic ducts or refers to multifocal bile duct tumors.
Type IV
Macroscopically, CC is classified into intra- and extrahepatic (Figure 60-3). The extrahepatic variety constitutes about two-thirds of all CCs and is subdivided into: (1) hilar or Klatskin tumor; (2) middle; and (3) distal tumors. Hilar tumors represent approximately 60% of extrahepatic CC. The Bismuth–Corlette classification of hilar CC is shown in Figure 60-4. Based on gross appearance, extrahepatic CC is grouped into sclerosing, nodular, and papillary. The sclerosing type is the most common and causes annular thickening of the bile ducts due to infiltration and fibrosis of the periductal tissues. Intrahepatic CC grows as a mass lesion, accounts for about one-third of bile duct tumors, and can be confused with hepatocellular carcinoma. Intrahepatic CC may be solitary or multinodular. It may also be well demarcated as a mass lesion or as a diffuse, infiltrating neoplasm growing along the intrahepatic bile ducts.
CLINICAL PRESENTATION AND DIAGNOSIS EXTRAHEPATIC CC Patients with extrahepatic CC present with symptoms, signs, and biochemical laboratory tests of obstructive cholestasis. They often complain of jaundice, dark urine, pale stools, pruritus, malaise, and weight loss. Laboratory tests reveal an increased alkaline phosphatase and bilirubin. The serum level of CA19-9 can be elevated. CA19-9 is the most commonly tested marker for pancreatobiliary malignancies.56 CA19-9 detects circulating high-molecular-weight mucin glycoproteins coated with sialylated blood group epitopes (i.e., sialyl Lewis).57 Therefore, CA19-9 blood levels depend on the Lewis phe-
1137
Section X. Tumors of the Liver
notype.58 Indeed, approximately 7% of the population are Lewisnegative, and thus, these individuals will have an undetectable CA199 in the presence of malignancy.59 CA19-9 is not specific for CC. In fact, pancreatic, gastric, colorectal, and gynecological cancers as well as bacterial cholangitis, and smoking can cause elevation of CA19-9. In extrahepatic CC, imaging studies demonstrate dilatation of the biliary tree and often localize the level of biliary obstruction. Unilobular bile duct obstruction usually causes atrophy of the affected lobe along with hypertrophy of the non-affected lobe, a phenomenon known as the atrophy–hypertrophy complex.60 Endoscopic retrograde cholangiopancreatography (ERCP) is used to define the position and extent of CC along the biliary tree. In addition, brush cytology and endoscopic biopsies of the bile ducts can be obtained during the procedure for histopathological diagnosis. An ERCP of a patient with a hilar CC is shown in Figure 60-5. Nevertheless, tissue-proven diagnosis can be challenging because CC is a highly desmoplastic tumor consisting of excessive fibrous tissue with few aggregations of malignant cholangiocytes. This desmoplastic reaction surrounds the bile ducts and extends into the submucosal tissue, making the diagnosis of CC based on positive biliary cytology findings possible only in ~30% of cases.2 Combined brush cytology and endoscopic biopsy may increase the yield of positive findings for CC ~40%.2 However, novel approaches, including single-cell techniques such as digitized image analysis (DIA) and fluorescence in situ hybridization (FISH), are promising methods to assess cellular aneuploidy and chromosomal duplication in CC.61 DIA generates a digital image of the cell nucleus with the assistance of a camera that captures pictures of light as it is transmitted via a slide specimen. This laboratory tool permits cellular quantification of DNA content, chromatin distribution, and nuclear morphology. DIA is feasible for
specimens of limited cellularity compared to flow cytometry.62 In a prospective study, DIA was compared to routine brush cytology for identification of malignancy in suspicious biliary tract strictures.64 DIA was significantly more sensitive than routine brush cytology (39.3% versus 17.9; P = 0.014). The specificity of DIA and cytology was 77.3% and 97.7%, respectively (P = 0.003). The lower specificity of DIA in this study was attributed to the high proportion of patients with PSC.64 Nevertheless, the accuracy of DIA was comparable to cytology (56% versus 53%). FISH utilizes fluorescently labeled DNA probes to detect cholangiocytes with chromosomal alterations (Figure 60-6). The presence of significant populations of cells with chromosomal gains indicates the possibility of biliary malignancy. A positive test is defined when five or more cells display gains of two or more chromosomes, or 10 or more cells demonstrate a gain of a single chromosome. A positive biliary FISH study does not identify the position or type of bile duct malignancy. To perform a biliary FISH study, bile duct brushing and aspirate specimens are collected at the time of ERCP and cells are fixed on a slide. Then four fluorescently labeled DNA probes hybridize to the centromere of chromosomes 3, 7, and 17 and the p16 gene on chromosome 9 (9p21). After hybridization, the slide is stained with the nuclear counterstain 4¢,6-diamidino-2phenylindole (DAPI) and fluorescence microscopy is used to scan the slide for atypical cells (i.e., gains of chromosomes 3, 7, 9 and 17) (Figure 60-6). If the number of cells with chromosomal gains (i.e., polysomy or trisomy) observed is adequate to declare the test positive, then the percentage of abnormal cells is calculated. In addition to biliary DIA and FISH studies, endoscopic ultrasound (EUS) may have applicability in determining the nature of biliary strictures. In a recent study, EUS-guided fine-needle aspiration biopsy of suspected CC demonstrated specificity, sensitivity, and positive predictive value of 86%, 100%, and 100%, respectively.65 Overall, EUS with fine-needle aspiration was reported to have a positive effect on the clinical management of 84% patients with CC.65 The contribution of magnetic resonance imaging (MRI) in CC is to aid the diagnosis and to evaluate resectability. In hilar CC, magnetic resonance cholangiopancreatography (MRCP) images may show moderately irregular thickening of the bile duct wall associated with proximal biliary dilatation.66 In clinical practice, it is not uncommon to make the diagnosis of extrahepatic CC based solely on clinical, laboratory, and imaging findings in the absence of tissue-proven diagnosis. A pragmatic challenge is the diagnosis of CC in patients with PSC. Indeed, the patient may present with a dominant biliary stricture, which is difficult to differentiate between a benign lesion and CC. In a patient with PSC, sudden and unexpected clinical deterioration, coupled with progressive elevation of alkaline phosphatase and serum CA19-9 values greater than 100 U/mL, in the absence of bacterial cholangitis, strongly indicates the development of CC. Figure 60-7 provides an algorithm for evaluation of these patients.
INTRAHEPATIC CC Figure 60-5. An endoscopic retrograde cholangiopancreatography demonstrating a hilar cholangiocarcinoma causing obstruction of the biliary bifurcation.
1138
Intrahepatic CC presents with non-specific symptoms and signs of a liver mass lesion, such as abdominal pain, anorexia, weight loss, malaise, and night sweats. An incidental abdominal mass during
Chapter 60 CHOLANGIOCARCINOMA
A
B
Figure 60-6. Fluorescently labeled DNA probes decorating genomic loci on four separate chromosomes. Red color probe indicates chromosome 3, green color probe specifies chromosome 7, gold color probe points to chromosome 9, and aqua color identifies chromosome 17. Normal cholangiocytes (panel A) have two duplicates of each probe, as expected for normal diploid cells. Malignant cholangiocytes (panel B) reveal gains of chromosomal probes, suggesting polysomy.
ERCP (obtain biliary brushings for standard cytology, DIA, FISH and biliary biopsies)
Positive cytology and histopathology findings
Negative cytology and histopathology findings
MR imaging Liver mass tumor, bile ducts dilation, vascular encasement
No liver mass, normal bile ducts and hepatic vessels
PET scan
Diagnosis of CC
Abnormal
Normal
“Hot spot”
Observation
Figure 60-7. An algorithm for the management of a primary sclerosing cholangitis patient presenting with a dominant biliary stricture. ERCP, endoscopic retrograde cholangiopancreatography; DIA, digitized image analysis; FISH, fluorescence in situ hybridization; OLT, orthotopic liver transplantation; CC, cholangiocarcinoma.
1139
Section X. Tumors of the Liver
physical exam or imaging study may be the sole finding in asymptomatic patients. Laboratory tests reveal an elevated alkaline phosphatase with a normal bilirubin. CA19-9 is likely increased.2 The diagnosis of intrahepatic CC is commonly made by the exclusion of other primary or metastatic hepatic mass lesions, which in many occasions could imitate the former. Using cross-sectional MR, intrahepatic CC appears hypo- and hyperintense on T1- and T2weighted images, respectively. At times, liver biopsy of the mass lesion is the definitive approach to make the correct diagnosis.
STAGING Staging of CC is necessary to identify potential candidates for surgical resection. Tables 60-3 and 60-4 describe the tumor node metastasis (TNM) classification for intra- and extrahepatic CC, respectively. Nevertheless, the value of the TNM categorization system for extrahepatic CC is limited. This is because the TNM system relates to histopathology but not to clinical extension of the disease, and the latter is more important in making decisions for sur-
gical resection of the tumor (Table 60-5). It is therefore critical that, during the clinical staging of extrahepatic CC, the proximal and distal margins of the tumor are clearly identified. Obtaining highquality imaging studies of the bile ducts and liver can attain this aim. ERCP, MRCP, and percutaneous transhepatic cholangiography (PTC) can be used, at times in combination, to map the tumor boundaries. Second, it is crucial to rule out vascular encasement of the contralateral (i.e., non-affected) liver lobe by the CC before committing to partial hepatectomy as well as to verify patency of the portal vein and hepatic artery. Third, regional metastases should be excluded. To this end, EUS is superior compared with conventional cross-sectional abdominal imaging (i.e., computed tomography, MRI) to exclude metastatic disease. This is particularly important for doubtful-appearing regional lymph nodes, which can be biopsied during EUS to rule out metastatic disease. Indeed, 15–20% of CC patients with unremarkable abdominal imaging studies have metastatic lymph node involvement based on EUS evaluation.61
THERAPY Table 60-3. Tumor Node Metastasis (TNM) Pathological Classification of Intrahepatic Cholangiocarcinoma Stage
Tumor
Node
Metastasis
I II III A III B III C IV
T1 T2 T3 T4 Any T Any T
N0 N0 N0 N0 N1 Any N
Mo M0 M0 M0 M0 M1
T1, solitary tumor without vascular invasion; T2, solitary tumor with vascular invasion or multiple tumors, none more than 5 cm; T3, multiple tumors more than 5 cm or tumor involving a major branch of the portal or hepatic vein(s); T4, tumor(s) with direct invasion of adjacent organs other than gallbladder or with perforation of visceral peritoneum; N0, no regional lymph node metastasis; N1, regional lymph node metastases; M0, no distant metastasis; M1, distant metastasis. Greene FL, ed. American Joint Committee on Cancer Staging. 6th edn. NewYork: SpringerVerlag; 2002.
Table 60-4. Tumor Node Metastasis (TNM) Pathological Classification of Extrahepatic Cholangiocarcinoma Stage
Tumor
Node
Metastasis
0 IA IB II A II B III IV
Tis T1 T2 T3 T1–T3 T4 Any T
N0 N0 N0 N0 N1 Any N Any N
Mo M0 M0 M0 M0 M0 M1
Tis , carcinoma in situ; T1, tumor confined to the bile duct histologically; T2, tumor invades beyond the wall of the bile duct; T3, tumor invades the liver, gallbladder, pancreas, and/or unilateral branches of the portal vein (right or left) or hepatic artery (right or left); T4, tumor invades any of the following: main portal vein or its branches bilaterally, common hepatic artery, or other adjacent structures, such as the colon, stomach, duodenum, or abdominal wall; N0, no regional lymph node metastasis; N1, regional lymph node metastasis; M0, no distant metastasis; M1, distant metastasis. Greene FL, ed. American Joint Committee on Cancer Staging. 6th edn. NewYork: SpringerVerlag; 2002.
1140
Surgical resection is potentially curative treatment for both intraand extrahepatic CC. Nevertheless, for all intrahepatic and most extrahepatic CCs, complete surgical removal of the malignancy necessitates major hepatectomy. To this end, many patients do not qualify for surgery because of other comorbidities. Unfortunately, more than half of CC patients present with advanced unresectable disease. In such cases, palliative therapies (i.e., biliary stenting, PDT) provide symptom relief and possibly have a positive effect on survival, as it was recently reported for PDT. To date, chemo- and/or radiation therapy for CC have not been evaluated in randomized controlled trials, including adjuvant treatment after surgical resection to diminish the risk of CC recurrence.2 A small fraction of selected patients with hilar CC may undergo OLT with curative intent at selected liver transplant centers. An algorithm for the overall management of CC is provided in Figure 60-8.
Table 60-5. Proposed, Preoperative T-Stage Criteria for Hilar Cholangiocarcinoma Stage
Criteria
T1
Tumor involving biliary confluence ± unilateral extension to second-order biliary radicles Tumor involving biliary confluence ± unilateral extension to second-order biliary radicles and ipsilateral portal vein involvement ± ipsilateral hepatic lobar atrophy Tumor involving biliary confluence + bilateral extension to second-order biliary radicles; or unilateral extension to secondorder biliary radicles with contralateral portal vein involvement; or unilateral extension to second-order biliary radicles with contralateral hepatic lobar atrophy; or main or bilateral portal vein involvement
T2
T3
Modified from Jarnagin WR, Fong Y, DeMatteo RP, et al. Staging, resectability, and outcome in 225 patients with hilar cholangiocarcinoma. Ann Surg 2001; 234:507–517; discussion 517–519.
Chapter 60 CHOLANGIOCARCINOMA Figure 60-8. An overview for the clinical management of cholangiocarcinoma. OLT, orthotopic liver transplantation.
Diagnosis of CC
Intra-hepatic
Extra-hepatic
Non-resectable
Resectable
Palliative therapies (preferably within the context of randomized clinical trials) • Radiotherapy • Chemotherapy
Non-resectable (Consider OLT only for hilar CC at selected liver transplant centers)
Surgical resection
Post resection Consider neoadjuvant/ adjuvant therapies (preferably within the context of randomized clinical trials) • Radiotherapy • Chemotherapy
No OLT
Cholestasis
No cholestasis
Biliary stenting
Palliative therapies (preferably within the context of randomized clinical trials) • Radiotherapy • Chemotherapy
SURGICAL THERAPY Extrahepatic CC Most extrahepatic CCs involve the biliary bifurcation (i.e., hilar tumors). CC implicating the distal bile ducts usually requires pancreaticoduodenectomy. In the section below we discuss exclusively surgical therapy for hilar tumors. Staging of patients with extrahepatic CC is key prior to consideration for surgical resection. Jarnagin and colleagues have proposed clinical staging criteria for hilar CC (Table 60-5). These criteria do correlate with tumor resectability (i.e., 60% in stage T1 and 0% in stage T3) and patient survival.67 The evaluation for resectability of extrahepatic CC demands appropriate patient selection and careful assessment of imaging studies. During this vigorous assessment process, about one-third of patients will be deemed unresectable. Yet, 25–30% of those who had been felt to be good candidates for surgical resection will be found to have unresectable CC during laparotomy.68 To this extent, laparoscopy prior to resection of extrahepatic CC has become the standard surgical approach. In general, patients with resectable extrahepatic CC require partial hepatic resection to have tumor-free margins. In fact, cases with positive surgical margins have survival comparable to those receiving palliative therapy alone.67 Patients with tumor-free margins have a 20–40% 5-year survival rate.67,69 Therefore, the primary aim of surgical resection is biopsy-proven negative margins. To achieve this goal for extrahepatic CC, however, resection of the tumor/extrahepatic bile ducts and subhilar lymphadenopathy is usually needed.69 Indeed, concurrent en bloc partial hepatectomy is
associated with higher degree of negative resection margins.67,70 Of note, extrahepatic CC involving the biliary confluence almost always engages the main caudate duct and demands caudate lobe removal.71 Resection of extrahepatic CC is major surgery, with 5–10% mortality and notable morbidity, even at medical centers with significant expertise.69 Postoperative mortality is dominated by infective complications.67 Regional lymph node metastases are associated with reduced 3- and 5-year survival.72 However, it is not clear whether extended lymph node resection improves patient survival. In addition to curative intent, surgery can be used to palliate obstructive jaundice in extrahepatic CC. This goal can be achieved by choledochojejunostomy or hepaticojejunostomy. Nonetheless, improvements of current endoscopic modalities (i.e., biliary stenting, PDT, brachytherapy) and the high cost as well as morbidity of surgical approaches have rendered the former more favorable for palliation of jaundice.
Intrahepatic CC Surgery is also the best option for treatment of intrahepatic CC. In general, intrahepatic CCs are large tumors at the time of diagnosis and patients need major liver resections (i.e., right or left hepatectomy). The prognostic factors indicating poor outcome after surgical resection of intrahepatic CC are shown in Table 60-6. Although tumor metastasis to regional lymph nodes predicts survival, the effect of surgical node dissection on survival is unclear. Following surgical resection of intrahepatic CC, the median and 5-
1141
Section X. Tumors of the Liver
Table 60-6. Prognostic Factors Associated with Unfavorable Outcome after Surgical Treatment of Intrahepatic Cholangiocarcinoma (CC) Preoperative CA19-9 levels >1000 U/mL Multifocal disease Liver capsule invasion Lack of cancer-free surgical margins Regional lymph node metastases Mass-forming or periductal infiltrating-type CC growth Expression of mucin 1 (MUC1) by CC cells
year survival rate range from 12 to 28 months and from 29% to 36%, respectively.
direction to the endoscopist in choosing the optimal bile ducts for stenting and in avoiding atrophied hepatic segments, thus minimizing the risk of postprocedure cholangitis.75 Performing an MRCP prior to ERCP not only improves the success rate of biliary stenting but also affects positively the resolution of jaundice.76 Both plastic and metallic biliary stents have been used to alleviate extrahepatic CC. Plastic stents are of smaller diameter and become occluded more easily compared to self-expandable metal stents.73 The latter are also cost-effective in patients with CC who survive for at least 3–6 months. Using MRCP guidance prior to ERCP to improve the success of unilateral biliary drainage, biliary metal stents will likely become the choice for palliation of obstructive jaundice in CC when the expected survival is longer than 6 months.
Palliative Therapeutic Approaches Many patients with intra- or extrahepatic CC present with unresectable disease. Moreover, because CC usually affects the elderly, a notable proportion of surgical candidates have other comorbidities which preclude them from surgical tumor resection. Palliative therapy focuses on relief of obstructive jaundice, which improves patient symptoms with less likely expansion in survival. Surgical bypass approaches aimed to offer palliation of jaundice have been successful but have high operative mortality, morbidity, and cost. Regarding the non-surgical candidate, palliative therapies for obstructive jaundice include: (1) biliary stenting; (2) PDT; (3) intraluminal brachytherapy; and (4) high-intensity intraductal ultrasound. Although the non-surgical approaches to treat obstructive jaundice can be performed endoscopically or percutaneously, the following discussion will mainly concentrate on endoscopy-based therapies.
Biliary Stents Amelioration of obstructive jaundice in CC improves patient symptoms and quality of life but not survival. Biliary stents are an effective modality in relieving malignant bile duct obstruction with subsequent decline of jaundice. Studies have shown that biliary drainage of only 25–30% of the hepatic parenchyma is required to accomplish adequate palliation of obstructive jaundice. In patients with CC there are a number of issues to be considered prior to palliative therapy of obstructive jaundice using endoscopic biliary stents.73 Host parameters (i.e., patient life expectancy, position and degree of biliary obstruction, hepatic lobe atrophy) and the appropriate use of endoscopic biliary stents (i.e., plastic versus metallic; single versus double drainage) are important issues to consider in optimizing treatment.73 In general, the type and number of biliary stents and endoscopic approach to treat obstructive jaundice in CC have to be individualized. For example, a jaundiced patient with Bismuth type I hilar CC can be successfully palliated with a single biliary stent. However, there is a lack of consensus among experts for biliary stenting of hilar CC Bismuth types II, III, and IV. In a prospective, randomized controlled trial of patients with hilar CC (Bismuth I–III), it was concluded that unilateral drainage was adequate to alleviate biliary obstruction and that efforts to place a second biliary stent may result in early complications (i.e., bacterial cholangitis) without any survival benefits.74 In current practice, MRCP is utilized to assist with the disease diagnosis and the endoscopic placement of biliary stents to relieve malignant obstruction. Obtaining an MRCP prior to ERCP offers
1142
Photodynamic Therapy This therapy involves the systemic preadministration of a non-toxic, photosensitizer drug, which accumulates principally within the CC. Subsequently, the patient undergoes ERCP to activate the photosensitizer via direct non-thermal laser application on CC. During activation, the drug reaches an excited reactive state (triplet state) and subsequently, energy is transferred from the triplet state of the photosensitizer to molecular oxygen. This event creates singlet molecular oxygen (1O2), which results in direct or indirect photodamage of the targeted tissue. The photosensitizer agent localizes to mitochondria and induces apoptosis of malignant cholangiocytes and other supporting tissues, resulting in tumor regression. There are several types of photosensitizer. The most commonly used are derivatives of hematoporphyrin, namely, porfimer sodium, Photofrin, HpD, and Photosan II. Sodium porfimer is given intravenously at 2 mg/kg. Studies have shown that porfimer enrichment in CC is adequate for employing PDT between days 1 and 4 following intravenous administration. Intraluminal photoactivation is achieved by laser light (wavelength: 630 nm; light dose: 180 J/cm2). The tumoricidal depth penetration of PDT with porfimer is about 4–6 mm. A plethora of parameters determine the depth and extent of tissue damage, including the kind and quantity of photosentizer used, the oxygen concentration in the affected tissue, and the intensity, absorption, and distribution of activating laser light. Indications and contraindications of PDT are shown in Table 60-7. Following injection of porfimer, patients can tolerate artificial room light; however, they must keep out of bright direct or indirect sunlight to prevent phototoxicity of the skin. PDT can produce mild to moderate epigastric pain for up to 3 days after the ERCP. A temporary elevation of aspartate aminotransferase and leukocytosis is expected 2–3 days after PDT. Complications of PDT include biloma and hemobilia. Biliary perforations have not been reported and the rate of cholangitis is not increased following PDT compared with biliary stents placement alone. The 30-day mortality following intraductal PDT has been reported as 2% due to two fatal episodes of pulmonary embolism, which were likely related to paraneoplastic thromboses. Patients should follow specific instructions to “bleach out” the accumulated porfimer in the skin. This goal can be achieved by stepby-step short exposures (5–10 minutes) of the skin after day 4 to mild evening sunlight prior to sunset. If this exposure is endured without skin sunburn, them slow progression of re-exposure is suggested until bright sunlight can be tolerated.
Chapter 60 CHOLANGIOCARCINOMA
Table 60-7. Indications and Contraindications of Photodynamic Therapy for Hilar Cholangiocarcinoma (CC) Indications Indications Preliminary indication: non-resectable hilar CC with unrelieved cholestasis Relative indications (i.e., within clinical trials) Non-resectable hilar CC with successful biliary drainage Inoperable comorbid patient with resectable hilar CC Borderline resectability of hilar CC (neoadjuvant photodynamic therapy for purging of intrahepatic ducts from tumor cells beyond the tumor margins) Contraindications Porphyria (all genetic types) Recent use of photosensitizing or dermatotoxic drugs (e.g., bleomycin) Insertion of a coated metal stent Severe hepatic or renal failure Relative contraindications Peritoneal carcinomatosis (cholestasis palliated) Karnofsky performance status < 30% Biliary empyema or liver abscess Modified from Berr F. Photodynamic therapy for cholangiocarcinoma. Semin Liver Dis 2004; 24:177–187.
Pilot non-randomized studies of PDT for unresectable CC have demonstrated promising results on: (1) prompt improvement of cholestasis; (2) stabilization of the Karnofsky performance status at almost normal or moderately diminished rates; and (3) improvement or preservation of quality of life.77 Recently, a multicenter prospective randomized trial evaluated in patients with unresectable CC the effect of biliary stenting followed by PDT (group A) compared with biliary stenting alone (group B).78 Patients in group A demonstrated prolonged survival (n = 20, median survival 493 days; 95% confidence index, 276–710) compared to patients in group B (n = 19, median survival 98 days; 95% confidence index, 87–107) (P < 0.0001).78 Of interest, this study failed to improve biliary obstruction and jaundice in the second group (sole bile duct stenting), as expected. Thus, it is likely that the survival benefit reported in the first group (biliary stenting and PDT) relates to amelioration of cholestasis rather than tumor burden.79 More prospective randomized trials on PDT for unresectable CC – including patients with concurrent PSC – are needed to assess further the utility of this promising approach.
Intraluminal Brachytherapy In the course of intraluminal brachytherapy, premounted 192iridium seeds are deployed within a cathether and placed across the neoplastic biliary stricture(s) during ERCP or PTC.80,81 The assumption is that brachytherapy permits focal, higher, and more effective doses of radiation compared to external-beam radiation therapy (EBRT). Therefore, it is anticipated to extend the palliation of obstructive jaundice and to avoid unnecessary radiation damage of the surrounding tissues/organs. To date, the results of intraluminal brachytherapy are unclear.81,82 It is apparent that more studies are needed to evaluate this palliative approach.
High-Intensity Intraductal Ultrasound This is another novel approach for palliative therapy of CC. During endoscopy, high-intensity intraductal ultrasound induces local necro-
sis of the malignant tissue (i.e., biliary obstruction/stricture).83 Although there were promising results in a pilot study, more clinical trials are necessary to assess this therapeutic method.83
Liver Transplantation for Unresectable Cholangiocarcinoma Former experience with OLT for CC has been disappointing.84–90 Recurrence of the malignancy was very frequent and the 5-year survival rates were only 5–15%. As a result, most liver transplant centers consider CC a contraindication for OLT. Nevertheless, a selected group of patients who underwent OLT and who had negative surgical resection margins and negative regional lymph nodes demonstrated long-term survival.91 Moreover, in a small number of patients who were treated with radiation plus brachytherapy plus 5fluorouracil, the observed 5-year survival rate was 22%.92 As a result of these favorable observations, we developed an experimental liver transplant protocol for therapy of selected patients with early-stage unresectable hilar CC or CC arising in the background of PSC. For patients to be enrolled to the protocol they have to be diagnosed with CC based on either biopsy, biliary brush cytology or aneuploidy of biliary cell epithelia (i.e., cholangiocytes). If the malignancy forms a mass lesion, the diameter should be less than 3 cm on cross-sectional imaging studies. Moreover, the CC have to be unresectable after clinical, laboratory, and imaging evaluation by an experienced hepatobiliary team. Tumor vascular encasement causing absence of blood flow without evidence of vessel invasion is not a contraindication to enrollment in this protocol. Enrolled patients have to be suitable for radiation therapy, chemotherapy, and liver transplantation, as determined by the interdisciplinary team. Patients who meet eligible criteria receive neoadjuvant chemo- and radiation therapy. EBRT (at a total dose of 4500 cGy in 30 sessions) is completed in 3 weeks. 5-fluorouracil is given intravenously at 500 mg/m2 daily as bolus for three consecutive days in the beginning of EBRT. ERBT can cause nausea, vomiting, leukopenia, cholangitis, gastrointestinal ulceration, and liver abscess. Following completion of ERBT, patients receive brachytherapy of the tumor using a transcatheter loaded with 192iridium seeds (total dose of 2000–3000 cGy). In the course of brachytherapy they are also treated with 5-fluorouracil at 225 mg/m2 daily. Subsequently, patients continue to receive the above dose of 5fluorouracil or capecitabine (2000 mg/m2 daily) for 2 out of 3 weeks until liver transplantation is performed. Once neoadjuvant chemoradiation therapy is finished, patients undergo staging laparotomy, which includes biopsy of regional hepatic lymph nodes. Patients with negative staging proceed with OLT. Cadaveric livers or living donor right liver grafts can be used. Transplanted patients receive standard immunosuppression therapy. Between 1993 and 2003, 56 patients with CC were enrolled in this liver transplant protocol at the Mayo Clinic in Rochester, MN. Forty-eight patients underwent staging laparotomy, of whom 14 (29%) were found to have progression of CC and were excluded from the protocol. During the decade (1993–2003), 28 patients have undergone OLT of whom 6 died following OLT and 22 are currently alive. The actuarial patient survival following liver transplantation was 88% at 1 year and 82% at 5 years. The time from enrollment to OLT was 4 (± 2.2) months and 10 (± 3.6) months
1143
Section X. Tumors of the Liver
during the first 5 and the second 5 years of the protocol, respectively. Following the above neoadjuvant chemoradiation protocol, the outcome of OLT in selected patents with CC is similar to liver transplantation for other chronic liver diseases. Moreover, the outcome of our liver transplant protocol for CC exceeds the outcome of surgical resection with curative intent, which is considered the gold-standard therapy for CC. A comparable liver transplant protocol for CC has been developed at the University of Nebraska, which has also shown a favorable outcome regarding patient survival.93
Future Therapeutic Directions At present, no survival benefit has been attained in patients with CC treated with chemotherapy and/or radiation therapy.1,2 Randomized, controlled clinical trials are needed to evaluate the usefulness of novel chemotherapeutic agents and/or radiation therapy. As the understanding of the biology of CC improves, it is expected that better pharmacological inhibitors will be developed and used against these tumors alone or in combination with existing and future chemotherapeutic agents. Given the increasing incidence of intrahepatic CC, we are in great need of such developments to improve the survival of patients afflicted with this devastating disease.
REFERENCES 1. Gores GJ. Cholangiocarcinoma: current concepts and insights. Hepatology 2003; 37: 961–969. 2. Khan SA, Davidson BR, Goldin R, et al. Guidelines for the diagnosis and treatment of cholangiocarcinoma: consensus document. Gut 2002; 51 (Suppl 6): VI1-9. 3. Shaib YH, Davila JA, McGlynn K, et al. Rising incidence of intrahepatic cholangiocarcinoma in the United States: a true increase? J Hepatol 2004; 40:472–477. 4. Jemal A, Murray T, Samuels A, et al. Cancer statistics, 2003. CA Cancer J Clin 2003; 53: 5–26. 5. Khan SA, Taylor-Robinson SD, Toledano MB, et al. Changing international trends in mortality rates for liver, biliary and pancreatic tumours. J Hepatol 2002; 37:806–813. 6. Patel T. Increasing incidence and mortality of primary intrahepatic cholangiocarcinoma in the United States. Hepatology 2001; 33:1353–1357. 7. Carriaga MT, Henson DE. Liver, gallbladder, extrahepatic bile ducts, and pancreas. Cancer 1995; 75 (Suppl):171–190. 8. Bergquist A, Broome U. Hepatobiliary and extrahepatic malignancies in primary sclerosing cholangitis. Best Pract Res Clin Gastroenterol 2001; 15: 643–656. 9. Broome U, Olsson R, Loof L, et al. Natural history and prognostic factors in 305 Swedish patients with primary sclerosing cholangitis. Gut 1996; 38:610–615. 10. Rosen CB, Nagorney DM, Wiesner RH, et al. Cholangiocarcinoma complicating primary sclerosing cholangitis. Ann Surg 1991; 213:21–25. 11. Simeone DM. Gallbladder and biliary tree: anatomy and structural anomalies. In Yamada T, ed: Textbook of gastroenterology. Philadelphia: Lippincott, Williams & Wilkins, 1999:2244–2257. 12. Chen MF, Jan YY, Wang CS, et al. A reappraisal of cholangiocarcinoma in patient with hepatolithiasis. Cancer 1993; 71:2461–2465.
1144
13. Liu D, Momoi H, Li L, et al. Microsatellite instability in thorotrast-induced human intrahepatic cholangiocarcinoma. Int J Cancer 2002; 102:366–371. 14. Matsumoto K, Fujii H, Michalopoulos G, et al. Human biliary epithelial cells secrete and respond to cytokines and hepatocyte growth factors in vitro: interleukin-6, hepatocyte growth factor and epidermal growth factor promote DNA synthesis in vitro. Hepatology 1994; 20:376–382. 15. Park J, Tadlock L, Gores GJ, et al. Inhibition of interleukin 6mediated mitogen-activated protein kinase activation attenuates growth of a cholangiocarcinoma cell line. Hepatology 1999; 30:1128–1133. 16. Yokomuro S, Tsuji H, Lunz JG 3rd, et al. Growth control of human biliary epithelial cells by interleukin 6, hepatocyte growth factor, transforming growth factor beta1, and activin A: comparison of a cholangiocarcinoma cell line with primary cultures of non-neoplastic biliary epithelial cells. Hepatology 2000; 32:26–35. 17. Yokomuro S, Lunz JG 3rd, Sakomoto T, et al. The effect of interleukin-6 (IL-6)/gp130 signaling on biliary epithelial cell growth, in vitro. Cytokine 2000; 12:727–730. 18. Heinrich PC, Behrmann I, Haan S, et al. Principles of interleukin (IL)-6-type cytokine signaling and its regulation. Biochem J 2003; 15:1–20. 19. Sugawara H, Yasoshima M, Katayanagi K, et al. Relationship between interleukin-6 and proliferation and differentiation in cholangiocarcinoma. Histopathology 1998; 33:145–153. 20. Boccaccio C, Gaudino G, Gambarota G, et al. Hepatocyte growth factor (HGF) receptor expression is inducible and is part of the delayed-early response to HGF. J Biol Chem 1994; 269:12845–12851. 21. Lai GH, Radaeva S, Nakamura T, et al. Unique epithelial cell production of hepatocyte growth factor/scatter factor by putative precancerous intestinal metaplasias and associated “intestinal-type” biliary cancer chemically induced in rat liver. Hepatology 2000; 31:1257–1265. 22. Aishima SI, Taguchi KI, Sugimachi K, et al. c-erbB-2 and c-Met expression relates to cholangiocarcinogenesis and progression of intrahepatic cholangiocarcinoma. Histopathology 2002; 40:269–278. 23. Harada K, Terada T, Nakanuma Y. Detection of transforming growth factor-alpha protein and messenger RNA in hepatobiliary diseases by immunohistochemical and in situ hybridization techniques. Hum Pathol 1996; 27:787–792. 24. Ito Y, Takeda T, Sasaki Y, et al. Expression and clinical significance of the erbB family in intrahepatic cholangiocellular carcinoma. Pathol Res Pract 2001; 197:95–100. 25. Schlessinger J. Ligand-induced, receptor-mediated dimerization and activation of EGF receptor. Cell 2002; 110:669–672. 26. Kiguchi K, Carbajal S, Chan K, et al. Constitutive expression of ErbB-2 in gallbladder epithelium results in development of adenocarcinoma. Cancer Res 2001; 61:6971–6976. 27. Collier JD, Guo K, Mathew J, et al. c-erbB-2 oncogene expression in hepatocellular carcinoma and cholangiocarcinoma. J Hepatol 1992; 14: 377–380. 28. Chariyalertsak CS, Sirikulchayanonta V, Mayer D, et al. Aberrant cyclooxygenase isozyme expression in human intrahepatic cholangiocarcinoma. Gut 2001; 48:80–86. 29. Endo K, Yoon BI, Pairojkul C, et al. ERBB-2 overexpression and cyclooxygenase-2 up-regulation in human cholangiocarcinoma and risk conditions. Hepatology 2002; 36:439–450. 30. Shimonishi T, Isse K, Shibata F, et al. Up-regulation of fas ligand at early stages and down-regulation of Fas at progressed stages of intrahepatic cholangiocarcinoma reflect evasion from immune surveillance. Hepatology 2000; 32:761–769. 31. Kang YK, Kim WH, Lee HW, et al. Mutation of p53 and K-ras, and loss of heterozygosity of APC in intrahepatic cholangiocarcinoma. Lab Invest 1999; 79:477–483.
Chapter 60 CHOLANGIOCARCINOMA
32. Tannapfel A, Benicke M, Katalinic A, et al. Frequency of p16INK4A alterations and k-ras mutations in intrahepatic cholangiocarcinoma of the liver. Gut 2000; 47:721–727. 33. Tada M, Omata M, Ohto M. High incidence of ras gene mutation in intrahepatic cholangiocarcinoma. Cancer 1992; 69:1115–1118. 34. Yoon JH, Werneburg NW, Higuchi H, et al. Bile acids inhibit Mcl-1 protein turnover via an epidermal growth factor receptor/Raf-1-dependent mechanism. Cancer Res 2002; 62:6500–6505. 35. Yoon JH, Higuchi H, Werneburg NW, et al. Bile acids induce cyclooxygenase-2 expression via the epidermal growth factor receptor in a human cholangiocarcinoma cell line. Gastroenterology 2002; 122:985–993. 36. Isa T, Tomita S, Nakachi A, et al. Analysis of microsatellite instability, K-ras gene mutation and p53 protein overexpression in intrahepatic cholangiocarcinoma. Hepatogastroenterology 2002; 49:604–608. 37. Evan GI, Vousden KH. Proliferation, cell cycle and apoptosis in cancer. Nature 2001; 411:342–348. 38. Levine AJ. p53, the cellular gatekeeper for growth and division. Cell 1997; 88:323–331. 39. Cong WM, Bakker A, Swalsky PA, et al. Multiple genetic alterations involved in the tumorigenesis of human cholangiocarcinoma: a molecular genetic and clinicopathological study. J Cancer Res Clin Oncol 2001; 127:187–192. 40. Ahrendt SA, Eisenberger CF, Yip L, et al. Chromosome 9p21 loss and p16 inactivation in primary sclerosing cholangitisassociated cholangiocarcinoma. J Surg Res 1999; 84:88–93. 41. Hengartner MO. The biochemistry of apoptosis. Nature 2000; 407:770–776. 42. Que FG, Phan VA, Phan VH, et al. Cholangiocarcinomas express Fas ligand and disable the Fas receptor. Hepatology 1999; 30:1398–1404. 43. Torok NJ, Higuchi H, Bronk S, et al. Nitric oxide inhibits apoptosis downstream of cytochrome C release by nitrosylating caspase 9. Cancer Res 2002; 62:1648–1653. 44. Nzeako UC, Guicciardi ME, Yoon JH, et al. COX-2 inhibits Fasmediated apoptosis in cholangiocarcinoma cells. Hepatology 2002; 35:552–559. 45. Buys CH. Telomeres, telomerase, and cancer. N Engl J Med 2000; 342:1282–1283. 46. Shay JW, Bacchetti S. A survey of telomerase activity in human cancer. Eur J Cancer 1997; 33:787–791. 47. Itoi T, Shinohara Y, Takeda K, et al. Detection of telomerase activity in biopsy specimens for diagnosis of biliary tract cancers. Gastrointest Endosc 2000; 52:380–386. 48. Itoi T, Shinohara Y, Takeda K, et al. Detection of telomerase reverse transcriptase mRNA in biopsy specimens and bile for diagnosis of biliary tract cancers. Int J Mol Med 2001; 7:281–287. 49. Benckert C, Jonas S, Cramer T, et al. Transforming growth factor beta 1 stimulates vascular endothelial growth factor gene transcription in human cholangiocellular carcinoma cells. Cancer Res 2003; 63:1083–1092. 50. Endo K, Ashida K, Miyake N, et al. E-cadherin gene mutations in human intrahepatic cholangiocarcinoma. J Pathol 2001; 193:310–317. 51. Terada T, Okada Y, Nakanuma Y. Expression of immunoreactive matrix metalloproteinases and tissue inhibitors of matrix metalloproteinases in human normal livers and primary liver tumors. Hepatology 1996; 23:1341–1344. 52. Lavaissiere L, Jia S, Nishiyama M, et al. Overexpression of human aspartyl (asparaginyl) beta-hydroxylase in hepatocellular carcinoma and cholangiocarcinoma. J Clin Invest 1996; 98:1313–1323. 53. Ince N, de la Monte SM, Wands JR. Overexpression of human aspartyl (asparaginyl) beta-hydroxylase is associated with malignant transformation. Cancer Res 2000; 60:1261–1266.
54. Maeda T, Sepe P, Lahousse S, et al. Antisense oligodeoxynucleotides directed against aspartyl (asparaginyl) beta-hydroxylase suppress migration of cholangiocarcinoma cells. J Hepatol 2003; 38:615–622. 55. Tanaka S, Sugimachi K, Kameyama T, et al. Human WISP1v, a member of the CCN family, is associated with invasive cholangiocarcinoma. Hepatology 2003; 37:1122–1129. 56. Koprowski H, Steplewski Z, Mitchell K, et al. Colorectal carcinoma antigens detected by hybridoma antibodies. Somatic Cell Genet 1979; 5:957–972. 57. Magnani JL, Steplewski Z, Koprowski H, et al. Identification of the gastrointestinal and pancreatic cancer-associated antigen detected by monoclonal antibody 19-9 in the sera of patients as a mucin. Cancer Res 1983; 43:5489–5492. 58. Vestergaard EM, Hein HO, Meyer H, et al. Reference values and biological variation for tumor marker CA 19-9 in serum for different Lewis and secretor genotypes and evaluation of secretor and Lewis genotyping in a Caucasian population. Clin Chem 1999; 45:54–61. 59. Steinberg W. The clinical utility of the CA 19-9 tumorassociated antigen. Am J Gastroenterol 1990; 85:350–355. 60. Hadjis NS, Adam A, Gibson R, et al. Nonoperative approach to hilar cancer determined by the atrophy-hypertrophy complex. Am J Surg 1989; 157:395–399. 61. Gores GJ. Early detection and treatment of cholangiocarcinoma. Liver Transpl 2000; 6 (Suppl 2):S30–S34. 62. Rumalla A, Baron TH, Leontovich O, et al. Improved diagnostic yield of endoscopic biliary brush cytology by digital image analysis. Mayo Clin Proc 2001; 76:29–33. 64. Baron TH, Harewood GC, Rumalla A, et al. A prospective comparison of digital image analysis and routine cytology for the identification of malignancy in biliary tract strictures. Clin Gastroenterol Hepatol 2004; 2:214–219. 65. Eloubeidi MA, Chen VK, Jhala NC, et al. Endoscopic ultrasound-guided fine needle aspiration biopsy of suspected cholangiocarcinoma. Clin Gastroenterol Hepatol 2:209–213. 66. Manfredi R, Barbaro Masseli G, et al. Magnetic resonance imaging of cholangiocarcinoma. Semin Liver Dis 2004; 24:155–164. 67. Jarnagin WR, Fong Y, DeMatteo RP, et al. Staging, resectability, and outcome in 225 patients with hilar cholangiocarcinoma. Ann Surg 2001; 234:507–517; discussion 517–519. 68. Weber SM, DeMatteo RP, Fong Y, et al. Staging laparoscopy in patients with extrahepatic biliary carcinoma. Analysis of 100 patients. Ann Surg 2002; 235:392–399. 69. Rea DJ, Munoz-Juarez M, Farnell MB, et al. Major hepatic resection for hilar cholangiocarcinoma. Arch Surg 2004; 139:514–525. 70. Burke EC, Jarnagin WR, Hochwald SN, et al. Hilar cholangiocarcinoma: patterns of spread, the importance of hepatic resection for curative operation, and a presurgical clinical staging system. Ann Surg 1998; 228:385–394. 71. Mizumoto R, Suzuki H. Surgical anatomy of the hepatic hilum with special reference to the caudate lobe. World J Surg 1988; 12:2–10. 72. Kitagawa Y, Nagino M, Kamiya J, et al. Lymph node metastasis from hilar cholangiocarcinoma: audit of 110 patients who underwent regional and paraaortic node dissection. Ann Surg 2001; 233:385–392. 73. Levy MJ, Baron TH, Gostout CJ, et al. Palliation of malignant extrahepatic biliary obstruction with plastic versus expandable metal stents: an evidence-based approach. Clin Gastroenterol Hepatol 2004; 2:273–285. 74. De Palma GD, Galloro G, Siciliano S, et al. Unilateral versus bilateral endoscopic hepatic duct drainage in patients with malignant hilar biliary obstruction: results of a prospective, randomized, and controlled study. Gastrointest Endosc 2001; 53:547–553.
1145
Section X. Tumors of the Liver
75. De Palma GD, Pezzullo A, Rega M, et al. Unilateral placement of metallic stents for malignant hilar obstruction: a prospective study. Gastrointest Endosc 2003; 58:50–53. 76. Hintze RE, Abou-Rebyeh H, Adler A, et al. Magnetic resonance cholangiopancreatography-guided unilateral endoscopic stent placement for Klatskin tumors. Gastrointest Endosc 2001; 53:40–46. 77. Berr F. Photodynamic therapy for cholangiocarcinoma. Semin Liver Dis 2004; 24:177–187. 78. Ortner M, Caca K, Berr F, et al. Successful photodynamic therapy for nonresectable cholangiocarcinoma: a randomized prospective study. Gastroenterology 2003; 125:1355–1363. 79. Gores GJ. A spotlight on cholangiocarcinoma. Gastroenterology 2003; 125:1536–1538. 80. Montemaggi P, Costamagna G, Dobelbower RR, et al. Intraluminal brachytherapy in the treatment of pancreas and bile duct carcinoma. Int J Radiat Oncol Biol Phys 1995; 32:437–443. 81. Bruha R Petrtyl J, Kubecova M, et al. Intraluminal brachytherapy and selfexpandable stents in nonresectable biliary malignancies – the question of long-term palliation. Hepatogastroenterology 2001; 48:631–637. 82. Gerhards MF, van Gulik TM, Gonzalez D, et al. Results of postoperative radiotherapy for resectable hilar cholangiocarcinoma. World J Surg 2003; 27:173–179. 83. Prat F Lafon C, De Lima DM, et al. Endoscopic treatment of cholangiocarcinoma and carcinoma of the duodenal papilla by intraductal high-intensity US: results of a pilot study. Gastrointest Endosc 2002; 56:909–915. 84. Pichlmayr R, Weimann A, Klempnauer J, et al. Surgical treatment in proximal bile duct cancer. A single-center experience. Ann Surg 1996; 224:628–638. 85. Goldstein RM, Stone M, Tillery GW, et al. Is liver transplantation indicated for cholangiocarcinoma? Am J Surg 1993; 166:768–771; discussion 771–772. 86. Iwatsuki S, Todo S, Marsh JW, et al. Treatment of hilar cholangiocarcinoma (Klatskin tumors) with hepatic resection or transplantation. J Am Coll Surg 1998; 187:358–364. 87. Jeyarajah DR, Klintmalm GB. Is liver transplantation indicated for cholangiocarcinoma? Pancreat Surg 1998; 5:48–51.
1146
88. Loinaz C, Abradelo M, Gomez R, et al. Liver transplantation and incidental primary liver tumors. Transplant Proc 1998; 30:3301–3302. 89. Meyer CG, Penn I, James L. Liver transplantation for cholangiocarcinoma: results in 207 patients. Transplantation 2000; 69:1633–1637. 90. Jonas S, Kling N, Guckelberger O, et al. Orthotopic liver transplantation after extended bile duct resection as treatment of hilar cholangiocarcinoma. First long-term results. Transplant Int 1998; 11 (Suppl 1):S206–S208. 91. Shimoda M, Farmer DG, Colquhoun SD, et al. Liver transplantation for cholangiocellular carcinoma: analysis of a single-center experience and review of the literature. Liver Transpl 2001; 7:1023–1033. 92. Foo ML, Gunderson LL, Bender CE, et al. External radiation therapy and transcatheter iridium in the treatment of extrahepatic bile duct carcinoma. Int J Radiat Oncol Biol Phys 1997; 39:929–935. 93. Sudan D, DeRoover A, Chinnakotla S, et al. Radiochemotherapy and transplantation allow long-term survival for nonresectable hilar cholangiocarcinoma. Am J Transplant 2002; 2:774–779. 94. Furubo S, Harada K, Shimonishi T, et al. Protein expression and genetic alterations of p53 and ras in intrahepatic cholangiocarcinoma. Histopathology 1999; 35:230–240. 95. Harnois DM, Que FG, Celli A, et al. Bcl-2 is overexpressed and alters the threshold for apoptosis in a cholangiocarcinoma cell line. Hepatology 1997; 26:884–890. 96. Okaro AC, Deery AR, Hutchins RR, et al. The expression of antiapoptotic proteins Bcl-2, Bcl-X(L), and Mcl-1 in benign, dysplastic, and malignant biliary epithelium. J Clin Pathol 2001; 54:927–932. 97. Ashida K, Terada T, Kitamura Y, et al. Expression of E-cadherin, alpha-catenin, beta-catenin, and CD44 (standard and variant isoforms) in human cholangiocarcinoma: an immunohistochemical study. Hepatology 1998; 27:974–982. 98. Shaib Yh, El-Serag HB, Davila JA, et al. Risk factors of intrahepatic cholangiocarcinoma in the United States: a casecontrol study. Gastroenterology 2005; 128:620-626.
Section X: Tumors of the Liver
61
BENIGN LIVER TUMORS Massimo Colombo and Riccardo Lencioni Abbreviations AFP alfafetoprotein CEA carcinoembrionic antigen CT computed tomography Gd-BOPTA gadobenate dimeglumine
FNH focal nodular hyperplasia MnDPDP mangafodipir trisodium MR magnetic resonance
INTRODUCTION Benign liver tumors are a heterogeneous group of nodular lesions originating from different cell lines. These tumors include hemangiomas, which are the most common benign nodes found in the liver, and hepatocellular neoplasms, which are clinically more relevant lesions (Table 61-1). For hepatocellular lesions, a descriptive nomenclature was set forward by an International Panel of Experts sponsored by the World Congress of Gastroenterology 1994.1 This chapter focuses on hemangioma, focal nodular hyperplasia (FNH), hepatocellular adenoma, and nodular regenerative hyperplasia (NRH). These lesions have gained popularity following the widespread use of imaging tests which has led to the recognition of an increased number of affected patients.
HEPATIC HEMANGIOMA Hepatic hemangiomas are benign, vascular tumors of the liver, the second most common liver mass after metastatic cancer.
EPIDEMIOLOGY The prevalence of hemangiomas ranges from 1 to 2% in the general population, the tumor being typically discovered incidentally during evaluation of non-specific abdominal complaints.2,3 The prevalence of hemangiomas has been overestimated in autopsy series (from 2% to as high as 20%), because of overrepresentation of elderly patients with comorbid illnesses4–7 (Table 61-2). The reported female-tomale gender ratios range from 1.6:1 to 6:1. The majority of hemangiomas are small (<4 cm). Liver nodes greater than 4 cm in size are defined as cavernous hemangiomas.
PATHOGENESIS The pathogenesis of hemangioma is unknown. Congenital hamartoma is a likely candidate. A pathogenic role of sex hormones has been postulated, because of consistent female predominance in larger tumors, and tumor enlargement/recurrence in hysterectomized women under estrogen replacement therapy and in patients with a long use of oral contraceptives.8 Cavernous hemangiomas may have accelerated growth during pregnancy and often display estrogen receptors. However, tumor growth was also induced or influenced by drugs such as metaclopramide.9
NRH SPECT SPIO
nodular regenerative hyperplasia single-photon emission CT superparamagnetic iron oxide
PATHOLOGY Macroscopically, the tumors are ovoid, soft, reddish-purple or blue masses separated from the surrounding parenchyma by a fibrous pseudocapsule. Varying degrees of fibrosis, hyalinization, calcification, thrombosis, and shrinking are seen. Extensive fibrosis and hyalinization, with narrowing or obliteration of vessels, are typical for sclerosed hemangiomas. Microscopically, hemangiomas are vascular abnormalities characterized by multiple blood-filled sinusoidal spaces and vascular lakes lined by endothelial cells. Vascular channels are separated by a fibrous tissue. They are fed by hepatic artery branches and their internal circulation is slow. Blood vessels and arteriovenous shunting may be seen in large septa. The tumor seems to grow by ectasia rather than by hyperplasia or hypertrophy.8
CLINICAL FEATURES The most common clinical presentation of a hemangioma is an incidental finding on hepatic imaging (Table 61-3) and the majority of patients will have a single tumor node (Table 61-4). A few patients may present with isolated diffuse hemangiomatosis in association with Rendu–Osler–Weber’s disease or skeletal hemangiomatosis.8 Cavernous hemangiomas are typically silent, clinically benign, and rarely expanding or symptomatic tumors. The few patients with symptoms complain of one of the following: abdominal swelling, abdominal pain, early satiety, anorexia, abdominal mass, and hepatomegaly. The presence of symptoms correlated with the size of hemangiomas in one study11 but not in another one.5 In addition, there seems to be no correlation between symptoms and number of tumors. Atypical hemangiomas exist that form arteriovenous shunts causing severe symptoms, including heart failure.12 Other unusual clinical presentations of a hepatic hemangioma include hemobilia, inflammatory pseudotumor caused by recurrent intranodal thrombosis, caval thrombosis, portal hypertension, and torsion of a pedunculated tumor.
DIAGNOSIS Liver function tests are typically normal in patients with hemangiomas. A few patients with cavernous hemangiomas may present with coagulopathy, i.e., thrombocytopenia and hypofibrinogenemia.13 Fine-needle aspiration biopsy is considered reasonably safe in hepatic hemangiomas as long as the aspiration route is through a layer of normal liver tissue. The aspirate consists mainly of blood with only a few non-characteristic spindle cells of benign appear-
1147
Section X. Tumors of the Liver
Table 61-1. Classification of Benign Nodular Lesions of the Liver
Table 61-3. Clinical Findings Leading to Diagnosis in Patients with a Liver Hemangioma
Hepatocellular
Clues to diagnosis
Biliary
Vascular
Mesenchymal
Regenerative lesions Monoacinar regenerative nodule Diffuse nodular hyperplasia without fibrous septa (nodular regenerative hyperplasia) Diffuse nodular hyperplasia with fibrous septa or in cirrhosis Multiacinar regenerative nodule Lobar or segmental hyperplasia Focal nodular hyperplasia Dysplastic or neoplastic lesions Hepatocellular adenoma Dysplastic focus Dysplastic nodule Bile duct adenoma Biliary hamartoma Biliary cystoadenoma Biliary papillomatosis Hemangioma Infantile hemangioendothelioma Hereditary hemorrhage telangiectasia Lymphoangiomatosis Leiomyoma Lipoma Myelolymphoma Angiomyolipoma Pseudolymphoma Fibrous mesothelioma Hamartoma Benign teratoma
Table 61-2. Prevalence of Hemangiomas in Population-Based Screening Programs with Abdominal Ultrasound and Autopsy Series Type of study
Author, year
No. of patients
Hemangioma prevalence
Population studies
Lu (1990)2 Gandolfi (1991)3 Karhunen (1986)4 Reddy (1993)5 Rubin (1996)6 Dodd (1999)7
923 21 280 95 256 284 508
1.4% 1.4% 20.0% 6.1% 5.2% 1.7%
Autopsy series
ance. In 36 consecutive fine-needle aspiration biopsies of liver hemangiomas in Helsinki, a cytological diagnosis of hemangiomas was obtained in 21 (58%). One patient (3%) had uneventful intraperitoneal bleeding after the aspiration.14 The sensitivity of percutaneous biopsy (microhistology) for the diagnosis of hemangioma is higher (75–91%), with a specificity of 100%.15
IMAGING FINDINGS Ultrasound and Contrast Ultrasound The most common ultrasound appearance of hemangioma is that of a sharply demarcated lesion with uniformly increased echogenicity relative to normal liver (Figure 61-1). This pattern is observed in about 70% of hemangiomas detected by ultrasound. The remaining cases show atypical ultrasound patterns, and appear either as hypoechoic lesions with a hyperechoic border or as lesions with heterogeneous internal structure.16 Heterogeneity is commonly observed
1148
Incidental Abdominal pain Suspected metastases Palpable mass Non-specific complaints
Terkivatan (2001)10 (n = 103)
Weimann (1997)30 (n = 238)
Farges (1995)31 (n = 163)
64 (62%) 24 (23%) 4 (4%) 2 (2%) 9 (9%)
114 (48%) 99 (42%) 7 (3%) 5 (2%) 13 (5%)
38 (23%) 87 (53%) 25 (16%) 0 13 (8%)
Table 61-4. Clinical Features and Characteristics of Hemangiomas Detected in 123 Patients Undergoing Screening with Abdominal Ultrasound3 No. of males Patient’s age, years
41 (33%) 56 (20–79)
No. of tumors 1 2 >2
93 (75%) 26 (21%) 4 (4%)
Tumor size (cm) <2 2–5 >5
75 (61%) 31 (25%) 17 (14%)
in large hemangiomas. While there are no vascular patterns that can be used to diagnose reliably hemangioma with conventional color or power Doppler ultrasound, early clinical experience with ultrasound contrast agents has suggested that contrast-enhanced ultrasound can provide useful information.17 Most liver hemangiomas (78–93%) show peripheral nodular enhancement during the early phase of the contrast-enhanced study, with progressive centripetal fill-in.17 Diffuse contrast enhancement with homogeneous fill-in or persistent hypoechoic appearance due to absent contrast enhancement can be observed in small, high-flow hemangiomas or thrombosed hemangiomas, respectively.17
Computed Tomography The standard spiral computed tomography (CT) protocol for suspected hemangioma includes baseline and contrast-enhanced scanning in the arterial, the portal venous, and the delayed phase. Most hemangiomas are hypoattenuating on baseline scans, show peripheral nodular enhancement of vascular attenuation on arterial and portal phase imaging, and are hyperattenuating with possible central hypoattenuation or isoattenuation to vascular space in the delayed phase. This pattern has a sensitivity of 67–86% and a specificity of 99–100% for the diagnosis of hemangioma.18,19 Atypical CT features are observed in hemangiomas with either high flow or very slow flow. High-flow hemangiomas show rapid filling after contrast administration, resulting in homogeneous enhancement during the hepatic arterial or portal venous phase.19 This feature is relatively common in small hemangiomas. Differentiation of high-flow hemangioma from hypervascular malignant tumors may be difficult, and relies on attenuation equivalent to that of the aorta during all phases of CT imaging, including the delayed phase. Very slow-flow hemangiomas appear either as non-enhancing lesions or as lesions with weak peripheral enhancement without centripetal progression. These fea-
Chapter 61 BENIGN LIVER TUMORS
A
Figure 61-1. Hemangioma, ultrasound. The lesion shows typical features, and appears as a round, well-defined, hyperechoic nodule (arrow).
tures may be related to thrombosis or abundant fibrosis, and mimic a hypovascular malignant tumor.
Magnetic Resonance Imaging The magnetic resonance (MR) imaging protocol for characterizing suspected hemangioma includes gradient-echo T1-weighted sequences, fast spin-echo T2-weighted sequences with short and long (>200 ms) echo times, and serial dynamic gadolinium-enhanced gradient-echo T1-weighted sequences. Hemangioma appears as a homogeneous focal lesion with smooth, well-defined margins. The lesion is hypointense to liver parenchyma on T1-weighted MR images and strongly hyperintense to liver parenchyma on T2weighted MR images. The high signal intensity on heavily T2weighted (long echo time) MR images gives to hemangiomas a consistent “light-bulb” pattern with 100% sensitivity and 92% diagnostic specificity.20 Dynamic contrast-enhanced MR imaging shows quite a typical perfusion pattern in hemangioma, i.e., peripheral nodular enhancement in the early phase with centripetal progression to uniform or almost uniform enhancement during the portal venous and the delayed phase (Figure 61-2). Such a characteristic
B
Figure 61-2. Hemangioma, magnetic resonance (MR) imaging. (A) On baseline T2-weighted MR image, the lesion shows very high signal intensity. (B) On baseline T1-weighted MR image, the hemangioma is hypointense to surrounding liver parenchyma.
enhancement pattern has a sensitivity of 77–91% and a specificity of 100% for the diagnosis of hemangioma.21,22 However, very small (less than 1.5 cm), high-flow hemangiomas frequently exhibit a hypervascular pattern, with uniform enhancement in the arterial phase, that may persist in the portal venous and delayed phases (Figure 61-3).22 In these cases, diagnostic assessment may be difficult, and requires careful analysis of baseline and contrast-enhanced images. Hemangioma shows a peculiar feature after the injection of reticuloendothelial system-targeted MR agents, that is, lesion hyper-
1149
Section X. Tumors of the Liver
A
C
B
Figure 61-3. Hemangioma, magnetic resonance (MR) imaging. (A) On baseline T2-weighted image, a tiny lesion is detected as a hyperintense nodule (arrow). (B) On baseline T1-weighted image, the hemangioma is hypointense to liver (arrow).
D
Figure 61-2, cont’d. (C) During the arterial phase of the contrast-enhanced dynamic study, the lesion shows peripheral globular enhancement. (D) In the delayed phase, complete homogeneous enhancement is observed within the lesion.
intensity on T1-weighted postcontrast MR images. This is due to the T1 effect of superparamagnetic iron oxide particles trapped within the slow-flow vascular channels of the lesion.23
Tc-Labeled Red Blood Cell Scintigraphy
99m
99m
Tc-pertechnetate-labeled red blood cell scintigraphy is a relatively specific examination for characterizing hemangioma. Using this method, there is decreased activity on early dynamic images and
1150
increased activity on delayed blood pool images. Comparison between 99mTc-pertechnetate-labeled red blood cell single-photon emission CT (SPECT) and MR imaging has shown that MR imaging had higher sensitivity and specificity than SPECT, especially for lesions smaller than 2 cm in diameter.24
DIAGNOSTIC WORKUP The recommended diagnostic workup for suspected hemangioma is dependent on the clinical scenario. If a hemangioma with typical ultrasound features is incidentally detected in a patient with neither history of malignancy nor chronic liver disease, no additional investigation may be required. It has been shown that in this clinical setting the risk of misinterpreting a malignant tumor for a hemangioma is negligible (0.5%). Conversely, in incidental lesions with atypical ultrasound features for hepatic hemangioma, further diagnostic workup is recommended. Additional investigation is also mandatory – regardless of the ultrasound features – for any lesion detected in a patient at increased risk of malignancy. MR imaging is
Chapter 61 BENIGN LIVER TUMORS
giomas should meet strict CT or MR imaging criteria whereas percutaneous biopsy can be used to solve uncertain diagnosis.
TREATMENT
C
D
Figure 61-3, cont’d. (C) In the arterial phase of the contrast-enhanced dynamic MR study, the small lesion shows uniform enhancement (arrow). (D) In the delayed phase, the lesion remains hyperintense due to persistent enhancement (arrow).
currently the most accurate technique for diagnostic confirmation of suspected hemangioma.25 Despite promising results obtained in recent investigations, contrast-enhanced ultrasound is at an early stage of clinical application.26 On the other hand, spiral CT has limitations in achieving a reliable diagnosis of small lesions, especially in the setting of cirrhosis.19,27 High-flow hemangiomas that enhance homogeneously in arterial-phase CT scans may not be confidently distinguished from small hypervascular hepatocellular carcinoma.19 Because MR imaging has greater sensitivity to small differences in contrast enhancement and because several fast MR sequences can be used to track the passage of a small, tight bolus of contrast material, MR images may show the characteristic enhancement patterns of hemangioma and hepatocellular carcinoma better than CT images.27 In addition, MR imaging, besides the information provided by the dynamic gadolinium-enhanced study, can offer improved capability in lesion characterization through the analysis of lesion signal intensity on baseline sequences, especially heavily T2weighted sequences.25 In the setting of cirrhosis, diagnosis of heman-
Treatment is not indicated for asymptomatic tumors that are less than 5 cm in size. Current indications for surgical management of these benign liver masses include uncertain diagnosis with a suspicion of malignancy, severe or progressive symptoms due to size and, less commonly, risk of hemorrhage or rupture.11 Indeed, elective surgical resection has been advocated in selected young patients with asymptomatic, larger than 10 cm hemangiomas to eliminate the risk of hemorrhage, thrombosis, and rupture. Enlarging tumors and those that became symptomatic have been successfully treated by resection, with relief of symptoms in 90% of the cases.11 Patients with multiple hemangiomas and those with extensive hilar involvement may be considered for either angiographic embolization or liver transplantation. The former treatment is indicated for patients with one or few tumors that have favorable vascular anatomy, as a debulking therapy before surgery, to reduce blood loss at the time of surgery, or to treat atypical tumors with arteriovenous shunts.12 When local ethanol injection therapy was delivered to 37 patients with symptomatic hemangiomas (41% with multiple nodes, 60% with cavernous tumors), the tumors shrunk in 27 (73%) of the patients and pain disappeared in 10 (35%) of the 29 patients with symptoms. Liver transplantation may be indicated for large, unresectable tumors, extensive multiple tumors, or when surgical resection is not feasible. Hepatic resection and transcatheter hepatic embolization are effective treatment modalities for the Kasabach–Merritt syndrome, though in a few instances unresectable cavernous hemangiomas with this complication may be an indication for liver transplantation.28 There is little evidence for efficacy of radiotherapy, which carries a risk of radiation hepatitis. In a Pub Med Medline search,29 32 patients with a spontaneously ruptured hemangioma (range 6–25 cm in diameter) were identified. Thirteen (59%) underwent hepatic resection. Five (23%) were sutured and 4 (18%) underwent tamponade. Three of the 13 resected patients, 2 of 5 sutured patients, and 3 of 4 patients who underwent tamponade died.
PROGNOSIS AND NATURAL HISTORY The risk that a patient with an asymptomatic hemangioma will develop abdominal pain is less than 5%.11,30 In a series of patients presenting with abdominal pain, pain disappeared in most patients (n = 47) after treatment of comorbidities or without specific treatment. Interestingly, pain persisted in two-thirds of patients who underwent treatment of a cavernous hemangioma with hepatic resection, embolization, or artery ligation.31 The mechanism of development of symptoms or pain is unclear but may include expansion of tumor size with pressure effects on adjacent hepatic parenchyma or Glisson’s capsule. In a few cases, symptoms relate to intralesional hemorrhage, localized thrombosis, or torsion of a pedunculated hemangioma. Kosabach–Merritt syndrome, i.e. disseminated intravascular coagulation in the setting of a cavernous hemangioma of the liver and cutaneous hemangiomas, presents with abdominal pain and signs of bleeding. During long-term follow-up,
1151
Section X. Tumors of the Liver
a tiny (10%) minority of patients with a hemangioma showed a decrease in tumor size, whereas in a similar percentage of patients the tumor grew in size (Table 61-5). The risk of rupture of hepatic hemangiomas is negligible: the only reports have involved patients with cavernous hemangiomas undergoing trauma-induced or spontaneous rupture. Rupture is associated with sudden onset of severe abdominal pain, abdominal distention, hypotension, or shock, and increased serum levels of transaminases and prothrombin time. A Pub Med Medline search29 has identified up to 32 cases of spontaneous rupture of hepatic hemangioma (mostly cavernous hemangiomas) in adults without a history of trauma.
FOCAL NODULAR HYPERPLASIA FNH of the liver is a rare, completely benign lesion characterized by nodular hyperplasia of hepatic parenchyma around a central scar containing an anomalous artery.
EPIDEMIOLOGY FNH is the second most common benign tumor of the liver, with an estimated prevalence of 0.4–0.8% in unselected autopsy series. The tumor has a female-to-male ratio of between 2 and 26:1 and the average age at presentation is between the ages of 35 and 50 years.8,33
PATHOGENESIS FNH appears to be the result of a hyperplastic response of the hepatic parenchyma to an arterial lesion and/or portal venous ischemia. A congenital vascular malformation is suggested by the presence of a central fibrous scar containing abundant arteries with spider-like malformations and the association with other vascular abnormalities, such as hepatic hemangioma and hereditary hemorrhagic telangiectasia.34–36 The finding of an unbalanced expression of angiopoietin 1 and angiopoietin 2 genes coupled with expression of angiopoietin 1 protein by the endothelial cells of dystrophic vessels suggests a role of angiopoietin genes in FNH.37 Clonality studies and overexpression of important genes involved in cell homeostasis, such as Bcl-2 and transforming growth factor-b, support the important role for hepatocellular proliferation in FNH.38 Conversely, the role of oral contraceptives in FNH is disputed. A hospital-based casecontrol study in women with histologically proven FNH showed a quantitatively proportional increase in the risk of FNH in patients who ever used oral contraceptives.39 Their use has been associated with increase in size and vascularity of FNH nodes and tumor regres-
sion was observed after drug withdrawal.34 However, the association between pregnancy, estrogen, and FNH was negated by an 8-year study in 216 women in Paris.40
PATHOLOGY Macroscopically, the vast majority of the patients have a single, paletan to light-brown lesion causing a central scar radiating into the liver tissue. The majority of solitary tumors are located in the right lobe. In a pathologic study of 305 lesions in Paris (mostly symptomatic), 21% of patients had two to five nodules in the liver and 3% had 15–30 nodular lesions.41 The size of the lesions ranged from 1 mm to 19 cm (median 3 cm), and occasionally lesions were either pedunculated or encapsulated. A large vascular pedicle was observed in a tiny minority (6%) of patients. FNH may present with classical and non-classical forms. The latter include telangiectatic, mixed hyperplastic/adenomatous, with cytologic atypia and multiple FNH syndrome forms. Eighty percent of patients present with the classical form, which is characterized by abnormal nodular architecture, malformed-appearing vessels, and bile duct proliferation. The majority of classical forms contain one to three macroscopic scars. Microscopically, classical FNH lesion shows nodular hyperplastic parenchyma, the nodules being completely or incompletely surrounded by fibrous septa. The central scar contains malformed vessels of various caliber, mostly large and tortuous arteries showing intimal or muscular fibrohyperplasia. Dense bile duct hilar proliferation accompanies the vascular structures both in the central scars and in the radiating septa with histologic cholestasis. A mild degree of macrovascular steatosis is often present. Non-classical forms of FNH show atypical histology and bile ductular proliferation, but lack either nodular architecture or malformed vessels. The multiple FNH syndrome is the presence of at least two FNH lesions and one or more of liver hemangioma, central nervous system vascular malformation, meningioma, and astrocytoma.1
CLINICAL FEATURES FNH is usually an incidental finding but a few patients may have symptoms as a palpable mass or hepatomegaly (Table 61-6).
DIAGNOSIS Liver chemistry is usually unaltered. In a few patients, FNH may cause slight elevations in serum g-glutamyltransferase levels. Serum tumor markers alfafetoprotein (AFP), Ca 19.9 and carcinoembrionic antigen (CEA) are invariably negative. Lesions lacking a central scar and smaller lesions with indeterminate vascular characteristics may be diagnosed by ultrasound-guided thin biopsy.42 In patients with
Table 61-5. Longitudinal Studies Assessing Changes of Volume of Hemangiomas During Follow-Up Author, year
No. of patients
3
Gandolfi (1991) Farges (1995)31 Weimann (1997)30 Terkivatan (2001)10 Okano (2001)32
1152
123 86 104 78 64
Months of follow-up Median (range) a
22 (12–60) 92 (N/A) 32 (7–123) 45 (24–72) 19 (6–58)
Tumor size Decreased
Increased
0 7 (8%) 7 (7%) 0 1 (2%)
1 (1%) 9 (10%) 11 (11%) 1 (1%) 0
Chapter 61 BENIGN LIVER TUMORS
Table 61-6. Clinical Findings Leading to Diagnosis in Patients with Focal Nodular Hyperplasia Clues to diagnosis
Weimann 199730 (n = 150)
Nguyen 199941 (n = 130)
Incidental Abdominal pain Palpable mass Abnormal liver function tests
66 (44%) 49 (37%) 3 (2%) 18 (12%)
46 (35%) 75 (58%) 5 (4%) 17 (13%)
atypical FNH, a scoring system is available to categorize lesions into definite FNH, possible FNH, and negative for FNH.
IMAGING FINDINGS Ultrasound and Contrast Ultrasound FNH may have variable ultrasound features. It usually appears as a round mass that is slightly hypoechoic or slightly hyperechoic compared to liver parenchyma. Some lesions may be isoechoic to liver and may only be detected because of vascular displacement. The lesion is frequently homogeneous. In fact, detection of the central scar at baseline ultrasound is uncommon. Typical findings at color or power Doppler ultrasound include the presence of a central feeding artery with a stellate or spoke-wheel pattern determined by vessels running into radiating fibrous septa originating from the central scar. Doppler spectral analysis can show an intralesional pulsatile waveform with high diastolic flow and low resistive index.43 The specificity of ultrasound in the diagnosis of FNH has improved following the introduction of ultrasound contrast agents. At contrast-enhanced ultrasound, FNH shows central vascular supply with centrifugal filling in the early arterial phase, followed by homogeneous enhancement in the late arterial phase. In the portal phase the lesion remains hyperechoic relative to normal liver tissue, and becomes isoechoic in the late phase. This pattern has been observed in 85–100% of FNH.44 The central scar becomes detectable as a hypoechoic area in the portal phase of the contrast-enhanced study.
Computed Tomography FNH is usually isoattenuating or slightly hypoattenuating to surrounding liver at baseline CT scanning. The detection rate of the central scar – that appears as a hypoattenuating structure – is related to the size of the lesion. It may be identified in 35% of lesions smaller than 3 cm in diameter and in 65% of those exceeding 3 cm.45 FNH shows strong homogeneous enhancement during the arterial phase of the contrast-enhanced CT study. The central scar is typically hypoattenuating during the arterial phase. In the portal venous and delayed phases, FNH becomes isoattenuating to the hepatic parenchyma. On delayed images, the central scar may become hyperattenuating because of contrast distribution within its fibrous stroma. CT features may enable correct characterization of FNH in 78% of cases.
Magnetic Resonance Imaging MR imaging is the most accurate imaging method to characterize FNH. Owing to the affinity of its cells with normal hepatocytes, FNH is usually slighly hypointense or isointense with respect to
normal liver parenchyma on T1-weighted images and slightly hyperintense or isointense on T2-weighted images.46 The hallmark of the lesion, the central stellate scar, is usually depicted because of its hypointensity on T1-weighted images and hyperintensity on T2weighted images, reflecting its pathologic substratum of a vascularized connective tissue.46 At baseline MR imaging, however, the mentioned typical features – homogeneous structure, isointensity to liver, and presence of the central scar – are observed in only 22% of cases.47 Diagnostic confirmation requires a contrast-enhanced MR study. This is usually performed through serial dynamic imaging following the administration of a gadolinium chelate.48 FNH shows strong, homogeneous enhancement in the arterial phase sparing the central scar, while it becomes isointense to liver parenchyma in the portal venous and delayed phases (Figure 61-4). The central scar may show contrast uptake in the delayed phase owing to the interstitial distribution of the contrast agent. These features have a specificity of more than 95% for the diagnosis of FNH.48 However, even with the administration of gadolinium chelates, the central scar may not be detectable in as many as 22% of the FNHs, including 80% of those smaller than 3 cm.48 Liver-specific MR contrast agent provides an alternate strategy to diagnose FNH. Owing to the affinity of its cells with hepatocytes, FNH takes up hepatocyte-targeted agents, like normal parenchyma. These agents are then trapped within the lesion, since FNH is unable to eliminate the compound via biliary excretion effectively. Hence, it appears hyperintense to normal parenchyma on T1-weighted images. Also, the central scar – that does not take up the hepatocyte-targeted agent – becomes well-delineated in up to 90% of cases (Figure 61-5).49 This approach may enable diagnosis of 90% of FNHs with atypical features at the baseline and conventional contrastenhanced dynamic study.48 The diagnosis of FNH has also been achieved with the use of reticulo-endothelial system (RES)-targeted agents. Because of its rich Kupffer cell population, FNH takes up iron oxide particles, and shows marked signal intensity decrease on T2-weighted images. The central scar is usually well-delineated on postcontrast images as it does not contain RES cells and therefore keeps a high signal intensity.47 99m
Tc Sulfur Colloid Scintigraphy
99m
Tc sulfur colloid scintigraphy has long been used to characterize FNH. In fact, up to 80% of these lesions show uptake owing to their Kupffer cell population.50 Unfortunately, the uptake of sulfur colloid is not highly specific. In a series of 20 lesions, sulfur colloid studies were diagnostic in only 16% of FNHs greater than 3.5 cm and in 14% of lesions smaller than 3.5 cm.51
DIAGNOSTIC WORKUP FNH is usually detected incidentally. Diagnostic confirmation can rely solely on imaging findings, provided that typical features are shown in the proper clinical setting. CT can be used to characterize lesions of medium-to-large size, but has limitations in the diagnosis of small lesions.19,52 Although promising results have recently been reported with the use of contrast-enhanced ultrasound, MR imaging is the most accurate technique to diagnose FNH.25 Besides the information provided by the baseline and dynamic gadoliniumenhanced study, MR imaging can offer improved capability in lesion
1153
Section X. Tumors of the Liver
Figure 61-4. Focal nodular hyperplasia, magnetic resonance (MR) imaging. (A) On baseline T2weighted image, the lesion that is focal nodular hyperplasia appears as isointense to surrounding liver parenchyma (i.e., undetectable); the central scar is detected as a hyperintense area (arrow). (B) On baseline T1-weighted image, the lesion is isointense to liver and the scar is detected as a hypointense zone (arrow).
A
B
1154
Chapter 61 BENIGN LIVER TUMORS Figure 61-4, cont’d. (C) In the arterial phase of the contrast-enhanced dynamic MR study, a rapid homogeneous enhancement sparing the central scar is observed. (D) In the delayed phase, the lesion becomes isointense to surrounding parenchyma.
C
D
1155
Section X. Tumors of the Liver
Figure 61-5. Focal nodular hyperplasia, magnetic resonance (MR) imaging. (A) On baseline T2weighted MR image, the lesion is slightly hyperintense to liver (arrow).
A
1156
Chapter 61 BENIGN LIVER TUMORS Figure 61-5, cont’d. (B) On baseline T1-weighted MR image, the lesion is hypointense to surrounding parenchyma (arrow).
B
1157
Section X. Tumors of the Liver
Figure 61-5, cont’d. (C) After the administration of a hepatobiliary contrast agent, mangafodipir trisodium (MnDPDP), the lesion appears hyperintense to normal parenchyma on the T1-weighted image (arrow) and the central scar is detected as a hypointense area.
C
characterization through the use of liver-specific agents, especially in small lesions.25 The use of percutaneous biopsy should be restricted to cases with questionable findings.52
ASSOCIATED CONDITIONS The syndrome of multiple FNH is the presence of FNH, hepatic hemangiomas, and disorders of the central nervous system, such as meningioma, astrocytoma, and arterial malformations. This syndrome has been described in association with the Kippel– Trenaunay–Weber syndrome, a non-hereditary congenital condition characterized by capillary malformations, hemihypertrophy, and venous stasis.34 The association between FNH and fibrolamellar carcinoma is disputed.
TREATMENT Treatment is rarely indicated. Hemorrhage, clinically important symptoms, and uncertain diagnosis are indications for surgical resec-
1158
tion. Treatment of FNH should also be reserved for patients with a lesion that demonstrates growth on sequential imaging. In a series of 150 patients with FNH in Hanover, 5 (3%) underwent hepatic resection because of the onset of symptoms. Recurrence or persistence of presenting symptoms following resection may exceed 20%.30
PROGNOSIS AND NATURAL HISTORY FNH is a completely benign condition with the potential of changing in size. In one study,30 4% of lesions decreased in size, whereas FNH increased in size in 3% of 136 patients who were monitored for 9 years. In recent reports, the size of the lesions did not increase during oral contraceptive treatment and pregnancy,40 nor did it expand in patients receiving immunosuppressive therapy.53 The risk of bleeding of FNH seems remote5,54 and neoplastic transformation has never been reported. In a few patients, FNH has been reported to progress to develop clinically important symptoms.54 In a series of 53 patients observed in Hanover for 3 years, upper abdominal
Chapter 61 BENIGN LIVER TUMORS
symptoms developed in 21 (40%), and these symptoms were severe in 2 cases (4%).30
Table 61-7. Clinical Findings Leading to Diagnosis of Hepatocellular Adenoma Clues to diagnosis
Weimann (1997)30 (n = 44)
Herman (2000)63 (n = 10)
Terkivatan (2001)10 (n = 33)
Incidental Abdominal pain Elevated GGTP Bleeding
12 (27%) 19 (43%) 3 (7%) 6 (13%)
2 (20%) 8 (80%) 4 (40%) n.a.
10 (30%) 10 (30%) 0 12 (36%)
HEPATOCELLULAR ADENOMA Hepatocellular adenomas are rare, frequently capsulated nodular lesions of the liver characterized by the benign proliferation of liver cells.
GGTP, g-glutamyl transpeptidase.
EPIDEMIOLOGY In a series of patients collected between 1989 and 1992 the adenoma-to-FNH ratio was 1:10.55 The lesions occur predominantly in young women, more commonly in women of reproductive age, who use oral contraceptives. In one study,56 the relative risk of hepatocellular adenoma was 25 for women who used oral contraceptives for more than 109 months compared to those who used contraceptives for less than 12 months. Though the female-to-male ratio is 4:1,30 the incidence of hepatocellular adenoma appears to have increased in males, because the use of anabolic drugs has become widespread in sports.57 In a few patients, the tumor may present as 10 adenomas in an otherwise normal liver (liver adenomatosis) with a male-to-female distribution of approximately 3:1.58
PATHOGENESIS The pathogenesis of adenomas is not fully understood. Clonality studies indicate that this tumor results from benign proliferation of a clone of hepatocytes.38 Limited genetic alterations have been found in patients with hepatocellular adenoma, such as gain of chromosome 7p, 17q, and 20q, and deletions from exon 3 to exon 4 of the beta-catenin gene. The deletion of chromosome 8p and loss of heterozygosity on the mannose 6-phosphate insulin-like growth factor II receptor locus and Mut-L-homologue 1 have been reported.8 One study also described heterogeneous expression of the membrane glycoprotein cadherins.59 At variance with hepatocellular carcinoma, no genetic changes for the p53, axin, and adenomatous polyposis colon genes were detected in any of the adenomas. Adenoma, therefore, is not pathogenically linked to the sequence adenoma–carcinoma as it is in intestinal polyposis. Oral contraceptives have a role in the evolution of some forms of liver adenomatosis, and vascular liver damage due to altered liver cell proliferation, intrahepatic vascular shunts, or an association with FNH have been postulated.60
PATHOLOGY Hepatic adenomas are soft, yellow lesions, often with a highly vascularized surface and a capsule, and focal areas of hemorrhage in the parenchyma. The histologic features are two or more cell-thick sheets of hepatocytes without cellular atypia (to differentiate from adenocarcinoma), portal tracts (to differentiate from liver cell regeneration), and biliary ductules and fibrosis (to differentiate from FNH). There may be fatty infiltration at the periphery of small areas of liver cell proliferation (adenomatous hyperplasia) seen between large adenomas. Some degree of liver cell dysplasia may be present in adenomas, particularly in those with a pseudofollicolar pattern. Foci of malignant transformation have been described that may
escape detection in small specimens obtained with a thin-needle liver biopsy.61
CLINICAL FEATURES Hepatocellular adenomas are usually solitary. Approximately 30% of the patients have multiple nodules and the presence of 10 adenomas defines liver adenomatosis.62 Approximately half of the cases of adenomas have been discovered incidentally, whereas the remainder have had symptoms such as pain or abdominal mass (Table 61-7). In the literature, 10 patients (26%) with liver adenomatosis became symptomatic because of intraperitoneal bleeding and 9 of these patients were taking oral contraceptives.62 Patients with hepatocellular adenoma have a higher prevalence of symptoms at first presentation compared to patients with hemangioma or FNH, probably caused by the high rate of intratumoral or intra-abdominal hemorrhage.10,61 The diagnosis of liver adenomatosis is made because of complications of adenomas (intraperitoneal bleeding, intratumoral hemorrhage, or necrosis producing acute pain), because of hepatomegaly with or without symptoms, or as an incidental finding. While the massive form of liver adenomatosis is rare and can be unilobular, most patients have multifocal liver adenomatosis spread in both lobes.62
DIAGNOSIS Laboratory tests are not helpful during the diagnostic workup. However, negative tests for serum alpha-fetoprotein and hepatitis B and C corroborate the exclusion of malignant disease. Percutaneous liver biopsy is of little value because of the possible lack of specific features in a small specimen, while the procedure carries the risk of needle-induced bleeding in hypervascular nodes. In liver adenomatosis, two- or threefold increases of alkaline phosphatase or gglutamyltranspeptidase levels have been described.62 Diagnosis of liver adenomatosis is better obtained by exploration of the liver by laparoscopy or laparotomy, allowing the operator to obtain biopsy specimens of several different lesions without the risk of hemorrhage.62
IMAGING FINDINGS Ultrasound and Contrast Ultrasound Hepatocellular adenoma has variable sonographic appearances. The lesion may appear as slightly hypoechoic, isoechoic, or hyperechoic. When necrotic or hemorrhagic changes occur, adenoma appears as a complex mass with a large cystic component. Using color or power Doppler, the arterial hypervascularity is well demonstrated by arterial vessels running along the border of the lesion in a “basket”
1159
Section X. Tumors of the Liver
pattern.64,65 At contrast-enhanced ultrasound, adenoma shows intense enhancement during the arterial phase. During the portal venous and equilibrium phases adenomas may appear as isoechoic or slightly hyperechoic mass.66 None of these features, unfortunately, is specific enough for the diagnosis.
Computed Tomography Baseline CT scans can easily detect the presence of fat or recent hemorrhage within the lesion, features that can suggest the diagnosis of adenoma. During dynamic contrast-enhanced CT scanning, non-complicated adenomas may enhance rapidly and appear homogeneously hyperdense compared to the liver. The enhancement usually does not persist in adenomas because of arteriovenous shunting within the lesion. Larger or complicated adenomas may be highly heterogeneous because of necrotic phenomena or intralesional hemorrhage (Figure 61-6).
Magnetic Resonance Imaging On MR images, hepatocellular adenoma can show variable signal intensity. It has been reported that 59–77% of adenomas are hyperintense on T1-weighted images due to intralesional content of fat or glycogen, or because of recent hemorrhage.67 In contrast, low-signalintensity areas are related to necrosis. The neoplasm may appear homogeneously or heterogeneously hyperintense on T2-weighted images. With dynamic MR study, adenoma shows early enhancement during the arterial phase and becomes isointense to liver in the portal venous and delayed phases, features that overlap with those of other hypervascular tumors.68 On the other hand, the usefulness of tissue-specific MR agents is not fully established. After administration of mangafodipir trisodium (MnDPDP), adenoma shows positive enhancement,69 but contrast uptake may not occur with gadobenate dimeglumine (Gd-BOPTA).67 In some cases, adenomas may take up superparamagnetic iron oxide particles, resulting in
Figure 61-6. Hepatocellular adenoma, computed tomography (CT). (A) Baseline CT scan shows a large, heterogeneous mass in the left liver lobe. (B) In the arterial phase of the contrast-enhanced CT study, rapid, inhomogeneous enhancement of the lesion is detected.
A
B
1160
Chapter 61 BENIGN LIVER TUMORS Figure 61-6, cont’d. (C) In the portal venous phase, the lesion shows heterogeneous features, with hypo- and hyperattenuating areas. (D) Resected specimen shows hepatocellular adenoma with intralesional hemorrhage and necrosis.
C
D
decreased signal intensity on T2-weighted images. However, the uptake of superparamagnetic iron oxide (SPIO) in adenoma is variable, depending of Kupffer cells content or function.
ASSOCIATED CONDITIONS
DIAGNOSTIC WORKUP
TREATMENT
Radiologic diagnosis of adenoma is very difficult. Unfortunately, CT and MR imaging features of adenoma are non-specific, and the only finding that may suggest the diagnosis is the detection of intratumoral hemorrhage. Even with the use of MR and liver-specific contrast agents, a final diagnosis can hardly be made in non-hemorrhagic adenoma.
The management of patients with adenoma is evolving. For patients taking oral contraceptives with an adenoma that is incidentally discovered and less than 4 cm, the best option would be treatment withdrawal and close monitoring with sequential ultrasound examinations of the abdomen. While patients with adenoma regression may escape or delay surgical treatment or locoregional ablation,
Adenoma has been associated with glycogen storage disease, anabolic drug use, diabetes,62 and oral contraceptives.
1161
Section X. Tumors of the Liver
those with adenomas that are larger, symptomatic and unresponsive to contraceptive withdrawal should be considered for surgical treatment. Tumor ablation should also be considered for patients who for medical reasons cannot stop oral contraceptives or for those who plan to become pregnant.11 Treatment of patients with ruptured adenoma requires stabilization with selective arterial embolization prior to resection.10 The management of liver adenomatosis is problematic. The unilobular, massive form can be treated with hepatic resection. For patients with multifocal liver adenomatosis, surgery of the largest or of the complicated adenoma is an option. Patients with malignant transformation are candidates for liver transplantation.70
PROGNOSIS AND NATURAL HISTORY The potential of adenomas to increase in size is not well established for women taking oral contraceptives, whereas regression of the lesion has been documented on cessation of oral contraceptive use.8 The potential of hepatocellular adenoma to progress to cancer is also uncertain. There have been reports suggesting that the risk of malignancy is higher in patients with multiple adenomas than in those with solitary lesions, but this interpretation is not universally accepted.71 The other clinical concern of adenoma, bleeding, involves approximately one-quarter of all patients. In 12 patients with hepatocellular adenoma collected over 10 years in Taiwan, 4 had pathologic findings of intratumor hemorrhage upon resection and 1 patient had in fact had hemoperitoneum due to rupture of the adenoma. In another series of 33 patients in Rotterdam, 14 were managed through observation and 1 of these presented with a ruptured adenoma.10
NODULAR REGENERATIVE HYPERPLASIA NRH of the liver is a condition in which nodules of proliferating hepatocytes develop in a liver with a preserved architecture without fibrous septa. The lesion is commonly seen at the hepatic hilum or around the large portal tract and may be associated with features of portal hypertension and sublinical cholestasis.72
EPIDEMIOLOGY NRH occurs in both males and females. Two autopsy series of approximately 3000 patients demonstrated NRH in approximately 2.5% of patients,73,74 with a female-to-male ratio of approximately 1–2:1. The disease predominantly affects patients who are older than 60 years and patients with portal hypertension or portal vein thrombosis.73 However, the disease has also been reported in children as young as 7 months as well as in young adults.
PATHOGENESIS The pathogenesis of NRH is still unknown. Two theories have been proposed to explain the pathogenesis of this disease. The basic pathologic lesion leading to NRH may be obliteration or thrombosis of the portal vein system causing ischemic atrophy in the central zones of the hepatic acinus,75 the centrolobular atrophy being compensated by proliferation of liver cells from periportal areas which
1162
form regenerative nodules. This sequence of events is accounted for by several portal venous abnormalities such as venous thrombi and phlebosclerosis.73,74 NRH could also be a generalized proliferative disorder of the liver with the potential to progress to hepatocellular carcinoma,76 an interpretation that is corroborated by the frequent occurrence of liver cell dysplasia in NRH.
PATHOLOGY NRH has been defined as a secondary and non-specific tissue adaptation to heterogeneous distribution of blood flow, occurring as part of a spectrum of architectural changes known as nodular transformation.74 Nodular transformation is recognized by regions of atrophy of liver tissue juxtaposed with normal or hyperplastic areas with a curved contour and no intervening fibrous septa. Diffuse nodular hyperplasia may also be associated with fibrous septa or may be superimposed on a previously cirrhotic liver.1 Macroscopically, the liver is normal in size and shows nodules of 1–10 mm diameter that are centered on a larger portal tract and may mimic micronodular cirrhosis. Partial nodular transformation of the liver is no longer used to define NRH that forms confluent masses in association with high-grade obstruction of the mediumsized or larger portal veins. Microscopically, the nodes show liver cell plates that are one or two cells wide, whereas the sinusoids are narrow. In the intranodular regions the cell plates are one-cell-thick, hepatocytes may be atrophic, and sinusoids are usually dilated. NRH differs from a multiacinar regenerative nodule which contains more than one portal tract located in a cirrhotic liver or with severe disease of portal veins, hepatic veins, or sinusoids.1
CLINICAL FEATURES The clinical presentation of NRH is heterogeneous and may range from asymptomatic to end-stage liver disease. Hepatomegaly and splenomegaly occur in less than half of patients with NRH.73 NRH is often an incidental finding in patients presenting with features of hepatic disease or lymphoproliferative disorders. In patients with a clinical history of vasculitis, hepatic artery fibrosis is found. Portal hypertension is present in half of patients with NRH but only in a minority of the patients as a presenting feature.74 Life-threatening bleeds and deaths are uncommon. In a few patients liver failure requiring liver transplantation was the presenting feature.77
DIAGNOSIS The typical presentation of NRH is bleeding from esophageal varices due to portal hypertension. However, most patients with NRH have no specific signs or symptoms related to their hepatic disease. Liver function tests are normal in most patients. Alkaline phosphatase is elevated to more than 1.5 times the upper limit of normal in onethird of patients.74 A liver biopsy may allow diagnosis of NRH. However, since NRH, incomplete cirrhosis, and complete cirrhosis may occur in different regions of the same liver, large quantities of liver tissue have to be examined for complete diagnosis.74 Portal pressure deterioration is an additional key to the diagnosis of NRH. The procedure of transjugular liver biopsy allows both histopathological diagnosis and hepatic venography and hepatic vein pressure measurement to be performed in parallel.
Chapter 61 BENIGN LIVER TUMORS
IMAGING FINDINGS
DISEASE COMPLICATIONS
Sonography may show multiple isoechoic or hyperechoic nodules. They may become hypoechoic or anechoic as a result of intratumoral hemorrhage.78 Findings at color or power Doppler ultrasound include the presence of intratumoral vessels and sometimes the presence of a central feeding artery.79 On baseline CT, the lesions are usually isodense to the liver. Subcapsular lesions may distort the surface of the liver. During the dynamic study, both hypervascular lesions and nodules with the same attenuation of the normal parenchyma are visible.80 The lesions are frequently isointense to normal liver on T1- and T2-weighted images. However, some nodules may show high signal intensity on T1-weighted images, probably correlated with copper deposits74,81,82 (Figure 61-7). Following the administration of hepatobiliary contrast agents, nodules become hyperintense to the liver as they contain normal hepatocytes and a dysfunctional biliary system. Because of its Kupffer cell population, nodules take up iron oxide particles, and show signal intensity decrease on T2-weighted images like the normal parenchyma.83 Unfortunately, the imaging findings of NRH are nonspecific. The diffuse nature of involvement, the associated portal hypertension, and the clinical history are all features that may allow the characterization of this entity.
Portal hypertension with symptoms of variceal bleeding and ascites may develop in 5–13% of patients.74 Compression of the intrahepatic portal radicles by the regenerating nodules and thrombosis of the portal vein/venules are likely pathogenetic mechanisms. Hepatic failure rarely occurs.
TREATMENT In most patients, NRH is a slow-evolving, asymptomatic condition requiring no treatment. The few patients with symptomatic portal hypertension presenting with variceal bleeding have been treated with recurrent endoscopic therapy or portacaval shunt.8 Three patients presenting with progressive liver failure from NRH underwent orthotopic liver transplantation and in at least 1 of these patients NRH recurred after transplantation.77
PROGNOSIS AND NATURAL HISTORY NRH tends to run an indolent course. The few patients developing clinical decompensation will have a slow progression toward endstage liver disease. A rare complication reported in the literature is disease exacerbation in patients receiving chemotherapy for associated myeloproliferative disease or hemoperitoneum due to rupture of a regenerative nodule.84
ASSOCIATED CONDITIONS NRH is associated with lymphoproliferative, rheumatologic, vascular, and storage disease. The condition has been seen in patients receiving anabolic steroids, chemotherapeutic agents, and azathioprine.8
CONCLUSIONS Benign liver tumors are usually identified by imaging assays and in a few instances only require treatment. Single or multiple imaging
Figure 61-7. Nodular regenerative hyperplasia, computed tomography (CT) and magnetic resonance imaging. (A) CT shows inhomogeneous liver, with subtle hypoattenuating nodules.
A
1163
Section X. Tumors of the Liver
Figure 61-7, cont’d. (B) On T2-weighted image, no focal abnormality is detected. (C) On T1-weighted image, multiple small hyperintense nodules are detected thoroughout the liver parenchyma.
B
C
1164
Chapter 61 BENIGN LIVER TUMORS
methods may be used to accomplish acceptable diagnostic specificity. Lesions failing to fulfill specific diagnostic criteria by radiologic imaging require either biopsy or surgical resection. On suspicion of cavernous hemangioma, asymptomatic tumors require surveillance with sequential imaging and, if stable, no further testing is required. Conversely, treatment is advisable for patients with a typical hemangioma that is symptomatic or more than 15 cm in diameter. Patients presenting with atypical hemangioma require enhanced follow-up or biopsy. If the tumor enlarges, resection should be considered. Patients with FNH require follow-up with imaging assays whenever typical imaging of FNH is obtained with demonstration of a central scar by CT scan or MR. Biopsy or resection is advocated in patients with expanding large masses or those presenting with atypical features. Masses with typical features of hepatocellular adenoma are often advocated to surgical resection or ablation, because of concern about bleeding and progression to hepatocellular carcinoma. Biopsy is also required to distinguish NRH from cirrhosis, which requires no treatment in the majority of patients.
REFERENCES 1. International working party. Terminology of nodular hepatocellular lesions. Hepatology 1995; 22:983–993. 2. Lu SN, Wang LY, Chang WY, et al. Abdominal sonographic screening in a single community. Gaoxiong Yi Xue Ke Xue Za Zhi 1990; 6:643–646. 3. Gandolfi L, Leo P, Solmi L, et al. Natural history of hepatic haemangiomas: clinical and ultrasound study. Gut 1991; 32:677–680. 4. Karhunen PJ. Benign hepatic tumours and tumour like conditions in men. J Clin Pathol 1986; 39:183–188. 5. Reddy KR, Schiff ER. Approach to a liver mass. Semin Liver Dis 1993; 13:423–435. 6. Rubin RA, Mitchell DG. Evaluation of the solid hepatic mass. Med Clin North Am 1996; 80:907–928. 7. Dodd GD 3rd, Baron RL, Oliver JH 3rd, et al. Spectrum of imaging findings of the liver in end-stage cirrhosis: part II, focal abnormalities. AJR Am J Roentgenol 1999; 173:1185–1192. 8. Trotter JF, Everson GT. Benign focal lesions of the liver. Clin Liver Dis 2001; 5:17–42. 9. Feurle GE. Arteriovenous shunting and cholestasis in hepatic hemangiomatosis associated with metoclopramide. Gastroenterology 1990; 99:258–262. 10. Terkivatan T, de Wilt JH, de Man RA, et al. Indications and longterm outcome of treatment for benign hepatic tumors: a critical appraisal. Arch Surg 2001; 136:1033–1038. 11. Charny CK, Jarnagin WR, Schwartz LH, et al. Management of 155 patients with benign liver tumours. Br J Surg 2001; 88:808–813. 12. Tanaka A, Morimoto T, Yamamori T, et al. Atypical liver hemangioma with shunt: long-term follow-up. J Hepatobiliary Pancreat Surg 2002; 9:750–754. 13. Suzuki H, Nimura Y, Kamiya J, et al. Preoperative transcatheter arterial embolization for giant cavernous hemangioma of the liver with consumption coagulopathy. Am J Gastroenterol 1997; 92:688–691. 14. Taavitsainen M, Airaksinen T, Kreula J, et al. Fine-needle aspiration biopsy of liver hemangioma. Acta Radiol 1990; 31:69–71. 15. Heilo A, Stenwig AE. Liver hemangioma: US-guided 18-gauge core-needle biopsy. Radiology 1997; 204:719–722.
16. Moody AR, Wilson SR. Atypical hepatic hemangioma: a suggestive sonographic morphology. Radiology 1993; 188:413–417. 17. Quaia E, Calliada F, Bertolotto M, et al. Characterization of focal liver lesions with contrast-specific US modes and a sulfur hexafluoride-filled microbubble contrast agent: diagnostic performance and confidence. Radiology 2004; 232:420–430. 18. Nino-Murcia M, Olcott EW, Jeffrey RB Jr, et al. Focal liver lesions: pattern-based classification scheme for enhancement at arterial phase CT. Radiology 2000; 215:746–751. 19. Kim T, Federle MP, Baron RL, et al. Discrimination of small hepatic hemangiomas from hypervascular malignant tumors smaller than 3 cm with three-phase helical CT. Radiology 2001; 219:699–706. 20. McFarland EG, Mayo-Smith WW, Saini S, et al. Hepatic hemangiomas and malignant tumors: improved differentiation with heavily T2-weighted conventional spin-echo MR imaging. Radiology 1994; 193:43–47. 21. Whitney WS, Herfkens RJ, Jeffrey RB, et al. Dynamic breathhold multiplanar spoiled gradient-recalled MR imaging with gadolinium enhancement for differentiating hepatic hemangiomas from malignancies at 1.5 T. Radiology 1993; 189:863–870. 22. Semelka RC, Brown ED, Ascher SM, et al. Brown JJ. Hepatic hemangiomas: a multi-institutional study of appearance on T2weighted and serial gadolinium-enhanced gradient-echo MR images. Radiology 1994; 192:401–406. 23. Montet X, Lazeyras F, Howarth N, et al. Specificity of SPIO particles for characterization of liver hemangiomas using MRI. Abdom Imaging 2004; 29:60–70. 24. Birnbaum BA, Weinreb JC, Megibow AJ. Definitive diagnosis of hepatic hemangiomas: MR vs Tc-99m labeled red blood cell SPECT. Radiology 1990; 176:95–105. 25. Lencioni R, Cioni D, Crocetti L, et al. Magnetic resonance imaging of liver tumors. J Hepatol 2004; 40:162–171. 26. Albrecht T, Blomley M, Bolondi L, et al. Guidelines for the use of contrast agents in ultrasound. Ultraschall Med 2004; 25:249–256. 27. Brancatelli G, Federle MP, Blachar A, et al. Hemangioma in the cirrhotic liver: diagnosis and natural history. Radiology 2001; 219:69–74. 28. Longeville JH, de la Hall P, Dolan P, et al. Treatment of a giant haemangioma of the liver with Kasalbach–Merritt syndrome by ortotopic liver transplant: a case report. HPB Surg 1997; 10:159–162. 29. Corigliano N, Mercantini P, Amodio PM, et al. Hemoperitoneum from a spontaneous rupture of a giant hemangioma of the liver: report of a case. Surg Today 2003; 33:459–463. 30. Weimann A, Ringe B, Klempnauer J, et al. Benign liver tumors: differential diagnosis and indication for surgery. World J Surg 1997; 21:983–990. 31. Farges O, Daradkeh S, Bismuth H. Cavernous hemangiomas of the liver: are there any indications for resection? World J Surg 1995; 19:19–24. 32. Okano H, Shiraki K, Inoue H, et al. Natural course of cavernous hepatic hemangioma. Oncol Rep 2001; 8:411–414. 33. Wanless IR, Albrecht S, Bilbao J, et al. Multiple focal nodular hyperplasia of the liver associated with vascular malformations of various organs and neoplasia of the brain: a new syndrome. Mod Path 1989; 2:456–462. 34. Haber M, Reuben A, Burrell M, et al. Multiple focal nodular hyperplasia of the liver associated with hemihypertrophy and vascular maiformations. Gastroenterology 1995; 108:1256–1262. 35. Mathieu D, Zafrani ES. Anglade MC, et al. Association of focal nodular hyperplasia and hepatic hemangioma. Gastroenterology 1989; 97:154–157. 36. Buscarini E, Danesino C, Plauchu H, et al. High prevalence of hepatic focal nodular hyperplasia in subjects with hereditary
1165
Section X. Tumors of the Liver
37.
38. 39.
40.
41.
42.
43. 44.
45.
46. 47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
1166
hemorrhagic telangiectasia. Ultrasound Med Biol 2004; 30:1089–1097. Paradis V, Bieche I, Dargere D, et al. A quantitative gene expression study suggests a role for angiopoietins in focal nodular hyperplasia. Gastroenterology 2003; 124:651–659. Gaffey MJ, Iezzoni JC, Weiss LM. Clonal analysis of focal nodular hyperplasia of the liver. Am J Pathol 1996; 4:1089–1096. Scalori A, Tavani A, Gallus S, et al. Oral contraceptives and the risk of focal nodular hyperplasia of the liver: a case-control study. Am J Obstet Gynecol 2002; 186:195–197. Mathieu D, Kobeiter H, Maison P, et al. Oral contraceptive use and focal nodular hyperplasia of the liver. Gastroenterology 2000; 118:560–564. Nguyen BN, Flejou JF, Terris B, et al. Focal nodular hyperplasia of the liver: a comprehensive pathologic study of 305 lesions and recognition of new histologic forms. Am J Surg Pathol 1999; 23:1441–1454. Bioulac-Sage P, Balabaud C, Wanless IR. Diagnosis of focal nodular hyperplasia not so easy. Am J Surg Pathol 2001; 25:1322–1325. Gaiani S, Piscaglia F, Serra C, et al. Hemodynamics in focal nodular hyperplasia. J Hepatol 1999; 31:576. Kim MJ, Lim HK, Kim SH, et al. Evaluation of hepatic focal nodular hyperplasia with contrast-enhanced gray scale harmonic sonography: initial experience. J Ultrasound Med 2004; 23:297–305. Brancatelli G, Federle MP, Grazioli L, et al. Focal nodular hyperplasia: CT findings with emphasis on multiphasic helical CT in 78 patients. Radiology 2001; 219:61–68. Bartolozzi C, Cioni D, Donati F, et al. Focal liver lesions: MR imaging–pathologic correlation. Eur Radiol 2001; 11:1374–1388. Ba-Ssalamah A, Schima W, Schmook MT, et al. Atypical focal nodular hyperplasia of the liver: imaging features of nonspecific and liver-specific MR contrast agents. AJR Am J Roentgenol 2002; 179:1447–1456. Grazioli L, Morana G, Federle MP, et al. Focal nodular hyperplasia: morphologic and functional information from MR imaging with gadobenate dimeglumine. Radiology 2001; 221:731–739. Savellano DH, Kostler H, Baus S, et al. Assessment of sequential enhancement patterns of focal nodular hyperplasia and hepatocellular carcinoma on mangafodipir trisodium enhanced MR imaging. Invest Radiol 2004; 39:305–312. Welch TJ, Sheedy PF, Johnson CM, et al. Focal nodular hyperplasia and hepatic adenoma: comparison of angiography, CT, US and scintigraphy. Radiology 1985; 156:593–595. Broglia L, Bezzi M, Massa R, et al. Computerized tomography, magnetic resonance, and nuclear medicine in the non-invasive diagnosis of focal nodular hyperplasia of the liver. Radiol Med 1998; 96:218–225. Choi CS, Freeny PC. Triphasic helical CT of hepatic focal nodular hyperplasia: incidence of atypical findings. AJR Am J Roentgenol 1998; 170:391–395. Tan M, Di Carlo A, Robinson P, et al. Successful outcome after transplantation of a donor liver with focal nodular hyperplasia. Liver Transpl 2001; 7:652–655. Sadowski DC, Lee SS, Wanless IR, et al. Progressive type of focal nodular hyperplasia characterized by multiple tumors and recurrence. Hepatology 1995; 61:210–214. Cherqui D, Rahmouni A, Charlotte F, et al. Management of focal nodular hyperplasia and hepatocellular adenoma in young women: a series of 41 patients with clinical, radiological, and pathological correlations. Hepatology 1995; 22:1674–1681. Rooks IB, Ory HW, Ishak KG, et al. Epidemiology of hepatocellular adenoma. The role of oral contraceptive use. JAMA 1979; 242:644–648. Creagh TM, Rubin A, Evans DJ. Hepatic tumours induced by anabolic steroids in an athlete. J Clin Pathol 1988; 41:441–443.
58. Flejou JF, Barge J, Menu Y, et al. Liver adenomatosis. An entity distinct from liver adenoma? Gastroenterology 1985; 89:1132–1138. 59. Kozyraki R, Scoazec JY, Flejou JF, et al. Expression of cadherins and alpha-catenin in primary epithelial tumors of the liver. Gastroenterology 1996; 110:1137–1149. 60. Oberti F, Rifflet H, Flejou JF, et al. Association of hepatic adenomatosis and hepatoportal sclerosis in a woman with incontinentia pigmenti. Gastroenterol Clin Biol 1997; 21:147–151. 61. Kerlin P, Davis GL, McGill DB, et al. Hepatic adenoma and focal nodular hyperplasia: clinical, pathologic, and radiologic features. Gastroenterology 1983; 84:994–1002. 62. Chiche L, Dao T, Salame E, et al. Liver adenomatosis: reappraisal, diagnosis, and surgical management: eight new cases and review of the literature. Ann Surg 2000; 231:74–81. 63. Herman P, Pugliese V, Machado MA, et al. Hepatic adenoma and focal nodular hyperplasia: differential diagnosis and treatment. World J Surg 2000; 24:372–376. 64. Bartolozzi C, Lencioni R, Paolicchi A, et al. Differentiation of hepatocellular adenoma and focal nodular hyperplasia of the liver: comparison of power Doppler imaging and conventional color Doppler sonography. Eur Radiol 1997; 7:1410–1415. 65. Golli M, Van Nhieu JTV, Mathieu D, et al. Hepatocellular adenoma: color Doppler US and pathologic correlation. Radiology 1994; 190:741–744. 66. Nicolau C, Bru C. Focal liver lesions: evaluation with contrastenhanced ultrasonography. Abdom Imaging 2004; 29:348–359. 67. Grazioli L, Federle MP, Brancatelli G, et al. Hepatic adenomas: imaging and pathologic findings. Radiographic 2001; 21:877–894. 68. Chung KY, Mayo-Smith WW, Saini S, et al. Hepatocellular adenoma: MR imaging features with pathologic correlation. AJR Am J Roentgenol 1995; 165:303–308. 69. Coffin CM, Diche T, Mahfouz AE, et al. Benign and malignant hepatocellular tumours: evaluation of tumoral enhancement after mangafodipir trisodium injection on MR imaging. Eur Radiol 1999; 9:444–449. 70. Leese T, Farges O, Bismuth H. Liver cell adenomas. A 12-year surgical experience from a specialist hepato-biliary unit. Ann Surg 1988; 208:558–564. 71. Foster JH, Berman MM. The malignant transformation of liver cell adenomas. Arch Surg 1994; 129:712–717. 72. Hoso M, Terada T, Nakanuma Y. Partial nodular transformation of liver developing around intrahepatic portal venous emboli of hepatocellular carcinoma. Histopathology 1996; 29:580–582. 73. Nakanuma Y. Nodular regenerative hyperplasia of the liver: retrospective survey in autopsy series. J Clin Gastroenterol 1990; 12:46–50. 74. Wanless IR. Micronodular transformation (nodular regenerative hyperplasia) of the liver: a report of 64 cases among 2500 autopsies and a new classification of benign hepatocellular nodules. Hepatology 1990; 11:787–797. 75. McEntee MF, Wright KN, Wanless I, et al. Non cirrhotic portal hypertension and nodular regenerative hyperplasia of the liver in dogs with mucopolysaccharidosis type I. Hepatology 1998; 28:385–390. 76. Nzeako UC, Goodman ZD, Ishak KG. Hepatocellular carcinoma and nodular regenerative hyperplasia: possible pathogenetic relationship. Am J Gastroenterol 1996; 91:879–884. 77. Loinaz C, Colina F, Musella M, et al. Orthotopic liver transplantation in 4 patients with portal hypertension and noncirrhotic nodular liver. Hepatogastroenterology 1998; 45:1787–1794. 78. Dachman AH, Ros PR, Goodman ZD, et al. Nodular regenerative hyperplasia of the liver: clinical and radiologic observations. AJR Am J Roentgenol 1987; 148:717–722. 79. Patriarche C, Pelletier G, Attali P, et al. Ultrasonography, angiography, computed tomography and magnetic resonance in
Chapter 61 BENIGN LIVER TUMORS
nodular regenerative hyperplasia of the liver: report of a pseudotumoral case. Radiat Med 1988; 6:111–114. 80. Brancatelli G, Federle MP, Grazioli L et al. Large regenerative nodules in Budd–Chiari syndrome and other vascular disorders of the liver: CT and MR imaging findings with clinicopathologic correlation. AJR Am J Roentgenol 2002; 178: 877–883. 81. Soler R, Rodriguez E, Pombo F, et al. Benign regenerative nodules with copper accumulation in a case of chronic Budd–Chiari syndrome: CT and MR findings. Abdom Imaging 2000; 25:486–489.
82. Siegelman ES, Outwater EK, Furth EE, et al. MR imaging of hepatic nodular regenerative hyperplasia. J Magn Reson Imaging 1995; 5: 730–732. 83. Reimer P, Balzer T. Ferucarbotran (Resovist): a new clinically approved RES-specific contrast agent for contrast-enhanced MRI of the liver: properties, clinical development, and applications. Eur Radiol 2003; 13:1266–1276. 84. Al-Mukhaizeem KA, Rosenberg A, Sherker AH. Nodular regenerative hyperplasia of the liver: an under-recognized cause of portal hypertension in haematological disorders. Am J Hepatol 2004; 75:225–230.
1167
Section X: Tumours of the Liver
62
SURGERY OF LIVER TUMORS Henri Bismuth and Jina Krissat Abbreviations CT computed tomography IVC inferior vena cava
PVE RFAs
portal vein embolization radiofrequency ablations
INTRODUCTION Liver resection has evolved into the treatment of choice for selected benign lesions and malignant hepatobiliary disease. Since Wilson and Adson who, in 1976, showed that resection of colorectal liver metastases improve survival, it has been universally accepted that surgical resection is the only curative treatment option for patients with colorectal liver metastases. The widespread acceptance of liver resection for malignant tumors is due to two factors: (1) the reduction in operative mortality and morbidity; and (2) the proven impact on prognosis.1–4 As a result, over the past decade many large series have documented better results with operative mortality rates less than 5% in high-volume centers, even for the most extensive resections (Table 62-1). This has led to the expansion of indications of liver resection in metastatic disease other than colorectal cancer. The selection of the operative approach in every patient must therefore guarantee a clear margin and preserve a sufficient amount of liver parenchyma.
TLV
total liver volume
VASCULAR ISOLATION TECHNIQUES Prevention of intraoperative blood loss is of prime concern in any type of liver resection. Intraoperative blood loss has been shown to influence adversely the short-term prognosis and may be associated with an increased risk of recurrence in patients operated for a hepatobiliary malignancy through impairment of the patient’s immune response.8–10 Although some liver resections may be safely performed without vascular clamping (this is a basic requirement in living related liver transplantation), the main important principle in liver surgery is to minimize blood loss through control of the major vascular structures. This may be achieved in several ways that range from segmental portal control to total hepatic vascular occlusion (Figure 62-2). The common drawback of any clamping method is ischemic (and/or reperfusion) injury to the liver parenchyma. The type of vascular occlusion is usually selected according to the site of the tumor, the presence of underlying liver disease, and the patient’s cardiovascular status.
Hemihepatic Vascular Clamping
GENERAL PRINCIPLES Criteria of resectability in metastatic disease are not based on either the number or the size of tumors, but must guarantee complete tumor removal with a safe margin. This essential prerequisite determines the extent of the resection, whether simple excision, atypical resection, segmental resection, or extended resection. Advances in the methods of vascular control have limited morbidity and mortality to postoperative complications related to blood loss. Furthermore, improved understanding of hepatic physiology and segmental anatomy of the liver has led to advances in surgical techniques for liver resection.
ANATOMY AND CLASSIFICATION Precise knowledge of the surgical anatomy of the liver, its blood vessels, and biliary ducts is essential for the performance of any hepatectomy.5,6 The liver is divided in two lobes, each divided into segments supplied by branches of the portal triads and drained by hepatic veins (Figure 62-1). The preferred nomenclature is based on the anatomical descriptions of Couinaud in 1957 and then by Bismuth in 1982, and, more recently, this terminology has been reviewed by an international committee.7 In general, major resections involve more than three segments, extended resection four segments and more, and limited resections include fewer than three segments (Table 62-2).
This technique involves selective clamping of the arterial and venous inflow to the right or left hemi-liver. First described by Lortat-Jacob in 1952, the classic intrafascial or extrahepatic hilar approach involves dissection of the appropriate branch of the portal vein, hepatic artery, and the bile duct outside the liver parenchyma; after isolation, the vascular and biliary structures are cut and ligated individually. In the whole, sheath of a pedicle is dissected directly and isolates the portal elements of the supplied territory. The intrahepatic posterior approach or glissonean approach includes the dissection of the whole sheath of pedicle directly after division of a substantial amount of the hepatic tissue to reach the pedicle. The hilar plate has to be lowered to expose the left hepatic duct and the confluence of the bile ducts.11,12 The main advantage of this technique is that it avoids any ischemia to the remnant liver; there is no splanchnic congestion. The main drawback is the persistence of bleeding in the resection plane from the non-occluded liver.
Clamping of the Hepatic Pedicle This so-called Pringle maneuver, since its first description in 1908, involves total occlusion of hepatic blood inflow using either a tourniquet or a vascular clamp. Intermittent occlusion is the preferred method used by surgeons because it minimizes the ischemic injury during the liver surgery despite periods of ischemia. This involves in general 15–20 minutes of clamping followed by 5 minutes of unclamping. However, this procedure may still cause ischemic
1169
Section X. Tumours of the Liver
Table 62-1. Large Series of Resection of Colorectal Liver Metastases Authors
Country
Year
No. of patients
Perioperative mortality (%)
5-year survival (%)
Recurrence rate (%)
Fong et al30 Sugihara and Yamamoto4 DeMatteo et al.25 Seifert et al.26 Minagawa et al.27 Choti et al.3 Jarnagin et al.1
USA Japan USA Germany Japan USA USA
1999 2000 2000 2000 2000 2002 2002
1001 330 267 120 235 226 1021
2.8 0 0.4 5.8 0 0.9 2.6
37 44 43 31 38 40 —
71 51 62 73 62 —
Inferior vena cava
Median hepatic vein
Right hepatic vein
Left hepatic vein
Table 62-2. Nomenclature of Major Hepatic Resections Segments
Couinaud (1999)5
Brisbane (2000)7
5, 6, 7, 8 4, 5, 6, 7, 8
Right hepatectomy Right lobectomy
2, 3
Left lobectomy
2, 3, 4
Left hepatectomy
2, 3, 4, 5, 8
Extended left hepatectomy
Right hemihepatectomya Right trisectionectomy or extended right hepatectomy or extended right hemihepatectomya Left lateral sectionectomy or bisegmentectomy 2, 3 Left hepatectomy or left hemihepatectomy Left trisectionectomy or extended left hepatectomy or extended left hemihepatectomya
II VII
I
VIII
IV III
a
Stipulate ± segment 1.
VI V
Portal vein
Inferior vena cava
Faliciform ligament
Figure 62-1. Schematic illustration of the anatomy of the liver.
damage to the remaining liver, with a risk of poor postoperative outcome. Another drawback inherent to intermittent clamping is the blood loss during each period of reperfusion and the increased operative time.13–16 An alternative to intermittent clamping is ischemic preconditioning, in which a brief period of ischemia and reperfusion is applied prior to the prolonged ischemic insult.17–19 This ischemic preconditioning technique has its main advantages in patients with abnormal liver parenchyma. Whichever method is used, clamping of the hepatic pedicle has to be associated with reduced central venous pressure (usually below 5 mmHg) to allow outflow control by reducing the pressure gradient that promotes bleeding through the hepatic veins during parenchymal dissection. The main limit of this technique is in patients with associated hazards such as right heart failure and pulmonary artery hypertension.
Inflow and Outflow Control Hepatic vascular occlusion combines total inflow and outflow vascular occlusion of the liver.20 This procedure completely isolates the
1170
liver from the circulation and aims to prevent bleeding and air embolism from injuries to major hepatic veins and/or the inferior vena cava (IVC). In this technique the liver has to be totally mobilized and the IVC dissected above and below the liver (Figure 623). An initial portal trial exclusion must be undertaken for up to 5 minutes, after adequately expanding the blood volume to ensure that the procedure will be well tolerated. The hemodynamic intolerance to total hepatic vascular exclusion is defined as a decrease of systemic blood pressure below 100 mmHg. Clamps are placed in the following order: portal pedicle, the infrahepatic IVC, and suprahepatic IVC. During the procedure it is recommended that the central venous pressure should be maintained below 12 mmHg to avoid excessive backflow bleeding once the clamps are released. After completing the liver resection the clamps are then removed in the reverse order to which they were initially placed. This technique allows a blood-free transection of the liver parenchyma. Main indications of this technique are major resections for lesions involving the main hepatic veins or cavohepatic junction and when there is an intolerance to lower the central venous pressure. The main limit of this procedure is the low tolerance rate to clamping mainly due to the inability of the myocardium to maintain an adequate cardiac output. A way of avoiding this problem is by clamping the hepatic veins achieving a total vascular exclusion of the liver without interruption of the IVC.21 Although this is technically more demanding, it allows preservation of hemodynamic stability and the possibility of intermittent inflow occlusion. However, for lesions involving the cavohepatic junction, it can be difficult and dangerous to expose the hepatic veins.
Chapter 62 SURGERY OF LIVER TUMORS Figure 62-2. Modalities of vascular clamping. A outflow exclusion or Pringle maneuver; B selective hemihepatic clamping (extrafascial or intrafascial); C suprahilar clamping.
C
A B
Segmental Vascular Clamping This procedure has two advantages: (1) to minimize ischemic injury to the liver parenchyma; and (2) to delineate the portal territory of the tumor. It involves dissection of the arterial branch corresponding to the segment to be resected, and the segmental portal branch is identified by intraoperative ultrasound and punctured through the liver parenchyma.22 Next, a flexible guidewire is introduced into the portal branch to allow introduction of a balloon catheter down to the origin of the portal branch to be occluded (Figure 62-4). The corresponding arterial branch is clamped and the ischemic segment delineated; in some cases injection of methylene blue into the portal catheter will allow more precise identification of the segment. The main advantage of this technique is when a limited resection is indicated, especially in patients with a cirrhotic liver, and it allows a carcinogenic resection by removing the portal territory in patients presenting with hepatocellular carcinoma.
VALUE OF INTRAOPERATIVE ULTRASOUND The routine use of intraoperative ultrasound ensures the detection of all tumoral disease.22,23 A systematic anatomical study of the liver should be performed in all cases, beginning with the three hepatic veins, the portal bifurcation, and its branches. Only when the segmental architecture has been understood are the known lesions are examined, paying attention to their location and relation to the vascular structures. Scanning of the remaining parenchyma for occult lesions ends the examination of the liver. Today intraoperative ultra-
sound is essential for the planning of an adapted surgical procedure. It shows the relationship between a lesion and the neighboring vascular structures and may demonstrate that no clear margin is available without sacrificing a vessel to achieve a radical excision. The proportion of cases in which the operative strategy is modified by intraoperative ultrasound varies between 0 and 32%.24
SEGMENTAL LIVER RESECTION Because intraoperative ultrasound is the only method that allows the display of the vascular architecture of the liver, this has led to the evolution of segment-based resection, which is preferable from an oncological standpoint, avoiding the need of extensive “blind” resections, thus limiting the amount of resected non-tumoral parenchyma.25 With intraoperative ultrasound, the line of resection and the safety margin can be determined and verified as the resection proceeds. The application of the principle of segmental liver resection guided by introperative ultrasound has the advantage of diminishing the risks of leaving residual ischemic liver tissue that may necrose and become infected, and of leaving undrained bile ducts that may create a biliary fistula. Furthermore, if the blood supply to the territory to be resected is controlled before the resection, the blood loss may be significantly reduced. Additional reasons to practice segmental resections are the need to spare the maximum amount of liver tissue because of the reduced hepatic reserve in patients with underlying liver disease, and the increased opportunity to perform repeat resection in case of recurrent malignancy.
1171
Section X. Tumours of the Liver
Figure 62-4. Segmental vascular clamping by portal balloon occlusion.
EXTENDED RESECTIONS
Figure 62-3. Total vascular exclusion.
This approach is supported by the fact that a lesser resection does not adversely impact on the long-term outcome.21,25–27
CHANGING DEFINITION OF RESECTABILITY The concept of what constitutes “resectable” disease continues to evolve. Lesions that would previously be deemed too advanced may now considered resectable. However, despite significant improvement in techniques and outcomes of liver resection, only 20% of patients who present with liver metastases have resectable disease and this remains a major limiting factor.27–30 Lessons learned from liver transplantation and the development of innovative surgical techniques have made a curative surgical approach to virtually all liver tumors possible. Unresectability of liver lesions can be related to technical considerations: • large or ill-located tumors; • multiple bilateral lesions; • when resection of a large area, although necessary, cannot be performed because the amount of the remnant liver is likely to result in liver failure. Several methods can now be used to allow resection.
1172
A non-cirrhotic healthy liver may tolerate a resection of up to 80% of its volume. The enormous regenerative capacity enables a functional compensation within a few weeks and regeneration to 75% of the preoperative liver volume within 1 year.31,32 It is suggested that the safe limit of resection should leave 35% of functional liver parenchyma intact. Computed tomography (CT) scans now provide an accurate and reproducible method for preoperative liver volume calculation using three-dimensional CT volumetry. CT volumetry accurately predicts total liver volume (TLV) on the basis of body surface area or body weight. Complications associated with technical aspects of liver surgery have shifted towards those related on the basis of liver remnant volume after major hepatic resection. Acceptable outcomes have been reported with living donor and partial liver transplantation by use of graft-volume-to-standard-liver-volume ratios of 30%.33 Similarly, extended resections of up to 80% of the functional hepatic parenchyma can be performed with acceptable morbidity rates in patients with primary and metastatic disease. Recent studies revealed that future liver remnants less than 25% of TLV were associated with an increased incidence of postoperative hepatic dysfunction (increased bilirubin and prothrombin time) in patients with normal liver.34,35 The lessons learned from transplant volumetry and their application in liver surgery have widened the indications for extended hepatectomy for primary and secondary liver malignancies. Using the valuable information provided by careful calculation of future liver remnant liver volume, extended hepatectomy can nowadays be performed with a near-zero mortality rate in specialized centers.36 The better understanding of liver physiology and its tolerance to warm ischemia has led to the development of a more aggressive approach towards lesions involving both the liver and the IVC. The resected IVC can be repaired primarily or reconstructed with synthetic or autogenous grafts. In our series of 22 patients who needed a reconstruction of the IVC, the operative mortality was 4.5% with an actuarial 5-year survival rate of 34% (unpublished data). These results are compatible with those reported by recent studies.37–39 In some cases the tumor might involve all the major hepatic veins and the IVC. Techniques of in-situ perfusion (in which total vascular exclusion of the liver and
Chapter 62 SURGERY OF LIVER TUMORS
in-situ cooling of the liver by portal infusion of a cold hepatic preservation solution are performed with or without systemic venous bypass) or ex-vivo liver resection (involving removal of the entire liver, extracorporeal resection, and reimplantation) can be performed with early promising results in selected cases.40
PORTAL VEIN EMBOLIZATION One of the prerequisites for hepatic resection is that there is sufficient remaining parenchyma to avoid postoperative liver failure. Preoperative portal vein embolization (PVE) is a technique which induces an atrophy of the liver to be resected, leading to a compensatory hypertrophy of the remnant liver.41 There are two routine methods by which PVE can be performed: first, transileocolic portal embolization, in which a minilaparotomy is carried out so that a catheter can be introduced into the portal vein through a branch of the ileocolic vein; the percutaneous transhepatic portal embolization is performed under ultrasound guidance by puncturing a small portal vein branch so that a catheter can be inserted (Figure 62-5). Second, control venous portography is performed under fluoroscopy and a
guidewire is then placed into the main portal branch ipsilateral to the tumor. For embolization material we prefer to use a mixture of enbucrilate (histoacryl, Braun, Melsungen Laboratories, Melsungen, Germany) and lipiodol (lipiodol Ultrafluide, Guerbet, Aulnay-sousBois, France). After injection of the material we use a balloon occlusion to avoid reflux into the contralateral portal vein. The catheter is then removed while injecting 2 ml fibrin glue (Tissucol, Immuno, Vienna, Austria) into the needle tract under ultrasound guidance. The pressure in the main portal trunk is measured before and after embolization. The ratio of the volumes of the embolized and nonembolized parts of the liver is monitored by CT scans taken before and after embolization. In our previous study41 we have shown that PVE has increased the resectability rate by 63% in patients who were considered unresectable, with a chance of 5-year survival of 40%. Recent studies have achieved similar results.42,43 PVE is a safe, modern, minimally invasive technique which may not only increase the pool of patients who are candidates for liver resection but may also increase the safety of resection.44,45 Liver hypertrophy relies on the ability of the hepatocytes to dedifferentiate, expand clonally,
A
B
C D
Figure 62-5. Portal vein embolization. (A, B) Pre-embolization angiograms; (C, D) postembolization.
1173
Section X. Tumours of the Liver
and respond to hepatotrophic stimuli, which are predominantly portal in origin.46–48 A period of 4–5 weeks before the planned operation, is usually enough to achieve a sufficient increase in volume of the prospective remnant liver. In livers with impaired function this time interval can be increased up to 8 weeks.
TWO-STEP HEPATECTOMY The ability of the liver to regenerate may also be used in situations where extensive or multifocal disease would have previously precluded surgery. Staged resections can be performed without inducing liver insufficiency, by resecting large amounts of liver parenchyma in two separate sessions, allowing the liver time to regenerate between resections. This procedure of two-step hepatectomy is planned at the first hepatectomy and re-resection is usually performed within 3–6 months. During this time interval patients should be submitted to chemotherapy to limit the growth and spread of the remaining tumors.49 This strategy has been developed in our unit since 1992 and only has been systematically considered in recent years. The median survival was 44 months after resection from the time of diagnosis of liver metastases.
A
NEOADJUVANT CHEMOTHERAPY The fact that patients with liver metastases initially considered unsuitable for radical surgery can be rendered resectable after chemotherapy is a recent concept in the management of these patients. To date there are no randomized trials of such an approach but several groups have reported evidence of improved survival after different chemotherapeutic regimens.50–53 Over the past decade this has been the policy at Paul Brousse hospital, France. Patients have been managed in a protocol of neoadjuvant chemotherapy using a chronomodulated chemotherapy with 5-fluorouracil, folinic acid, oxaliplatin, and/or irinotecan. This has allowed the administration of doses of medication that could be given with fewer toxic side effects and larger than those usually administered. This protocol was given in an outpatient setting, minimizing the discomfort to these patients. Using this aggressive approach, liver resection has been routinely reconsidered in all cases of objective response to the treatment, permitting the tumor resectability rate to be increased from 20 to 30% (Figure 62-6). The overall 5-year patient survival was 34%.52 These results emphasize the importance of reconsidering liver resection in patients who were not candidates for primary surgery at first, but who responded well to chemotherapy. Independent of the type of chemotherapy used, the main point is that this approach offers a chance of cure to patients initially thought to have unresectable disease.
RESULTS AND PROGNOSIS The long-term outcome achieved after liver resection is often compromised by tumor recurrence. About 60% of patients develop recurrence and approximately one-third of them have isolated liver recurrence.52,54 Treatment of recurrent cancer after liver resection is usually palliative, and include chemotherapy, with disappointing results.53 However, with the improvement of surgical techniques and with the increased range of sophisticated non-operative modalities, it is possible to have a more radical approach.
1174
B
Figure 62-6. Downstaging by neoadjuvant chemotherapy. Metastatic deposit in the liver before A and after B systemic neoadjuvant chemotherapy with 5fluorouracil and oxaliplatin.
REPEAT HEPATECTOMY During the past decade, repeat resection has been more frequently performed in patients with isolated liver recurrence. This is due to recent advances in the early detection of recurrences, the efficacy of chemotherapy, and improvements in the expertise of liver surgery. Published series have shown that patients submitted to a repeat resection for liver recurrence can achieve similar results to those of initial liver resection, with a reported survival benefit of 30–40% at 5 years.55,56 However, tumor recurrence is frequent after either a
Chapter 62 SURGERY OF LIVER TUMORS
first or a second resection. Is a third hepatectomy indicated in cases of isolated liver recurrence? We have recently reported the results achieved by this approach.57 The benefit provided by a third resection (whenever technically possible) is demonstrated by the 5-year survival rate of 32% with no operative mortality. These results compare favorably with those achieved after a first hepatectomy. Repeat hepatectomy can provide long-term survival rates similar to those of primary liver resections, and can be carried out safely without additional perioperative mortality.
RADIOFREQUENCY Repeat resection is only possible in about one-third of patients. In the rest, either the liver is diffusely involved, or there is extrahepatic relapse. Such patients may be offered further systemic chemotherapy. Non-operative ablative modalities such as radiofrequency ablations (RFAs)58–61 are now available, and may be very useful in selected cases. RFA is a new modern technique, which allows localized thermal treatment, designed to produce tumor destruction by heating tumor tissue to temperatures that exceed 50°C. RFA can be performed percutaneously, laparoscopically, or at the time of laparotomy. Percutaneous RFA has the advantage of minimizing morbidity and is performed under local anesthetic and conscious sedation. It is currently the preferred approach in our institution. After a small nick is made in the skin, a needle is introduced in a selected area of the tumor under ultrasound guidance. In our experience we have initially used both the RITA Medical System unit and, more recently, the RF 2000 generator system produced by Radiotherapeutics.62 The device consists of a generator that supplies up to 100 W power, a LeVeen monopolar array needle electrode, and a dispersive electrode pad applied to the patient’s skin. The electrode is a 15-gauge, 12–15-cm long insulated cannula that contains 10 individual hook-shaped electrode arms that are
deployed in situ after ultrasound-guided placement of the needle electrode into the liver tumor (Figure 62-7). The maximum deployment diameter for the hooks is 3–5 cm. Optimal positioning of the electrode allows complete destruction of tumor. Lesions less than 3 cm require only one or two treatments. Lesions larger than 3 cm may require multiple overlapping fields to encompass the entire tumor. At the completion of the ablation process the prongs are retracted and the needle tract is cauterized as it is removed, to minimize the risk of tumor spread and bleeding. CT scan is used to assess the completeness of the treatment. Our current practice is to recommend retreatment for patients who have evidence of residual tumor based on a 1-month post-RFA CT scan. Early results have shown that RFA appears to be a promising technique with few complications and a low mortality rate.61 Whether RFA is curative is still unknown.62–66 Elias et al.64 have reported a series of 47 patients with liver recurrence after hepatectomy treated by percutaneous radiofrequency. The 55% 2-year survival obtained was similar to that obtained after repeat hepatectomy. However, the median follow-up was only 14.4 months and the number of patients is still too small. Current RFA technology limits its use to patients with four tumors or fewer, which are less than 5 cm in diameter. In addition to tumor size, other factors influence the efficacy of this technique, including tumor location and proximity to large vessels. Some lesions may be difficult to access. Proximity to large vessels dissipates flow of heat generation and may result in incomplete necrosis. This may be solved by repositioning the probe and completing repetitive ablation sessions until cytodestructive temperatures are obtained in all areas of the lesion or by temporarily occluding hepatic inflow. At the present time, RFA appears to be a safe, minimally invasive treatment option for patients with inoperable tumors. However, long-term results and future large randomized trials are necessary to define its precise indications. We do
Figure 62-7. Radiofrequency ablation. (A) A 3-cm lesion in segment 2.
A
1175
Section X. Tumours of the Liver
Figure 62-7, cont’d. B sonographic image of an echogenic wire guided into the lesion; C radiofrequency ablation needle open: note the tip of the hook implanted inside the lesion.
B
C
not think that RFA should be used as an alternative treatment for patients who would otherwise be candidates for curative hepatic resection.
CONCLUSION At the present time a surgical approach, whenever possible, seems to be the most favorable choice of treatment for liver tumors. The
1176
perioperative mortality rate is low and very extensive resections can be carried out safely. Advances in surgical techniques, better knowledge of liver anatomy, improved capabilities of identifying liver tumors, and improved critical care management have allowed many patients to undergo liver resections with minimal morbidity and virtually no mortality. However, patients presenting with liver tumors are best managed in specialized centers where surgeons are familiar with all aspects of both complex hepatobiliary surgery and liver transplantation and cared for by multidisciplinary teams that possess surgical and oncological expertise.
Chapter 62 SURGERY OF LIVER TUMORS
REFERENCES 1. Jarnagin WR, Gonen M, Fong Y, et al. Improvement in perioperative outcome after hepatic resection. Analysis of 1803 consecutive cases over the past decade. Ann Surg 2002; 236:397–407. 2. Holt DR, Van Thiel D, Edelstein S, Brems JJ. Hepatic resections. Arch Surg 2000; 135:1353–1358. 3. Choti M, Sitzmann JV, Tiburi MF, et al. Trends in long-term survival following liver resection for hepatic colorectal metastases. Ann Surg 2002; 235:759–766. 4. Sugihara K, Yamamoto J. Surgical treatment of colorectal liver metastases. Ann Chirurg Gynaecol 2000; 89:221–224. 5. Couinaud C. Liver anatomy: portal (and suprahepatic) or biliary segmentation. Digest Surg 1999; 16:449–467. 6. Sutherland F, Harris J. Claude Couinaud: a passion for the liver. Arch Surg 2002; 137:1305–1310. 7. The Terminology Committee of the IHPBA. The Brisbane 2000 terminology of hepatic anatomy and resections. Hepatopancreatobiliary Surg 2000; 2:333–339. 8. Kwon AH, Matsui Y, Kamiyama Y. Perioperative blood transfusion in hepatocellular carcinomas: influence of immunologic profile and recurrence free survival. Cancer 2001; 91:771–778. 9. Kooby DA, Stockman J, Ben-Porat L, et al. Influence of transfusions on perioperative and long-term prognosis outcome in patients following hepatic resection for colorectal metastases. Ann Surg 2003; 237:860–870 10. Arnoletti JP, Brodsky J. Reduction of transfusion requirements during major hepatic resection for metastatic disease. Surgery 1999; 125:166–171. 11. Launois B, Tay KH, Meunier B. The posterior intrahepatic approach to liver resection. Hepatopancreatobiliary Surg 1999; 1:209–214. 12. Figueras J, Lopez-Ben S, Llado L, et al. Hilar dissection versus the “Glissonean” approach and stapling of the hepatic pedicle for major hepatectomies: a prospective randomized trial. Ann Surg 2003; 238:111–119. 13. Belghiti J. Vascular isolation techniques in liver resection. In: Blumgart LH, Fong Y, eds. Surgery of the liver and biliary tract, 3rd edn. Philadelphia: Churchill Livingstone; 2000: 1715–1724. 14. Buell JF, Koffron A, Yoshida A, et al. Is any method of vascular control superior in hepatic resection of metastatic cancers? Longmire clamping, Pringle maneuver, and total vascular isolation. Arch Surg 2001; 136:569–575. 15. Wu CC, Yeh DC, Ho WM, et al. Occlusion of hepatic blood inflow for complex central liver resections in cirrhotic patients. Arch Surg 2002; 137:1369–1376. 16. Muratore A, Ribero D, Ferrero A, et al. Prospective randomized study of steroids in the prevention of ischemic injury during hepatic resection with pedicle clamping. Br J Surg 2003; 90:17–22. 17. Clavien PA, Yadav S, Sindram D, Bentley RC. Protective effects of ischemic preconditioning for liver resection performed under inflow occlusion in humans. Ann Surg 2000; 232:155–162. 18. Chen H, Merchant NB, Didolkar MS. Hepatic resection using intermittent vascular inflow occlusion and low central venous pressure anesthesia improves morbidity and mortality. J Gastrointest Surg 2000; 4:162–167. 19. Clavien PA, Selzner M, Rudiger HA, et al. A prospective randomized study in 100 consecutive patients undergoing major liver resection with versus without ischemic preconditioning. Ann Surg 2003; 238:843–852. 20. Zografos GN, Kakaviatos ND, Skiathitis S, Habib N. Total vascular exclusion for liver resections: pro and cons. J Surg Oncol1999; 72:50–55.
21. Cherqui D, Malassagne B, Colau PI. Hepatic vascular exclusion with preservation of the caval flow for liver resections. Ann Surg 1999; 230:24–30. 22. Bismuth H, Majno PE, Castaing D, Azoulay D. Intraoperative ultrasound and liver resection and liver transplantation. In: Blumgart LH, Fong Y, eds. Surgery of the liver and the biliary tract, 3rd edn. Philadelphia: Churchill Livingstone; 2000:1785–1795. 23. Luck AJ, Maddern GJ. Intraoperative abdominal ultrasonography. Br J Surg 1999; 86:5–16. 24. Oehme J, Bernhardt H, Fritze F, et al. The value of intraoperative ultrasound in liver surgery. In: Kockerling F, Schwartz SI, eds. Liver surgery. Operative techniques and avoidance of complications. Heidelberg: JA Barth; 2001:65–68. 25. De Matteo RP, Palese C, Jarnagin WR, et al. Anatomic segmental hepatic resection is superior to wedge resection as an oncologic operation for colorectal liver metastases. J Gastrointest Surg 2000; 4:178–184. 26. Seifert JK, Bottger TC, Weigel TF, et al. Prognostic factors following liver resection for hepatic metastases from colorectal cancer. Hepatogastroenterology 2000; 47:239–246. 27. Minagawa M, Makuuchi M, Torzilli G, et al. Extension of the frontiers of surgical indications in the treatment of liver metastases from colorectal cancer. Ann Surg 2000; 4:487–499. 28. Chouillard E, Cherqui D, Tayar C, et al. Anatomical bi- and trisegmentectomies as alternatives to extensive liver resections. Ann Surg 2003; 238:29–34. 29. Bismuth H, Krissat J, Adam R. Evidence for metastasectomy for colorectal liver metastases. In: Cunningham D, Allen-Mersh TG, Miles A, eds. The effective management of metastatic colorectal cancer. Oxford: Radcliffe Medical Press; 1999:12–24. 30. Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer. Analysis of 1001 consecutive cases. Ann Surg 1999; 230:309–321. 31. Miyagawa S, Kawasaki S, Noike T, et al. Liver regeneration after extended right hemihepatectomy in patients with hilar or diffuse bile duct carcinoma. Hepatogastroenterology 1999; 46:364–368. 32. Abdalla EK, Denys A, Chevalier P, et al. Total and segmental liver volume variations: implications for liver surgery. Surgery 2004; 135:404–409. 33. Shoup M, Gonen M, D’Angelica M, et al. Volumetric analysis predicts hepatic dysfunction in patients undergoing major liver resection. J Gastrointest Surg 2003; 7:325–330. 34. Leelaudomlipi S, Sugawara Y, Kaneko J, et al. Volumetric analysis of liver segments in 155 living donors. Liver Transplant 2002; 8:612–614. 35. Varotti G, Gondolesi GE, Goldman J, et al. Anatomic variations in right liver living donors. J Am Coll Surg 2004; 198:577–582. 36. Vauthey JN, Pawlik TM, Abdalla EK, et al. Is extended hepatectomy for hepatobiliary malignancy justified? Ann Surg 2004; 239:722–732. 37. Hemming AW, Reed AI, Langham MR, et al. Hepatic vein reconstruction for resection of hepatic tumors. Ann Surg 2002; 235:850–858. 38. Arii S, Teramoto K, Kawamura T, et al. Significance of hepatic resection combined with inferior vena cava resection and its reconstruction with expanded polytetrafluoroethylene for treatment of liver tumors. J Am Coll Surg 2003; 2:243–249. 39. Hemming AW, Reed AI, Langham MR, et al. Combined resection of the liver and inferior vena cava for hepatic malignancy. Ann Surg 2004; 239:712–721. 40. Lodge J PA, Ammori BJ, Prasad KR, Bellamy MC. Ex vivo and in situ resection of inferior vena cava with hepatectomy for colorectal metastases. Ann Surg 2000; 231:471–479. 41. Azoulay D, Casataing D, Smail A, et al. Resection of nonresectable liver metastases from colorectal cancer after
1177
Section X. Tumours of the Liver
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
1178
percutaneous portal vein embolization. Ann Surg 2000; 231:480–486. Elias D, Cavalcanti A, DeBaere T, Roche A. Resultats carcinologiques à long terme des hepatectomies realisées après embolisation portale selective. Ann Chir 1999; 53:559–564. Abdalla EK, Barnett CC, Doherty D, et al. Extended hepatectomy in patients with hepatobiliary malignancies with and without preoperative portal vein embolization. Arch Surg 2002; 137:675–681. Shimada R, Imamura H, Nikayama A, et al. Changes in blood flow and function of the liver after right portal vein embolization. Arch Surg 2002; 137:1384–1388. Wakabayashi H, Ishimura K, Okano K, et al. Application of preoperative portal vein embolization before major hepatic resection in patients with normal or abnormal liver parenchyma. Surgery 2002; 131:26–33. Tominaga YK, Sugimoto T, Iwasaki T, et al. Preoperative hepatic venous embolization for partial hepatectomy combined with segmental resection of major hepatic vein. Br J Surg 2002; 89:63–69. Seymour K, Charnley RM, Rose JDG, et al. Preoperative portal vein embolization for primary and metastatic liver tumours: volume effects, efficacy, complications and short-term outcome. Hepatopancreatobiliary Surg 2002; 4:21–28. Goffette P, Lerut J, Laterre PF, Gigot JF. Preoperative portal vein embolization for extension of hepatectomy indications. Vasc Intervent Radiol 2000; 11:265. Adam R, Laurent A, Azoulay D, et al. Two-stage hepatectomy: a planned strategy to treat irresectable liver tumors. Ann Surg 2000; 232:777–784. Link KH, Pillasch J, Formentini A, Sunelaitis E. Downstaging by regional chemotherapy of non resectable isolated colorectal liver metastases. Eur J Surg Oncol 1999; 25:381–388. Shankar A, Renaut AJ, Lederman J, et al. Neoadjuvant therapy improves respectability rates for colorectal cancer. Br J Cancer 1999; 80:110. Adam R, Avisar E, Ariche A, et al. Five-year survival following hepatic resection after neoadjuvant therapy for nonresectable colorectal liver metastases. Ann Oncol 2001; 8:347–353. Tanaka K, Adam R, Shimada H, et al. Role of neoadjuvant chemotherapy in the treatment of multiple colorectal metastases to the liver. Br J Surg 2003; 90:963–969.
54. Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer. Analysis of 1001 consecutive cases. Ann Surg 1999; 230:309–321. 55. Hobday TJ, Kugler JW, Mahoney MR. Long term survivors of metastatic colorectal cancer treated with chemotherapy only: a North Central Cancer Treatment group review. J Clin Oncol 2002; 20:4574–4580. 56. Petrowsky H, Gonen M, Jarnagin W, et al. Second liver resections are safe and effective treatment for recurrent hepatic metastases from colorectal cancer. A bi-institutional study. Ann Surg 2002; 6:863–871. 57. Adam R, Pascal G, Azoulay D, et al. Liver resection for colorectal metastases. The third hepatectomy. Ann Surg 2003; 238:871–884. 58. Curley SA, Izzo F, Delrio P, et al. Radiofrequency ablation of unresectable primary and metastatic hepatic malignancies. Results in 123 patients. Ann Surg 1999; 1:1–8. 59. Taylor I, Gillams AR. Colorectal liver metastases: alternatives to resection. J Roy Soc Med 2000; 93:576–579. 60. Iannitti DA, Dupuy DE, Mayo-Smith WW, Murphy B. Hepatic radiofrequency ablation. Arch Surg 2002; 137:422–427. 61. Curley SA. Radiofrequency ablation of malignant liver tumours. Oncologist 2001; 6:14–23. 62. Adam R, Hagopian EJ, Linhares M, et al. A comparison of percutaneous cryosurgery and percutaneous radiofrequency for unresectable hepatic malignancies. Arch Surg 2002; 137 : 1332–38 63. Mulier S, Mulier P, Miao Y, et al. Complications of radiofrequency coagulation of liver tumours. Br J Surg 2002; 89:1206–1222. 64. Elias D, DeBaere T, Smayra T, et al. Radiofrequency ablation of liver tumour recurrence. Br J Surg 2002; 89:752–756. 65. Oshowo A, Gillams A, Harrison E, et al. Comparison of resection and radiofrequency ablation for treatment of solitary colorectal liver metastases. Br J Surg 2003; 90:1240–1243. 66. Kaneko T, Sugimoto H, Tezel E, et al. Radiofrequency ablation therapy for malignant hepatic tumours: comparison of two procedures. Hepatopancreatobiliary Surg 2003; 5:19–26.
Section XI: Diseases of the Biliary Tract
63
THE MEDICAL MANAGEMENT OF GALLSTONES Veronica A. Arteaga, Hans Fromm Abbreviations AP alkaline phosphatase CT computed tomography ERCP endoscopic retrograde cholangiopancreatography ESWL extracorporeal shock-wave lithotripsy
EUS FDA GGT HIDA
endoscopic US US food and drug administration g-glutamyl transpeptidase cholecystokinin scintigram
INTRODUCTION As a common medical condition that is increasing worldwide, gallstone disease presents a significant burden for the public health system. About 10–15% of the population of the US and western Europe has gallstones,1–3 which seem to be even more common in Latin America.2 About 90% of gallstones are composed predominantly of cholesterol. The remaining 10% are pigment stones, made up mainly of calcium bilirubinate. The growing prevalence of cholelithiasis appears to be related mainly to the propagation of cholesterol gallstones resulting from the spreading influence of the socalled western lifestyle.3 Overeating and a sedentary lifestyle, with the consequent development of obesity and diabetes mellitus, promote disturbances in cholesterol metabolism. Bile becomes supersaturated in cholesterol and gallstones develop. Our understanding of the metabolic disturbances underlying the pathogenesis of cholesterol gallstones and the recognition of its epidemiologic link to lifestyle provide opportunities for prevention. Therefore the discussion of gallstone management will consider preventive measures, in addition to the more traditional methods of management.
PATHOGENESIS AND RISK FACTORS OF GALLSTONES CHOLESTEROL GALLSTONES Cholesterol gallstones form on the basis of cholesterol-supersaturated bile.4 The pathophysiologic chain of events leading to cholesterol cholelithiasis is thought to be initiated by a decrease in propulsive intestinal motility, possibly because of an increased deposition of dietary cholesterol in small intestinal smooth muscle membrane.5,6 It is interesting to note that this pathogenic mechanism of decreased smooth muscle contractility appears to be identical to that underlying the development of decreased gallbladder emptying.7 Decreased intestinal transit leads to prolonged exposure of bile acids to bacteria, with increased dehydroxylation of the trihydroxy bile acid cholic acid to its more hydrophobic dihydroxy metabolite, deoxycholic acid. Because of its hydrophobicity, deoxy-
MRC MTBE UDCA US
magnetic resonance cholangiography methyl-tert butyl ether ursodeoxycholic acid ultrasonography
cholic acid stimulates the hepatic secretion of cholesterol, mainly by facilitating its solubilization in the canalicular membrane. Deoxycholic acid also increases the proportion of arachidonate-rich species of phosphatidylcholine in bile. As a result, more mucin, which has strong nucleating properties, is secreted by the gallbladder mucosa. The increase of cholesterol in gallbladder bile, in the form of both a supersaturated solution and crystals, sets the stage for deposition of cholesterol in gallbladder smooth muscle membranes, with the consequent impairment of gallbladder contractility. In addition to gallbladder mucin and decreased gallbladder emptying, cholesterol crystallization and gallstone growth are facilitated by the secretion of a number of pronucleating proteins, such as apoproteins A1 and A2, aminopeptidase N, phospholipase A2 and C, haptoglobin, and immunoglobulins. The main risk of cholesterol gallstones relates to an increase in the biliary secretion of cholesterol.4 Risk factors include obesity, rapid weight loss, hypertriglyceridemia, low high-density lipoprotein levels, increasing age, the presence of the apo E4 isoform, and parity. Another important risk factor that is also based primarily on the presence of increased cholesterol secretion is genetic predisposition in certain ethnic groups, such as Native Americans and northern Europeans, and in patients with a family history of gallstones.8–11 The power of gallstone (lith) genes is also evidenced by recent studies in inbred mice, which are characterized by a very high incidence of cholesterol gallstones.12 Under certain conditions nutritional changes, which are associated with diminished stimulation of gallbladder contraction, present major risks for cholesterol gallstone formation. Examples are both parenteral nutrition and rapid weightloss diets that contain less than 10 g of fat per day.13
PIGMENT STONES Pigment stones form when calcium bilirubinate precipitates in bile because of an increased presence of unconjugated bilirubin, which is less soluble than is its conjugated moiety. Risk factors are hemolysis, cirrhosis of the liver, inflammatory bowel disease, and ileal resection. The increased risk of pigment stones in the latter two conditions appears to be related to an increased intestinal absorption of unconjugated bilirubin.
1181
Section XI. Diseases of the Biliary Tract
ACALCULOUS GALLBLADDER DISEASE Although this review concerns the management of gallstones, a few words should be said about acalculous gallbladder disease. The reason for directing attention to this entity relates to the problem of managing patients with symptoms indistinguishable from those of biliary pain in whom no gallstones can be found. The pain may be caused by passage of the gallstone through the bile ducts, by microlithiasis, or by a recently defined muscle defect in the gallbladder.14–17 Recent studies by Amaral and colleagues17 suggest that acalculous gallbladder disease may be related to a dysfunction of the contractile apparatus of the gallbladder. The contraction induced by cholecystokinin was markedly reduced both in vivo and in vitro. This contractility defect could not be corrected by G-protein activators or with the second messenger 1,2-dioctanoyl-Sn-glycerol. The clinical significance of these findings remains uncertain, although patients with this disorder have been reported to experience relief of pain after cholecystectomy.18 How decreased motility in an acalculous gallbladder could cause biliary pain is not clear. Therefore further studies are needed, and it appears premature to endorse the recommendation made by surgeons to subject patients to cholecystectomy if their gallbladder ejection fraction is less than 35%.18
CLINICAL PRESENTATION SYMPTOMS Most gallstones cause no symptoms or complications. In 70–80% of patients cholelithiasis is asymptomatic (i.e. biliary pain, the only symptom characteristic for active gallstone disease, is absent).19 Therefore, a careful history from the patient is essential for distinguishing biliary pain from non-specific symptoms. Although often referred to as biliary colic, biliary pain is not colicky. Rather, it presents as a steady, severe pain in the epigastrium or right upper abdomen that lasts at least 30 minutes, typically with a crescendo–plateau–decrescendo sequence of intensity and with radiation into the right scapula. The pain often wakes the patient at night and is frequently associated with nausea and vomiting.
COMPLICATIONS The main complications of gallstones are acute cholecystitis, common bile duct stones, acute pancreatitis, and ascending cholangitis. The diagnosis of acute cholecystitis is based largely on clinical symptoms and signs, a careful physical examination, and certain laboratory findings. Murphy’s sign, a distinct tenderness in the area of the gallbladder on inspiration, is probably the most reliable finding pointing toward an acute inflammation of the gallbladder. Although ultrasonographic changes, such as increased thickness of the gallbladder wall or a pericholecystic fluid collection, are consistent with the diagnosis of acute cholecystitis, they are not diagnostic per se. Similar caution should be used in the interpretation of nucleotide imaging studies of the gallbladder, such as a cholecystokinin scintigram (HIDA) scan. Although a normal study does not rule out acute gallbladder disease, an abnormal finding (i.e. no appearance of the nucleotide in the gallbladder) is diagnostically useful only in conjunction with the described typical symptoms and signs. Acute cholecystitis with or without the additional complication of common bile
1182
duct obstruction must be suspected if biliary pain develops in association with fever, leukocytosis with a left shift, and a transient rise in alanine aminotransferase, aspartate aminotransferase, and gglutamyl transpeptidase (GGT). These enzyme abnormalities may be followed by alkaline phosphatase (AP) and bilirubin elevations. Very high serum amylase and lipase values may also be present and are indicative of acute pancreatitis, which can be serious, often requiring endoscopic retrograde cholangiopancreatography (ERCP) and, if a stone is found in the bile duct, endoscopic sphincterotomy and stone extraction. In a small percentage of cases choledocholithiasis can evolve into ascending cholangitis, which characteristically presents with rigors, fever, and jaundice. With few exceptions, complications of gallstones are preceded by repeated episodes of biliary pain.20 Exceptions are elderly patients or patients immunocompromised by conditions such as poorly controlled diabetes mellitus or chronic renal insufficiency, in whom biliary complications can develop without pain or typical findings on physical examination. In these cases liver test abnormalities, usually consisting of bilirubin, AP, GGT, and transaminase elevations, prompt the appropriate diagnostic studies (i.e. ultrasonography (US), computed tomography (CT) scan, magnetic resonance imaging, or ERCP). Unfortunately, cholecystectomy is too often performed unnecessarily for symptoms not related to gallstones, such as abdominal discomfort resulting from irritable bowel syndrome or dyspepsia, which is easily distinguishable from biliary pain.
DIAGNOSIS So-called real-time transabdominal US, which is about 90% sensitive in detecting stones larger than 2 mm in the gallbladder, represents the standard procedure for the diagnosis of gallbladder stones and dilated bile ducts. Oral cholecystography, which involves visualization of the gallbladder with an orally administered iodine-based contrast medium, has a sensitivity for stones similar to that of US. However, it is barely used any longer, as it has several disadvantages in comparison to US. These include the lack of visualization of the gallbladder if there is significant impairment of hepatic excretory function (cholestasis with a serum bilirubin level above 2 mg%), the association with a moderate degree of radiation exposure, and a small incidence of hypersensitivity reactions to iodine. However, the sensitivity of US to detect stones in the common bile duct is low, i.e. in the order of only about 30%. Endoscopic US (EUS), in which a US device can be brought endoscopically into close proximity to the gallbladder and common bile duct, is extremely sensitive for stones in both these structures. In a recent study, the sensitivity for the detection of gallbladder stones was 100% with EUS versus 84% with conventional transabdominal US. EUS was also found to be as sensitive, specific and accurate as ERCP for the diagnosis of common bile duct stones.21 Magnetic resonance cholangiography (MRC) has emerged as another very reliable non-invasive method for the detection and exclusion of common bile duct stones. In one representative study, the sensitivity was 100%, the specificity 95.6%, the positive predictive value 92.6% and the negative predictive value 100%.22
Chapter 63 THE MEDICAL MANAGEMENT OF GALLSTONES
MANAGEMENT EXPECTANT MANAGEMENT AND PREVENTIVE MEASURES The annual risk of patients with asymptomatic cholelithiasis to develop biliary pain is 1–4%. In the majority of cases significant symptoms or complications from the gallstones never develop. No treatment, therefore, represents an important management strategy for carriers of asymptomatic gallstones.22,23 The risk of symptomatic gallstones may be decreased by physical activity and by increased coffee consumption.24,25 Studies also suggest that long-term treatment with ursodeoxycholic acid (UDCA) may markedly decrease the recurrence of biliary pain, the development of complications, and the consequent need for cholecystectomy.26
CHOLECYSTECTOMY Cholecystectomy, which in the vast majority of cases is carried out laparoscopically (laparoscopic cholecystectomy), is the main treatment for symptomatic gallstones. The question is whether surgery should be considered after the first episode of biliary pain or after the second attack. The life expectancy has been estimated to be virtually the same with either strategy. Even if surgery would be delayed until a complication occurs, the life expectancy would not decrease by more than 23–25 days compared to operative treatment immediately after the first episode of biliary pain. The strategy of delaying surgery until the second episode of biliary pain appears particularly appropriate for middle-aged or older patients, because in approximately 30% biliary symptoms do not recur for 10 years.27 The age of the patient is a factor that needs to be weighed carefully in the decision between surgical and medical therapy for symptomatic gallstones. The cumulative death rate for a 30-year-old undergoing immediate cholecystectomy has been estimated to be 0.11%. Elderly patients, particularly those 65 or over, are at increased risk from surgical treatment. About 20% of abdominal procedures in patients aged 80 or more are hepatobiliary. In 6.7% of about 5900 laparoscopic cholecystectomies performed at a Veterans’ Administration Medical Center from 1991 to 2001 the patients were 65 years or older.28 Interestingly, 64% of the patients operated on in this age group were male. Therefore, biliary complications may be more common in men than in women. The risk of cholecystectomy, performed mainly laparoscopically, is increased in the elderly. In the above-cited Veterans’ Administration study the mortality was 0.6% in patients aged 65–69, 2.2% in those 70–79 years of age, and 2.0% in those aged 80 and older.28 Complications, mainly of a cardiac and a pulmonary nature requiring intervention, increased with age from 1.8% (65–69 years) to 3.3% (70–79 years) and 11.5% (80 years or older). The cholecystectomy-related mortality in elderly patients reported in the literature varies, and has been often higher than that indicated in the above described study by Bingener et al.28 The cumulative lifetime risk of mortality from gallstones is 2.3%. The experience of the surgeon influences the risk of cholecystectomy. Intraoperative complications are both more common and more serious in laparoscopic than in open cholecystectomy. For example, the incidence of bile duct injury in laparoscopic cholecys-
tectomy, which on average is around 0.25%, has been reported to be as high as 0.95%.29 As far as long-term sequelae of cholecystectomy are concerned, studies suggest that it increases the risk of the development of adenocarcinoma of the esophagus, small bowel and right colon.30,31 The enhanced risk may be related to an increased exposure of the mucosa to bile, in particular to the bile acid deoxycholic acid.
ORAL BILE ACID DISSOLUTION THERAPY Medical management plays a major role in gallstone patients who have biliary pain for the first time, who are at increased risk from surgery, or who do not want to have surgery. The choice of medical treatment is based on selection criteria related to the patient’s condition and gallstone features, to the familiarity of the treating physician with the different therapeutic options, and to the availability of the treatment method that appears most promising. Oral dissolution therapy with UDCA can be very effective in selected patients with small, non-calcified gallstones.23,32–34 UDCA is therapeutically attractive for at least two reasons. First, it is extraordinarily safe (i.e. virtually free of significant side effects). Second, as has already been mentioned, UDCA appears to markedly decrease the incidence of both gallstone complications and recurrent biliary pain, effects that are independent of its cholelitholytic action. Patients considered for UDCA dissolution treatment should have uncomplicated gallstone disease with infrequent episodes of biliary pain. Gallstone calcifications can be excluded with a plain Xray of the abdomen, although a CT scan is substantially more sensitive. However, a CT scan may not be cost-effective because minor calcifications not detected by conventional radiography do not preclude gallstone dissolution.35 UDCA is taken twice daily with meals at a total dose of 8–10 mg/kg/day. The best treatment results – complete dissolution in up to 90% of patients – are obtained in gallstones not exceeding 5 mm in diameter. The success rate decreases with increasing stone size. Gallstones larger than 1 cm rarely dissolve. If stones do dissolve, they do so at an average rate of 1 mm per month. A drawback of UDCA therapy is that gallstones recur in about 30% of patients who had single stones versus approximately 50% of those with multiple stones. Most recurrences are seen during the first years after dissolution, at an annual rate of about 10%. Fortunately, most recurrences are asymptomatic. Although it decreases gallstone recurrence, maintenance treatment with UDCA is not considered cost-effective.
EXTRACORPOREAL SHOCK-WAVE LITHOTRIPSY Extracorporeal shock-wave lithotripsy (ESWL) for gallstones, which is being used in a few medical centers in Europe, is not approved by the US Food and Drug Administration (FDA). Similar to oral bile acid dissolution therapy, ESWL is indicated in selected patients with uncomplicated symptomatic gallstones.36 The gallstones must be non-calcified. The highest success rate is achieved in single stones up to 2 cm in size. Although patients receive adjuvant therapy with UDCA at a dose of 8–10 mg/kg/day to promote the dissolution of the stone fragments, studies have shown that stones that are broken up to very small fragments often clear the gallbladder spontaneously. Optimal fragmentation of the gallstones to particles not larger than 1–2 mm considerably improves the success of ESWL. With obser-
1183
Section XI. Diseases of the Biliary Tract
vation of the described optimal selection criteria, about 80% of patients become stone free within 12 months. The rate of complications, such as acute pancreatitis and common bile duct obstruction by gallstone fragments, is extremely low.
TOPICAL DISSOLUTION THERAPY Gallstones can be topically dissolved using either methyl-tert butyl ether (MTBE) or ethyl propionate. The solvent is infused into the gallbladder through a pigtail catheter, which can be placed either endoscopically or transhepatically. Although the success rate can be high in properly selected patients with non-calcified gallstones, the treatment is experimental and rarely used. In addition to the special precautions required, because of the fire hazard and the risk of other side effects caused by the volatility of MTBE, the topical dissolution procedure is labor intensive unless a specially designed pump and computerized delivery system, currently under study, are used.37 Although ethyl propionate is considered to be safer than MTBE, experience with this compound in human studies is limited.
GALLSTONE RECURRENCE A drawback of all medical methods is that gallstones recur in about 30% of patients who had single stones versus approximately 50% of those with multiple stones.38 Most recurrences are seen during the first years after successful treatment, at an annual rate of about 10%. If gallstones recur after UDCA therapy, most do so without biliary pain. However, the incidence of symptomatic recurrences appears to be higher after ESWL. This may possibly be because patients selected for ESWL were more symptomatic than those chosen for UDCA therapy. Although it reduces gallstone recurrence, maintenance treatment with UDCA is not considered cost-effective.
MICROLITHIASIS Microlithiasis constitutes a clinically very important stage of gallstone disease. It occurs either in the form of biliary crystals that can be detected only by microscopic examination of a bile sample, or as sonographically evident sludge in the gallbladder.14–16 Although probably involved in the pathogenesis of biliary pain and complications in gallstone disease, in general biliary microcrystals have been shown to be the cause of so-called idiopathic acute pancreatitis in about 70% of cases. Gallbladder sludge is often found in hospitalized patients in association with prolonged fasting and parenteral nutrition. As with typical cholelithiasis, these crystals can be birefringent cholesterol monohydrate crystals, birefringent calcium carbonate microspheroliths, or calcium bilirubinate granules. The management of microlithiasis-related acute pancreatitis is controversial. Although removal of the gallbladder as the locus of crystal formation is likely to be curative, expectant management, endoscopic sphincterotomy or ursodiol treatment may be reasonable alternatives in view of the lack of data regarding the recurrence rate of microlithiasis-induced acute pancreatitis. The latter is likely to vary in different patient populations and associated disease conditions.
1184
GALLBLADDER POLYPS Gallbladder polyps encompass all mucosal projections into the gallbladder lumen, i.e. cholesterolosis, adenomyomas, inflammatory polyps, adenomas, and other miscellaneous lesions of polypoid appearance.39 Most polyps are non-neoplastic and are formed by cholesterolosis and adenomyomatosis. Cholesterolosis is characterized by mucosal villous hyperplasia with excessive accumulation of cholesterol esters within epithelial macrophages. Adenomyomatosis consists of hyperplastic changes marked by excessive proliferation of surface epithelium, with invaginations into a thickened muscularis propria, which sonographically gives the appearance of a thickened gallbladder wall with intramural diverticula. Usually, neither condition causes biliary pain or idiopathic pancreatitis. In most cases cholesterolosis and adenomyomatosis requires no specific treatment. However, cholecystectomy is considered advisable if polyps are larger than 10 mm because of their increased risk to become either adenomatous or malignant. Recent studies suggest that EUS may be helpful in evaluating gallbladder polyps.40 Correct diagnoses were obtained by EUS in 49 of 50 (98%) patients with cholesterol polyps, in all six patients with adenomyomatosis, and in 22 of 24 (91.7%) of those with carcinoma. However, EUS proved to be less reliable in distinguishing between hyperplastic, inflammatory, adenomatous and malignant polyps: 77.8% of benign lesions were misdiagnosed as carcinomas. A disturbing finding in this Japanese study (although not specifically discussed in the paper) was that six out of 38 patients (16%) with gallbladder polyps in the 6–10 mm size category were carcinomas (in the remaining 32 cases, 30 were cholesterol polyps, one inflammatory and one adenomatous). This report contained only two patients who had polyps that were 5 mm or smaller. In both cases they were cholesterol polyps. Gallbladder polyps in the 6–10 mm size category therefore require either careful follow-up evaluation or a cholecystectomy. In a recent review, the difficulty in differentiating between benign and malignant polyps was further emphasized.41 Although the majority of gallbladder polyps are benign, malignant transformation was considered to be a concern. Several features, including patient age, polyp size and number, and rapid growth of polyps, were described as important discriminating features between benign and malignant polyps. Based on the evidence highlighted in their review, the authors recommended cholecystectomy in symptomatic patients, as well as in asymptomatic individuals over 50 years of age, or those whose polyps are solitary, more than 10 mm in diameter, or associated with gallstones or polyp growth on serial US. Novel imaging techniques, including EUS and enhanced CT, were considered to be helpful in the differential diagnosis of these lesions and to permit expectant management.
SUMMARY Gallstones are not only very prevalent in most parts of the world, but also a significant burden for the public health system. Considerable progress has been made both in the understanding of the pathogenesis of gallstones and in their management. Although cholecystectomy is standard therapy for symptomatic gallstones,
Chapter 63 THE MEDICAL MANAGEMENT OF GALLSTONES
medical treatment alternatives are appropriate in selected patients who either do not want surgery or are at increased risk from cholecystectomy. Gallstone patients are often subjected unnecessarily to cholecystectomy for gastrointestinal symptoms not related to gallstones. Most gallstones are asymptomatic. If symptoms are absent or non-specific, cholelithiasis should be managed expectantly. Selected patients with non-calcified small gallstones may benefit from oral dissolution therapy with ursodiol. ESWL, which may soon be available in the US, is particularly successful in 5–20-mm solitary non-calcified gallbladder stones. The role of potentially effective topical dissolution methods in the management of gallstones requires further study. Biliary microlithiasis represents a clinically very important stage of gallstone disease. Its management as the most common cause of so-called idiopathic acute pancreatitis is controversial and in need of further investigation. Although usually benign and asymptomatic, gallbladder polyps require cholecystectomy under certain conditions that increase the risk of malignancy.
REFERENCES 1. Sama C, Morselli Labate AM, et al. Epidemiology and natural history of gallstone disease. Semin Liver Dis 1990;10:149–158. 2. Diehl AK. Epidemiology and natural history of gallstone disease. Gastroenterol Clin North Am 1991;20:1–19. 3. Everhart JE, Khare M, Hill M, et al. Prevalence and ethnic differences in gallbladder disease in the United States. Gastroenterology 1999;117:632–639. 4. Apstein MD, Carey MC. Pathogenesis of cholesterol gallstones: a parsimonious hypothesis. Eur J Invest 1996;26:343–352. 5. Xu Q, Scott RB, Tan DTM, et al. Slow intestinal transit: a motor disorder contributing to cholesterol gallstone formation in the ground squirrel. Hepatology 1996;23:1664–1672. 6. Van Erpecum KJ, Van Berge-Henegouwen GP. Gallstones: an intestinal disease? Gut 1999;44:435–438. 7. Chen Q, Amaral J, Biancani P, et al. Excess membrane cholesterol alters human gallbladder muscle contractility and membrane fluidity. Gastroenterology 1999;116:678–685. 8. Sampliner RE, Bennett PH, Comess LJ, et al. Gallbladder disease in Pima Indians. Demonstration of high prevalence and early onset by cholecystography. N Engl J Med 1970;283: 1358–1364. 9. Van der Linden W, Simonsen N. The familial occurrence of gallstone disease. Incidence in parents of young patients. Hum Hered 1973;23:123–127. 10. Gilat T, Feldman C, Halpern Z, et al. An increased familial frequency of gallstones. Gastroenterology 1983;84:242–246. 11. Jorgensen T. Gallstones in a Danish population: familial occurrence and social factors. J Biosoc Sci 1988;20:11–120. 12. Wang DQ-H, Paigen B, Carey MC. Phenotypic characterization of Lith genes that determine susceptibility to cholesterol cholelithiasis in inbred mice: physical chemistry of gallbladder bile. J Lipid Res 1997;38:1395–1411. 13. Festi D, Colecchia A, Orsini M, et al. Gallbladder motility and gallstone formation in obese patients following very low calorie diets. Use it (fat) to lose it (well). Int J Obesity Relat Metab Disord 1998;22:592–600. 14. Ros E, Navarro S, Bru C, et al. Occult microlithiasis in idiopathic acute pancreatitis: prevention of relapses by cholecystectomy or ursodeoxycholic acid therapy. Gastroenterology 1991;101:1701–1709. 15. Lee SP, Nicholls JF, Park HZ. Biliary sludge as a cause of acute pancreatitis. N Engl J Med 1992;326:589–593.
16. Ko CW, Sekijima JH, Lee SP. Biliary sludge. Ann Intern Med 1999;130:301–311. 17. Amaral J, Xiao Z-L, Chen Q, et al. Gallbladder muscle dysfunction in patients with chronic acalculous disease. Gastroenterology 2001;120:506–511. 18. Mulholland MW. Progress in understanding of acalculous gallbladder disease. Gastroenterology 2001;120:570–572. 19. Ricci G, the GREPCO Group (Rome Group for the Epidemiology and Prevention of Cholelithiasis). The GREPO research programmes: aims and prevalence data. In: Cappocaccia L, Ricci G, Angelico F, et al., eds. Epidemiology and prevention of gallstone disease. Lancaster: MTP Press, 1984: 9. 20. Ransohoff DF, Gracie WA. Treatment of gallstones. Ann Intern Med 1993;119:606–619. 21. Liu CL, Lo CM, Chan JK, et al. Detection of choledocholithiasis by EUS in acute pancreatitis: a prospective evaluation in 100 consecutive patients. Gastrointest Endosc 2001; 54:325–330. 22. Demartines N, Eisner L, Schnabel K, et al. Evaluation of magnetic resonance cholangiography in the management of bile duct stones. Arch Surg 2000;135:148–152. 23. Howard DE, Fromm H. Overview of non-surgical therapy of gallstones. Gastroenterol Clin North Am 1999;28:133–144. 24. Leitzmann MF, Rimm EB, Willett WC, et al. Recreational physical activity and the risk of cholecystectomy in women. N Engl J Med 1999;341:777–784. 25. Leitzmann MF, Willett WC, Rimm EB, et al. A prospective study of coffee consumption and the risk of symptomatic gallstone disease in men. JAMA 1999;281:2106–2112. 26. Tomida S, Abei M, Yamaguchi T, et al. Long-term ursodeoxycholic acid therapy is associated with reduced risk of biliary pain and acute cholecystitis in patients with gallbladder stones: a cohort analysis. Hepatology 1999;30:6–13. 27. Ransohoff D. Management of patients with symptomatic gallstones: a quantitative analysis. Am J Med 1990; 88:154–160. 28. Bingener J, Richards ML, Schwesinger WH, et al. Laparoscopic cholecystectomy for elderly patients: gold standard for golden years? Arch Surg 2003;138:531–535. 29. Targarona EM, Marco C, Balague C, et al. How, when, and why bile duct injury occurs: a comparison between open and laparoscopic cholecystectomy. Surg Endosc 1998;12:322. 30. Lagergren J, Weimin YE, Ekbom A. Intestinal cancer after cholecystectomy: Is bile involved in carcinogenesis? Gastroenterology 2001;121:542–547. 31. Freedman J, Weimin YE, Naslund E, et al. Association between cholecystectomy and adenocarcinoma of the esophagus. Gastroenterology 2001;121:548–553. 32. Fromm H, Roat JW, Gonzalez V, et al. Comparative efficacy and side effects of ursodeoxycholic and chenodeoxycholic acids in dissolving gallstones: a double blind controlled study. Gastroenterology 1983;85:1257–1264. 33. Fromm H. Gallstone dissolution therapy. Current status and future prospects. Gastroenterology 1986;91:1560–1567. 34. Levenson DE, Fromm H. Management of gallbladder disease. In: Zakim D, Boyer TD, eds. Hepatology. A textbook of liver disease. Philadelphia: WB Saunders, 1996: 1877–1897. 35. Sarva RP, Farivar S, Fromm H, et al. Study of the sensitivity and specificity of computerized tomography in the detection of calcified gallstones which appear radiolucent by conventional roentgenography. Gastrointest Radiol 1981;6:165–167. 36. Mulagha E, Fromm H. Extracorporeal shock wave lithotripsy of gallstones revisited: current status and future promises. J Gastroenterol Hepatol 2000;15:239–243s. 37. Zakko S, Hofmann AF. Microprocessor-assisted solvent-transfer system for gallstone dissolution. In vitro and in vivo validation. Gastroenterology 1990;99:1807–1813.
1185
Section XI. Diseases of the Biliary Tract
38. Fromm H, Malavolti M. Gallstone recurrence after medical therapy. Viewpoints Dig Dis 1992;24:1–7. 39. Owen CC, Bilhartz LE. Gallbladder polyps, cholesterolosis, adenomyomatosis, and acute acalculous cholecystitis. [Review]. Semin Gastrointest Dis 2003;14:178–188.
1186
40. Azuma T, Yoshikawa T, Araida T, et al. Differential diagnosis of polypoid lesions of the gallbladder by endoscopic ultrasonography. Am J Surg 2001;181:65–70. 41. Myers RP, Shaffer EA, Beck PL. Gallbladder polyps: epidemiology, natural history and management. Can J Gastroenterol 2002;6:187–194.
Section XI: Diseases of the Biliary Tract
64
NON-SURGICAL MANAGEMENT OF GALLSTONES John T. Cunningham Abbreviations CDS common duct stones CT computed tomography EBD Endoscopic balloon dilation ERC endoscopic retrograde cholangiography
ERCP ES
endoscopic retrograde cholangiopancreatography endoscopic biliary sphincterotomy
INTRODUCTION The management of gallstones is changing due to a better appreciation of the natural history of gallstone disease and the development of new technologies now available for the management of gallstones, whether in the gallbladder or the biliary tree. The emphasis of this chapter will be on both endoscopic and percutaneous approaches to stone management as other chapters will handle the medical and surgical methods. Since the first almost simultaneous reports of endoscopic retrograde cholangiopancreatography (ERCP) and endoscopic sphincterotomy (ES) by Classen et al.1 and Kawai et al.2 in 1974, there has been an explosion in techniques to extract biliary stones.
CHOLELITHIASIS It is well known that gallbladder stones will develop in 10–15% of the western population and that the majority of these will be clinically silent.3,4 Patients with symptomatic cholelithiasis are best managed by laparoscopic cholecystectomy. The population in which there is no consensus is the patient with both gallbladder and common duct stones (CDS) in whom the gallbladder stones are silent. If the initial event was either symptomatic choledocholithiasis or pancreatitis in a patient with known gallbladder stones, and the common duct is cleared of any residual stones, is cholecystectomy mandatory? Does elective cholecystectomy reduce the future risk of biliary symptoms? Two studies found no difference in the incidence in recurrent biliary symptoms, whether the gallbladder was removed prior to endoscopic therapy, just after, or left in situ.5,6 Removing the gallbladder in patients with CDS does not reduce the rate of recurrence of CDS, suggesting that the primary source of the stones is the common duct rather than the gallbladder.5 The major issue is whether the risk of cholecystitis in patients whose gallbladders are not removed is great enough to warrant surgery in all patients. In one series of 371 patients studied retrospectively, the incidence of cholecystitis was 5.9% and choledocholithiasis 9.7%.7 Hence, in patients with a high operative risk, delaying cholecystectomy is a reasonable approach, whereas in those who are otherwise healthy, an elective cholecystectomy is probably the preferred approach.
EUS GGT ULN
Endoscopic ultrasonography g-glutamyltranspeptidase upper limit of normal
There may be a limited role for ERCP in the management of symptomatic gallbladder stones. Patient with end-stage liver disease with a high model for end-stage liver disease score who present with recurrent biliary colic or acute cholecystitis are high-risk patients for any surgical intervention. An alternative to surgery is endoscopy to access the common bile duct and place a hydrophilic tipped guidewire through the cystic duct into the gallbladder. Over the guidewire double pigtail stents, either 5F or 7F, of sufficient length to leave the proximal tip in the gallbladder and the distal end in the duodenum, are passed leading to successful decompression of the gallbladder8 (Figure 64-1).
PERCUTANEOUS CHOLECYSTOSTOMY The morbidity and mortality of laparoscopic and open cholecystectomy increase with age, comorbid conditions, and in the presence of acute cholecystitis:9–12 the mortality of emergent cholecystectomy reaches 19% in the elderly.9 The conversion rate from laparoscopic to open cholecystectomy also increases with age.10 Interventional radiologic techniques have allowed direct access to the gallbladder by direct puncture through the abdominal wall using either ultrasound or computed tomography (CT) guidance.11,13–15 The technique of puncture (Figure 64-2) is to enter through the anterior abdominal wall with a 16–22-gauge needle either directly into the gallbladder or by entering the gallbladder by passing through the liver. After entering the gallbladder a guidewire is then inserted through the needle and catheters of varying size can then be passed over the guidewire and the gallbladder decompressed. The goal is to allow stabilization of the patient and later elective cholecystectomy, preferably laparoscopic. In one series of high-risk patients researchers were able to perform delayed laparoscopic surgery in 50%, with a 15% conversion to open surgery. The mortality in this high-risk group was 26%.15 For patients who remain at a high operative risk it is possible to allow the cholecystostomy tract to mature and use electrohydraulic or mechanical lithotripsy to fragment and extract the calculi via the tract.13,15 Tract maturation usually requires
1187
Section XI. Diseases of the Biliary Tract
Transperitoneal approach
Transhepatic puncture
Stones Gallbladder
Liver
Common bile duct Spine
Figure 64-1. Double pigtail stent with lateral pigtail in the gallbladder and medial pigtail in the duodenal lumen.
Figure 64-2. Schematic demonstrating the transhepatic versus the transperitoneal approach to cholecystostomy. After needle access, a guidewire is placed though the needle into the gallbladder, and pigtail catheters are inserted over the wire.
2 weeks for the transhepatic route or 3 weeks for the transperitoneal approach.16
CHOLEDOCHOLITHIASIS DIAGNOSIS The first issue in patients with biliary symptoms is whether or not there is a CDS present and how will that patient be best managed. For the patient with an intact gallbladder and who will undergo an operative procedure, the major issue is whether there should be an intervention directed toward the common duct before the cholecystectomy or afterward. In the patient with ongoing cholangitis or acute biliary pancreatitis with persistent elevation of liver tests and ongoing pain, there are several prospective studies that demonstrate an improvement in morbidity and mortality with early intervention with ERCP versus conservative management.17–19 For the less acutely ill patient there is no hard-and-fast rule and some of these decisions will depend upon the expertise of the interventionalist and the ease and comfort of the surgeon in the performance of operative cholangiography, if indicated. Multiple investigators have tried to stratify the relative risk of finding a CDS based on a variety of non-invasive clinical or laboratory parameters, including liver tests alone20 or in conjunction with ultrasonographic findings.21–23 The finding of an alkaline phosphatase greater than three times the upper limit of normal (ULN), elevation of the g-glutamyltranspeptidase (GGT) or common bile duct ≥ 8 mm were the best predictors of finding a CDS, with the incidence of stones in the high-risk group ranging from 77% to 95%.22,23 In another study, using a CBD caliber of ≤ 5 mm the incidence was 6% and if >5 mm the incidence was 37.5%.24
1188
Transabdominal ultrasonography, though reasonable for detection of duct caliber, has a poor detection rate for choledocholithiasis,25,26 with false-negative rates of 70% depending on the stone size. Endoscopic ultrasonography (EUS) has been used as a predictor for CDS (Figure 64-3) in patients presenting with a suspected diagnosis of biliary pancreatitis.27–30 EUS has a sensitivity of 97%, which is very similar to ERCP in the detection of CDS and overall accuracy of 98%.28 A negative EUS accurately predicted the absence of stones in 98% of cases. Most of the original EUS studies have involved the radial endoscope, but the results being reported with the linear array endoscopes suggest a similar sensitivity.31,32 ERCP is used as the gold standard for choledocholithiasis but most experienced endoscopists know that ERCP can miss stones, particularly small stones in a large duct, and this was confirmed in studies use a catheter probe as an adjunct to ERCP.33,34 With this approach 55 stones seen by ERCP were identified plus four missed stones, with two false-positives.29 Another study looked at patients from high to low risk, who underwent ERCP with the use of C-arm fluoroscopy. Intraductal ultrasound was done in all patients when the ERCP was negative or questionable for stones. They were able to eliminate the need for sphincterotomy in 10% of the cohort and because of positive findings 15% received an ES that they would otherwise not have had. The use of a catheter probe adds only a few minutes to the procedure, and can be done safely over a guidewire after the common bile duct has been accessed during the diagnostic portion of the ERCP. Endoscopic and catheter ultrasound is not universally available. An approach that might be considered would be to stratify the
Chapter 64 NON-SURGICAL MANAGEMENT OF GALLSTONES Figure 64-3. Endoscopic ultrasound demonstrating a large stone with an acoustic shadow in a dilated common bile duct.
patient according to relative risk and base the diagnostic approach accordingly (Figure 64-4). Patients with ongoing cholangitis or gallstone pancreatitis could go directly to ERCP. Those at intermediate risk could either have a preoperative EUS, or go to cholecystectomy with intraoperative cholangiography and postoperative ERCP if choledocholithiasis is found. Many of these decision paths will be dependent on the skill of the therapeutic endoscopist and the surgeon. The issue of the cost of an additional procedure can be justified if used to exclude the use of the higher-risk ERCP in a population of patients at intermediate or low risk.35
ENDOSCOPIC RETROGRADE CHOLANGIOPANCREATOGRAPHY Since the first description of endoscopic biliary sphincterotomy (ES), endoscopic retrograde cholangiography (ERC) has become one of the primary methods for management of CDSs.1,2,17,18,19,36 Sphincterotomy with stone extraction has been accepted as the primary method for stone removal and is relatively safe in the hands of the experienced practitioner with an acceptable morbidity and mortality.36 The management of a stone crowning in the papilla is easily approached by either repeated suctioning and allowing the stone to pass spontaneously37 or by performing a papillotomy over the stone with a needle knife, thus avoiding the pancreatic duct (Figure 64-5). The basic ES is carried out after selective cannulation of the biliary tree has revealed a retained stone, either with a diagnostic catheter with a guidewire left in place or after direct cannulation with a sphincterotome. Standard ES is carried out using a monopolar
Symptomatic gallstones Level of Risk
Low Risk Normal LFTs CBD < 6 mm Pancreatitis resolved
Intermediate Risk CBD 6–8 mm Abnormal LFTs Pancreatitis resolved
Laparoscopic cholecystectomy
High Risk cholangitis Pancreatitis Abnormal LFTs & CBD > 8 mm
Pre-operative ERCP
Endoscopic ultrasound
Laparoscopic cholecystectomy with operative cholangiogram + ERCP
Laparoscopic cholecystectomy
Figure 64-4. Suggested flow chart for use of endoscopic retrograde cholangiopancreatography (ERCP) in suspected choledocholithiasis. LFTs, liver function tests; CBD, common bile duct.
1189
Section XI. Diseases of the Biliary Tract
Figure 64-5. A A stone is impacted in the papillary orifice. The needle knife will incise the papilla over the stone. B The sphincterotomy has been completed and the stone has been released.
A
B
current, incising the roof of the biliary sphincter to a sufficient degree to extract the largest stone, or until the incision is carried to the point where the common bile duct enters the duodenal wall (Figure 64-6). There is an old dictum that it is safe to cut to the first transverse fold crossing the papilla, but the magnitude of the intraduodenal portion of the papillary mechanism is highly variable and is the limiting factor in the length of the incision. Once the incision is complete, the stone removal is done with an extraction balloon or basket (Figure 64-7). Care must be exercised during the initial extraction sequence to avoid stone impaction in the papilla. Testing the magnitude of the incision with the balloon will give the operator an idea of the adequacy of the sphincterotomy relative to the size of the stone to be extracted. Stones that are too large to be extracted intact will need to be fragmented. Mechanical lithotripsy is successful in fragmenting stones in a high percentage of patients38–40 (Figure 64-7). In one series over 5 years involving 698 cases, lithotripsy was required in 15%. The success rate was 89%, and 15 of the patients required multiple sessions to clear the duct.38 Stones that are not successfully extracted can have stents placed to decompress the biliary tree. Some authors report that this technique will be associated with spontaneous stone passage or additional fragmentation, allowing stone extraction at follow up ERCP in 44% of patients.41,42 Using a stent as a long-term solution in patients at very high operative risk is also associated with a significant incidence of cholangitis – 40–63%.43,44 Several problems arose in attempting to interpret these studies: the expertise of the examiner, and the size of stones which were not extractable, which were >12 mm in one study with high subsequent extraction rate41 versus median size of 30 mm in another study.44 A second problem is that the series with the highest failure rate is prospective and has a longer duration of follow-up.44
ENDOSCOPIC BALLOON DILATION Sphincterotomy is no longer the sole technique for alteration of the papillary mechanism to allow for stone extraction. Endoscopic balloon dilation (EBD) of the papilla has become an alternative to facilitate stone removal.45–48 The results of four prospective comparative studies are included in Table 64-1.45,46,49,50 The supposition
1190
Figure 64-6. Sphincterotomy with standard papillotome with purulent bile passing through the incision after the stone has been disimpacted.
is that with EBD there will be less long-term impairment of the papilla, as demonstrated by transient pancreatobiliary reflux which had decreased by 3 months and then resolved.51 Similar improvement has been noted after ES also,52 but duodenobiliary reflux persists longer after ES than EDB.51,52 Maintenance of sphincter function was demonstrated by normal hilum-to-duodenal transit of a radioactive tracer in EBD patients, versus accelerated transit in ES patients at 1-year follow-up.53 Normal smooth-muscle architecture has been observed in 9/10 patients after EDB, but residual inflammation and fibrosis were common.54 The technique for EDB is to cannulate the papilla in the standard fashion, and to place a guidewire deep into the biliary tree. Over the wire a dilating balloon is inserted into a papilla and inflated to an appropriate size to allow stone extraction (Figure 64-8). The size of the dilation is dependent on stone size and varies from 6 to 12 mm. Stones larger than 12 mm may require lithotripsy and the method of extraction is the same as with sphincterotomy. The question is whether this technique is at least as safe as or safer than endo-
Chapter 64 NON-SURGICAL MANAGEMENT OF GALLSTONES Figure 64-7. Combined procedure. Percutaneous transhepatic cholangiography guidewire used to pull the endoscope to the papilla, after sphincterotomy. The stone has been trapped in a four-wire lithotripsy basket, and the metal sheath has closed the basket on to the stone B; the basket is subsequently closed over the stone, crushing it into fragments A.
A
B
Table 64-1. Results of the Consensus Conference on the Complications of Endoscopic Retrograde Cholangiopancreatography Complication
Mild
Moderate
Severe
Bleeding
Hemoglobin fall <3 g No transfusion Amylase >3 times normal and hospitalization <3 days Possible, slight, medical therapy for <4 days >38°C or 24–48 h
Transfusion <5 units, no other intervention Clinical pancreatitis and hospitalization 4–10 days
Transfusion >5 units or angiographic or surgical intervention Clinical pancreatitis and hospitalization >10 days or other intervention
Definite treatment for 7–10 days
Definite treatment >10 days or intervention required Septic shock or surgical intervention
Pancreatitis
Perforation Infection
Basket impaction
Spontaneous release or repeat endoscopy
Febrile or septic illness for >3 days or endoscopic or radiologic intervention Radiologic intervention
Surgical intervention
(Data from Cotton PB, Lehman G, Vennes J, et al. Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 1991; 37:383–393.56)
scopic sphincterotomy. Multiple studies suggest that it is safe;48,49 however, most of the reports are either retrospective or not prospective controlled trials. One prospective study showed a significantly higher incidence of hemorrhage with endoscopic sphincterotomy (26%) versus balloon dilation (2%) with no appreciable pancreatitis.45 However this is a much higher bleeding rate than reported in the largest prospective study on the complications of endoscopic sphincterotomy and a study looking at the complication of bleeding associated with sphincterotomy.36,55 The largest prospective study comparing these two modalities found the incidence of bleeding as a complication as defined by the consensus conference was 0%, but “self-limited” or bleeding which required some form of endoscopic intervention occurred in 27% of the ES group and 10.5% of the balloon dilation group. Unfortunately, the definition of a “complication” is not specifically delineated in many of these studies and they fail to use the definitions of complications that were developed by a consensus conference on complications of endoscopic sphincterotomy (Table 64-1).56 The largest US study, which included academic and private-practice endoscopists, and the European study
both came to the conclusion that balloon dilation should be avoided in routine practice.46,50 EBD may have a role in several clinical situations, such as in patients with Billroth II anatomy where complete sphincterotomy is technically more difficult due to the inverted position of the major papilla, and in patients with cirrhosis who have a high rate of hemorrhage at endoscopic sphincterotomy.36,55 One study performed EBD in 21 patients with cirrhosis and coagulopathy and compared them to 20 historical control patients undergoing ES at the same institution.57 The bleeding rate was 30% in the ES group and 0% in the EBD group and all but one of the bleeding cases occurred in Child–Pugh class C patients.
COMPLEX STONE EXTRACTION In any large series, there will be a small percentage of intraductal calculi which are either inaccessible due to failed cannulation or prior surgery, such as a Billroth II or Roux en Y anastomosis. Another common cause is failure of standard methods of stone extraction. These patients will either have to undergo surgical extraction, either
1191
Section XI. Diseases of the Biliary Tract
Figure 64-8. Endoscopic papillary balloon dilation. The stone is located above the balloon next to the wire, and the balloon waist is in the papilla.
laparoscopically or open, or use a variety of specialized techniques to access and/or fragment the stones. One of the commonest techniques is the combined percutaneous–endoscopic approach, where the interventional radiologist will access the biliary tree by a percutaneous transhepatic puncture using fluoroscopic guidance and place a guidewire through the needle and across the major papilla. The endoscopist can then grasp the guidewire, pull it through the endoscope, and use it to perform a sphincterotomy if access to the biliary tree was the only issue.58–60 More complex cases, such as Roux en Y bilioenteric anastomoses or presence of large stones with prior failed endoscopic extraction, present more challenging issues.61,62 Extracorporeal lithotripsy is a proven modality, but is not approved for use in the USA.5,62 Direct ductoscopy, either with a mother–baby scope37,41,62,63 or percutaneous cholangioscopy with a small-caliber endoscope, provides an attractive, non-surgical alternative for extraction of large or complex stones which fail standard endoscopic or percutaneous extraction.61,64,65 An interventional radiologist can use mechanical methods to fragment the stone and push the fragments into the duodenum through a prior sphincterotomy or balloon-dilate the sphincter during the same procedure.66 Stones that fail mechanical lithotripsy can be fragmented with a piezoelectric lithotripsy probe or with a holmium-YAG laser37,61,64–66 (Figure 64-9).
Figure 64-9. A Transhepatic cholangiogram in a patient with left intrahepatic duct stones and a choledochoduodenostomy due to duct trauma during laparoscopic cholecystectomy. B Ureteroscope passed through a 12 French biliary catheter. C Endoscopic view of stone and electrohydraulic lithotripsy probe. D Duct cleared of all stones.
A
B
C
D
1192
Chapter 64 NON-SURGICAL MANAGEMENT OF GALLSTONES
ERCP IN PREGNANCY Gallbladder crystals form “sludge” which is common in pregnancy, is usually asymptomatic, but is associated with a higher incidence of stone formation, especially if it appears during the first trimester.67 Symptomatic gallstone disease is a not uncommon occurrence in pregnancy and the physician has additional concerns, including the risk of any intervention on both the mother and the fetus, as well as the effects of the radiation exposure on the fetus if fluoroscopy is needed during ERCP. Surgical cholecystectomy is considered relatively safe for the mother with both the open and laparoscopic technique, but increased fetal wastage is reported with open procedures,68 whereas the relative safety of laparoscopy has been demonstrated in multiple studies.69,70 Management of CDS presents another level of complexity. There are data on the amount of radiation exposure that is considered safe.71 There are only a few large series on ERCP in pregnancy.70,72,73 The patients in these series who underwent ERCP and ES for presumed or proven gallstone disease had no increased incidence of complications or apparent untoward effects on the fetus. The administration of standard medication for conscious sedation was well tolerated by mother and fetus. Fetal monitoring was used in only one study.73 The patient is placed in the standard prone position or the left lateral decubitus position and a lead shield is placed between the cathode-ray tube and the patient to protect the fetus from ionizing radiation (Figure 64-10). We place a dosimeter over the sacrum and anterior abdomen to document the radiation dose to the skin surface. The recorded dose was 59 mRem, which was well within the current recommended guidelines.71,74 A phantom model was used in another study73 to estimate the fetal dose which ranged from 310 ± 160 mrad. It is best to minimize fluoroscopy time and avoid fixed exposures to reduce radiation to the minimal amount necessary to complete the procedure successfully.
ERCP IN PEDIATRICS ERCP has been an accepted modality in adults for 30 years. Its use in the pediatric population has been a little slower to evolve but guide-
lines have been developed.75 The indication for ERCP in the pediatric population is so infrequent that most pediatric gastroenterologists either have not been trained or cannot maintain proficiency for this examination, and the procedures are performed by adult endoscopists. There are several situations that are unique in this population. Infants may need to be done with a pediatric-size endoscope with 8.0-mm outside diameter and 2.0-mm operating channel.76 The only problem is that the channel size limits the range of therapeutic instruments that can be used. The procedure is technically similar to adults, with the exception that a higher percentage of cases are performed with deep sedation or general anesthesia and general anesthesia can be safely administered in the ERCP suite.77 Most reports include patients 18 years of age and under,78–80 and do not separate adolescents, who may act more like the adult population, from children. However, one study found no difference in success or complications rates between children and adolescents.81 None of these studies report a complication rate any higher than in the adult population, especially when the indication was for suspected choledocholithiasis. Another series advises caution, reporting post-ERCP pancreatitis in 28.6% of pediatric patients, and 40% when the procedure was performed for suspected biliary pathology. They defined pancreatitis as any pain and elevated amylase post-ERCP but the clinical significance of the pancreatitis was not clear.82 Endoscopic therapy in this patient population is best left to the expert endoscopists who work closely with the pediatrician in each instance.
ENDOSCOPY IN CHOLECYSTECTOMY COMPLICATIONS The incidence of complications after cholecystectomy has not changed markedly in the era of laparoscopic cholecystectomy, although the number of duct injuries may have increased slightly.83 The management is best approached through the cooperative effort of the operating surgeon, endoscopist, and interventional radiologist. The preferred approach is established and may depend on the local expertise available. Complicated injuries are best palliated as much Figure 64-10. Endoscopic retrograde cholangiopancreatography positioning of the dosimeters and lead shielding for the pregnant patient who requires fluoroscopy.
Fluoroscopy tower
Dosimeters Endoscope
Lead shielding
Radiation source
1193
Section XI. Diseases of the Biliary Tract
as possible locally and referred to a center with specific expertise in complex biliary problems.
BILIARY LEAKS Postoperative biliary leaks are the more common complication of cholecystectomy.83 The typical presentation is abdominal pain and tenderness, with nausea and vomiting less frequent. Mild elevation of AST, ALT, and alkaline phosphatase occur but bilirubin level is normal or minimally elevated.84 The diagnosis can be suggested by demonstrating a fluid collection on transabdominal ultrasound, but this finding is not sensitive and may be non-specific. Radionuclide scanning is the most sensitive for demonstration of leaks.84 ERCP can correctly identify the source of the leak in most instances and offers a modality for therapy.85–89 The method of endoscopic approach has been variable, with the common denominator being obliteration of the transpapillary gradient.87 High levels of success have been reported with ES alone, 7 F90,91 or 10 F stenting alone88,91 or in combination with ES89,91 and nasobiliary stents.89 The largest retrospective study found only a 66% success with sphincterotomy alone.91 None of these options has been approached in a prospective systematic fashion, and in some instances small leaks may resolve spontaneously.88,91 The drawback to stent placement is that a second procedure is required for stent removal. Nasobiliary stenting has the advantage of allowing re-opacification of the biliary tree and then removal of the stent once closure is documented. The major difficulty with nasobiliary stents is accidental dislodgement and this problem has made us reserve this technique for complicated leaks or leaks that have failed conventional management, with evidence for persistent leak despite prior stent or ES. Success is high with endoscopic therapy for simple leaks that are demonstrable at ERCP, especially for cystic duct stump or duct of Luschka leaks – 100% success was reported in four series.85,87,89,91,92 Biliary leaks with large biloma should have the biloma drained by an interventional radiologist to prevent infection within the peritoneal cavity. Complex leaks and leaks associated with biliary strictures present a more complex problem92 and will be addressed below.
DUCT INJURIES One large series92 classifies bile duct injuries as follows. Type A has a bile leak from the cystic duct stump or from a peripheral radicle and management is typically endoscopic. Type B is a leak from a major biliary radicle with or without duct stricture. Type C is a duct stricture with no leak, and type D is complete transection of a duct. It is extremely rare that endoscopy can manage a type D lesion93 and this is left to the surgeon and the interventional radiologist. The management of type B and C lesions is possible endoscopically if the lesion involves the common bile duct and/or common hepatic duct and does not extend into the bifurcation. The endoscopic approach is to place a 10 F stent through the stenosis/leak area at the first ERCP, and 2–3 months later reassess the lesion. If the leak has closed and the stenosis has resolved then no further treatment is needed (Figure 64-11). If the stricture is still evident, then two92 or three 10 F stents94 are placed for 6–12 months, usually with good long-term results.92,94,95 Strictures at the level of the bifurcation have
1194
a high failure rate for endoscopic management92,94,95 and are more amenable to surgical intervention.96,97 One of the rare complications of cholecystectomy is a documented leak by radionuclide scanning and a “normal” ERCP which may represent an injury to the right hepatic duct and, depending on the anatomy of the injury, can have a complex presentation (Figure 64-12). The common situation is a left-sided dominant system and the lateral aspect of the right lobe is supplied by an aberrant duct. If the cystic duct is very close to or actually inserts into the aberrant duct, then the anatomy for the injury to occur is present.98,99 The patient can present with pain and positive HIDDA scan, and the ERCP reveals no leak; however the endoscopist should be alert to this lesion if there is a paucity of ducts in the lateral aspect of the right lobe (Figure 64-12a). The lesion can be demonstrated by transhepatic cholangiography or by injecting a subhepatic drain if it is present. Management is a surgical Roux en Y hepaticojejunostomy.99 A second presentation is right upper quadrant pain, abnormal liver tests with a normal bilirubin, and again the ERCP is “normal,” and a magnetic resonance imaging scan can demon-strate a clipped segment with dilated ducts in the right lobe (Figure 64-12b). The most complex presentation would be represented by Figure 64-12d. The patient presents with pain and abnormal LFTs and ERCP demonstrates a leak (Figure 64-13a) which resolves with stenting. However, pain or fever may occur later and MRCP shows a dilated segment in the lateral aspect of the right lobe (Figure 64-13b).
COMPLICATIONS The complications of ERCP are well known, and the prevalence can depend on the definition of the investigator but the criteria shown in Table 64-1 should be used.56 Some complications are related to the diagnostic portion of the procedure and others only if therapy is performed. Major predictive factors for complications are expertise of the endoscopist, and performance of a precut to gain access to the bile duct. A complication rate of 4.6% was reported in patients undergoing ERCP within 30 days of cholecystectomy for suspected choledocholithiasis, significantly lower than the overall 9.5% rate for sphincterotomy in general.36 The incidence of “nuisance” bleeding, defined as bleeding during or immediately postprocedure, which requires endoscopic therapy has decreased with the introduction of microprocessor-controlled electrosurgical units. Significant bleeding with > 3 g fall in hemoglobin or need for transfusions still occurs and most such cases are delayed bleeds after the patient has been discharged.55 Many of these bleeds resolve with supportive measures or re-endoscopy, and the need for angiography or surgery for control is rare. Pancreatitis remains the commonest complication of sphincterotomy, but is much more common if the indication is for sphincter of Oddi dysfunction as compared to gallstones.36 Elevation of serum amylase and lipase levels is common postprocedure and is not predictive of evolving pancreatitis. The more relevant laboratory data are an elevation of the white blood cell count, and AST, which is the most predictive of prolongation of hospitalization.100 Cholangitis is primarily related to failure to extract stones. One should avoid high-pressure injection of contrast in a duct that has been obstructed as significant bacteremia occurs if biliary pressures exceed 30 cm of H2O.101 If there is a concern about the amount of contrast to be injected, one should aspirate
Chapter 64 NON-SURGICAL MANAGEMENT OF GALLSTONES Figure 64-11. A Endoscopic retrograde cholangiopancreatography demonstrating a large leak from the common bile duct beside a T-tube with stenosis at the level of the take-off of the cystic duct. B 10 French biliary stent placed with the proximal end above the T-tube. C At 3-month follow-up ERCP there is resolution of the narrowing and leak.
B
A
C
1195
Section XI. Diseases of the Biliary Tract
Figure 64-12. Most isolated right hepatic duct (RHD) injuries occur if the cystic duct inserts in close proximity or from the right hepatic duct. A common association with this anomaly is the dominant left hepatic duct system where the first bifurcation of the left system supplies the middle portion of the liver. The three lower figures demonstrate the various presentation of this form of duct injury.
Left hepatic duct Aberrant right duct Common bile duct
A
Leak from RHD– Cystic stump clipped
B
Both ducts clipped
C
No clips Both ducts leaking
D
RHD clippedCystic stump Leaking
Figure 64-13. A Endoscopic retrograde cholangiopancreatography shows a cholangiogram with contrast leaking from the cystic duct to a subhepatic drain. Note the lack of ducts in the lateral aspect of the right lobe. B Magnetic resonance imaging cholangiogram showing a dilated duct system in the same region.
A
B
some of the bile prior to contrast injection. Perforation is the least common complication, but is more common when CDS is the indication for the procedure. Perforations can be secondary to endoscope trauma and are usually large and require immediate surgical closure.102 Instrumentation perforation at the papilla related to the sphincterotomy or guidewire perforations in the bile duct can be managed conservatively initially as most are microperforations, and are treated with nothing by mouth, intravenous fluids, nasogastric suction and broad-spectrum antibiotics.102,103 A surgical consultation should be obtained in all instances, but surgical exploration is only needed for failure to respond to conservative treatment.102,103 The patients should be followed with frequent CT scans to document the course of the disease.104 Indications for surgery include large
1196
intraperitoneal or retroperitoneal fluid collections and deterioration in clinical status.102
REFERENCES 1. Classen M, Demling L. Endoskopische Sphinkterotomie der Papilla vateri und Steinextraktion aus dem Ductus choledochus. Dtsch Med Wochenschr 1974; 99:496–497. 2. Kawai K, Akasaka Y, Murakami K, et al. Endoscopic sphincterotomy of the ampulla of Vater. Gastrointest Endosc 1974; 20:148–151. 3. Gracie WA, Ransohoff DF. The natural history of silent gallstones: the innocent is not a myth. N Engl J Med 1982; 307:798–800.
Chapter 64 NON-SURGICAL MANAGEMENT OF GALLSTONES
4. Bennion LJ, Grundy SM. Risk factors for the development of cholelithiasis in man. N Engl J Med. 5. Adamek HE, Kudis V, Jakobs R, et al. Impact of gallbladder on the outcomes in patients with retained bile duct stones treated with extracorporeal shockwave lithotropsy. Endoscopy 2002; 34:624–627. 6. Lai KH, Lin LF, Lo GH, et al. Does cholecystectomy after endoscopic sphincterotomy prevent the recurrence of biliary complications? Gastrointest Endosc 1999; 49:483–487. 7. Saito M, Tsuyuguchi T, Yamaguchi T, et al. Long-term outcome of endoscopic papillotomy for choledocholithiasis with cholecystolithiasis. Gastrointest Endosc 2000; 51:540–545. 8. Conway JD, Russo MW, Shrestha R. Endoscopic stent insertion into the gallbladder for symptomatic gallbladder disease in patients with end-stage liver disease. Gastrointest Endosc 2005; 61:32–36. 9. Houghton PW, Jenkinson LR, Donaldson LA. Cholecystectomy in the elderly: a prospective study. Br J Surg 1985; 72:220–222. 10. Firilas A, Duke BE, Max MH. Laparoscopic cholecystectomy in the elderly. Surg Endosc 1996; 10:33–35. 11. Spira RM, Nissan A, Zamir O, et al. Percutaneous transhepatic cholecystectomy and delayed laparoscopic cholecystectomy in critically ill patients with acute calculus cholecystitis. Am J Surg 2002; 183:62–66. 12. Pigott JP, Williams GB. Cholecystectomy in the elderly: a prospective study. Am J Surg 1988; 155:408–411. 13. Wong SKH, Yu SCH, Lam YH, Chung SSC. Percutaneous cholecystostomy and endoscopic cholecystolithotripsy in the management of acute cholecystitis. Surg Endosc 1999; 13:48–52. 14. Sosna J, Kruskal JB, Copel I, et al. US-guided percutaneous cholecystostomy: features predicting culture-positive bile and clinical outcome. Radiology 2004; 230:785–791. 15. Boland GW, Lee MJ, Mueller PR, et al. Gallstones in critically ill patients with acute calculous cholecystitis treated by percutaneous cholecystostomy: nonsurgical therapeutic options. Am J Roentgenol 1994; 162:1101–1103. 16. Hatjidakis AA, Karampekios S, Prassopoulos P, et al. Maturation of the tract after percutaneous cholecystostomy with regard to the access route. Cardiovasc Interv Radiol 1998; 21:36–40. 17. Fan ST, Lai ECS, Mok FPT, et al. Early treatment of acute biliary pancreatitis by endoscopic papillotomy. N Engl J Med 1993; 328:228–232. 18. Pezzilli RP, Billi P, Barakat B, et al. Effects of early ductal decompression in human biliary acute pancreatitis. Pancreas 1998; 16:165–168. 19. Neoptolemos JP, Carr-Locke DL, London NJ, et al. Controlled trial of urgent endoscopic retrograde cholangiopancreatography and endoscopic sphincterotomy versus conservative treatment for acute pancreatitis due to gallstones. Lancet 1988; 2:979–983. 20. Wang CH, Mo LR, Lin RC, et al. Rapid diagnosis of choledocholithiasis using biochemical tests in patients undergoing laparoscopic cholecystectomy. HepatoGastroenterology 2001; 48:619–621. 21. Prat F, Meduri B, Ducot B, et al. Prediction of common bile duct stones by noninvasive tests. Ann Surg 1999; 229:362–368. 22. Barr LL, Frame BC, Coulanjon A. Proposed criteria for preoperative endoscopic retrograde cholangiography in candidates for laparoscopic cholecystectomy. Surg Endosc 1999; 13:778–781. 23. Santucci L, Natalini G, Sarpi L, et al. Selective endoscopic retrograde cholangiography and preoperative bile duct stone removal in patients scheduled for laparoscopic cholecystectomy: a prospective study. Am J Gastro 1996; 91:1326–1330. 24. Majeed AW, Ross B, Johnson AG, Reed MW. Common duct diameter as an independent predictor of choledocholithiasis: is it useful? Clin Radiol 1999; 54:170–172.
25. Lichtenbaum RA, McMullen HF, Newman RM. Preoperative abdominal ultrasound may be misleading in the risk stratification for presence of common bile duct abnormalities. Surg Endosc 2000; 14:254–257. 26. Pezzilli R, Billi P, Barakat B, et al. Ultrasonographic evaluation of the common bile duct in biliary acute pancreatitis patients: comparison with endoscopic retrograde cholangiopancreatography. J Ultrasound Med 1999; 18:391–394. 27. Prat F, Edery J, Meduri B, et al. Early EUS of the bile duct before endoscopic sphincterotomy for acute biliary pancreatitis. Gastrointest Endosc 2001; 54:724–729. 28. Liu CL, Lo CM, Chan JK, et al. Detection of choledocholithiasis by EUS in acute pancreatitis: a prospective evaluation in 100 consecutive patients. Gastrointest Endosc 2001; 54:325–330. 29. Prat F, Amouyal G, Amouyal P, et al. Prospective controlled study of endoscopic ultrasonography and endoscopic retrograde cholangiography in patients with suspected common bile duct lithiasis. Lancet 1996; 347:75–79. 30. Amouyal P, Amouyal G, Levy P, et al. Diagnosis of choledocholithiasis by endoscopic ultrasonography. Gastroenterol 1994; 106:1062–1067. 31. Kohut M, Nowak A, Nowakowska-Dulawa E, et al. Endosonography with linear array instead of endoscopic retrograde cholangiography as the diagnostic tool in patients with moderate suspicion of common bile duct stones. World J Gastroenterol 2003; 9:612–614. 32. Kohut M, Nowakowsky-Dulawa E, Marek T, et al. Accuracy of linear endoscopic ultrasonography in the evaluation of patients with suspected common bile duct stones. Endoscopy 2002; 34:299–303. 33. Tseng LJ, Jao YT, Mo LR, Lin RC. Over-the-wire US catheter probe as an adjunct to ERCP in the detection of choledocholithiasis. Gastrointest Endosc 2001; 54:720–723. 34. Catanzaro A, Pfau P, Isenberg GA, et al. Clinical utility of intraductal US for evaluation of choledocholithiasis. Gastrointest Endosc 2003; 57:648–652. 35. Buscarini E, Tansini P, Vallisa D, et al. EUS for suspected choledocholithiasis: do benefits outweigh costs? A prospective, controlled study. Gastrointest Endosc 2003; 54:510–518. 36. Freeman ML, Nelson DB, Sherman S, et al. Complications of endoscopic biliary sphincterotomy. N Engl J Med 1996; 335:909–918. 37. Borgaonkar MR. Passage of a bile duct stone. Gastrointest Endosc 2003; 57:721. 38. Cunningham JT, Cotton PB, Hawes RH, et al. Mechanical lithotripsy for biliary stones: a 5 year university experience. Am J Gastroenterol 2000; 95:2474–2475. 39. Cipolletta L, Costamagna G, Bianco MAA, et al. Endoscopic mechanical lithotripsy of difficult common bile duct stones. Br J Surg 1997; 84:1407–1409. 40. Sorbi D, Van OS E, Aberger FJ, et al. Clinical application of a new disposable lithotripter: a prospective multicenter study. Gastrointest Endosc 1999; 49:210–213. 41. Katsinelos P, Galanis I, Pilpilidis I, et al. The effect of indwelling endoprosthesis on stone size for fragmentation after long-term treatment with biliary stenting for large stones. Surg Endosc 2003; 17:1552–1555. 42. Cotton PB, Forbes A, Leung FW, Dineen L. Endoscopic stenting for long term treatment of large bile duct stones. Gastrointest Endosc 1987; 33:411–412. 43. Bergman JJ, Rauws EA, Tijssen JG, et al. Biliary endoprosthesis in elderly patients with endoscopically irretrievable common bile duct stones: report on 117 patients. Gastrointest Endosc 1995; 42:195–201. 44. Hui CK, Lai KC, Wong WM, et al. Retained common bile duct stones: a comparison between biliary stenting and complete clearance of stone by electrohydraulic lithotripsy. Aliment Pharmacol Ther 2003; 17:289–296.
1197
Section XI. Diseases of the Biliary Tract
45. Lin CK, Lai KH, Chan HH, et al. Endoscopic balloon dilation is a safe method in the management of common bile duct stones. Digest Liver Dis 2004; 36:68–72. 46. Arnold JC, Benz C, Martin WR, et al. Endoscopic papillary balloon dilation vs. sphincterotomy for removal of common bile duct stones: a prospective randomized pilot study. Endoscopy 2001; 33:563–567. 47. Bergman JJ, Rauws EA, Fockens P, et al. Randomised trial of endoscopic balloon dilation versus endoscopic sphincterotomy for removal of bile duct stones. Lancet 1997; 349: 1124–1129. 48. Mathuna PM, White P, Clarke E, et al. Endoscopic balloon sphincteroplasty (papillary dilation) for bile duct stones: efficacy, safety, and follow-up in 100 patients. Gastrointest Endosc 1995; 42:468–474. 49. Tanaka S, Sawayama T, Yoshioka T. Endoscopic papillary balloon dilation and endoscopic sphincterotomy for bile duct stones: long-term outcomes in a prospective randomized controlled trial. Gastrointest Endosc 2004; 59:614–618. 50. DiSario JA, Freeman ML, Bjorkman DJ, et al. Endoscopic balloon dilation compared with sphincterotomy for extraction of bile duct stones. Gastroenterology 2004; 127:1291–1299. 51. Sugiyama M, Atomi Y. Endoscopic papillary balloon dilation causes transient pancreatobiliary and duodenobiliary reflux. Gastrointest Endosc 2004; 60:186–190. 52. Sugiyama M, Atomi Y. Does endoscopic sphincterotomy cause prolonged pancreatobiliary reflux? Am J Gastroenterol 1999; 94:795–798. 53. Isayama H, Komatsu Y, Inoue Y, et al. Preserved function of the Oddi sphincter after endoscopic papillary balloon dilation. Hepato-Gastroenterology 2003; 50:1787–1791. 54. Kawabe T, Komatsu Y, Isayama H, et al. Histological analysis of the papilla after endoscopic papillary balloon dilation. Hepato-Gastroenterology 2003; 50:919–923. 55. Perini RF, Sadurski R, Cotton BP, et al. Post-sphincterotomy bleeding after the introduction of microprocessor-controlled electrosurgery: does the new technology make the difference? Gastrointest Endosc 2005; 61:53–57. 56. Cotton PB, Lehman G, Vennes J, et al. Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 1991; 37:383–393. 57. Park DH, Kim MH, Lee SK, et al. Endoscopic sphincterotomy vs. endoscopic papillary dilation for choledocholithiasis in patients with liver cirrhosis and coagulopathy. Gastrointest Endosc 2004; 60:180–185. 58. Ponchon T, Valette PJ, Bory R, et al. Evaluation of a combined percutaneous-endoscopic procedure for the treatment of choledocholithiasis and benign papillary stenosis. Endoscopy 1987; 19:164–166. 59. Chespak LW, Ring EJ, Shapiro HA, et al. Multidisciplinary approach to complex endoscopic biliary intervention. Radiology 1989; 173:995–997. 60. Calvo MM, Bujanda L, Heras I, et al. The rendezvous technique for the treatment of choledocholithiasis. Gastrointest Endosc 2001; 54:511–513. 61. van der Velden JJ, Berger MY, Bonjer JH, et al. Percutaneous treatment of bile duct stones in patients treated unsuccessfully with endoscopic retrograde procedures. Gastrointest Endosc 2000; 51:418–422. 62. Adamek HE, Maier M, Jakobs R, et al. Management of retained bile duct stones: a prospective open trial comparing extracorporeal and intracorporeal lithotripsy. Gastrointest Endosc 1996; 44:40–47. 63. Weikert U, Muhlen E, Janssen J, et al. The holmium-YAG laser: a suitable instrument for stone fragmentation in choledocholithiasis. The assessment of the results of its use under babyscopic control. Dtsche Med Wochensch 1999; 124:514–518.
1198
64. Yamakawa T. Percutaneous cholangioscopy for the management of retained biliary tract stones and intrahepatic stones. Endoscopy 1989; 21:333–337. 65. Kusano T, Masato F, Isa T, et al. Percutaneous transhepatic cholangioscopic lithotripsy and change of biliary manometry patterns. Hepato-Gastroendterology 1999; 46:2153–2158. 66. Chikamori F, Nishio S, LeMaster JC. Percutaneous papillary balloon dilation as a therapeutic option for cholecystocholedocholithiasis in the era of laparoscopic cholecystectomy. Surg Today 1999; 29:856–861. 67. Maringhini A, Ciambra M, Baccelliere R, et al. Biliary sludge and gallstones in pregnancy: incidence, risk factors, and natural history. Ann Intern Med 1993; 119:116–120. 68. Hiatt JR, Gordon-Hiatt JC, Williams RA, Klein SR. Biliary disease in pregnancy: strategy for surgical management. Am J Surg 1986; 151:263–265. 69. Glasgow RE, Visser BC, Harris HW, et al. Changing management of gallstones disease during pregnancy. Surg Endosc 1998; 12:241–246. 70. Sungler P, Heinerman PM, Steiner H, et al. Laparoscopic cholecystectomy and interventional endoscopy for gallstone complications during pregnancy. Surg Endoscopy 2000; 14:267–271. 71. Medical radiation exposure of pregnant and potentially pregnant women. NCRP report no. 54. Washington, DC: National Council on Radiation Protecton and Measurements; 1977. 72. Jamidar PA, Beck GJ, Hoffman BJ, et al. Endoscopic retrograde cholangiopancreatography in pregnancy. Am J Gastroenterol 1995; 90:1263–1267. 73. Tham TCK, Vandervoort J, Wong RCK, et al. Safety of ERCP during pregnancy. Am J Gastroenterol 2003; 98:308–311. 74. Hoffman BJ, Cunningham JT. Radiation exposure to the pregnant patient during ERCP. Gastrointest Endosc 1992; 38:253. 75. Fox VL, Werlin SL, Heyman MB. Subcommittee on endoscopy procedures of the patient care committee of the North American Society for Pediatric Gastroenterology Nutrition. Endoscopic retrograde cholangiopancreatography in children. J Pediatr Gastroenterol Nutr 2000; 30:335–342. 76. Kato S, Kamagata S, Asakura T, et al. A newly developed smallcaliber videoduodenoscope for endoscopic retrograde cholangiopancreatography in children. J Clin Gastroenterol 2003; 37:102–104. 77. Wengrower D, Gozal D, Gozal Y, et al. Complicated endoscopic pediatric procedures using deep sedation and general anesthesia are safe in the endoscopy suite. Scand J Gastroenterol 2004; 39:283–286. 78. Guelrud M, Mendoza S, Jaen D, et al. ERCP and endoscopic sphincterotomy in infants and children with jaundice due to common bile duct stones. Gastrointest Endosc 1992; 38:450–453. 79. Newman KD, Powell DM, Holcomb III GW. The management of choledocholithiasis in children in the era of laparoscopic cholecystectomy. J Pediatr Surg 1997; 32:1116–1119. 80. Tarnasky PR, Tagge EP, Hebra A, et al. Minimally invasive therapy for choledocholithaisis in children. Gastrointest Endosc 1998; 47:189–192. 81. Pfau PR, Chelimsky GG, Kinnard MF, et al. Endoscopic retrograde cholangiopancreatography in children and adolescents. J Pediatr Gastroenterol Nutr 2002; 35:619–623. 82. Prasil P, Laberge JM, Barkun A, Flageole H. Endoscopic retrograde cholangiopancreatography in children: a surgeon’s perspective. J Pediatr Surg 2001; 36:733–735. 83. The Southern Surgeons Club. A prospective analysis of 1518 laparoscopic cholecystectomies. N Engl J Med 1991; 324:1075–1078. 84. Brugge WR, Rosenberg DJ, Alavi A. Diagnosis of postoperative bile leaks. Am J Gastroenterol 1994; 89:2178–2183.
Chapter 64 NON-SURGICAL MANAGEMENT OF GALLSTONES
85. Foutch PG, Harlan JR, Hoefer M. Endoscopic therapy for patients with a post-operative biliary leak. Gastrointest Endosc 1993; 39:416–421. 86. Kozarek RA, Ball TJ, Patterson DJ, et al. Endoscopic treatment of biliary injury in the era of laparoscopic cholecystectomy. Gastrointest Endosc 1994; 40:10–16. 87. Bjorkman DJ, Carr-Locke DL, Lichtenstein DR, et al. Postsurgical bile leaks: endoscopic obliteration of the transpapillary pressure gradient is enough. Am J Gastroenterol 1995; 90:2128–2133. 88. Wootton FT, Hoffman BJ, Marsh WH, Cunningham JT. Biliary complications following laparoscopic cholecystectomy. Gastrointest Endosc 1992; 38:183–1185. 89. Christoforidis E, Goulimaris I, Tsalis K, et al. The endoscopic management of persistent bile leakage after laparoscopic cholecystectomy. Surg Endosc 2002; 16:843–846. 90. Hoffman BJ, Cunningham JT, Marsh WH. Ensdoscopic management of biliary fistulas with small caliber stents. Am J Gastroenterol 1990; 85:705–707. 91. Kaffes AJ, Hourigan L, De Luca N, et al. Impact of endoscopic intervention in 100 patients with suspected postcholecystectomy bile leaks. Gastrointest Endosc 2005; 61:269–275. 92. Bergman JJHG, van de Brink GR, Rauws EAJ, et al. Treatment of bile duct lesions after laparoscopic cholecystectomy. Gut 1996; 38:141–147. 93. Baron TH, Feitoza AB, Nagorney DM. Successful endoscopic treatment of a complete transaction of the bile duct complicating laparoscopic cholecystectomy. Gastrointest Endosc 2003; 57:765–769. 94. Borowicz MR, Adams DB, Simpson JP, Cunningham JT. Management of biliary strictures due to laparoscopic cholecystectomy. J Surg Res 1995; 58:86–89.
95. Draganov P, Hoffman BJ, Marsh WH, et al. Long-term outcome in patients with benign biliary strictures treated endoscopically with multiple stents. Gastrointest Endosc 2002; 55:680–685. 96. De Wit LH, Rauws EAJ, Gouma DJ. Surgical management of iatrogenic bile duct injury. Scand J Gastroenterol 1999; 34:89–94. 97. Adams DB, Borowicz MR, Wootton III FT, Cunningham JT. Bile duct complications after laparoscopic cholecystectomy. Surg Endoscopy 1993; 7:79–83. 98. Kalayci C, Aisen A, Canal D, et al. Magnetic resonance cholangiopancreatography documents bile leak site after cholecystectomy in patients with aberrant right hepatic duct where ERCP fails. Gastrointest Endosc 2000; 52:277–281. 99. Lillemoe KD, Petrofski JA, Choti MA, et al. Isolated right segmental hepatic duct injury: a diagnostic and therapeutic challenge. J Gastrointest Surg 2000; 4:168–177. 100. Wojtun S, Gil M, Gil J. Recognition of ERC-induced pancreatitis in patients with choledocholithiasis by an analysis of laboratory findings. Hepato-Gastroenterol 2000; 47:550–553. 101. Yoshimoto H, Ikeda S, Tanaka M, Matusmoto S. Relationship of biliary pressure to cholangiovenous reflux during endoscopic retrograde balloon catheter cholangiography. Digest Dis Sci 1989; 34:16–20. 102. Stapfer M, Selby RR, Stain SC, et al. Management of duodenal perforations after endoscopic retrograde cholangiopancreatography and sphincterotomy. Ann Surg 2000; 232:191–198. 103. Enns R, Eloubeidi MA, Mergener K, et al. ERCP-related perforations: risk factors and management. Endoscopy 2002; 34:293–298. 104. Zissen R, Shapiro-Feinberg M, Oscadchy A, et al. Retroperitoneal perforation during endoscopic sphincterotomy: imaging findings. Abdom Imaging 2000; 25:279–282.
1199
Section XI: Disease of the Biliary Tract
65
SURGERY OF THE BILIARY TRACT Daniel Tseng and John Hunter Abbreviations ERCP endoscopic retrograde cholangiopancreatography LC laparoscopic cholecystectomy
MRCP
magnetic resonance cholangiopancreatography
INTRODUCTION The extrahepatic biliary tract extends from the right and left hepatic ducts exiting the liver to the ampulla of Vater. It has always been an area troublesome to surgeons because of its tenuous blood supply, difficult anatomy, and unforgiving complications when disturbed. As surgeons enter a new age of minimally invasive surgery and endoscopic techniques, the limits are constantly tested and the old ways are redefined. Fortunately, patients have benefited from technologic advances with decreased hospital stays, diminished pain, small incisions, shorter recovery period, and a lower morbidity and mortality compared with the past. The purpose of this chapter is to provide an update on the management of problems of the biliary tract, specifically addressing the surgical options.
GALLBLADDER Cholecystectomy accounted for 442 000 of the nearly 40 million surgical procedures performed in the USA in 2001. It has surpassed appendectomy as the most frequently performed operation by the general surgeon.1 Since the first laparoscopic cholecystectomy (LC) was performed in 1985, the use of minimally invasive techniques in the treatment of gallbladder problems has expanded dramatically. The surgeon is now able not only to remove the gallbladder laparoscopically, but also to perform many other interventions that in the past required open intervention. While LC has dramatically improved the general surgeon’s ability to care for patients, it has also brought new dilemmas to surgery. Despite innovative surgical techniques and equipment, major complications still occur and may do so in a different manner than occurred previously. Therefore surgeons are required to manage challenges that did not exist prior to the development of minimally invasive surgery.
CHOLELITHIASIS In 2001, the Centers for Disease Control reported in their National Hospital Discharge Survey that cholelithiasis accounted for 367 000 of the 3.2 million hospital admissions, with an average length of stay for these patients of nearly 4 days.1 With approximately 10–15% or 20 million of adults in the USA carrying gallstones, this disease costs individuals and society a great deal.
PTD
percutaneous transhepatic biliary drainage
Gallstone composition consists primarily of cholesterol and bile pigments. An imbalance of the natural production of bile acids, phospholipids, and cholesterol predisposes individuals to the formation of lithogenic bile anywhere in the biliary system. The mnemonic of the four Fs (female, fat, forty, and fertile) characterizes patients who are classically found to have cholelithiasis associated with cholesterol stones. Other factors attributable to the formation of cholelithiasis include gallbladder dysmotility, nutrition, prior bowel surgery, hormonal changes, blood dyscrasias, medications, and infection. Genetic factors may play a role in the unusually high 80% incidence of gallstones in women of the Pima Indians in the USA, as well as the increased risk of gallstones in first-degree relatives.2,3 Fortunately, the majority of patients with cholelithiasis never develop symptoms. A review of the major published literature on the natural history of gallstones reveals that symptoms occur at approximately 1–2% per year, with complications developing in only 2–6% (Table 65-1). Most biliary symptoms result from pressure within the biliary system or result from inflammatory processes in or around the biliary system such as cholecystitis or pancreatitis (Table 65-2). Visceral sensation from the gallbladder can mimic pathologic states such as gastroesophageal reflux disease, angina, and appendicitis. Most commonly the visceral pain is referred to the right shoulder. Parietal irritation from either an acute or chronic condition results from overdistension of the gallbladder and irritation of the peritoneum causing pain located in the right upper quadrant.
INDICATIONS FOR CHOLECYSTECTOMY Once symptoms from cholidocholithiasis develop, recurrence can be expected in approximately 50% of patients over the course of 1 year.4 More recent data demonstrate a significant improvement in quality of life after cholecystectomy for biliary colic in the majority of patients and can be an indication for cholecystectomy in goodrisk patients.5 Although symptoms can be controlled medically, complications can be life-threatening. Complications occur in most series at less than 1% per year and represent an indication for surgery (Table 65-3).6,7 The diminished pain, smaller incisions, shorter hospitalizations, and faster recovery associated with LC have prompted surgeons to advise patients to have their gallbladder removed before complications occur.
1201
Section XI. Diseases of the Biliary Tract
Table 65-1. Natural History of Gallstones Author (year)
Gracie (1982)8 McSherry (1985)9 Friedman (1989)10 Cucchiaro (1990)11 Wada (1993)12 Juhasz (1994)13 Attili (1995)14 Angelico (1997)15 Halldestam (2004)16
Table 65-3. Complications of Cholelithiasis
No. of patients
Length of disease (years)
Biliary pain (%)
Biliary complications (%)
123 135 123 125 680 110 118 426 120
15 4 5 5 13 6 10 10 7
18 17
2 0 6 2 3 5 3
11 20 9 26 38 7
5
Table 65-2. Symptoms of Cholelithiasis Epigastric or right upper quadrant abdominal pain Fatty food intolerance Heartburn Bloating Belching Jaundice Flatulence
Acute Cholecystitis Simply defined, acute cholecystitis is inflammation of the gallbladder as a result of obstruction of the outflow of bile from the cystic duct. Causes for cystic duct obstruction include gallstones, biliary sludge, or neoplasms that result in overgrowth of native bacteria present within the gastrointestinal tract. The most common Gram-negative organisms identified are Escherichia coli and Klebsiella spp., and the anaerobes Clostridium and Bacteroides are second most common; and Gram-positive organisms are rare. While infection of the bile is not uncommon, most acute cholecystitis occurs with sterile bile. Signs and symptoms of acute cholecystitis include abdominal pain and tenderness, fever, nausea, vomiting, positive Murphy’s sign, and a palpable fullness in the right upper quadrant. Unlike biliary colic, these symptoms can be somewhat insidious and are constant in nature, worsening over time. Whereas colic subsides over time, true acute cholecystitis does not, and therefore often requires admission to hospital for treatment. Laboratory examination often reveals an elevated white blood cell count with or without a significant shift in the differential towards polymorphonuclear cells. Ultrasonographic description of acute cholecystitis includes a thickened edematous gallbladder wall, pericholecystic fluid, and sonographic Murphy’s sign. Once the diagnosis of acute cholecystitis has been made, initial therapy is resuscitative in nature. Initiation of intravenous fluid resuscitation and broad-spectrum antibiotics for Gram-negative and anaerobic bacteria is paramount in preparation for surgery. Appropriate response to therapy allows the patient time to recover from the initial insult prior to proceeding to the operating room. Although cholecystitis is not an emergent operation in most instances, urgent removal of the gallbladder should be performed during the same hospitalization. The earlier and more edematous phases of acute cholecystitis facilitate LC.
1202
Cholecystitis Cholangitis Gallstone pancreatitis Gallstone ileus Pyoderma gangrenosum Mirizzi’s syndrome
In the past, it was suggested that those patients who develop complications such as acute cholecystitis have an increased morbidity with early surgical intervention, which prompted many to propose that those with cholecystitis wait for up to 6 weeks before undergoing cholecystectomy. That information was mainly based on observational studies and applied more generally to open procedures. Recent data, however, have demonstrated that outcomes are generally equivalent and may be slightly better if surgery is performed during the same hospitalization rather than waiting 4–6 weeks.17–19 Recently, a meta-analysis of three randomized prospective clinical trials revealed that patients who had undergone early LC had similar rates of complications, a trend toward fewer conversions to open procedures, and a trend toward decreased hospital stay (P < 0.06) compared with delayed LC.17 Of significance, a failure rate requiring urgent operation during the initial hospitalization in 9.3% of those who were randomized to delayed surgery was observed. Of those who were able to leave the hospital, almost 15% suffered a recurrence of their symptoms requiring readmission, and 26% of those required urgent operation. Overall, 23% of patients failed nonoperative therapy during the study period, and over half of these patients required urgent surgery. The current available literature clearly supports early surgery in the setting of acute uncomplicated cholecystitis.18,19
Cholangitis Normally bile flowing from the liver is “sterile.” Bacterial contamination of bile is normally prevented by a number of factors. The sphincter of Oddi acts as a barrier to stop the reflux of duodenal bacteria into the biliary system. Hepatic sinusoids contain Kupffer cells which remove organisms and secrete immunoglobulin A and bile salts that help sterilize the bile. The flow of bile naturally removes debris and bacterial contamination helping to create the sterile environment. Once obstruction occurs, those natural protective mechanisms are lost, and this results in backflow of organisms into the biliary tree with subsequent development of cholangitis.20 In the USA, gallstones are the major etiologic cause for the development of cholangitis. Approximately 10% of those with gallstones may go on to develop cholangitis if left untreated.21 However, when evaluating someone with cholangitis, other causes must be considered (Table 65-4). Typical organisms cultured in the blood and bile are the usual bacteria found in the gastrointestinal tract, including E. coli, Enterobacter, Enterococcus, and Klebsiella. However, instrumentation may allow Pseudomonas, skin, and oral flora to be introduced into the biliary system as well.22 Cholangitis is characterized by Charcot’s triad, which consists of fever (90%), right upper quadrant pain (70%), and jaundice (60%).23
Chapter 65 SURGERY OF THE BILIARY TRACT
Table 65-4. Causes of Cholangitis Cholelithiasis Primary biliary stones Malignant neoplasms Strictures Instrumentation Infection Primary sclerosis cholangitis
Sicker patients have associated mental status changes (10–20%), and hypotension (30%), resulting in Reynolds’ pentad.24 Laboratory abnormalities include leukocytosis, hyperbilirubinemia, and mild elevations of transaminases and alkaline phosphatase.25 Severe liver dysfunction with coagulation abnormalities can result from prolonged untreated disease, with a mortality rate of nearly 100%. Initial resuscitation with intravenous fluids, broad-spectrum antibiotics, and correction of any underlying coagulopathy is critical in the initial management of this disease. After stabilization of the patient, biliary decompression must be performed in order to allow the infection to resolve. Non-operative drainage procedures are offered first, and these include endoscopic retrograde cholangiopancreatography (ERCP) or percutaneous transhepatic biliary drainage (PTD). ERCP can be successfully performed in over 90% of patients with cholangitis and is the treatment of choice for decompression of the biliary system. In one published series of 898 patients with cholangitis treated with ERCP, the mortality rate was only 0.42% and the complication rate was 6%.26 Inability to perform endoscopic decompression usually results from difficulty in identifying and cannulating the ampulla of Vater. In patients with prior upper gastrointestinal surgery, it may prove too difficult or impossible to cannulate the ampulla. Even if cannulation is completed, debris or stones within the common bile duct may not be cleared completely. Also, individuals may have intrahepatic stones that are not amenable to removal. In these situations, PTD may be a better option. However, PTD is associated with a higher complication rate than ERCP. A number of studies have been completed demonstrating lower morbidity and mortality rates with ERCP compared to surgery in the setting of acute severe cholangitis. Lai et al.27 performed a prospective randomized trial directly comparing the use of ERCP versus surgical decompression in the setting of acute cholangitis. Eighty-two patients were randomized over a 43-month period. The results favor endoscopic drainage over surgical drainage, demonstrating a diminished need for ventilatory support (30% versus 63%), fewer complications (34% versus 66%), and fewer retained stones (7% versus 29%). Mortality was also significantly decreased (10% versus 32%) in the endoscopic drainage versus surgery groups, respectively. This study and others have demonstrated the superiority of early endoscopic decompression, reserving surgery for those who have failed the endoscopic approach.28
PERCUTANEOUS TRANSHEPATIC BILIARY DECOMPRESSION An alternative therapy for the initial treatment of obstructive cholangitis that was frequently utilized in the 1980s was PTD. This method allowed decompression of the biliary system without the
Table 65-5. Complications from Percutaneous Transhepatic Decompression Sepsis Hemobilia Intra-abdominal bleeding Cholangitis Bile leak Catheter malfunction/misplacement Hypotension Pancreatitis Pneumothorax/hemothorax
need for heavy sedation or general anesthesia. It was found to be especially helpful in the elderly and frail with multiple comorbidities. Besides being less invasive, PTD allowed access to the intrahepatic biliary system after decompression, for removal of stones through the percutaneous tract, and treatment of intrahepatic strictures via balloon dilation. Currently PTD is used most frequently when ERCP is unsuccessful. PTD-related complication rates vary between 6 and 30% (Table 65-5).22,25,29,30 However, retrospective studies have shown that PTD can be performed with much lower morbidity and mortality than early surgery for cholangitis. Chen et al.22 reported a decline in mortality of 13% with early surgery to 2% in those who had a good response to PTD followed by surgery. In this series there was no mortality related to the PTD alone. Rapid improvement in both clinical and laboratory abnormalities is expected, certainly within 72 hours if successful drainage is achieved.31
SURGICAL DECOMPRESSION IN CHOLANGITIS DUE TO CHOLELITHIASIS Open surgical drainage procedures have been associated with a high morbidity and mortality rate in the setting of acute cholangitis. The morbidity and mortality rate in the 1970s and 1980s ranged from 20 to 60% when surgical treatment was used as the first line of treatment.32 However, results are dismal if no therapeutic intervention is performed, with mortality close to 100%. Welch et al.33 in 1976 reviewed the charts of 20 patients admitted for cholangitis. Sixteen had undergone surgical exploration while the other 4 were managed non-operatively: the latter all subsequently died. Those who were explored within 24 hours had a mortality of 17%, whereas those who were explored within 24–72 hours had a mortality of 50%. Other risk factors that have been identified on univariate analysis that increase mortality following surgery include low pH, thrombocytopenia, hypoalbuminemia, multiple medical problems, and high bilirubin.34 There are several surgical options and the choice of operative approach is dependent on the location of the gallstones. The simplest approach is open common bile duct exploration with placement of a T-tube for drainage. This approach is the fastest and least morbid operation available. It allows complete external drainage of the infected bile and should be the surgical treatment of choice, especially in those who are unstable.35,36 Other additional procedures to be considered at the time of the operation that may be helpful are cholecystectomy, cholangiography, and choledochoscopy. However these procedures may add precious time to the surgery and are not necessary in the unstable patient.
1203
Section XI. Diseases of the Biliary Tract
If the patient is stable, clearance of the biliary system by choledochoscopy or cholangiography is helpful in identifying retained common bile duct or intrahepatic stones. Choledochoscopy is superior to conventional intraoperative cholangiography in identification of retained common bile duct stones, as demonstrated by Lau et al. in 1991.37 Their group performed a large prospective randomized trial comparing choledoscopy to routine cholangiography for the demonstration of common bile duct stones. Retained common bile duct stones at 2 weeks postoperatively were more common in the group that received cholangiogram alone (17%) versus those undergoing choledochoscopy (1%). Choledochoscopy was performed safely in the setting of the hemodynamically stable patient with acute cholangitis but was time-consuming and not recommended for unstable septic patients. Alternatively, internal drainage via choledochoenteric bypass can be considered in those who have an obstructed ampulla or very large stones that would not otherwise pass on their own. This is rather time-consuming and should only be performed in the stable patient. Options include choledochoduodenostomy or Roux en Y choledochojejunostomy. Side-to-side choledochoduodenostomy is faster and allows for a single anastomosis. Extensive kocherization of the duodenum allows tension to be removed from the anastomosis, but this may not be possible in those with severe inflammation. Roux en Y choledochojejunostomy is preferred for those who have a rather fixed and immobile duodenum.
are multiple treatment options with no clear consensus as to the best approach for the management of this problem. Currently pancreatitis is generally classified as mild, moderate, or severe forms. There are multiple scoring systems that have been developed for grading pancreatitis. Among these scoring systems are Ranson’s criteria,31 APACHE II,45 Glasgow,34 and Atlanta criteria.35 These scoring systems offer prognostic criteria of anticipated morbidity and mortality. Despite the ability to classify the degree of pancreatitis, initial therapy for pancreatitis is based on severity at presentation. All patients share a common presentation of abdominal pain, with different degrees of organ compromise detected by physical and laboratory examination. Patients admitted to the hospital are placed on strict bowel rest and require fluid and electrolyte replacement. Depending on the degree of pancreatitis, intensive care unit monitoring, antibiotics, and parenteral nutrition may be initiated. Identifying the cause of pancreatitis is paramount in preventing its recurrence. Ultrasound and computed tomography are used to identify gallstones and biliary sludge, determine biliary ductal dilation, assess the degree of pancreatitis, and detect any complications from the disease. If gallstones or biliary sludge are identified as the source of pancreatitis, multiple options currently exist as to the further treatment and are tailored to the severity of the pancreatitis.
Mild/Moderate Gallstone Pancreatitis CHOLECYSTECTOMY AFTER DRAINAGE In the past, after endoscopic drainage was completed, not all patients underwent subsequent cholecystectomy. The high-risk patients, the elderly, those with multiple comorbid conditions, or those who refused cholecystectomy were managed expectantly. Unfortunately a recurrence rate for cholangitis as high as 19% occurred if the gallbladder was left in situ.38 Multiple prospective randomized trials have demonstrated that the recurrence rate of complicated gallstone disease requiring cholecystectomy is between 14 and 40%, even after endoscopic sphincterotomy (Table 65-6). Therefore it is our policy routinely to remove the gallbladder in patients who have been cleared of common duct gallstones endoscopically once patients have been stabilized.
GALLSTONE PANCREATITIS Acute pancreatitis may be a severe disease process with a mortality rate between 5 and 15%.39,40 The presence of gallstones accounts for approximately 30–50% of patients who develop pancreatitis and remains the most common cause for acute pancreatitis. The severity of resulting gallstone pancreatitis is extremely variable, with some patients requiring prolonged intensive care unit stays. There Table 65-6. Recurrence of Gallstone Complications Requiring Cholecystectomy after Endoscopic Sphincterotomy Author (year) Tanaka (1987)41 Hammarstrom (1995)42 Targarona (1996)43 Suc (1998)44
1204
Patients
Follow-up
103 39 50 97
3 years 5 years 17 months 5 years
Later cholecystectomy (%) 33% 14% 37%
For those who do not have severe pancreatitis based upon available criteria, multiple therapeutic alternatives are available that depend on the local expertise (Table 65-7). Management strategy is dependent on the available resources and skills of the endoscopist, radiologist, and surgeon. Laparoscopic cholangiography is now a routine procedure; however, laparoscopic common bile duct exploration remains technically challenging, timeconsuming, and vexing for surgeons unfamiliar with the technique. Because of this, UK consensus guidelines for the treatment of biliary pancreatitis advocate the use of selective preoperative ERCP followed by interval LC (two-stage approach).46 Alternatively, for those individuals of adequate training and skill, a more technically demanding one-stage approach with LC and common bile duct exploration results in excellent outcomes.47,48 Optimal timing for LC in those with mild or moderate form of pancreatitis is still debated. Some authors advocate that cholecystectomy after ERCP and stone clearance is unnecessary because the incidence of recurrent gallstone pancreatitis is low and morbidity
Table 65-7. Treatment Alternatives for Gallstone Pancreatitis in Mild or Moderate Form ERCP with sphincterotomy alone ± interval cholecystectomy Preoperative ERCP with sphincterotomy followed by laparoscopic cholecystectomy Laparoscopic cholecystectomy alone Laparoscopic cholecystectomy with intraoperative cholangiogram and possible common bile duct exploration Laparoscopic cholecystectomy with intraoperative cholangiogram and postoperative ERCP with sphincterotomy ERCP, endoscopic retrograde cholangiopancreatography.
Chapter 65 SURGERY OF THE BILIARY TRACT
and mortality are not different.49 Still, there are significant data showing that recurrence rates of gallstone pancreatitis without cholecystectomy range between 30 and 61%.50,51 Moreover, if pancreatitis does recur, it has a much higher morbidity and mortality.52 There is a decrease in recurrent episodes of pancreatitis if the gallbladder is removed early.53 Unfortunately, the risks of complications during LC are higher in the setting of gallstone pancreatitis compared with uncomplicated gallstone disease, but studies have shown that it can still be performed safely.54 The ability to perform the operation laparoscopically successfully ranges between 90 and 100% in most series.55–63 As with cholecystitis, early surgery (<7 days) is as safe as if not safer than delayed surgery.64 Therefore, ideally it is best to perform cholecystectomy during the same hospitalization once laboratory parameters have normalized.
Severe Gallstone Pancreatitis There is no one measure that is able to predict the natural course of acute pancreatitis. Subsequently, we are left with a host of measurable laboratory anomalies that may help separate patients into mild/moderate or severe pancreatitis. Whichever classification scheme is used to identify those with severe pancreatitis, the general treatment regimen has been unchanged for the past two decades, as described previously. The most significant advancement in care of these patients is the advent of ERCP. A meta-analysis of the four randomized controlled trials of ERCP with sphincterotomy versus conservative management in gallstone pancreatitis has recently been performed.52 Two of these trials specifically addressed mild and severe pancreatitis separately.55,65 Results from the meta-analysis revealed less morbidity (38.2% versus 25%) and mortality (9.1% versus 5.2%) in those who were treated conservatively compared to those who underwent endoscopic therapy. However, others have provided conflicting data that early ERCP and papillotomy may not be beneficial in those patients with acute gallstone pancreatitis without obstructive jaundice.66 Despite this divergence in study outcomes, a 2002 National Institutes of Health consensus statement stated that, “In patients with severe biliary pancreatitis, early intervention with ERCP reduces morbidity and mortality compared with delayed ERCP.”67 Cholecystectomy in the presence of severe pancreatitis is fraught with hazard. The intense inflammatory reaction with increasingly severe pancreatitis makes the dissection dangerous. Most authors uniformly agree that cholecystectomy is safer a few weeks after the resolution of an episode of severe pancreatitis.65,68–71 Therefore it is our policy to allow the inflammation to subside for a few weeks and perform an elective cholecystectomy on an outpatient basis.
Mirizzi Syndrome The functional hepatic syndrome, or Mirizzi syndrome, was first described in 1948 as an obstruction of the common hepatic duct secondary to an impacted gallstone in the cystic duct.72 The definition has expanded to include not only gallstones, but also gallbladder malignancy and chronic inflammation as a source for obstruction. This is a rather rare complication of gallstone disease that occurs with an incidence ranging from 0.05 to 1%.73 Classification schemes have been provided by both McSherry et al.68 and Csendes et al.74 (Tables 65-8 and 65-9).
Table 65-8. McSherry Classification Type I Type II
External compression of common hepatic duct External compression of common hepatic duct with cholecystocholedochal fistula
Table 65-9. Csendes Classification Type I Type II Type III Type IV
Simple pressure on common hepatic duct More severe disease with cholecystobiliary fistula, less than one-third the circumference of the duct wall More severe disease with cholecystobiliary fistula, two-thirds of duct wall involved More severe disease with cholecystobiliary fistula, entire circumference of wall involved
Patients with Mirizzi’s syndrome are generally in their 60s and 70s, and present with classical findings of cholecystitis, including fever, right upper quadrant pain, nausea, and vomiting. Jaundice associated with hyperbilirubinemia, elevated transaminases, and leukocytosis is also rather typical in these patients, but certainly may be absent.75 Diagnosis of Mirizzi syndrome can be made by many methods. Most often these patients undergo ERCP, which establishes the diagnosis of suspected Mirizzi syndrome with a sensitivity close to 100%.76 It is sometimes difficult to discriminate between benign and malignant causes of Mirizzi’s without surgical exploration. Other imaging modalities that improve diagnostic accuracy include ultrasound, computed tomography, and magnetic resonance cholangiopancreatography (MRCP). There is evidence to suggest that the Mirizzi syndrome may be associated with a slightly higher incidence of biliary tract cancer. Proposed theories revolve around the notion that chronic inflammation as a result of chronic cholelithiasis may result in a higher propensity for cancer formation in the surrounding structures. Incidence of associated gallbladder cancer ranges between 10 and 28% in published reports.69,76–78 Some authors have suggested that frozensection biopsy should be performed in all individuals when evidence of Mirizzi syndrome is discovered at the time of surgery due to the possible presence of cancer so that liver resection can be performed if necessary. Optimal treatment of Mirizzi syndrome is highly dependent on the degree of inflammation present within the porta hepatis. Success rates of LC for Mirizzi syndrome vary widely throughout the literature. Although the procedure can be accomplished laparoscopically, the operation can be treacherous, especially in the exposure of Calot’s triangle. Sixteen studies reporting LC for Mirizzi’s syndrome in 82 patients required conversion from laparoscopic to open cholecystectomy in 32%.76 Successful cholecystectomy is more likely in the type I situation; types II–IV often require more complex procedures. Management of Mirizzi type I can be performed by simple cholecystectomy with closure of the enlarged cystic duct. Proper closure of a large cystic duct cannot always be accomplished with traditional clip placement; therefore the surgeon must be prepared to place sutures or staples to close the duct. Cholangiography is helpful to identify individuals who may develop common hepatic duct stenosis as a result of the severe inflammation. T-tube decompression is
1205
Section XI. Diseases of the Biliary Tract
appropriate for a few weeks while the inflammation resolves. Treatment of Mirizzi types II and III are individualized to the location and size of the biliary fistula. Partial cholecystectomy with closure of the remnant of gallbladder or cystic duct is initially performed. Placement of a T-tube through the fistula tract can lead to prolonged bile leak, so some advocate flap closure of the fistulous opening using gallbladder wall or cystic duct. Choledochotomy and T-tube placement are performed remote to the biliary fistula.79 Another alternative is to perform a choledochoduodenostomy to repair the fistula.80 Mirizzi type IV cases are best repaired with Roux en Y hepaticojejunostomy.
Gallstone Ileus Cholecystenteric fistulas are a rare complication of gallstone disease, with an incidence of only 0.3–0.5%.81–83 Fistulization results from the severe inflammatory process that occurs with chronic cholecystitis and is similar to Mirizzi syndrome in this respect. The large gallstone then passes through the fistula tract into the gastrointestinal tract at various locations. The fistula may join the gallbladder to the duodenum, stomach, or colon.84,85 The primary morbidity resulting from cholecystenteric fistula is bowel obstruction. Gallstones account for 1–4% of all bowel obstructions and up to 25% in those over the age of 65 who have no evidence of strangulation.76,84,86–88 The most common site of obstruction by gallstones in the gastrointestinal tract is the ileocecal valve (60%), followed by jejunum
(16%), stomach (14%), colon (4%), and duodenum (Bouveret’s syndrome: 3.5%).89 Iatrogenic causes from the passage of gallstones from sphincterotomy and biliary-enteric bypasses have also been reported.86,90–92 Typically patients are female, elderly (mean age in the 70s), and have multiple comorbidities and long-standing gallbladder disease.84,89,94 Early diagnosis remains a challenge, and the findings of Rigler’s triad94 of pneumobilia, ectopic gallstones, and bowel dilation is found in only 10% of those later diagnosed with gallstone ileus84 (Figure 65-1). In fact, the majority of patients with gallstone ileus are not identified preoperatively.86,95 Because of the difficulty in diagnosis, care is delayed and patients are treated later than would otherwise be expected, resulting in a mortality between 10 and 20%.84,89,96 Initial management is directed toward aggressive resuscitation of the patient. Often these patients have electrolyte abnormalities and fluid imbalances that need to be corrected prior to any definitive treatment. Nasogastric tube placement is advised for those with significant ongoing emesis for decompression of the gastrointestinal tract. Antibiotics may be administered for those who manifest signs of perforation as a result of their obstruction. Definitive treatment alternatives have included surgical and nonsurgical approaches, all with variable success. Non-surgical treatments have attempted to utilize shock-wave lithotripsy,89,97–100 laser lithotripsy,101 endoscopy,102–104 or a combination.105 However, these
Figure 65-1. Gallstone ileus. Thin arrow, opacified gallstone; thick arrows, pneumobilia. (Reproduced from Vaidya JS, Lalude O, Grant D, Mukhtar H. Gallstone ileus. Lancet 2003; 362:1105, with permission.)
1206
Chapter 65 SURGERY OF THE BILIARY TRACT
approaches remain somewhat anecdotal and are only used on a very selective basis in low-risk patients, whereas surgery remains the mainstay of therapy. A raging debate has occurred in the literature regarding the optimal surgical approach in this disease. A two-stage operation has been taught as the “gold standard” and, at the first operation, the gallstone is removed via an enterotomy and the gallbladder and fistula are left alone. Numerous case reports have been published regarding the benefits of laparoscopic enterolithotomy, including a much simpler re-operation with fewer adhesions and avoidance of a laparotomy with its associated risks.106–114 The second operation is performed weeks later, once the patient has recovered from the bowel obstruction and enterotomy, to remove the gallbladder and close the fistula opening. Alternatively, a single-stage operation with enterolithotomy, cholecystectomy, and fistula closure has been proposed in those who are good-risk patients. Critics of the two-stage approach cite the potential for recurrence of gallbladder symptoms, cholecystitis, cholangitis, and recurrent gallstone ileus as some reasons why a two-stage approach is less advantageous.84,115,116 Moreover, recent case series have shown that mortality is equivalent between the one- and two-stage approaches, with mortality ranging between 0 and 33%.108,117,118 The overwhelming body of literature and surgical community still favor a two-stage approach.89,97,119,120 The goal for the first operation is relief of the obstruction and whether performed via the endoscope, laparoscope, or open laparotomy is simply a matter of preference. There are no data available to support one method over the other. For those patients who are at an extreme operative risk, endoscopy offers an alternative that saves the risk of anesthesia and surgery. However, the durability and success of this procedure remain unproven at this time and it has been used in only a small number of cases.
LAPAROSCOPIC CHOLECYSTECTOMY The first human LC has been attributed to a French general surgeon, Phillipe Mouret, back in 1987.121 Soon afterward, in 1988, McKerman and Saye performed the first LC in the USA.122 Later Reddick and Olsen published the first reports of LC and introduced the procedure of laparoscopic cholangiography.123 The procedure begins by creating a pneumoperitoneum of carbon dioxide with a spring-loaded safety needle (Veress needle) or open peritoneal access with a Hasson trocar, or use of an optical trocar. Once the pneumoperitoneum is established to a pressure of 15 mmHg, four ports are inserted into the abdomen (Figure 65-2). Mobilization of the gallbladder starts at its base, removing peritoneal attachments and exposing the gallbladder infundibulum. Posterolateral retraction of the gallbladder is essential to expose Calot’s triangle and assist in defining the biliary and arterial anatomy (Figure 65-3).124 Once the cyst duct and artery are identified, the proximal ends are ligated with clips and divided. Final dissection of the gallbladder off the surface of the liver is carried out with monopolar cautery in a bottom-up manner. The gallbladder specimen can be removed through either the subxiphoid or periumbilical port site. Although up to 95% of laparoscopic cholecystectomies are performed without incident, the complications, when they occur, can be devastating (Table 65-10). The portion of the procedure most
15cm
Figure 65-2. Port site placements for laparoscopic cholecystectomy. The gallbladder infundibulum sits at the apex of a diamond 15 cm from the umbilicus. The surgeon’s left and right hands occupy the trocars on the left and right corners of the diamond.
Table 65-10. Complications of Laparoscopic Cholecystectomy Wound infection Bowel injury Hemorrhage Bile duct injury Hepatic artery injury Cystic duct stump leak Biloma Abscess Gallstone spillage Bile duct stricture Retained common bile duct stone
disconcerting to the surgeon is appropriate identification of bile duct structures. Intense scarring due to chronic inflammation can make this rather difficult and it is often cited as a reason for converting to an open procedure. Intraoperative cholangiography can assist by providing a roadmap of biliary anatomy and help with the identification and prevention of major injury.125 Others find that intraoperative cholangiogram itself does not prevent bile duct injury, but mitigates against extension and worsening of a bile duct injury.126,127 In order to perform this procedure, dissection of the cystic duct is essential. Once control of the cystic duct is obtained, a ductotomy is created and a soft catheter is passed into the common bile duct. Water-soluble contrast is then injected and real-time fluoroscopy is performed to trace the flow of contrast proximally into the liver and distally to the duodenum. Abnormalities may appear as filling defects or extravasation. The incidence of major complications during the learning phase of LC was high and has steadily decreased over time. Due to its
1207
Section XI. Diseases of the Biliary Tract
Ultrasound/CT
(+) Fluid
(-) Fluid
Aspirate and leave drain
Pus
Blood
Antibiotics
Resolved
Unstable
Stable
Bile ducts not dilated
LFTs abnormal Drain stopped
Persistent
Bile ducts dilated
Bile
LFTs normal
Drain persistent ERCP
Surgery No further therapy
No further therapy
ERCP
Surgery Negative
Positive Cystic duct stump leak
Bile duct injury
Sphincterotomy and stent
Refer to biliary injury center
Stops
No further therapy
Negative No surgical complication
Observe (Duct of Luschka injury?)
Retained stone
Bile duct injury
Continues
Surgery
Extract stone +/– sphincterotomy
Sphincterotomy if leak Image biliary tract with PTC if ERCP not adequate Refer or repair
Figure 65-3. Algorithm for evaluation and treatment of bile duct injury. ERCP, endoscopic retrograde cholangiopancreatography; LFTs, liver function tests; PTD.
severe nature, the injury that concerns the general surgeon the most is injury to the bile ducts. Reflecting this severity, the Physician Insurers Agency of America reported in 2000 that common bile duct injury from LC was the leading source of medical malpractice claims against general surgeons. Recent retrospective analysis of the Medicare database over 7 years and 1.5 million patients has shown the incidence of common bile duct injuries to be stable at 0.5%.128 Furthermore, in the Medicare population the adjusted hazard risk for death was 2.8 times greater for those who had sustained a bile duct injury compared to those who had routine cholecystectomy. Therefore, despite the revolutionary improvement that LC confers to the patients, morbidity and mortality remain substantial risks for those who have complications.
1208
The study of bile duct injury has developed into an entirely new field of research since the entrance of LC. In order to characterize the injury, classification schemes have been developed (Tables 6511–65-13). Based on these data, the most common bile duct injury occurs when the common bile duct or common hepatic duct is misidentified as the cystic duct and erroneously ligated and transected. In fact, up to 75% of the time, the surgeon does not realize an injury has occurred until problems occur in the postoperative period.129–131 Even if an intraoperative cholangiogram is performed, proper interpretation of the images by the surgeon fails to recognize a bile duct injury in a majority of the cases.129,130,132 If the injury is recognized, further management in the operating room is dictated by the skill and training of the operating surgeon. Definitive repair
Chapter 65 SURGERY OF THE BILIARY TRACT
Table 65-11. Olsen Classification of Bile Duct Injury127
Table 65-14. Symptoms of Bile Duct Injury137
Type I (17%) Type II (6%) Type III (65%) Type IV (11%) Type V (1%)
Refractory pain Nausea Vomiting Jaundice Fever Ileus Anorexia
Simple laceration or hole in duct Ligation without transection Ligation with transection Right hepatic duct injury Stricture of unknown cause
Table 65-12. Csendes Classification of Bile Duct Injury133 Type I (31%) Type II (30%) Type III (24%) Type IV (15%)
Small tear of hepatic duct during dissection Injury at cysticocholedochal junction Partial or complete transection of bile duct Resection of 1 cm of bile duct
Table 65-13. Stewart–Way Classification of Bile Duct Injury134 Type I (7%) Type II (22%) Type III (61%) Type IV (10%)
Common bile duct incision Clip or cautery to common hepatic duct Common bile duct, common hepatic duct, right/left hepatic duct transection/resection Right hepatic duct and artery transected
at the time of the operation is ideal if the surgeon is capable of performing a proper repair. Calling for assistance from an experienced hepatobiliary surgeon would be ideal, and certainly transferring the patient to a tertiary care center would be an excellent alternative if the operating surgeon has not had the proper training to deal with the problem.117,119,135,136 Unfortunately the majority of bile duct injuries are not identified at the time of surgery, and can go unrecognized for days to weeks.
Suspected Bile Duct Injury By far the majority of laparoscopic cholecystectomies performed in the USA are uneventful, with most patients leaving the hospital within a day of surgery. The subset of patients who are unable to be discharged because of continued problems are those that surgeons must be most suspicious of having a bile duct injury, and therefore must have a low threshold to proceed with further evaluation. Biliary tract injury symptoms are often complex and non-specific, sometimes very difficult to distinguish from normal postoperative problems (Table 65-14). These symptoms may mimic the usual side effects of laparoscopy, surgery, anesthesia, and medication, which may make diagnosis of bile duct injury difficult. Once the common causes of postoperative recovery failures are excluded, ruling out bile duct injury is mandatory. Initially, laboratory studies are obtained, specifically, liver tests and a complete blood count. A significant decrease in hemoglobin may indicate ongoing hemorrhage or a large hematoma. A significant elevation in white blood cell count may signify an infectious process such as abscess formation. Rising total bilirubin may demonstrate the presence of a retained common bile duct stone, biloma, or occlusion of the common bile duct from a clip.
Ultrasound or computed tomography is helpful in identifying abnormal fluid collections and intrahepatic bile duct dilation. Once a fluid collection is identified, it is imperative to discover the type of fluid that is present – blood, bile, ascites, pus, or simple irrigation fluid. The fluid should be aspirated percutaneously by an experienced interventional radiologist and sent for bilirubin analysis, amylase, and Gram stain. The presence of bilirubin above serum levels is indicative of biliary tract injury or intestinal injury. Duodenal perforation or pancreatic injury can result in extremely high amylase. Organisms identified by Gram stain can help diagnose the presence of peritoneal contamination. Additionally, cholescintigraphy (HIDA) can be useful as a noninvasive screening test for the presence of a bile leak and is approximately 85% accurate.138 Recent reports have suggested that MRCP is another useful non-invasive alternative for diagnosis of bile duct injury; however, its main limitation is that it is not therapeutic.139,140 The gold standard for imaging in search of a bile duct injury is still (PTC) or ERCP. PTC may be limited by the absence of bile duct dilation, but it does provide a roadmap of the intrahepatic biliary system when ERCP cannot be performed or the contrast does not go past an obstruction. ERCP is considered the optimal method of ruling out bile duct injury because it is both diagnostic and potentially therapeutic. Once complete evaluation is performed and all imaging and laboratory analyses return to normal, then the likelihood of a surgical complication is low (Figure 65-4).
BILE DUCT INJURY MANAGEMENT The ideal setting to perform repair of a bile duct injury is at the same time the injury was created. Conversion to an open procedure is certainly advisable, but is somewhat dependent on the skill of the surgeon and the extent of the injury. In most instances laparoscopic placement of a clip on the transected ends of the bile duct (if visible) and laparoscopic placement of a drain in the porta hepatis is the best primary management. The patient should be transferred at the earliest convenience to a tertiary biliary referral center. In tertiary referral centers, most experts feel that a primary repair should only be performed infrequently, and only if there is a clean cut less than 50% of the circumference of the bile duct. Placement of a T-tube through a separate choledochotomy is then mandatory. If there is any debate as to the quality of the repair, multiple series have identified that the Roux en Y hepaticojejunostomy is a superior operation and should be used most frequently.141–144 Mucosal-to-mucosal approximation without tension is paramount to success of this repair. Once performed, long-term success rates can be greater than 90% in many cases.
1209
Section XI. Diseases of the Biliary Tract
Figure 65-4. Calot’s triangle. Exposed cystic duct, cystic artery, and gallbladder.
The majority of surgeons performing LC are not trained hepatobiliary surgeons and do not work at tertiary care referral centers. For these surgeons, in these locales, transfer of the patient to a specialty care center with simple placement of a drainage catheter in the gallbladder fossa is advisable, with no attempt at biliary exploration. The highest likelihood for successful operation is performed on the first attempt. Urgent referral can then be made, and the patient can be transferred to the appropriate institution for proper repair. When referred, a patient with a known bile duct injury must have a proper workup prior to undergoing a major reconstruction. Sepsis must be controlled with antibiotics and drainage of bile. Biliary congestion should be relieved with PTC, which allows for simultaneous imaging of the biliary system. Computed tomography or ultrasound is necessary to evaluate for the presence of any suspicious fluid collections that require drainage. ERCP may be helpful in the setting of an ongoing biliary fistula or suspected retained common bile duct stone. Because hepatic artery injury occurs in approximately 20–61% of bile duct injuries,129 any suspicion of vascular injury requires performance of angiography to elucidate proper biliary blood flow necessary for successful repair. Once the patient is properly evaluated and anatomy defined, the optimal timing of repair is soon after the injury occurred, if recognized early. Emergency operations are limited to situations in which the patient is succumbing from overwhelming bile peritonitis and sepsis. Most patients are clinically stable, and allowing the inflam-
1210
mation to subside from bile peritonitis facilitates the dissection and permits sewing to more healthy tissue.
Biloma Postoperative bile leaks occur in 0.65–2%130,131,145,146 of laparoscopic cholecystectomies. It may signify a bile duct injury that was not recognized at the time of surgery, a cystic duct stump leak, a draining cholecystohepatic duct (duct of Luschka), or may never be elucidated. Once a biloma is identified, percutaneous drainage is followed by ERCP because it allows the gastroenterologist to diagnose and potentially treat the problem. If a bile duct injury is identified on ERCP, sphincterotomy and stenting are helpful in reducing the outflow of bile into the peritoneal cavity by creating a pathway of least resistance through the enlarged common bile duct. However, referral to an experienced hepatobiliary surgeon is certainly warranted. Leakage from the cystic duct can be identified by ERCP as well. These injuries can occur from misplacement of clips across a large cystic duct, clip malfunction, or a missed hole in the cystic duct. Regardless of the etiology of cystic duct stump leaks, the treatment remains essentially the same. ERCP with sphincterotomy alone or with stent placement is a non-surgical method of treating this problem, with success rates of nearly 100%.141,147–149 However, if the leakage persists, surgical exploration may be required to identify the source of the leak and directly repair it.
Chapter 65 SURGERY OF THE BILIARY TRACT
Ducts of Luschka are small bile ducts found in the gallbladder fossa that branch from the right hepatic or common hepatic ducts. These ducts are blind ends and do not drain the liver parenchyma. They are distinct from the cystohepatic ducts that drain directly into the gallbladder or cystic duct.150 During dissection and removal of the gallbladder from its bed, these ducts can be injured, resulting in persistent bile leak. These injuries can usually be prevented by careful dissection staying along the wall of the gallbladder.151 If they are identified at the time of surgery, they should be ligated appropriately. However, often they are too small to see and are divided. Subsequent biloma requires ERCP evaluation which may identify the ducts. Treatment is similar to a cystic duct stump leak in that sphincterotomy with or without a stent is usually sufficient, with success rates exceeding 90%.152–154
Table 65-15. Bismuth Classification and Repairs Classification
Type of repair
I: ≥ 2 cm of common bile duct II: < 2 cm of common bile duct
Roux en Y side-to-side hepaticojejunostomy Extend ductotomy on to left hepatic duct and perform Roux en Y side-toside hepaticojejunostomy Lower hilar plate via Hepp–Couinaud159 technique and expose left hepatic duct for anastomosis to Roux en Y hepaticojejunostomy Hepp–Couinaud technique and separate left and right hepatic duct anastomosis to Roux en Y hepaticojejunostomy As above, with anastomosis of right hepatic duct
III: At bifurcation of hepatic ducts
IV: Confluence destroyed, hepatic ducts separated V: Type I, II, or III and separate right hepatic duct
BILE DUCT STRICTURES Development of narrowing of the bile ducts can occur as a result of benign or malignant disease. Chronic pancreatitis, cholecystitis, trauma, prior liver transplant, and abscesses can result in severe scarring and narrowing of the bile duct. Unfortunately, the most common reason why surgeons see and evaluate bile duct strictures is iatrogenic injury of the bile duct during cholecystectomy. Bile duct injury during surgery predisposes patients to stricture formation due to the tenuous blood supply of the bile duct. The portion of the common bile duct that is most likely to be injured during cholecystectomy is supplied mainly from branches from below (60%), commonly the cystic artery and right hepatic artery. There are two main arteries that run at the 3 o’clock and 9 o’clock positions along the sides of the common bile duct. When the duct is injured, especially close to the hilar plate, the blood supply from below is lost, thus leading to a high rate of ischemia and subsequent stricture formation.150,155 Patients who have suffered a bile duct injury from cholecystectomy, and undergone attempted repair via primary end-to-end anastomosis or Roux en Y hepaticojejunostomy, are at risk for later stricture formation. End-to-end anastomosis for complete transection has a very high stricture rate – as much as 100% – whereas if there is only a partial transection this decreases the stricture rate to 57%.141 Therefore most authors choose Roux en Y hepaticojejunostomy as the best alternative, with a long-term stricture rate of less than 10% in qualified hands.149,156,157 The location of bile duct strictures is essential to planning the proper repair. Bismuth classified155 these strictures and offered solutions to each of these158 (Table 65-15). Once a stricture does occur from an attempted bile duct injury repair, reoperative management can be extremely challenging, especially in patients who are poor operative candidates. Endoscopic and percutaneous techniques have been developed to treat these strictures as an alternative to surgery. Balloon dilation is usually best performed through a PTD access. Successful treatment of strictures has been demonstrated best in dealing with troublesome narrowing of bile duct anastomoses in liver transplant recipients. Endoscopic stent placement in these patients is associated with long-term success rates of upwards of 90%.160 For those with strictures from biliary reconstruction due to bile duct injury, the results are not as favorable. The process involves two sep-
arate approaches that depend on the anatomy of the patients. First, if a stricture occurs after primary end-to-end anastomosis of the common bile duct without biliary-enteric bypass, endoscopy with dilation and stenting is possible. Only plastic stents should be used in these situations, and recurrent strictures or stent occlusions are indications for re-repair. Otherwise, if a stricture occurs at the site of a biliary-enteric anastomosis, most commonly a Roux en Y hepaticojejunostomy then endoscopy is not possible, and percutaneous techniques are employed. Problems occur in maintaining stent patency as they often occlude with debris, requiring frequent changes and increasing the risk for overwhelming cholangitis.161 Limited long-term data are available, but primary patency is reported at 50–72% up to 6 years, with most failures occurring within 2–3 years.162–164
CHOLECYSTECTOMY IN CIRRHOSIS The National Institutes of Health consensus conference in 1992 for LC regarded those with advanced cirrhosis with portal hypertension as having a relative contraindication to LC.165 Since then there have been multiple case series of successful LC in all stages of cirrhosis as long as pericholecystic varices are not present. Recently, Puggioni and Wong in 2003 performed a meta-analysis of publications between 1993 and 2001 specifically addressing this issue.166 The main questions to be answered are: (1) Is LC as safe as open cholecystectomy in the cirrhotic patient? (2) Is LC as safe in cirrhotic as non-cirrhotic patients? The data are summarized in Tables 65-16 and 65-17 for patients with good liver function (Child’s A). There is limited experience with LC in Child’s B patients, and Child’s C patients should undergo transplantation for severe recurrent symptoms of gallstone disease. Another consideration is the possibility of future liver transplantation in those with cirrhosis. Open cholecystectomy may result in more adhesion formation, especially in the region of the porta hepatis, making future transplantation more challenging. In those circumstances LC is a much more attractive option, despite its pitfalls. Preoperative planning is required to perform the operation successfully and safely with minimal morbidity and mortality. Liver tests, coagulation tests, and nutrition should be optimized prior to
1211
Section XI. Diseases of the Biliary Tract
Table 65-16. Laparoscopic versus Open Cholecystectomy in Cirrhotics
Blood loss Operative time Length of stay Morbidity Mortality Wound infection
Laparoscopic cholecystectomy
Open cholestectomy
= = =
+ + + = = =
Table 65-17. Cirrhotic versus Non-Cirrhotic Undergoing Laparoscopic Cholecystectomy
Conversion to open Operative time Blood loss Morbidity Mortality Wound infection Length of stay
Cirrhosis
No Cirrhosis
+ + + + = = +
= = -
any surgical intervention. This will help prevent excessive blood loss and worsening of ascites, and ensure adequate wound-healing. Communication with the anesthesiologist preoperatively is critical to ensure a safe perioperative period.
CHOLECYSTECTOMY IN PREGNANCY The physiology of pregnancy has significant biliary ramifications. Among these is alteration of the enterohepatic circulation, resulting in bile stasis and increases in cholesterol in bile with decreased chenodeoxycholic acid presumably due to hormonal changes with pregnancy.167–169 Progesterone in particular is implicated as the hormone that results in decreased gallbladder emptying and has been found in animal studies to promote gallstone formation when given exogenously.170,171 A large controlled study investigating the presence of gallstones in the immediate postpartum period compared with a control population revealed an incidence of gallstones in 12% of postpartum women versus 1.3% of controls.172 However, most series report approximately 2% incidence of gallstones during pregnancy, with most forming gallstones in the second and third trimester.168,173–174 Fortunately, most gallstones are asymptomatic but symptoms can occur in up to 30% of individuals.172,174 Once symptoms occur, medical management usually fails, with some authors reporting as high as 70% recurrence of symptoms or complications during pregnancy.175,176 Because of the significant risk to both the fetus and the mother of complications from gallstones, surgery is an important therapeutic and preventive measure. Indications for surgery include failure of non-operative management to control symptoms, and complications from gallstones such as pancreatitis and cholecystitis. These patients are at high risk for premature labor, and fetal and maternal demise, and require surgery. Weber et al. in 1991 performed the first LC in a pregnant patient successfully.177 Since then many surgeons have adapted this technique to their practice in caring for the pregnant patient with gall-
1212
Table 65-18. SAGES 2000 Guidelines for Laparoscopic Surgery During Pregnancy 1. Obstetrical consultation should be obtained preoperatively 2. When possible, operative intervention should be deferred until the second trimester, when fetal risk is lowest 3. Pneumoperitoneum enhances lower-extremity venous stasis already present in the gravid patient and pregnancy induces a hypercoagulable state. Therefore pneumatic compression devices should be utilized whenever possible 4. Fetal and uterine status, as well as maternal end-tidal carbon dioxide and/or arterial blood gases, should be monitored 5. The uterus should be protected with a lead shield if intraoperative cholangiography is a possibility. Fluoroscopy should be utilized selectively 6. Given the enlarged gravid uterus, abdominal access should be attained using an open technique 7. Dependent positioning should be utilized to shift the uterus off the inferior vena cava 8. Pneumoperitoneum pressures should be minimized (to 8–12 mmHg) and not allowed to exceed 15 mmHg
bladder disease. There have been some concerns initially with the use of laparoscopy in pregnant patients due to the CO2 pneumoperitoneum, introduction of the Veress needle, and exposure secondary to the gravid uterus. However, a recent study observed that hemodynamic changes were not significantly different between pregnant and non-pregnant patients with CO2 pneumoperitoneum.176 Of the many possible effects of the CO2 pneumoperitoneum, including changes in uterine blood flow, increased fetal acidosis, and increased abdominal pressure on the fetus, only fetal acidosis was shown to be concerning in fetal sheep models.162 Hyperventilation of the mother to a PCO2 of 30–35 mmHg protected the fetus from acidosis and should be the standard protocol for humans as well. Due to the enlarged uterus, placement of Veress needle may be changed to the supraumbilical or right upper quadrant location. Alternatively the Hasson open technique can be performed. Care must be taken during trocar insertion to avoid uterine injury so direct-access techniques are safest. The bowel is often displaced superiorly out of the pelvis so care must be taken to avoid injury. Although the gravid uterus in the third trimester takes up the majority of the abdomen, establishment of pneumoperitoneum does allow for enough working room to perform a safe LC. The patient should be placed in a slight left lateral position to remove the pressure off the inferior vena cava by the gravid uterus. Maintaining a lower limit for pneumoperitoneum in the 10–12 mmHg range may decrease CO2 absorption, but should not be too low to compromise exposure. Hyperventilation should be performed to keep end-tidal PCO2 in the 25–30 mmHg range. Deep-vein thrombosis prophylaxis in the form of sequential compression devices helps mitigate against the risk of forming deep-vein thrombosis in this hypercoagulable patient. The surgical dictum for timing of safe cholecystectomy offered by both the National Institutes of Health consensus conference in 1992 and SAGES guidelines in 2000 has been that the second trimester is the safest, as it avoids the teratogenic effects of medications in the first trimester, and the risk of premature labor in the third trimester (Table 65-18).165,179 However, a few studies have
Chapter 65 SURGERY OF THE BILIARY TRACT
reported safe and successful LC in the first and third trimester.176,180,181 We believe that patients should be managed nonoperatively if possible to either the second trimester or after the baby’s birth rather than risk an operation during the first or third trimesters. Fortunately, if complications from gallstones arise during the first or third trimester requiring an operation, it can be performed safely.
REFERENCES 1. Hall MJ, DeFrances CJ. 2001 National hospital discharge survey. No. 332, April 9, 2003. 2. Sarin SK, Negi VS, Dewan R, et al. High familial prevalence of gallstones in the first-degree relatives of gallstone patients. Hepatology 1995; 22:138–141. 3. Sampliner RE, Bennett PH, Comess LJ, et al. Gallbladder disease in Pima indians. Demonstration of high prevalence and early onset by cholecystography. N Engl J Med 1970; 283:1358–1364. 4. Browning JD, Horton JD. Gallstone disease and its complications. Semin Gastrointest Dis 2003; 14:165–177. 5. Mentes BB, Akin M, Irkorucu O, et al. Gastrointestinal quality of life in patients with symptomatic or asymptomatic cholelithiasis before and after laparoscopic cholecystectomy. Surg Endosc 2001; 15:1267–1272. 6. Lawrentschuk N, Hewitt PM, Pritchard MG. Elective laparoscopic cholecystectomy: implications of prolonged waiting times for surgery. Aust NZ J Surg 2003; 73:890–893. 7. Howard DE. Nonsurgical management of gallstone disease. Gastroenterol Clin North Am 1999; 28:133–144. 8. Gracie WA, Ransohoff DF. The natural history of silent gallstones: the innocent gallstone is not a myth. N Engl J Med 1982; 307:798–800. 9. McSherry CK, Ferstenberg H, Calhoun WF, et al. The natural history of diagnosed gallstone disease in symptomatic and asymptomatic patients. Ann Surg 1985; 202:59–63. 10. Friedman GD, Raviola CA, Fireman B. Prognosis of gallstones with mild or no symptoms: 25 years of follow-up in a health maintenance organization. J Clin Epidemiol 1989; 42:127–136. 11. Cucchiaro G, Rossitch JC, Bowie J, et al. Clinical significance of ultrasonographically detected coincidental gallstones. Dig Dis Sci 1990; 35:417–421. 12. Wada K, Wada K, Imamura T. (Natural course of asymptomatic gallstone disease.) Nippon Rinsho 1993; 51:1737–1743. 13. Juhasz ES, Wolff BG, Meagher AP, et al. Incidental cholecystectomy during colorectal surgery. Ann Surg 1994; 219:467–472; discussion 472–474. 14. Attili AF, De Santis A, Capri R, et al. The natural history of gallstones: the GREPCO experience. The GREPCO Group. Hepatology 1995; 21:655–660. 15. Angelico F, Del Ben M, Barbato A, et al. Ten-year incidence and natural history of gallstone disease in a rural population of women in central Italy. The Rome Group for the Epidemiology and Prevention of Cholelithiasis (GREPCO). Ital J Gastroenterol Hepatol 1997; 29:249–254. 16. Halldestam I, Enell EL, Kullman E, et al. Development of symptoms and complications in individuals with asymptomatic gallstones. Br J Surg 2004; 91:734–738. 17. Papi C, Catarci M, D’Ambrosio L, et al. Timing of cholecystectomy for acute calculous cholecystitis: a metaanalysis. Am J Gastroenterol 2004; 99:147–155. 18. Lo CM, Liu CL, Fan ST, et al. Prospective randomized study of early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Ann Surg 1998; 227:461–467.
19. Uchiyama K, Onishi H, Tani M, et al. Timing of laparoscopic cholecystectomy for acute cholecystitis with cholecystolithiasis. Hepatogastroenterology 2004; 51:346–348. 20. Hanau LH, Steigbigel NH. Acute (ascending) cholangitis. Infect Dis Clin North Am 2000; 14: 521–546 21. Yusoff IF, Barkun JS, Barkun AN. Diagnosis and management of cholecystitis and cholangitis. Gastroenterol Clin 2003; 32:1145–1168. 22. Chen MF, Jan YY, Lee TY. Percutaneous transhepatic biliary drainage for acute cholangitis. Int Surg 1987; 72:131–133. 23. Charcot JM. Leçons sur les maladies du foie des voices bilars et des reins. Paris: Faculté de Medicine de Paris, Bournville et Sevestre; 1877. 24. Reynolds BM, Dargan EL. Acute obstructive cholangitis: a distinct clinical syndrome. Ann Surg 1959; 150:299–303. 25. Pessa ME, Hawkins IF, Vogel SB. The treatment of acute cholangitis. Percutaneous transhepatic biliary drainage before definitive therapy. Ann Surg 1987; 205:389–392. 26. Siegel JH, Rodriquez R, Cohen SA, et al. Endoscopic management of cholangitis: critical review of an alternative technique and report of a large series Am J Gastroenterol 1994; 89:1142–1146. 27. Lai EC, Mok FP, Tan ES, et al. Endoscopic biliary drainage for severe acute cholangitis. N Engl J Med 1992; 326:1582–1586. 28. Leese T, Neoptolemos JP, Baker AR, et al. Management of acute cholangitis and the impact of endoscopic sphincterotomy. Br J Surg 1986; 73:988–992. 29. Mueller PR, van Sonnenberg E, Ferrucci JT Jr. Percutaneous biliary drainage: technical and catheter-related problems in 200 procedures. AJR Am J Roentgenol 1982; 138:17–23. 30. Gould RJ, Vogelzang RL, Neiman HL, et al. Percutaneous biliary drainage as an initial therapy in sepsis of the biliary tract. Surg Gynecol Obstet 1985; 160:523–527. 31. Ranson J, Rifkind K, Roses D. Prognostic signs and the role of operative management in acute pancreatitis. Surg Gynecol Obstet 1974; 139:69–81. 32. Lai EC, Tam PC, Paterson IA, et al. Emergency surgery for severe acute cholangitis. The high-risk patients. Ann Surg 1990; 211:55–59. 33. Welch JP, Donaldson GA. The urgency of diagnosis and surgical treatment of acute suppurative cholangitis. Am J Surg 1976; 131:527–532. 34. Blamey S, Imrie C, O’Neill J, et al. Prognostic factors in acute pancreatitis. Gut 1984; 25:1340–1346. 35. Bradley EL 3rd. A clinically based classification system for acute pancreatitis. Summary of the International Symposium on Acute Pancreatitis, Atlanta, GA, September 11 through 13, 1992. Arch Surg 1993; 128:586–590. 36. Ostermiller W Jr, Thompson RJ Jr, Carter R, et al. Acute obstructive cholangitis. Arch Surg 1965; 90:392–395. 37. Lau WY, Chu KW, Yuen WK, et al. Operative choledochoscopy in patients with acute cholangitis: a prospective, randomized study. Br J Surg 1991; 78:1226–1229. 38. Poon RT, Liu CL, Lo CM, et al. Management of gallstone cholangitis in the era of laparoscopic cholecystectomy. Arch Surg 2001; 136:11–16. 39. Banks PA. Practice guidelines in acute pancreatitis. Am J Gastroenterol 1997; 92:377–386. 40. Toh SK, Phillips S, Johnson CD. A prospective audit against national standards of the presentation and management of acute pancreatitis in the South of England. Gut 2000; 46:239–243. 41. Tanaka M, Ikeda S, Yoshimoto H, et al. The long-term fate of the gallbladder after endoscopic sphincterotomy. Complete follow-up study of 122 patients. Am J Surg 1987; 154:505–509. 42. Hammarstrom LE, Holmin T, Stridbeck H, et al. Long-term follow-up of a prospective randomized study of endoscopic
1213
Section XI. Diseases of the Biliary Tract
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57. 58.
59.
60.
61.
1214
versus surgical treatment of bile duct calculi in patients with gallbladder in situ. Br J Surg 1995; 82:1516–1521. Targarona EM, Ayuso RM, Bordas JM, et al. Randomised trial of endoscopic sphincterotomy with gallbladder left in situ versus open surgery for common bile duct calculi in high-risk patients. Lancet 1996; 347:926–929. Suc B, Escat J, Cherqui D, et al. Surgery vs endoscopy as primary treatment in symptomatic patients with suspected common bile duct stones: a multicenter randomized trial. French Associations for Surgical Research. Arch Surg 1998; 133:702–708. Knaus W, Draper E, Wagner D, et al. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13:818–829. Glazer G, Mann DV. United Kingdom guidelines for the management of acute pancreatitis. British Society of Gastroenterology. Gut 1998; 42 (Suppl 2):S1–S13. Isla A, Griniatsos J, Rodway A. Single-stage definitive laparoscopic management in mild acute biliary pancreatitis. J Laparoendosc Adv Surg Tech A 2003; 13:77–81. Cuschieri A, Lezoche E, Morino M, et al. EAES multicenter prospective randomized trial comparing two-stage vs single-stage management of patients with gallstone disease and ductal calculi. Surg Endosc 1999; 13:952–957. Kaw M, Al-Antably Y, Kaw P. Management of gallstone pancreatitis: cholecystectomy or ERCP and endoscopic sphincterotomy. Gastrointest Endosc 2002; 56:61–65. Alimoglu O, Ozkan OV, Sahin M, et al. Timing of cholecystectomy for acute biliary pancreatitis: outcomes of cholecystectomy on first admission and after recurrent biliary pancreatitis. World J Surg 2003; 27:256–259, 2003. Epub 2003 Feb 27. Uhl W, Warshaw A, Imrie C, et al. International Association of Pancreatology. IAP guidelines for the surgical management of acute pancreatitis. Pancreatology 2002; 2:565–573. Sharma VK, Howden CW. Metaanalysis of randomized controlled trials of endoscopic retrograde cholangiography and endoscopic sphincterotomy for the treatment of acute biliary pancreatitis. Am J Gastroenterol 1999; 94:3211–3214. Ong SK, Christie PM, Windsor JA. Management of gallstone pancreatitis in Auckland: progress and compliance. Aust NZ J Surg 2003; 73:194–199. Ammori BJ, Davides D, Vezakis A, et al. Laparoscopic cholecystectomy: are patients with biliary pancreatitis at increased operative risk? Surg Endosc 2003; 17:777–780. Epub 2003 Feb 18. Fan ST, Lai EC, Mok FP, et al. Early treatment of acute biliary pancreatitis by endoscopic papillotomy. N Engl J Med 1993; 328:228–232. Rhodes M, Armstrong CP, Longstaff A, et al. Laparoscopic cholecystectomy with endoscopic retrograde cholangiopancreatography for acute gallstone pancreatitis. Br J Surg 1993; 80:247. Tate JJ, Lau WY, Li AK. Laparoscopic cholecystectomy for biliary pancreatitis. Br J Surg 1994; 81:720–722. Canal DF, Broadie TA. Results of laparoscopic cholecystectomy for the treatment of gallstone pancreatitis. Am Surg 1994; 60:495–498; discussion 498–499. Soper NJ, Brunt LM, Callery MP, et al. Role of laparoscopic cholecystectomy in the management of acute gallstone pancreatitis. Am J Surg 1994; 167:42–50; discussion 50–51. DeIorio AV Jr, Vitale GC, Reynolds M, et al. Acute biliary pancreatitis. The roles of laparoscopic cholecystectomy and endoscopic retrograde cholangiopancreatography. Surg Endosc 1995; 9:392–396. Tang E, Stain SC, Tang G, et al. Timing of laparoscopic surgery in gallstone pancreatitis. Arch Surg 1995; 130:496–499; discussion 499–500.
62. Ballestra-Lopez C, Bastida-Vila X, Bettonica-Larranaga C, et al. Laparoscopic management of acute biliary pancreatitis. Surg Endosc 1997; 11:718–721. 63. Ricci F, Castaldini G, de Manzoni G, et al. Minimally invasive treatment of acute biliary pancreatitis. Surg Endosc 1997; 11:1179–1182. 64. Uhl W, Muller CA, Krahenbuhl L, et al. Acute gallstone pancreatitis: timing of laparoscopic cholecystectomy in mild and severe disease. Surg Endosc 1999; 13:1070–1076. 65. Neoptolemos JP, Carr-Locke DL, London NJ, et al. Controlled trial of urgent endoscopic retrograde cholangiopancreatography and endoscopic sphincterotomy versus conservative treatment for acute pancreatitis due to gallstones. Lancet 1988; 2:979–983. 66. Folsch UR, Nitsche R, Ludtke R, et al. Early ERCP and papillotomy compared with conservative treatment for acute biliary pancreatitis. The German study group on acute biliary pancreatitis. N Engl J Med 1997; 336:237–242. 67. NIH state-of-the-science statement on endoscopic retrograde cholangiopancreatography (ERCP) for diagnosis and therapy. NIH Consens State Sci Statements 2002; 19:1–26. 68. McSherry CK, Ferstenberg H, Virshup M. The Mirizzi syndrome: suggested classification and surgical therapy. Surg Gastroenterol 1982; 1:219–225. 69. Nishimura A, Shirai Y, Hatakeyama K. High coincidence of Mirizzi syndrome and gallbladder carcinoma. Surgery 1999; 126:587–588. 70. Tondelli P, Stutz K, Harder F, et al. Acute gallstone pancreatitis: best timing for biliary surgery. Br J Surg 1982; 69:709–710. 71. Runkel NS, Buhr HJ, Herfarth C. Outcome after surgery for biliary pancreatitis. Eur J Surg 1996; 162:307–313. 72. Mirizzi PL. Sindrome del conducto hepatico. J Int Chir 1948; 8:731–735. 73. Yeh CN, Jan YY, Chen MF. Laparoscopic treatment for Mirizzi syndrome. Surg Endosc 2003; 17:1573–1578. Epub 2003 Jul 11, 2003. 74. Csendes A, Diaz JC, Burdiles P, et al. Mirizzi syndrome and cholecystobiliary fistula: a unifying classification. Br J Surg 1989; 76:1139–1143. 75. Johnson LW, Sehon JK, Lee WC, et al. Mirizzi’s syndrome: experience from a multi-institutional review. Am Surg 2001; 67:11–14. 76. VanLandingham SB, Broders CW. Gallstone ileus. Surg Clin North Am 1982; 62:241–247. 77. Redaelli CA, Buchler MW, Schilling MK, et al. High coincidence of Mirizzi syndrome and gallbladder carcinoma. Surgery 1997; 121:58–63. 78. Schafer M, Schneiter R, Krahenbuhl L. Incidence and management of Mirizzi syndrome during laparoscopic cholecystectomy. Surg Endosc 2003; 17:1186–1190; discussion 1191–1192. Epub 2003 May 13. 79. Sandblom P, Tabrizian M, Rigo M, et al. Repair of common bile duct defects using the gallbladder or cystic duct as a pedicled graft. Surg Gynecol Obstet 1975; 140:425–432. 80. Baer HU, Matthews JB, Schweizer WP, et al. Management of the Mirizzi syndrome and the surgical implications of cholecystcholedochal fistula. Br J Surg 1990; 77:743–745. 81. Kasahara Y, Umemura H, Shiraha S, et al. Gallstone ileus. Review of 112 patients in the Japanese literature. Am J Surg 1980; 140:437–440. 82. Kurtz RJ, Heimann TM, Kurtz AB. Gallstone ileus: a diagnostic problem. Am J Surg 1983; 146:314–317. 83. Pavlidis TE, Atmatzidis KS, Papaziogas BT, et al. Management of gallstone ileus. J Hepatobiliary Pancreat Surg 2003; 10:299–302. 84. Schutte H, Bastias J, Csendes A, et al. Gallstone ileus. Hepatogastroenterology 1992; 39:562–565.
Chapter 65 SURGERY OF THE BILIARY TRACT
85. Deitz DM, Standage BA, Pinson CW, et al. Improving the outcome in gallstone ileus. Am J Surg 1986; 151:572–576. 86. Lennon GM, Browne R, Watson RG, et al. Gallstone ileus following cholecystectomy and side to side choledochoduodenostomy. Ir J Med Sci 1990; 159:287–288. 87. van Hillo M, van der Vliet JA, Wiggers T, et al. Gallstone obstruction of the intestine: an analysis of ten patients and a review of the literature. Surgery 1987; 101:273–276. 88. Reisner RM, Cohen JR. Gallstone ileus: a review of 1001 reported cases. Am Surg 1994; 60:441–446. 89. Elewaut A, Afschrift M, Barbier F. Gallstone ileus treated with extracorporeal shock wave lithotripsy. J Clin Ultrasound 1993; 21:343–345. 90. Williams IM, Hughes OD, Hicks E, et al. Gall stone ileus following multiple endoscopic retrograde cholangiopancreatographies. J R Coll Surg Edinb 1997; 42:423. 91. Gandhi A, Maxwell AJ, Wells S, et al. Gallstone ileus following endoscopic sphincterotomy. Br J Hosp Med 1995; 54:229–230. 92. Lancaster JF, Strong RW, McIntyre A, et al. Gallstone ileus complicating endoscopic sphincterotomy. Aust NZ J Surg 1993; 63:416–417. 93. Despland M, Clavien PA, Mentha G, et al. Gallstone ileus and bowel perforation after endoscopic sphincterotomy. Am J Gastroenterol 1992; 87:886–888. 94. Rigler LG, Borman CN, Noble JF. Gallstone obstruction. Pathogenesis and roentogen manifestations. JAMA 1941; 117:1753–1759. 95. Day EA, Marks C. Gallstone ileus. Review of the literature and presentation of thirty-four new cases. Am J Surg 1975; 129:552–558. 96. Rodriguez-Sanjuan JC, Casado F, Fernandez MJ, et al. Cholecystectomy and fistula closure versus enterolithotomy alone in gallstone ileus. Br J Surg 1997; 84:634–637. 97. Meyenberger C, Michel C, Metzger U, et al. Gallstone ileus treated by extracorporeal shockwave lithotripsy. Gastrointest Endosc 1996; 43:508–511. 98. Fujita N, Noda Y, Kobayashi G, et al. Gallstone ileus treated by electrohydraulic lithotripsy. Gastrointest Endosc 1992; 38:617–619. 99. Bourke MJ, Schneider DM, Haber GB. Electrohydraulic lithotripsy of a gallstone causing gallstone ileus. Gastrointest Endosc 1997; 45:521–523. 100. Sackmann M, Holl J, Haerlin M, et al. Gallstone ileus successfully treated by shock-wave lithotripsy. Dig Dis Sci 1991; 36:1794–1795. 101. Alsolaiman MM, Reitz C, Nawras AT, et al. Bouveret’s syndrome complicated by distal gallstone ileus after laser lithotropsy using Holmium: YAG laser. BMC Gastroenterol 2002; 2:15. 102. Shenoy VN, Limbekar S, Long PB, et al. Relief of small bowel obstruction following colonoscopy in a case of gallstone ileus. J Clin Gastroenterol 2000; 30:326–328. 103. Roberts SR, Chang C, Chapman T, et al. Colonoscopic removal of a gallstone obstructing the sigmoid colon. J Tenn Med Assoc 1990; 83:18–19. 104. Lopez Roses L, Toscano J, Iniguez F, et al. Successful endoscopic therapy in a case of Bouveret’s syndrome. Rev Esp Enferm Dig 1994; 85:483–485. 105. Lubbers H, Mahlke R, Lankisch PG. Gallstone ileus: endoscopic removal of a gallstone obstructing the upper jejunum. J Intern Med 1999; 246:593–597. 106. Lichten JB, Tehrani K, Sekons D. Laparoscopically assisted enterolithotomy for a gallstone ileus in an atypical location. Surg Endosc 2003; 17:1496–1497. Epub 2003 Jun 19. 107. Hagger R, Sadek S, Singh K. Recurrent small bowel obstruction after laparoscopic surgery for gallstone ileus. Surg Endosc 2003; 17:1679.
108. Ferraina P, Gancedo MC, Elli F, et al. Video-assisted laparoscopic enterolithotomy: new technique in the surgical management of gallstone ileus. Surg Laparosc Endosc Percutan Tech 2003; 13:83–87. 109. Allen JW, McCurry T, Rivas H, et al. Totally laparoscopic management of gallstone ileus. Surg Endosc 2003; 17:352. Epub 2002 Oct 31, 2003. 110. Malvaux P, Degolla R, De Saint-Hubert M, et al. Laparoscopic treatment of a gastric outlet obstruction caused by a gallstone (Bouveret’s syndrome). Surg Endosc 2002; 16:1108–1109. Epub 2002 May 03. 111. Soto DJ, Evan SJ, Kavic MS. Laparoscopic management of gallstone ileus. JSLS 2001; 5:279–285. 112. Sarli L, Pietra N, Costi R, et al. Gallstone ileus: laparoscopicassisted enterolithotomy. J Am Coll Surg 1998; 186:370–371. 113. Franklin ME Jr, Dorman JP, Schuessler WW. Laparoscopic treatment of gallstone ileus: a case report and review of the literature. J Laparoendosc Surg 1994; 4:265–272. 114. Montgomery A. Laparoscope-guided enterolithotomy for gallstone ileus. Surg Laparosc Endosc 1993; 3:310–314. 115. Zuegel N, Hehl A, Lindemann F, et al. Advantages of one-stage repair in case of gallstone ileus. Hepatogastroenterology 1997; 44:59–62. 116. Tan YM, Wong WK, Ooi LL. A comparison of two surgical strategies for the emergency treatment of gallstone ileus. Singapore Med J 2004; 45:69–72. 117. Huang CS, Lein HH, Tai FC, et al. Long-term results of major bile duct injury associated with laparoscopic cholecystectomy. Surg Endosc 2003; 17:1362–1367. Epub 2003 Jun 17, 2003. 118. Doko M, Zovak M, Kopljar M, et al. Comparison of surgical treatments of gallstone ileus: preliminary report. World J Surg 2003; 27:400–404. 119. Robinson TN, Stiegmann GV, Durham JD, et al. Management of major bile duct injury associated with laparoscopic cholecystectomy. Surg Endosc 2001; 15:1381–1385. 120. Syme RG. Management of gallstone ileus. Can J Surg 1989; 32:61–64. 121. Dubois F, Icard P, Berthelot G, et al. Celioscopic cholecystectomy: preliminary report of 36 cases. Ann Surg 1990; 211:60–62. 122. Filipi CJ, Fitzgibbons RJ Jr, Salerno GM. Historical review: diagnostic laparoscopy to laparoscopic cholecystectomy and beyond. In: Zucker KA, ed. Surgical laparoscopy. St Louis, MO: Quality Medical; 1991:3–21. 123. Reddick EJ, Olsen DO. Laparoscopic laser cholecystectomy: a comparison with non-laparoscopic cholecystectomy. Surg Endosc 1989; 3:131–133. 124. Hunter JG. Exposure, dissection, and laser versus electrosurgery in laparoscopic cholecystectomy. Am J Surg 1993; 165:492–496. 125. Flum DR, Dellinger EP, Cheadle A, et al. Intraoperative cholangiography and risk of common bile duct injury during cholecystectomy. JAMA 2003; 289:1639–1644. 126. Vezakis A, Davides D, Ammori BJ, et al. Intraoperative cholangiography during laparoscopic cholecystectomy. Surg Endosc 2000; 14:1118–1122. 127. Olsen D. Bile duct injuries during laparoscopic cholecystectomy. Surg Endosc 1997; 11:133–138. 128. Flum DR, Cheadle A, Prela C, et al. Bile duct injury during cholecystectomy and survival in Medicare beneficiaries. JAMA 2003; 290:2168–2173. 129. Schmidt SC, Settmacher U, Langrehr JM, et al. Management and outcome of patients with combined bile duct and hepatic arterial injuries after laparoscopic cholecystectomy. Surgery 2004; 135:613–618. 130. Jamshidi M, Obermeyer RJ, Garcia G, et al. Post-laparoscopic cholecystectomy bile leak secondary to an accessory duct of Luschka. Int Surg 1999; 84:86–88.
1215
Section XI. Diseases of the Biliary Tract
131. Slater K, Strong RW, Wall DR, et al. Iatrogenic bile duct injury: the scourge of laparoscopic cholecystectomy. Aust NZ J Surg 2002; 72:83–88. 132. Albasini JL, Aledo VS, Dexter SP, et al. Bile leakage following laparoscopic cholecystectomy. Surg Endosc 1995; 9:1274–1278. 133. Csendes A, Navarrete C, Burdiles P, et al. Treatment of common bile duct injuries during laparoscopic cholecystectomy: endoscopic and surgical management. World J Surg 2001; 25:1346–1351. 134. Way LW, Stewart L, Gantert W, et al. Causes and prevention of laparoscopic bile duct injuries: analysis of 252 cases from a human factors and cognitive psychology perspective. Ann Surg 2003; 237:460–469. 135. Wudel LJ Jr, Wright JK, Pinson CW, et al. Bile duct injury following laparoscopic cholecystectomy: a cause for continued concern. Am Surg 2001; 67:557–563; discussion 563–564. 136. Stewart L, Way LW. Bile duct injuries during laparoscopic cholecystectomy. Factors that influence the results of treatment. Arch Surg 1995; 130:1123–1128; discussion 1129. 137. Strasberg SM, Hertl M, Soper NJ. An analysis of the problem of biliary injury during laparoscopic cholecystectomy. J Am Coll Surg 1995; 180:101–125. 138. Brugge WR, Rosenberg DJ, Alavi A. Diagnosis of postoperative bile leaks. Am J Gastroenterol 1994; 89:2178–2183. 139. Bujanda L, Calvo MM, Cabriada JL, et al. MRCP in the diagnosis of iatrogenic bile duct injury. NMR Biomed 2003; 16:475–478. 140. Yeh TS, Jan YY, Tseng JH, et al. Value of magnetic resonance cholangiopancreatography in demonstrating major bile duct injuries following laparoscopic cholecystectomy. Br J Surg 1999; 86:181–184. 141. De Palma GD, Iuliano GP, Puzziello A, et al. Biliary leaks after laparoscopic cholecystectomy. Results of the endoscopic treatment. Minerva Chir 2002; 57:123–127. 142. Mercado MA, Chan C, Orozco H, et al. Acute bile duct injury. The need for a high repair. Surg Endosc 2003; 17:1351–1355. Epub 2003 Jun 19. 143. Chaudhary A, Chandra A, Negi SS, et al. Reoperative surgery for postcholecystectomy bile duct injuries. Dig Surg 2002; 19:22–27. 144. Johnson SR, Koehler A, Pennington LK, et al. Long-term results of surgical repair of bile duct injuries following laparoscopic cholecystectomy. Surgery 2000; 128:668–677. 145. Sefr R, Ochmann J, Kozumplik L, et al. The role of relaparoscopy in the management of bile leaks after laparoscopic cholecystectomy. Int Surg 1995; 80:356–357. 146. Brooks DC, Becker JM, Connors PJ, et al. Management of bile leaks following laparoscopic cholecystectomy. Surg Endosc 1993; 7:292–295. 147. Woods MS, Shellito JL, Santoscoy GS, et al. Cystic duct leaks in laparoscopic cholecystectomy. Am J Surg 1994; 168:560–563; discussion 563–565. 148. Himal HS. The role of ERCP in laparoscopic cholecystectomyrelated cystic duct stump leaks. Surg Endosc 1996; 10:653–655. 149. Barton JR, Russell RC, Hatfield AR. Management of bile leaks after laparoscopic cholecystectomy. Br J Surg 1995; 82:980–984. 150. Sharif K, de Ville de Goyet J. Bile duct of Luschka leading to bile leak after cholecystectomy – revisiting the biliary anatomy. J Pediatr Surg 2003; 38:E21–E23. 151. McQuillan T, Manolas SG, Hayman JA, et al. Surgical significance of the bile duct of Luschka. Br J Surg 1989; 76:696–698. 152. Aru GM, Davis CR Jr, Elliott NL, et al. Endoscopic retrograde cholangiopancreatography in the treatment of bile leaks and bile duct strictures after laparoscopic cholecystectomy. South Med J 1997; 90:705–708.
1216
153. Mergener K, Strobel JC, Suhocki P, et al. The role of ERCP in diagnosis and management of accessory bile duct leaks after cholecystectomy. Gastrointest Endosc 1999; 50:527–531. 154. Familiari L, Scaffidi M, Familiari P, et al. An endoscopic approach to the management of surgical bile duct injuries: nine years’ experience. Dig Liver Dis 2003; 35:493–497. 155. Bismuth H. Postoperative strictures of the bile ducts. In: Blumgart LH, ed. The biliary tract, clinical surgery international, vol. 5. Edinburgh: Churchill Livingstone; 1982:209–218. 156. Lillemoe KD, Melton GB, Cameron JL, et al. Postoperative bile duct strictures: management and outcome in the 1990s. Ann Surg 2000; 232:430–441. 157. Murr MM, Gigot JF, Nagorney DM, et al. Long-term results of biliary reconstruction after laparoscopic bile duct injuries. Arch Surg 1999; 134:604–609; discussion 609–610. 158. Bismuth H, Majno PE. Biliary strictures: classification based on the principles of surgical treatment. World J Surg 2001; 25:1241–1244. 159. Hepp J, Couinaud C. L’abord et l’utilisation du canal hépatique gauche dans les reparations de la vuie biliaire principale. Presse Med 1956; 64:947. 160. Morelli J, Mulcahy HE, Willner IR, et al. Long-term outcomes for patients with post-liver transplant anastomotic biliary strictures treated by endoscopic stent placement. Gastrointest Endosc 2003; 58:374–379. 161. Bonnel DH, Liguory CL, Lefebvre JF, et al. Placement of metallic stents for treatment of postoperative biliary strictures: long-term outcome in 25 patients. AJR Am J Roentgenol 1997; 169:1517–1522. 162. Hunter JG, Swanstrom L, Thornburg K. Carbon dioxide pneumoperitoneum induces fetal acidosis in a pregnant ewe model. Surg Endosc 1995; 9:272–277. 163. Pitt HA, Kaufman SL, Coleman J, et al. Benign postoperative biliary strictures. Operate or dilate? Ann Surg 1989; 210:417–425; discussion 426–427. 164. Misra S, Melton GB, Geschwind JF, et al. Percutaneous management of bile duct strictures and injuries associated with laparoscopic cholecystectomy: a decade of experience. J Am Coll Surg 2004; 198:218–226. 165. Proceedings of the NIH Consensus Development Conference on gallstones and laparoscopic cholecystectomy. Bethesda, Maryland, September 14–16, 1992. Am J Surg 1993; 165:387–548. 166. Puggioni A, Wong LL. A metaanalysis of laparoscopic cholecystectomy in patients with cirrhosis. J Am Coll Surg 2003; 197:921–926. 167. Ramin KD, Ramsey PS. Disease of the gallbladder and pancreas in pregnancy. Obstet Gynecol Clin North Am 2001; 28:571–580. 168. Scott LD. Gallstone disease and pancreatitis in pregnancy. Gastroenterol Clin North Am 1992; 21:803–815. 169. Braverman DZ, Johnson ML, Kern FJ. Effects of pregnancy and contraceptive steroids on gallbladder function. N Engl J Med 1980; 302:362–364. 170. Everson GT. Gastrointestinal motility in pregnancy. Gastroenterol Clin North Am 1992; 21:751–776. 171. Imamoglu K, Wangensteen SL, Root HP. Production of gallstones by prolonged administration of progesterone and estradiol in rabbits. Surg Forum 1959; 10:246. 172. Valdivieso V, Covarrubias C, Siegel F, et al. Pregnancy and cholelithiasis: pathogenesis and natural course of gallstones diagnosed in early puerperium. Hepatology 1993; 17:1–4. 173. Tsimoyiannis EC, Antoniou NC, Tsaboulas C, et al. Cholelithiasis during pregnancy and lactation. Prospective study. Eur J Surg 1994; 160:627–631. 174. Maringhini A, Ciambra M, Baccelliere P, et al. Biliary sludge and gallstones in pregnancy: incidence, risk factors, and natural history. Ann Intern Med 1993; 119:116–120.
Chapter 65 SURGERY OF THE BILIARY TRACT
175. Swisher SG, Schmit PJ, Hunt KK, et al. Biliary disease during pregnancy. Am J Surg 1994; 168:576–579. 176. Muench J, Albrink M, Serafini F, et al. Delay in treatment of biliary disease during pregnancy increases morbidity and can be avoided with safe laparoscopic cholecystectomy. Am Surg 2001; 67:539–542. 177. Weber AM, Bloom GP, Allan TR, et al. Laparoscopic cholecystectomy during pregnancy. Obstet Gynecol 1991; 78:958–959. 178. Steinbrook RA, Bhavani-Shankar K. Hemodynamics during laparoscopic surgery in pregnancy. Anesth Analg 2001; 93:1570–1571.
179. SAGES guidelines for laparoscopic surgery during pregnancy, 2000. 180. Cosenza CA, Saffari B, Jabbour N, et al. Surgical management of biliary gallstone disease during pregnancy. Am J Surg 1999; 178:545–548. 181. Barone JE, Bears S, Chen S, et al. Outcome study of cholecystectomy during pregnancy. Am J Surg 1999; 177:232–236.
1217
Section XII: Inherited and Pediatric Diseases of the Liver
66
WILSON DISEASE Eve A. Roberts and Diane W. Cox Abbreviations ALT alanine aminotransferases AST aspartate aminotransferase ATPases adenosine triphosphatases CHO Chinese hamster ovary
DHPLC DMT1 SNPs
denaturing high-performance liquid chromatography divalent cation transporter single nucleotide polymorphisms
INTRODUCTION COPPER DISORDERS Copper is essential for biological processes in plants, bacteria, yeast, and complex organisms. Copper can exist in two redox states (Cu+1 or Cu+2). This characteristic makes it a versatile cofactor for numerous enzymes but also permits it to participate in reactions which produce activated oxygen species. Although mammals absolutely require copper for normal cellular functions, they utilize trace amounts which are far lower than what is absorbed from the diet. Since excess copper is highly toxic in tissues, disposition of copper is tightly controlled. Copper-associated diseases occur when this control is disturbed. There are two human diseases of copper transport due to defective metal-transporting P-type adenosine triphosphatases (ATPases). Menkes disease is an X-linked defect in transport of copper from the intestine which results in copper deprivation in many organs. Wilson disease, an autosomal recessive disorder, leads to copper overload in the liver and other organs. Two other human diseases characterized by hepatic copper overload have been described: (1) Indian childhood cirrhosis; and (2) Tyrolean cirrhosis, but their possible genetic basis and disease mechanisms have not been determined and may not be identical. No human counterpart to the Bedlington terrier hereditary hepatic copper toxicosis associated with mutations in the MURR1 gene has yet been identified. Exogenous copper intoxication rarely occurs and hepatic accumulation of copper is a well-known consequence of severe chronic cholestasis, but neither of these conditions will be discussed further.
WILSON DISEASE: HISTORY Kinnear Wilson, an American neurologist working in England, first described Wilson disease in 1912 as a familial disorder characterized by progressive, lethal neurological disease along with chronic liver disease1 and a corneal change, known as the Kayser–Fleischer ring, an abnormal eye finding reported approximately 10 years earlier. Although numerous observations over the next few decades pointed to copper overload as being important in Wilson disease, the etiological role of copper was not firmly established until 1948. The importance of low concentrations of plasma ceruloplasmin, the major serum protein containing copper, was recognized in 1952.2 By the mid-1950s, the main diagnostic criteria for Wilson disease were clear, and in 1956 oral chelation therapy with penicillamine,3 a sulfhydryl-containing metabolite of penicillin, was first used. This
SOD1 SSCP XIAP
superoxide dismutase single-strand conformation polymorphism X-linked inhibitor of apoptosis
treatment radically changed the outlook for the disease because patients could be restored to good health in many, if not most, cases. Although Wilson disease could be treated successfully, knowledge of its pathobiology remained incomplete. A defect in biliary excretion of copper appeared to be the most likely mechanism of disease, but this remained unproven. An autosomal recessive pattern of inheritance was confirmed in 1960.4 In 1985 the gene was localized to chromosome 13. In 1993, the abnormal gene in Wilson disease was identified.5–7 The gene ATP7B, but sometimes still carrying the conventional name of WND, encodes a metal-transporting P-type ATPase ATP7B, the Wilson ATPase, which has six copper-binding motifs. The amino-terminus of the Wilson ATPase, including all the copper-binding units, was expressed in 1997, and the copperbinding action of this domain was characterized.8 The structure of the Wilson ATPase was demonstrated by homology modeling in 2002.9 Since the abnormal gene in Wilson disease was identified more than 12 years ago, approximately 300 mutations have been described (see www.uofa-medical-genetics.org/wilson/index.php or www.uofa-medical-genetics.org/wilson/index.php).
COPPER PATHWAY The average diet includes ample amounts of copper, and ordinary daily intake of copper ranges from 1 to 10 mg/day, usually around 2–5 mg/day depending on the mix of meat, legumes, shellfish, and chocolate. The recommended daily intake is 0.9 mg/day.10 Most (85%) of the dietary copper is excreted, and only 15% is retained in body tissues (Figure 66-1). The predominant route for excretion of dietary copper is hepatobiliary. There is no enterohepatic recirculation of copper. Typically the kidneys account for <5% of copper excretion, unless renal tubular reabsorption capacity is exceeded, as occurs in Wilson disease. Thus, in the normal individual close regulation of excretion of copper in bile is critically important for whole-body copper homeostasis. Dietary copper, as well as copper found in bile, salivary and gastric secretions, and pancreatic juice, is absorbed in the small intestine, mainly in the duodenum and proximal jejunum. Absorption is probably via human CTR1 expressed on enterocytes, although the divalent cation transporter (DMT1), which is mainly involved in iron uptake, may play a small role.11 Once absorbed, copper binds reversibly to serum albumin and to various amino acids, of which histidine is the most important. Copper albumin and copper histidine distribute copper to various tissues, mainly to the liver. Copper loosely bound to amino acids is filtered in the kidneys and
1221
Section XII. Inherited and Pediatric Diseases of the Liver
Oral intake (1.5–4 mg/day)
A
Other tissues, proteins: Brain Connective tissue Muscle Enzymes
Plasma albumin/ histidine Rapid clearance
Figure 66-1. Simplified overview of the pathways for copper ion transport, with excretion predominantly via bile. MT, metallothionein (copper storage) represents copper. Two sites of transport disorders are shown: A, Menkes disease (ATP7A gene); B, Wilson disease (ATP7B gene).
Protein synthesis Apoceruloplasmin B (Holo)ceruloplasmin (ferroxidase) MT (storage) ferritin Fe2+
Kidney urinary excretion, reabsorption
transferriin Fe3+ iron mobilization
fecal excretion (1–4) mg/day)
reabsorbed in the tubules. Trace amounts of copper are required for the action of essential enzymes such as lysyl oxidase involved in connective tissue production and elastin cross-linking, Cu/Zn superoxide dismutase (SOD1), a cytoplasmic free-radical scavenger, cytochrome c oxidase integral to mitochondrial electron transfer, tyrosinase required for pigment production, dopamine b-monooxygenase involved in neurotransmission, and peptidyl a-aminating mono-oxygenase which plays a role in processing neurotransmitters. Molecular copper does not exist freely within cells. In hepatocytes and other cells it is always bound to low-molecular-weight proteins, called metallochaperones, which each deliver copper to different specific target molecules within the cell for use in cellular processes or further protein synthesis. In hepatocytes, copper is incorporated into the ferroxidase ceruloplasmin, a 132-kDa molecular weight a2-glycoprotein containing six molecules of copper. As a ferroxidase, ceruloplasmin oxidizes iron for transport from ferritin to transferrin. Ceruloplasmin lacking copper (apoceruloplasmin) is devoid of ferroxidase activity; its half-life within the plasma compartment is comparatively brief, of the order of 24 h. When copper is inserted into apoceruloplasmin, the resulting protein is ceruloplasmin (technically known as holoceruloplasmin), which is enzymatically active. The Wilson ATPase appears to be essential for synthesizing ceruloplasmin. Approximately 95% of the copper in plasma is within ceruloplasmin, and therefore this copper is not exchangeable. Both enterocytes and hepatocytes contain metallothioneins, a class of low-molecular-weight cysteine-rich proteins,
1222
which are inducible and can sequester copper in a non-toxic form. Copper can also form complexes with intracellular glutathione.
EPIDEMIOLOGY INCIDENCE AND GEOGRAPHIC DISTRIBUTION Wilson disease is found worldwide. Wilson disease occurs on average in 30 individuals per million (i.e., 1 per 30 000) population, with a corresponding carrier frequency of approximately 1 in 90. The frequency is higher in some populations, such as Sardinians and Chinese, and is relatively infrequent in populations of African origin. Mass screening of newborn infants might reveal a higher incidence of disease.
PATHOGENESIS BASIC DEFECT Cloning the genes for the two known human metal-transporting P-type ATPases provided the first key insights into the basic cellular defects in Wilson disease. ATP7A, the abnormal gene in Menkes disease, an X-linked disorder, was cloned by using a chromosomal breakpoint in an affected female. It was found to be related to bacterial copper resistance genes. The abnormal gene in Wilson disease (ATP7B) is found on chromosome 13. It was cloned by a combination of conventional linkage analysis,12 physical mapping of the
Chapter 66 WILSON DISEASE
relevant region of chromosome 13q14, and finally by capitalizing on its extensive homology with the Menkes disease gene.5–7 The coding region of ATP7B is 4.1 kilobases in length, with mRNA of approximately 8 kilobases. It comprises 21 exons; its 5¢-untranslated region has also been defined. All functionally important regions of ATP7B are conserved as compared with bacteria and yeasts. Although ATP7A is expressed in many tissues, including those of muscle, kidney, heart, and intestine, ATP7B has a different, and to some extent complementary, pattern of tissue expression. ATP7B is expressed predominantly in the liver and kidneys, with minor expression in brain, lungs, and placenta. The product, known as the Wilson or copper-transporting ATPase, is an intracellular membrane-spanning P-type ATPase, consisting of 1411 amino acid residues, with a molecular mass ~165 kDa. P-type ATPases are functionally diverse, but all have a cation channel and phosphorylation domains with a highly conserved aspartate-containing motif, DKTGT, in which the aspartate residue is transiently phosphorylated during the transport cycle. The Wilson ATPase retains this feature and also a highly characteristic CPC motif in a transmembrane segment. The Wilson ATPase is notable for having six tandem-arranged Cu-binding domains each consisting of the motif MXCXXC, in addition to eight transmembrane segments which form a pore and an ATPase loop region. In the absence of crystallographic data, its structure has recently been determined by homology mapping using SERCA1, the sarcoplasmic Ca2+ P-type ATPase, as a model.9,13 The Wilson ATPase N-terminal Cu-binding region binds copper with the stoichiometry of one copper molecule per metal-binding domain through a cooperative mechanism.8 Bound copper is in the +1 oxidation state and is coordinated by two cysteines in a distorted linear geometry, and the N-terminal region undergoes secondary and tertiary conformational changes in response to Cu-binding.14 Conformational changes may influence
the function of the Wilson ATPase depending on copper concentration, in which case the N-terminal region would exert a regulatory role.
COPPER TRANSPORT AND HOMEOSTASIS IN HEPATOCYTES Copper loosely bound to albumin or histidine is available for uptake into hepatocytes across the sinusoidal plasma membrane (Figure 662). Since copper associated with these transporters is in the cupric form (+2 valence state), it must be reduced to the cuprous form (+1 valence state) prior to hepatocellular uptake. After the available copper is reduced by a reductase presumed to be on the outer surface of the hepatocyte membrane or possibly by dietary reductants, it is taken up into the hepatocytes via transmembrane transporter, CTR1, a member of the solute ligand carrier superfamily, encoded by the gene SLC31A1. Human CTR1 is a membranespanning protein that appears to exist as a trimer to form a channel in the hepatocellular plasma membrane. Somewhat analogous to the Wilson ATPase, human CTR1 has a copper-binding domain near the amino-terminus, but these domains consist of a different motif: a methionine cluster (MXXM), as opposed to a cysteine cluster (CXXC). Copper uptake may be linked to potassium transport. Although CTR1 is not the regulatory control point for copper homeostasis, it appears to be degraded in the presence of high concentrations of copper.15,16 A second copper transporter, known as human CTR2, mediates low-affinity copper uptake.17 The role of DMT1 in hepatocellular copper uptake, apart from that mediated by CTR1, is controversial. Once inside the hepatocyte, small proteins called metallochaperones or copper chaperones coordinate the movement of copper to specific sites in the cell.18 CCS1 guides copper to SOD1. Cox17 supplies copper to cytochrome c oxidase in mitochondria.19 Sco1
Secretion into plasma
(Holo)ceruloplasmin
Nucleus Apoceruloplasmin
metallothionein (copper storage) ATP7B
Bile canaliculus
ATOX1 Cu2+ Cu1+
CTR1
CCS
Cox17
Cu/Zn superoxide dismutase
TGN
ATP7B
Mitochondria SCO1
MURR1?
Cytochrome oxidase
Figure 66-2. Proposed model of intracellular pathways of copper trafficking within the hepatocyte, showing major proteins involved. Low-molecular-weight copper “chaperones” (ATOX1, SCO1, Cox17, and CCS) each deliver copper to a specific target protein (ATP7B, cytochrome oxidase, and superoxide dismutase, respectively). SCO1 transports across the mitochondrial membrane. ATP7B (shown as channel) traffics from the trans-Golgi network (TGN) to cytoplasmic vesicles that deliver copper to the bile canaliculus. MURR1 is proposed to be involved in excretion of copper from the cell.
1223
Section XII. Inherited and Pediatric Diseases of the Liver
and Sco2 mediate the transfer of copper to subunit II of cytochrome c oxidase, and other metallochaperones might be involved.20 The chaperone ATOX1 transports copper to the Wilson ATPase, located in the trans-Golgi network region.21,22 ATOX1 has a single MXCXXC copper-binding unit; it interacts directly with the Wilson ATPase to transfer copper, although not all details of this interaction have been established.21,22 The intracellular action of the Wilson ATPase is twofold: it has a role both in incorporating copper into ceruloplasmin and in facilitating excretion of copper into the bile. In-vitro studies using various continuous cell lines have shown that the intracellular location of the Wilson ATPase changes with increased intracellular Cu concentration. When intracellular Cu concentration is elevated, the Wilson ATPase redistributes from the trans-Golgi network to the vesicle in the apical region of the hepatocyte, that is, to the vicinity of the bile canalicular membrane.23–25 The sixth copper-binding unit appears to be essential for normal intracellular trafficking.26 Lysosomal function relating biliary excretion in the Long-Evans cinnamon (Wilsonian) rat is normal. Many details of the cellular mechanism of biliary excretion of copper remain unknown. Recent discoveries regarding the inherited copper toxicosis in Bedlington terriers have identified a possible new component of the copper transport system. This hepatic copper toxicosis is recessively inherited. Clinical expression is variable, with death at 2–3 years of age, or a long chronic course. Through genetic studies, the abnormal gene in this disorder was identified in a region equivalent to human chromosome 2, and not on chromosome 13 where the Wilson disease gene is located. The causative gene for this condition was proposed to be MURR1, in which one of the three coding exons was found to be completely deleted in affected dogs.27 However, in at least two studies, there are affected dogs lacking the deletion.28,29 This raises the question of whether MURR1 is truly the defective gene responsible for canine copper toxicosis. In addition to interacting with the copper-binding region of ATP7B,30 it affects the activity of the human delta epithelial sodium channel,31 interacts with the X-linked inhibitor of apoptosis (XIAP)32 and inhibits nuclear factor NFkB, a transcription factor affecting immune responses and cell cycle regulation. The latter effect of MURR1, however, is not through binding with NFkB, but by affecting association of NFkB with chromatin in the nucleus. Since the effect upon cellular copper distribution is modest, MURR1 cannot yet be regarded as taking a proven major role in copper transport. A family of related genes has now been identified, named “COMMD genes,” containing the “copper metabolism gene MURR1” domain.33 Some aspects of cellular metal transport are not specific. Of interest, platinum, including organic platinum compounds used as chemotherapy for neoplasia, appears to utilize all of the hepatocellular machinery evolved for the disposition of copper.34 This is not the case for zinc, which has its own family of transporters.35,36
CONSEQUENCES OF COPPER STORAGE Since copper is a pro-oxidant, liver damage in Wilson disease is attributed to oxidative stress.37,38 Accumulated copper is probably a source of activated oxygen species (such as superoxide O2∑, hydrogen peroxide, and hydroxyl radicals) through a Fenton-type reaction.39 Generation of peroxides and hydroxyl radicals leads to nuclear DNA damage.40,41 Studies using primary cultures of rat
1224
hepatocytes showed that copper generates more activated oxygen species and causes more lipid peroxidation than cadmium.42 In the human hepatoma line Hep G2, high copper concentrations in the medium were cytotoxic and impaired cell proliferation.43 Incubation of Hep G2 cells with increasing concentrations of Cu led to generation of activated oxygen species in a dose-responsive pattern, whereas incubation with zinc did not have this effect.44 Copper toxicity to the brain appears to involve oxidative stress. The amyloid precursor protein may regulate copper action in the brain,45 but it is not clear if it functions as a metallochaperone. Increased concentrations of copper in the milieu surrounding neurons or altered activities of key copper-containing enzymes in the brain may contribute to neuronal damage and possibly account for the selectivity of damage. Some clinical data support a role for oxidative stress in the disease mechanism. Patients with Wilson disease may develop focal deletions in mitochondrial DNA typical of aging.37,46 Subnormal levels of antioxidants such as vitamin E are found in untreated Wilson disease, suggesting overutilization.47,48 Oxyradical damage may also be reflected in the increased mutations in the p53 tumor suppressor gene and increased nitric oxide synthase reported in the liver of patients with Wilson disease.49
CLINICAL FEATURES CLINICAL DIAGNOSTIC FEATURES The clinical presentation of Wilson disease is extremely variable.50 Age at onset of symptoms is usually from 6 to about 45 years. Patients presenting outside these age limits continue to be reported and pose important diagnostic challenges. Wilson disease with hepatic involvement has been identified in patients less than 5 years old,51 and in patients over 50.52–54 Wilson disease may present clinically as chronic or fulminant liver disease, as a progressive neurological disorder without evident hepatic dysfunction, or as rather nondescript psychiatric illness. Some patients have one or more episodes of isolated, self-limited acute hemolysis. This degree of clinical variability makes confirming the diagnosis often very difficult. A diagnostic algorithm and clinical classification have been proposed but require further validation.55
Hepatic Presentation Wilson disease presents as liver disease more commonly in children, but it should be considered as the cause of any acute or chronic liver disease in adults. Wilson disease must always be considered as a possible diagnosis in any child, whether symptomatic or not, who has hepatomegaly, persistently elevated serum aminotransferases, or evidence of fatty liver. Symptoms may be vague and non-specific: fatigue, anorexia, or abdominal pain. Occasionally patients of any age present with a self-limited clinical illness resembling acute hepatitis, with malaise, anorexia and nausea, jaundice, elevated serum aminotransferases, and abnormal coagulation tests. Some patients have a history of self-limited jaundice, apparently caused by unexplained hemolysis. Patients with hepatic Wilson disease may present with severe, established chronic liver disease as shown by hepatosplenomegaly, ascites, congestive splenomegaly, low serum albumin, and persistently abnormal coagulation. Many of these
Chapter 66 WILSON DISEASE
findings relate more to portal hypertension as a consequence of Wilson disease than to the metabolic disorder itself. A few patients have been encountered with isolated splenomegaly, without hepatomegaly or indeed any evidence of hepatic dysfunction. Presentation with splenic rupture has been reported.56 Wilson disease may present in children and young adults with clinical liver disease that looks exactly like autoimmune hepatitis.57–59 As with actual autoimmune hepatitis, patients frequently present with acute onset. In Wilson disease mimicking autoimmune hepatitis, fatigue, malaise, arthropathy, and rashes may occur; laboratory findings include elevated aminotransferases, greatly increased serum immunoglobulin G (IgG) concentration and positive non-specific autoantibodies such as antinuclear antibody and anti-smooth muscle (anti-actin) antibody. Wilson disease must be specifically investigated since the treatment of the two diseases is entirely different. With appropriate treatment, the long-term outlook for patients with Wilson disease manifesting as autoimmune hepatitis appears to be favorable even if cirrhosis is present. Recurrent bouts of hemolysis may lead to cholelithiasis with bilirubinate stones. Cirrhosis, if present, may be a further predisposing factor. Children with unexplained cholelithiasis, particularly with small bilirubinate stones, should be tested for Wilson disease. Hepatocellular carcinoma is considered rare in Wilson disease, compared with other chronic liver diseases, but recent reports suggest it is not as rare as generally thought.60,61
Neurological Presentation Neurological Wilson disease tends to occur in the second and third decades or later, but it has been reported in children as young as 6–10 years old. If Kayser–Fleischer rings and prominent cupruria are not present or if the neuropsychiatric syndrome is either atypical or non-specific, diagnosis can be difficult.62,63 Most, but not all, patients with the neurological presentation have hepatic involvement; however, the hepatic disease may be entirely asymptomatic. Approximately 40% of patients with neurological presentation of Wilson disease may have cirrhosis.64 Neurological Wilson disease follows two main patterns: (1) increased or abnormal movement disorder which may be characterized by tremor or dystonia; or (2) a relative loss of movement, which resembles parkinsonian rigidity.65,66 Movement disorders tend to occur earlier and include tremors, poor coordination, and loss of fine motor control. The earliest symptoms may be subtle and attributed to other factors; for example, familial tremor has been implicated erroneously in some patients. Dystonia involves sustained focal movement disorder; patients may also have a facial grimace. The disorders with reduced movement and rigidity generally develop later. These strongly resemble a parkinsonian phenotype with mask-like facies, gait disturbance, and pseudobulbar involvement such as dysarthria, drooling, and swallowing difficulty. These features suggesting pseudobulbar involvement may, however, occur in any patient with neurological Wilson disease. Various speech disorders may occur with any of these neurological presentations: speech abnormalities include slurred or scanned speech, garbled speech described as difficulty getting the words out, and hypophonia, a soft whispery voice. Intellect is not impaired. In patients who have predominantly hepatic disease, neurological involvement is often subtle. Direct questioning may be required to determine whether the patient has significant mood disturbance,
recent deterioration in academic or occupational performance, clumsiness, and disorganized or cramped handwriting.
Psychiatric Presentation As many as 20% of patients may present with purely psychiatric symptoms.67–69 These are highly variable, although depression is common. Neurotic behavior such as phobias or compulsive behaviors have been reported; aggressive or antisocial behavior may occur. Psychosis may occur.
Ocular Features The classic Kayser–Fleischer ring is due to copper deposition in Desçemet’s membrane (Figure 66-3). Although copper is distributed throughout the cornea, fluid streaming favors accumulation of copper here, especially at the superior and inferior poles, and eventually circumferentially around the iris. Kayser–Fleischer rings are visible on direct inspection only when the iris is lightly pigmented and copper deposition heavy. A careful slit-lamp examination is mandatory. Kayser–Fleischer rings may be absent in 15–50% of patients with exclusively hepatic involvement, and in presymptomatic patients. They are less commonly found in children. Most patients with a neurological or psychiatric presentation of Wilson disease have Kayser–Fleischer rings; however, approximately 5% do not. Kayser–Fleischer rings are not specific for Wilson disease. They may be found in patients with other types of chronic liver disease, usually with a prominent cholestatic component, such as primary biliary cirrhosis. With effective chelation, Kayser–Fleischer rings disappear: the lateral edges resorb first and the superior and inferior poles last. Disappearance of Kayser–Fleischer rings on treatment should not be interpreted as casting doubt on the original diagnosis of Wilson disease. Copper can be deposited in the lens, leading to the so-called “sunflower cataract” (Figure 66-4) These cataracts do not interfere with vision. They may be found on slit-lamp examination and, like Kayser–Fleischer rings, disappear with effective chelation therapy.
Involvement of Other Systems Wilson disease can be accompanied by various extrahepatic disorders, apart from neurological disease. Episodes of hemolytic anemia can result from sporadic release of copper into the blood. Renal disease, mainly Fanconi syndrome, may be prominent. Abnormalities include microscopic hematuria, aminoaciduria, phosphaturia, and defective acidification of the urine, and nephrolithiasis has been reported.70 Arthritis, affecting mainly the large joints, may occur, due to synovial copper accumulation.71 Other musculoskeletal problems include osteoporosis and osteochondritis dissecans. Vitamin Dresistant rickets may develop as a result of the renal damage. Copper deposition in the heart can lead to cardiomyopathy or cardiac arrhythmias.72,73 Sudden death in Wilson disease has been attributed to cardiac involvement, but this is rare. Copper deposition in skeletal muscle can cause rhabdomyolysis. Pancreatitis, possibly due to copper deposition in the pancreas, may also occur.74 Endocrine disorders can occur. Hypoparathyroidism has been attributed to copper deposition.75 Amenorrhea and testicular problems appear to be due to Wilson disease itself, not as a consequence of cirrhosis. Infertility or repeated spontaneous abortions occur with untreated Wilson disease.76,77
1225
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 66-3. Kayser–Fleischer ring, a goldenbrown deposit seen at the outer rim of the cornea, in Desçemet’s membrane. Kayser–Fleischer rings are present in 50–95% of patients, depending on the type of onset. (Courtesy of the late Professor Dame S. Sherlock, London.)
Figure 66-4. Sunflower cataract, an infrequent ocular manifestation, due to copper deposits in the lens. (Courtesy of I. Sternlieb, New York.)
BIOCHEMICAL DIAGNOSTIC FEATURES The two main disturbances in hepatocellular copper disposition in Wilson disease are absent or decreased incorporation of copper into ceruloplasmin and greatly decreased biliary excretion of copper. Abnormalities in clinical biochemistry revolve around serum ceruloplasmin concentration. The normal serum concentration of ceruloplasmin in adults is 200–600 mg/l.78 In the neonatal period serum ceruloplasmin levels are low (50–260 mg/l) and in the first few years of life levels run in the high end of the normal range (300–500 mg/l), reaching adult serum concentrations by adolescence. Ceruloplasmin is an acute-phase protein, whose serum levels are increased by inflammatory hepatic disease, pregnancy, or estrogen supplements. In Wilson disease serum ceruloplasmin concentration is typically
1226
decreased, and therefore serum copper is also decreased. Serum ceruloplasmin <50 mg/l is a significant diagnostic finding. It is now recognized that many patients with hepatic Wilson disease, up to 30–40% depending on the reported series, have a normal or nearnormal serum ceruloplasmin concentration.52,79,80 This is in part methodological. Immunological methods routinely used in most clinical laboratories measure both apoceruloplasmin and holoceruloplasmin: these methods overestimate the true amount of ceruloplasmin in plasma. Enzymatic assays, which measure the ferroxidase activity of ceruloplasmin, provide a more reliable measure of ceruloplasmin for the diagnosis of Wilson disease patients by measuring ceruloplasmin which is enzymatically active by virtue of containing copper. The enzymatic assay permits a more accurate estimate of non-ceruloplas-
Chapter 66 WILSON DISEASE
min-bound copper,81 and may indicate early copper deficiency when oxidized activity is completely absent.82 In some patients with hepatic Wilson disease, hepatic inflammation may be sufficient to increase serum ceruloplasmin levels to normal range. In very young children the level may appear normal but is actually low for age. Serum ceruloplasmin by itself is not a sufficient diagnostic test for Wilson disease, since protein synthesis may be decreased in other types of severe chronic liver disease. Protein loss due to intestinal malabsorption, nephrotic syndrome, or severe malnutrition can cause a reduction in concentration. Furthermore, subnormal serum ceruloplasmin concentrations are found in at least 10% of heterozygotes for Wilson disease. Near-absence of ceruloplasmin is found in hereditary aceruloplasminemia, a rare autosomal recessive disease due to mutations in the structural gene for ceruloplasmin on chromosome 3. Aceruloplasminemia is associated with neurological, retinal, and pancreatic degeneration, due to iron accumulation in the brain, retina, and pancreas.83,84 Anemia and increased serum ferritin are observed. Excessive hepatic iron storage occurs, as in hereditary hemochromatosis. Not only has hereditary aceruloplasminemia indicated the critical function of ceruloplasmin as a ferroxidase, a targeted disruption of the ceruloplasmin gene in a mouse model has confirmed its key role in transporting iron out of cells.85 Theoretically, Wilson disease patients undergoing prolonged, aggressive chelation therapy could show similar effects if copper deprivation reduces the ferroxidase activity of ceruloplasmin to undetectable levels. This has been found in a few patients after prolonged treatment, and in those with particularly low ceruloplasmin.86,87 Typically the serum copper concentration parallels the serum ceruloplasmin level. With severe or untreated Wilson disease the non-ceruloplasmin-bound copper concentration is elevated. This concentration can be estimated by subtracting the amount of copper associated with ceruloplasmin from the total serum copper. The arithmetic method is to multiply the serum ceruloplasmin concentration (in milligrams per liter) by 3.15 (the amount of copper, in micrograms, per milligram ceruloplasmin) to determine the amount of copper (in micrograms per liter) associated with ceruloplasmin. If total serum copper is reported in micromoles per liter, it must be converted to micrograms per liter by multiplying that value by 63.5, the molecular weight of copper. The ceruloplasmin-bound copper is then subtracted from the total serum copper. In normal individuals the non-ceruloplasmin-bound copper concentration is approximately 50–100 mg/l. In Wilson disease, the concentration is more than 200 mg/l, or even 10 times higher in the presence of fulminant hepatic failure and intravascular hemolysis. The usefulness of this calculation is highly dependent on the accuracy of the copper and ceruloplasmin measurements. It has not been validated as a diagnostic criterion. Studies of urinary copper excretion, preferably with three separate 24-h collections, prove useful for diagnosis. It is critically important to ensure that collection is complete and precautions are taken against contamination with copper in the collection process; in general, problems with urine collection are not sufficient to invalidate the usefulness of testing. “Spot” urine samples are not suitable as a basis for diagnosis. The volume and urine creatinine concentration should be measured to show completeness of the 24-h collection. The conventional cut-off for diagnosing Wilson disease has been >100 mg/24 h (>1.6 mmol/24 h) in symptomatic patients.
Recent studies suggest that this threshold for diagnosis is too high in that basal 24-h urinary copper excretion may be <100 mg/24 h in 16–23% of patients.79,80 Using 40 mg/24 h (equivalent to 0.6 mmol/ 24 h) appears to be a better diagnostic cut-off for Wilson disease.52,88 Reference values for normal urinary copper excretion may vary from laboratory to laboratory. Heterozygotes usually have a normal 24-h urinary copper excretion rate, although this may rise to borderline abnormal in some cases. A provocative test of urinary copper excretion in which penicillamine (500 mg orally every 12 h) is given while a 24-h urinary collection is obtained sometimes provides useful information.89 Although a normal person may excrete as much as 20 times the baseline level after penicillamine administration, a patient with Wilson disease will excrete considerably more. Urinary copper excretion of 25 mmol or more of copper per 24 h is taken as diagnostic of Wilson disease. This provocative test has been suggested to be more reliable than measurement of hepatic tissue content of copper.89 Although well validated for children, this test has not been validated in adults. Some patients with Wilson disease do not reach the 25 mmol threshold, and therefore sensitivity of this test is uncertain. From a practical point of view, it is worthwhile running this test on 3 successive days and making three separate 24-h urine collections during penicillamine administration. Serum uric acid and phosphate concentrations may be low, reflecting renal tubular dysfunction of untreated Wilson disease, but these findings are not specific for Wilson disease or particularly sensitive. Urinalysis may show microscopic hematuria; if possible, aminoaciduria, phosphaturia, and proteinuria should be quantified. Hepatic tissue copper concentration, usually measured by neutron activation analysis or atomic absorption spectrometry, may provide important diagnostic information. Hepatic copper content greater than 250 mg/g dry weight is considered diagnostic of Wilson disease. Conversely, hepatic parenchymal copper concentrations <40 mg/g dry weight are taken as strong evidence against the diagnosis of Wilson disease. A recent study placed the upper limit of normal for hepatic parenchymal copper at 55 mg/g dry weight.90 Liver biopsy samples must be collected without extraneous copper contamination, but disposable liver biopsy needles can be used. In early stages of Wilson disease, when copper is distributed diffusely in the liver cell cytoplasm and therefore not detected by histochemical stains, this measurement may clearly indicate hepatic copper overload. In later stages of hepatic Wilson disease, the measurement of hepatic copper is less reliable because copper is distributed unequally in the liver.91 However, liver biopsy may not be safe in some patients because of coagulopathy or ascites. The use of a transjugular liver biopsy may get around this contraindication to percutaneous biopsy. Some patients with Wilson disease have a hepatic tissue copper concentration intermediate between normal and definitely elevated (between 55 and 250 mg/g dry weight). The 250 mg/g dry weight threshold may be too high, but further data are needed. Alternatively, there may be a sampling problem. An adequate core of the liver biopsy, at least 1 cm in length, should be analyzed.92 Paraffinembedded tissue can be used for retrospective analysis. An elevated hepatic copper concentration is not specific: patients with chronic cholestasis or diseases such as Indian childhood cirrhosis may have elevated hepatic copper levels. Some heterozygotes have similar moderate elevations of liver tissue copper.
1227
Section XII. Inherited and Pediatric Diseases of the Liver
Since impaired production of ceruloplasmin is an important aspect of the hepatocellular pathology of Wilson disease, measuring incorporation of copper into apoceruloplasmin has been used as a diagnostic test. Radioactive isotopes (64Cu, 67Cu) or, preferably, a stable (65Cu) copper isotope93 can be used. Patients with Wilson disease show little or no incorporation of labeled copper into plasma ceruloplasmin, following oral or intravenous administration of copper isotope. However, heterozygotes cannot always be differentiated from presymptomatic homozygous patients. This test is obsolete and rarely used diagnostically at present.
IMAGING STUDIES Sonography of the liver may reveal features associated with fatty infiltration of the liver or advanced chronic liver disease with portal hypertension and splenomegaly. Magnetic resonance imaging of the brain may be informative in the neurological presentation of Wilson disease.94–96
HISTOPATHOLOGY Wilson disease can present a broad range of histological findings on liver biopsy, many of which are non-specific.97,98 In the earliest stages features are non-specific, and include steatosis, focal necrosis, glycogenated nuclei in hepatocytes, and occasional apoptotic bodies. Mallory hyaline has occasionally been reported. As parenchymal damage progresses, possibly by repeated episodes of lobular necrosis, periportal fibrosis develops. Cirrhosis develops late: it is usually macronodular but may be micronodular. Early in the disease, hepatocellular copper is mainly bound to metallothionein and distributed diffusely in the cytoplasm of hepatocytes. Histochemical stains for copper are negative. With disease progression, hepatocellular copper exceeds the storage capacity of available metallothionein and is deposited in lysosomes. Staining techniques for copper or copper-binding protein (such as rubeanic acid or orcein) detect these lysosomal aggregates of copper. Copper is usually distributed throughout the lobule or nodule, but in the cirrhotic liver, some nodules may have no stainable copper. If the clinical presentation resembles autoimmune hepatitis, findings on liver biopsy may also suggest an autoimmune process. Inflammation is sometimes severe, with classic piecemeal necrosis. Additionally, features not typical for autoimmune hepatitis such as Mallory hyaline and hepatocellular copper aggregates may be found. When Wilson disease presents as fulminant hepatic failure, histological findings confirm pre-existing liver disease. Cirrhosis is found but parenchymal copper is mainly in Kupffer cells, rather than in hepatocytes. Changes in hepatocellular mitochondria, identified by electron microscopy, are an important feature in Wilson disease.99 The mitochondria vary in size; dense bodies in mitochondria may be increased in number. The most striking change is dilatation of the tips of the mitochondrial cristae as a result of separation of the inner and outer membranes of the cristae. The intercristal space is widened so that it looks bulbous or, if more extreme, is an irregular cystic shape. This finding, although not absolutely specific, can be helpful diagnostically, even in quite young and minimally affected patients. Not all hepatocytes in a given lobule are affected, and some lobules may be more severely affected than others. The mitochondrial changes
1228
are probably a consequence of oxidative damage from excessive liver copper.
DIAGNOSIS BY MUTATION ANALYSIS Characteristic clinical features, accompanied by typical biochemical parameters, can provide a diagnosis of Wilson disease in some cases. However, gene mutation analysis has highlighted that many of these traditional parameters can often be misleading, due to overlap with other liver diseases or borderline biochemical results. Diagnosis by DNA mutation analysis plays an important role in providing a reliable diagnosis. DNA diagnosis is possible from liver biopsy samples or blood. Mutation analysis is valuable in any cases where the biochemical and clinical features are atypical, in early stages of the disease when biochemistry may be borderline, and in cases of fulminant hepatic failure undergoing liver transplant. In the latter case, DNA analysis will not alter treatment, but is essential for appropriate follow-up of sibs. Even when the original patient is deceased, accurate diagnosis is important for other family members. Almost 300 mutations in the ATP7B gene have been reported from a variety of populations worldwide, and are listed in the HUGO Wilson disease database maintained at the University of Alberta (www.uofa-medical-genetics.org/wilson/index.php), along with all relevant references. Although the gene has 21 exons, with a coding region of 4.1 kilobases, modern technologies, arising particularly from the human genome sequencing project, make DNA diagnosis feasible. High-throughput analytical methods include denaturing high-performance liquid chromatography (DHPLC),100 single-strand conformation polymorphism (SSCP) analysis101,102 adapted to automated methods, and direct sequencing.103 Certain mutations in a limited number of exons are often characteristic of specific populations, and thus feasible methods for mutation analysis can be devised to improve cost-effectiveness. Most patients are compound heterozygotes, carrying two different mutations of ATP7B. This is helpful for DNA diagnosis, as one mutation may be sufficient to confirm the diagnosis, where the patient has typical clinical signs and at least one biochemical assay indicating increased copper storage. The identification of only one mutation is not adequate when copper assays are borderline and clinical features are ambiguous (such as psychiatric features, slight abnormalities in liver function). Mutation identification is very likely to become the diagnostic aid of choice when clinical and biochemical features arouse suspicion of Wilson disease. Specific mutations are typical for certain ethnic origins of patients. Different populations may have a typical spectrum of mutations. The most common mutation, histidine1069glutamine (His1069Gln or H1069Q),6 is present at least in the heterozygous state (i.e., one of the two mutations) in 35–75% of affected patients of European origin, particularly those from Eastern Europe.104 This mutation is not found in Chinese and related populations, where arginine778leucine (Arg778Leu, R778L) is common.101,105 Patients from most populations have a large number of mutations, with none in a particularly high frequency. One of the exceptions is the island of Sardinia, where the majority of patients have a 15 basepair deletion in the promoter region.106 This is the only population in which a promoter mutation has been identified to date.107 In some populations, feasible mutation identification schemes can be developed for rapid mutation assessment for 90% of patients, as in
Chapter 66 WILSON DISEASE
Sardinia108 and eastern Germany.109 Examples of other populations with a characteristic limited spectrum of mutations include those in Iceland,110 Korea,111 Japan,112 and the Canary Islands.88 Of interest, the spectrum of mutations in Brazil is similar to that in the Canary Islands.113 Mutations have been identified throughout the gene and in the promoter region. Most mutations identified to date in ATP7B, as recorded in the mutation database, are missense (57%), small deletions and insertions (28%) nonsense (7%), and splice site (8%) mutations. Large gene deletions, found in approximately 20% of patients with Menkes disease, are apparently rare. The mutation spectrum for ATP7A, abnormal in Menkes disease, is different from that of ATP7B.114 Missense mutations tend to lie predominantly within functional domains. Deletions, duplications, and nonsense mutations can be predicted to affect the production of the ATP7B transporter severely. The effects of missense mutations are difficult to predict. Given the large numbers of missense mutations, an important challenge is to ensure that mutations identified within ATP7B are actually responsible for defective function causing Wilson disease. When only one mutation is used to support the diagnosis, that one mutation must be definitely established as disease-causing. Prediction of the disruption of the molecule, conservation of the specific residue between species, and absence in at least 50 controls are among the features examined to attempt to identify which mutations are disease-causing. Changes in size, shape, and hydrophobicity are compared. Examination of mutation location, using a model of ATP7B based on the crystallized calcium transporter SERCA1, can assist in assessing which mutations lie in critical regions of the protein.9,13 Functional assays are useful but have been carried out for very few mutations. The components of the copper transport system are completely conserved between yeast and humans. Therefore a yeast assay is useful to determine whether the mutant Wilson ATPase is capable of transporting copper, a requirement for its normal function.115 Cell culture assays in Chinese hamster ovary (CHO) or hepatoma cell lines can indicate whether ATP7B is distributed normally116 and can traffic within the cell from the trans-Golgi network to cytoplasmic vesicles in response to elevated copper concentrations.117,118 With certain mutations, this trafficking required for biliary excretion of copper does not occur.
DIAGNOSIS OF PRESYMPTOMATIC SIBS Since Wilson disease is a recessive condition, brothers and sisters of a patient have a 25% chance of also being affected and 50% risk of being heterozygotes. Reliable diagnosis is essential for presymptomatic individuals before embarking on a lifetime treatment. Because of the variability of biochemical tests, diagnosis in the presymptomatic stage can be difficult. However, initiation of treatment before tissue damage occurs offers the best outlook for a normal span and quality of life. DNA diagnosis performed properly offers the only secure diagnosis. If both mutations have been identified in the patient, sib diagnosis can be carried out directly by mutation analysis. In cases where the diagnosis of Wilson disease is secure in the proband by clinical and biochemical criteria and DNA analysis has not been undertaken, or where only one mutation has been identified, DNA marker analysis can be used without knowledge of the specific mutations present in the initial patient. Markers are variable regions in the DNA within the gene or in regions closely flanking the gene. The specific markers allow tracking of the disease gene, along with its accompanying markers, from each parental chromosome (Figure 66-5). When using this approach, the flanking markers must be appropriate and informative on both sides of the gene to avoid possible error due to a recombination event. Many single nucleotide polymorphisms (SNPs) have been identified throughout the genome, and these can also be used for marker analysis. The importance of DNA analysis for sib diagnosis has been demonstrated by the occurrence of occasional heterozygotes with borderline normal results in copper parameters such as ceruloplasmin, urinary copper, and even liver copper.81,93
11
D13S314
7
8
10
D13S301
7
15
4
6
WND
N
W
N
W
D13S316
7
9
8
6
D13S137
2
3
4
7
11
7
10
SPECIFIC MUTATIONS AND CLINICAL FEATURES
D13S314
8
Because most patients have two different mutations of the ATP7B gene, correlation of clinical features with specific mutations is difficult. Data from a number of laboratories indicate that there is not a high correlation between clinical disease (phenotype) and specific mutation (genotype). However, generally the more severe mutations which interfere with production of intact Wilson ATPase result in an earlier age of onset and tend to be associated with a hepatic presentation of disease.51,119 For the most common His1069Gln mutation, age of onset in homozygous patients ranges from approximately 10 to 50 years of age (mean age: approximately 20 yearsold), somewhat more frequently in those with neurological onset.120 In addition to variations due to specific mutations in the patient, disease severity and clinical features may be influenced by other modifying factors.
D13S301
15
6
7
4
WND
W
W
N
N
D13S316
9
6
7
8
D13S137
3
7
2
4
Affected (Proband)
Normal
11
11
8
7
6
15
6
N
W
W
W
7
6
9
6
2
7
3
7
7
Heterozygote Affected
Figure 66-5. Polymorphic DNA microsatellite markers reliably diagnose the status of sibs of a confirmed patient. One or preferably two informative markers should flank the gene on each side. DNA markers are listed in centromeric to telomeric order. Numbers represent alleles of each marker listed. The proband (arrow) and presymptomatic sib, confirmed as affected, are shown as a filled circle or square. Markers indicate the genotype of each sib. The specific chromosome segment can be followed through the family. Mutation identification is not necessary when the patient has a firm diagnosis of Wilson disease.
1229
Section XII. Inherited and Pediatric Diseases of the Liver
Table 66-1. Treatment of Wilson Disease Drug D-Penicillamine
Trientine
Zinc
a
Dose
Assessment of efficacy
Monitoring of side effects
Initial: 1–1.5 g/day (adult) or 20 mg/kg per day (child), divided into two or three equal oral doses Maintenance: 0.75–1 g/day (in two doses) orally to maintain cupruresis Initial: 1–1.5 g/day (adult) or approximately 20 mg/kg per day (child), divided into two or three equal doses orally Maintenance: same Initial: 50 mg elemental zinc/day (adult), or approximately 25 mg elemental zinc/day (child 5–12 years), three times/day orally, away from meals if possible Maintenance: Titrate dose as required to achieve indicators of efficiency
24-h urinary copper: 200–500 mg (3–8 mmol)/day as target
Complete blood count; urinalysis; skin examination
24-h urinary copper: 200–500 mg (3–8 mmol)/day as target
Complete blood count; urinalysis; serum iron and iron-binding capacity
24-h urinary copper: <75 mg (1.2 mmol)/day as targetb
Serum creatinine, urinalysis; serum zinc (also to monitor compliance)
a
Plus pyridoxine 25 mg daily by mouth. 24-h urinary copper excretion reflects total body copper load and can be used to monitor zinc treatment even though zinc does not cause cupruresis; alternatively, estimated serum nonceruloplasmin-bound copper <100 mg/l can be used. b
When sib diagnosis by haplotype is carried out without mutation analysis, the initial patient must have an unequivocal diagnosis of Wilson disease. This analysis assumes that there is no other similar disease of copper storage. This is the case, given our present knowledge. Of considerable importance is whether a defect in MURR1 could cause copper storage in humans. No mutations of MURR1 were identified in a series of patients with early-onset childhood cirrhosis or in 24 patients with some features of copper storage.121 In another study of 63 patients diagnosed with WND, a polymorphic codon change, with no change of amino acid, was suggested to influence age of onset.122 There is currently no rigorous evidence that mutations in MURR1 cause a human hereditary copper toxicosis. In the absence of marker analysis, or until these results are available, screening should include physical examination, liver function tests, serum copper and ceruloplasmin, 24-h urinary copper measurement, and a careful slit-lamp examination of the eyes. Children 6 years-old or under who appear to be unaffected should be rechecked at annual intervals over the next 5–10 years. Genetic screening, with the use of flanking markers, or by direct mutation analysis, remains the most reliable way of identifying affected sibs when the patient’s DNA is available. For deceased patients, tissue from autopsy or biopsy can be used. Thus far, no known heterozygotes have developed clinical Wilson disease. Liver tissue from a heterozygote can be used for livingdonor liver transplantation. Confirmation of genotype is highly recommended before treatment is initiated because a heterozygote should not be unnecessarily consigned to lifelong treatment with the inherent risk of adverse side effects. On the other hand, some mutations may be functionally more severe than others. Reassessing heterozygotes for hepatic copper accumulation at approximately 50 years of age may have merit. Tests should include liver function tests, serum copper and ceruloplasmin, basal 24-h urinary copper excretion, and a liver sonogram. If total body copper overload is detected, treatment with zinc might be considered, although no clinical studies are currently available to determine whether treatment is ever warranted. Heterozygotes should be counseled to
1230
maintain good liver health by avoiding truncal obesity and excessive use of alcohol.
TREATMENT Medical treatment for Wilson disease involves either chelation or induction of metallothioneins (Table 66-1). There are two generally accepted oral chelating agents: penicillamine and trientine. The potent chelator tetrathiomolybdate is relatively new and remains an experimental treatment modality. Zinc interferes with copper uptake from the intestinal contents and stabilizes hepatic copper by inducing metallothioneins. Based on extensive clinical experience with chronic chelator treatment, it is evident that most patients live normal, healthy lives with effective treatment. Early institution of treatment is critical to the overall success of treatment. The outcome is best for patients who begin treatment when their disease is diagnosed before the onset of symptoms (“presymptomatic”). The role of adjunctive treatments such as antioxidants is formally unproven. The potential utility of gene transfer therapy is currently undetermined.
CHELATION Penicillamine, introduced in 1956 by Walshe, has been the first-line treatment for Wilson disease for decades, and is effective in most patients. Penicillamine is the sulfhydryl-containing amino acid cysteine substituted with two methyl groups (b,b-dimethylcysteine). It can be administered orally, and is rapidly absorbed with a bioavailability in the order of 50%. Penicillamine greatly increases urinary excretion of copper. Studies in the Long-Evans cinnamon rat model of Wilson disease indicate that penicillamine inhibits the accumulation of copper in hepatocellular lysosomes, and, once accumulated, solubilizes copper for mobilization from these particles, but not from cytoplasmic metallothionein.123 The action of penicillamine involves reductive chelation in which copper bound to proteins as Cu(II) is reduced to Cu(I), thus diminishing the affinity of the
Chapter 66 WILSON DISEASE
protein for the copper and permitting chelation to penicillamine.124 Copper is incompletely removed from the liver by penicillamine. In addition to its chelating action, penicillamine inhibits collagen crosslinking and acts as an immunosuppressant. Penicillamine, although effective, can have serious adverse side effects. Some patients develop a febrile reaction with rash and proteinuria within the first 7–10 days after beginning treatment. Although penicillamine can be restarted slowly, along with corticosteroids, changing to an alternate chelator may be safer and has become the customary management. The neurological status of patients presenting with neurological Wilson disease may deteriorate initially after penicillamine treatment is started; this appears to occur in 30–40% of such patients, although a few series suggest the incidence is higher.125 Structural changes in the central nervous system have been reported.126 Most, but certainly not all patients, recover despite continued use of penicillamine. Recent evidence supports discontinuing penicillamine and substituting an alternate treatment, usually zinc.127 Use of penicillamine as a first-line drug for treatment of neurological Wilson disease is currently being re-evaluated. A host of other side effects can occur at any time during chronic treatment. Some side effects are minor (loss of taste, gastrointestinal upset, and arthralgias), whereas others are severe (proteinuria, leukopenia, or thrombocytopenia). Aplastic anemia occurs rarely and does not always reverse when penicillamine is stopped. Nephrotic syndrome, Goodpasture syndrome, myasthenia syndrome, and a systemic disease resembling lupus erythematosus have all been reported. All these severe side effects require immediate discontinuation of penicillamine and use of a different chelator. Up to 30% of patients with Wilson disease develop a severe enough side effect of penicillamine that necessitates a change of treatment.128 Hepatotoxicity has been suspected.129 Adverse reactions involving the skin include an assortment of rashes, but more importantly, pemphigus and elastosis perforans serpiginosa. D-penicillamine is intrinsically less toxic than L-penicillamine and this is why the racemic mixture is not used. Some chronic toxicities such as optic neuritis may be due to pyridoxine insufficiency, and accordingly pyridoxine supplementation is routine with penicillamine use. The overall safety of lifelong treatment with penicillamine is unknown. Quality of life in patients with Wilson disease may be compromised by drug toxicity. Patients who have taken penicillamine for 30–40 years may have chronic skin changes with loss of elastic tissue.130,131 Whether the antifibrotic effect weakens vascular connective tissues is not known. Anecdotal observations suggest that damage to collagen may accrue over decades in patients maintained indefinitely on penicillamine, but risk has not been adequately assessed. Complete absence of enzymatically active ceruloplasmin suggests that chronic depletion of copper could occur.132 Trientine, or triethylene tetramine dihydrochloride (2,2,2tetramine, official short name: trien), introduced by Walshe in the early 1980s,133 is the usual second-line treatment for patients intolerant of penicillamine. A major logistic problem with its clinical use is that it remains an orphan drug, neither universally available nor absolutely assured to stay in pharmaceutical production indefinitely. Trientine differs chemically from penicillamine: as a polyamine chelator, it has a different structure and lacks sulfhydryl groups.
Copper is chelated by forming a stable complex with the four constituent nitrogens in a planar ring. Trientine increases urinary copper excretion and may interfere with intestinal absorption of copper.132 Trientine is a less potent chelator than penicillamine, but this difference is not clinically important. Trientine is relatively non-toxic, apart from causing occasional gastritis and inducing iron deficiency, apparently by chelating dietary iron. Bone marrow suppression has been reported rarely. Most importantly, adverse effects of penicillamine resolve and do not recur when patients intolerant of penicillamine are converted to treatment with trientine.134 Neurological worsening after beginning treatment with trientine has rarely been reported. Trientine appears to be an attractive first-line treatment, but sufficient clinical data are lacking to permit this recommendation as a routine. It is well tolerated in adults135 and in children.136 Its use in pregnancy is based on a favorable, but extremely limited, experience. Ammonium tetrathiomolybdate may be especially suitable for the treatment of severe neurological Wilson disease because, unlike penicillamine, it is not associated with early neurological deterioration.137 Tetrathiomolybdate interferes with copper absorption from the intestine and binds to plasma copper with high affinity. Studies in Long-Evans cinnamon rats indicate that, unlike penicillamine, tetrathiomolybdate removes copper from metallothionein at low doses; at higher doses, an insoluble copper complex is deposited in the liver.138 Although tetrathiomolybdate is regarded as non-toxic, bone marrow suppression is an important and potentially serious adverse side effect, although it may be due to copper deficiency.139 Serum aminotransferase elevations may occur with treatment. Little is known about where mobilized copper, and molybdate, might be deposited. Cerebral deposition of molybdate is associated with an organic brain syndrome. Dose and length of treatment, as well as long-term side effects, must still be determined, but short-term use for acute neurological presentation may be the role for this drug. Systemic copper deficiency might be associated with prolonged use of this potent copper-binding drug. Tetrathiomolybdate also has antiangiogenic action.140
INDUCTION OF METALLOTHIONEINS AND INTERFERENCE WITH ABSORPTION Zinc has been used in Europe since the late 1960s141 and extensively developed as a treatment modality in North America in the past 20 years. Its mechanism of action differs from that of chelators. In pharmacological doses, zinc interferes with the absorption of copper from the gastrointestinal tract and increases copper excretion in the stools. The principal mechanism of action is through its induction of metallothionein in enterocytes. This metallothionein has a greater affinity for copper than for zinc and preferentially binds copper from the intestinal contents. Once bound, the copper is not absorbed but is lost in the feces as enterocytes are shed in normal turnover.142 Since some non-ceruloplasmin-bound copper is secreted into the intestinal contents through gastric and pancreatic secretions and in bile, chronic treatment can lead to an overall decrement in total body copper stores. Although copper induces hepatic metallothionein, it is possible that zinc also has this effect since hepatic parenchymal copper concentration does not decrease during chronic treatment with zinc.143
1231
Section XII. Inherited and Pediatric Diseases of the Liver
Zinc treatment appears to be effective, with few adverse side effects.144 In a randomized controlled trial comparing penicillamine and zinc sulfate, both were equally effective in stabilizing the clinical disease, but adverse side effects were more numerous with penicillamine. However, 2 patients deteriorated when zinc was used as primary therapy. Gastritis is the most common side effect with zinc. Using salts other than sulfate may minimize gastritis, but any zinc salt is equally acceptable for treatment of Wilson disease. Food interferes with effectiveness, and some authors recommend taking no food for 1 h before or after taking the zinc dose. This tends to increase the severity of gastritis and may be sufficiently inconvenient to compromise compliance, for example, in adolescents. An alternative approach is to be less strict about taking zinc away from mealtimes but to titrate the dose against the estimated serum nonceruloplasmin-bound copper concentration. Rare patients experience deterioration in hepatic Wilson disease when started on zinc. Zinc may have immunosuppressant effects and reduce leukocyte chemotaxis. Studies in rats suggest possible interference with bone formation. Zinc excess is associated with increased serum cholesterol, but in doses used for Wilson disease, the adverse effects on blood lipids appear to be very modest.145 Zinc interferes with the absorption of quinolone antibiotics.146 The longterm effectiveness and adverse side effects of zinc require further investigation, but present data indicate that zinc is effective as maintenance therapy and has low toxicity.147 Combination of zinc with a conventional chelator (penicillamine or trientine) has recently become a popular treatment strategy despite a lack of laboratory studies to provide a rationale or systematic clinical evidence as validation. If zinc and a chelator are combined, the two types of treatment must be temporally dispersed through the day, with at least 4–5 h between administrations of either drug, or else they may neutralize each other. In general the routine is to alternate the drugs through the day at 6-hourly intervals such that zinc 50 mg elemental is given as the first and third doses, and penicillamine or trientine (250 or 500 mg) given as the second and fourth doses. This intensive treatment may thus be best suited to very severe hepatic or neurological disease, indeed where it has mainly been used.148,149 Limited clinical data favor the combination of trientine with zinc, but further data are required before this strategy can be recommended confidently. Prompt diagnosis and early institution of chelator treatment in the patient with severe hepatic decompensation but without encephalopathy may be critical.51,150
dietary management can be designed, the main foods to be avoided are organ meats, shellfish, nuts, chocolate, and mushrooms. In the North American diet, the main problem is eliminating chocolate. Vegetarians require specific dietary counseling. If there is reason to believe that the drinking water is high in copper, the water should be analyzed and a copper-removing device may be installed in the plumbing system. Dietary management alone is not sufficient to establish control of the disease.
LIVER TRANSPLANTATION The role of liver transplantation in Wilson disease is limited, but it can be life-saving. Fulminant hepatic failure in Wilson disease necessitates liver transplantation for recovery. Some patients with severe liver disease that is unresponsive to drug therapy may also proceed to early transplantation. The potential for rescue by combination therapy with temporally dispersed trientine plus zinc, used along with vitamin E and other antioxidants, has not been well explored. It is not established that liver transplantation improves severe neurological disease. Favorable151 and unfavorable152,153 results have been reported, but experience is limited and subject to subtle bias. On balance, this intervention cannot be recommended, especially as its rationale is not apparent. Therefore liver transplantation should be reserved for patients who present with severe, decompensated liver disease unresponsive to therapy, or for those with fulminant failure. Living-related transplants, where the graft may be from a heterozygote, have been found to function adequately, albeit with minor defects in copper disposition, as encountered in heterozygotes.154,155 Survival after liver transplantation for Wilson disease is highly favorable, on the order of 70–88% or more.156–159 Renal failure which often accompanies the fulminant liver failure presentation of Wilson disease resolves in the post-transplant period. Kayser–Fleischer rings resorb after liver transplantation.
GENE THERAPY Liver cell transplantation has shown promising results in various animal models.160–162 The practical utility of gene transfer therapy requires further investigation, although good outcomes have been reported in experimental models.163–165 With further refinements these could be effective curative strategies in the future.
SPECIAL PROBLEMS PREGNANCY
ANTIOXIDANTS Antioxidants such as a-tocopherol may be important adjunctive treatment for Wilson disease, either to prevent or reverse progressive liver damage. Patients have been reported to have low serum levels of vitamin E in the presence of high concentrations of nonceruloplasmin-bound copper.47 Anecdotal evidence favors their use in patients with severe hepatic decompensation; however, rigorous clinical data relating to this strategy are not available. Likewise the role of N-acetylcysteine as treatment for severe hepatic Wilson disease has not yet been investigated systematically.
DIETARY MANAGEMENT Most patients should eliminate copper-rich foods from their diet, especially during the first year of treatment. Although detailed
1232
Treatment must be continued throughout pregnancy. There is a risk of postpartum hepatic decompensation if treatment is stopped altogether.166 Many successful pregnancies have been carried out during penicillamine treatment.167–169 Nevertheless, penicillamine is regarded as carrying 5% risk of fetotoxicity.170,171 Severe collagen defects reported occasionally in the offspring could be in part due to copper deficiency from prolonged aggressive treatment, as well as from teratogenic effects of penicillamine.172 Zinc treatment may provide less opportunity for adverse effects on developing collagen in the fetus.173,174 Trientine has also been used as treatment during pregnancy.175,176 Judicious reduction of the dose of penicillamine or trientine, by approximately 25% of the prepregnancy dose, is advisable in the third trimester, especially if delivery by cesarean section is anticipated.50,177 Meticulous attention to diet, iron sufficiency, and
Chapter 66 WILSON DISEASE
folic acid supplementation is important in the pregnant woman with Wilson disease. The absence of enzymatically active ceruloplasmin may be a possible early sign of copper depletion, a problem of uncertain significance for the fetus. Women with undiagnosed and thus untreated Wilson disease may have difficulty conceiving and maintaining pregnancy if they do conceive.169,178 Wilson disease should be part of the differential diagnosis of recurrent spontaneous abortion. Copper accumulates in the placenta in untreated Wilson disease;179 the Wilson ATPase is found in syncytiotrophoblasts in the placenta and plays a role in copper, and possibly iron, homeostasis in the maternal–fetal unit.180 Parous women with Wilson disease require counseling as to contraception since some contraceptive drugs are not tolerated well with chronic liver disease. They should not use copper-containing intrauterine devices. No firm consensus exists as to whether a mother who has Wilson disease should breast-feed her infant. The disease itself does not appear to impose a contraindication. Penicillamine can enter the breast milk and is potentially toxic to the infant. Therefore taking penicillamine, even in relatively low doses, is regarded as a contraindication to breast-feeding. The safety of trientine is not known, and it is not clear whether trientine is secreted into breast milk. Zinc is secreted into breast milk and, although its safety is uncertain, breast-feeding seems to be acceptable. Currently available data are too limited to make any firm recommendations. Caution should be exercised.
FULMINANT LIVER FAILURE Wilson disease may present as fulminant hepatic failure: liver dysfunction with coagulopathy unresponsive to vitamin K supplementation and encephalopathy.181,182 With at least 80 cases reported and additional cases in liver transplantation databases, this presentation of Wilson disease is more common than was initially supposed. Although chronic liver disease (usually cirrhosis) is present, the diagnosis is not initially suspected. Wilsonian fulminant hepatic failure occurs predominantly in females. Most patients are in the 10–30-year-old age bracket but a few very young children have been reported with Wilsonian fulminant liver failure.183 The fulminant liver failure presentation of Wilson disease has clinical and biochemical features which, taken together, distinguish it from acute liver failure associated with acute viral infection or drug hepatotoxicity. Acute intravascular hemolysis occurs; it is Coombsnegative. Renal failure is often present and progresses rapidly. Because the patient has not been suspected of having underlying liver disease, fulminant viral hepatitis is usually the working diagnosis. Unlike fulminant viral hepatitis, fulminant Wilson disease is usually characterized by serum aminotransferases that are disproportionately low (usually <2000 U/l) for the severity of liver failure from the onset of clinically apparent disease. Serum alkaline phosphatase levels are strikingly reduced: in the normal range or even low for age.184,185 Serum bilirubin is often elevated as a result of hemolysis.186 Urinary copper excretion is greatly elevated and the estimated non-ceruloplasmin-bound copper concentration is also very high; however, these results may not be available soon enough to be diagnostically useful. Slit-lamp examination, if it can be performed, may reveal Kayser–Fleischer rings.
Simple biochemical indices are not effective for the diagnosis or prognostication of Wilsonian fulminant hepatic failure. In some patients the aspartate aminotransferase (AST) is higher than the alanine aminotransferases (ALT), suggesting a primary mitochondrial lesion, but this is not a reliable feature of the disorder.157,187,188 The pattern described by the ratio of serum alkaline phosphatase (in IU/l) to serum bilirubin (in mg/dl) may be somewhat more reliable (<2 in Wilsonian liver failure).188 In children with Wilsonian fulminant hepatic failure this ratio, in IU/mmol, was <2 and distinguished Wilson-disease patients from those with fulminant hepatic failure due to other cause.189 An index based on serum AST, total serum bilirubin, and prolongation of prothrombin time identifies cases of Wilsonian fulminant hepatic failure with reasonably good accuracy but fails to prognosticate cases of decompensated chronic hepatic Wilson disease.190 A recent revision of that index comprising total serum bilirubin, white blood count, international normalized ratio, serum albumin, and serum AST showed excellent predictive value where the threshold for bad prognosis was 11.191 Very rarely, patients with Wilson disease present with acute liver failure due to an intercurrent viral hepatitis.192,193 Clinical and biochemical features may then be those typical of viral hepatitis. Concurrent chronic hepatitis C and Wilson disease have also been reported.193 A few patients with Wilsonian fulminant hepatic failure have developed intractable liver disease due to failure to adhere to the treatment regimen. Liver biopsy findings in Wilsonian fulminant hepatic failure usually reveal cirrhosis with intracellular copper detectable in both hepatocytes and Kupffer cells.194 Extensive hepatocellular apoptosis may account for this disease pattern.195 Patients with Wilsonian fulminant hepatic failure require liver transplantation urgently. They do not respond to chelation treatment, although this may be instituted. Plasmapheresis and exchange transfusion196 or hemofiltration,197,198 or hemodialysis can be used to minimize renal injury prior to liver transplantation. Albumin dialysis199 and related techniques, including the molecular adsorbent recycling system,200 may serve as temporizing measures until liver transplantation can be performed. These measures may delay, but do not eliminate, the need for liver transplantation.
DISEASE COMPLICATIONS Patients who stop taking all treatment have a poor prognosis. New neurological abnormalities, such as dysarthria, may develop. Rapidly progressive hepatic decompensation has been observed, occurring on average within 3 years, but reported as early as 8 months after stopping treatment. This reactivated liver damage is usually refractory to reinstitution of chelator therapy. These patients require liver transplantation.
CONSIDERATION OF RISK FOR COPPER DEFICIENCY Deficiencies in trace metals may develop with the use of any chelator, although it is not yet clear whether these are clinically important. Abnormal iron metabolism, leading to hepatic iron overload and anemia, can be predicted if ceruloplasmin oxidase activity is reduced to zero.82
1233
Section XII. Inherited and Pediatric Diseases of the Liver
PROGNOSIS AND NATURAL HISTORY Patients with Wilson disease are generally regarded as having a good prognosis, if the disease is diagnosed promptly and treated consistently. The presymptomatic sib, diagnosed on biochemical or genetic grounds before any sign of clinical impairment, has the best longterm outlook with prompt institution of treatment. The best time to begin treatment in asymptomatic young children has not been established. Treatment of infants and very young children must be carefully evaluated against the risks of copper depletion during a critical period of growth. Patients with early hepatic disease have a generally favorable prognosis as long as treatment is consistent and well tolerated. Severe neurological disease may not resolve entirely on treatment.
CONCLUSION Wilson disease is a comparatively rare genetic disorder of copper disposition affecting mainly the liver and brain. The pattern of inheritance is autosomal recessive. Distribution is worldwide, with an average incidence of 30 affected individuals per million population; the incidence is higher in some more isolated populations. Numerous mutations have been identified since the gene was first cloned in 1993. Compound heterozygotes predominate. This may account in part for the clinical heterogeneity and has certainly made it more difficult to find correlations between genotype and phenotype. One important feature of Wilson disease is that it can be treated effectively. Without treatment, it is either lethal or extremely handicapping. Therefore accurate diagnosis is of the utmost importance. Screening all sibs of a patient is an important preventive measure and is mandatory when Wilson disease is diagnosed. Although there can be problems with adverse side effects from drug treatment, effective drug therapy converts Wilson disease from a lethal disease to one which permits near-normal longevity and lifestyle. Clinical illness frequently improves on treatment, or at least stabilizes. Child-bearing is possible. Liver transplantation plays an important, but limited, role in the management of this disease.
ACKNOWLEDGMENTS The authors’ copper studies are supported by grants from Canadian Institutes of Health Research (MOP 67100 to EAR; MOP 67061 to DWC) and Natural Science and Engineering Research Council of Canada (200762-05 to DWC).
REFERENCES 1. Wilson SAK. Progressive lenticular degeneration: a familial nervous disease associated with cirrhosis of the liver. Brain 1912; 34:295–507. 2. Scheinberg IH, Gitlin D. Deficiency of ceruloplasmin in patients with hepatolenticular degeneration (Wilson’s disease). Science 1952; 116:484–485. 3. Walshe JM. Wilson’s disease. New oral therapy. Lancet 1956; i:25–26.
1234
4. Bearn AG. A genetical analysis of thirty families with Wilson’s disease (hepatolenticular degeneration). Ann Hum Genet 1960; 24:33–43. 5. Bull PC, Thomas GR, Rommens JM, et al. The Wilson disease gene is a putative copper transporting P-type ATPase similar to the Menkes gene. Nat Genet 1993; 5:327–337. 6. Tanzi RE, Petrukhin K, Chernov I, et al. The Wilson disease gene is a copper transporting ATPase with homology to the Menkes disease gene. Nat Genet 1993; 5:344–350. 7. Petrukhin K, Fischer SG, Pirastu M, et al. Mapping, cloning and genetic characterization of the region containing the Wilson disease gene. Nat Genet 1993; 5:338–343. 8. DiDonato M, Narindrasorasak S, Forbes JR, et al. Expression, purification, and metal binding properties of the N-terminal domain from the Wilson disease putative copper-transporting ATPase (ATP7B). J Biol Chem 1997; 272:33279–33282. 9. Fatemi N, Sarkar B. Structural and functional insights of Wilson disease copper-transporting ATPase. J Bioenerg Biomembr 2002; 34:339–349. 10. Trumbo P, Yates AA, Schlicker S, Poos M. Dietary reference intakes: vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. J Am Diet Assoc 2001; 101:294–301. 11. Sharp PA. Ctr1 and its role in body copper homeostasis. Int J Biochem Cell Biol 2003; 35:288–291. 12. Frydman M, Bonne-Tamir B, Farrer LA, et al. Assignment of the gene for Wilson disease to chromosome 13: linkage to the esterase D locus. Proc Natl Acad Sci USA 1985; 82:1819–1821. 13. Efremov RG, Kosinsky YA, Nolde DE, et al. Molecular modelling of the nucleotide-binding domain of Wilson’s disease protein: location of the ATP-binding site, domain dynamics and potential effects of the major disease mutations. Biochem J 2004; 382:293–305. 14. DiDonato M, Hsu HF, Narindrasorasak S, et al. Copper-induced conformational changes in the N-terminal domain of the Wilson disease copper-transporting ATPase. Biochemistry 2000; 39:1890–1896. 15. Klomp AE, Juijn JA, van der Gun LT, et al. The N-terminus of the human copper transporter 1 (hCTR1) is localized extracellularly, and interacts with itself. Biochem J 2003; 370:881–889. 16. Guo Y, Smith K, Lee J, et al. Identification of methionine-rich clusters that regulate copper-stimulated endocytosis of the human Ctr1 copper transporter. J Biol Chem 2004; 279:17428–17433. 17. Lee J, Petris MJ, Thiele DJ. Characterization of mouse embryonic cells deficient in the ctr1 high affinity copper transporter. Identification of a Ctr1-independent copper transport system. J Biol Chem 2002; 277:40253–40259. 18. Field LS, Luk E, Culotta VC. Copper chaperones: personal escorts for metal ions. J Bioenerg Biomembr 2002; 34:373–379. 19. Amaravadi R, Glerum DM, Tzagoloff A. Isolation of a cDNA encoding the human homolog of COX17, a yeast gene essential for mitochondrial copper recruitment. Hum Genet 1997; 99:329–333. 20. Horng YC, Cobine PA, Maxfield AB, et al. Specific copper transfer from the Cox17 metallochaperone to both Sco1 and Cox11 in the assembly of yeast cytochrome C oxidase. J Biol Chem 2004; 279:35334–35340. 21. Larin D, Mekios C, Das K, et al. Characterization of the interaction between the Wilson and Menkes disease proteins and the cytoplasmic copper chaperone, HAH1p. J Biol Chem 1999; 274:28497–28504. 22. Hamza I, Schaefer M, Klomp LW, Gitlin JD. Interaction of the copper chaperone HAH1 with the Wilson disease protein is essential for copper homeostasis. Proc Natl Acad Sci USA 1999; 96:13363–13368. 23. Nagano K, Nakamura K, Urakami KI, et al. Intracellular distribution of the Wilson’s disease gene product (ATPase7B)
Chapter 66 WILSON DISEASE
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36. 37.
38.
39.
40.
41.
42. 43.
44.
45.
after in vitro and in vivo exogenous expression in hepatocytes from the LEC rat, an animal model of Wilson’s disease. Hepatology 1998; 27:799–807. Schaefer M, Roelofsen H, Wolters H, et al. Localization of the Wilson’s disease protein in human liver. Gastroenterology 1999; 117:1380–1385. Harada M, Kumemura H, Sakisaka S, et al. Wilson disease protein ATP7B is localized in the late endosomes in a polarized human hepatocyte cell line. Int J Mol Med 2003; 11:293–298. Cater MA, Forbes J, La Fontaine S, et al. Intracellular trafficking of the human Wilson protein: the role of the six N-terminal metal-binding sites. Biochem J 2004; 380:805–813. van De Sluis B, Rothuizen J, Pearson PL, et al. Identification of a new copper metabolism gene by positional cloning in a purebred dog population. Hum Mol Genet 2002; 11:165–173. Coronado VA, Damaraju D, Kohijoki R, Cox DW. New haplotypes in the Bedlington terrier indicate complexity in copper toxicosis. Mamm Genome 2003; 14:483–491. Hyun C, Lavulo LT, Filippich LJ. Evaluation of haplotypes associated with copper toxicosis in Bedlington terriers in Australia. Am J Vet Res 2004; 65:1573–1579. Tao TY, Liu F, Klomp L, et al. The copper toxicosis gene product MURR1 directly interacts with the Wilson disease protein. J Biol Chem 2003; 278:41593–41596. Biasio W, Chang T, McIntosh CJ, McDonald FJ. Identification of MURR1 as a regulator of the human delta epithelial sodium channel. J Biol Chem 2004; 279:5429–5434. Burstein E, Ganesh L, Dick RD, et al. A novel role for XIAP in copper homeostasis through regulation of MURR1. EMBO J 2004; 23:244–254. Burstein E, Hoberg JE, Wilkinson AS, et al. COMMD proteins: a novel family of structural and functional homologs of MURR1. J Biol Chem 2005;280:22222–22232. Safaei R, Howell SB. Copper transporters regulate the cellular pharmacology and sensitivity to Pt drugs. Crit Rev Oncol Hematol 2005; 53:13–23. Dufner-Beattie J, Langmade SJ, Wang F, et al. Structure, function, and regulation of a subfamily of mouse zinc transporter genes. J Biol Chem 2003; 278:50142–50150. Eide DJ. The SLC39 family of metal ion transporters. Pflugers Arch 2004; 447:796–800. Sokol RJ, Twedt D, McKim JM Jr, et al. Oxidant injury to hepatic mitochondria in patients with Wilson’s disease and Bedlington terriers with copper toxicosis. Gastroenterology 1994; 107:1788–1798. Pena MM, Lee J, Thiele DJ. A delicate balance: homeostatic control of copper uptake and distribution. J Nutr 1999; 129:1251–1260. Deng DX, Ono S, Koropatnick J, Cherian MG. Metallothionein and apoptosis in the toxic milk mutant mouse. Lab Invest 1998; 78:175–183. Oikawa S, Kawanishi S. Site-specific DNA damage induced by NADH in the presence of copper(II): role of active oxygen species. Biochemistry 1996; 35:4584–4590. Yamamoto H, Hirose K, Hayasaki Y, et al. Mechanism of enhanced lipid peroxidation in the liver of Long–Evans cinnamon (LEC) rats. Arch Toxicol 1999; 73:457–464. Pourahmad J, O’Brien PJ. A comparison of hepatocyte cytotoxic mechanisms for Cu2+ and Cd2+. Toxicology 2000; 143:263–273. Aston NS, Watt N, Morton IE, et al. Copper toxicity affects proliferation and viability of human hepatoma cells (HepG2 line). Hum Exp Toxicol 2000; 19:367–376. Seth R, Yang S, Choi S, et al. In vitro assessment of copperinduced toxicity in the human hepatoma line, Hep G2. Toxicol In Vitro 2004; 18:501–509. Barnham KJ, McKinstry WJ, Multhaup G, et al. Structure of the Alzheimer’s disease amyloid precursor protein copper
46.
47.
48.
49.
50. 51.
52.
53.
54.
55. 56.
57.
58. 59.
60.
61. 62.
63. 64. 65.
66.
67.
68. 69.
binding domain. A regulator of neuronal copper homeostasis. J Biol Chem 2003; 278:17401–17407. Mansouri A, Gaou I, Fromenty B, et al. Premature oxidative aging of hepatic mitochondrial DNA in Wilson’s disease. Gastroenterology 1997; 113:599–605. von Herbay A, de Groot H, Hegi U, et al. Low vitamin E content in plasma of patients with alcoholic liver disease, hemochromatosis and Wilson’s disease. J Hepatol 1994; 20:41–46. Ogihara H, Ogihara T, Miki M, et al. Plasma copper and antioxidant status in Wilson’s disease. Pediatr Res 1995; 37:219–226. Hussain SP, Raja K, Amstad PA, et al. Increased p53 mutation load in nontumorous human liver of Wilson disease and hemochromatosis: oxyradical overload diseases. Proc Natl Acad Sci USA 2000; 97:12770–12775. Roberts EA, Schilsky ML. A practice guideline on Wilson disease. Hepatology 2003; 37:1475–1492. Wilson DC, Phillips MJ, Cox DW, Roberts EA. Severe hepatic Wilson’s disease in preschool-aged children. J Pediatr 2000; 137:719–722. Gow PJ, Smallwood RA, Angus PW, et al. Diagnosis of Wilson’s disease: an experience over three decades. Gut 2000; 46:415–419. Danks DM, Metz G, Sewell R, Prewett EJ. Wilson’s disease in adults with cirrhosis but no neurological abnormalities. Br Med J 1990; 301:331–332. Ala A, Borjigin J, Rochwarger A, Schilsky M. Wilson disease in septuagenarian siblings: raising the bar for diagnosis. Hepatology 2005; 41:668–670. Ferenci P, Caca K, Loudianos G, et al. Diagnosis and phenotypic classification of Wilson disease. Liver Int 2003; 23:139–142. Ahmed A, Feller ER. Rupture of the spleen as the initial manifestation of Wilson’s disease. Am J Gastroenterol 1996; 91:1454–1455. Scott J, Gollan JL, Samourian S, Sherlock S. Wilson’s disease, presenting as chronic active hepatitis. Gastroenterology 1978; 74:645–651. Schilsky ML, Scheinberg IH, Sternlieb I. Prognosis of Wilsonian chronic active hepatitis. Gastroenterology 1991; 100:762–767. Milkiewicz P, Saksena S, Hubscher SG, Elias E. Wilson’s disease with superimposed autoimmune features: report of two cases and review. J Gastroenterol Hepatol 2000; 15:570–574. Walshe JM, Waldenstrom E, Sams V, et al. Abdominal malignancies in patients with Wilson’s disease. Q J Med 2003; 96:657–662. Iwadate H, Ohira H, Suzuki T, et al. Hepatocellular carcinoma associated with Wilson’s disease. Intern Med 2004; 43:1042–1045. Prashanth LK, Taly AB, Sinha S, et al. Wilson’s disease: diagnostic errors and clinical implications. J Neurol Neurosurg Psychiatry 2004; 75:907–909. Carlson MD, Al-Mateen M, Brewer GJ. Atypical childhood Wilson’s disease. Pediatr Neurol 2004; 30:57–60. Ferenci P. Review article: diagnosis and current therapy of Wilson’s disease. Aliment Pharmacol Ther 2004; 19:157–165. Oder W, Prayer L, Grimm G, et al. Wilson’s disease: evidence of subgroups derived from clinical findings and brain lesions. Neurology 1993; 43:120–124. Brewer GJ. Neurologically presenting Wilson’s disease: epidemiology, pathophysiology and treatment. CNS Drugs 2005; 19:185–192. Oder W, Grimm G, Kollegger H, et al. Neurological and neuropsychiatric spectrum of Wilson’s disease: a prospective study of 45 cases. J Neurol 1991; 238:281–287. Dening TR. The neuropsychiatry of Wilson’s disease: a review. Int J Psychiatry Med 1991; 21:135–148. Rathbun JK. Neuropsychological aspects of Wilson’s disease. Int J Neurosci 1996; 85:221–229.
1235
Section XII. Inherited and Pediatric Diseases of the Liver
70. Azizi E, Eshel G, Aladjem M. Hypercalciuria and nephrolithiasis as a presenting sign in Wilson disease. Eur J Pediatr 1989; 148:548–549. 71. Golding DN, Walshe JM. Arthropathy of Wilson’s disease. Study of clinical and radiological features in 32 patients. Ann Rheum Dis 1977; 36:99–111. 72. Factor SM, Cho S, Sternlieb I, et al. The cardiomyopathy of Wilson’s disease. Myocardial alterations in nine cases. Virchows Arch [Pathol Anat] 1982; 397:301–311. 73. Hlubocka Z, Maracek Z, Linhart A, et al. Cardiac involvement in Wilson disease. J Inher Metab Dis 2002; 25:269–277. 74. Weizman Z, Picard E, Barki Y, Moses S. Wilson’s disease associated with pancreatitis. J Pediatr Gastroenterol Nutr 1988; 7:931–933. 75. Carpenter TO, Carnes DL Jr, Anast CS. Hypoparathyroidism in Wilson’s disease. N Engl J Med 1983; 309:873–877. 76. Kaushansky A, Frydman M, Kaufman H, Homburg R. Endocrine studies of the ovulatory disturbances in Wilson’s disease (hepatolenticular degeneration). Fertil Steril 1987; 47:270–273. 77. Tarnacka B, Rodo M, Cichy S, Czlonkowska A. Procreation ability in Wilson’s disease. Acta Neurol Scand 2000; 101:395–398. 78. Gaffney D, Fell GS, O’Reilly DS. ACP best practice no 163. Wilson’s disease: acute and presymptomatic laboratory diagnosis and monitoring. J Clin Pathol 2000; 53:807–812. 79. Steindl P, Ferenci P, Dienes HP, et al. Wilson’s disease in patients presenting with liver disease: a diagnostic challenge. Gastroenterology 1997; 113:212–218. 80. Sanchez-Albisua I, Garde T, Hierro L, et al. A high index of suspicion: the key to an early diagnosis of Wilson’s disease in childhood. J Pediatr Gastroenterol Nutr 1999; 28:186–190. 81. Walshe JM. Wilson’s disease: the importance of measuring serum caeruloplasmin non-immunologically. Ann Clin Biochem 2003; 40:115–121. 82. Macintyre G, Gutfreund KS, Martin WR, et al. Value of an enzymatic assay for the determination of serum ceruloplasmin. J Lab Clin Med 2004; 144:294–301. 83. Harris ZL, Takahashi Y, Miyajima H, et al. Aceruloplasminemia: molecular characterization of this disorder of iron metabolism. Proc Natl Acad Sci USA 1995; 92:2539–2543. 84. Miyajima H. Aceruloplasminemia, an iron metabolic disorder. Neuropathology 2003; 23:345–350. 85. Harris ZL, Durley AP, Man TK, Gitlin JD. Targeted gene disruption reveals an essential role for ceruloplasmin in cellular iron efflux. Proc Natl Acad Sci USA 1999; 96:10812–10817. 86. Luca P, Demelia L, Lecca S, et al. Massive hepatic haemosiderosis in Wilson’s disease. Histopathology 2000; 37:187–189. 87. Shiono Y, Wakusawa S, Hayashi H, et al. Iron accumulation in the liver of male patients with Wilson disease. Am J Gastroenterol 2001; 96:3147–3151. 88. Garcia-Villarreal L, Daniels S, Shaw SH, et al. High prevalence of the very rare Wilson disease gene mutation Leu708Pro in the island of Gran Canaria (Canary Islands, Spain): a genetic and clinical study. Hepatology 2000; 32:1329–1336. 89. Martins da Costa C, Baldwin D, Portmann B, et al. Value of urinary copper excretion after penicillamine challenge in the diagnosis of Wilson’s disease. Hepatology 1992; 15:609–615. 90. Nuttall KL, Palaty J, Lockitch G. Reference limits for copper and iron in liver biopsies. Ann Clin Lab Sci 2003; 33:443–450. 91. Faa G, Nurchi V, Demelia L, et al. Uneven hepatic copper distribution in Wilson’s disease. J Hepatol 1995; 22:303–308. 92. Song YM, Chen MD. A single determination of liver copper concentration may misdiagnose Wilson’s disease. Clin Biochem 2000; 33:589–590. 93. Lyon TD, Fell GS, Gaffney D, et al. Use of a stable copper isotope (65Cu) in the differential diagnosis of Wilson’s disease. Clin Sci (Colch) 1995; 88:727–732.
1236
94. Saatci I, Topcu M, Baltaoglu FF, et al. Cranial MR findings in Wilson’s disease. Acta Radiol 1997; 38:250–258. 95. Alanen A, Komu M, Penttinen M, Leino R. Magnetic resonance imaging and proton MR spectroscopy in Wilson’s disease. Br J Radiol 1999; 72:749–756. 96. Kozic D, Svetel M, Petrovic B, et al. MR imaging of the brain in patients with hepatic form of Wilson’s disease. Eur J Neurol 2003; 10:587–592. 97. Stromeyer FW, Ishak KG. Histology of the liver in Wilson’s disease: a study of 34 cases. Am J Clin Pathol 1980; 73:12–24. 98. Ludwig J, Moyer TP, Rakela J. The liver biopsy diagnosis of Wilson’s disease. Methods in pathology. Am J Clin Pathol 1994; 102:443–446. 99. Sternlieb I. Mitochondrial and fatty changes in hepatocytes of patients with Wilson’s disease. Gastroenterology 1968; 55:354–367. 100. Weirich G, Cabras AD, Serra S, et al. Rapid identification of Wilson’s disease carriers by denaturing high-performance liquid chromatography. Prev Med 2002; 35:278–284. 101. Thomas GR, Forbes JR, Roberts EA, et al. The Wilson disease gene: spectrum of mutations and their consequences. Nat Genet 1995; 9:210–217. 102. Loudianos G, Dessi V, Lovicu M, et al. Mutation analysis in patients of Mediterranean descent with Wilson disease: identification of 19 novel mutations. J Med Genet 1999; 36:833–836. 103. Waldenstrom E, Lagerkvist A, Dahlman T, et al. Efficient detection of mutations in Wilson disease by manifold sequencing. Genomics 1996; 37:303–309. 104. Caca K, Ferenci P, Kuhn HJ, et al. High prevalence of the H1069Q mutation in East German patients with Wilson disease: rapid detection of mutations by limited sequencing and phenotype–genotype analysis. J Hepatol 2001; 35:575–581. 105. Gu YH, Kodama H, Du SL, et al. Mutation spectrum and polymorphisms in ATP7B identified on direct sequencing of all exons in Chinese Han and Hui ethnic patients with Wilson’s disease. Clin Genet 2003; 64:479–484. 106. Loudianos G, Dessi V, Lovicu M, et al. Molecular characterization of Wilson disease in the Sardinian population – evidence of a founder effect. Hum Mutat 1999; 14:294–303. 107. Cullen LM, Prat L, Cox DW. Genetic variation in the promoter and 5’ UTR of the copper transporter, ATP7B, in patients with Wilson disease. Clin Genet 2003; 64:429–432. 108. Lovicu M, Dessi V, Zappu A, et al. Efficient strategy for molecular diagnosis of Wilson disease in the Sardinian population. Clin Chem 2003; 49:496–498. 109. Huster D, Weizenegger M, Kress S, et al. Rapid detection of mutations in Wilson disease gene ATP7B by DNA strip technology. Clin Chem Lab Med 2004; 42:507–510. 110. Thomas GR, Jensson O, Gudmundsson G, et al. Wilson disease in Iceland: a clinical and genetic study. Am J Hum Genet 1995; 56:1140–1146. 111. Kim EK, Yoo OJ, Song KY, et al. Identification of three novel mutations and a high frequency of the Arg778Leu mutation in Korean patients with Wilson disease. Hum Mutat 1998; 11:275–278. 112. Nanji MS, Nguyen VT, Kawasoe JH, et al. Haplotype and mutation analysis in Japanese patients with Wilson disease. Am J Hum Genet 1997; 60:1423–1429. 113. Deguti MM, Genschel J, Cancado EL, et al. Wilson disease: novel mutations in the ATP7B gene and clinical correlation in Brazilian patients. Hum Mutat 2004; 23:398. 114. Hsi G, Cox DW. A comparison of the mutation spectra of Menkes disease and Wilson disease. Hum Genet 2004; 114:165–172. 115. Forbes JR, Cox DW. Functional characterization of missense mutations in ATP7B: Wilson disease mutation or normal variant? Am J Hum Genet 1998; 63:1663–1674.
Chapter 66 WILSON DISEASE
116. Huster D, Hoppert M, Lutsenko S, et al. Defective cellular localization of mutant ATP7B in Wilson’s disease patients and hepatoma cell lines. Gastroenterology 2003; 124:335–345. 117. Payne AS, Kelly EJ, Gitlin JD. Functional expression of the Wilson disease protein reveals mislocalization and impaired copper-dependent trafficking of the common H1069Q mutation. Proc Natl Acad Sci USA 1998; 95:10854–10859. 118. Forbes JR, Cox DW. Copper-dependent trafficking of Wilson disease mutant ATP7B proteins. Hum Mol Genet 2000; 9:1927–1935. 119. Panagiotakaki E, Tzetis M, Manolaki N, et al. Genotype– phenotype correlations for a wide spectrum of mutations in the Wilson disease gene (ATP7B). Am J Med Genet A 2004; 131A:168–173. 120. Stapelbroek JM, Bollen CW, van Amstel JK, et al. The H1069Q mutation in ATP7B is associated with late and neurologic presentation in Wilson disease: results of a meta-analysis. J Hepatol 2004; 41:758–763. 121. Coronado VA, Bonneville JA, Nazer H, et al. COMMDI (MURR1) as a candidate in patients with copper storage disease of undefined etiology. 2005, submitted. 122. Stuehler B, Reichert J, Stremmel W, Schaefer M. Analysis of the human homologue of the canine copper toxicosis gene MURR1 in Wilson disease patients. J Mol Med 2004; 82:629–634. 123. Klein D, Lichtmannegger J, Heinzmann U, Summer KH. Dissolution of copper-rich granules in hepatic lysosomes by Dpenicillamine prevents the development of fulminant hepatitis in Long–Evans cinnamon rats. J Hepatol 2000; 32:193–201. 124. Sarkar B. Treatment of Wilson and Menkes diseases. Chem Rev 1999; 99:2535–2544. 125. Brewer GJ, Terry CA, Aisen AM, Hill GM. Worsening of neurologic syndrome in patients with Wilson’s disease with initial penicillamine therapy. Arch Neurol 1987; 44:490–493. 126. Huang CC, Chu NS. Acute dystonia with thalamic and brainstem lesions after initial penicillamine treatment in Wilson’s disease. Eur Neurol 1998; 39:32–37. 127. Huang CC, Chu NS. Wilson’s disease: resolution of MRI lesions following long-term oral zinc therapy. Acta Neurol Scand 1996; 93:215–218. 128. Walshe JM. Wilson’s disease presenting with features of hepatic dysfunction: a clinical analysis of eighty-seven patients. Q J Med 1989; 70:253–263. 129. Deutscher J, Kiess W, Scheerschmidt G, Willgerodt H. Potential hepatotoxicity of penicillamine treatment in three patients with Wilson’s disease. J Pediatr Gastroenterol Nutr 1999; 29:628. 130. Coatesworth AP, Darnton SJ, Green RM, et al. A case of systemic pseudo-pseudoxanthoma elasticum with diverse symptomatology caused by long-term penicillamine use. J Clin Pathol 1998; 51:169–171. 131. Amichai B, Rotem A, Metzker A. D-penicillamine-induced elastosis perforans serpiginosa and localized cutis laxa in a patient with Wilson’s disease. Isr J Med Sci 1994; 30:667–669. 132. Siegemund R, Lossner J, Gunther K, et al. Mode of action of triethylenetetramine dihydrochloride on copper metabolism in Wilson’s disease. Acta Neurol Scand 1991; 83:364–366. 133. Walshe JM. Treatment of Wilson’s disease with trientine (triethylene tetramine) dihydrochloride. Lancet 1982; 1:643–647. 134. Scheinberg IH, Jaffe ME, Sternlieb I. The use of trientine in preventing the effects of interrupting penicillamine therapy in Wilson’s disease. N Engl J Med 1987; 317:209–213. 135. Dahlman T, Hartvig P, Lofholm M, et al. Long-term treatment of Wilson’s disease with triethylene tetramine dihydrochloride (trientine). Q J Med 1995; 88:609–616. 136. Dubois RS, Rodgerson DO, Hambidge KM. Treatment of Wilson’s disease with triethylene tetramine hydrochloride (Trientine). J Pediatr Gastroenterol Nutr 1990; 10:77–81. 137. Brewer GJ, Hedera P, Kluin KJ, et al. Treatment of Wilson disease with ammonium tetrathiomolybdate: III. Initial therapy
138.
139.
140.
141.
142.
143.
144.
145.
146. 147.
148.
149.
150. 151.
152.
153.
154.
155.
156.
157.
158.
in a total of 55 neurologically affected patients and follow-up with zinc therapy. Arch Neurol 2003; 60:379–385. Ogra Y, Suzuki KT. Targeting of tetrathiomolybdate on the copper accumulating in the liver of LEC rats. J Inorg Biochem 1998; 70:49–55. Karunajeewa H, Wall A, Metz J, Grigg A. Cytopenias secondary to copper depletion complicating ammonium tetrathiomolybdate therapy for Wilson’s disease. Aust NZ J Med 1998; 28:215–216. Brewer GJ. Tetrathiomolybdate anticopper therapy for Wilson’s disease inhibits angiogenesis, fibrosis and inflammation. J Cell Mol Med 2003; 7:11–20. Hoogenraad TU, Van Hattum J, Van den Hamer CJ. Management of Wilson’s disease with zinc sulphate. Experience in a series of 27 patients. J Neurol Sci 1987; 77:137–146. Yuzbasiyan-Gurkan V, Grider A, Nostrant T, et al. Treatment of Wilson’s disease with zinc: X. Intestinal metallothionein induction. J Lab Clin Med 1992; 120:380–386. Lee DY, Brewer GJ, Wang YX. Treatment of Wilson’s disease with zinc. VII. Protection of the liver from copper toxicity by zinc-induced metallothionein in a rat model. J Lab Clin Med 1989; 114:639–645. Brewer GJ, Dick RD, Johnson VD, et al. Treatment of Wilson’s disease with zinc: XV long-term follow-up studies. J Lab Clin Med 1998; 132:264–278. Brewer GJ, Yuzbasiyan-Gurkan V, Johnson V. Treatment of Wilson’s disease with zinc. IX: response of serum lipids. J Lab Clin Med 1991; 118:466–470. Lomaestro BM, Bailie GR. Absorption interactions with fluoroquinolones. Drug Safety 1995; 12:314–333. Anderson LA, Hakojarvi SL, Boudreaux SK. Zinc acetate treatment in Wilson’s disease. Ann Pharmacother 1998; 32:78–87. Askari FK, Greenson J, Dick RD, et al. Treatment of Wilson’s disease with zinc. XVIII. Initial treatment of the hepatic decompensation presentation with trientine and zinc. J Lab Clin Med 2003; 142:385–390. Santos Silva EE, Sarles J, Buts JP, Sokal EM. Successful medical treatment of severely decompensated Wilson disease. J Pediatr 1996; 128:285–287. Durand F, Bernuau J, Giostra E, et al. Wilson’s disease with severe hepatic insufficiency: beneficial effects of early administration of D-penicillamine. Gut 2001; 48:849–852. Stracciari A, Tempestini A, Borghi A, Guarino M. Effect of liver transplantation on neurological manifestations in Wilson disease. Arch Neurol 2000; 57:384–386. Bax RT, Hassler A, Luck W, et al. Cerebral manifestation of Wilson’s disease successfully treated with liver transplantation. Neurology 1998; 51:863–865. Kassam N, Witt N, Kneteman N, Bain VG. Liver transplantation for neuropsychiatric Wilson disease. Can J Gastroenterol 1998; 12:65–68. Asonuma K, Inomata Y, Kasahara M, et al. Living related liver transplantation from heterozygote genetic carriers to children with Wilson’s disease. Pediatr Transplant 1999; 3:201–205. Komatsu H, Fujisawa T, Inui A, et al. Hepatic copper concentration in children undergoing living related liver transplantation due to Wilsonian fulminant hepatic failure. Clin Transplant 2002; 16:227–232. Eghtesad B, Nezakatgoo N, Geraci LC, et al. Liver transplantation for Wilson’s disease: a single-center experience. Liver Transpl Surg 1999; 5:467–474. Emre S, Atillasoy EO, Ozdemir S, et al. Orthotopic liver transplantation for Wilson’s disease: a single-center experience. Transplantation 2001; 72:1232–1236. Sutcliffe RP, Maguire DD, Muiesan P, et al. Liver transplantation for Wilson’s disease: long-term results and quality-of-life assessment. Transplantation 2003; 75:1003–1006.
1237
Section XII. Inherited and Pediatric Diseases of the Liver
159. Wang XH, Cheng F, Zhang F, Li XC, et al. Copper metabolism after living related liver transplantation for Wilson’s disease. World J Gastroenterol 2003; 9:2836–2838. 160. Yoshida Y, Tokusashi Y, Lee GH, Ogawa K. Intrahepatic transplantation of normal hepatocytes prevents Wilson’s disease in Long–Evans cinnamon rats. Gastroenterology 1996; 111:1654–1660. 161. Malhi H, Irani AN, Volenberg I, et al. Early cell transplantation in LEC rats modeling Wilson’s disease eliminates hepatic copper with reversal of liver disease. Gastroenterology 2002; 122:438–447. 162. Allen KJ, Cheah DM, Wright PF, et al. Liver cell transplantation leads to repopulation and functional correction in a mouse model of Wilson’s disease. J Gastroenterol Hepatol 2004; 19:1283–1290. 163. Terada K, Nakako T, Yang XL, et al. Restoration of holoceruloplasmin synthesis in LEC rat after infusion of recombinant adenovirus bearing WND cDNA. J Biol Chem 1998; 273:1815–1820. 164. Ha-Hao D, Merle U, Hofmann C, et al. Chances and shortcomings of adenovirus-mediated ATP7B gene transfer in Wilson disease: proof of principle demonstrated in a pilot study with LEC rats. Z Gastroenterol 2002; 40:209–216. 165. Meng Y, Miyoshi I, Hirabayashi M, et al. Restoration of copper metabolism and rescue of hepatic abnormalities in LEC rats, an animal model of Wilson disease, by expression of human ATP7B gene. Biochim Biophys Acta 2004; 1690:208–219. 166. Shimono N, Ishibashi H, Ikematsu H, et al. Fulminant hepatic failure during perinatal period in a pregnant woman with Wilson’s disease. Gastroenterol Jpn 1991; 26:69–73. 167. Walshe JM. Pregnancy in Wilson’s disease. Q J Med 1977; 46:73–83. 168. Sternlieb I. Wilson’s disease and pregnancy. Hepatology 2000; 31:531–532. 169. Sinha S, Taly AB, Prashanth LK, et al. Successful pregnancies and abortions in symptomatic and asymptomatic Wilson’s disease. J Neurol Sci 2004; 217:37–40. 170. Mjolnerod IK, Rasmussen K, Dommerud SA, Gjeruldsen ST. Congenital connective-tissue defect probably due to Dpenicillamine treatment in pregnancy. Lancet 1971; i:673–675. 171. Rosa FW. Teratogen update: penicillamine. Teratology 1986; 33:127–131. 172. Pinter R, Hogge WA, McPherson E. Infant with severe penicillamine embryopathy born to a woman with Wilson disease. Am J Med Genet A 2004; 128:294–298. 173. Lao TT, Chin RK, Cockram CS, Leung NW. Pregnancy in a woman with Wilson’s disease treated with zinc. Asia Oceania J Obstet Gynaecol 1988; 14:167–169. 174. Brewer GJ, Johnson VD, Dick RD, et al. Treatment of Wilson’s disease with zinc. XVII: treatment during pregnancy. Hepatology 2000; 31:364–370. 175. Walshe JM. The management of pregnancy in Wilson’s disease treated with trientine. Q J Med 1986; 58:81–87. 176. Devesa R, Alvarez A, de las Heras G, Ramon de Miguel J. Wilson’s disease treated with trientine during pregnancy. J Pediatr Gastroenterol Nutr 1995; 20:102–103. 177. Berghella V, Steele D, Spector T, et al. Successful pregnancy in a neurologically impaired woman with Wilson’s disease. Am J Obstet Gynecol 1997; 176:712–714. 178. Schagen van Leeuwen JH, Christiaens GC, Hoogenraad TU. Recurrent abortion and the diagnosis of Wilson disease. Obstet Gynecol 1991; 78:547–549. 179. Oga M, Matsui N, Anai T, et al. Copper disposition of the fetus and placenta in a patient with untreated Wilson’s disease. Am J Obstet Gynecol 1993; 169:196–198. 180. Hardman B, Manuelpillai U, Wallace EM, et al. Expression and localization of Menkes and Wilson copper transporting ATPases in human placenta. Placenta 2004; 25:512–517.
1238
181. Roche-Sicot J, Benhamou JP. Acute intravascular hemolysis and acute liver failure associated as a first manifestation of Wilson’s disease. Ann Intern Med 1977; 86:301–303. 182. McCullough AJ, Fleming CR, Thistle JL, et al. Diagnosis of Wilson’s disease presenting as fulminant hepatic failure. Gastroenterology 1983; 84:161–167. 183. Kalach N, Seidman EG, Morin C, et al. Acute liver failure from Wilson’s disease in a five year-old child. Can J Gastroenterol 1993; 7:610–612. 184. Shaver WA, Bhatt H, Combes B. Low serum alkaline phosphatase activity in Wilson’s disease. Hepatology 1986; 6:859–863. 185. Willson RA, Clayson KJ, Leon S. Unmeasurable serum alkaline phosphatase activity in Wilson’s disease associated with fulminant hepatic failure and hemolysis. Hepatology 1987; 7:613–615. 186. Hoshino T, Kumasaka K, Kawano K, et al. Low serum alkaline phosphatase activity associated with severe Wilson’s disease. Is the breakdown of alkaline phosphatase molecules caused by reactive oxygen species? Clin Chim Acta 1995; 238:91–100. 187. Berman DH, Leventhal RI, Gavaler JS, et al. Clinical differentiation of fulminant Wilsonian hepatitis from other causes of hepatic failure. Gastroenterology 1991; 100:1129–1134. 188. Sallie R, Katsiyiannakis L, Baldwin D, et al. Failure of simple biochemical indexes to reliably differentiate fulminant Wilson’s disease from other causes of fulminant liver failure. Hepatology 1992; 16:1206–1211. 189. Tissieres P, Chevret L, Debray D, Devictor D. Fulminant Wilson’s disease in children: appraisal of a critical diagnosis. Pediatr Crit Care Med 2003; 4:338–343. 190. Nazer H, Ede RJ, Mowat AP, Williams R. Wilson’s disease: clinical presentation and use of prognostic index. Gut 1986; 27:1377–1381. 191. Dhawan A, Taylor RM, Cheeseman P, et al. Wilson’s disease in children: 37-year experience and revised King’s score for liver transplantation. Liver Transpl 2005; 11:441–448. 192. Sallie R, Chiyende J, Tan KC, et al. Fulminant hepatic failure resulting from coexistent Wilson’s disease and hepatitis E. Gut 1994; 35:849–853. 193. Lembowicz K, Kryczka W, Walewska-Zielecka B, Kubicka J. Wilson’s disease coexisting with viral hepatitis type C: a case report with histological and ultrastructural studies of the liver. Ultrastruct Pathol 1999; 23:39–44. 194. Davies SE, Williams R, Portmann B. Hepatic morphology and histochemistry of Wilson’s disease presenting as fulminant hepatic failure: a study of 11 cases. Histopathology 1989; 15:385–394. 195. Strand S, Hofmann WJ, Grambihler A, et al. Hepatic failure and liver cell damage in acute Wilson’s disease involve CD95 (APO-1/Fas) mediated apoptosis. Nat Med 1998; 4:588–593. 196. Valbonesi M, Valente U, Andorno E, et al. Role of intensive PEX in a patient with fulminant hepatic failure due to Wilson’s disease (WD) in preparation for orthotopic liver transplantation (OLT). Int J Artif Organs 2003; 26:965–966. 197. Rakela J, Kurtz SB, McCarthy JT, et al. Fulminant Wilson’s disease treated with postdilution hemofiltration and orthotopic liver transplantation. Gastroenterology 1986; 90:2004–2007. 198. Nagata Y, Uto H, Hasuike S, et al. Bridging use of plasma exchange and continuous hemodiafiltration before living donor liver transplantation in fulminant Wilson’s disease. Intern Med 2003; 42:967–970. 199. Kreymann B, Seige M, Schweigart U, et al. Albumin dialysis: effective removal of copper in a patient with fulminant Wilson disease and successful bridging to liver transplantation: a new possibility for the elimination of protein-bound toxins. J Hepatol 1999; 31:1080–1085. 200. Manz T, Ochs A, Bisse E, et al. Liver support – a task for nephrologists? Extracorporeal treatment of a patient with fulminant Wilson crisis. Blood Purif 2003; 21:232–236.
Section XII: Inherited and Pediatric Diseases of the Liver
67
HEMOCHROMATOSIS Paul C. Adams Abbreviations ALT alanine aminotransferase AST aspartate aminotransferase HFE hemochromatosis gene
HIV HJV HLA
human immunodeficiency virus juvenile hemochromatosis human leukocyte antigen
INTRODUCTION Hemochromatosis is the most common genetic disease in populations of European ancestry. Despite estimates based on genetic testing in Caucasian populations of 1 in 227,1 many physicians consider hemochromatosis to be a rare disease. The diagnosis can be elusive because of the non-specific nature of the symptoms. With the discovery of the HFE gene in 19962 came new insights into the pathogenesis of the disease and new diagnostic strategies. A fundamental issue that has arisen since the discovery of the HFE gene is whether the disease hemochromatosis should be defined strictly on phenotypic criteria such as the degree of iron overload (transferrin saturation, ferritin, liver biopsy, hepatic iron concentration, iron removed by venesection therapy), or whether the condition should be defined as a familial disease in Europeans most commonly associated with the C282Y mutation of the HFE gene and varying degrees of iron overload. It is important to realize that there are many causes of iron overload other than hemochromatosis (Table 67-1) and there are a growing number of new non-HFE genetic mutations that can be associated with iron overload.3
HISTORY OF HEMOCHROMATOSIS In 1865, Trousseau described an association between cirrhosis of the liver, diabetes, pancreatic disease, and pigmentation of the skin. These clinical features are associated with advanced disease and are now uncommonly seen by clinicians. The term hemochromatosis was used by von Recklinghausen in 1889 because he suspected that the increased pigment came from the blood. Many variations in the nomenclature have been suggested (HFE-linked hemochromatosis, hereditary hemochromatosis, iron overload disease). In 1935, Sheldon published his classic monograph on 311 cases. The relative roles of genes and alcoholism were debated for many years but the familial aspects of the disease were clearly elucidated in pedigree studies by Simon in Brittany, France, in 1977, including a close linkage of the disease to the human leukocyte antigen (HLA) complex on chromosome 6. Genetic studies concentrated on this region of the chromosome but informative genetic recombinations were rarely found. The hemochromatosis gene (HFE) was finally discovered a long distance from the HLA complex by Mercator Genetics in 1996.2 This discovery led to the development of a simple genetic blood test which has demonstrated that more than 90% of typical hemochromatosis patients are homozygous for the C282Y mutation of the HFE gene.
MRI SQUID
magnetic resonance imaging superconducting quantum interference device
Most patients with hemochromatosis can trace their ancestry to northern Europe. Because of the high prevalence of hemochromatosis in Brittany, it has been hypothesized that the original couple with hemochromatosis were born in Brittany around 800 AD. Many people from Brittany crossed the English Channel during the Norman Conquest in 1066, and resettled in the UK, and currently residents of Wales and Ireland share traditional language and music with Brittany. Since the Vikings were exploring Brittany extensively during those years, it is possible that some of the original genes were of Nordic ancestry. Further evidence for this hypothesis comes from the high prevalence of the HFE gene in Iceland, which was explored by the Vikings but not by French explorers. However, Irish women may have been taken by the Vikings to Iceland. Furthermore, there has been evidence presented that the gene originated much earlier in mainland Europe before 4000 BC.4 The hemochromatosis gene can be used as a genetic marker to study the migration of Europeans to North and South America, South Africa, and Australia. Most studies have linked HFE-related hemochromatosis to northern Europe but there is also a high prevalence of this condition in Portugal.5
EPIDEMIOLOGY The typical genetic pattern (C282Y homozygote) is found in approximately 1 in 227 Caucasians.1 However, the clinical symptoms of hemochromatosis are much less common and many of the symptoms are non-specific. The regions with the highest prevalence of hemochromatosis are Ireland, northern Portugal, western France, and regions where migration from these countries has occurred (Australia, Canada, USA, South Africa).6 Iron overload can occur from a variety of genetic and environmental factors and the geographical distribution of iron overload is much broader than for HFE-linked hemochromatosis.
PATHOGENESIS OF HEMOCHROMATOSIS In a normal human, most dietary iron is absorbed from the proximal duodenum. Both ionic iron and heme iron are absorbed across the enterocyte at the tip of the intestinal villi. An intestinal heme iron transporter (HCP1) has been identified. In regard to non-heme iron absorption, it is becoming clearer that there are a cascade of
1239
Section XII. Inherited and Pediatric Diseases of the Liver
Table 67-1. Differential Diagnosis of Iron Overload
H63D Mutation
HFE-related hemochromatosis C282Y homozygotes (95%) C282Y/H63D compound heterozygotes (4%) H63D homozygotes (1%)
a Heavy Chain a1
Non-HFE-related hemochromatosis Ferroportin disease Transferrin receptor-2 mutation Juvenile hemochromatosis (young adults with cardiac and endocrine dysfunction)
iron-related proteins that are involved in normal and abnormal iron absorption. This concept is similar to the cascade of events that occurs with blood coagulation or complement activation. In hemochromatosis, there is an increased iron absorption from the intestine and an inappropriate absorption in the presence of total body iron overload. There is a disruption of the regulation of iron absorption and whether this occurs at the level of the intestinal enterocyte or at a more distant site is a subject of current research. Iron proteins at the brush border include DMT1 (divalent metal transport protein 1) and duodenal cytochrome B (Dcytb), a ferrireductase. Iron transport within the enterocyte has not been well defined but a number of iron-related proteins, including transferrin, transferrin receptor, ferritin, iron-regulatory peptide, and hepcidin, have been described. The transfer of iron from the enterocyte into the portal circulation involves another series of transport proteins including hephaestin, and ferroportin (IREG1) (Figure 67-1).7 The HFE gene produces a major histocompatibility complex class 1 protein (Figure 67-2) that is expressed in many cells but has a high concentration in the duodenal crypts. It interacts with transferrin receptor to facilitate iron uptake into cells. In hemochromatosis, patients have a mutated HFE protein, resulting in a conformational change in the protein, and this interaction is impaired. The C282Y mutation has a more pronounced effect on HFE protein function than the H63D mutation. Since the HFE protein is not abundant at the intestinal villus where iron absorption occurs, it has been necessary to implicate other iron proteins in the pathogenesis of hemochromatosis. One theory suggests that duodenal crypt cells develop a relative iron deficiency and when they migrate to the tip of the villi another gene is stimulated (DMT1) which increases iron absorption. However, the data at the present time on HFE protein function and DMT1 expression have not built a conclusive case for the pathogenesis of hemochromatosis.8 Hepcidin is a hepatic peptide that may be the key regulator of iron metabolism. Hepcidin deficiency, as demonstrated in knockout mice, results in severe iron overload.9 The relationship of hepcidin
1240
a2
S
NH2 NH2 b2-microglobulin Exracellular
Miscellaneous iron overload African-American iron overload African iron overload Transfusional iron overload Insulin resistance-related iron overload Aceruloplasminemia Alcoholic siderosis Iron overload secondary to end-stage cirrhosis Porphyria cutanea tarda Post-portacaval shunt
S
S S
S S
COOH
a3 C282Y Mutation
Plasma membrane Intracellular COOH Figure 67-1. A schematic representation of the HFE protein, a class 1 major histocompatibility complex protein that is involved in iron metabolism with a cascade of other iron proteins. Typical hemochromatosis patients are homozygous for the C282Y mutation of the HFE which causes conformational changes in the protein which impairs intracellular trafficking.
to other iron proteins, such as the HFE protein, and transferrin receptor, is under investigation but it appears that hepcidin can down-regulate the release of iron by enterocytes and macrophages. Hepcidin may also have a direct effect on the uptake of iron by intestinal epithelia.10 Conversely, a relative deficiency of hepcidin can result in increased iron efflux of iron from macrophages. This is consistent with the paradoxical observation in HFE-related hemochromatosis of relative iron deficiency in the spleen and within macrophages. Hepcidin expression in liver tissue, serum, and urine has suggested a relative deficiency in HFE-linked hemochromatosis.11,12 However, at the present time it is unclear whether the defect in hepcidin is the primary abnormality in hemochromatosis, or a downstream effect of abnormalities in the HFE protein. Genetic mutations in hepcidin in humans have been documented in severe cases of iron overload which have a clinical diagnosis of juvenile hemochromatosis.13 It is intriguing to speculate whether alterations in non-HFE iron proteins could explain the wide range of clinical expression seen in hemochromatosis. The concept is that the most severe cases of HFE-related hemochromatosis may be heterozygotes for a mutation in another iron-related protein.14 These mutations would need to be very common to fit this hypothesis. A tantalizing theory has been suggested that the hemochromatosis gene has not yet been found and that the HFE gene is only a modifying gene.15
CLINICAL FEATURES OF HEMOCHROMATOSIS A major problem has been the attribution of clinical symptoms in hemochromatosis to iron overload. Earlier studies did not utilize control populations and recent population-screening studies have
Chapter 67 HEMOCHROMATOSIS
Normal
HFE-Related Hemochromatosis Lumen
Liver
Ferritin-iron
Ferritin-iron
Villus enterocyte
Villus enterocyte
Ferroportin
Ferroportin
Hepcidin
Ferritiniron
Lumen
Liver
Hepcidin
Iron Ferroportin
Ferritiniron
Plasma Macrophage
Uncontrolled release of iron from macrophage and duodenal enterocytes Ferroportin
Iron
Plasma
Macrophage
Figure 67-2. A schematic of normal iron metabolism, and iron metabolism in hemochromatosis. (Adapted from Pietrangelo A. Hereditary hemochromatosis: a new look at an old disease. N Engl J Med 2004; 350:2383–2397.)
drawn increasing attention to the non-specific nature of the clinical symptoms such as arthralgia, fatigue, and even diabetes. It has been assumed that the symptoms of hemochromatosis are related to iron overload causing tissue injury. However, iron depletion does not reverse many of the symptoms of hemochromatosis. Of all the putative symptoms of hemochromatosis, liver disease is the most consistent. Although the disease was originally called “bronze diabetes,” several studies have now suggested that the prevalence of diabetes is similar between C282Y homozygotes and a control population.1,16,17 The study of the possible symptoms in C282Y homozygotes without biochemical iron overload may provide more information on the relationship of symptoms to the degree of iron overload. Another unresolved issue is whether early iron depletion in an asymptomatic C282Y homozygote will prevent the development of symptoms.
LIVER DISEASE Although hemochromatosis is often classified as a liver disease, it should be emphasized that it is a systemic genetic disease with multisystem involvement. The liver has a great capacity to accumulate iron within hepatocytes initially without any obvious sequelae both in terms of clinical symptoms and abnormal liver biochemistry. Since hepatic iron presumably accumulates from birth in this genetic disease, you would predict a relationship between iron and age. However, this may only apply serially within an individual patient since the corrleation coefficient between age and hepatic iron concentration was not significant in the 410 homozygotes (r = 0.07, P = 0.12).18 Hepatomegaly remains one of the more common physical signs in hemochromatosis but may not be present in the young asymptomatic homozygote. In older studies in which patients pre-
sented with clinical features of chronic liver disease in the fifth or sixth decade, cirrhosis was invariably present. As patients are detected as young adults through pedigree studies or populationscreening studies, the prevalence of cirrhosis is much lower. In a study of 410 referred homozygotes from Canada and France, 22% had cirrhosis of the liver at the time of diagnosis. The mean aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were within the normal range within these 410 patients. Cirrhotic patients and patients with concomitant alcohol abuse were more likely to have abnormal liver enzymes.18 A clinical presentation with marked elevations in liver enzymes and elevated iron tests should suggest an alternate diagnosis such as alcoholic liver disease, chronic viral hepatitis, or non-alcoholic steatohepatitis. A study of the critical hepatic iron concentration associated with cirrhosis in C282Y homozygotes using receiver operating characteristic curve analysis suggested that the critical iron concentration was >283 mmol/g (normal 0–35 mmol/g).19 However, there were many patients with much higher liver iron concentrations that did not have cirrhosis. Liver damage at lower iron concentrations was usually associated with other risk factors such as alcohol abuse or chronic viral hepatitis. Therefore, it seems likely that there are factors other than iron overload that contribute to cirrhosis in hemochromatosis.20 It should also be emphasized that women can have significant liver involvement in hemochromatosis. In a study of 176 women, matched with males for year of birth, there was no difference in the hepatic iron concentration.21 Many of these studies are subject to referral bias, with only the sickest patients being sent for medical evaluation. Population-screening studies have rarely identified patients with cirrhosis but an increased prevalence of liver disease has been detected in C282Y homozygotes.1,16,22
1241
Section XII. Inherited and Pediatric Diseases of the Liver
The effect of iron depletion therapy has usually been stabilization of the liver disease. This may account for the relatively small number of C282Y homozygotes who have required liver transplantation.23 Reversal of cirrhosis has rarely been described with iron depletion. This has previously been questioned on the basis of sampling errors but evidence is increasing in other liver diseases (treatment of chronic viral hepatitis) that cirrhosis can be reversed. Post-treatment liver biopsies have been uncommonly reported,24 and are not recommended. Hepatocellular carcinoma has been described in 18.5% of cirrhotic patients with hemochromatosis.25 It has rarely been described in non-cirrhotic hemochromatosis patients. The relative risk is approximately 200-fold, which is similar to cirrhotic patients with chronic viral hepatitis. Screening for hepatocellular carcinoma remains a controversial topic because of the costs of the screening program, organ shortages allowing for liver transplantation, and the small number of candidates who are cured by surgical resection. In the case of hemochromatosis patients with cirrhosis, it could be considered, but this is not an evidence-based recommendation.
DIABETES IN HEMOCHROMATOSIS Many patients with cirrhosis of any etiology have glucose intolerance or diabetes and this is true for hemochromatosis as well. The presence of diabetes in hemochromatosis is usually related to the presence of liver disease18 and glucose intolerance or diabetes has been found in 85% of hemochromatosis patients with cirrhosis.24 There has also been a higher prevalence of diabetes found in family members of hemochromatosis patients with diabetes. Earlier morphological studies of iron deposition in the pancreas suggested pancreatic damage as the cause of diabetes. However subsequent studies have demonstrated that high circulating insulin levels (insulin resistance secondary to liver disease) is more common than low circulating insulin levels related to islet cell damage. Metabolic studies of insulin and glucose have not clearly demonstrated a reversal of these changes with iron depletion.26 This is consistent with the clinical observation that diabetes rarely resolves with therapy.26 Insulin resistance in obese patients has also been associated with iron overload in patients without the typical genetic profile for hemochromatosis.27 Most of these cases have a moderate elevation in serum ferritin with a normal transferrin saturation. However, most obese patients with an elevated ferritin have normal liver iron concentrations and likely have a mild elevation in serum ferritin (<1000 mg/l) secondary to steatohepatitis.
CARDIAC DISEASE IN HEMOCHROMATOSIS Cardiac disease in hemochromatosis includes both cardiomyopathy and arrhythmias. In a series of 410 patients, cardiac disease was only present in 10% of probands and 3% of discovered cases.18 Dyspnea is the most common symptom associated with the dilated cardiomyopathy. Echocardiography is the preferred initial diagnostic test and abnormalities can be seen in up to 35% of referred cases.24 The cardiac iron concentration is significantly lower than the liver iron concentration. Transvenous cardiac biopsies have occasionally missed the diagnosis of hemochromatosis and should not be considered to have excluded the diagnosis. There have been uncommon cases of young adults presenting with life-threatening cardiac disease. These cases have been called juvenile hemochromatosis and
1242
the gene for juvenile hemochromatosis (HJV) has been localized to chromosome 1. Some of these cases have been found to have mutations in the hepcidin gene.28 These patients can also have lifethreatening ventricular arrhythmias requiring implantable defibrillators or potentially hepatotoxic medications such as amiodarone. Cardiac dysfunction is more common in secondary iron overload such as thalassemia and is a common cause of death.
ARTHROPATHY OF HEMOCHROMATOSIS Arthralgias are perhaps the most common symptom of hemochromatosis. Joint complaints have been found in 32% of probands and 21% of discovered cases.18 The classical description is in the proximal interphalangeal joints of the hands but wrist, shoulder, knees, and feet are commonly affected. The features of the arthropathy are more similar to osteoarthritis and less commonly chondrocalcinosis. Radiological features are often non-specific but on occasion the diagnosis is suggested by an astute radiologist. Subchondral cyst formation, osteopenia, squaring of the metacarpophalangeal joint heads, and chondrocalcinosis have all been described. Joint complaints are particularly common in women with hemochromatosis. Arthritis has been demonstrated to be the major factor affecting quality of life in hemochromatosis.29 Since arthritis is common, it has been difficult to attribute the arthritis to hemochromatosis in an aging population.
ENDOCRINE ABNORMALITIES The most common symptom of endocrine dysfunction is impotence in men with hemochromatosis. This has been estimated to be present in 40% of cases and is almost universal in male cirrhotic patients. Most cases will have low luteinizing and folliclestimulating hormones and testosterone. Amenorrhea and infertility can be seen in women. Hypothyroidism is a less common endocrine manifestation of hemochromatosis. Hemochromatosis patients are predisposed to osteoporosis. This is likely related to the low testosterone and an independent effect of cirrhosis. Many patients will receive testosterone supplementation without a major effect on their impotence but possibly an improvement in bone density. The effect of androgens potentially increasing the risk of hepatocellular carcinoma in a cirrhotic hemochromatosis patient is another consideration in the clinical management.
PIGMENTATION OF THE SKIN The characteristic pigmentation of hemochromatosis is caused by melanin, not iron. In the late stages of hemochromatosis, iron can be seen in the sweat glands and in the basal layers of the epidermis. The prevalence of pigmentation depends to an extent on the age of the patient and the severity of the iron overload. It was estimated to be present in 47% of proband cases and 18% of discovered cases,18 but, like all signs and symptoms of hemochromatosis, is decreasing with earlier diagnosis.
INFECTIONS Yersinia infections have been described in cases of hemochromatosis and secondary iron overload. Other infections have included Pasteurella pseudotuberculosis, Vibrio vulnificus, and Listeria monocytogenes. It has been implied that the bacterial iron metabo-
Chapter 67 HEMOCHROMATOSIS
lism of these species encourages their growth in a high-iron environment.
depending on age and gender but fatigue is the most common complaint. Women are more likely to have fatigue, arthralgia, and pigmentation rather than liver disease.
FATIGUE Although a non-specific symptom, fatigue is one of the most common reported symptoms of hemochromatosis. The presence of fatigue often leads to biochemical iron studies and the physician expecting a low value is surprised to find iron overload. Since more than half of patients attending a general medical clinic complain of fatigue, this has been one of the most difficult symptoms to attribute to hemochromatosis. Screening projects within chronic fatigue syndrome clinics have not demonstrated an enhanced case detection for hemochromatosis.
DIAGNOSIS OF HEMOCHROMATOSIS A paradox of genetic hemochromatosis is the observation that the disease is underdiagnosed in the general population, and overdiagnosed in patients with secondary iron overload. A case definition that appeals to all experts has been elusive. Minimum criteria for the diagnosis of hemochromatosis are increased iron stores in the absence of a cause for secondary iron overload. Genetic testing for HFE mutations (C282Y and H63D) has been a major advance, with a single mutation explaining most typical cases, but there are a growing number of new genetic mutations in other genes that make a HFE-specific case definition unsatisfactory. The original descriptions of 311 cases of iron overload described by Sheldon in 1935 were likely C282Y-related, and many experts continue to use the presence of homozygosity for the C282Y mutation of the HFE gene as the cornerstone of the diagnosis of hemochromatosis. A broader case definition dependent on a degree of iron overload in the liver or as iron mobilized by phlebotomy allows for the consideration of an expanding number of newer genetic mutations which may result in iron overload. An increasing awareness that iron overload can be a sequela of a wide range of chronic liver diseases has led to many previous cases of “hemochromatosis” being reclassified as iron overload secondary to cirrhosis. From a practical perspective, the clinician should not become too immersed in the debate about whether an individual case has hemochromatosis, since this is largely a semantic debate. The focus should be on identifying treatable causes of iron overload and initiating the appropriate diagnostic tests and therapies in the patient and other affected family members.
UNDERDIAGNOSIS OF HEMOCHROMATOSIS The fact that many physicians consider hemochromatosis to be rare implies either a lack of penetrance of the gene (non-expressing homozygote) or a large number of patients who remain undiagnosed in the community. It is likely that both of these factors are contributory. A major problem in the diagnosis of hemochromatosis is the lack of symptoms and the non-specific nature of symptoms. An elderly patient who presents with joint symptoms and diabetes is not often considered to have genetic hemochromatosis. Many patients assume that the routine blood tests done frequently at ambulatory clinics would have tested their iron status but this is uncommonly done unless iron deficiency is suspected. The presenting features vary
DIAGNOSTIC TESTS FOR HEMOCHROMATOSIS A number of diagnostic algorithms based on laboratory tests have been proposed for the diagnosis of hemochromatosis30 (Figure 673). These should be used as guidelines for the clinician and do not replace clinical judgment based on history and physical examination, imaging studies, pathology, and pedigree studies.
TRANSFERRIN SATURATION The transferrin saturation is often elevated in patients with HFElinked hemochromatosis (Figure 67-4). It is a two-step test that is widely available. It can be calculated using the serum iron and one of the following: total iron-binding capacity, unsaturated ironbinding capacity (UIBC), or transferrin. The serum iron, although often elevated in hemochromatosis, has been a less reliable test than the transferrin saturation. The transferrin saturation has been reported to have a sensitivity of greater than 90% for hemochromatosis. However, this has previously been part of the diagnostic criteria. The sensitivity and specificity of transferrin saturation have usually been established at referral centers, where most of the cases have HFE-linked hemochromatosis. The sensitivity of transferrin saturation is lower in population-screening studies designed to detect C282Y homozygotes and was only 52% (threshold >50%) in a large screening study from San Diego, California, which included a significant number of cases with a normal serum ferritin.31 The transferrin saturation continues to have a high sensitivity for the detection of an iron-loaded hemochromatosis patient. A fasting value has even greater predictive value but may not always be practical. The fasting transferrin saturation is most useful in excluding false-positive cases. The transferrin saturation is often elevated in young adults with hemochromatosis before the development of iron overload and a rising ferritin. The threshold to pursue further diagnostic studies has varied from 45 to 62% in previous studies. A lower threshold picks up more patients with hemochromatosis but also leads to more investigations in patients without hemochromatosis. A higher threshold leads to fewer investigations overall with a greater possibility of missing some patients. A common threshold used in screening studies is >45% in women and >50% in men. An elevated transferrin saturation in the presence of a normal serum ferritin rarely indicates significant iron overload but may be a marker that iron overload may develop over time in that patient. Newer genetic mutations may not share this typical pattern of an elevated transferrin saturation and ferritin. A marked elevation (>1000 μg/l) in the presence of a normal transferrin saturation may still represent significant iron overload and further investigations may be indicated to differentiate iron overload from an inflammatory elevation in ferritin.
UNSATURATED IRON-BINDING CAPACITY The UIBC is a one-step colorimetric assay that has been used in many reference laboratories to calculate the transferrin saturation. It is an inexpensive test compared to transferrin saturation and has been demonstrated to be a promising initial screening test for
1243
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 67-3. A proposed diagnostic algorithm for patients with suspected hemochromatosis.
Patient has clinical features suggesting hemochromatosis and elevated transferrin saturation and/or serum ferritin
C282Y and H63D genotyping
C282Y homozygote
C282Y/H63D H63D homozygote
Ferritin < 1000 mg/L AST, ALT normal
Ferritin > 1000 mg/L AST, ALT elevated
C282Y heterozygote
No HFE mutations H63D heterozygote
Reassess other causes of iron overload
Liver biopsy with elevated liver iron concentration
Weekly phlebotomy until serum ferritin = 50 mg/L Genetic counseling Family studies
Figure 67-4. The relationship between transferrin saturation and serum ferritin in C282Y homozygotes (n = 411, r = 0.37 , P < 0.0001). The blue circles represent patients with cirrhosis.
8000 *p<0.05
7000
Serum ferritin (mg/L)
6000 5000 4000 3000 2000 1000 0 0
20
40
60
80
100
Transferrin saturation (%)
hemochromatosis.32,33 A UIBC <26 mmol/l in men and <33 mmol/l in women has been proposed as a screening threshold for HFElinked hemochromatosis with a similar sensitivity and specificity to transferrin saturation.34 The major value is the lower cost and ease of automation and it may be the ideal initial test for high-volume population-screening studies.
SERUM FERRITIN The relationship between serum ferritin and total body iron stores has been clearly established by strong correlations with hepatic
1244
iron concentration and amount of iron removed by venesection. However, ferritin can be elevated secondary to obesity, chronic alcohol consumption, steatohepatitis, chronic inflammation, including viral hepatitis, and histiocytic neoplasms. A major diagnostic dilemma in the past was whether the serum ferritin is related to hemochromatosis or another underlying liver disease such as alcoholic liver disease, chronic viral hepatitis, or non-alcoholic steatohepatitis (with or without insulin resistance). It is likely that some of these difficult cases will now be resolved by genetic testing. As the ferritin increases, the risk of significant liver disease also
Chapter 67 HEMOCHROMATOSIS
300
1000 Obesity/Fatty liver Daily alcohol consumption Inflammation Early hemochromatosis Unknown
5000 Hemochromatosis Alcoholic liver disease Hepatitis C Hepatitis B Other iron overload
100,000 Histiocytosis Still's disease Fulminant liver failure HIV HCC
increases. This has led to the recommendation in the American Association for the Study of Liver Diseases Practice Guidelines that liver biopsy be considered when the serum ferritin is greater than 1000 mg/l.30,35,36 The investigation of hyperferritinemia can be challenging. It may be helpful to classify the most common causes of an elevated ferritin according to the degree of ferritin elevation (Figure 67-5). Extreme elevations of serum ferritin (>50 000 mg/l) are usually not related to hemochromatosis. Since serum ferritin is a glycosylated protein secreted by macrophages, neoplasms of macrophages such as histiocytosis can produce extreme elevations in ferritin. These diseases can cause elevations of liver enzymes associated with infiltration of the liver and a variant called histiocytic medullary reticulocytosis can present with acute liver failure. Elevations of serum ferritin in the range of 1000–5000 mg/l can be associated with clinical hemochromatosis but a careful investigation for hepatitis B and C, alcoholic liver disease, and non-alcoholic steatohepatitis should be considered. If another liver disease is the predominant clinical diagnosis, it is more likely that the ferritin elevation is secondary to that disease rather than the concomitant presence of genetic hemochromatosis. In this range of serum ferritin, liver biopsy is often recommended and the review of the liver pathology, iron staining, and liver iron concentration is often diagnostic. The most common problem is the assessment of mild elevations in serum ferritin in the range of 300–1000 mg/l. The prevalence of an increased serum ferritin is so common in men that the origin and appropriateness of a reference range of up to 300 mg/l have been questioned. Genetic testing can be useful to detect early hemochromatosis but most cases will have normal HFE testing, and heterozygosity for a HFE mutation is an unlikely explanation for an elevated ferritin.37 Although it had been considered that heterozygotes could have mild iron overload, since genetic testing has become available it has become apparent that most of these apparent heterozygotes with mild iron overload were actually compound heterozygotes (C282Y/H63D).38 The epidemic of obesity has likely contributed to this high prevalence of ferritin elevations since fatty liver may be the most common cause of an elevated serum ferritin. This is assumed to be related in most cases to inflammation secondary to steatohepatitis and not to iron overload. However, many liver biopsies do not show large amounts of inflammatory cells and correlations with inflammatory markers such as C-reactive protein or erythrocyte sedimentation rate have been inconsistent. A clinical approach to the assessment of these patients is shown in Table 672. A liver biopsy is often unappealing to the physician and the patient in the setting of a mild elevation in ferritin. Serial monitoring is often done and a ferritin rising above 1000 mg/l is an indica-
Figure 67-5. The differential diagnosis of an elevated serum ferritin differs according to the degree of elevation (reference range 15–200 μg/l women, 30–300 μg/l).
Table 67-2. A Clinical Approach to an Elevated Serum Ferritin
• • • • • • •
Is this a new or chronic observation? Is the ferritin rising? Is the transferrin saturation elevated? Is obesity present? Is daily alcohol consumption present? Is inflammation present? Is there a family history of iron overload?
tion for more investigations or empirical phlebotomy therapy. Noninvasive investigations such as magnetic resonance imaging (MRI) scanning may be most useful in this clinical setting; however the current techniques are not ideally suited for the evaluation of mild iron overload (<2 ¥ upper limit of normal, <60 mmol/g).39,40 Phlebotomy therapy can be considered but since the ferritin is more likely to be inflammatory rather than an indicator of iron overload, the patient will likely become anemic after several phlebotomies without a marked decrease in the serum ferritin. Voluntary blood donation may also be an alternative in healthy patients.41 There are patients with marked elevations in serum ferritin without iron overload. The presence of cataracts at a young age is a clue to the diagnosis of the hyperferritinemia–cataract syndrome.42 Still’s disease and human immunodeficiency virus (HIV) disease are two other conditions in which marked elevations of serum ferritin without iron overload can occur.
IRON REMOVED BY VENESECTION AS DIAGNOSTIC CRITERIA Since hemochromatosis has usually been diagnosed when symptoms developed in the fifth or sixth decade, patients had significant iron overload at the time of diagnosis. The weekly removal of 500 ml of blood (0.25 gram iron) was well tolerated, often for years, without the development of significant anemia. If a patient became anemic (hemoglobin <10 g/dl) after only six venesections, this suggested mild iron overload incompatible with the diagnosis of hereditary hemochromatosis. These guidelines may no longer apply as population and pedigree studies uncover patients in the second and third decade. This is another historical diagnostic criterion for hemochromatosis that will no longer be as relevant in the era of genetic testing. Furthermore, these guidelines were established in C282Y-linked hemochromatosis and patients with other types of iron overload such as aceruloplasminemia and ferroportin mutations commonly become anemic with phlebotomy without achieving iron depletion.
1245
Section XII. Inherited and Pediatric Diseases of the Liver
LIVER BIOPSY Liver biopsy has previously been the gold-standard diagnostic test for hemochromatosis. Liver biopsy has shifted from a major diagnostic tool to a method of estimating prognosis and concomitant disease. The need for liver biopsy seems less clear now in the young asymptomatic C282Y homozygote where there is a low clinical suspicion of cirrhosis based on history, physical examination, and liver biochemistry. A large study conducted in France and Canada suggested that C282Y homozygotes with a serum ferritin of <1000 mg/l, a normal AST, and without hepatomegaly have a very low risk of cirrhosis.43 C282Y homozygotes with a ferritin >1000 mg/l, an elevated AST, and a platelet count < 2.0 ¥ 109/mm3 had a 77–81% chance of having cirrhosis.35 Patients with cirrhosis have a 5.5-fold relative risk of death compared to the non-cirrhotic hemochromatosis patients.24,44 Liver biopsy has not been widely accepted by hematologists, public health physicians, geneticists, and some hemochromatosis patients, and should always be considered optional. Liver biopsy is considered in typical C282Y homozygotes with liver dysfunction; however, it is most useful in the patient without HFE mutations since it may demonstrate that iron overload is not present and therefore phlebotomy therapy is not required. The distribution of iron within the liver can still provide clues to the cause of the iron overload. Typical C282Y-linked hemochromatosis has a portal to central vein gradient of iron distribution within hepatocytes. A predominance of iron within macrophages in the absence of transfusions may suggest a ferroportin mutation. If the liver biopsy suggests another diagnosis, such as alcoholic hepatitis, or chronic viral hepatitis with patchy iron distribution in macrophages, the iron is likely secondary to the primary disease. Simple C282Y heterozygotes, compound heterozygotes (C282Y/H63D) and patients with other risk factors (alcohol abuse, chronic viral hepatitis) with moderate to severe iron overload (ferritin >1000 mg/l) may be considered for liver biopsy.
HEPATIC IRON CONCENTRATION AND HEPATIC IRON INDEX The traditional method of assessing iron status by liver biopsy utilizes the semiquantitative staining method of Perls. This is adequate when there is no iron staining or massive parenchymal iron overload. However, when moderate iron overload is present, the degree of iron overload can be difficult to interpret. There have been more comprehensive methods developed to analyze liver iron staining and distribution but these are not widely utilized. Hepatic iron concentration can be measured using atomic absorption spectrophotometry. This can be done on a piece of paraffin-embedded tissue so special preparation is not required at the time of the biopsy. An advantage of cutting the tissue from the block is that you are more certain that the tissue assayed is the same as the tissue examined under the microscope. For example, a piece of tissue set aside for iron analysis at the time of the biopsy may represent intercostal muscle rather than liver tissue. The normal reference range for hepatic iron concentration is 0–35 mmol/g dry weight (<2000 mg/g). The hepatic iron concentration (mmol/g) divided by age (years) is the hepatic iron index. This was demonstrated to be a useful test in differentiating the patient with genetic hemochromatosis
1246
from the patient with alcoholic siderosis. The index remains a useful test in this clinical setting but it has been extrapolated to be a diagnostic criterion for hemochromatosis. A threshold of 1.9 for the hepatic iron index had a 91% sensitivity for hemochromatosis and area under the receiver operating characteristic curve was 0.94 (0.9–0.99, 95% confidence interval).45 Early diagnosis in populationscreening and pedigree studies have demonstrated many homozygotes with a hepatic iron index <1.9.18 Increasing awareness of the concept of moderate iron overload in cirrhosis of any etiology has demonstrated many patients without hemochromatosis with a hepatic iron index >1.9. The hepatic iron index has become less useful with the advent of genetic testing. The comment on liver biopsy reports that the hepatic iron index is greater than 1.9, confirming a diagnosis of genetic hemochromatosis, should be strongly discouraged. It will remain a tool to aid the clinician with clinical judgment about an individual case. It may be most useful in the unusual hemochromatosis patient that is negative by conventional genetic testing but clinically seems to have genetic hemochromatosis. There are patients with 1–2 + iron staining on liver biopsy who have a normal liver iron concentration and the liver iron concentration can be used to help to decide the need for phlebotomy therapy.
IMAGING STUDIES OF THE LIVER MRI can demonstrate moderate to severe iron overload of the liver. Proponents of MRI to alleviate the need for liver biopsy have emphasized the non-invasive nature of the test for the diagnosis. In a study of 174 patients, Gandon et al. demonstrated that a simple MRI protocol could detect hepatic iron overload >60 mmol/g (normal range 0–36 mmol/g) with a sensitivity of 89%.39 A modification of this protocol was used by a second group in the study of 112 patients in which a correlation coefficienct of 0.94 was demonstrated between liver iron concentration and MRI estimation of liver iron concentration.40 These techniques are improving and may be ideally suited to exclude iron overload in a patient in whom an inflammatory condition may be responsible for the elevations in iron tests such as severe alcoholic hepatitis. The MRI can also demonstrate the clinical features of cirrhosis such as nodularity of the liver, ascites, portal hypertension, and splenomegaly, as well as hepatocellular carcinoma, but these features can be more readily assessed by abdominal ultrasound at a lower cost. Magnetic susceptibility has been studied using a superconducting quantum interference device (SQUID) machine to estimate liver iron concentration but the technology is not widely available.46
GENETIC TESTING FOR HEMOCHROMATOSIS A major advance which stems from the discovery of the hemochromatosis gene (HFE) is the use of a diagnostic genetic test. Clinical studies in well-defined hemochromatosis pedigrees reported that 90–100% of typical hemochromatosis patients were homozygous for the C282Y mutation of the HFE gene. The presence of a single mutation in most patients is in marked contrast to other genetic diseases in which multiple mutations were discovered (cystic fibrosis,
Chapter 67 HEMOCHROMATOSIS
Serum Ferritin Levels in Men and Women According to Genotype 6400 Men Women
3200
Serum ferritin (mg/liter)
1600 800 400
100 50 25
+/ +
D/ + 63 H
2Y 28 C
D/ 63 H
2Y 28 C
/+
D H
63 /H
28 /C 2Y
63
D
2Y
7
28
C282Y homozygote This is the classical genetic pattern which is seen in > 90% of typical cases. Expression of disease ranges from no evidence of iron overload to massive iron overload with organ dysfunction. Siblings have a 1 in 4 chance of being affected and should have genetic testing. For children to be affected, the other parent must be at least a heterozygote. If iron studies are normal, false-positive genetic testing or a non-expressing homozygote should be considered. C282Y/H63D – compound heterozygote This patient carries one copy of the major mutation and one copy of the minor mutation. Most patients with this genetic pattern have normal iron studies. A small percentage of compound heterozygotes have been found to have mild to moderate iron overload. Severe iron overload is usually seen in the setting of another concomitant risk factor (alcoholism, viral hepatitis).
200
C
Table 67-3. Interpretation of Genetic Testing for Hemochromatosis
HFE Genotype
Figure 67-6. Serum ferritin (μg/l) by genotype in men and women. An elevated serum ferritin was considered to be >200 μg/l in women and >300 μg/l in men. Data are presented as box plots and means (*) are geometric means. The box stretches from the 25th to the 75th percentile. The median is shown as a line across the box and the mean as (*). The whiskers indicate 1.5 times the interquartile range above the third and below the first quartiles, or to the upper or lower extreme values, whichever is closer. Previously diagnosed cases of hemochromatosis are excluded. (Reproduced from Adams PC, Reboussin DM, Barton JC, et al. Hemochromatosis and iron-overload screening in a racially diverse population. N Engl J Med 2005; 352:1769–1778, with permission.)
Wilson’s disease, a1-antitrypsin deficiency). A second minor mutation, H63D, was also described in the original report.2 This mutation does not cause the same intracellular trafficking defect of the HFE protein. Compound heterozygotes (C282Y/H63D) and, less commonly, H63D homozygotes47 may resemble C282Y homozygotes with mild to moderate iron overload. These genotypes are much more common than C282Y homozygotes in the general population yet are not commonly reported in large series of typical hemochromatosis patients. Large population studies have demonstrated that most patients with C282Y/H63D or H63D/H63D have normal iron studies1,22 (Figure 67-6).16 A polymorphism on intron 4 of the HFE gene (5569A) was independently reported by several laboratories to lead to false-positive genetic testing in which a C282Y heterozygote appears to be a homozygote.48,49 This should be considered during the evaluation of a “non-expressing” C282Y homozygote. The correct diagnosis can be confirmed by direct DNA sequencing. Other hemochromatosis HFE mutations have not been clearly established in large studies to explain iron overload in nonC282Y homozygotes. It is likely that more mutations will be found but they will only be relevant to a minority of patients. The interpretation of the test in several settings is shown in Table 67-3. Genetic discrimination is a major concern with the widespread use of genetic testing. A positive genetic test even without iron overload could disqualify a patient for health or life insurance.50–52 In the
C282Y heterozygote This patient carries one copy of the major mutation. This pattern is seen in about 10% of the Caucasian population and is usually associated with normal iron studies. In rare cases the iron studies are high, in the range expected in a homozygote rather than a heterozygote. These cases may carry an unknown hemochromatosis mutation and liver biopsy is helpful to determine the need for venesection therapy. H63D homozyote This patient carries two copies of the minor mutation. Most patients with this genetic pattern have normal iron studies. A small percentage of these cases have been found to have mild to moderate iron overload. Severe iron overload is usually seen in the setting of another concomitant risk factor (alcoholism, viral hepatitis). H63D heterozygote This patient carries one copy of the minor mutation. This pattern is seen in about 20% of the Caucasian population and is usually associated with normal iron studies. This pattern is so common in the general population that the presence of iron overload may be related to another risk factor. Liver biopsy may be required to determine the cause of the iron overload and the need for treatment in these cases. No HFE mutations There are other iron overload diseases associated with mutations in other iron-related genes (transferrin receptor-2, ferroportin, HJV). Genetic testing is not widely available for these conditions. There will likely be other hemochromatosis mutations discovered in the future. If iron overload is present without any HFE mutations, a careful history for other risk factors must be reviewed and liver biopsy may be useful to determine the cause of the iron overload and the need for treatment. Most of these cases are isolated, non-familial cases.
case of hemochromatosis, the advantages of early diagnosis in young adulthood of a treatable disease outweigh the disadvantages of genetic discrimination. The widespread use of genetic testing for hemochromatosis has also led to misinterpretation of the test results by the patient and physician. For example, a H63D heterozygote, which is seen in 1 in 5 of the Caucasian population, may be interpreted as evidence of hemochromatosis. This can occur because of the complexity of the genetic test report, and is also commonly seen when patients prefer to attribute their lifestyle-induced liver disease to a genetic problem. In this setting, the patient often attributes every symptom from
1247
Section XII. Inherited and Pediatric Diseases of the Liver
Table 67-4. Genetic diseases of iron metabolism Clinical feature
HFE-linked hemochromatosis
Ferroportin disease
Juvenile hemochromatosis
Transferrin receptor 2
Aceruloplasminemia
Gene location
6p21.3
2q32
7q22
3q25
Inheritance pattern Organs involved Hepatic iron distribution Anemia Response to phlebotomy
Autosomal recessive Liver, endocrine, heart Hepatocytes
Autosomal dominant Liver, spleen
1q21 (HJV), 19q13.1 (hepcidin) Autosomal recessive
Macrophages
Heart, endocrine, liver Hepatocytes
Autosomal recessive Liver, endocrine, heart Hepatocytes
Autosomal recessive Retina, basal ganglia, pancreas, liver Hepatocytes
No Excellent
Yes Anemia
No Excellent
No Excellent
Yes Anemia
head to toe to this genetic test result and may be seeking disability benefits. This is clearly an adverse effect of genetic testing. Genetic testing is not recommended for children since organ damage is not typically seen in childhood and early detection may lead to insurance discrimination, labeling, or stigmatization. Most cases of familial iron overload are C282Y homozygotes but there are a growing number of less common genetic mutations that may result in iron overload3,53 (Table 67-4). Some of these conditions have been labeled HFE2, HFE3, and HFE4 but they do not refer to mutations in the HFE gene, so this nomenclature is not recommended. In an isolated case with hepatic iron overload, consideration should first be given to a secondary cause of iron overload. The identification of an iron-loaded family member is a powerful clinical clue to the presence of a genetic disorder in iron metabolism. Siblings have the highest yield in an autosomal recessive disease. Pedigree studies have become difficult in some countries because of geographical separation, differences in health care providers, and new privacy legislation.
FERROPORTIN DISEASE Ferroportin-associated iron overload is an autosomal dominant disorder in iron metabolism resulting from a mutation of the SLC40A1 on chromosome 2q32.53 The ferroportin protein is normally involved in iron transport out of macrophages and a ferroportin mutation causes an accumulation in hepatic macrophages. The typical clinical findings include a progressive elevation in serum ferritin and a mild hypochromic anemia. Transferrin saturation may be normal or rise as the iron overload increases. In an original report from the Netherlands, several cases were described with severe hepatic iron overload with a normal transferrin saturation and ferritin.54 Liver disease can include hepatic fibrosis but this is usually less severe than HFE-linked hemochromatosis. MRI scanning can demonstrate iron overload in the liver and spleen, in contrast to HFE-linked hemochromatosis, in which there is a relative paucity of iron in the spleen. Ferroportin mutations have been described in a wide range of ethnic groups and countries, including Caucasians, AfricanAmericans, and Asians. A ferroportin mutation has been found in a patient from the Solomon Islands, suggesting that the previously described Polynesian iron overload may be related to a ferroportin mutation.55 Treatment is by a slow phlebotomy protocol (500 ml, once per month) because anemia develops with weekly phlebotomy.
1248
ACERULOPLASMINEMIA The original description of this genetic mutation on chromosome 3q25 was of progressive retinal and basal ganglia degeneration, diabetes, and ataxia. Hepatic iron overload has been demonstrated and serum ceruloplasmin is absent or very low. Treatment of this condition has been problematic because of the development of anemia with phlebotomy.56
JUVENILE HEMOCHROMATOSIS The typical clinical profile of a patient with juvenile hemochromatosis was a male in the second or third decade presenting with severe congestive heart failure, ventricular arrhythmias, and hypogonadism. The disease was often fatal and severe iron overload was demonstrated in the liver, heart, and endocrine organs. Although considered to be a single disease, recent advances in the genetics of juvenile hemochromatosis suggest that this is genetically heterogeneous. A gene on chromosome 1 (HJV) produces hemojuvelin.28 This iron protein appears to interact with hepcidin to regulate body iron metabolism. A growing number of genetic mutations have been described in the HJV gene. Mutations in the hepcidin gene on chromosome 19 have also been described to produce a similar clinical picture to juvenile hemochromatosis. Treatment of juvenile hemochromatosis may require twice-weekly phlebotomy and careful cardiac supportive care. Phlebotomy can lead to a marked improvement in symptoms and cardiac function. Arrhythmia management can be challenging since chronic amiodarone therapy can be hepatotoxic and many patients have cirrhosis at the time of diagnosis. Pacemakers and implantable defibrillators have been used to manage arrhythmias. Testosterone replacement therapy is often required.
TRANSFERRIN RECEPTOR-2 MUTATION Several European family studies have identified mutations in the transferrin receptor-2 gene (chromosome 7q22) associated with hepatic iron overload. The liver findings were similar to HFE-linked hemochromatosis, with a predominance of iron in the hepatocytes. This is an autosomal recessive condition and can be treated by phlebotomy.57
Chapter 67 HEMOCHROMATOSIS
GENETIC TESTING FOR NON-HFE IRON OVERLOAD The genetic tests for mutations in ferroportin, hepcidin, hemojuvelin, and transferrin receptor-2 are not likely to become widely available commercially because of the low prevalence of these mutations. Research centers are more likely to be interested in a pedigree of iron overload before initiating newer genetic tests, which can be expensive and often are non-informative. As more polymorphisms are described, the genotypic–phenotypic correlations have been difficult to establish and it important to establish the prevalence of these new polymorphisms in larger populations of affected and unaffected cases. Newer advances in microarrays and gene chips may allow for the simultaneous identification of multiple iron genes.58
NON-EXPRESSING HOMOZYGOTES As genetic testing becomes more widespread, an increasing number of individuals have been found with the hemochromatosis gene without iron overload. This includes siblings within well-defined hemochromatosis pedigrees.59 Pooled estimates from 14 studies have suggested that 50% of C282Y homozygotes may not have iron overload.60 The prevalence of an elevated ferritin in C282Y homozygotes from screening studies is shown in Figure 67-7. The term “nonexpressing” homozygote has been used for C282Y homozygotes with a normal ferritin and transferrin saturation, a normal transferrin saturation, and an elevated ferritin, and in asymptomatic homozygotes with elevated iron tests. Patients who are homozygous for the C282Y mutation should be considered at risk of developing iron overload but, if there are no abnormalities in transferrin saturation or ferritin in adulthood, it seems more likely that they are a non-expressing homozygote rather than a patient who will develop iron overload later in life. The follow-up of adult C282Y homozygotes with a normal serum ferritin has not demonstrated a marked increase in the serum ferritin.15,61–63 It seems appropriate at the present time to repeat the serum ferritin and transferrin saturation every 5 years in non-iron-loaded C282Y homozygotes to understand
more about their natural history. The study of the non-expressing homozygote may provide additional information about new modifying genes that counteract the effect of the hemochromatosis gene.
FAMILY STUDIES IN HEMOCHROMATOSIS Once the proband case is identified and confirmed with the genetic test for the C282Y mutation, family testing is strongly recommended.64 Siblings have the highest chance of carrying the gene and should be screened with the genetic test (C282Y and H63D mutation), transferrin saturation, and serum ferritin (Figure 67-8). Phenotypic expression can vary widely between siblings, suggesting that environmental factors are contributory. Patients are usually very concerned about their children and may have difficulty with the concept of autosomal recessive transmission. The risk to a child is dependent on the prevalence of heterozygotes in the community and is probably greater than 1 in 20 and much lower if the spouse is nonCaucasian.65 A cost-effective strategy now possible with the genetic test is to test the spouse for the C282Y mutation to assess the risk in the children. If the spouse is not a C282Y heterozygote or homozygote, the children will be obligate heterozygotes. This assumes paternity and excludes another gene or mutation causing hemochromatosis. This strategy is particularly advantageous where the children are geographically separated or may be under a different health care system. If an isolated heterozygote is detected by genetic testing, it is recommended to test siblings. Extended family studies are less revealing than a family study with a homozygote but more likely to uncover a homozygote than random population screening. The risk to family members is illustrated in Table 67-5. The impact of the H63D mutation is far less than the C282Y mutation. Genetic counseling for an autosomal recessive disease like HFElinked hemochromatosis has been within the realm of the gastroenterologist or hepatologist. However, as additional mutations are discovered, new polymorphisms are identified that may or may not
Figure 67-7. The percentage of C282Y homozygotes (men, red boxes, women, blue boxes) in population-screening studies with an elevated serum ferritin at the time of discovery.
Percentage with elevated ferritin
100
80
60
40
20
0 Olynyk 1999 3,011
Asberg 2001 65,238
Beutler 2002 41,038
Deugnier 2002 9,396
Heirs 2004 101,168
1249
Section XII. Inherited and Pediatric Diseases of the Liver
1B 64 years 3487 mg/L
1A 68 years 287 mg/L
2A 47 years 134 mg/L
2B 46 years 179 mg/L
2C 41 years 12 mg/L
2D 38 years 14 mg/L
2E 35 years 21 mg/L
1C 59 years 1106 mg/L
2F 34 years 171 mg/L
2G 25 years 30 mg/L
2H 32 years 203 mg/L
2I 23 years 35 mg/L
Figure 67-8. A pedigree in a hemochromatosis family carrying the C282Y mutation. Homozygotes are the black squares and heterozygotes are the half-filled symbols. Age and serum ferritin are displayed. Note the presence of a non-expressing 68-year-old C282Y homozygous male with a normal serum ferritin (1A) with two expressing younger siblings (1B and 1C).
Table 67-5. Predicting risk to family members with hemochromatosis A. Risks for iron overload in relatives of C282Y/C282Y probanda Chance of: Risk to:
Father Mother Brother Sister Son Daughter
C282Y/C282Y
C282Y/C282Y with iron overload
C282Y/H63D
C282Y/H63D with iron overload
H63D/H63D
H63D/H63D with iron overload
5.6% 5.6% 26.4% 26.4% 5.5% 5.5%
5.0% 2.8% 23.8% 13.2% 4.9% 2.7%
28.5% 28.5% 7.9% 7.9% 15.0% 15.0%
0.6% 0.3% 0.2% 0.1% 0.3% 0.2%
Doesn’t occur Doesn’t occur 0.0% 0.0% Doesn’t occur Doesn’t occur
0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
B. Risks for iron overload in relatives of C282Y/H63D proband Chance of: Risk to:
Father Mother Brother Sister Son Daughter
C282Y/C282Y
C282Y/C282Y with iron overload
C282Y/H63D
C282Y/H63D with iron overload
H63D/H63D
H63D/H63D with iron overload
2.8% 2.8% 1.4% 1.4% 2.7% 2.7%
2.5% 1.4% 1.3% 0.7% 2.5% 1.4%
20.1% 20.1% 27.9% 27.9% 10.2% 10.2%
0.4% 0.2% 0.6% 0.3% 0.2% 0.1%
8.1% 8.1% 3.8% 3.8% 7.5% 7.5%
0.1% 0.0% 0.0% 0.0% 0.1% 0.0%
a The risks to family members are based on Hardy–Weinberg equilibrium. This assumes that a Caucasian marries another Caucasian and that there are no paternity issues. The risk to a sibling of a C282Y homozygote is slightly higher than 1 in 4 (25%), at 26.4%, because the parents have a rare chance of being C282Y homozygotes or compound heterozygotes. Allele frequency C282Y = 0.05, H63D = 0.15. Risk of iron overload was based on available estimates from population-screening studies and assumes that 90% of male and 50% of female C282Y homozygotes will have elevated iron tests. Risks of iron overload in C282Y/H63D and H63D/H63D homozygotes are much lower since most participants with these genotypes have normal iron studies (Adapted from Adams P, Acton R, Walker A. A primer for predicting risk of disease in HFE-linked hemochromatosis. Genetic Testing 2001; 5:311–316.)
1250
Chapter 67 HEMOCHROMATOSIS
contribute to iron overload, and multiple genes are tested simultaneously, genetic counseling becomes far more complex and may require additional support from medical geneticists.
DIAGNOSIS OF NON-HFE HEMOCHROMATOSIS AND SECONDARY IRON OVERLOAD It is important to remember that there will be patients with a clinical picture indistinguishable from genetic hemochromatosis who will be negative for the C282Y mutation. In non-Caucasians, iron overload is not commonly associated with HFE mutations.53 Therefore, screening for HFE mutations in African-Americans and Asians has not been clinically useful. Many Asians have been found to have elevations in serum ferritin and transferrin saturation, although documented cases of iron overload remain rare.66 Hispanic populations have been found to have HFE mutations, and this is likely related to their Spanish heritage. Iron overload and ferroportin mutations have been described in African-Americans67 but many of these cases had other risk factors for iron overload and pedigree studies have not been commonly reported. The iron overload described in subSaharan Africans may be related to another iron-loading gene68 and a linkage to iron overload in African-Americans is an intriguing hypothesis that awaits the identification of that gene. A negative C282Y test should alert the physician to question the diagnosis of genetic hemochromatosis and reconsider secondary iron overload related to cirrhosis, alcohol, viral hepatitis, or iron-loading anemias. If no other risk factors are found, the patient should begin venesection treatment similar to any other hemochromatosis patient. The decision to classify this group of patients as non-HFE hemochromatosis or idiopathic iron overload is a matter of semantics and ideology surrounding case definition. Quantification of iron burden by hepatic iron concentration or quantitative phlebotomy will be important to characterize this group of patients further. In most cases of secondary iron overload secondary to blood diseases, the patient has anemia with iron overload from increased iron absorption and/or multiple transfusions. These patients will not tolerate venesections and will require parenteral chelation therapy. A liver biopsy and echocardiogram may be helpful to determine if the secondary iron overload is causing organ damage. Since parenteral desferrioxamine therapy has chronic side effects, clinical judgment is required in each case to assess whether the benefits outweigh the toxicity of chelation therapy. Oral iron chelators such as deferiprone have been controversial because of potential hepatotoxicity.69
TREATMENT OF HFE-LINKED HEMOCHROMATOSIS The therapy of hemochromatosis continues to use the medieval therapy of periodic bleeding. The goal of therapy is to remove excess iron to prevent any further tissue damage. Phlebotomy therapy has never been subjected to a randomized clinical trial and this has hindered our understanding of the natural history of untreated disease. Although most experts believe that iron depletion can stabilize liver
disease, improve cardiac function and dyspnea, and reduce skin pigmentation, there are still skeptics who have suggested that there is no evidence to support phlebotomy therapy.15 At our center, patients attend an ambulatory care facility and the venesections are peformed by a nurse using a kit containing a 16-gauge straight needle and collection bag (Blood Pack MR6102, Baxter, Deerfield, IL) Blood is removed with the patient in the reclining position over 15–30 minutes. A hemoglobin is done at the time of each venesection. If the hemoglobin has decreased to less than 10 g/dl, the venesection schedule is modified to 500 ml every other week. The concomitant administration of a salt-containing sport beverage (e.g. Gatorade) is a simple method of maintaining plasma volume during the venesection. Maintenance venesections after iron depletion of 3–4 venesections per year are done in most patients, although the rate of iron reaccumulation is highly variable.70,71 The transferrin saturation will remain elevated in many treated patients and will not normalize unless the patient becomes iron-deficient. In some cases, a component of the serum ferritin elevation is related to inflammation rather than iron overload, so the ferritin does not decrease with treatment and the patient becomes anemic. In these cases, a careful review of the liver biopsy, including hepatic iron concentration, may be helpful in deciding to discontinue treatment or decrease the frequency of phlebotomies. There are different ways to follow the progess of phlebotomy therapy. At our center, patients are treated until the serum ferritin is approximately 50 μg/l. This is at the low end of the normal range and allows some room for iron reaccumulation into the normal range. Patient support groups have advocated for more intensive phlebotomy but fatigue begins to intervene as iron deficiency is approached. The transferrin saturation may not decrease until the patient is on the brink of iron deficiency and therefore, we will discontinue phlebotomy therapy in some patients with a low ferritin but an elevated transferrin saturation (Figure 679). Other approaches to monitoring phlebotomy therapy include the continuation of weekly phlebotomy until anemia develops (hemoglobin <10 g/dl) or monitoring the mean corpuscular volume during therapy. Many patients enjoy the concept of maintenance therapy, particularly if they can be voluntary blood donors.72 The evidence supporting the need for maintenance therapy is lacking70 and it may be useful to repeat a serum ferritin in 6 months following cessation of therapy to estimate the risk of iron reaccumulation. Repeat liver biopsy is rarely done and is not recommended. Chelation therapy with desferrioxamine is not recommended for hemochromatosis. The therapy is expensive, inefficient, cumbersome, and potentially toxic. Oral chelators such as deferiprone have side effects such as agranulocytosis and possible hepatotoxicity. Erythrocytophoresis has been used but is more expensive than simple phlebotomy therapy. Patients are advised to avoid oral iron therapy and alcohol abuse but there are no proven dietary restrictions. Patient support groups are discouraged by the practice of iron fortification of foods, but much of this iron is in an inexpensive form with poor bioavailability. Iron fortification has been removed from food in Sweden and a decrease in the mean serum ferritin has been demonstrated in the normal population.73 Tea consumption has been shown to decrease intestinal iron absorption. Many patient support groups recommend avoidance of shellfish because of the increased risk of Vibrio infections in iron overload.
1251
80 60 40 20
Cumulative iron removal (g)
Transferrin saturation (%)
100
25
2500
20
2000
15
1500
10
1000
5
500
Serum ferritin (ng/ml)
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 67-9. Response of indirect markers of iron overload to phlebotomy therapy. This figure illustrates the reciprocal relationship between the serum ferritin level and the removal of excess iron stores.
0
0 0
2
4
6
8
10
12
14
16
18
20
Time (mos)
LIVER TRANSPLANTATION FOR HEMOCHROMATOSIS In a patient with decompensated cirrhosis from hemochromatosis, liver transplantation can be done. Despite the high prevalence of hemochromatosis, it remains an uncommon indication for liver transplant, and many cases that we labeled as hemochromatosis are more likely to have had iron overload secondary to cirrhosis from other causes. A consistent finding with liver transplantation in the setting of iron overload is a higher mortality postoperatively from sepsis and heart disease. Pretransplant phlebotomy may improve cardiac function and is recommended if tolerated. Chelating agents have been suggested in several case reports. Recurrence of iron overload after liver transplantation has been infrequent.23 Transplantation of iron-loaded livers from C282Y homozygotes into recipients has usually resulted in the mobilization of hepatic iron over time, and the transplantation of a small intestine and liver into a recipient resulted in the development of iron overload in the recipient.74 These transplant experiments have been fertile ground for speculation on the pathogenesis of hemochromatosis.23
POPULATION SCREENING FOR HEMOCHROMATOSIS Identification of individuals at risk of developing the sequelae of hemochromatosis is preferred. The plan is to screen a population of asymptomatic individuals, with no personal or family history to suggest that they are at higher risk of the disease than the rest of the population. It aims to detect disease in presymptomatic individuals in order to provide more effective treatment in the early stages of disease. Since screening programs are initially associated with increased health care costs, it is imperative that planning of screening protocols considers all the risks and benefits, and the diagnostic strategy relevant to the disease being considered prior to implementation. Population screening for hemochromatosis meets most of the criteria established by the World Health Organization for screening for medical disease. However, the area of greatest concern is the uncertainty about the natural history of untreated hemochromatosis. An important aspect of a screening project is the demonstration of a difference between screened and unscreened cases. If unscreened cases would never develop significant morbidity, the utility of population screening is greatly decreased. If most
1252
C282Y homozygotes are asymptomatic and are not progressing to cirrhosis or diabetes, then the early identification and treatment of patients may not be as cost-effective as had initially been projected.15,75 Sporadic screening studies have been performed to establish the utility of various tests and the prevalence of hemochromatosis in a target population prior to the use of genetic testing. Target populations have included blood donors, hospital inpatients, outpatients, employees, diabetics, and army recruits. Initial testing and test thresholds also varied and have included serum iron, UIBC, transferrin saturation, ferritin, and combinations of these tests. Since the advent of genetic testing in 1996, many studies have used an iron test (transferrin saturation, ferritin, UIBC) as the initial screening test with follow-up genetic testing in those with elevated iron tests. This strategy minimizes the potential for genetic discrimination in C282Y homozygotes with normal iron tests and it also has the potential to detect non-HFE iron overload.76,77 An example of this approach is a study in 65 238 Norwegians attending a health appraisal clinic.22 Genetic testing was only done in cases with an elevated transferrin saturation. In this study, liver biopsies were performed in 149 patients and cirrhosis of the liver was detected in 3.7% of men and none of the women. The symptoms in the C282Y homozygotes did not differ significantly from a control population.78 The screening of 41 038 primary care patients in San Diego, California, detected 152 C282Y homozygotes. In an analysis of clinical symptoms only liver disease was significantly more common in C282Y homozygotes compared to participants without HFE mutations. There was no difference in diabetes, arthritis, impotence, or pigmentation between these two groups.16 The screening of 101 168 primary care participants of diverse racial background found an increase in self-reported liver disease in C282Y homozygotes and compound heterozygotes.1 In general, the screening studies to date have demonstrated a high prevalence of HFE mutations in Caucasian populations but low morbidity.79 These studies have also highlighted the non-specific nature of the symptoms that have been historically attributed to hemochromatosis. The observation that there is significant disease in referred cases and much less disease in screened cases is not unique to hemochromatosis and occurs in most screening studies. Another example of this phenomenon can be seen in the study of hepatitis
Chapter 67 HEMOCHROMATOSIS
C. The unresolved issue is whether or not these asymptomatic C282Y homozygotes would develop cirrhosis or diabetes if untreated. A clinical trial randomizing C282Y homozygotes to treatment or non-treatment seems unlikely to be done because of ethical concerns of withholding treatment. Periodic health examination in asymptomatic people has been discouraged by many health care systems. Screening in diabetic and arthritis clinics has also been studied. By the time that diabetes is present, organ damage is also evident and arthritis is a common cause of an elevation in serum ferritin which will increase the number of false-positive patients who will go on to have invasive testing. The use of genetic testing within an arthritis clinic would likely improve the screening algorithm but several studies in diabetes and arthritis clinics have not shown an increased prevalence of HFE mutations.80 At the present time, it seems likely that mass populationscreening studies will not be recommended. Targeted screening in young males could be considered in countries with a homogeneous Caucasian population.
Porphyria cutanea tarda is a cutaneous complication of chronic liver disease, including hepatitis C, alcoholic liver disease, and hemochromatosis. HFE mutations have been found to be more common in porphyria cutanea tarda patients in many studies.90 Phlebotomy therapy is useful in these cases to improve the skin rash. The presence of anemia should immediately suggest that HFElinked hemochromatosis is not the correct diagnosis. Mild anemia has been seen in other iron overload genetic diseases such as ferroportin mutations and aceruloplasminema.91 The initial investigations include review of the peripheral blood smear which may provide clues to the diagnosis of spherocytosis, sickle-cell disease, or thalassemia. Further investigations may include hemoglobin electrophoresis and bone marrow examination. Hepatic iron overload can occur even in the absence of blood transfusions, and the liver biopsy in an untransfused thalassemia patient can be indistinguishable from HFE-linked hemochromatosis. Transfusional iron overload is characterized by iron deposits in the macrophages in the liver.
PROGNOSIS IN HEMOCHROMATOSIS END-STAGE CIRRHOSIS It is well recognized that iron overload can complicate many forms of end-stage liver disease.81 The most common liver diseases associated with secondary iron overload are alcoholic liver disease, hepatitis B or C, and non-alcoholic steatohepatitis. Iron overload is much less common in chronic cholestatic liver diseases. It is important to recognize that most patients with these liver diseases have elevations in serum ferritin and/or transferrin saturation without iron overload. This is a source of diagnostic confusion and liver biopsy and liver iron concentration are often helpful in this clinical setting. The role of HFE mutations (C282Y heterozygotes, compound heterozygotes C282Y/H63D, H63D homozygotes and heterozygotes) in the pathogenesis of mild iron overload in these other liver diseases has been controversial but a study of liver explants for non-hemochromatosis demonstrated no increase in prevalence of HFE mutations compared to a screened population.82 Alcoholic siderosis is an uncommon manifestation of alcoholic liver disease and historical reports of an increase in alcoholism in hemochromatosis were likely related to inclusion of these cases as hemochromatosis.83 Hepatitis C has also commonly been associated with elevated iron tests but has uncommonly been associated with significant elevations in liver iron concentrations. Phlebotomy therapy has shown a decrease in liver enzymes but no consistent effect on hepatitic C virus RNA has been demonstrated in controlled trials studying the effects of phlebotomy and antiviral therapy.84 The role of HFE mutations and their contribution to fibrogenesis remain controversial in hepatitis C but theories have had to invoke non-iron-related mechanisms since many cases have normal liver iron concentrations.85–88 The prevalence of HFE mutations has also been studied in non-alcoholic steatohepatitis, with conflicting results.89 Insulin resistance has been associated with mild to moderate iron overload in non-alcoholic steatohepatitis independent of HFE mutations.27 A typical patient is obese, with an elevated ferritin and normal transferrin saturation.
The major factor affecting the long-term outcome in hemochromatosis is the presence of cirrhosis. Cirrhotic patients have a 5.5fold relative risk of death compared to non-cirrhotic patients. Diabetes is also a risk factor affecting long-term prognosis but most diabetic patients with hemochromatosis also have cirrhosis. Cirrhotic patients are also at risk of hepatocellular carcinoma, which is a common cause of death.24,72 The long-term survival in cirrhotic hemochromatosis patients compared to non-cirrhotic patients is shown in Figure 67-10. These findings are the major stimulus for screening programs that attempt to identify and treat hemochromatosis patients at a precirrhotic stage of disease. However, there are likely a large group of hemochromatosis patients that never progress to cirrhosis or diabetes without phlebotomy therapy.
1.0
0.8
Probability
SECONDARY IRON OVERLOAD
0.6
0.4
0.2 Non-cirrhosis Cirrhosis 0.0 0
10
20
30
Cumulative survival (years) Figure 67-10. The long-term survival in C282Y homozygotes. Cirrhotic patients had a significantly decreased survival compared to non-cirrhotic patients (n = 392, P < 0.0001, log-rank test.)
1253
Section XII. Inherited and Pediatric Diseases of the Liver
CONCLUSIONS The clinical profile of a typical hemochromatosis patient has not changed significantly since the scholarly descriptions of Sheldon in 1935. However, the inclusion of control patients in populationbased studies has brought into question the attribution of symptoms to hemochromatosis and iron overload. Of all the symptoms, liver disease has the most consistent relationship to hemochromatosis and the prognosis of hemochromatosis is most closely linked to the degree of iron overload. The discovery of the HFE gene in 1996 was a major advance in the field, and most Caucasian patients with typical hemochromatosis can be diagnosed with a commercially available genetic test. However, a growing number of new iron-related genes have been discovered and linked to other iron overload syndromes.
REFERENCES 1. Adams PC, Reboussin DM, Barton JC, et al. Hemochromatosis and iron-overload screening in a racially diverse population. N Engl J Med 2005; 352:1769–1778. 2. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in patients with hereditary hemochromatosis. Nature Genetics 1996; 13:399–408. 3. Pietrangelo A. Hereditary hemochromatosis: a new look at an old disease. N Engl J Med 2004; 350:2383–2397. 4. Distante S, Robson K, Graham-Campbell J, et al. The origin and spread of the HFE-C282Y haemochromatosis mutation. Hum Genet 2004; 115:269–279. 5. Lucotte G, Dieterlen F. A European allele map of the C282Y mutation of hemochromatosis: Celtic versus Viking origin of the mutation? Blood Cells Mol Dis 2003; 31:262–267. 6. Merryweather-Clarke A, Pointon J, Jouanolle A, et al. Geography of HFE C282Y and H63D mutations. Genet Test 2000; 4:183–198. 7. Hentze M, Muckenthaler M, Andrews N. Balancing acts: molecular control of mammalian iron metabolism. Cell 2004; 117:285–297. 8. Andrews N, Fleming M, Levy J. Molecular insights into mechanisms of iron transport. Curr Opin Hematol 1999; 6:61–64. 9. Muckenthaler M, Roy C, Custodio A, et al. Regulatory defects in liver and intestine implicate abnormal hepcidin and Cybrd1 in mouse hemochromatosis. Nat Genet 2003; 34:102–106. 10. Yamaji S, Sharp P, Ramesh B, et al. Inhibition of iron transport across human intestinal epithelial cells by hepcidin. Blood 2004;104:2178–2180. 11. Bridle K, Frazer D, Wilkins S, et al. Disrupted hepcidin regulation in HFE-associated haemochromatosis and the liver as a regulator of body iron homeostasis. Lancet 2003; 361:661–673. 12. Gehrke S, Kulaksiz H, Hermann T, et al. Expression of hepcidin in hereditary hemochromatosis: evidence for a regulation in response to the serum transferrin saturation and to nontransferrin bound iron. Blood 2003; 102:371–376. 13. Merryweather-Clarke A, Cadet E, Bomford A, et al. Digenic inheritance of mutations in HAMP and HFE result in different types of hemochromatosis. Hum Mol Genet 2003; 12:2241–2247. 14. LeGac G, Scotet V, Ka C, et al. The recently identified type 2A juvenile haemochromatosis gene (HJV), a second candidate modifer of the C282Y homozygous phenotype. Hum Mol Genet 2004; 13:1913–1918.
1254
15. Beutler E. Natural history of hemochromatosis. Mayo Clin Proc 2004; 79:305–306. 16. Beutler E, Felitti V, Koziol J, et al. Penetrance of the 845G to A (C282Y) HFE hereditary haemochromatosis mutation in the USA. Lancet 2002; 359:211–218. 17. Adams P, Reboussin D, Acton R, et al. Prevalence of self-reported symptoms in the HEIRS study: differences by HFE genotype. Hepatology (abstract) 2003; 38:230A. 18. Adams PC, Deugnier Y, Moirand R, et al. The relationship between iron overload, clinical symptoms and age in 410 patients with genetic hemochromatosis. Hepatology 1997; 25:162–166. 19. Adams P. Is there a threshold of hepatic iron concentration for the development of cirrhosis in C282Y hemochromatosis? Am J Gastro 2001; 96:567–569. 20. Fletcher L, Dixon J, Purdie D, et al. Excess alcohol greatly increases the prevalence of cirrhosis in hereditary hemochromatosis. Gastroenterology 2002; 122:563–565. 21. Moirand R, Adams PC, Bicheler V, et al. Clinical features of genetic hemochromatosis in women compared to men. Ann Int Med 1997; 127:105–110. 22. Asberg A, Hveem K, Thorstensen K, et al. Screening for hemochromatosis – high prevalence and low morbidity in an unselected population of 65 238 persons. Scand J Gastroenterol 2001; 36:1108–1115. 23. Crawford D, Fletcher L, Hubscher S, et al. Patient and graft survival after liver transplantation for hereditary hemochromatosis: implications for pathogenesis. Hepatology 2004; 39:1655–1662. 24. Niederau C, Fischer R, Purschel A, et al. Long-term survival in patients with hereditary hemochromatosis. Gastroenterology 1996; 110:1107–1119. 25. Adams PC. Hepatocellular carcinoma in hereditary hemochromatosis. Can J Gastroenterol 1993; 7:37–41. 26. Hramiak I, Finegood D, Adams PC. Factors affecting glucose tolerance in hereditary hemochromatosis. Clin Invest Med 1997; 20:110–118. 27. Mendler MH, Turlin B, Moirand R, et al. Insulin resistanceassociated hepatic iron overload. Gastroenterology 1999; 117:1155–1163. 28. Papanikolaou G, Samuels M, Ludwig E, et al. Mutations in HFE2 cause iron overload in chromosome 1q-linked juvenile hemochromatosis. Nature Genetics 2004; 36:77–82. 29. Adams PC, Speechley M. The effect of arthritis on the quality of life in hereditary hemochromatosis. J Rheumatol 1996; 23:707–710. 30. Tavill AS. Diagnosis and management of hemochromatosis. Hepatology 2001; 33:1321–1328. 31. Beutler E, Felitti V, Gelbart T, et al. The effect of HFE genotypes on measurement of iron overload in patients attending a health appraisal clinic. Ann Intern Med 2000; 133:329–337. 32. Adams PC, Kertesz AE, McLaren C, et al. Population screening for hemochromatosis: a comparison of unbound iron binding capacity, transferrin saturation and C282Y genotyping in 5211 voluntary blood donors. Hepatology 2000; 31:1160–1164. 33. Hickman P, Hourigan L, Powell LW, et al. Automated measurement of unsaturated iron binding capacity is an effective screening strategy for C282Y homozygous hemochromatosis. Gut 2000; 46:405–409. 34. Adams P, Zaccaro D, Moses G, et al. A comparison of the unsaturated iron binding capacity to transferrin saturation as a screening test to detect C282Y homozygotes for hemochromatosis in 101 168 participants in the HEIRS study. Clin Chem 2005; 51:1048–1052. 35. Beaton M, Guyader D, Deugnier Y, et al. Non-invasive prediction of cirrhosis in C282Y-linked hemochromatosis. Hepatology 2002; 36:673–678. 36. Morrison E, Brandhagen D, Phatak P, et al. Serum ferritin levels predicts advanced hepatic fibrosis among US patients with
Chapter 67 HEMOCHROMATOSIS
37.
38.
39. 40. 41. 42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53. 54.
55.
56. 57.
58.
phenotypic hemochromatosis. Ann Intern Med 2003; 138:627–633. Moirand R, Jouanolle A-M, Brissot P, et al. Phenotypic expression of HFE mutations: a French study of 1110 unrelated ironoverloaded patients and relatives. Gastroenterology 1999; 116:372–377. Rossi E, Olynyk J, Cullen D, et al. Compound heterozygous hemochromatosis genotype predicts increased iron and erythrocyte indices in women. Clin Chem 2000; 46:162–166. Gandon Y, Olivie D, Guyader D, et al. Non-invasive assessment of hepatic iron stores by MRI. Lancet 2004; 363:357–360. Alustiza J, Artetxe J, Castiella A, et al. MR quantification of hepatic iron concentration. Radiology 2004; 230:479–484. Power T, Adams PC. Hemochromatosis patients as voluntary blood donors. Can J Gastro 2004; 18:393–396. Roetto A, Bosio S, Gramaglia E, et al. Pathogenesis of hyperferritinemia cataract syndrome. Blood Cell Mol Dis 2002; 29:532–535. Guyader D, Jacquelinet C, Moirand R, et al. Non-invasive prediction of fibrosis in C282Y homozygous hemochromatosis. Gastroenterology 1998; 115:929–936. Adams PC, Speechley M, Kertesz AE. Long-term survival analysis in hereditary hemochromatosis. Gastroenterology 1991; 101:368–372. Adams PC, Bradley C, Henderson AR. Evaluation of the hepatic iron index as a diagnostic criterion in hereditary hemochromatosis. J Lab Clin Med 1997; 130:509–514. Nielsen P, Engelhardt R, Dullman J, et al. Non-invasive liver iron quantification by SQUID-biosusceptometry and serum ferritin iron as new diagnostic parameters in hereditary hemochromatosis. Blood Cells Mol Dis 2002; 29:451–458. Aguilar-Martinez P, Bismuth M, Picot M, et al. Variable phenotypic presentation of iron overload in H63D homozygotes: are gene modifiers the cause? Gut 2001; 6:836–842. Jeffrey G, Chakrabarti S, Hegele R, et al. Polymorphism in intron 4 of HFE may cause overestimation of C282Y homozygote prevalence in haemochromatosis. Nat Genet 1999; 22:325–326. Somerville M, Sprysak K, Hicks M, et al. An HFE intronic variant promotes misdiagnosis of hereditary hemochromatosis. Am J Hum Genet 1999; 65:924–926. Barash C. Genetic discrimination and screening for hemochromatosis: then and now. Genet Testing 2000; 4:213–218. Shaheen N, Lawrence L, Bacon B, et al. Insurance, employment, and psychosocial consequences of a diagnosis of hereditary hemochromatosis in subjects without end-organ damage. Am J Gastroenterol 2003; 98:1175–1180. Power T, Adams PC. Psychosocial impact of genetic screening for hemochromatosis in population screening and referred patients. Genet Testing 2001; 5:107–110. Pietrangelo A. Non-HFE hemochromatosis. Hepatology 2004; 39:21–29. Njajou O, Dejong G, Berghuis B, et al. Dominant hemochromatosis due to N144H mutation of SLC11A3: clinical and biological characteristics. Blood Cell Mol Dis 2002; 29:439–443. Wallace D, Pedersen P, Dixon J, et al. Novel mutation in ferroportin 1 is associated with autosomal dominant hemochromatosis. Blood 2002; 100:692–694. Hellman N, Schaefer M, Gehrke S, et al. Hepatic iron overload in aceruloplasminemia. Gut 2003; 47:858–860. Camaschella C, Roetto A, Cali A, et al. The gene TFR2 is mutated in a new type of haemochromatosis mapping to 7q22. Nat Genet 2000; 25:14–15. Kotze M, DeVillers J, Bouwens C, et al. Molecular diagnosis of hereditary hemochromatosis: application of a newly-developed reverse-hybridization assay in the South African population. Clin Genet 2004; 65:317–321.
59. Adams PC, Chakrabarti S. Genotypic/phenotypic correlations in genetic hemochromatosis: evolution of diagnostic criteria. Gastroenterology 1998; 114:319–323. 60. Bacon BR. Hemochromatosis: diagnosis and management. Gastroenterology 2001; 120:718–725. 61. Yamashita C, Adams PC. Natural history of the C282Y homozygote of the hemochromatosis gene (HFE) with a normal serum ferritin level. Clin Gastroenterol Hepatol 2003; 1:388–391. 62. Olynyk J, Hagan S, Cullen D, et al. Evolution of untreated hereditary hemochromatosis in the Busselton population: a 17year study. Mayo Clin Proc 2004; 79:309–313. 63. Andersen R, Tybjaerg-Hansen A, Appleyard M, et al. Hemochromatosis mutations in the general population: iron overload progression rate. Blood 2004; 103:2914–2919. 64. Gleeson F, Ryan E, Barrett S, et al. Clinical expression of haemochromatosis in Irish C282Y homozygotes identified through family screening. Eur J Gastroenterol Hepatol 2004; 16:859–863. 65. Adams P, Acton R, Walker A. A primer for predicting risk of disease in HFE-linked hemochromatosis. Genet Testing 2001; 5:311–316. 66. McLaren G, Barton J, Gordeuk V, et al. High HFE C282Y homozygote frequency in Caucasians but not in Hispanics, African Americans, or Asians: an analysis of 50 290 primary care patients in the hemochromatosis and iron overload screening study (HEIRS). Blood 2002; 100: 447a. 67. Beutler E, Barton J, Felitti V, et al. Ferroportin 1 (SCL40A1) variant associated with iron overload in African-Americans. Blood Cells Mol Dis 2003; 31:305–309. 68. Gordeuk V, Mukiibi J, Hasstedt SJ, et al. Iron overload in Africa. Interaction between a gene and dietary iron content. N Engl J Med 1992; 326:95–100. 69. Olivieri N, Brittenham G, McLaren C, et al. Long-term safety and effectiveness of iron-chelation therapy with deferiprone for thalassemia major. N Engl J Med 1998; 339:417–423. 70. Adams PC, Kertesz AE, Valberg LS. Rate of iron reaccumulation following iron depletion in hereditary hemochromatosis. Implications for venesection therapy. J Clin Gastroenterol 1993; 16:207–210. 71. Adams PC. Factors affecting rate of iron mobilization during venesection therapy for hereditary hemochromatosis. Am J Hematol 1998; 58:16–19. 72. Wojcik J, Speechley M, Kertesz A, et al. Natural history of C282Y homozygotes for haemochromatosis. Can J Gastro 2002; 16:297–302. 73. Olsson K, Vaisanen M, Konar J, et al. The effect of withdrawal of food iron fortification in Sweden as studied with phlebotomy in subjects with genetic hemochromatosis. Eur J Clin Nutr 1998; 51:782–786. 74. Adams PC, Alanen K, Preshaw R, et al. Transplantation of haemochromatosis liver and intestine into a normal recipient. Gut 1999; 45:783. 75. Adams PC, Valberg LS. Screening blood donors for hereditary hemochromatosis: decision analysis model comparing genotyping to phenotyping. Am J Gastroenterol 1999; 94:1593–1600. 76. Adams PC. Population screening for haemochromatosis. Gut 2000; 46:301–303. 77. McCullen MA, Crawford D, Hickman P. Screening for hemochromatosis. Clin Chim Acta 2002; 315:169–186. 78. Asberg A, Hveem K, Kruger O, et al. Persons with screeningdetected haemochromatosis: as healthy as the general population? Scand J Gastroenterol 2002; 37:719–724. 79. Hanson E, Imperatore G, Burke W. HFE gene and hereditary hemochromatosis. Am J Epidemiol 2001; 154:193–206. 80. DuBois S, Kowdley K. Targeted screening for hereditary haemochromatosis in high-risk groups. Aliment Pharm Ther 2004; 20:1–14.
1255
Section XII. Inherited and Pediatric Diseases of the Liver
81. Ludwig J, Hashimoto E, Porayko M, et al. Hemosiderosis in cirrhosis: a study of 447 native livers. Gastroenterology 1997; 112:882–888. 82. Alanen K, Chakrabarti S, Rawlins J, et al. Prevalence of the C282Y mutation of the hemochromatosis gene in liver transplant recipients and donors. Hepatology 1999; 30:665–669. 83. Adams PC, Agnew S. Alcoholism in hereditary hemochromatosis revisited: prevalence and clinical consequences among homozygous siblings. Hepatology 1996; 23:724–727. 84. Fontana RJ, Israel J, LeClair P, et al. Iron reduction before and during interferon therapy of chronic hepatitis C: results of a multicenter, randomized, controlled trial. Hepatology 2000; 31:730–736. 85. Thorburn D, Curry G, Spooner R, et al. The role of iron and haemochromatosis gene mutations in the progression of liver disease in chronic hepatitis C. Gut 2002; 50:248–252. 86. Tung B, Edmond M, Bronner M, et al. Hepatitis C, iron status, disease severity: relationship with HFE mutations. Gastroenterology 2003; 124:318–326.
1256
87. Bataller R, North K, Brenner D. Genetic polymorphisms and the progression of liver fibrosis: a critical appraisal. Hepatology 2003; 37:493–503. 88. Hezode C, Cazeneuve C, Coue O, et al. Liver iron accumulation in patients with chronic active hepatitis C: prevalence and role of hemochromatosis gene mutations and relationship with hepatic histological lesions. J Hepatol 1999; 31:979–984. 89. Bugianese E, Manzini P, D’Antico S, et al. Relative contributions of iron burden, HFE mutations, and insulin resistance to fibrosis in non-alcoholic fatty liver. Hepatology 2004; 39:179–187. 90. Bonkovsky HL, Poh-Fitzpatrick M, Pimstone N, et al. Porphyria cutanea tarda, hepatitis C, and HFE gene mutations in North America. Hepatology 1998; 27:1661–1669. 91. Xu X, Pin S, Gathinji M, et al. Aceruloplasminemia: an inherited neurodegenerative disease with impairment of iron homeostasis. Ann NY Acad Sci 2004; 1012:299–305.
Section XII: Inherited and Pediatric Diseases of the Liver
a1-ANTITRYPSIN DEFICIENCY
68
David H. Perlmutter Abbreviations a1AT a1-antitrypsin MPT mitochondrial permeability transition
PBA
phenylbutyric acid
INTRODUCTION a1-Antitrypsin (a1AT) deficiency is an autosomal co-dominant disorder associated with premature development of pulmonary emphysema, chronic liver disease, and hepatocellular carcinoma. A point mutation results in a protein that is retained in the endoplasmic reticulum (ER) of liver cells rather than secreted into the blood and body fluids. Loss of function permits uninhibited proteolytic destruction of the connective tissue matrix of the lung, ultimately leading to emphysema. Cigarette smoking exacerbates the development of emphysema because residual a1AT molecules are functionally inactivated by the increased load of active oxygen intermediates that are produced by smokers’ alveolar macrophages. In contrast, liver disease results from a gain-of-function mechanism whereby the retention of mutant a1AT in liver cells somehow triggers a series of events that lead to liver injury and a predilection for hepatocellular carcinoma. Although it is the most common genetic cause of liver disease in children, only ~10% of affected homozygotes develop clinically significant liver disease, an observation that has provided the basis for the notion that genetic modifiers and environmental factors play a role in susceptibility to and/or protection from liver disease in a1AT deficiency. Although there are several attractive new concepts for chemoprophylaxis and treatment of liver disease in this deficiency, the only effective treatment currently available is liver transplantation.
EPIDEMIOLOGY The incidence of the deficiency is 1 in 1800 to 1 in 2000 live births in most populations that have been carefully studied.1 Most of the studies of lung and liver disease in this population have been biased in ascertainment because they involve patients referred to specialty clinics. The only unbiased study comes from nationwide screening of all newborns in Sweden in the 1970s.2,3 Over 200 000 newborns were screened and 127 homozygotes for the a1AT Z alleles were identified and most of them have now been followed for almost 30 years. The results show that only 14 of these individuals (11%) had prolonged obstructive jaundice in infancy and only 9 (7%) have developed clinically significant liver disease. Of the remaining a1ATdeficient population, 85% have had persistently normal transaminase levels as they have aged. Liver biopsies have not been done in this study and therefore it is not known whether some of these seemingly unaffected individuals have subclinical histological abnormalities and will develop clinical signs of liver disease as they reach
SERPIN
serine protease inhibitor
the fourth and fifth decades of life. Although it is widely believed that lung disease affects a higher percentage of homozygotes,4 a valid determination of incidence will not be possible until the Swedish prospective cohort reaches the peak age range for emphysema, 40–60 years of age.
PATHOGENESIS As the archetype of the serine protease inhibitor (SERPIN) family, a1AT mainly functions as a blood-borne inhibitor of destructive neutrophil proteases, including elastase, cathepsin G, and proteinase 3. a1AT is predominantly derived from parenchymal liver cells, constituting the most abundant glycoprotein secreted by the liver. a1AT is also considered a positive acute-phase reactant because its plasma concentration increases during the host response to inflammation/tissue injury. The classical deficiency mutant, a1ATZ, is characterized by a point mutation that results in the substitution of lysine for glutamate 342 and by itself accounts for defective secretion. The mutant a1ATZ molecule is retained in the ER of liver cells (Figure 68-1). Studies by Carrell and Lomas have shown that this substitution reduces the stability of a1ATZ as a monomer and increases the formation of polymers by a “loop-sheet” insertion mechanism.5 Although polymers have been detected by electron microscopy in hepatocytes of a liver biopsy specimen from an a1AT-deficient individual6 and by sucrose density gradient centrifugation in transfected cell lines that express mutant a1ATZ,7,8 it is still not clear whether polymerization is responsible for ER retention of mutant a1ATZ. The strongest evidence for this comes from studies in which the fate of a1ATZ is examined after the introduction of additional mutations into the classical mutant molecule. For instance, Kim et al. introduced a mutation, F51L, on the back side of the a1ATZ molecule remote from the Z mutation, E342K, but predicted to impede loop-sheet polymerization.9 This double mutant was less prone to polymerization and more efficiently secreted in microinjected Xenopus oocytes10 and transfected yeast.11 A number of other observations appear to militate against the notion that polymerization is the cause of ER retention. For one, we recently found that a novel, naturally occurring variant of a1AT bearing both the E342K substitution that characterizes the Z molecule and a mutation that results in truncation of the carboxyl-terminus that is retained in the ER for at least as long as a1ATZ, even though it does not polymerize.7 These results could indicate that there are mechanisms other than polymerization that can result in ER retention and that each of the mechanisms could affect a particular a1AT variant. It is also possible that a single, as yet
1257
Section XII. Inherited and Pediatric Diseases of the Liver
Nucleus
a1 AT gene
Factor B gene
a1 AT transcript Plasma membrane
Factor B transcript
Rough ER / Golgi
a1 AT Protein Factor B Protein
Figure 68-1. Secretory defect in a1AT deficiency. The mutant a1AT protein is retained in the endoplasmic reticulum, whereas another hepatic secretory protein, complement factor B, is distributed throughout the secretory pathway and is efficiently secreted. (Reproduced from Perlmutter DH. Alpha-1-antitrypsin deficiency. In: Walker WA, Durie PR, Hamilton JR, et al., eds. Pediatric gastrointestinal disease. Philadelphia, PA: BC Decker; 1991:979, with permission).
undetermined mechanism causes ER retention of a1ATZ as well as other non-polymerogenic mutants and that polymerization of a1ATZ in the ER is an effect rather than a cause of retention. There is relatively limited information about how ER retention of mutant a1ATZ causes liver injury by a gain-of-toxic function mechanism. Recent studies have shown that liver from a1AT-deficient patients is characterized by significant mitochondrial injury and mitochondrial autophagy as well as activation of caspases-3 and -9.12,13 These mitochondrial changes and caspase activation were also seen in transfected cell line and transgenic model systems. Indeed, treatment of one transgenic mouse model with cyclosporin A, which inhibits the mitochondrial permeability transition (MPT) pore, resulted in diminished hepatic mitochondrial injury and caspase activation and improved survival.12 This may mean that ER retention of a1ATZ causes mitochondrial injury by a direct interaction between the distended ER and adjacent mitochondria. In fact, this author’s most recent studies in model cell lines with inducible expression have shown that ER retention of a1ATZ activates BAP31,14 an ER protein that mediates direct interactions between ER and mitochondria.15 It is also possible that mitochondria are injured as innocent bystanders of an overexuberant autophagic response that is activated by ER retention of a1ATZ. Autophagy is a ubiquitous, highly conserved cellular mechanism by which senescent and/or denatured constituents in the cytoplasm and intracellular organelles or whole organelles are sequestered from the rest of the cytoplasm within newly formed vacuoles that then fuse with lysosomes for
1258
degradation (Figure 68-2). It is believed to be a mechanism for turnover of cellular constituents during nutritional deprivation, stress states, morphogenesis, differentiation, and aging. Our previous studies have shown that ER retention of a1ATZ is a powerful stimulus for the autophagic response.12,16,17 A marked increase in autophagosomes has been observed in several different model cell lines genetically engineered to express a1ATZ, including human fibroblasts, murine hepatoma, and rat hepatoma cell lines.16 Moreover, in a HeLa cell line engineered for inducible expression of a1ATZ, autophagosomes appear as a specific response to the expression of a1ATZ and its retention in the ER. There is a marked increase in autophagosomes in hepatocytes in transgenic mouse models of a1AT deficiency and a disease-specific increase in autophagosomes in liver biopsies from patients with the deficiency.16 Mutant a1ATZ molecules can be detected in autophagosomes by immune electron microscopy, often together with the ER chaperone calnexin. Intracellular degradation of a1ATZ is partially abrogated by chemical inhibitors of autophagy16 and, when it is expressed in autophagydeficient (atg5-knockout) cell lines (N Mitzushima, T Yoshimura, DH Perlmutter, manuscript in preparation), indicating that autophagy also contributes to the quality control mechanism for disposal of a1ATZ. Recently, the autophagic response to ER retention of a1ATZ in vivo was examined by testing the effect of fasting on the liver of the PiZ mouse model of a1AT deficiency.17 Starvation is a well-defined physiologic stimulus of autophagy, as well as a known environmen-
Chapter 68 a1-ANTITRYPSIN DEFICIENCY
Normal Liver Cell
Autophagic body
Autophagosome Starvation signal
1. Induction
2. Formation
3. Docking and fusion
Figure 68-2. Autophagosomes in (A) normal and (B) a1-antitrypsin (AT)-deficient liver cells. In the normal liver cell, a stress, such as starvation, leads to induction of vacuolar membranes, formation of an autophagosome that envelops cytoplasm and organelles, docking and fusion of the autophagosome with lysosome, and breakdown and fusion of its contents within the lysosome. In the a1ATdeficient cell, accumulation of mutant a1ATZ in the endoplasmic reticulum (ER) leads to induction of vacuolar membranes and pinching-off of ER to form autophagosomes. (Adapted from Klionsky DJ, Emr SD. Autophagy as a regulated pathway of cellular degradation. Science 2000; 290:1717–1721, with permission.)
4. Breakdown and recycling
Lysosome
Plasma membrane A
a1 AT Deficient Liver Cell
ER
Autophagosome
1. Induction
2. Formation
Autophagic body
3. Docking and fusion
Plasma membrane
4. Breakdown and recycling Lysosome
B
tal stressor of liver disease in children. The results showed that there is a marked increase in fat accumulation and in a1ATZ-containing, ER-derived globules in the liver of the PiZ mouse induced by fasting. These changes were particularly exaggerated at 3–6 months of age. Three-month-old PiZ mice had a significantly decreased tolerance for fasting compared to non-transgenic C57/BL6 littermates. Although fasting induced a marked autophagic response in wild-type mice, the autophagic response was already activated in PiZ mice to levels that were more than 50% higher than those in the liver of fasted wild-type mice. In contrast to wild-type mice, there was no increase in autophagosomes in the liver of PiZ mice during fasting. These results indicate that autophagy is constitutively activated in the liver in a1AT deficiency and that the liver is unable to mount an increased autophagic response to physiologic stressors.
There is almost nothing known about the pathogenesis of hepatocellular carcinoma in a1AT deficiency. Results of recent studies by Rudnick et al.13 have suggested an interesting hypothesis. Using bromodeoxyuridine labeling in the PiZ transgenic mouse model of a1AT deficiency, these authors show that there is an increase in proliferation of hepatocytes in the liver of these mice under resting conditions and that this increase is proportional to the amount of a1AT that has accumulated in the ER in globules. However, the cells that proliferate are the ones that do not have the globules. Previous studies of transgenic mouse models of a1AT deficiency have shown that the number of globule-devoid hepatocytes and the area of the liver occupied by globule-devoid hepatocytes increase with age and this is the area where adenomas and carcinomas arise.18 Interestingly, autophagosomes are predominantly found in liver cells
1259
Section XII. Inherited and Pediatric Diseases of the Liver
that contain globules in the PiZ mouse.16 Several studies have shown that autophagic activity inhibits tumorigenesis in vivo.19 Taken together, these observations suggest that hepatocytes that have accumulated a lot of a1AT and have formed globules are sick but not dead, that globule-devoid hepatocytes therefore have a selective proliferative advantage and, most importantly, that adenomas and eventually carcinomas evolve in the globule-devoid hepatocytes because they are chronically stimulated to regenerate in an ‘injury milieu’ by signals generated from the sick, globule-containing hepatocytes. A similar mechanism appears to occur in the mouse model of hereditary tyrosinemia.20 In the latter model, hepatocytes which have genetically reverted or transplanted wild-type hepatocytes have a selective proliferative advantage and are chronically stimulated by damaged cells that are unable to undergo complete cell death. To address the question of how a subpopulation of a1AT-deficient individuals become susceptible to liver disease and how the remainder of the population is apparently protected from liver disease, a number of years ago it was hypothesized that genetic modifiers or environmental factors which affected the fate of the a1ATZ molecule once it accumulated in the ER or the protective cellular response pathways activated by accumulation of a1ATZ in the ER would play an important role. This hypothesis was validated by a series of experiments in which the fate of a1ATZ in skin fibroblast cell lines from deficient individuals with liver disease (susceptible hosts) was compared to its fate in skin fibroblast cell lines from deficient individuals without liver disease (protected hosts) after expression was established in each by stable gene transduction techniques.21 The results showed that a1ATZ was retained in the ER in
each case, but it was degraded more efficiently in cell lines from the protected hosts (Figure 68-3). The mechanisms for the degradation of mutant a1ATZ once it has accumulated in the ER are very complex. There appear to be ubiquitin-dependent proteosomal22 and ubiquitin-independent proteosomal23 as well as non-proteosomal pathways involved. At least two non-proteosomal pathways, a tyrosine phospatase-dependent pathway24 and autophagy,16 have already been implicated. Genetic modifiers that affect the function of any of these pathways would theoretically increase susceptibility to liver disease. The protective cellular response pathways that are activated by ER retention of mutant a1ATZ are also being elucidated. We know from our studies that the autophagic response is activated by this state.12,16,17 In recent work, using cell line and transgenic mouse models with inducible expression of a1ATZ, optimally designed to determine the effect of this state on a naive cell or a naive organ, it was discovered that there is activation of ER-specific caspases, caspase-12 in the mouse and caspase-4 in the human, and activation of NFkB.14 Taken together with previous studies,16 these observations suggest that both the ER and the mitochondrial caspase pathways are activated in the a1AT-deficient liver. Because NFaB mediates increased expression of interleukin-8, a potent neutrophil chemotactic factor, its activation is likely to be very important in the development of inflammation within the liver. Genetic modifiers that affect these response pathways could also theoretically increase or even decrease susceptibility to liver disease. Chemical or genetic manipulation of these cellular responses could also constitute strategies for chemoprophylaxis or treatment for liver disease in a1AT deficiency.
Protected Synthesis
Susceptible Secretion
Synthesis
Secretion
a1 AT Z Protein in Globules
a1 AT Z Protein in Globules
RER
RER Transcription
Transcription Golgi
Golgi
Translation
Translation
DNA RNA nucleus
DNA RNA nucleus
ER Degradation
ER Degradation Cellular Protective Responses • UPR • Autophagy • ? others
Cellular Protective Responses • UPR • Autophagy • ? others
Figure 68-3. Fate of mutant a1ATZ in liver cells from protected and susceptible hosts. In the susceptible hosts, there is a subtle block in endoplasmic reticulum degradation of a1ATZ or cellular protective responses to a1ATZ.
1260
Chapter 68 a1-ANTITRYPSIN DEFICIENCY
CLINICAL FEATURES In many cases this liver disease first becomes apparent at 4–8 weeks of age because of persistent jaundice (Table 68-1). Conjugated bilirubin and transaminase levels in the blood are mildly to moderately elevated. The liver may be enlarged but rarely is there more in the way of symptoms, signs, or laboratory abnormalities that suggest significant liver injury. It is very difficult to differentiate clinically these infants from infants affected by many other causes of neonatal liver disease, including infections, metabolic diseases, and even the destructive hepatobiliary lesions of biliary atresia, and so a1AT deficiency is usually thought of as one of the causes of a broad diagnostic category termed “neonatal hepatitis syndrome.” Occasionally the diagnosis will be discovered in a newborn with bleeding symptoms such as hematemesis, melena, bleeding from the umbilical stump, or bruising.25 In some cases there may be a cholestatic picture with icterus, pruritus, and laboratory abnormalities such as hypercholesterolemia. Indeed, this subgroup of a1ATdeficient infants may have severe biliary epithelial cell damage and even paucity of the intrahepatic bile ducts detected in their liver biopsies.26 Rarely a1AT deficiency manifests itself with severe progressive liver disease in the first year of life.27 The liver disease of a1AT deficiency may also be diagnosed later in childhood because of asymptomatic hepatomegaly, elevated transaminases detected incidentally, or jaundice that develops during an intercurrent illness. Finally, this disease can first present in childhood, adolescence, or adult life with complications of portal hypertension, including splenomegaly, hypersplenism, gastrointestinal bleeding from varices, ascites, and/or hepatic encephalopathy (Table 68-1). It should be considered in any adult patient with chronic liver disease, cryptogenic cirrhosis, or hepatocellular carcinoma. An autopsy study done in Sweden suggested that as many as 25% of a1AT-deficient men who die between the ages of 40 and 60 have evidence of inflammation, necrosis, and/or carcinoma in their liver.28 The natural history of liver disease in a1AT deficiency is quite variable. Most infants who present with prolonged jaundice are asymptomatic by the time they reach 1 year of age. In the majority of these cases there is no further evidence of liver disease for many years. Because this diagnosis has only been known for 35 or so years,
Table 68-1. Antitrypsin Deficiency-Associated Liver Disease Clinical manifestations Infancy
Early childhood
Late childhood/adolescence
Prolonged obstructive jaundice Elevated transaminases Symptoms of cholestasis Elevated transaminases Asymptomatic hepatomegaly Severe liver dysfunction Chronic active hepatitis Cryptogenic cirrhosis Portal hypertension Hepatocellular carcinoma
Diagnostic features Diminished serum levels of ATZ (10–15% normal levels) Abnormal mobility of antitrypsin in isoelectric focusing (PIZ) Periodic acid–Schiff-positive, diastase-resistant globules in liver cells
it is not yet clear what proportion of these individuals go on to develop liver disease and/or hepatocellular carcinoma. The only prospective data on the course of a1AT deficiency comes from the Swedish nationwide screening study started by Sveger in the early 1970s.2 In that study, 200 000 newborn infants were screened, and 127 were found to have the classical form of a1AT deficiency. Fourteen of the 127 had prolonged obstructive jaundice and 9 of these 14 had clinically significant liver disease. Another 8 of the 127 had hepatomegaly with or without elevated bilirubin or transaminase levels. Approximately 50% of the remaining population had elevated transaminase levels alone. The long-term outcome for these infants was last published when their mean age was 18 years of age,3 but unpublished observations by Sveger indicate that little has changed for the population well into their third decade. There has been no evidence for the development of clinically significant liver disease in any of the patients since the first year of life. This means that only 8% of the population at most have encountered clinically significant liver disease to date. Of the remaining population of a1AT-deficient children identified in this cohort, 85% have had persistently normal transaminase levels as they have aged. Liver biopsies have not been done in this study and therefore it is not known whether some of these seemingly unaffected individuals have subclinical histological abnormalities and will develop clinical signs as they reach the fourth and fifth decades of life. Even patients with severe liver disease caused by a1AT deficiency may have a stable or relatively slowly progressing course. In one retrospective review of a pediatric hepatology service, 9 of 17 patients with a1AT deficiency and cirrhosis, portal hypertension, or both, had a prolonged, relatively uneventful course for at least 4 years after the diagnosis of cirrhosis or portal hypertension was made.29 Two of these patients eventually underwent liver transplantation but 7 were leading relatively healthy lives for as long as 23 years while carrying a diagnosis of severe a1AT deficiencyassociated liver disease. It has not yet been possible to identify specific clinical and/or laboratory signs that can be used to predict a poor prognosis for liver involvement in a1AT deficiency. Results of one early study suggested that persistence of hyperbilirubinemia, hard hepatomegaly, development of splenomegaly, and progressive prolongation of the prothrombin time were indicators of poor prognosis.30 In another study, elevated transaminase levels, prolonged prothrombin time, and lower trypsin inhibitory capacity correlated with a worse prognosis.31 In the author’s experience the first definitive evidence of poor prognosis comes in the form of a complication that affects the overall life functioning of the patient. It is still unclear whether heterozygotes for the classical form of a1AT deficiency are predisposed to liver disease. Early studies of liver biopsy collections suggested that there was a relationship between heterozygosity and the development of liver disease.32 This has been confirmed by later studies of liver biopsy collections. In particular, the liver biopsies from patients who have undergone liver transplantation show a higher than expected prevalence of heterozygosity for the classical form of a1AT deficiency without another diagnostic explanation for severe liver disease.33 However, these studies and others like them have an inherent bias in ascertainment. Results of one cross-sectional study of patients with a1AT deficiency in a referral-based Austrian university hospital, who were re-
1261
Section XII. Inherited and Pediatric Diseases of the Liver
examined with recently developed more sensitive and sophisticated diagnostic assays, suggested that liver disease in heterozygotes could be accounted for, to a large extent, by infections with hepatitis C virus or by autoimmune disease.34 Unfortunately neither type of study has provided convincing evidence for or against a predisposition to liver disease in PiMZ individuals. Nevertheless, the author’s experience with numerous PiMZ individuals with severe liver disease and no other plausible explanation leads me to believe that the predisposition does exist. Liver disease has been described for several other allelic variants of a1AT. Children with compound heterozygosity for the S and Z alleles are affected by liver disease in a manner similar to PIZZ children.2,3 Interestingly, a1AT S forms heteropolymers with a1ATZ35 and this compound heterozygous state is probably associated with retention of the mutant protein in a polymeric form within the ER of liver cells.36 There have been several reports of liver disease in a1AT deficiency PiM Malton.37,38 This is an interesting association, because the abnormal a1AT Malton molecule has been shown to undergo polymerization and retention in the ER.37 Because it has only been reported in single patients with other allelic variants of a1AT,39 it is not clear whether liver disease is causally related to those variants. Destructive lung disease/emphysema caused by a1AT deficiency probably does not become clinically manifest until late in the third decade. Although there are a few reports of younger individuals with lung disease, the diagnosis of a1AT deficiency in these cases was not convincing.40 There is still limited information about the incidence of liver disease in a1AT-deficient individuals with established emphysema. In one study of 22 PIZZ patients with emphysema, there were elevated transaminase levels in 10 patients and cholestasis in 1 patient.41 Liver biopsies were not done in this study and so it might underestimate the extent and incidence of liver disease in adults with emphysema as their predominant clinical problem.
DIAGNOSIS This deficiency should be considered in anyone with elevated transaminases, elevated conjugated bilirubin levels, asymptomatic hepatomegaly, signs or symptoms of portal hypertension, signs or symptoms of cholestasis, as well as bleeding/bruising with a prolonged prothrombin time. It should be considered in adults with chronic idiopathic hepatitis, cryptogenic cirrhosis, and hepatocellular carcinoma. The diagnosis is established by means of serum a1AT phenotype (PI type) determination in isoelectric focusing electrophoresis or agarose electrophoresis at acid pH (Figure 68-4). Serum concentrations can be used for screening with follow-up PI typing of any values below normal (85–215 mg/dl). A retrospective study of all pediatric patients who had both serum concentrations and PI typing done at one center indicated that the serum concentration determination had a positive predictive value of 94% and a negative predictive value of 100% for homozygous PIZZ a1AT deficiency,42 but, because of the inherent limitations of retrospectively defining a patient population for the analysis, the results of the study are not necessarily applicable to each diagnostic situation that might be
1262
Figure 68-4. Isoelectric focusing gel. Blood specimens from three individuals are shown – normal with M1 and M2 alleles; heterozygote with M2 and Z allele (arrow); normal with 2 M1 alleles. (Reproduced from Perlmutter DH. Alpha-1antitrypsin deficiency. In: Snape WJ, ed. Consultations in gastroenterology. Philadelphia, PA: WA Saunders; 1996:791–801, with permission.)
encountered. In my experience it is wise to get both the serum concentration and PI typing when seriously considering this diagnosis. Serum concentrations increase during the host response to inflammation and therein may reach normal levels in heterozygotes and near-normal levels in homozygotes. Both the serum concentrations and the PI type will be needed to confirm the homozygous, compound heterozygous as well as the heterozygous states that can be present at the a1AT locus. In some cases, phenotype determination of parents and other relatives are necessary to confirm the diagnosis if there is any discrepancy and to ensure the distinction between the ZZ and SZ allotypes, for which isoelectric focusing may not be straightforward. This distinction and the others will be important for genetic counseling. The PI type is particularly important in the neonatal period because it may be very difficult to distinguish patients with a1AT deficiency from those with biliary atresia. Moreover, it is not uncommon for neonates with a PIZZ phenotype to have no biliary excretion on scintigraphic studies.43 There is one report of a1AT deficiency and biliary atresia in a single patient.44 I have seen several patients with homozygous PIZZ a1AT deficiency and cholestasis for whom there was no biliary excretion of technetium-labeled mebrofenin, but with more prolonged observation, in each of these cases, cholestasis remitted so that it was then obvious that the patients did not have biliary atresia. The distinctive histological feature of homozygous PIZZ a1AT deficiency, periodic acid–Schiff-positive, diastase-resistant globules in the ER of hepatocytes, substantiates the diagnosis (Figure 68-5). According to some observers, these globules are not as easy to detect in the first few months of life.45 The presence of these inclusions
Chapter 68 a1-ANTITRYPSIN DEFICIENCY
Figure 68-5. Liver biopsy from an a1-antitrypsin (a1-AT)-deficient patient. Cells with periodic acid–Schiff-positive, diastase-resistant globules are shown with arrows. (Reproduced from Consultations in Gastroenterology 1996, 793, with permission).
should not be interpreted as diagnostic of a1AT deficiency. Similar structures are occasionally present in other liver diseases.46 The inclusions are eosinophilic, round to oval and 1–40 μm in diameter. They are most prominent in periportal hepatocytes but may also be seen in Kupffer cells and biliary epithelial cells.47 The liver biopsy may also be characterized by variable degrees of hepatocellular necrosis, inflammatory cell infiltration, periportal fibrosis, and/or cirrhosis. There is often evidence of biliary epithelial cell destruction. Recent studies have also shown evidence for autophagosomes and mitochondrial injury16 (Figure 68-6).
TREATMENT There is no specific therapy for a1AT deficiency-associated liver disease and so clinical care involves avoidance of cigarette smoking to prevent exacerbation of destructive lung disease/emphysema, supportive management of symptoms caused by liver dysfunction, and prevention of complications of liver disease. Cigarette smoking markedly accelerates the lung disease associated with a1AT deficiency, reduces the quality of life, and significantly shortens the longevity of these patients.48 Progressive liver dysfunction and failure in children with a1AT deficiency has been managed successfully with liver transplantation, with survival rates well over 92% for 5 years.49,50 Nevertheless, a number of homozygotes with severe liver disease, even cirrhosis or portal hypertension, may have a relatively low rate of disease progression and lead a relatively normal life for extended periods. With the availability of living-related donor transplantation techniques it may be possible to observe these patients for some time before transplantation becomes necessary. Patients with a1AT deficiency and emphysema have been treated with purified plasma or recombinant a1AT administered intravenously or by means of aerosol as replacement therapy.51 This therapy is associated with improvement in serum and bronchoalveolar lavage fluid a1AT concentrations and neutrophil elastase inhibitory capacity in lavage fluid without significant side effects.
Although results of initial studies have suggested that there is a slower decline in forced expiratory volume in patients undergoing replacement therapy, this only occurred in a subgroup of patients and the study was not randomized.52 This therapy is designed for established and progressive emphysema. Protein replacement therapy is not being considered for patients with liver disease because there is no evidence that deficient serum levels of a1AT play a role in the development of liver injury. A number of patients with severe emphysema from a1AT deficiency are being treated with lung transplantation. Over a 13-year experience, 86 patients with a1AT deficiency underwent lung transplantation in St. Louis with a ~60% 5-year survival.53 A number of novel strategies for chemoprophylaxis and treatment of a1AT deficiency have been proposed recently. Because treatment of the PiZ mouse with cyclosporin A resulted in reduced hepatic mitochondrial damage, absence of hepatic caspase-3 activation and improved tolerance of starvation,13 cyclosporin A and other drugs that prevent the mitochondrial permeability transition appear to be candidates for treatment of a1AT deficiency-associated liver disease. This strategy is particularly attractive because it involves a mechanism of action at a distal step in the pathway of liver damage that is still effective, even though the primary pathologic phenomenon, mutant a1ATZ, continues to accumulate in the ER. Several studies have shown that a class of compounds called chemical chaperones can reverse the cellular mislocalization or misfolding of mutant plasma membrane, lysosomal, nuclear, and cytoplasmic proteins including CFTRaF508, prion proteins, mutant aquaporin molecules associated with nephrogenic diabetes insipidus, and mutant galactosidase A associated with Fabry disease.54–56 These compounds include non-specific chaperones such as glycerol, trimethylamine oxide, deuterated water and 4-phenylbutyric acid (PBA) as well as specific chaperones that have antagonistic or agonistic pharmacological bases. Burrows et al. found that glycerol and PBA mediate a marked increase in the secretion of a1ATZ in a model cell line.57 Moreover, oral administration of PBA was well tolerated by the PiZ mouse and consistently mediated an increase in blood levels of human a1AT, reaching 20–50% of the levels present in PiM mice and normal humans. PBA did not affect the synthesis or intracellular degradation of a1ATZ. The a1ATZ secreted in the presence of PBA was functionally active, in that it could form an inhibitory complex with neutrophil elastase. Because PBA has been used safely as an oral drug in children with urea cycle disorders, it was considered an excellent candidate for chemoprophylaxis of a1AT deficiency. However, PBA did not mediate an increase in blood levels of a1AT in a recent pilot human trial.58 Studies of other candidate chemical chaperones will be important because the approach has the potential both to prevent liver damage by reducing the burden of a1ATZ that accumulates in the cell and by increasing the secretion of a1AT and therein the amount of a1AT that reaches the lung to inhibit elastases. It also appears that several imino sugar compounds may potentially prove useful for the chemoprophylaxis of liver and lung disease in a1AT deficiency. These compounds are designed to interfere with oligosaccharide side-chain trimming of glycoproteins and are now being examined as potential therapeutic agents for viral hepatitis and other types of infection.59,60 We have examined several of these compounds initially to determine the effect of inhibiting glucose or
1263
Section XII. Inherited and Pediatric Diseases of the Liver
A
D
B
C
E
Figure 68-6. Electron micrographs of liver biopsy from an a1-antitrypsin (a1-AT)-deficient patient. (A) Rough endoplasmic reticulum (rER) from a normal liver cell; (B) rER distended with proteinaceous material; (C) autophagic vacuole pinching-off rER; (D) early autophagosomes (AVi) adjacent to rER; (E) degradative autophagosomes (AVd) adjacent to rER. M, mitochondrion. (Reproduced from Teckman JH, Perlmutter DH. Retention of the mutant secretory protein a1antitrypsin Z in the endoplasmic reticulum induces autophagy. Am J Physiol 2000; 279:G961–G974, with permission.)
mannose trimming from the carbohydrate side chain of mutant a1ATZ on its fate in the ER, but to our surprise we found that one glucosidase inhibitor, castanospermine and two mannosidase I inhibitors, kifunensine and deoxymannojirimicin, actually mediate increased secretion of a1ATZ.61 Importantly, the a1ATZ that is secreted is partially functional as an elastase inhibitor. Kifunensine and deoxymannojirimicin are less attractive candidates for chemoprophylactic trials because they delay degradation of a1ATZ in addition to increasing its secretion and therefore have the potential to exacerbate susceptibility to liver disease. However, castanospermine has no effect on the degradation of a1ATZ and, therefore, may be targeted for development as a chemoprophylactic agent. Alternative strategies for at least partial correction of a1AT deficiency may result from further studies of the fate of a1ATZ in the
1264
ER. For instance, delivery of synthetic peptides to the ER to insert into the gap in the A sheet or into a particular hydrophobic pocket of the a1ATZ62 and prevent polymerization of a1ATZ might result in release into the extracellular fluid and prevent accumulation in the ER. Although it is not yet entirely clear, there is some evidence from studies on the assembly of major histocompatibility complex class I molecules that synthetic peptides may be delivered to the ER from the extracellular medium of cultured cells.63 There is also evidence that certain molecules may be transported retrograde to the ER by receptor-mediated endocytosis.64 Even if polymerization results from, rather than causes, the secretory defect, this strategy may be effective if peptide insertion leads to a change in conformation that is associated with translocation competence. Second, elucidation of the biochemical mechanism by which abnormally
Chapter 68 a1-ANTITRYPSIN DEFICIENCY
folded a1ATZ undergoes intracellular degradation might allow the development of pharmacological agents that can manipulate this degradative system, such as agents like interferon-a that enhance proteosomal activity, for prophylactic application to the subpopulation of deficient individuals predisposed to liver disease. Replacement of a1AT by means of somatic gene therapy has been discussed in the literature for a number of years.65 This strategy is potentially less expensive than replacement therapy with purified protein and may avert the need for weekly or even monthly administration. As a form of replacement therapy, however, this strategy will only be useful for lung disease in a1AT deficiency. There are still significant issues that need to be addressed before gene therapy becomes a realistic alternative.66 The most important prerequisite will be demonstration that replacement therapy with purified plasma a1AT is truly associated with an ameliorative effect. Several novel types of gene therapy, such as repair of mRNA by transsplicing ribozymes,67,68 chimeric RNA/DNA oligonucleotides,69–71 triplex-forming oligonucleotides,72 small fragment homologous replacement,73 or by RNA silencing74,75 are theoretically attractive alternative strategies for the prevention of liver disease associated with a1AT deficiency because they would prevent the synthesis of mutant a1ATZ and ER retention. Other studies have shown that transplanted hepatocytes can repopulate the diseased liver in several mouse models76,77 including a mouse model of the childhood metabolic liver disease hereditary tyrosinemia. Replication of transplanted hepatocytes only occurs when there is injury and/or regeneration in the liver. The results provide evidence that it may be possible to use hepatocyte transplantation techniques to treat hereditary tyrosinemia and, perhaps, other metabolic liver diseases in which the underlying defect is cell-autonomous. For instance, a1AT deficiency involves a cellautonomous defect and would be an excellent candidate for this strategy.
REFERENCES 1. de Serres FJ, Blanco I, Fernandez-Bustillo E. Genetic epidemiology of alpha-1 antitrypsin deficiency in North America and Australia/New Zealand: Australia, Canada, New Zealand and the United States of America. Clin Genet 2003; 64:382–397. 2. Sveger T. Liver disease in a1-antitrypsin deficiency detected by screening of 200 000 infants. N Engl J Med 1976; 294:1316–1321. 3. Sveger T. The natural history of liver disease in a1-antitrypsin deficient children. Acta Paediatr Scand 1995; 77:847–851. 4. DeMeo DL, Silverman EK. Alpha1-antitrypsin deficiency. 2: genetic aspects of alpha(1)-antitrypsin deficiency: phenotypes and genetic modifiers of emphysema risk. Thorax 2004; 59:259–264. 5. Carrell RW, Lomas DA. Conformational disease. Lancet 1997; 350:134–138. 6. Lomas DA, Evans DL, Finch JT, et al. The mechanism of Z a1antitrypsin accumulation in the liver. Nature 1992; 357:605–607. 7. Lin L, Schmidt B, Teckman J, et al. A naturally occurring nonpolymerogenic mutant of a1-antitrypsin characterized by prolonged retention in the endoplasmic reticulum. J Biol Chem 2001; 276:33893–33898. 8. Schmidt B, Perlumutter DH. GRP78, GRP94 and GRP170 interact with a1AT mutants that are retained in the endoplasmic reticulum. Am J Physiol 2005; 289:6444–6455.
9. Kim J, Lee KN, Yi G-S, et al. A thermostable mutation located at the hydrophobic core of a1-antitrypsin suppresses the folding defect of the Z-type variant. J Biol Chem 1995; 270:8597–8601. 10. Sidhar SK, Lomas DA, Carrell RW, et al. Mutations which impede loop-sheet polymerization enhance the secretion of human a1-antitrypsin deficiency variants. J Biol Chem 1995; 270:8393–8396. 11. Kang HA, Lee KN, Yu M-H. Folding and stability of the Z and Siiyama genetic variants of human a1-antitrypsin. J Biol Chem 1997; 272:510–516. 12. Teckman JH, An JK, Blomenkamp K, et al. Mitochondrial autophagy and injury in the liver in alpha 1-antitrypsin deficiency. Am J Physiol Gastrointest Liver Physiol 2004; 286:G851–G862. 13. Rudnick DA, Liao Y, An JK, et al. Analyses of hepatocellular proliferation in a mouse model of alpha-1-antitrypsin deficiency. Hepatology 2004; 39:1048–1055. 14. Hidvegi T, Schmidt B, Hale P, et al. Accumulation of mutant a1ATZ in the ER activates caspases-4 and -12, NFkB and BAP31 but not the unfolded protein response. J Biol Chem (in press). 15. Wang B, Nguyen M, Breckenridge DG, et al. Uncleaved BAP31 in association with A4 protein at the endoplasmic reticulum is an inhibitor of fas-initiated release of cytochrome c from mitochondria. J Biol Chem 2002; 278:14461–14468. 16. Teckman JH, Perlmutter DH. Retention of the mutant secretory protein a1-antitrypsin Z in the endoplasmic reticulum induces autophagy. Am J Physiol 2000; 279:G961–G974. 17. Teckman JH, An JK, Loethen S, et al. Effect of fasting on liver in a mouse model of a1-antitrypsin deficiency: constitutive activation of the autophagic response. Am J Physiol 2003; 283:G1117–G1124. 18. Geller SA, Nichols WS, Kim S, et al. Hepatocarcinogenesis is the sequel to hepatitis in Z#2 a1-antitrypsin transgenic mice: histopathological and DNA ploidy studies. Hepatology 1994; 9:389–397. 19. Liang X, Jackson S, Seaman M, et al. Induction of autophagy and inhibition of tumorigenesis by beclin 1. Nature 1999; 402:672–676. 20. Vogel A, van den Berg I, Al-Dhalimy M, et al. Chronic liver disease in murine hereditary tyrosinemia type 1 induces resistance to cell death. Hepatology 2004; 39:433–443. 21. Wu T, Whitman I, Molmenti E, et al. A lag in intracellular degradation of mutant a1-antitrypsin correlates with the liver disease phenotype in homozygous PiZZ a1-antitrypsin deficiency. Proc Natl Acad Sci 1994; 91:9014–9018. 22. Qu D, Teckman JH, Omura S, et al. Degradation of mutant secretory protein, a1-antitrypsin Z, in the endoplasmic reticulum requires proteasome activity. J Biol Chem 1996; 271:22791–22795. 23. Teckman JH, Gilmore R, Perlmutter DH. The role of ubiquitin in proteasomal degradation of mutant a1-antitrypsin Z in the endoplasmic reticulum. Am J Physiol 2000; 278:G39–G48. 24. Cabral CM, Liu Y, Moremen KW, et al. Organizational diversity among distinct glycoprotein endoplasmic reticulum-associated degradation programs. Mol Biol Cell 2002; 13:2639–2650. 25. Hope PL, Hall MA, Millward-Sadler GH, et al. Alpha-1antitrypsin deficiency presenting as a bleeding diathesis in the newborn. Arch Dis Child 1982; 57:68–70. 26. Hadchouel M, Gautier M. Histopathologic study of the liver in the early cholestatic phase of alpha-1-antitrypsin deficiency. J Pediatr 1976; 89:211–215. 27. Grishan FR, Gray GF, Green HL. a1-antitryspin deficiency presenting with ascites and cirrhosis in the neonatal period. Gastroenterology 1983; 85:435–438. 28. Eriksson S, Carlson J, Velez R. Risk of cirrhosis and primary liver cancer in a1-antitrypsin deficiency. N Engl J Med 1986; 314:736–739.
1265
Section XII. Inherited and Pediatric Diseases of the Liver
29. Volpert D, Molleston JP, Perlmutter DH. a1-antitrypsin deficiency-associated liver disease may progress slowly in some children. J Pediatr Gastroenterol Nutr 2001; 32:265–269. 30. Nebbia G, Hadchouel M, Odievre M, et al. Early assessment of evolution of liver disease associated with a1-antitrypsin deficiency in childhood. J Pediatr 1983; 102:661–665. 31. Ibarguen E, Gross CR, Savik SK, et al. Liver disease in a1antitrypsin deficiency: prognostic indicators. J Pediatr 1990; 117:864–870. 32. Hodges JR, Millward-Sadler GH, Barbatis C, et al. Heterozygous MZ a1-antitrypsin deficiency in adults with chronic active hepatitis and cryptogenic cirrhosis. N Engl J Med 1981; 304:357–360. 33. Gradziadei IW, Joseph JJ, Wiesner RH, et al. Increased risk of chronic liver failure in adults with heterozygous a1-antitrypsin deficiency. Hepatology 1998; 28:1058–1063. 34. Propst T, Propst A, Dietze O, et al. High prevalence of viral infection in adults with homozygous and heterozygous a1antitrypsin deficiency and chronic liver disease. Ann Intern Med 1992; 117:641–645. 35. Mahadeva R, Chang W-S, Dafforn T, et al. Heteropolymerization of S, I, and Z 1-antitrypsin and liver cirrhosis. J Clin Invest 1999; 103:999–1006. 36. Teckman JH, Perlmutter DH. The endoplasmic reticulum degradation pathway for mutant secretory proteins 1-antitrypsin Z and S is distinct from that for an unassembled membrane protein. J Biol Chem 1996; 271:13215–13220. 37. Curiel DT, Holmes MD, Okayama H, et al. Molecular basis of the liver and lung disease associated with the a1-antitrypsin deficiency allele Mmalton. J Biol Chem 1989; 264:13938– 13945. 38. Reid CL, Wiener GJ, Cox DW, et al. Diffuse hepatocellular dysplasia and carcinoma associated with the Mmalton variant of a1antitrypsin. Gastroenterology 1987; 93:181–187. 39. Lomas DA, Elliott PR, Sidhar SK, et al. a1-antitrypsin Mmalton (Phe52 deleted) forms loop-sheet polymers in vivo: evidence for the C-sheet mechanism of polymerization. J Biol Chem 1995; 270:16864–16870. 40. Perlmutter DH. Alpha-1-antitrypsin deficiency. In: Schiff ER, Sorrell MF, Maddrey WD, eds: Schiff ’s disease of liver, 9th edn, vol. 2. Philadelphia: Lippincott-Raven, 2002:1206– 1229. 41. von Schonfeld J, Breuer N, Zotz R, et al. Liver function in patients with pulmonary emphysema due to severe alpha-1antitrypsin deficiency (PIZZ). Digestion 1996; 57:165–169. 42. Steiner SJ, Gupta SK, Croffie JM, et al. Serum levels of a1antitrypsin predict phenotypic expression of the a1-antitrypsin gene. Dig Dis Sci 2003; 48:1793–1796. 43. Johnson K, Alton HM, Chapman S. Evaluation of mebrofenin hepatoscintigraphy in neonatal-onset jaundice. Pediatr Radiol 1998; 28:937–941. 44. Nord KS, Saad S, Joshi VV, et al. Concurrence of a1-antitrypsin deficiency and biliary atresia. J Pediatr 1987; 111:416–418. 45. Mowat AP. Hepatitis and cholestasis in infancy: intrahepatic disorders. In: Mowat AP, ed: Liver disorders in children. London: Butterworths, 1982:50. 46. Qizibash A, Yong-Pong O. Alpha-1-antitrypsin liver disease: differential diagnosis of PAS-positive diastase-resistant globules in liver cells. Am J Clin Pathol 1983; 79:697–702. 47. Yunis EJ, Agostini RM, Glew RH. Fine structural observations of the liver in a1-antitrypsin deficiency. Am J Pathol 1976; 82:265–286. 48. Wilson-Cox D. Alpha-1-antitrypsin deficiency. In Scriver CB, Beaudet AL, Sly WS, et al, eds: The metabolic basis of inherited disease. New York: McGraw-Hill, 1989:2409–2437. 49. Francavilla R, Castellaneta S, Hadzic N, et al. Prognosis of alpha1-antitrypsin deficiency-related liver disease in the era of pediatric liver transplantation. J Hepatol 2000; 32:986–992.
1266
50. Kayler LK, Merion RM, Lee S, et al. Long-term survival after liver transplantation in children with metabolic disorders. Pediatr Transplant 2002; 6:295–300. 51. Wewers MD, Casolaro MA, Sellers SE, et al. Replacement therapy for alpha 1-antitrypsin deficiency associated with emphysema. N Engl J Med 1987; 316:1055–1062. 52. The Alpha-1-Antitrypsin Deficiency Registry Study Group. Survival and FEV1 decline in individuals with severe deficiency of a1-antitrypsin. Am J Respir Crit Care Med 1998; 158:49–59. 53. Cassivi SD, Meyers BF, Battafarano RJ, et al. Thirteen-year experience in lung transplantation for emphysema. Ann Thorac Surg 2002; 74:1663–1670. 54. Sato S, Ward CL, Krouse ME, et al. Glycerol reverses the misfolding phenotype of the most common cystic fibrosis mutation. J Biol Chem 1996; 271:635–638. 55. Tamarappoo BK, Verkman AS. Defective aquaporin-2 trafficking in nephrogenic diabetes insipidus and correction by chemical chaperones. J Clin Invest 1998; 101:2257–2267. 56. Brown CR, Hong-Brown LQ, Welch WJ. Correcting temperaturesensitive protein folding defects. J Clin Invest 1997; 99:1432–1444. 57. Burrows JA, Willis LK, Perlmutter DH. Chemical chaperones mediate increased secretion of mutant alpha 1-antitrypsin (a1-AT) Z: a potential pharmacological strategy for prevention of liver injury and emphysema in a1-AT deficiency. Proc Natl Acad Sci USA 2000; 97:1796–1801. 58. Teckman JH. Lack of effect of oral 4-phenylbutyrate on serum alpha-1-antitrypsin in patients with a1-antitrypsin deficiency: a preliminary study. J Pediatr Gastroenterol Nutr 2004; 39:34–37. 59. Jacob GS. Glycosylation inhibitors in biology and medicine. Curr Opin Struct Biol 1995; 5:605–611. 60. Zitzmann N, Mehta AS, Carrouee S, et al. Imino sugars inhibit the formation and secretion of bovine viral diarrhea virus, a pestivirus model of hepatitis C virus: implications for the development of broad spectrum anti-hepatitis virus agents. Proc Natl Acad USA 1999; 96:11878–11882. 61. Marcus NY, Perlmutter DH. Glucosidase and mannosidase inhibits mediate increased secretion of mutant a1-antitrypsin Z. J Biol Chem 2000; 275:1987–1992. 62. Elliott PR, Abrahams J-P, Lomas DA. Wild-type a1-antitrypsin is in the canonical inhibitory conformation. J Mol Biol 1998; 275:419–425. 63. Day PM, Yewdell JW, Porgador A, et al. Direct delivery of exogenous MHC class I molecule-binding oligopeptides to the endoplasmic reticulum of viable cells. Proc Natl Acad Sci USA 1997; 94:8064–8069. 64. Lord JM, Roberts LM. Toxin entry: retrograde transport through the secretory pathway. J Cell Biol 1998; 140: 733–736. 65. Crystal RG. Alpha-1-antitrypsin deficiency, emphysema and liver disease: genetic basis and strategies for therapy. J Clin Invest 1990; 95:1343–1352. 66. Anderson WF. The current status of clinical gene therapy. Hum Gen Ther 2002; 13:1261–1262. 67. Long MB, Jones JP, Sullenger BA, et al. Ribozyme-mediated revision of RNA and DNA. J Clin Invest 2003; 112:312– 318. 68. Garcia-Blanco MA. Messenger RNA reprogramming by spliceosome-mediated RNA trans-splicing. J Clin Invest 2003; 112:474–480. 69. Kren BT, Bandyopadhyay P, Steer CJ. In vivo site-directed mutagenesis of the factor IX gene by chimeric RNA/DNA oligonucleotides. Natl Med 1998; 4:285–290. 70. Metz R, Dicola M, Kurihara T, et al. Mode of action of RNA/DNA oligonucleotides. Chest 2002; 121:915–925. 71. Kmiec EB. Targeted gene repair – in the arena. J Clin Invest 2003; 112:632–626.
Chapter 68 a1-ANTITRYPSIN DEFICIENCY
72. Seidman MM, Glazer PM. The potential for gene repair via triple helix formation. J Clin Invest 2003; 114:487–494. 73. Gruenert DC, Bruscia E, Novelli G, et al. Sequence-specific modification of genomic DNA by small DNA fragments. J Clin Invest 2003; 112:637–641. 74. Davidson BL. Hepatic diseases – hitting the target with inhibitor RNAs. N Engl J Med 2003; 349:2357–2359. 75. Rubinson DA, Dillon CP, Kwiatkowski AV, et al. A lentivirusbased system to functionally silence genes in primary mammalian
cells, stem cells and transgenic mice by RNA interference. Nat Gen 2003; 33:401–406. 76. Rhim JA, Sandgen EP, Degen JL, et al. Replacement of diseased mouse liver by hepatic cell transplantation. Science 1994; 263:1149–1152. 77. Overturf K, Al-Dhalimy M, Tanguay R, et al. Hepatocytes corrected by gene therapy are selected in vivo in a murine mouse model of hereditary tyrosinemia type I. Nat Genet 1996; 12:226–273.
1267
Section XII: Inherited and Pediatric Diseases of the Liver
69
INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY Fayez K. Ghishan Abbreviations ADP adenosine diphosphate ALT alanine aminotransferase AMP adenosine monophosphate AST aspartate aminotransferase ATP adenosine triphosphate BSEP bile-salt export pump BSP sulfobromophthalein CFTR cystic fibrosis transconductor regulator CoA coenzyme A DHCA 3a,7a-dihydroxy-5b-cholestan-26-oic acid DHA docosahexaenoic acid
FAH FDPase GBE GGTP GSD H&E HFI MDR-3 NADH NADPH
fumaryl acetoacetate hydrolyase fructose-1,6-diphosphatase glycogen branching enzyme g-glutamyl transpeptidase glycogen storage disease hematoxylin and eosin hereditary fructose intolerance multidrug-resistant 3 nicotinamide-adenine dinucleotide nicotinamide-adenine dinucleotide phosphate
INTRODUCTION The liver is often affected by inborn errors of metabolism, but only a few of these injure the liver severely enough to cause permanent damage. Percutaneous liver biopsy has proved safe in infants and children, and this has allowed for the histologic and biochemical evaluation of liver specimens from living patients allowing for rapid progess in the study of metabolic diseases during the past few years. Advances in molecular genetics promise greater advances in diagnosis and treatment of metabolic illnesses as our understanding of the pathophysiology of such disorders improves. Table 69-1 lists the major metabolic diseases that involve the liver. Those marked with asterisks may lead to progressive disease and are discussed here or elsewhere in this book.
EVALUATION OF HEPATIC METABOLIC DISORDERS The clinical history and physical examination are the first essentials in evaluating infants and children with metabolic liver disorders. Of particular importance is a family history of any metabolic liver disease. Symptoms that may be associated with metabolic liver disorders are usually non-specific and include vomiting, diarrhea, jaundice, seizures, and abnormal urinary odor. Clinical findings include hepatosplenomegaly, hypotonicity or hypertonicity, coarse facial features, and respiratory distress. In general, storage diseases usually cause marked hepatomegaly. By contrast, disorders resulting in hepatocellular damage cause only modest hepatomegaly. The physical examination should include adequate ophthalmologic examina-
NTBC PAF-1 PAS PFIC Pi PRPP THCA TPN UDP
2-(2-nitro-4-trifluoro-methylbenzoyl)-1,3cyclohexanedione peroxisome assembly factor-1 periodic acid–Schiff progressive familial intrahepatic cholestasis inorganic phosphate phosphoribosyl pyrophosphate 3a,7a,12a-trihydroxy-5b-cholestan-26oic acid total parenteral nutrition uridine diphosphate
tion by slit lamp for corneal, lenticular, and retinal alterations. Psychomotor evaluation to detect developmental delays is important in identifying those diseases that may involve the central nervous system. Initial laboratory screening tests such as analysis of the urine for reducing substances may help in early diagnosis of galactosemia. The peripheral blood smear may reveal vacuolation of the lymphocytic cytoplasm, which signifies deposition of storage material. Skeletal X-rays may reveal changes consistent with certain storage diseases, as in the case of mucopolysaccharidosis. Confirmatory tests depend on assays of appropriate enzymes in tissues such as leukocytes, skin fibroblasts, intestine, and liver, and DNA analysis for mutations in those disorders with known genetic defects.
DISORDERS OF CARBOHYDRATE METABOLISM Three inborn errors in carbohydrate metabolism that may result in permanent liver damage are: (1) disorders of fructose metabolism; (2) disorders of galactose metabolism; and (3) certain of the glycogen storage diseases (GSDs) (Figure 69-1). These three abnormalities are discussed separately. Inborn errors resulting from abnormal metabolism of lipids are mostly untreatable at present, and most of the errors in protein metabolism that respond to treatment require relatively stringent dietary restrictions. In contrast, most errors in carbohydrate metabolism which respond favorably to treatment do so with relatively modest dietary restrictions. Liver diseases that occur in untreated patients with inborn errors of carbohydrate metabolism such as fructose intolerance and galactosemia are usually preventable. Although these particular conditions are relatively rare (about 1 in 30 000 live
1269
Section XII. Inherited and Pediatric Diseases of the Liver
Table 69-1. Inborn Errors of Metabolism Resulting in Injury to the Liver Inborn errors of carbohydrate metabolism Glycogen storage disease, types I–XIIa Galactosemiaa Fructose-1-phosphate aldolase deficiency Fructose-1,6-diphosphatase deficiency Inborn errors of protein metabolism Tyrosinemiaa Urea cycle enzymic defects Inborn errors of lipid metabolism Gaucher’s disease Niemann–Pick disease Gangliosidosis Acid cholesterol ester hydrolase deficiencya Wolman’s disease Cholesterol ester storage disease Lipodystrophy Inborn errors of mucopolysaccharide metabolism Inborn errors of porphyrin metabolism Protoporphyria (discussed in Chapter 73)a Inborn errors of bile acid metabolism: Progressive familial intrahepatic cholestasisa Type I (Byler’s disease)a Type II (Byler’s syndrome)a Type IIIa Hereditary lymphedema with recurrent cholestasis (Aagenaes’s syndrome) THCA syndrome Disorders of peroxisome biogenesis Zellweger’s syndromea Alagille’s syndrome (arteriohepatic dysplasia)a Inborn errors of copper metabolism Wilson’s disease (discussed in Chapter 67)a Unclassified Alpha1-antitrypsin deficiencya Cystic fibrosisa a
Diseases that lead to progressive disease and are discussed in this chapter or other sections of this book. THCA, 3a,7a,12a-trihydroxy-5b-cholestan-26-oic acid.
births for fructose intolerance and 1 in 20 000 live births for galactosemia), their combined incidence is similar to that of phenylketonuria (about 1 in 14 000 births in the USA). Considering the outcome in untreated patients and the simplicity of either measuring urinary reducing sugar or assaying the blood sample that is obtained to screen for phenylketonuria, it seems reasonable to implement the practice of routine screening of all newborn infants for both galactosemia and fructose intolerance.
DISORDERS OF FRUCTOSE METABOLISM FRUCTOSE PHOSPHATE ALDOLASE DEFICIENCY (HEREDITARY FRUCTOSE INTOLERANCE) AND FRUCTOSE-1,6-DIPHOSPHATASE DEFICIENCY Until the mid-1950s, the only identified defect in fructose metabolism was the benign disorder essential fructosuria.1 It was recog-
1270
nized in 1956, in some patients, that ingestion of fructose was followed by vomiting, severe hypoglycemia, and liver disease.2 A year later, this illness was characterized and named hereditary fructose intolerance (HFI).3 A third disorder of fructose metabolism was identified in 1970. It was associated with fasting-induced as well as diet-induced hypoglycemia, but more strikingly, both fasting and dietary fructose caused lactic acidosis.4 Each of these three disorders is distinct from the others both clinically and biochemically. Essential fructosuria is due to deficient activity of hepatic fructokinase; HFI is due to deficiency of fructose-1-phosphate aldolase; and the third condition results from deficiency of fructose-1,6diphosphatase (FDPase). Essential fructosuria does not cause liver injury. Patients with FDPase deficiency may show transient fatty infiltration of the liver. In contrast, liver injury may be a significant feature of HFI.
HEREDITARY FRUCTOSE INTOLERANCE In 1957, Froesch and colleagues described the typical syndrome of HFI in two siblings and two relatives.3 They recognized that the disorder, like essential fructosuria, was inherited and was associated with urinary excretion of fructose. They also realized that it was due to a different enzymatic defect and had different prognostic implications.
Molecular Basis of Hereditary Fructose Intolerance HFI occurs with a frequency of 1 in 20 000 individuals. It has a recessive mode of inheritance, and is caused by a deficiency of fructose-1-phosphate aldolase (aldolase B), which is normally present in the liver, kidneys, and small intestine. The enzymatic activities of the two other aldolase isoenzymes, aldolase A in muscle and aldolase C in the brain, are normal. The three isoenzymes are related and are derived from a single ancestral gene. The aldolase B gene has been sequenced and is mapped to human chromosome 9q13Æq32.5 The gene has nine exons encoding 364 amino acids5,6 The gene consists of 14 500 base pairs. The first mutation described was a GÆC transversion in exon 5, which resulted in an amino acid substitution (alanineÆproline) at position 149 of the protein within a region critical for substrate binding (A149P). The GÆC transversion created a new recognition site for the restriction enzyme Aha II.7 The alanine at position 149 is a conserved amino acid because it is present in the aldolase B gene of humans, rats, and chickens.8 The substitution of proline is likely to disrupt the spatial configuration of juxtaposed residues in aldolase B and adversely affect its catalytic activity. The mutation alanineÆproline was found in 67% of 50 HFI patients studied by Cross and colleagues.9 This mutation is encountered more frequently in patients from northern than from southern Europe. Several other mutations have been described, such as alanine 174Æaspartic acid (A174D) and asparagine 334Ælysine (N334K).10 Tolan and Brooks characterized the molecular defects in the aldolase B gene in 31 North Americans with HFI. Fifty-nine percent of mutant North American alleles were alanine 149Æproline. Alanine 174Æaspartic acid and asparagine 334Ælysine represented 11% and 2% of North American alleles, respectively. Nine subjects (32%) had HFI alleles that were not these common, missense mutations.11 So far, 25
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Site of metabolic block Omission in the metabolic sequence
Galactose
Galactose-1-phosphate
Glycogen
Glucose-1-phosphate
Glucose
Glucose-6-phosphate
Fructose-1,6-diphosphate
Fructose
Fructose-1-phosphate
Triose phosphate
Pyruvate
Lactate
Aerobic pathway Figure 69-1. Metabolic relationship between disorders in glycogen, galactose, glucose, and fructose metabolism. Solid rectangle indicates site of metabolic block; . . . indicates omission in the metabolic sequence.
mutations have been reported. The missense mutations could be classified into two groups: catalytic mutants with retained tetrameric structure but altered kinetic properties (W147R, R303W, and A337V) and structural mutations in which heterotetramers dissociate into subunits with impaired enzymatic activity (A149P, A174D, N334K).12
Clinical Features Patients with HFI may be extremely ill and may die after continuous exposure to fructose. However, affected patients are generally healthy and symptom-free so long as they do not ingest fructose or fructose-containing foods.13,14 For this reason, symptoms do not arise until breast milk or cow’s milk formulas are supplemented with fructose-containing foods. In fortunate cases, fructose is not introduced until after an affected infant is 5–6 months of age. By this time, the child is likely to associate nausea, vomiting, and symptoms of hypoglycemia with sweet-tasting food. In such cases, aversion to sweets is probably life-saving, and the diagnosis may go undetected until adulthood.15 When this occurs, the diagnosis may be suspected on the basis of a careful history that recognizes the extreme aversion to dietary “sweets.” The largest single collection of patients consists of 55 patients diagnosed between 1961 and 1977 as having HFI.16 Fifty of the patients had become symptomatic because of dietary fructose, and 5 were diagnosed shortly after birth because an older sibling of each infant was known to have HFI. Fourteen patients received fructose in their first feedings, and symptoms usually appeared within a few days. The remaining patients received a fructose-free diet (breast milk or cow’s milk formula). Their symptoms began immediately after introduction of dietary fructose or sucrose. Thirty-two
(64%) were diagnosed as having HFI at less than 6 months of age, 12 (24%) between 6 and 12 months of age, and 6 (12%) after 1 year of age. The younger patients were usually admitted to the hospital on an emergency basis with acute liver impairment, sepsis, bleeding diathesis, shock, or dehydration. Patients less than 6 months old developed a triad of jaundice, edema, and bleeding tendency. Older patients were admitted more often because of liver enlargement, ascites, or both. Vomiting and hepatomegaly were observed in all patients, and about half had anorexia, weight retardation, and bleeding tendency. About a third had jaundice, diarrhea, edema or ascites or both, and growth retardation. Thirteen of the 50 had developed an aversion to sweet foods. This aversion had developed as early as 3 months of age and in 2 patients had resulted in continued breast-feeding until 9 months of age. Vomiting and diarrhea in the young children were sometimes severe enough to cause dehydration. The laboratory findings were variable, but liver tests were severely deranged in the younger patients in this series.16 Deficiency of clotting factors and elevated alanine aminotransferase (ALT) were present in all but one of the patients less than 6 months old. Two patients also had serum albumin levels of 2.8 g/dl. Fifteen patients had aminoaciduria; the predominant amino acids were tyrosine and methionine in 3 of the 15. All patients showed complete resolution of laboratory abnormalities in response to removal of fructose from the diet during a succeeding 2-week period. Histologic abnormalities were found in the livers of all patients. Fibrosis without inflammation was present in either periportal or intralobular areas in most; all but 3 patients had some steatosis. These 3 patients were older and had voluntarily restricted themselves to diets with small amounts of fructose.16
1271
Section XII. Inherited and Pediatric Diseases of the Liver
Treatment of the symptomatic patients consisted of immediate cessation of fructose intake. Those with normal liver tests were given fructose-free diets normal in protein content. Infants who had acute liver dysfunction were given a glucose–electrolyte mixture intravenously and were given exchange transfusion when they had a serious bleeding tendency. Thereafter, a fructose-free, low-protein diet was fed; when the liver dysfunction had been corrected, a diet normal in protein was begun. Clinical and biochemical improvement was dramatic after exchange transfusion. Vomiting ceased immediately. The bleeding tendency disappeared in less than 24 hours, and renal tubular dysfunction disappeared within 3 days. All patients showed resolution of symptoms and normalization of laboratory findings within 2 weeks. Catch-up growth occurred within 2–3 years. The only persistent abnormality was hepatomegaly with steatosis, which was present from birth in the 5 patients restricted in dietary fructose.16 Although there were no deaths in the series reported by Odievre and co-workers,16 the continuous intake of fructose may cause death due to hypoglycemic seizures or progressive liver failure and inanition. The second child of such a family may profit from experience with the first child by more rapid recognition of the illness. With greater awareness, more cases of HFI in children are being diagnosed, and the condition is arrested by feeding fructoserestricted diets. One cautionary note is that a number of proprietary milks, primarily the soy-based formulas, contain sucrose as a significant source of the carbohydrate calories. The remaining carbohydrate is usually a glucose oligosaccharide. Hypoglycemia and seizures may not be a problem in affected infants fed these formulas because the remaining carbohydrate is glucose. The liver disease due to fructose ingestion may be progressive, however, and infants fed these formulas may simply fail to thrive, have hepatomegaly and vomit-
ing, or progress to chronic liver failure and death. Acute liver failure due to fructose intolerance is exceedingly unusual, and the absence of hepatomegaly in an infant who has severe liver disease and has reducing sugar in the urine should make one doubt the diagnosis of HFI. Follow-up studies of infants and recognition of older patients with HFI indicate a normal life expectancy. Patients retain their sensitivity to dietary fructose as adults, but the hypoglycemic response to fructose may be somewhat more delayed in adults than in infants (45–60 minutes in infants; 60–90 minutes in adults).14 The sensitivity to fructose may be life-threatening for adult patients. For example, patients with known HFI have been given sorbitol intravenously after surgery. Because sorbitol is metabolized to fructose, the patients died of complications from the sorbitol infusion.17,18
Biochemical Characteristics The clinical and biochemical abnormalities seen in patients with HFI result from decreased activity of hepatic fructose-1-phosphate aldolase (aldolase B). This enzyme is normally present in the liver, renal tubular cells, and intestinal mucosa.19 It catalyzes the conversion of fructose-1-phosphate to D-glyceraldehyde and dihydroxyacetone phosphate (reaction [1]). The metabolic consequences of this enzymatic deficiency are accumulation of large amounts of fructose-1-phosphate in the liver and depletion of inorganic phosphate (Pi) and adenosine triphosphate (ATP). The inability to metabolize fructose-1-phosphate in cells of affected patients leads to sequestration of large amounts of Pi. One of the many effects secondary to this sequestration of Pi is an inability to regenerate ATP, a process that depends on the presence of Pi. The clinical and laboratory features of HFI can be understood on the basis of this simple scheme (Figure 69-2).
Glucose-6-phosphate Hexose monophosphate shunt Fructose-6-phosphate Phosphoribose pyrophosphate Fructose-1,6-diphosphate 4 Fructose
Fructose-1-phosphate
ATP
ADP
2
Purine biosynthesis
Triose
Cellular HPO–4
1 ATP depletion
3
Purine degradation
Xanthine
Uric acid
Figure 69-2. Mechanism of hyperuricemia and hypophosphatemia in hereditary fructose intolerance. After fructose intake, there is rapid phosphorylation to fructose-1-phosphate, causing adenosine triphosphate (ATP) depletion (1) because of the aldolase block. The accumulated fructose-1-phosphate inhibits the aldolase step, preventing ATP generation from anaerobic glycolysis. Phosphate (HOP4–) is not released from the sugar, causing depletion of intracellular phosphate (2). Low ATP and HOP4– levels favor degradation of preformed purines to uric acid (3). There is a compensatory increase in purine biosynthesis (4). Solid rectangle indicates site of metabolic block. ADP, adenosine diphosphate.
1272
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Patients with HFI have been shown to have levels of activity of fructose-1-phosphate aldolase ranging from 0 to 12% of normal.14,20,21 In addition, most patients have reduced levels of activity of hepatic fructose-1,6-diphosphate aldolase ranging from 25 to 85% of normal.20,22 The differential between the activities of the two aldolase enzymes suggests that they are separate protein moieties. However, Gurtler and Leuthardt have crystallized human liver aldolase and have shown that both enzymatic activities are attributable to a single liver aldolase.23 In addition, slight alterations of the aldolase molecule, such as splitting off an end-terminal residue, may change the ratio of its affinity for fructose-1-phosphate or fructose1,6-diphosphate.24,25 Patients with HFI produce a protein that has the immunologic properties of fructose-1-phosphate aldolase but is biologically inactive.26 On the basis of these findings, it seems probable that a mutation of the structural gene is responsible for the enzyme defect in HFI. More recent studies suggest that the mutation in HFI affects aldolase B function by decreasing substrate affinity, maximal velocity, and/or enzyme activity.27 Accumulation of fructose-1-phosphate apparently causes the major manifestations of the disease through inhibition of other enzymatic reactions. Two metabolic pathways studied most extensively are gluconeogenesis and glycogenolysis. Their inhibition by fructose1-phosphate explains fructose-induced hypoglycemia. Concentrations of fructose-1-phosphate in excess of 10 mmol/l completely inhibit fructose-1,6-diphosphate aldolase activity in vitro.28 This finding suggests that fructose-1-phosphate inhibits gluconeogenesis at this enzymatic step. Inhibition at this site is further supported by the finding that fructose-induced hypoglycemia is not prevented by simultaneous infusions of gluconeogenic precursors such as dihydroxyacetone or glycerol. In addition, liver specimens from patients with HFI do not form glucose from 14C-glycerol when fructose is present, whereas oxidation of 14C-glycerol is apparently unaffected by fructose (reaction [2]).29 Patients with HFI apparently also have inhibition of glycogenolysis after fructose intake. This inhibition occurs above the level of phosphoglucomutase. In support of this idea was the finding that when galactose is administered together with fructose, hypoglycemia is less pronounced and does not last as long as when fructose is given alone (reaction [3]).13 Thus, a defect in the phosphorylation of glycogen to glucose-1-phosphate is incriminated. Several studies of normal liver indicate that depletion of Pi as well as accumulation of fructose-1-phosphate may contribute to an almost complete failure of glycogen mobilization.30–32 In addition, depletion of intracellular ATP levels may contribute to the lack of glycogen degradation to glucose-1-phosphate.32–34 A variant of fructose intolerance has been described in which the red cell galactose-1-phosphate uridyl transferase activity was normal but galactose as well as fructose caused hypoglycemia.35 The nature of this finding is unclear, because a re-evaluation of these patients 12 years later showed normal blood glucose responses to fructose and galactose.36 Results of studies by Schwartz and colleagues suggest that newborn infants delivered at term have less capacity for fructose metabolism in the first few days of life as compared with later in life.37 This is believed to be caused by the immaturity of the enzymes for handling fructose. In their studies, a rapid infusion of
fructose caused a prompt but transient decrease in blood glucose and suppressed the glucagon-induced elevation of blood glucose. Although hepatic aldolase was not measured, the findings suggest that until further studies are completed, the use of fructose or sorbitol as a calorie source (as in total parenteral nutrition (TPN)) for term infants in the first few days of life may not be justified. Enzymatic maturation may take longer in premature infants, although definitive studies have not been reported.
Laboratory Features The primary laboratory features of HFI are fructose-induced hypoglycemia and hypophosphatemia and/or chronic liver disease. In addition, serum and urinary urate levels may be increased.
HYPOGLYCEMIA Hypoglycemia induced by dietary or intravenous fructose is a characteristic of the illness. The hypoglycemia is not due to excess circulating insulin.28,38 Sorbitol provokes hypoglycemia before substantial amounts of fructose are released into the circulation and is evidence against a direct effect of fructose on blood glucose levels.14 More likely, the adverse effects of fructose result from intracellular accumulation of a fructose metabolite, such as fructose-1phosphate.28,39 This metabolite impairs both gluconeogenesis and glycogenolysis (reactions [2] and [3]). Studies with 14C-glucose indicate a complete cessation of hepatic glucose release after fructose infusion.38 Also, glucagon does not increase blood glucose after fructose-induced hypoglycemia, even in the presence of normal to slightly elevated hepatic glycogen content.40
HYPOPHOSPHATEMIA This is the second prominent feature of fructose-induced hypoglycemia. The reduction of inorganic phosphorus precedes that of glucose and may be the only abnormal finding when a small dose of fructose is administered.14 Hypophosphatemia appears to be a consequence of binding and sequestration of phosphorus in the form of fructose-1-phosphate within the hepatocytes.39,41 The first step in fructose metabolism is phosphorylation of the sugar by ATP. With large doses of fructose, ATP is depleted rapidly. With deficient activity of aldolase, as occurs in HFI, Pi is not released back into the cell. To compensate, phosphate from the serum is sequestered by the liver, with a resulting reduction in available circulating phosphate levels.14 Phosphorylation of fructose decreases intracellular phosphate in normal individuals, but the phosphate sequestered in normal liver is made available by further metabolism of fructose-1phosphate. Thus, changes in serum levels of phosphate are extremely transient in a normal individual and depend on the amount of fructose ingested.
HEPATIC ENZYME ELEVATION This appears to be the direct effect of increased hepatic fructose-1phosphate levels. Within 11/2 hours of a large dose of fructose, aminotransferase levels may increase more than twofold.34 The mechanism of liver cell damage is not clear, but it may result from a combination of depletion of ATP and a direct toxic effect of elevated levels of the phosphorylated hexose.
1273
Section XII. Inherited and Pediatric Diseases of the Liver
HYPERURICEMIA AND INCREASED URATE EXCRETION These conditions appear to result from depletion of intracellular ATP and Pi. This depletion of ATP and Pi increases the rate of purine degradation to uric acid (Figure 69-2).32,33 Other laboratory findings are less consistent. Some patients show substantial decreases in serum potassium and increases in serum magnesium after fructose intake.14,40 Some have increases in serum lactate and pyruvate levels.13,14,20 These changes appear to be related to the extent of liver damage and the severity of hypoglycemia. Granulocytosis may be noted with chronic fructose ingestion. A galactose infusion may relieve the hypoglycemia, but this is also an inconsistent finding. As blood glucose declines after fructose ingestion, insulin and insulin-like activity decrease and levels of glucagon, epinephrine, and growth hormone increase. In response to these hormonal changes, the non-esterified fatty acids in plasma increase more than twofold, a response not observed in normal subjects.13,20,42,43 Renal tubular acidosis and a Fanconi-like syndrome with renal tubular reabsorptive defects have been reported.16,44,45 In one patient, renal tubular acidosis persisted despite restriction of dietary fructose.15 The renal tubular acidosis is normalized in most patients as soon as fructose intake ceases.46,47 Fructose-1-phosphate aldolase is normally present in the renal tubules but it is absent in patients with HFI. Hence, the transient renal disturbance in affected patients may be due to accumulation of fructose-1-phosphate in renal tubular cells after fructose intake.19
Pathology Within 11/2–2 hours of a single dose of fructose are “glycogenassociated membrane arrays” and cytolysosomes in various stages of development are seen in the liver. The latter may represent lysosomes ingesting the accumulated fructose-1-phosphate in an attempt to get rid of it by acid hydrolysis. With chronic ingestion of fructose, the primary histologic changes are lipid accumulation and varying stages of hepatocellular necrosis and bile duct proliferation. Liver disease may progress to cirrhosis with impairment of liver-dependent coagulation factors.48 Despite extremely severe hepatic disease, the liver shows a remarkable ability to regenerate once dietary fructose is excluded. For example, a 3-month-old child who had cirrhosis, hypoalbuminemia, hypoprothrombinemia, and ascites had normal liver function and disappearance of ascites after 3 weeks of a fructose-free diet. Except for slight fibrosis, the hepatic architecture was normal 3 years later.40 The pathogenesis of acute and chronic liver cell injury following fructose intake has not been studied in detail. By analogy with galactosemia, a six-carbon sugar phosphorylated at carbon 1 may have a direct cytotoxic effect, whereas phosphorylation of the carbon-6 position apparently has no obvious hepatic toxicity. In addition, the severe alterations in energy metabolism seen in patients with HFI may have a role in derangement of liver cell function and may thereby result in the pathologic changes observed after ingestion of dietary fructose. The reason for lipid accumulation is also unclear, but it may represent general cellular dysfunction seen in a number of unrelated conditions.13,20,42,43 Although renal function may be severely impaired, little histologic change occurs in the kidneys other than some increase in medullary
1274
lipid. The teeth of patients who have HFI are unusually free of caries. This has been taken to indicate that fructose and sucrose are important cariogenic substrates.49 Despite the recurrent episodes of hypoglycemia and seizures that are common in undiagnosed cases, the brain is remarkably free of abnormalities. In contrast to patients with galactosemia, who commonly show psychomotor retardation, HFI patients surviving even the most severe forms of liver disease appear to have normal intelligence after being given fructose-free diets.
Genetics The genetic findings are compatible with an autosomal recessive trait. Levels of hepatic fructose-1-phosphate aldolase activity in five parents of patients with HFI were normal.50 In addition, carriers usually metabolize substantial fructose loads without difficulty. This is in contradistinction to carriers of galactosemia, who metabolize galactose at slower rates than normal and who may develop lenticular opacification with chronic galactose ingestion. As already mentioned, the molecular basis of HFI is missense mutations in the gene. Testing for these mutations in amplified DNAs by the polymerase chain reaction with a limited panel of allele-specific oligonucleotides identifies more than 95% of patients. A reverse dot blot method is available as a screening tool, and can detect the two most common mutations (A149P and A174D).51
Differential Diagnosis During infancy, various conditions may be associated with hypoglycemia,52,53 but most of them are associated with fasting. Hypoglycemia after eating should be a clue to the possibility of HFI, particularly in the presence of urinary reducing sugar. Other diseases that are associated with hypoglycemia follow ingestion of food include deficiency of fructose diphosphatase, galactosemia, and leucine intolerance. In addition, premature infants may have transient fructose intolerance due to immaturity of hepatic aldolase activity.37 Some patients with Tay–Sachs disease have deficient levels of fructose-1-phosphate aldolase.54,55 Three such patients were given fructose loads but did not develop hypoglycemia, hypophosphatemia, or the hypermagnesemia. The relationship of the decreased enzymatic activity to the pathogenesis of Tay–Sachs disease is unknown, although measurement of fructose-1-phosphate aldolase activity has been used as a marker to detect the carrier state of Tay–Sachs disease.55,56 The diagnosis of HFI can be made by an intravenous fructose tolerance test. The smallest dose that produces the typical symptoms without causing nausea and vomiting is 0.25 g/kg body weight.14 A dose of 0.25 g/kg body weight or 3 g/m2 surface area is given as a single rapid injection. Marked, prolonged reductions of blood glucose and phosphate levels occur regularly with this dose. However, at least one infant with marked hepatomegaly and severely deranged hepatic function did not have the typical changes in blood glucose and phosphate until the intravenous dose was doubled. Thus, with severe liver disease, blood and urinary levels of fructose are inconsistently elevated with the lesser fructose load. The lower dose of fructose may be a valuable aid as an initial screen for HFI, but a negative result, as described earlier, should not be used to rule out the diagnosis of HFI. Because the fructose load may
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
cause symptomatic hypoglycemia, vomiting, and derangement of liver function, measurement of intestinal aldolase activity may be preferable in patients strongly suspected of having HFI.57 The presumptive diagnosis made from tolerance tests should be confirmed by assay of enzyme activity in percutaneous liver biopsy specimens.
Treatment A diet containing no fructose alleviates all the symptoms and liver dysfunction associated with HFI.3,11–15 It is important that children and their parents receive detailed dietary counseling about which foods contain fructose. Older children commonly may associate discomfort with specific foods and regularly avoid them. However, infants are completely dependent on dietary selections made by their parents. The common practice of adding small amounts of sugar to processed foods demands almost constant attention to avoid substantial fructose intake. One patient with HFI who continued to consume small amounts of fructose had chronic slight elevations in aspartate aminotransferase (AST) levels as well as hepatic fat accumulation. Because pharmacologic doses of folate are known to cause non-specific increases in both aldolase activities, the patient was treated with 5 mg folate daily, with a resultant 53% increase in hepatic fructose-1-phosphate aldolase activity. With no change in dietary intake there were decreases in AST levels and hepatic lipid content. Tolerance tests, however, showed that folate treatment did not increase this patient’s ability to handle a large dietary intake of fructose.57
Prognosis Patients maintained on fructose-free diets have developed entirely normally, with normal lifespans, although most continue to have slight hepatomegaly with hepatic steatosis.16 Even infants with severely deranged liver function and substantial hepatic fibrosis can achieve remarkable recoveries once fructose is removed from their diets.
FRUCTOSE DIPHOSPHATASE DEFICIENCY In 1970, Baker and Winegrad described a patient who had a third type of genetic defect in fructose metabolism.4 The predominant clinical findings were hepatomegaly and fasting-induced hypoglycemia with lactic acidosis. The patient was shown later to have deficient hepatic FDPase activity. Other cases with similar clinical and laboratory findings have subsequently been reported.58 The primary difference between patients with FDPase deficiency and patients with HFI is that fasting as well as dietary fructose induces symptoms in patients with deficiencies of FDPase. As a result, an occasional patient with FDPase deficiency has been incorrectly diagnosed as having type IB GSD. Several patients have been found to have “partial” FDPase deficiencies. These patients do not have lactic acidosis but develop hypoglycemia during fasting or secondary to dietary intake of fructose or glycerol. One such patient had many of the characteristics of the syndrome ketotic hypoglycemia.59 Others have been erroneously diagnosed during infancy as having acute tyrosinosis. The deficiency of FDPase is inherited as an autosomal recessive trait. The diagnosis can be made by measurement of FDPase in cultured lymphocytes60 and confirmed by detections of mutations in FDPase gene.61
DISORDERS OF GALACTOSE METABOLISM In 1935, Mason and Turner provided the first detailed description of a patient intolerant of galactose.62 Since then, numerous case reports have described the constellation of nutritional failure, liver disease, cataracts, and mental retardation that results from a deficiency of hepatic galactose-1-phosphate uridyl transferase activity.63–65 The defect in galactosemia was initially thought to be a lack of synthesis of the transferase protein.66 Advances in molecular cloning have shown, however, that there are missense mutations in the gene coding for the transferase enzyme in the majority of patients with galactosemia (discussed later). Subsequently, Gitzelmann described the case of a 44-year-old patient with galactosuria and early-onset cataracts.67 Later reports indicated that this patient represented a second defect in galactose metabolism, which was a deficiency of galactokinase.68,69 The latter defect apparently does not result in progressive liver disease and mental retardation. In 1972, Gitzelmann discovered a third type of galactosemia, caused by uridine diphosphate (UDP) galactose-4-epimerase deficiency.70 This condition has been considered benign to the extent that the deficiency is limited to leukocytes and erythrocytes and affected individuals show no other laboratory or clinical abnormalities.71 More recently, generalized epimerase deficiency has been described.72–75 These patients show signs and symptoms identical to those of transferase-deficiency galactosemia. Because each of these conditions results in milk-induced galactosemia but represents three distinct biochemical entities, it has been suggested that the term galactosemia be supplemented by the specific enzymatic defect – that is, transferase-deficiency galactosemia, galactokinase-deficiency galactosemia, and epimerase-deficiency galactosemia.
TRANSFERASE-DEFICIENCY GALACTOSEMIA Human beings are capable of metabolizing large quantities of galactose, as demonstrated by the rapid elimination of galactose from blood.76,77 Elevation of the level of plasma glucose occurs shortly after galactose infusion as a result of conversion of galactose to glucose. Tracer studies indicate that as much as 50% of galactose may be found in body glucose pools within 30 minutes of injection.77 Under normal circumstances, plasma galactose is removed so rapidly by the liver that the rate of galactose clearance has been used as an index of hepatic blood flow78 and liver function.79 The removal mechanism is saturated at plasma levels of about 50 mg/dl, a level corresponding to the limits of the ability of galactokinase to phosphorylate the sugar. With a load of galactose that increases blood levels by 30–40 mg/dl, urinary losses may be substantial, because the kidney threshold is at plasma levels of 10–20 mg/dl.80 Almost half of the calorie source in most mammalian milks is from hydrolysis of lactose to its two monosaccharides, glucose and galactose. Consequently, the series of enzymatic steps involved in conversion of galactose to glucose are most stressed during infancy. As a consequence, enzymatic defects of this pathway are most likely to produce clinical signs and symptoms as well as marked elevations of blood and urinary galactose levels during this crucial period of development.
1275
Section XII. Inherited and Pediatric Diseases of the Liver
The first described defect resulting in galactosemia results from deficient activity of the enzyme required for the second of four steps in galactose metabolism (Figure 69-3). The consequences of this defect are much more severe than are those of the other defects in galactose metabolism, galactokinase deficiency and UDP-galactose4-epimerase deficiency.
Molecular Basis of Transferase-Deficiency Galactosemia Transferase deficiency is an autosomal recessive disorder. The sequence of the homologous protein from Escherichia coli,81 from Saccharomyces cerevisiae,82 and from humans shows overall sequence identity of 35%. The cDNA coding for the human transferase enzyme is 1295 nucleotide bases in length and predicts a 43kDa protein.83 The gene has been mapped to chromosome 9p18 and spans 4 kb with 11 exons. The amino acids histidine (164)-prolinehistidine (166) form an active site sequence that is essential for activity of the enzyme.84 Southern, Northern, and Western blotting experiments suggest that the majority of galactosemia mutations are missense mutations that result in low or undetectable enzyme activity. The two most commonly characterized mutations are glutamine 188Æarginine (Q188R) and arginine 333Ætryptophan (R333W). The arginine 188 mutation is the most common, accounting for onefourth of the galactosemia alleles studied.85 The second mutation, arginine 333Ætryptophan, occurs at a highly conserved domain in the homologous enzymes from E. coli, yeast, and humans. Several other mutations have been described, such as valine 44Æmethionine (M) and methionine 142Ælysine (K). Mutation S135L involving leucine substitution by serine occurs mostly in AfricanAmericans. Mutation N314D involves asparagine to aspartate change as the basis for the Durate variant. This variant is benign as the transferase expresses diminished but adequate enzyme activity.86 These other mutations result in low or total loss of activity of the transferase. Therefore, it appears that transferase-deficiency galactosemia results from missense mutations that tend to occur in regions that are highly conserved throughout evolution.87 So far more than 150 mutations have been described.88
Clinical Features Since transferase-deficiency galactosemia was first described in 1935, numerous patients with the disorder have been monitored for
Table 69-2. Common Clinical Findings in 43 Symptomatic Galactosemic Patients
Anorexia and weight loss Hepatomegaly Jaundice Ascites and/or edema Vomiting Abdominal distention Cataracts Infection Sepsis
Galactose
Galactose-1-phosphate
ADP
UDP-glucose
UDP-galactose
Glucose-1phosphate
3
4 Glucose-1-phosphate
Uridine triphosphate
1276
Number of patients
% Incidence
23 39 34 7 17 9 21 10 5
53 91 79 16 40 21 49 23 12
Reproduced from Koch R, Donnell GN, Fishler K, et al. Galactosemia. In: Kelley VC, ed. Practice of pediatrics. Hagerstown, MD: Harper & Row; 1979:14, with permission.
2
1
ATP
long periods. In 1970, Komrower and Lee reported the results of long-term follow-up of the 60 known cases of galactosemia in the UK.89 Long-term follow-up studies from 47 families have been reported from Los Angeles as well.90 The disease varies in severity. Some patients may present with an acute, fulminant illness after the first milk feedings or more commonly as a subacute illness with gastrointestinal symptoms (i.e., jaundice and failure to thrive). In milder cases, moderate intestinal upset after galactose ingestion may be the only manifestation. In severe cases, anorexia, abdominal distention, diarrhea, vomiting, and hypoglycemic attacks may occur shortly after birth. The most common initial symptoms are failure to thrive and vomiting, usually starting within the first few days of milk ingestion. Table 69-2 lists the more common findings in 43 symptomatic patients.74 Within the first week of life, hepatomegaly and jaundice are usually noted. In fact, prolonged obstructive jaundice in a neonate is a common presenting feature, and examination of urine for reducing sugar should be done in all infants with clinical jaundice. The jaundice from intrinsic liver disease may be accentuated in some infants with galactosemia by severe hemolysis and erythroblastosis. With continued galactose ingestion, ascites may develop as early as 2–5 weeks after birth. Cataracts may be seen within a few days after birth, or, if the mother has consumed large amounts of milk late during gestation, they may be present at delivery. The cataracts, consisting of punctate lesions in the nucleus of the fetal lens, may be so small that they can only be seen with slit-lamp examination. Retardation of mental development may be
UDP-glucose
Pyrophosphate
Figure 69-3. Reactions in the conversion of galactose to glucose. 1, galactokinase; 2, galactose1-phosphate uridyltransferase; 3, uridine diphosphate (UDP) galactose-4-epimerase; 4, UDP glucose pyrophosphorylase.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
apparent after the first several months. A few patients homozygous for the disorder have been entirely asymptomatic while ingesting milk. These patients, who are usually black, may be capable of metabolizing moderate amounts of galactose.75,91 Since milk-substitute formulas have become easily accessible, infants who have recurrent vomiting and poor weight gain during the first few weeks of life are often given trials of one of the lactosefree formulas. An occasional infant with galactosemia may be unwittingly changed to such a formula with resulting improvement without any awareness of the child’s underlying defect. In such instances, the patient may have the disease undetected until months or years later, when he or she may have motor retardation, hepatomegaly, and cataracts.92 An important clinical observation about galactosemia is that of Levy and associates, who showed the strong association between galactosemia and E. coli sepsis.93 In routine screening of more than 700 000 infants during a 12-year period, 8 infants with classic transferase-deficiency galactosemia were identified. Four of these infants had septicemia at the time galactosemia was detected (second week of life), and 3 of the 4 died. A review of results from eight other states that routinely screen newborns for galactosemia indicates that in screening more than 2.5 million newborns, 35 such patients were identified. Of the 35 patients, 10 are known to have had systemic infection, and 9 of the 10 died of bacteremia despite therapy. Infections usually develop at the end of the first week or during the second week of life, and their incidence appears to correlate directly with continued intake of galactose. These findings suggest that infants with E. coli sepsis should be considered possibly galactosemic and that infants recently diagnosed as galactosemic should be considered possibly infected with E. coli. With initiation of a galactose elimination diet, all acute manifestations of the disease usually improve within 72 hours, and hepatic dysfunction begins to normalize by 1 week. During the first year of life, small amounts of dietary galactose may cause symptoms; however, around puberty, most patients show an improved tolerance to dietary galactose. To account for this improved tolerance, an alternative metabolic pathway for galactose has been postulated to develop around the time of puberty. This pathway is thought to bypass the deficient transferase step by forming UDP-galactose from the interaction of galactose-1-phosphate and uridine triphosphate
(Figure 69-4).94 This reaction in liver and brain would reduce the concentration of galactose-1-phosphate to normal levels and would thus protect against the effects of the defective pathway. Although this hypothesis represents an attractive explanation for the increased ability to tolerate galactose later in life, tracer studies do not support an increased rate of galactose metabolism. A third pathway, involving the formation of xylulose, is unimportant in normal humans but may permit survival of some patients who continue to ingest galactose (Figure 69-4).
Laboratory Findings Routine laboratory findings may be varied but include elevated blood and urinary levels of galactose, hyperchloremic acidosis, albuminuria, aminoaciduria, hypoglycemia, and blood changes reflecting deranged liver function. Occasionally, infants may have severe and prolonged hypoglycemia. The galactosuria may be intermittent because of poor food intake or may disappear within 3 or 4 days of intravenous feeding. Thus, if the urine is not tested for reducing sugar during a period of galactosuria, the diagnosis may not be suspected. The finding of a urinary reducing substance that does not react with the glucose oxidase test should alert one to the possibility of galactosemia. This finding does not establish the diagnosis, because several other conditions such as fructosuria, lactosuria (from deficient intestinal lactase), and severe liver disease of any origin may impair the clearance of blood galactose and result in the presence of urinary reducing sugar that is not glucose.92
Biochemical Characteristics and Pathogenesis of Galactose Toxicity The hepatic manifestations of transferase-deficiency galactosemia are entirely due to the abnormal metabolism of galactose. Toxicity apparently results from accumulation of the metabolic products of galactose rather than from galactose itself. The two compounds that have been studied most extensively are galactose-1-phosphate and galactitol, the product of an alternate pathway (Figure 69-4). The biochemical causes of toxicity in individual organs may differ, depending on the metabolic patterns and functions of the involved organs. For example, cataracts are apparently caused by galactitol accumulation, whereas this compound appears to have little, if any, relation to the renal abnormalities or hepatic dysfunction.
Figure 69-4. Alternate pathways of galactose metabolism. 1, aldose reductase or L-hexonate dehydrogenase; 2, galactose dehydrogenase; 3, uridine diphosphate galactose pyrophosphorylase.
1 Galactose
Galactitol
NAD NADPH
NADP
2 NADH Galactonolactone
Galactose-1-phosphate
3
Uridine triphosphate UDP-galactose + pyrophosphate
Galactonic acid
-Ketogalactonic acid
D-Xylulose
1277
Section XII. Inherited and Pediatric Diseases of the Liver
HYPOGLYCEMIA Hypoglycemia that may occur after galactose ingestion is apparently caused by inhibition of glucose release from glycogen. The mechanism responsible is postulated to be high levels of galactose-1phosphate that interfere with phosphoglucomutase, the enzyme that catalyzes the conversion of glucose-1-phosphate to glucose-6phosphate.95 In addition, there is an inhibition of glucose formation through gluconeogenesis.96 The galactose-1-phosphate may be toxic in other ways as well, although investigations aimed at answering this question depended primarily on changes induced in normal animals fed high-galactose diets and do not necessarily reflect the changes that occur in patients who have deficiencies of enzymatic activity.
Aminoaciduria can be induced in both normal humans and rats by large amounts of galactose.110,111 This could be because of an accumulation of galactose-1-phosphate that secondarily impairs amino acid accumulation by the tubules.112 If analogous to that in the human intestine, the inhibition is non-competitive.113
BRAIN
Changes in the lens appear to result primarily from accumulation of galactitol. Van Heyningen showed galactitol accumulation, and Kinoshita and associates showed that an increase in galactitol concentration caused a concomitant increase in water content secondary to the oncotic pressure from the galactitol.97–99 The poor diffusion of the alcohol from the lens makes this organ particularly susceptible. Reversing the osmotic effect of the galactitol accumulation with an osmotically balanced incubation medium prevents the lenticular opacification.98 Many biochemical alterations occur concurrently in the lens as it undergoes cataract formation. These include decreases in amino acid transport, protein synthesis, several enzymatic activities, and alteration of ion fluxes.99–104 Glycolysis and respiration of the lens are reduced by about 30% after about 2 days of galactose feeding and remain at this level until cataracts develop.103 It is not surprising, then, that nutrient supplements can alter the rate of cataract formation in galactose-fed animals.105 Thus, nutrient imbalances and alterations in lenticular water content from galactitol formation together are prime initiators of lenticular opacification. The causes of cellular damage in other organs are less clear.
Changes in the brains of patients with transferase-deficiency galactosemia may not be entirely reversible after galactose restriction. For this reason, substantial efforts have been made to delineate the pathogenesis of galactose-induced brain damage. In the brains of patients and those of rats fed galactose, galactitol accumulates in higher concentrations than in any other tissue except the lens.108,114 This observation suggests that galactitol accumulation may be important in the pathogenesis of the brain abnormality. However, in patients with galactokinase deficiencies, galactitol accumulation does not appear to damage tissues other than the lens. Studies in chick brain showed that galactose administration diminished ATP, reduced brain glucose and glycolytic intermediates, redistributed hexokinase, enhanced fragility of neural lysosomes, and decreased fast axoplasmic transport.115–119 The effects could be temporarily reversed by glucose.120 Changes in the chick brain appear to be related to several factors such as hyperosmolality,119 alterations in energy metabolism,116 abnormal serotonin levels,121 and interference with active uptake of glucose into the neurons. It remains to be determined whether these changes in chicks are similar to galactose-induced abnormalities in patients with transferase deficiencies. Although intestinal epithelium of patients with galactosemia is also deficient in transferase activity, this deficiency does not appear to alter intestinal transport of galactose. Many infants develop intestinal symptoms of vomiting and diarrhea after galactose ingestion, but it is unclear whether this is a direct effect on the intestine or secondary to the effects of galactose on the central nervous system.
LIVER
GONADS
The concentration in liver of both galactose-1-phosphate and galactitol is elevated in patients fed galactose.106,107 However, the liver damage seen in patients with transferase deficiencies does not occur in normal animals fed high-galactose diets despite high hepatic levels of galactose-1-phosphate and high hepatic galactitol levels.108,109 In addition, patients with galactokinase deficiencies form large amounts of galactitol but show no liver damage. This suggests that some other metabolite or metabolites accumulate to act singly or in concert to cause cellular toxicity. In this respect, galactosamine, which was increased in the liver of one patient, is known to induce hepatocellular changes in animals.109
The majority of galactosemic females have ovarian failure as manifested clinically by infertility and biochemically by hypergonadotropic hypogonadism.122 Males with galactosemia have normal testicular function. The mechanism underlying ovarian failure in galactosemic women is not known, although galactose toxicity has been implicated. It is interesting to note that tissues with the highest transferase activity are target organs for the dysfunction and are affected to the greatest extent in galactosemia. In this regard, the ovary has five times more transferase activity and transferase mRNA than the testis.123
LENTICULAR CHANGES
Pathology KIDNEYS Kidneys of transferase-deficient patients develop renal tubular dysfunction after galactose ingestion. They accumulate both galactose1-phosphate and galactitol.106,107 However, accumulation of galactitol alone does not appear toxic, because patients with galactokinase deficiencies do not develop renal dysfunction but do excrete large amounts of galactitol. This suggests that the alcohol is not the primary renal insult in patients with transferase deficiency.
1278
Early hepatic lesions, present in the first weeks of life, consist of cholestasis and diffuse fatty vacuolation with little or no inflammatory reaction. The fatty changes are extensive and generalized throughout the lobule. Later, disorganization of the liver cells with pseudoductular and pseudoglandular formation occurs. This tendency toward pseudoglandular orientation of cells has been described as characteristic of galactosemia but is relatively nonspecific. As the disease progresses, delicate fibrosis appears, first in
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
the periportal regions, eventually extending to bridge adjacent portal tracts. Regenerating nodules and hepatic fibrosis are late features that, with continued galactose ingestion, progress to cirrhosis similar in many respects to the cirrhosis of ethanol abuse. Death usually occurs in the first year of life unless galactose intake is decreased or curtailed. Frank cellular necrosis is unusual but may occur with large amounts of dietary galactose. Despite the severity of the hepatic lesion there is a remarkable lack of infiltration of inflammatory cells.124 Except for cataract formation in the lens, other tissues show only minor changes. Kidneys show dilation of tubules at the corticomedullary junction. The spleen enlarges as a result of portal hypertension. Lesions in the brain are subtle, with minor loss of nerve cells and gliosis in the dentate nucleus and gliosis in the cerebral cortex and gray matter.125
Genetics Investigations of red blood cell and leukocytic transferase activities in family members indicate that the disorder is transmitted as an autosomal recessive trait.126 Heterozygotes have about 50% of normal activity, and genotype detection is more accurate when the transferase to galactokinase ratio is determined.127 Population studies indicate that the incidence of heterozygosity for galactosemia is between 0.9 and 1.25% and that between 8 and 13% carry the Duarte gene.128 Incidences of transferase-deficiency galactosemia derived from large-scale screening in neonatal nurseries have been between 1:10 000 and 1:70 000 live births.129,130 At least one patient with transferase-deficiency galactosemia has delivered a normal heterozygous infant.90
Diagnostic Screening for Galactosemia The rationale for genetic screening is threefold: (1) to detect disease at its incipient stage and thereby offset harmful expression of the mutant gene through appropriate medical treatment; (2) to identify a variant genotype for which reproductive options (family planning) may be provided; and (3) to identify gene frequency or biologic significance and natural history of variant phenotypes. Various screening methods for galactosemia have been used.131 The original Guthrie test used filter-paper blood samples from which a microbiologic assay detected elevated galactose levels. The newer Beutler test assays the erythrocyte transferase activity directly from dried filter paper, and the Paigen assay is an improved bacteriologic method that includes detection of elevated levels of galactose and galactose-1-phosphate. Measurements of elevated galactose require that the infant receive sufficient dietary galactose or a false-negative test will result. Conversely, the normal enzyme may become inactive in a hot or humid climate, and a false-positive (negative enzyme activity) may be reported. The relatively common inaccuracies of the screening tests for galactosemia and the low prevalence of the illness, coupled with the observation that infants born into families without a known history of galactosemia may be affected at birth, have prompted some screening centers not to screen for galactosemia. For example, in Quebec, the spaces on the blood sample filter paper assigned to galactosemia were given over to screening for congenital hypothyroidism, with a striking increase in cost-effectiveness.
In-utero assay for galactosemia is indicated in pregnant women with a family history of galactosemia. Cultured fibroblasts from amniotic fluid can be assayed for transferase activity. Additionally, the technique of chorionic villus sampling has been used to detect galactosemia during the 10th week of gestation.132 More recently, cloning of the cDNA encoding for the transferase enzyme and the finding that the majority of galactosemic patients have missense mutations have allowed for rapid molecular approaches using the polymerase chain reaction to detect common mutations.
Diagnosis The presumptive diagnosis of galactosemia in an infant with vomiting and failure to thrive on milk feedings may be made by identification of a urinary reducing sugar that does not react with glucose oxidase reagents. It should be remembered that lactose, fructose, and pentose may give similar results, but if the formula is milk-based and there is no other dietary sugar, galactosemia is the presumed diagnosis, and restriction of dietary galactose should be initiated immediately. Identification of the sugar can be made by paper or gas–liquid chromatography. Paper impregnated with galactose oxidase makes screening for galactosuria easier. Normal premature and some normal term infants may excrete as much as 60 mg/dl urinary galactose during the first week or two of life. Unlike suspected fructose intolerance, which may be diagnosed by use of a fructose tolerance test, demonstration of galactosuria or galactosemia by a galactose tolerance test should never be used. Although not clearly documented, it has been suggested that a single large exposure to galactose may result in severe and prolonged hypoglycemia, with resulting in brain damage. For this reason, the definitive diagnosis should be made on the basis of direct measurement of transferase activity and not by a tolerance test.75,76 The red-cell UDP-glucose consumption test has been used extensively as a screening test for the past decade.92,129,130,133 It is based on the assay of UDP-glucose before and after incubation of galactose-1-phosphate with red cell hemolysate as the source of transferase. Conversion of UDP-glucose to UDP-glucuronic acid by UDP-glucose dehydrogenase forms NAD from nicotinamideadenine dinucleotide (NADH), which is measured spectrophotometrically. With this procedure, a complete absence of red-cell transferase activity is found in homozygous patients, and intermediate levels in heterozygous carriers. Infants identified as having 50% of normal activity should have further tests to rule out other variants that may be homozygous and provide 50% enzymatic activity (discussed later). With the advent of screening for galactosemia, multiple variants of this disease have become apparent, the variants being more prevalent than classic transferase-deficiency galactosemia.134 There are three homozygotic types: 1. “Classic” galactosemia is autosomal recessive, and there is no transferase activity in erythrocytes, fibroblasts, liver, and presumably in any other tissue. In heterozygotic, unaffected carriers, activity is 50% of normal. 2. The Duarte variant is the most common form of galactosemia and is only detected by enzyme screening, because these infants are asymptomatic. Red-cell activity is 50% of normal, and on starch gel electrophoresis the enzyme migrates faster
1279
Section XII. Inherited and Pediatric Diseases of the Liver
than normal. Red cells of patients who have this variant produce two distinct bands rather than the single normal transferase band. In addition, red cells of a parent of a Duartehomozygous patient have three bands for the variant enzyme. Homozygotic Duarte erythrocytes have 50% of normal enzyme activity; heterozygous Duarte erythrocytes, 75% of normal activity. Ten to 15% of the population may have Duarte-variant galactosemia. The Duarte gene is apparently allelic with the normal and galactosemic genes, because the most frequently detected abnormality on neonatal screening tests is the compound heterozygous state, consisting of classic galactosemia with the Duarte variant. Two protein bands are present on protein electrophoresis, and erythrocyte transferase activity is 25% of control. Although some of these infants appear asymptomatic at birth and remain so during infancy, others have systemic symptoms with metabolic manifestations of galactosemia. 3. In the “Negro” variant, erythrocytic transferase activity is absent, but 10% of normal activity is present in liver and intestine. The Duarte and the Negro variants may be asymptomatic despite galactose ingestion, although patients with the variant may develop a galactose toxicity syndrome in the neonatal period. In addition to the homozygotic variants, several heterozygotic variants have been identified: 1. an Indiana variant, in which erythrocytic transferase activity is approximately 35% of normal and is highly unstable (mobility on starch gel electrophoresis is slower than normal); 2. a Rennes variant, which has about 7–10% of normal transferase activity (this variant also travels more slowly than normal by electrophoresis); 3. the Los Angeles variant, which has erythrocytic transferase activity higher than normal (about 140%); this has been detected in six families. Electrophoretic mobility of this variant of the enzyme is similar to that of the Duarte variant. West German and Chicago variants have also been identified by screening procedures.
Treatment Although the cause of the entire toxicity syndrome in transferase deficiency is uncertain, there is no disagreement that elimination of galactose intake reverses the biochemical manifestations of transferase-deficiency galactosemia. Some patients seem to have increasing tolerance to galactose with advancing age; however, studies using 14 C-galactose do not support the clinical impression that alternate pathways of galactose metabolism develop at puberty,90–92 nor is there any indication that any drug will increase galactose oxidation, although some patients with variant forms of transferase deficiencies can oxidize limited amounts of galactose. The only acceptable treatment at present is elimination of dietary galactose. Permissible diets are described in at least two publications.77,90 Preparations used in treating infants are Pregestimil, Nutramigen, and the soybean milk preparations. Both Pregestimil and Nutramigen are prepared from casein and may contain small amounts of lactose, but this amount of lactose does not appear to be sufficient to impair therapeutic efficacy. The soybean formulas
1280
contain small amounts of galactose in raffinose and stachyose, and other dietary constituents contain small amounts of galactosides, but these carbohydrates are not digested by human intestinal enzymes and should not affect the efficacy of treatment.90 Because of the frequent addition of milk to a number of proprietary food items, strict attention must be given to the diet during and after weaning. Concern has been raised regarding the presence of galactose in grains, fruits, and vegetables.135 These foods contain significant amounts of soluble galactose, although newer information related to substantial endogenous production of galactose has minimized this concern.136 It is important to be aware that asymptomatic heterozygotic mothers may have elevated serum galactose levels after ingestion of diets high in milk. Infants delivered of such mothers may have the galactosemic syndrome at birth. For this reason, restriction of galactose during the pregnancies of women who have previously borne children with galactosemia is recommended.90–92 The use of uridine and aldose reductase inhibitors in galactosemic patients has not been shown to be effective despite their theoretical advantage.137,138
Prognosis When untreated, galactosemia results in early deaths of many affected children and is attended by the prospect of mental retardation of those who survive. In a series of 43 galactosemic patients described by Koch and colleagues, there were 13 neonatal deaths.90 The deaths occurred at an average age of 6 weeks and were usually attributed to infection. Levy and co-workers noted that 9 of 35 patients died of E. coli infections and strongly recommended early cultures and institution of antibiotics effective against E. coli in any infant with galactosemia who appears ill.93 Treatment of galactosemic patients with a galactose-free diet results in survival with reversal of the acute symptoms, normal growth, and complete recovery of liver function; however, the longterm outcome (particularly for intellectual development) is not entirely certain. Experience gained in the long-term follow-up of 59 patients in the Los Angeles area indicates that many patients have developed very well and have attained college-level educations.90 Others who were equally well treated with galactose restriction have had various intellectual deficiencies, including verbal dyspraxia, reduced intelligence, learning disabilities, and neurologic deficits.139 The causes of the variability in the responses to treatment need further exploration. Hypergonadotropic hypogonadism is another long-term disorder observed in galactosemic women.140,141 Although successful pregnancy is possible, 65% of galactosemic women develop ovarian failure with atrophic gonads. The mechanism is believed to be related to excess galactose exposure during fetal and childhood development142 or to galactose restriction and a specific galactose need during ovarian development. The male gonads, however, appear more resistant to the effects of galactosemia.143 Osteoporosis is a frequent complication among females with galactosemia. The mechanisms underlying this complication may relate to low calcium intake, lack of sex hormones associated with ovarian failure, and an independent defect in collagen synthesis resulting in disturbances in bone mineralization.144,145 Renner and coworkers have shown that treatment with hormone replacement and vitamin D therapy (1000 IU/day) resulted in the onset of menarche
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
and increased bone density in two 28-year-old galactosemic twins when treatment was started at 25 years of age.145 Although genetic and social factors may influence results of intelligence tests, such factors do not explain all the differences observed. The association of thyroid dysfunction with galactosemia may have some role in the outcome.146,147 A factor that definitely affects outcome is the age of the patient at diagnosis. Evidence supports the previous impression that a more favorable outcome can be expected when a patient is treated at an early age. For example, the mean IQ of 16 patients treated before 7 days of age was 99.5, whereas that of patients treated between 4 and 6 months of age was 62. It is generally desirable to institute treatment at the earliest possible age, and neonatal screening is an important step in this direction. Although diagnosis and treatment at birth are desirable objectives, it is also possible that homozygotic galactosemic infants may have experienced unfavorable intrauterine exposure to galactose or its metabolites. Even with maternal dietary restriction of lactose during pregnancy, levels of erythrocytic galactose-1-phosphate in samples of cord blood from 12 homozygotic infants still averaged 11.3 mg/dl. Thus, it appears that the intrauterine environment is not ideal for a homozygotic galactosemic fetus.90
GALACTOKINASE-DEFICIENCY GALACTOSEMIA Galactokinase deficiency is less common than classic transferase deficiency, with an incidence of about 1 in 10 000.148 It does not result in progressive liver disease and mental retardation, but galactose exposure may result in cataract formation.67–69 It is appropriate to compare this entity with transferase deficiency, because it affects the first reaction (kinase) and the transferase, the second reaction of the galactose pathway (Figure 69-3). Comparison of patients with these two defects and the defect involving the third reaction (epimerase) has helped to define some of the mechanisms of toxicity in several organs, including the development of cataracts. With galactokinase deficiency, there is no accumulation of galactose-1phosphate, no systemic manifestations, and no mental retardation. Cataract formation is related to synthesis of galactitol in the lens and osmotic disruption of lens fiber architecture, as discussed in the previous section. Maternal galactokinase deficiency may result in fetal cataract formation.148 Because of the potential for cataract formation, lifelong elimination of galactose is suggested. The gene has been mapped to chromosome 17q24, and 20 mutations have been described in galactokinase gene, resulting in loss of the activity of the enzyme.149
URIDINE DIPHOSPHATE GALACTOSE-4EPIMERASE-DEFICIENCY GALACTOSEMIA Galactose epimerase catalyzes the third reaction of galactose metabolism (Figure 69-3). Epimerase deficiency was discovered incidentally while screening for galactosemia and has an incidence of about 1 in 46 000 in Switzerland. Patients have normal erythrocyte transferase activity but elevated levels of galactose-1-phosphate.70,71 One form of this condition is apparently caused by a decreased stability of the epimerase and leads to enzyme deficiency in those cells in which its turnover is slow or absent, such as erythrocytes.150 It is therefore considered to be a benign illness in as much as the enzyme
deficiency is limited to leukocytes and red blood cells. Affected persons with the form limited to leukocytes and red blood cells have no symptoms; but patients with generalized epimerase deficiency have been described.72,74 The latter patients have signs and symptoms identical to transferase-deficiency galactosemia. By contrast with transferase deficiency, in which uridine diphosphogalactose can be formed from uridine diphosphoglucose, one patient with generalized epimerase deficiency was unable to synthesize the galactose precursor necessary for synthesis of glycoproteins and glycolipids. These glycosylated compounds are necessary for cell membrane integrity, especially in the central nervous system. Thus, in contrast to patients with transferase deficiency, the rare patient with systemic epimerase deficiency may require small quantities of galactose for normal growth and development. One patient with epimerase deficiency continued to show slightly elevated levels of galactose-1-phosphate in red cells even with dietary restriction of galactose. Appropriate treatment of this disorder therefore requires frequent monitoring of erythrocyte galactose-1-phosphate levels in order best to determine the optimal dietary level of galactose.
GLYCOGEN STORAGE DISEASES Clinical and pathologic recognition of GSD affecting the liver and kidneys was described by von Gierke in 1929.151 Three years later, Pompe recognized another type of GSD that involved not only the liver and kidneys but most other organs as well.152 In 1952, Cori and Cori showed that hepatic glucose-6-phosphatase activity was deficient in patients with von Gierke’s disease.153 This marked the beginning of a classification of glycogenoses by the types of enzymatic defects and the primary organs of involvement. In most types of GSD, the glycogen content of liver or muscle or both is excessive. In unusual cases, the glycogen content may be less than normal, the molecular structure of glycogen may be abnormal, or both may occur. Despite differences in the specific enzymatic defects, most of the syndromes are not readily distinguishable on clinical grounds alone, and tissue analyses for glycogen content and enzymatic activity are necessary to confirm the diagnoses.154,155 Deficiencies of enzymes involving almost every step of glycogen synthesis and degradation have been identified. Their locations in the sequence of glycogen synthesis and degradation are illustrated in Figure 69-5. Most enzymatic defects giving rise to hepatic glycogenosis involve degradation of glycogen to glucose-6-phosphate or, in rare instances, the synthesis of glycogen from glucose-6-phosphate.154–157 On the other hand, patients with von Gierke’s disease are deficient in the activity of glucose-6-phosphatase, a gluconeogenic enzyme. As a consequence of this enzymatic difference, many of the clinical and chemical features of von Gierke’s disease, or type I GSD (GSD-I), are unique among the GSDs. For example, the tetrad of a bleeding tendency from thrombasthenia, urate stones from hyperuricemia, hyperlipidemia, and lactic acidosis accompanies GSD-I and is not part of the aberrations seen with other glycogenoses. Despite the number of enzymatic deficiencies leading to the glycogenoses, only types 0 (glycogen synthetase deficiency) and IV (a-1,4-glucan:a-1,4-glucan 6-glycosyl transferase deficiency) invariably lead to cirrhosis and liver failure. Patients who have type I (glucose-6-phosphatase deficiency) may develop benign hepatic adenomas and hepatic adenocarcinomas, and patients with type III (amylo-1,6-glucosidase deficiency; debrancher enzyme deficiency)
1281
Section XII. Inherited and Pediatric Diseases of the Liver
cAMP
Glucagon
cAMP dependent kinase Glycogen synthetase (1)
Phosphorylase (b)
Figure 69-5. Pathway for glycogen synthesis and degradation to glucose. Broken lines indicate enzymic activation after glucagon stimulation; heavy arrows indicate glycogen degradation from glucagon infusion; and hatched boxes indicate points in the metabolic sequence where enzymic defects have been identified.
Phosphorylase kinase (active) Glucose acid maltase
Glycogen synthetase (D) (inactive)
(1,4
Glycogen 1,6 linkage)
Branching enzyme
Debranching enzyme
Phosphorylase (a) (active)
Glycogen (1,4 linkage) UDP-glucose
G-1-P uridyl transferase
Glucose-1-P mutase Glucose-6-P
Glucokinase
Glucose-6-P9tase
Glucose
may develop hepatic fibrosis or cirrhosis. Because only a few patients have been reported with type 0 GSD, only types I, III, and IV are discussed in detail in this chapter.
TYPE I GLYCOGENOSIS (GLUCOSE-6PHOSPHATASE DEFICIENCY) Classification of Type I Glycogenesis Type I glycogenosis represents a deficiency of glucose-6phosphatase. This enzyme is alleged to be a multicomponent system consisting of three transport proteins, T1, T2, and T3, which transport glucose-6-phosphate, phosphate/pyrophosphate, and glucose, respectively, across the membrane of the endoplasmic reticulum. Two other components of this enzyme system include the catalytic subunit and a regulatory, calcium-stabilizing protein (SP).158 A schematic representation of the glucose-6-phosphatase enzyme system is shown in Figure 69-6. Although this is the generally accepted concept for glucose-6-phosphatase, there is no proof for its existence, and other interpretations of the relevant data are possible. Several subtypes of GSD-I have been identified. Type Ia is the classic type and represents a complete absence of glucose-6phosphatase activity, of which the catalytic subunit is a polypeptide
1282
doublet with molecular mass of 36.5.159 Type IaSP subtype describes patients with clinically classic type Ia with normal activity of the enzyme but lacking a 21-kDa stabilizing polypeptide protein (SP).160 Type IIb has a similar clinical picture to type Ia; however, in these patients, the activity of glucose-6-phosphatase in fully disrupted microsomes is completely normal, whereas the activity in intact microsomes and in vivo is lacking. The patients have a putative defect in the transport protein T1.161 Clinically, patients with type Ib are similar to those with type Ia, except for the presence of neutropenia. Type Ic is characterized by a putative deficiency of T2, the microsomal phosphate/pyrophosphate transport protein.162 Some patients with type Ic have impaired insulin release in response to glucose,163 others have a normal response. Type Id is limited to defects in the putative T3 transport protein; however, there are no reports of a deficiency in T3.
Molecular Basis of Glycogen Storage Disease Type I The cDNA encoding the murine glucose-6-phosphatase enzyme was cloned by screening a mouse liver cDNA library differentially with mRNA populations representing the normal and the albino deletion mouse known to express markedly reduced level of glucose 6-
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Glucose-6-phosphate
Glucose
Phosphate
SP T1
Glucose-6phosphate
Enzyme
Glucose
T3
T2
Phosphate
Endoplasmic reticulum
Figure 69-6. Schematic model of hepatic microsomal glucose-6-phosphatase. Glucose-6-phosphate entry into the endoplasmic reticulum is via a transport protein (T1). Hydrolysis occurs by the catalytic subunit of the glucose-6phosphatase. SP is a stabilizing protein. Phosphate is returned to the cytosol via a transport protein (T2). Glucose is returned via T3. (Modified from Burchell A. Molecular pathology of glucose-6-phosphatase. Faseb J 1990; 4:2978–2988, with permission.)
phosphatase enzyme.164 This discovery allowed the cloning of the human glucose-6-phosphatase cDNA by homology screening. The human gene spans 12 kb, composed of five exons, and encodes for a 357-amino-acid protein.165–167 The gene has been localized to chromosome 17q21. The glucose-6-phosphatase mRNA is expressed in the liver, kidney, and intestine, however, it is not expressed in human neutrophils/monocytes.165 To date, more than 70 mutations have been identified in the glucose-6-phosphatase gene of patients with type Ia.167–173 The two most common mutations are R83C and Q347X, which account for more than 70% of mutations in Caucasian populations.174 The common mutation Q347X causes a protein truncation of the last 10 carboxy-terminal amino acids that contain the signal for retention of the enzyme in the endoplasmic reticulum.
Clinical Characteristics Type IA is the most commonly diagnosed type and represents about a fourth of all cases diagnosed. A general discussion of type IA (designated GSD-I) is presented, followed by a brief discussion of type IB. The clinical picture is one of severe hepatomegaly, which may be apparent within the first 2 weeks of life. Profound hypoglycemia may develop shortly after birth or may not be prominent for several weeks, depending on frequency of feeding, the presence of intercurrent infection, and the severity of the disease. Because glucose6-phosphatase activity is also deficient in the kidneys, patients have substantial enlargement of the kidneys. Serum transaminase levels may be slightly elevated but become normal with effective treatment that maintains blood glucose between 75 and 110 mg/dl at all times. Neither the liver nor the kidneys show functional abnormalities other than the inability to release free glucose into the circula-
tion. In this regard, patients who receive a renal transplant remain unable to maintain normal fasting blood glucose levels.175 Consanguinity of parents is common, and the disease is transmitted as an autosomal recessive. During the past decade, major improvements in therapy have been documented.176–179 The study of mechanisms whereby deficiencies of glucose-6-phosphatase activity can cause striking abnormalities in lipid, purine, and carbohydrate metabolism has been instrumental in these therapeutic advances. To place GSDI in metabolic perspective, the mechanisms by which this single enzyme deficiency can have such profound effects on other metabolic pathways are reviewed.
Biochemical Characteristics Blood Glucose Changes. The most consistent and life-threatening feature of GSD-I is the low blood glucose levels that result from relatively short periods of fasting. Fasting for as short a time as 2–4 hours is almost always associated with decreases in blood glucose to less than 70 mg/dl, and it is not uncommon to observe levels of 5– 10 mg/dl after 6–8 hours of fasting. In normal individuals, blood glucose levels are maintained within a relatively narrow range by hepatotropic agents such as glucagon, which releases glucose either from stored glycogen or by gluconeogenesis.180 In GSD-I, degradation of glycogen can occur, or lactate or other gluconeogenic precursors can be converted to glucose-6-phosphate, but in the absence of glucose-6-phosphatase, glucose is not released, and blood glucose levels continue to decline. Blood hormone measurements indicate that, during periods of hypoglycemia, insulin levels are appropriately low and glucagon levels are high. After a glucose load, there is a substantial although somewhat delayed insulin release, with concomitant decreases in glucagon and alanine levels.155,176,181 Thus, the hormonal response to changes in the blood glucose concentrations appears appropriate. Lactic Acid Changes. Under normal circumstances, most circulating lactate is generated by muscle glycolysis during exercise. Removal and metabolism of this lactate are efficiently performed by the liver.180 On the other hand, much of the circulating lactate in patients with GSD-I is generated by hepatic glycolysis.182 This phenomenon is apparently the result of hepatic stimulation to release glucose from glycogen in combination with inefficient gluconeogenesis. Excess glucose-6-phosphate formed during glycogenolysis cannot be hydrolyzed to free glucose because of the lack of glucose6-phosphatase activity. Instead, glucose-6-phosphate is diverted through the glycolytic pathway. This metabolic diversion appears to be the basis for enhancement of lactate formation, as illustrated in Figure 69-7. Hyperlipidemia. Elevation of plasma lipids is a consistent and striking abnormality.183,184 Levels of triglyceride may reach 6000 mg/ dl, with associated cholesterol levels of 400–600 mg/dl. Free fatty acid levels are also usually elevated. Around puberty, xanthomas can appear over extensor surfaces, but they may also appear in childhood, with involvement of the nasal septum. Those located on the septum may contribute to the frequency of prolonged nosebleeds seen in some patients. As with lacticemia, elevated levels of triglyceride and cholesterol appear to be a consequence of increased rates of glycogenolysis and
1283
Section XII. Inherited and Pediatric Diseases of the Liver
Type I glycogen storage disease
Glycogen
Low blood glucose Hepatic stimulus for glucose release Increased stimulus
+ Glucose-6-phosphate Adenine
Further decrease in blood glucose
"Phosphate trap"
+ Protein
Inosine Amino acids
Pyruvate Lactate
Triglycerides
Uric acid
Acetyl CoA
Cholesterol Figure 69-7. Biochemical basis for the primary laboratory findings in patients with glucose-6-phosphatase deficiency (indicated by the solid rectangle). The increased production of glucose-6-phosphate that results from continuous stimulation of glycogen breakdown apparently increases glycolysis, which in turn results in a net increase (indicated by dark arrows) in the production of lactate, triglyceride, cholesterol, and uric acid. Both glycogenolysis and gluconeogenesis are involved in the overproduction of substrate.
glycolysis. Observations by Sadeghi-Nejad and co-workers suggest that excess hepatic glycolysis increases hepatic levels of NADH, nicotinamide-adenine dinucleotide phosphate (NADPH), and acetyl coenzyme A (CoA), three compounds important in fatty acid and cholesterol synthesis.182 Thus, increases in glycerol-3-phosphate and acetyl CoA generated by the glycolytic pathway, together with high levels of reduced co-factors, could sustain an increased rate of triglyceride and cholesterol synthesis.185,186 In addition to this apparent increased rate of lipid synthesis, an event concomitant with hypoglycemia is lipolysis from peripheral lipid stores. This further augments the tendency for hyperlipidemia and hepatic steatosis to occur by increasing circulating free fatty acids.183–187
Hyperuricemia Although blood levels of uric acid and the tendency to develop gouty arthritis and nephropathy vary in different patients, those who survive puberty often have gouty complications.188,189 Hyperuricemia was originally attributed to the increased levels of serum lactate and lipid, which competitively inhibit urate excretion.184 However, the high level of urate excretion together with the rate of incorporation of 14C-L-glycine into plasma and urinary urate indicates that an increased rate of purine synthesis de novo is probably more important in the genesis of hyperuricemia than is a decrease in urate excretion.190,191 The rate of purine synthesis can be influenced by at least two mechanisms: (1) alteration of the substrate (precursor) concentration (i.e., phosphoribosyl pyrophosphate (PRPP) and glutamine levels); and (2) alteration of the end-product, or purine concentration (i.e., low intracellular purine levels increase purine synthesis).192–194 In support of the former, two substrates, PRPP and glutamine, are necessary for the first committed reaction. This reaction transfers the amine from L-glutamine to PRPP to form 5-phosphoribosyl-1-amine and is apparently rate-limiting for the
1284
entire sequence of purine synthesis (reaction [4]). Although tissue glutamate and glutamine levels have not been measured, blood levels of the two substrates obtained from hyperuricemic patients with GSD-I are three- to eightfold higher than are values obtained after urate is normalized by glucose infusion.186 In addition to the possibility of increased availability of glutamine, the high levels of glucose-6-phosphate produced during periods of hypoglycemia and excessive glycogenolysis may increase synthesis of the second important substrate in purine synthesis, ribose-5-phosphate.184,188,195 These findings suggests that an apparent increased availability of purine precursors, glutamine, and ribose-5-phosphate may cause a secondary increase in PRPP and thus increase the rate of purine synthesis. Studies using human leukocytes indicate, however, that an increase in availability of glutamine and ribose-5-phosphate alone will not increase the generation of PRPP.196 Assuming this is true in liver, the second mechanism, alteration of end-product concentration, should be more important in modulating the increased rate of purine synthesis in patients with GSD-I. In support of the second mechanism for hyperuricemia, a decreased concentration of purine ribonucleotides would favor an increase in the rate of purine biosynthesis by releasing the glutamine pyrophosphate-ribose-phosphate amidotransferase from endproduct inhibition.197 Although hepatic nucleotide levels during hypoglycemic episodes have not been determined directly, indirect evidence suggests that, in patients with GSD-I, hypoglycemia can reduce adenyl ribonucleotide levels. Such a conclusion is based on measured values of hepatic ATP before and after simulating the effects of hypoglycemia with intravenous glucagon administration.198 Seven patients had a threefold decrease in hepatic ATP levels with concomitant 1.3-fold decreases in adenosine diphosphate (ADP). Such a reduction in ATP has been shown to favor the rapid degradation of adenyl or guanyl ribonucleotides to xanthine and uric acid.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
The latter set of reactions is also favored by low intracellular phosphate levels, which apparently occur through phosphate trapping of the phosphorylated sugar. Normally, this accumulation of glucose-6-phosphate is prevented by the action of glucose-6phosphatase.199,200 These observations suggest that the increase in urate production is secondary to recurrent episodes of hypoglycemia (reaction [5]), which result in compensatory glucagon release. This hepatotropic agent stimulates glycogen degradation to glucose-6-phosphate. The absence of phosphatase activity results in a phosphate-trapping effect and lowering of ATP levels, which in turn promotes degradation of preformed purines to uric acid (reaction [5]).25,200,201 Finally, the decrease in end-product (purine) concentration promotes a high rate of purine biosynthesis (reaction [5]).
Hypophosphatemia Low serum phosphate levels are not an invariable finding but are more likely to be present during periods of hypoglycemia and acidosis. It has been noted that a glucagon injection is followed by an acute decrease in serum phosphate that spontaneously reverts to the preinjection level within 3 hours. This suggests that cellular [Pi] levels are rapidly depleted by the glucagon effects and that there is a compensatory shift of phosphate out of the circulating pool. A well-demonstrated corollary occurs after fructose ingestion in patients with HFI. This has been shown to result from phosphate trapping on the fructose moiety because of blockage of the aldolase step.14,25,200,201 As a result of the inability to release Pi from the sugar, the liver cell must take up phosphorus from serum (reaction [6]). For example, when 6.6 mmol of fructose is administered to an 8.8-kg infant, the fructose load exceeds the amount of Pi mole per mole, in the entire extracellular fluid.14 A phenomenon analogous to that in HFI apparently occurs because of the phosphate trap created as a result of glucose-6phosphatase deficiency. A relative metabolic block at the aldolase step would also be expected because of the progressive increase in NADH formed during the initial phase of the reaction cascade from glucose-6-phosphate to pyruvate. The demonstrated decrease in hepatic glycogen content by a mean of 2.3% after a glucagon injection provides a large amount of glucose-6-phosphate that cannot be hydrolyzed to release the bound Pi (reaction [7]).198 The series of reactions (reaction [7]) reflects the phosphate trapping by the accumulated sugars of the anaerobic pathway, thus causing an acute shift of circulating phosphate to the intracellular pool. As the phosphate is lost from the sugars, there will be a compensatory shift of phosphate back into the circulation. Recurrent Fever. A few patients have recurrent fever in association with acidosis and hypoglycemia. In these patients, the fever can be reproduced by intravenous injection of glucagon if the patient is already slightly hypoglycemic (blood glucose 35–55 mg/dl) and acidotic (arterial blood pH 7.28–7.36). The febrile response begins 8–12 minutes after glucagon injection (0.1 mg/kg given over 3 minutes), and usually peaks 12–16 minutes later. If the low blood glucose level is corrected by intravenous administration of glucose and the acidosis is corrected by sodium bicarbonate, the temperature usually returns to normal within 45 minutes of glucagon infusion.
If the explanation for hypophosphatemia just postulated is correct, the febrile response may represent an uncoupling of oxidative phosphorylation secondary to lack of Pi. For example, the glucagon results in the excessive formation of glucose-6-phosphate from glycogen. Because of the glucose-6-phosphatase deficiency, a burst of glycolysis results in excess production of reduced co-factors, which normally produce high-energy phosphates. Because of low intracellular phosphate levels, oxidative phosphorylation, were it to occur, would have to be uncoupled, leading to production of heat rather than chemical energy in the form of ATP (reaction [8]).
Platelet Dysfunction Patients with GSD-I usually have prolonged bleeding times secondary to abnormal platelet aggregation. Corby and co-workers examined platelet function in 13 patients, each with deficient hepatic activity of one of the following enzymes: glucose-6phosphatase, debrancher enzyme, phosphorylase, or phosphorylase kinase.201 Only the 7 patients with glucose-6-phosphatase deficiencies had abnormal platelet aggregation, and 4 of these also had abnormal platelet adhesiveness. The defect appears to be intrinsic, because cross-over and resuspension studies using patients’ platelets in normal plasma and normal platelets in patients’ plasma did not alter in vitro platelet function. Two such patients had the ADP content of affected platelets measured, and in both instances it was normal. Nevertheless, the release of ADP from platelets in response to added collagen and epinephrine was markedly impaired. These observations suggest that the functional defect is an impairment of the ability of the platelet membrane to release ADP. Cooper has shown a similar defect in ADP release from platelets with elevated cholesterol content.202 The elevated cholesterol content impaired fluidity of the membrane, causing secondary impairment of ADP and epinephrine-induced aggregation. Although platelet cholesterol levels of patients with GSD-I have not been measured, the elevated serum cholesterol content might reflect elevated platelet cholesterol content and therefore may contribute to the abnormality of platelet function in patients with GSD-I. If this postulate is correct, then treatment that lowers blood and platelet cholesterol levels should also normalize platelet function. One of the author’s patients was found to have abnormal platelet function but had normal serum cholesterol and triglyceride levels. This finding does not support the hypothesis. Growth Impairment. Children who have GSD-I are of short stature but without disproportionate head sizes or limb or trunk lengths. The abdomen is usually massively enlarged as a result of hepatomegaly. Bones may be osteoporotic, and some patients have delayed bone age. The mechanism leading to these changes is not clear. Growth hormone and thyroid hormone levels are normal or increased.155,176 Measurements of calorie–protein intake in 3 patients for 2 weeks indicated adequate caloric consumption. Observations suggest that chronic lactic acidosis and concomitant reversal of the insulin-to-glucagon ratio may be more important factors in preventing normal growth.
Hepatic Adenomas and Carcinomas Most patients with GSD-I who are more than 15 years old are now found to develop adenomas. This is at variance with the previously
1285
Section XII. Inherited and Pediatric Diseases of the Liver
held view that they occur only infrequently. Adenomas develop in most patients during the second decade of life, but they may be found in 3-year-old children. The nodules, which are best demonstrated by ultrasonography and radioisotopic scanning, show increased echodensity and decreased isotope uptake. At laparotomy, they appear as discrete, pale nodules that range in number from one to many and in size from 1 to 5 cm. A number of patients have been found to have solitary hepatocellular adenocarcinomas in individual nodules.203–206 The mechanism causing the adenomas or their malignant degeneration is unknown, but treatment with portacaval shunting does not prevent their development. The author now has 2 patients who had adenomas before nocturnal feedings, but after 3 years of treatment, nodules were no longer detectable by scanning (Figure 69-8). Another older patient (age 16 years) developed adenomas during treatment with nocturnal feedings; the feedings were later found to be inadequate to maintain blood glucose above 75 mg/dl. After readjustment of therapy, the nodules decreased in size but did not show complete resolution. This suggests that chronic stimulation of the liver by hepatotropic agents (glucagon and others) that increase blood glucose levels may be important in the genesis of the adenomas. The tendency for adenoma formation and malignant transformation is highest in young adult patients and appears to be a consequence of supportive therapy, which currently ensures survival into childhood and young adulthood. The mechanism leading to hepatic malignancy is unknown. A similar progression has been observed in experimental hepatocarcinogenesis from exposure to N-nitrosomorpholine. The progression from normal hepatocytes to malignancy appears to be as follows. First, multifocal areas of cells containing excessive glycogen develop. The cells in these areas also show decreased glucose-6-phosphatase activity. Second, the focal cluster of cells develops a gradual reduction in glycogen content and a concomitant increase in ribosomes, reflected as basophilia by hematoxylin and eosin (H&E) staining. Finally, the foci enlarge and acquire the phenotypic markers of hepatocellular carcinoma. These experimental observations, coupled with the findings in patients with GSD-I, have led Bannasch and associates to postulate that the metabolic disturbance leading to hepatocellular
glycogenesis is fixed at the genetic level in both the experimental animals and the patients and is causally related to the neoplastic transformation.207
Severity of Illness Despite the fact that there is no difference in activities of the phosphatase enzyme between patients, the expression of symptoms and of chemical anomalies varies substantially from one patient to another without detectable differences in management. Some patients may have only moderate abnormalities in blood chemistry and slightly decreased growth rates, whereas others may have marked alterations in blood lipids, require frequent hospitalizations for fever and lactic acidosis, or even die in infancy or early childhood. In addition, hypoglycemia seems to abate somewhat, after patients have reached adulthood. In fact, a few patients who had moderately severe symptoms during childhood improved so dramatically during adulthood that they were able to have successful pregnancies.208,209
Glucose Production In 1969, Havel and colleagues reported that 2 adult patients with GSD-I showed near-normal basal rates of glucose production using 14 C-glucose as a marker.210 This observation has been confirmed in patients of all ages by several investigators who used dideuteroglucose as the isotopic marker.211–214 These studies defined several features of the illness that have important therapeutic implications: 1. Patients with GSD-I can release glucose into the circulation at close to normal basal rates. 2. Patients cannot increase glucose release during hypoglycemia or after a pharmacologic dose of glucagon; therefore, their basal rates of glucose production are also their maximal rates of production. 3. Maximal glucose production is variable between patients but is not related to residual activity of hepatic glucose-6phosphatase. However, the tendency for fasting-induced hypoglycemia and severity of the clinical illness is directly related to maximal rates of glucose production. Figure 69-8. Technetium scan of the liver, showing regression of hepatic filling defect in response to dietary therapy. The arrow in (A) indicates a hepatic adenoma before therapy; (B) shows its disappearance, with reduction in the size of the liver after 4 years of dietary therapy.
A
1286
B
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
4. Endogenous glucose production is not inhibited unless an exogenous source of glucose is provided at a rate of 8 mg/kg per minute, an amount that maintains blood glucose levels at about 90 mg/dl. 5. The improvement in ability to fast for a longer time after the second decade of life appears to result from a decrease in glucose utilization rather than an increase in glucose production.
Diagnosis Characteristically, these patients have no increase in blood glucose levels after administration of glucagon or epinephrine, and usually they show substantial decreases in blood glucose within 20 minutes of the intravenous administration of glucagon. As mentioned in the previous section, a major product of glycogenolysis is lactate rather than glucose. In some patients who are already slightly acidotic, glucagon stimulation may cause severe acidosis. Aside from the glucagon tolerance test, other tolerance tests have been used as an aid in diagnosing the type of glycogenosis. For example, neither galactose nor fructose is converted to free glucose in patients with GSD-I, and a tolerance test with either of these sugars results in a flat blood glucose curve. These tolerance tests have the advantage of avoiding liver biopsy. On the other hand, a substantial volume of blood is required to complete all the tolerance testing, and not infrequently, the results are inconclusive. For example, the author has had several patients referred with erroneous diagnoses based on tolerance tests. Thus, for accurate diagnosis, the author feels that measurement of hepatic enzyme activity is necessary. To provide a basis for some selectivity in the enzymatic analyses of biopsy specimens, the author’s practice is to determine serial blood glucose and lactate measurements during a 4–6-hour fast, followed by the maximum glucose response to glucagon before liver biopsy. For example, a rise in glucose of more than 30 mg/dl generally indicates a defect in phosphorylase kinase, which is not routinely measured from needle biopsy material. Before an effective form of treatment was available, the need for accurate diagnosis was less important than it is today. Because of the effectiveness of treatment of patients with glucose-6-phosphatase deficiency, the suspected diagnosis of GSD should be confirmed by enzymatic assay of hepatic tissue. A diagnosis can be confirmed by examination of needle biopsy material, which avoids potential complications of surgery and general anesthesia.215
Pathology In GSD-I, the liver cells are distended with glycogen and often contain medium-sized to large lipid vacuoles. The lipid content in the liver of an untreated patient is substantially greater than that in the liver of a patient who has been treated, but in either instance, hepatic steatosis is a prominent morphologic feature. The liver cells are pale-staining and have prominent plasma membranes. Three notable features differentiate GSD-I from other glycogenoses: (1) the presence of substantial steatosis; (2) a lack of associated fibrosis; and (3) nuclear hyperglycogenosis. Nuclear glycogenosis is commonly noted in hepatocytes of normal children, but in GSD-I (and GSD-III), the nuclei are grotesquely enlarged – that is, hyperglycogenosis is present.
It is not possible to distinguish between normal and elevated levels of cytoplasmic glycogen in the liver in any of the forms of glycogenosis through the use of periodic acid–Schiff (PAS) stain.216 Thus, to make a diagnosis of excessive glycogen content, a quantitative determination is necessary. It is also important to note that the livers of normal individuals and those of patients with glucose-6-phosphatase deficiencies can degrade glycogen. Thus, postmortem analyses of surgical specimens that are not frozen promptly may give inappropriately low (e.g., “normal”) levels.
Treatment Patients with some glycogenoses – for example, those who have deficiencies of hepatic phosphorylase kinase activity – and some patients with GSD-III (debrancher deficiency) have an excellent prognosis without specific treatment. In fact, with the exception of those who have defects in glycogen synthesis, i.e., generalized glycogenosis (acid maltase deficiency; GSD-II; Pompe’s disease), or glucose6-phosphatase deficiencies, most patients who have hepatic glycogenoses have favorable prognoses and are successfully treated with attention to frequencies of food intake. This, however, has not been true of most patients who have GSD-I. Present recommendations for treatment of GSD-I stem primarily from the studies by Folkman and associates,217 who first illustrated the reversal of most biochemical abnormalities after TPN. Their observation that both TPN and portacaval shunting218,219 delivered nutrients primarily into the systemic circulation suggested that hepatic exposure to nutrients was important in the pathogenesis of many of the biochemical abnormalities. Nevertheless, it was later demonstrated that the same beneficial effect seen with TPN or portacaval shunting could be achieved with an intragastric infusion of a nutrient solution similar in content to that used for TPN.220 This suggested that bypassing the liver with nutrients was not the most important factor in reversing the abnormalities. The similarity in the three types of treatment (i.e., portacaval shunting, TPN, and continuous intragastric infusion of glucose) was that a hormonal stimulus to the liver to produce glucose was decreased or averted. Specifically, both TPN and continuous intragastric feeding prevented a hepatic stimulus for glucose release by maintaining blood glucose levels in the range of 90–150 mg/dl, whereas the portacaval shunt prevented such a stimulus by diverting pancreatic and enteric blood into systemic circulation. On this basis, the hypothesis for treatment illustrated in Figure 69-9 was formulated. The hypothesis states that, as blood glucose falls below a critical level, compensatory mechanisms cause glycogen degradation to glucose-6-phosphate. In the absence of glucose-6-phosphatase, glucose-6-phosphate is not hydrolyzed to release free glucose, and the hepatic stimulus for glycogenolysis results in formation of other intermediates such as lactate, triglycerides, and cholesterol. To interrupt the stimulus, treatment with an exogenous source of glucose inhibits the release of hepatotropic stimuli and thus the excess glycogenolysis. If this postulate is correct, any method of treatment that maintains blood glucose above a critical level should also prevent, or at least alleviate, biochemical manifestations of the illness. In addition, the hypothesis suggests that diversion of portal vein blood flow should dilute hepatotropic agents in the systemic circulation. This dilution should result in less stimulation of glycogenolysis.
1287
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 69-9. A biochemical basis for management of patients with glucose-6-phosphatase deficiency (indicated by solid rectangle). By preventing the decrease in blood glucose with an exogenous supply of glucose, excessive glycolysis and gluconeogenesis are prevented. This results in a net decrease in production of circulating triglyceride, cholesterol, lactate, and uric acid.
Glycogen
Low blood glucose Hepatic stimulus for glucose release
Glucose-6-phosphate Treat with exogenous
Glucose
Uric acid Pyruvate
Triglycerides
Lactate Cholesterol
Theoretically, either portacaval shunting or continuous infusion of a high-glucose diet should be effective in reversing most manifestations of the illness, with the exception that portacaval shunting should have little or no beneficial effect on hypoglycemia. Thus, portacaval shunting is not recommended as a sole form of treatment for those patients who are expected to have frequent episodes of very low blood glucose levels or for small children, in whom shunts may be more likely to close spontaneously. Although TPN and continuous intragastric infusion of glucose are effective treatment modalities for GSD-I, they are impractical on a long-term basis. A more practical method was devised to maintain blood glucose at physiologic concentrations or at levels that would prevent stimulation of excess glycogenolysis and glycolysis. This treatment consisted of a high-glucose diet given to simulate TPN. Thus, it was given enterally either by nasogastric tube or by gastrostomy during nighttime sleep, along with a high-starch diet, which was consumed at frequent intervals while the patient was awake. Such a regimen has successfully maintained a large number of patients relatively symptom-free for more than 10 years and has provided normal or near-normal growth and development.221,222 Chen and colleagues found that a number of patients can maintain normal blood glucose levels by taking cold, uncooked cornstarch (2 g/kg) at 6-hour intervals.223 This regimen has been used by a number of patients to avoid the continuous nocturnal feedings. A number of the younger children have not been able to maintain normal glucose and lactate levels with the cornstarch regimen as well as they had with the continuous nocturnal feeding regimen.224,225 In the author’s experience, growth rates were less with the raw starch regimen than with continuous feedings, and one patient consumed such large quantities of cornstarch that protein intake was insufficient to maintain normal secretory proteins (albumin, transferrin, and retinol-binding protein). Thus, although the cornstarch feedings can be beneficial as a time-release form of glucose in some patients, a dose–response to the starch preparation and careful monitoring of blood glucose levels should be carried out to ensure that treatment is appropriate for individual patients. The author believes that many patients require an intensive feeding regimen at least until they have stopped growing. This consists of high starch feedings at 2–3-hour intervals during the day, continuous nocturnal feeding of a complete, low-lipid-containing (<5% calories) formula, and periodic monitor-
1288
ing to ensure normal blood glucose and lactate levels throughout the day and night. As patients become fully grown and have a relatively lower requirement for glucose, the uncooked cornstarch regimen may allow discontinuation of the nighttime nasogastric feedings.
Prognosis Until the use of nocturnal feedings, the first few years of life were usually marked by frequent hospitalizations for treatment of hypoglycemia and acidosis, with a high rate of death or permanent central nervous system impairment from recurrent and prolonged episodes of hypoglycemia. Patients who survived puberty appeared to have fewer problems than they had when younger. Patients with persistent hyperuricemia had gouty complications during the second and third decades of life, and many patients had complications of hyperlipemia, with xanthomas, and with higher rates of cardiovascular disease and pancreatitis than those in the general population.154–156,180,226 Recognition of hepatic adenomas has been relatively recent, and the incidence of complications from benign hepatic adenomas is unclear, although several patients have developed hepatomas.203,205 Long-term follow-up of the results of portacaval shunting or nocturnal feedings is not complete. Ten-year follow-up of patients treated with nocturnal feedings indicates that infants so treated have many fewer problems than they had before treatment and that some patients with hepatic adenomas may show resolution after a few years of treatment. Too few patients have been monitored into the third decade of life to permit conclusions, but early observations indicate that, for optimal treatment, the nocturnal feedings are necessary for most young patients, whereas raw cornstarch administration may suffice for older patients. On the other hand, patients generally have less tendency to hypoglycemia after the age of 20. A 22-year-old patient, severely affected at the age of 15 years, had been treated with nocturnal feedings for 61/2 years, with complete resolution of all chemical manifestations of the illness except hypoglycemia (blood glucose 62 mg/dl) and lactate elevation (7 mmol/l) after a 9-hour fast. She was weaned off nocturnal feedings by gradually decreasing the hourly feeding over a period of a month. After 21/2 years of no nocturnal feedings, she continues to have completely normal blood chemistry results after an 8-hour fast. There is still no
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
hepatic glucose-6-phosphatase activity, however. So long as blood glucose is consistently maintained between 70 and 120 mg/dl, most children appear to lead fairly normal, healthy lives, with normal growth and development, although in at least 1 patient the disease has been unresponsive to this type of management.227 Chen and colleagues observed that older patients (>18 years) with suboptimal treatment have a high incidence of progressive renal disease. The affected individuals show progressive glomerular sclerosis with proteinuria as an early manifestation.228 Renal involvement appears initially with microalbuminuria and hyperfiltration progressing to frank proteinuria and hypertension.229 Although the cause of the lesion is unclear, it appears that the incidence is lowered by maintenance of good control of blood glucose and other blood abnormalities, because none of the author’s 9 patients aged 25–34 show renal abnormalities.
GLYCOGEN STORAGE DISEASE TYPE IB In 1968, Senior and Loridan described a patient with clinical and laboratory features identical to GSD-I except that no enzyme defect was identified from frozen liver.230 Subsequently, a number of similar patients have been identified.231–239 In addition to the hepatic abnormality, patients have repeated infections because of neutropenia and abnormal leukocyte migration,234–238 and some show decreased neutrophil phagocytosis-stimulated oxygen consumption, decreased nitroblue tetrazolium reduction,237 defective bactericidal activity, defective hexose monophosphate shunt activity, and increased incidence of inflammatory bowel disease.238,240 These patients have been diagnosed as having GSD-IB. There has been little mention of a familial occurrence in the reports of GSD-IB. This is curious, because a number of patients with GSD-I have affected siblings. One of the author’s patients with GSD-IB has seven unaffected siblings. The number of reported cases is small, however; and the mode of inheritance is presumed to be autosomal recessive. The reason for the discrepancy between the in vitro and in vivo activity of glucose-6-phosphatase in patients with GSD-IB is not known. Steady-state kinetic measurements led Arion and associates to conclude that the normal microsomal glucose-6-phosphatase enzyme is a two-compartment system consisting of a specific glucose-6-phosphate carrier on the outer bilayer of the membrane of the endoplasmic reticulum and a catalytic phosphorylase component located on the inner half of the membrane.238 Because disruption of isolated microsomes of patients with GSD-IB with cholate or freeze-thawing results in a marked increase in activity, the general assumption has been that the defect in GSD-IB is a deficiency of the translocase or carrier portion of the enzyme system,231–233,239 although the putative translocase has never been identified in microsomes from normal liver. Zakim and Edmonson, using methods to measure pre-steady-state kinetics,241 have shown that the limiting step in the reaction is not glucose release from the enzyme but the release of phosphate. These findings in normal liver microsomes are similar in the liver of a patient with GSD-IB, although the pre-steady-state kinetics are blunted. These findings question the general concept of a translocase and suggest that patients with GSD-IB may have a configurational abnormality in the enzyme–membrane interaction that can be overcome by alteration of the membrane lipid rather than by an opening of the microsomal vesicle.241,242
Molecular Basis of Glycogen Storage Disease Type IB GSD type IB is caused by mutation in the gene encoding microsomal glucose-6-phosphate transporter.243 The gene has been mapped to chromosome 11q23 and is composed of 9 exons spanning a genomic region of 4 kb. The gene is expressed in the liver, kidney, and leukocytes.244 More than 69 mutations have been described in the gene, resulting in functional deficiency of glucose-6-phosphate transporter, which explains the neutropenia and neutrophil–monocyte dysfunction characteristic of GSD type IB.243,245 No genotype–phenotype correlations have been described for type IA or IB disorders.246 Treatment of GSD-IB is identical to that of GSD-IA, with the possible exception that prophylactic antibiotics may lessen the tendency for frequent infection.234–238,247 Improvement in neutrophil function with treatment has been reported by some investigators.237,238 However, the authors’ experience was that the neutropenia and abnormal migration persisted even after 3 months of management that normalized all parameters of disease in the blood, and even after subsequent portacaval shunting.247,248 The lack of improvement in neutrophil function in some of the well-treated patients suggests that the defect in GSD-IB is intrinsic to both the liver and leukocytes. Because glucose-6-phosphatase is not known to have any importance for normal neutrophil function, the relationship between the hepatic enzyme defect and leukocyte dysfunction is not known. Improvement of neutropenia and neutrophil dysfunction occurs in response to granulocyte colony-stimulating factor.249
TYPE III GLYCOGEN STORAGE DISEASE (GSD-III) AMYLO-1,6-GLUCOSIDASE (DEBRANCHER) DEFICIENCY This glycogenosis accumulates a polysaccharide that has a structure similar to limit dextrin produced by degradation of glycogen with phosphorylase and oligo-1,4-1,4 glucantransferase but no debrancher activity (reaction [9]).250 As depicted, the terminal a1,4-glucosyl units are hydrolyzed by the combined activity of oligo1,4-1,4 glucantransferase and phosphorylase, but the inner branch points of a-1,6 linkages (MG) are not hydrolyzed by debranching enzyme. Thus, the glycogen molecule is abnormal, with an excessive number of branch points (1,6 linkages). The debrancher enzyme contains two catalytic subunits on a single polypeptide chain. The two activities are oligo-1,4-1,4-glucantransferase and amylo-1,6-glucosidase.
Molecular Basis of Type III Glycogen Storage Disease The human gene encoding for glycogen debranching enzyme is 85 kb is length and consists of 35 exons.251 The gene has been localized to chromosome 1p21.252 The cDNA includes a 4545 bp encoding region and 2371 bp 3¢-untranslated region. The predicted protein is ª172 kDa, consistent with the estimated size of the purified protein.253 Six mRNA isoforms have been identified.254 Isoform 1 is expressed in the liver; isoforms 2, 3, and 4 are muscle-specific. Isoforms 5 and 6 are minor isoforms. Mutations in the glycogen
1289
Section XII. Inherited and Pediatric Diseases of the Liver
debranching gene have been described in patients with type IIIa and IIIb. These mutations include missense, nonsense, splicing, and deletion insertion defects.255 Specific mutations in exon 3 such as 17delAG and Q6x are only seen in type IIIb.256 The splice mutation IVS32-12A>G was found to cause mild clinical symptoms, whereas mutations 3965delT and 4529insA are associated with a severe phenotype and early onset of clinical symptoms.257 Depending on the tissue(s) involved and enzymatic characteristics, subtypes of GSD-III have been described (i.e., GSD-IIIA, GSD-IIIB, and so on). Type IIIa describes patients who exhibit complete absence of debrancher enzyme activity in hepatic and muscle tissue, whereas type IIIb describes patients with liver involvement only.
Clinical Characteristics By physical examination alone, these patients cannot be readily distinguished from patients who have type I glycogenosis (GSD-I). Early in life, hepatomegaly and growth failure may be striking. However, a few patients may develop splenomegaly at 4–6 years of age.258 These patients usually have evidence of hepatic fibrosis but do not necessarily develop cirrhosis and liver failure. In addition to hepatic involvement, a number of patients with GSD-III have muscle weakness. Rapid walking and climbing result in increased weakness without cramps.259 Some patients may have a progressive myopathy. Glycogen may also accumulate in the heart, and moderate cardiomegaly with non-specific electrocardiographic changes is sometimes present.222 However, congestive heart failure and cardiac arrhythmias are not reported. There is no renal enlargement in GSD-III, in contradistinction to GSD-I. The clinical course in GSD-III is generally much milder than that of GSD-I, in that severe hypoglycemia is not a problem except with prolonged fasting. Some patients have shown relative decreases in liver size around puberty,250,260 but rare patients have shown evidence of progressive fibrosis and liver failure.154,261,262 The latter patients may have an additional phosphorylase or phosphorylase kinase deficiency.154 In a study of 41 patients with type III GSD, 31 patients had involvement of the liver and muscle (type IIIa), 4 patients had liver involvement only (type IIIb), 3 patients had unknown muscle status, and 3 patients had isolated deficiency of transferase activity with retention of glucosidase activity.263
Biochemical Characteristics and Laboratory Findings Lipid levels in plasma are variably elevated and to some extent appear to be related to the individual tendency toward fastinginduced hypoglycemia.260 That is, the patients who develop lower glucose levels with 6–8-hour fasts tend to have higher blood lipid levels. However, none of the patients approach the severe elevations of 4000–6000 mg/dl seen with GSD-I. Uric acid levels are generally normal, but rare patients reportedly have slight elevations. Serum transaminase levels are consistently moderately elevated (300–600 IU; normal <40 IU),64 although some patients show elevations of 900–2000 IU. Galactose and fructose are readily converted to glucose by these patients; similarly, protein and amino acid mixtures induce small and prolonged increases in blood glucose levels.181,260 Both glucagon and
1290
epinephrine increase blood glucose when given between 11/2 and 3 hours after a meal but elicit little response after a 14-hour fast (“double glucagon tolerance test”).264 This result is interpreted to indicate available 1,4 glucosyl linkages that can be degraded by phosphorylase shortly after a meal. The glycogen in this case is degraded only until 1,6 linkages are encountered. Access to 1,4 linkages is blocked by terminal 1,6 glycosyl linkages that would be present after a prolonged fast, and these would prevent glucose increase after glucagon administration. Unfortunately, the glucose response to the double glucagon tolerance test is not a consistent finding, and it should not be used as a definitive diagnostic test. Because glucagon also stimulates gluconeogenesis, the inconsistency of the test results is possibly due to glucose formation via this pathway. The liver content of glycogen is often markedly increased (to as much as 17.4 g/100 g tissue) in GSD-III.250 By various techniques, the glycogen is found to have abnormally short outer branch points. Many patients also show some depression of glucose-6-phosphatase activity. Hug has found several patients with combined defects in phosphorylase and phosphorylase kinase. These patients are generally more severely affected and tend to develop cirrhosis.154 A series of techniques is available for measuring debrancher activity: (1) liberation of glucose from phosphorylase-treated limit dextrin; (2) incorporation of 14C-glucose into glycogen; (3) 1,4Æ1,4 transfer of an oligoglucan (glucan transferase activity); and (4) hydrolysis of singly branched oligosaccharides. Using these various techniques, Hers and associates have identified a series of biochemical subtypes of GSD-III. These subtypes are also divided according to types of muscle glycogen.250,265,266
Pathology The liver in GSD-III is very similar in appearance to that in GSDI, with two notable exceptions: (1) the presence of fibrous septa or frank cirrhosis; and (2) the paucity of fat (Figure 69-10). Progression of the fibrosis to cirrhosis has been demonstrated. Hug reports that progression to cirrhosis is more likely to occur in patients who have combined enzymatic defects.154,264 The ultrastructural appearance of the liver is not distinguishable from that in GSD-I, except that in GSD-III lipid vacuoles are small and are less frequent.266 If muscle is affected, excessive glycogen is readily demonstrable in ethanol-fixed specimens by the PAS method. Glycogen accumulates between intact myofibrils and in the subsarcolemmal position,216 locations in which glycogen usually occurs but not in abundance. The diagnosis depends on demonstration of deficient activity of amylo-1,6-glucosidase (debrancher enzyme).
Treatment Treatment of this disorder remains investigative. Treatment should be restricted to patients who have obvious muscle involvement, progressive fibrotic changes in the liver, or both. An accurate correlation between the type of glycogen accumulation and progression of liver disease, so that a clear-cut prognosis could be assigned to each patient, would be helpful in showing a positive therapeutic response. Present investigative efforts combine the technique of nocturnal feedings with the known responses to protein and amino acids.181,260,267 Slonim and co-workers have shown improved growth and increased muscle strength in a patient given a high-protein diet
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Figure 69-10. Percutaneous liver biopsy specimen from a patient with glycogen storage disease type III, showing increased fibrosis. The hepatocytes do not contain fat deposition. ¥80.
during the day and continuous nocturnal intragastric feedings of a high-protein liquid formula (Sustacal) at night.268 Borowitz and Greene269 have found that growth and transaminase and blood glucose levels were more positively influenced by a high-starch diet with a standard (recommended dietary allowance) protein intake. This therapy is therefore virtually identical to that for GSD-I. This outcome is encouraging, but more extensive follow-up evaluation over a longer period of treatment is needed.
TYPE IV GLYCOGENOSIS (a-1,4 GLUCAN-6GLYCOSYL TRANSFERASE DEFICIENCY; GSD-IV) Type IV GSD, a rare form of glycogenosis, was first described clinically and pathologically in 1952 by Anderson.270 Illingsworth and Cori showed that the glycogen possessed abnormally long outer and inner chains of glucose units.271 Over 10 years later, Brown and Brown demonstrated the absence of branching enzyme activity in this disorder.272 The few descriptions of the disorder (about 20 documented cases) have illustrated its unusual clinical, biochemical, and pathologic aspects.273–279
Clinical Characteristics Infants are normal at birth and for some months thereafter. The onset of symptoms during the first year of life is insidious.
Symptoms may manifest as early as 3 months or as late as 15 months of age. The disorder is usually diagnosed because of hepatosplenomegaly, abdominal distention, signs and symptoms referable to hepatic dysfunction, non-specific gastrointestinal symptoms, and failure to thrive. Muscle hypotonia and wasting may be present. Superficial veins over the distended abdomen are prominent as the disease progresses. Patients who live beyond infancy develop cirrhosis with accompanying portal hypertension, ascites, and esophageal varices. The terminal course is usually due to chronic hepatic failure and jaundice with bleeding esophageal varices. Varients include a cardiopathic form of childhood in which patients develop cardiac failure, apparently from myofibrillar damage caused by polysaccharide deposits within myocardial cells. Intercurrent infection is a common terminal complication. The duration of survival after diagnosis is usually 2–37 months, although an occasional patient may survive 3–4 years. A neuromuscular variant has been reported that may involve the skeletal muscles, peripheral nervous system, and central nervous system. The level and extent of the neuromuscular lesions appear to vary from patient to patient, and at least 1 patient showed signs of involvement of muscles and nervous system simultaneously.279 A 59-year-old man with GSD-IV had deficient branching enzyme in skeletal muscle with normal muscle glycogen and low normal enzyme activity in leukocytes.280 The man had a 30-year history of
1291
Section XII. Inherited and Pediatric Diseases of the Liver
progressive, asymmetric limb-girdle weakness and a vacuolar myopathy. The vacuoles contained glycogen, which was partially resistant to diastase. Ultrastructural changes resembled the amylopectin polysaccharide deposits encountered in childhood GSD-IV. Three other adult patients have been reported with an amylopectin-like storage myopathy; 2 of these later developed involvement of the heart and brain, but none showed hepatic involvement.281–283 Approximately 50% of infants with GSD-IV have signs suggestive of involvement of the neuromuscular system and abnormal deposits of polysaccharide in skeletal muscle.277,279,284,285 The pathologic changes can occur in the absence of clinical changes.277 Because the majority of these patients have not had comparative measurements of branching enzyme activity in fibroblasts, liver, muscle, and leukocytes, it is not possible to correlate apparent clinical and pathologic findings with enzymatic changes. It has been suggested that two branching enzymes may be acting at different sites272,273 and that the clinical expression of this disease depends on deficiency of organ-specific brancher isoenzymes. The author has had experience in managing two unusual patients with deficient branching enzyme measured in both the liver and skin fibroblasts (by Dr. Barbara Brown, St. Louis, Missouri). The first patient, a male, showed the usual progression to cirrhosis and liver failure by 18 months of age.286 At age 2 years, he underwent liver transplantation and standard treatment with immunosuppressants. Six years post-transplant, the patient had normal liver function and shows no evidence of nerve, muscle, or cardiac abnormalities. Thus, despite the generalized nature of the defect, transplantation of a normal liver has not been associated with obvious progression in other organs. The second patient, a 6-year-old boy, was noted at 3 years of age to have hepatomegaly and chronically elevated serum aminotransferases.287 Open-liver biopsy showed moderate, generalized micronodular fibrosis and typical PAS-positive, diastaseresistant deposits in about 25% of the hepatocytes. Electron microscopy showed accumulation of Drochman fibrils characteristic of GSD-IV, but these fibrils were not present in all cells. During the succeeding 3 years, the boy showed normal growth and development with spontaneous normalization of liver tests. Liver biopsy at the age of 6 years showed minimal periportal fibrosis with no PASpositive, diastase-resistant deposits present in any hepatocytes and normal-appearing glycogen by electron microscopic analysis. Assays of hepatocytes and skin fibroblasts showed no detectable branching enzyme activity. These 2 patients plus previous reports of GSD-IV indicate that branching enzyme deficiency may present in various ways, the classic infantile variety primarily affecting the liver. As more patients have the opportunity for transplantation, it is anticipated that a more accurate classification system can be found for the various subgroups of this unusual disorder.
Laboratory Findings Blood electrolytes are usually within normal limits except in patients with renal tubular defects, who have low bicarbonate concentrations. Serum transaminase and alkaline phosphatase levels are usually elevated to three to six times normal. Except with malnutrition, late in the disease, serum cholesterol is often slightly elevated. Until liver failure develops, serum albumin, globulin, bilirubin, and ammonia are normal. All liver test results become abnormal as liver failure becomes severe. Glucagon and epinephrine
1292
tolerance tests cause a positive glucose response, with increases in levels of 15–23 mg/dl, the maximum response occurring about 30 minutes after hormone injection and 2 hours after a meal.274 Both hormones may allow detection of urinary ketone bodies. Hypoglycemia is not a characteristic feature of the illness until terminal liver failure occurs. Chronic and severe acidosis may occur secondary to a renal tubular defect in hydrogen ion excretion.275–277 Oral glucose and fructose tolerance tests show no abnormality, and serum lactate and pyruvate levels are normal.274
Biochemical Characteristics The stored polysaccharide in GSD-IV is a glucose polymer whose properties differ from those of normal mammalian glycogen. The normal level of liver glycogen in a child in the fed state is roughly 6% of the hepatic wet weight.288,289 Normal glycogen has the following characteristics: (1) at least 36% of the glucose units are susceptible to phosphorylase-catalyzed degradation; (2) chain lengths are 8–12 glycosyl units; (3) branch points (1,6 linkages) make up 8%; and (4) the KI:I2 absorption band is at 460 nm.21 Patients with GSD-IV usually have hepatic glycogen levels slightly lower than normal (3.5–5%), although 1 patient had a level of 10.7%. Hepatic glycogen in these patients is unusually susceptible to phosphorylase-induced hydrolysis (>40% phosphorylase degradation), suggesting that it has longer outer-chain lengths and fewer branch points (about 6%) than does normal glycogen. The polysaccharide is highly chromogenic, with a maximal KI:I2 absorption band at about 525 nm.290 About half the extracted polysaccharide may be insoluble and poorly characterized. Muscle glycogen appears normal; leukocytic glycogen is abnormal, as is brancher enzyme activity in these cells. Although the deficiency of branching enzyme explains the formation of an amylopectin-like polysaccharide, it does not account for the presence of an appreciable proportion of branch points in the abnormal glycogen. Brown and Brown suggest that normal liver contains two enzymes with branching activities of different specificities, only one of which is measured by the methods used.272 Another possibility is that the mutant gene in this disorder has produced a protein with substantially modified enzymatic specificity such that branching occurs mainly with long outer chains.273 The liver glycosyl deposits and myocardial deposits are believed to be biochemically the same, because they appear similar histochemically and are both resistant to digestion by a-amylase. This resistance to digestion by amylase is difficult to explain, because there is no other evidence for the existence of an a-amylase-resistant glycogen.
Pathology Examination of the liver shows a uniform micronodular cirrhosis with broad bands of fibrous tissue extending around and into the lobules. Portal veins, lymphatic channels, and hepatic arteries are normal with slight portal biliary duct proliferation. Liver cell plates are distorted, and, as the disease progresses, the lobules develop prominent sinusoidal channels with fibrous walls coursing between thick liver plates. Pale amphophilic or basophilic deposits occur in liver cells, cardiac and skeletal muscle, and brain.273,276,279 The deposits in cardiac muscle resemble cardiac colloid, and those in brain resemble
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Lafora’s bodies.218,221 Liver cell nuclei are frequently eccentric in position. They appear (with H&E stain) to be displaced by pale, slightly eosinophilic or colorless inclusions deposited in the cytoplasm. This is the most striking and characteristic finding and is generally limited to the periphery of the lobule. The inclusions vary from hyaline to reticulate and are usually sharply demarcated from normal cytoplasm. Clear haloes may surround the contents of the inclusions. In the late stages of the disease, nodular accumulations of a slightly different hyaline, fibrillar material are scattered throughout the hepatic lobules. This material is birefringent, appearing in polarized light as sheaths of crystals that cannot be easily distinguished from deposits that typify a1-antitrypsin deficiency, except perhaps on the basis of their greater frequency and larger size in GSD-IV. In both conditions, the peripheral lobular deposits are PAS-positive and diastase-resistant.288 Examination of myocardial tissue shows hypertrophied muscle fibers with large rectangular nuclei. Within the fibers, colorless discrete deposits similar to those in the liver are found. These deposits are uniformly distributed, with only a slight predilection for subendocardial regions. The epicardium and vessels are normal, without significant myocardial fibrosis, endocardial sclerosis, necrosis, or calcification. PAS-positive material may be present within foamy macrophages of the spleen and lymph nodes, smooth muscle of the gastrointestinal tract and large blood vessels, and skeletal muscle around the larynx, diaphragm, tongue, and esophagus. Peripheral skeletal muscle is usually free of any abnormality except for scant amounts of abnormal polysaccharide. Central nervous system abnormalities have been found in only 6 of 20 reported cases. In these instances, discrete, spherical PASpositive globules were widely scattered throughout the neuraxis. They were usually most prominent in white matter and in subependymal and subapical regions but were also numerous in gray matter. Peripheral nerves also contain PAS-positive globules in the endoneurium. The PAS-positive deposits in neuronal tissue are also resistant to amylase digestion.279 Electron microscopic studies show a decrease in cytoplasmic organelles. The organelles are found in tongues of cytoplasm extending between large, irregular aggregates of low electron density. The contents of these aggregates are variable, and they contain two or three components: glycogen rosettes or the alpha particles of Drochman fibrils and granular material. The fibrils are straight or curved and are about 65 nm wide.278 Some areas, which contain primarily glycogen, frequently fail to stain well despite the use of lead citrate. Myocardial fibers are distended by zones of material with low electron density, similar to those seen in the liver. Unlike liver cells, cardiac cells rarely contain material thought to be glycogen. Electron microscopic examination of skeletal muscle reveals granulofibrillar deposits similar to those in the viscera but less conspicuous; beta glycogen is more prominent in these deposits. Ultrastructural studies of the central nervous system tissue show the presence of large numbers of globules with a granulofibrillar appearance and of medium electron density. In some deposits, the fibrils are oriented radially, and in others, they assume a whorled appearance. The material is mostly restricted to astrocytic processes. It is not found within neuronal perikaryons or processes.274,279
Histochemical stains of the liver deposits indicate that the material is an abnormal glycogen with fewer branch points than usual. The deposits also have properties that are unusual for a glucose polymer: a positive colloidal iron stain and resistance to a-amylase digestion of conventional duration. Only pectinase was able to reduce the PAS and colloidal iron reactions significantly.
Molecular Basis of Type IV Glycogen Storage Disease GSD type IV is a rare autosomal recessive disorder caused by deficiency of glycogen branching enzyme (GBE), leading to accumulation of amylo-pectin-like compounds in various tissues. The human GBE cDNA is 3 kb in length, and encodes a 702-amino-acid protein. GBE gene is located on chromosome 3p14 and has 16 exons. Mutations in GBE have been reported.289
Treatment Three types of treatment, without decided improvement, were used for 1 patient.290 First, a high-protein, low-carbohydrate diet with corn oil added to the milk fat caused no change in weight gain or in the progression of cirrhosis. Second, with progression of the disease, treatment with purified a-glucosidase from Aspergillus niger was given for 6 days. This treatment resulted in a striking decrease in hepatic glycogen content, from 10 to 2%. Although no unfavorable reaction occurred, liver size did not decrease. Glycogen content was maintained at 3% by a third treatment, intramuscular injection of zinc-glucagon 1 mg three times a day for 24 days. Any positive effect of these treatments remains doubtful. On the other hand, the poor clinical results following the treatments might have been related to the advanced state of cirrhosis before their initiation. Other than supportive nutritional management286 for terminal cirrhosis, no specific treatment appears to be beneficial. The finding that the Aspergillus extract caused a striking decrease in hepatic glycogen suggests that, if the accumulation of abnormal glycogen is in some way hepatotoxic, this form of treatment might be studied in patients before the onset of severe cirrhosis. A study by Starzl’s group suggests that liver transplantation in GSD-IV results in resorption of extrahepatic deposits of amylopectin, possibly by systemic microchimerism (i.e., cells of the host organs became mixed with cells with the donor genomes that had migrated from the allograft into the recipient tissues and presumably serve as enzyme carriers).291
INBORN ERRORS OF AMINO ACID METABOLISM Hereditary tyrosinemia is the only inborn error of amino acid metabolism that results in permanent liver injury. This section summarizes the normal metabolic pathway of tyrosine, and then discusses the disorders associated with abnormal tyrosine metabolism.
TYROSINE METABOLISM The principal hepatic pathway for degradation of tyrosine is shown in Figure 69-11. These reactions normally catabolize 99% of tyrosine. The steady-state plasma concentration of tyrosine is
1293
Section XII. Inherited and Pediatric Diseases of the Liver
CH2CHOOH NH2
Phenylanine
Phenylanine hydroxylase
1
CH2CHOOH
HO
H Tyrosine transaminase
2
d
hy
De
HO
CH2C
De
ca
rbo
xy
las
CH2C
COOH
e
HO
p-Hydroxyphenylpyruvic oxidase
HO
HO
OH p-Hydroxyphenyllactic acid
COOH
O p-Hydroxyphenylpyruvic acid 3
se
na
e rog
CH2COOH
p-Hydroxyphenylacetic acid
OH CH2COOH
Homogentisic acid Homogentisic oxidase O
4
HOOC
C
C
C
CH2CCH2COOH
H H O Maleylacetoacetic acid Maleylacetoacetic isomerase H O HOOC
C
C
C
CH2CCH2COOH
H O Fumarylacetoacetic acid 5
H HOOC
C
C
Fumarylacetoacetase O
COOH
H Fumaric acid
CH2CCH2COOH Acetoacetic acid
Figure 69-11. Metabolic pathway of phenylalanine and tyrosine. //, block in the metabolic pathway; 1, block in phenylketonuria; 2, block in persistent hypertyrosinemia; 3, block in Medes’s tyrosinemia patient; 4, block in alkaptonuria; 5, block in hereditary tyrosinemia.
determined by two primary factors: (1) gastrointestinal uptake, which is regulated by an active transport system; and (2) the rate of production of tyrosine from phenylalanine and its subsequent catabolism to carbon dioxide and water. Most of the ingested phenylalanine is catabolized via tyrosine. The rate-limiting reaction in the tyrosine oxidation pathway is that catabolized by tyrosine aminotransferase (Figure 69-11, reaction [1]).292,293 Pyridoxal phos-
1294
phate is the coenzyme. Tyrosine aminotransferase is mainly found in the cytosol of the liver and its activity shows a circadian rhythm. Also, the activity of tyrosine aminotransferase is induced by various compounds, including corticosteroids.294–296 p-Hydroxyphenylpyruvic hydroxylase is the second enzyme involved in tyrosine catabolism. This enzyme is found in the cytosol of the human liver and kidney.297,298 It catalyzes the conversion of
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
p-hydroxyphenylpyruvic acid to homogentisic acid (Figure 69-11, reaction [2]) and converts phenylpyruvic acid to p-hydroxyphenylacetic acid.299 The enzyme requires a reducing agent and ascorbic acid can serve both in vivo and in vitro in this capacity.300 p-Hydroxyphenylpyruvic acid is not normally found in the urine. When its normal catabolism to homogentisic acid is blocked, the levels of both the a-keto acid and tyrosine in the plasma may be increased. At the same time, p-hydroxyphenyl lactic acid is formed through the action of lactic dehydrogenase. p-Hydroxyphenylacetic acid can also be formed from decarboxylation of p-hydroxyphenylpyruvic acid. Homogentisic acid oxidase catalyzes the conversion of homogentisic acid to maleylacetoacetic acid. This reaction also requires vitamin C for maximum activity in vivo. Fumaryl acetoacetase acts on maleylacetoacetic acid to yield fumaric acid and acetoacetic acid.
TRANSITORY TYROSINEMIA OF THE NEWBORN
DISORDERS OF TYROSINE METABOLISM
This condition affects approximately 30% of premature infants and as many as 10% of full-term infants.303–305 It is assumed that the hepatic enzymes catalyzing the early steps of tyrosine metabolism are not well developed in these infants.306,307 Administration of vitamin C, which is known to protect p-hydroxyphenylpyruvic oxidase from unphysiologic levels of its own substrate, usually corrects the transitory tyrosinemia.308 The transitory tyrosinemia in infants usually disappears within a few weeks but occasionally persists for several months. A reduction in protein intake to 1.5–2 g/kg per day with administration of vitamin C should correct the condition. Although it is generally assumed that transitory tyrosinemia is harmless, this may not be true, because persistent hypertyrosinemia (discussed later) is regularly associated with mental retardation. There is a definite need for prospective studies of infants – specifically premature infants with transitory hypertyrosinemia.
Several conditions associated with abnormalities in tyrosine metabolism have been described; however, only hereditary tyrosinemia (hepatorenal type) is associated with permanent injury to the liver. Other abnormalities in tyrosine metabolism, summarized in Table 69-3, are discussed briefly.
PERSISTENT HYPERTYROSINEMIA (TYROSINEMIA II, RICHNER–HANHART SYNDROME)
TYROSINOSIS The first known case of abnormal excretion of a tyrosine metabolite was described by Medes and colleagues in 1927.301 The patient, a 49-year-old Russian Jew who had myasthenia gravis, was found to have an unusual reducing substance in the urine. The reducing compound was later isolated and identified as p-hydroxyphenylpyruvic acid, the a-keto acid of tyrosine. The condition was named tyrosinosis by Medes.302 When the patient was fasting, the urine contained 1.6 g p-hydroxyphenylpyruvic acid in a 24-hour period. This quantity was considered to represent endogenous production from the catabolism of endogenous protein. When the patient was fed a regular diet, the amount of urinary p-hydroxyphenylpyruvic acid doubled, and tyrosine could also be isolated from his urine. One of the interesting findings was that feeding large amounts of tyrosine also led to the excretion of 3,4-dihydroxyphenylalanine (dopa), a product of the minor pathway of tyrosine metabolism. Medes postulated a defect in the conversion of p-hydroxyphenylpyruvic acid to homogentisic acid. No other case of tyrosinosis has been described.
Several patients with persistent tyrosinemia without associated hepatic or renal disease have been described.309–314 The patients were reported to have cataracts, corneal ulcers, keratotic skin lesions, and neuropsychiatric abnormalities. Enzymatic studies were carried out in 4 patients. A total deficiency of cytosolic tyrosine transaminase was found in 2;311 however, the activity was only reduced in 2 others.315 The keratotic lesions abated in response to low-phenylalanine, low-tyrosine diets in 2 cases.313,316 Large doses of vitamin C had no effect on the tyrosinemia in those patients.
HYPERTYROSINEMIA SECONDARY TO LIVER DISEASE Patients with hepatic cirrhosis have a reduced capacity to metabolize tyrosine and other amino acids.317 Cirrhotic patients have significantly increased fasting levels of plasma tyrosine and basal levels of p-hydroxyphenylpyruvic acid. They have impaired tolerance to oral loading doses of tyrosine, p-hydroxyphenylpyruvic acid, and homogentisic acid.318 These findings suggest generalized partial defects in tyrosine transaminase, p-hydroxyphenylpyruvic acid oxidase, and homogentisic acid oxidase enzymes in patients with
Table 69-3. Conditions Associated with Hypertyrosinemia Enzyme deficiency
Clinical features
Transitory tyrosinemia of the newborn Tyrosinosis (Medes’s patient)
p-Hydroxyphenylpyruvic acid oxidase p-Hydroxyphenylpyruvic acid oxidase
Persistent tyrosinemia Hereditary tyrosinemia (hepatorenal type)
Cytosol tyrosine transaminase in one case Fumaryl acetoacetase
None Mental retardation Skin and eye lesions No hepatic or renal disease Cirrhosis, hepatomas Renal tubular defects
Hypertyrosinemia secondary to liver disease
Generalized partial defects of tyrosine transaminase, p-hydroxyphenylpyruvic acid oxidase, and homogentisic acid oxidase
1295
Section XII. Inherited and Pediatric Diseases of the Liver
cirrhosis. Levels of tyrosine are also slightly elevated in other diseases, including cystic fibrosis,319 hypoxia with respiratory failure,320 and rheumatoid arthritis.321
HEREDITARY TYROSINEMIA (HEPATORENAL TYROSINEMIA) In 1956, Baber reported the case of a 9-month-old infant with failure to thrive, abdominal distention, and diarrhea. The child was found to have cirrhosis of the liver, a renal tubular defect with gross aminoaciduria of a distinct type, and vitamin D-resistant rickets.322 In 1957–1959, Sakai and co-workers reported the case of another patient with a similar clinical picture and marked p-hydroxyphenyl lactic aciduria. They drew attention to the abnormal metabolism of tyrosine in this patient and named the condition atypical tyrosinosis.323–325 Since then, many cases have been reported from Norway,326 Canada,327,328 Sweden,329 and the USA.330,331 Although many names have been given to the disorder in these cases, hereditary tyrosinemia is generally accepted.
Clinical Features Hereditary tyrosinemia may be either acute or chronic.327,332 Symptoms appear in the first month with acute disease, and the patient usually dies with hepatic failure within the first 3–9 months.327 In the chronic form of the disease, the symptoms appear later. The lifespans of these patients are longer than those of patients who have the acute form.327–332 Both forms can occur within the same family. The main symptoms are failure to thrive, vomiting, diarrhea, anorexia, hepatosplenomegaly, ascites, edema, jaundice, bruising, and rickets.326 Some patients have slight mental deficiencies. In the chronic form of the disease, hepatoma may develop. A review of the literature by Weinberg and colleagues in 1976 disclosed 16 cases of hepatoma in 43 patients surviving beyond 2 years of age. This incidence is higher than that in adults with macronodular cirrhosis.333
Laboratory Findings Elevated serum levels of AST and ALT are common and total and direct bilirubin were elevated in 19 of 20 patients so tested.326 However, total serum bilirubin levels were usually less than 10 mg/ dl until the appearance of signs of liver failure, at which time they became markedly elevated. Synthetic function of the liver is also disturbed, as evidenced by low levels of serum albumin and vitamin K-dependent clotting factors. Hematologic studies show mild anemia, elevated reticulocyte counts, and normal serum iron levels. Bone marrow aspirates show erythroid hyperplasia. These findings suggest a hemolytic anemia, which may be related to an intracorpuscular defect secondary to abnormal accumulation of tyrosine metabolites.334 In the chronic form of the disease, anemia, leukopenia, and thrombocytopenia are present secondary to hypersplenism. A tendency to develop hypoglycemia has been found in association with hereditary tyrosinemia.334,335 Marked hyperplasia of the islets of Langerhans is a rather constant finding, but insulin levels are reported to be normal. The responses to epinephrine and glucagon of 2 patients thus examined showed flat curves.334 These findings suggest a disturbance in the release of glucose from
1296
glycogen; however, hepatic glycogen content and structure and glycogenolytic enzymes are normal and the glycolytic enzymes are also normal. The laboratory findings related to the renal tubular defects include hyperphosphaturia, glucosuria, proteinuria, and gross aminoaciduria. Biochemical evidence of rickets secondary to hypophosphatemia is seen in almost all cases.327 The urinary excretion of amino acids follows a distinctive pattern.336 The aminoaciduria is more pronounced in the acute disease. The pattern of urinary excretion of tyrosine and phenolic acids (p-hydroxyphenyl lactic, p-hydroxyphenylpyruvic, and p-hydroxyphenylacetic acids) by tyrosinemic patients is similar to that described for premature infants with transitory tyrosinemia.337 Chromatography of the serum amino acids reveals slight hypertyrosinemia and, frequently, hypermethioninemia. Other amino acids, including phenylalanine, are occasionally slightly elevated. Some of these results must be interpreted with caution, because liver damage may cause modest elevations of these amino acids. In 6 patients with hereditary tyrosinemia, urinary d-aminolevulinic acid levels were elevated to as much as 100-fold above control values. In 2 of these cases, attacks of abdominal pain and paresis of a peripheral type, resembling acute intermittent porphyria, prompted investigation for porphyria. The amounts of porphobilinogen and porphyrins excreted were normal or slightly elevated, but d-aminolevulinic acid levels were markedly elevated.338 Similar abnormal pyrrole metabolism was found by Kang and Gerald in a patient with hereditary tyrosinemia. d-Aminolevulinic acid synthetase activity in liver tissue was increased331 and it was believed that the abnormality of pyrrole metabolism is a secondary process related to induction of d-aminolevulinic acid synthetase activity by one of the accumulated metabolites of tyrosine. However, Lindblad and colleagues339 and Melancon and associates340 have shown that accumulation of succinyl acetoacetate inhibits porphobilinogen synthetase (d-aminolevulinic acid dehydratase), leading to increased excretion of d-aminolevulinic acid. This observation has been confirmed by Berger and co-workers and thus the cause of the high levels of d-aminolevulinic acid is unclear.341
Biochemical Features Patients with hereditary tyrosinemia are reported to lack or to have markedly reduced activity of p-hydroxyphenylpyruvic acid oxidase in liver and kidneys.342,343 There is reason to question whether deficiency of this enzyme can account for the clinical manifestations in patients with hereditary tyrosinemia. Thus, activity of p-hydroxyphenylpyruvic acid oxidase is reduced in 30% of premature infants, 10% of full-term infants, and patients with persistent hypertyrosinemia (but these patients have no derangement of function of the liver or kidneys). Perry and colleagues induced experimental hypertyrosinemia in vitamin C-deficient newborn guinea pigs fed diets containing large amounts of tyrosine. There was no evidence of a liver or kidney disturbance.344 Gaull and associates therefore proposed that deficiency of p-hydroxyphenylpyruvic acid oxidase is not the primary defect in patients with hereditary tyrosinemia. These researchers believe that there are deficiencies in methionineactivating enzyme and cystathionine synthetase in affected patients and that the signs and symptoms of the disease reflect these deficiencies.345,346
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Feeding large amounts of methionine to guinea pigs in fact produced a syndrome similar to that in infants with hereditary tyrosinemia. The animals showed hypertyrosinemia, hypermethioninemia, generalized aminoaciduria, hypoglycemia, and pancreatic islet cell degeneration.344 Hence, some metabolite of methionine may be the toxic factor responsible for the pathologic and biochemical changes in hereditary tyrosinemia. Some findings, however, do not support this idea: 1. Hypertyrosinemia usually precedes hypermethioninemia in infants with hereditary tyrosinemia. 2. Transient neonatal hypermethioninemia is an apparently benign condition unassociated with tyrosinemia. 3. Weanling rats fed high-methionine diets have increases of methionine, taurine, and alanine, but not tyrosine, in the serum.347 An apparently identical clinical and biochemical picture resembling hereditary tyrosinemia has occasionally been found in patients with hereditary fructosemia348–350 but not in galactosemia, although severe liver damage is noted in both conditions. Plasma levels of tyrosine and methionine and the excretion of phenolic acid decreased markedly with the exclusion of fructose from the diets of patients with hereditary fructosemia.349 Lindblad and associates339 reported increased excretion of succinyl acetone and succinyl acetoacetate in the urine of patients with hereditary tyrosinemia. These compounds presumably originate from maleylacetoacetate or fumarylacetoacetate or both. Their accumulation indicates a block in metabolism of tyrosine at the fumarylacetoacetase reaction (Figure 69-11). These observations have been confirmed by Melancon and colleagues.340 Since then, several groups have confirmed the defect in fumarylacetoacetase in liver tissues from patients with hereditary tyrosinemia.351–355 Additionally, the activity of fumarylacetoacetase was found to be markedly decreased in patients with the acute form of hereditary tyrosinemia. It is reasonable to conclude that severe liver and kidney damage in hereditary tyrosinemia is secondary to accumulation of succinyl acetone and succinyl acetoacetate, which can bind to the SH group of proteins, thereby destroying their function. The liver and kidneys would be the organs principally affected by these metabolites, because these tissues are the only ones with p-hydroxyphenylpyruvate hydroxylase activity – that is, metabolism of tyrosine to potentially toxic metabolites depends on the presence of this enzyme. Of interest in this regard are the findings of Lindblad, indicating that patients with the more benign form of hereditary tyrosinemia have lower levels of activity of p-hydroxyphenylpyruvate hydroxylase in their livers than do patients who have more serious forms of the illness. The cause of the low activity of p-hydroxyphenylpyruvic hydroxylase in hereditary tyrosinemia remains to be determined. The possibility of a defect in a regulatory gene common to p-hydroxyphenylpyruvic hydroxylase and fumarylacetoacetase can be considered. Continued search for other possible biochemical defects that may be responsible for all the clinical and biochemical features is needed.
Molecular Basis of Hereditary Tyrosinemia The gene responsible for hereditary tyrosinemia fumaryl acetoacetate hydrolyase (FAH) has been cloned and mapped to chromo-
some 15q23-25. The cDNA encodes for a 419-amino-acid cytosolic homodimer protein, which is present in the liver, kidney, lymphocytes, erythrocytes, fibroblast, and chorionic villi.356 Several mutations in the FAH have been described. The IVS12+5GÆA allele accounts for over 95% of mutant FAH allele in the SaguenayLac St-Jean area of Quebec.357,358 Certain populations have specific mutations, including W262x in Finns359,360 and Q64H in Pakistanis.361
Pathology
The major pathologic findings are in the liver and kidneys.362–364 Macroscopically, the liver is enlarged, firm, and nodular. The kidneys are enlarged, with poor architectural demarcations. Microscopically, the architecture of the liver is distorted by extensive fibrosis and infiltration of the portal areas by lymphocytes and plasma cells. The liver cell cords have a pseudoglandular appearance. The hepatocytes show fatty metamorphosis, and some may undergo acidophilic degeneration and, occasionally, giant-cell transformation. Glycogen is either lacking or markedly decreased.362–364
Genetics Tyrosinemia has been found with increased frequency in French Canadians. Inheritance is autosomal recessive. The carrier rate in north-eastern Quebec is 1 : 14, for an estimated frequency of 14.6 cases per 10 000 population.365 An automated fluorometric test for hypertyrosinemia has been described,366 and in the province of Quebec, a neonatal screening program has been developed. The association of increased serum levels of alpha-fetoprotein with hypertyrosinemia distinguishes patients with hereditary tyrosinemia from patients with transient hypertyrosinemia of the newborn.367 Prenatal diagnosis has been accomplished by measurement of fumarylacetoacetase in cultured amniotic fluid cells.368
Treatment A diet low in phenylalanine and tyrosine has been used in management of the acute and chronic forms of the disease by several groups of investigators.326,330,369–371 This diet decreases serum tyrosine levels. It increases serum phosphorus secondary to enhanced renal tubular reabsorption of phosphorus.326 Similar beneficial effects on renal function are reflected by reductions of glycosuria, hyperaminoaciduria, and proteinuria. The effect of diet on hepatic dysfunction is uncertain. In only 1 acute case of tyrosinemia were reductions of fibrosis and infiltration of the liver by inflammatory cells found.372 Other patients did not respond to the diet in this manner.328,371,373 The available evidence suggests that a diet low in phenylalanine and tyrosine should be given to patients with the acute and chronic forms of the disease. Signs of deficiency of phenylalanine and tyrosine should be monitored in patients receiving this diet. The amounts of phenylalanine and tyrosine used in such diets are approximately 25 mg/kg per day of each; however, there is some variation in the optimal minimum requirements for individual patients. In one case, dietary treatment returned the elevated levels of phenylalanine and tyrosine to normal in the serum of a patient with hereditary tyrosinemia. However, the patient continued to have hypermethioninemia and clinical evidence of liver disease. Strict control of his dietary intake of methionine, as well as phenylalanine
1297
Section XII. Inherited and Pediatric Diseases of the Liver
and tyrosine, returned all serum amino acid levels to normal and eliminated the hepatic abnormalities.374 Large doses of vitamin D (10 000–15 000 U/day), together with dietary restrictions of phenylalanine and tyrosine, are necessary to correct the rickets seen in patients with hereditary tyrosinemia. Other recent treatment modalities pioneered by the Quebec investigators include utilization of 2-(2-nitro-4-trifluoro-methylbenzoyl)-1,3-cyclohexanedione (NTBC). NTBC appears to block the conversion of 4-hydroxyphenyl-pyruvate to homogentisate and to maleylaceto-acetate. The treatment protocol includes maintenance of plasma tyrosine level below 400 mmol/l by dietary phenylalanine and tyrosine restriction. NTBC is given at a starting dose of 1 mg/kg per day in two divided doses and adjusted to achieve plasma NTBC concentration of greater than 50 mmol/l and no detectable urinary and blood succinylacetone. In asymptomatic patients diagnosed by neonatal screening, none developed hepatic nodules and no acute liver crises have been seen. However, in patients who present with liver disease, progression to cirrhosis was noted, requiring transplantation.375 The major complication of NTBC treatment has been corneal crystals, photophobia, and ocular inflammation. Hepatocellular carcinoma has developed in FAH-deficient mice despite NTBC treatment.376 Liver transplantation has been carried out in 4 patients with chronic hereditary tyrosinemia and hepatoma. All were reported to be alive and well 3 months to 3 years after transplantation. Therefore, this modality of therapy needs to be considered in patients who develop hepatoma as a complication from their chronic hereditary tyrosinemia.377 A follow-up of 10 patients who underwent liver transplantation for tyrosinemia suggests the feasibility of this therapy.378 The indications for liver transplantation were hepatoma in 3, acute liver failure in 2, and progressive chronic liver disease in 5. One patient died during surgery. Of the remaining 9, 7 patients are alive 6 months to 61/2 years after transplantation. Two died of complications.
Prognosis Patients who have acute hereditary tyrosinemia die in the first few months of life. Unfortunately, this is the commonest form of the disease.327,332 In patients with the chronic form, the disease progresses slowly, with the eventual development of cirrhosis.
Hepatoma has been reported to occur in 37% of patients with chronic hereditary tyrosinemia.333
INBORN ERRORS OF LIPID METABOLISM INTRACELLULAR METABOLISM OF CHOLESTEROL Metabolism of cholesterol and cholesterol esters has been studied primarily in cultured human fibroblasts. Study of patients with inborn errors of lipid metabolism has provided information that tends to support most of the general hypotheses developed from growing fibroblasts in cultures. A brief discussion of the origin, transport, and degradation of free cholesterol and cholesterol esters is included to provide a basis for understanding the metabolic consequences of a deficiency of lysosomal acid lipase. Peripheral cells can synthesize free cholesterol and cholesterol esters, but they derive most cholesterol from exogenous sterols circulating in the serum as low-density lipoproteins. Intracellular levels of cholesterol are the primary regulators of the rate of synthesis of cholesterol and of uptake from low-density lipoproteins. The metabolism of cholesterol esters is illustrated schematically in Figure 69-12. The initial event in the intracellular metabolism of low-density lipoproteins (i.e., particles enriched in cholesterol esters) is binding to receptors on the cell surface. Low-density lipoprotein receptors bind only those human plasma lipoproteins that contain apolipoproteins B and E (e.g., low-density lipoprotein and very-low-density lipoprotein). After low-density lipoprotein is bound to its specific receptor site, the lipoprotein remains metabolically inactive until the endosomes fuse with lysosomes. At this point, the protein component is hydrolyzed by lysosomal enzymes to products that mostly consist of free amino acids and small peptides (molecular mass <1000). The cholesterol ester component of low-density lipoprotein is hydrolyzed by a lysosomal acid lipase. The liberated cholesterol is then available for metabolic use by the cell.379 A consequence of the uptake and storage of low-density lipoprotein cholesterol is that the synthesis of cholesterol in non-hepatic tissues is maintained at a low level.
Figure 69-12. Metabolism of cholesterol esters. 3-Hydroxy-3-methylglutaryl CoA reductase
Free cholesterol
Low-density lipoprotein cell binding
Low-density lipoprotein Cholesterol esters-
Lysosomal hydrolysis
1298
Acetyl CoA cholesterol acyltransferase Membrane formation
Amino acids Linoleate
Cholesterol oleate storage
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Patients with isolated deficiencies of acid hydrolase activity (Wolman’s disease) should develop predictable changes in tissue lipids. For example, cholesterol ester as cholesterol linoleate should accumulate in lysosomes of affected individuals. This has been shown to occur.310,311 Because of a reduced rate of generation of free cholesterol within cells, there should also be decreased suppression of the activity of 3-hydroxy-3-methylglutaryl CoA reductase and reduced activation of endogenous cholesterol acyltransferase. These two events result in increased cholesterol synthesis by the cell and decreased cholesterol esterified with oleic acid, respectively. These abnormalities have been shown to occur in patients with cholesterol ester storage disease.380,381 However, patients with Wolman’s disease have additional abnormalities such as accumulation of oxygenated steryl esters, which are difficult to explain solely on the basis of an isolated deficiency of acid hydrolase.382 Similarly, patients with Wolman’s disease have unexplained steatorrhea, which may be secondary to defects in the metabolism of bile acids.
Molecular Basis of Lysosomal Acid Lipase Disorders (Wolman’s Disease and Cholesterol Ester Storage Disease) Deficiency of lysosomal acid lipase leads to either Wolman’s disease or the more benign cholesterol ester storage disease. Both disorders are inherited as autosomal recessive diseases and are associated with reduced activity and genetic defects of lysosomal acid lipase. The gene for lysosomal acid lipase has been cloned and mapped to chromosome 10q24-q25. The gene is expressed in most tissues with high expression in the hepatocytes, splenic and thymic cells, small intestinal villus cells, and adrenal cortex.383 Homozygotes for exon-8 splice junction mutation resulting in incomplete exon skipping has been described in many Wolman’s disease patients.384 Cholesterol ester storage disease is distinct from Wolman’s disease in that at least one mutant allele has the potential to produce some residual enzymatic activity to ameliorate the phenotype. Thus, in the majority of cholesterol ester storage disease cases, a single splicing mutation occurs in one allele.385 A knockout mouse model of Wolman’s disease has been produced by targeted disruption of the lysosomal acid lipase gene.386
WOLMAN’S DISEASE Enzymatic Abnormalities In 1956, Abramov and colleagues reported an infant who died at 2 months of age after a short illness characterized by abdominal distention, severe vomiting, diarrhea, hepatosplenomegaly, and radiographic evidence of calcification of the adrenal glands.387 In 1961, Wolman and co-workers reported the cases of two siblings of their first patient who were found to have the same illness.388 Accumulations of cholesterol esters and triglycerides were found in the liver, spleen, lymph nodes, and adrenal glands. The disorder was reported under the title generalized xanthomatosis with calcified adrenals.388 In 1965, Crocker and associates described 3 patients with this disorder as having Wolman’s disease; since then, the latter term has been widely accepted.389 In 1969, Patrick and Lake confirmed that the accumulated cholesterol was esterified and demonstrated an acid hydrolase deficiency in the livers and spleens of those
patients.390 The activity of acid hydrolase was also deficient in cultured fibroblasts of patients with Wolman’s disease.391 Deficiency of acid hydrolase or acid esterase manifests in two phenotypic forms. Wolman’s disease represents the clinically acute and severe form. Affected patients die in the first year of life. Cholesterol ester storage disease represents a chronic form of deficiency of acid hydrolase. Patients who have this latter disorder do not have adrenal calcifications, and they live to adulthood.380 Forty patients with Wolman’s disease and more than 20 patients with cholesterol ester storage disease have been described so far.
Clinical Features The majority of patients have similar clinical courses: projectile vomiting, diarrhea, abdominal distention, and failure to thrive, with the first noticeable symptoms appearing in about the second week of life.387,388 Some patients are jaundiced.389,392 Neurologic development of these infants is not normal. A decrease in activity may be noticed in the second month of life. It is not clear whether these symptoms are related to severe malnutrition or reflect neurologic defects. Konno et al. reported the case of an infant who had exaggerated tendon reflexes, ankle clonus, and opisthotonos.392 Electroencephalography was reported to show no abnormality in this patient or several others.381,393 On physical examination, a patient is usually slightly feverish, irritable, and wasted. Nonspecific skin eruptions may be observed. Abdominal distention with hepatosplenomegaly is noted in all patients; lymphadenopathy is detected in some. Subsequently progressive hepatosplenomegaly and abdominal distention, fever, and vomiting and diarrhea occur, persisting until the death of the patient, usually in the first 6 months of life.387,388
Laboratory Findings Anemia usually appears early in the course of the disease, and hemoglobin levels progressively decrease to as low as 5 g/dl. Peripheral blood lymphocytes show intracytoplasmic and intranuclear vacuolation.389,392,393 Acanthocytosis was found in a Japanese infant.392 Lipidladen histiocytes or foam cells are seen in bone marrow aspirates and in the peripheral blood.387 Total serum bilirubin and the conjugated fraction are elevated in some patients.388,389,392 Some have elevated serum levels of AST and ALT.389,392 Steatorrhea is present, evidenced by a high fecal fat content389,392,393 and an abnormal 157I-triolein absorption test.393 The most consistent diagnostic finding, present in all reported cases of Wolman’s disease, is calcification of the adrenal glands. The adrenal glands are symmetrically enlarged and show extensive punctate calcifications (Figure 69-13).387,388 This is the only disease that causes such calcification of the adrenal glands. In all other conditions, the calcifications are scattered. In most patients, the responses of the adrenal glands to adrenocorticotropic hormone stimulation are depressed. Plasma cholesterol and triglycerides are usually normal in patients with Wolman’s disease.388,389,393–395 Triglycerides and very-low-density lipoproteins were elevated in 2 patients.394 It is expected that low-density lipoprotein levels should be high, because the acid hydrolase deficiency decreases cellular metabolism of low-density lipoproteins. Some patients, however, had hypolipoproteinemia.393 The severe inanition and reduced production of
1299
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 69-13. Radiograms showing calcification of the enlarged adrenal glands in a patient with Wolman’s disease. (A) Patient in the supine position; (B) the adrenals after removal at autopsy. (Reproduced from Crocker AC, Vawter GF, Neuhauser EB, et al. Wolman’s disease: three new patients with a recently described lipidosis. Pediatrics 1965; 35:627–640, with permission.)
B
A
lipoproteins by the liver probably offset the tendency toward hyperlipoproteinemia.
Table 69-4. Tissue Lipid Analysis in Patients with Wolman’s Disease
Biochemical Characteristics Most of the enzymes that hydrolyze fatty acid esters are active at neutral or alkaline pH. An ester hydrolase active at acidic pH (4.0–5.6) has been found in lysosomes and cell membranes.390 This enzyme requires no co-factors and uses triglycerides and cholesterol ester as substrates. Lysosomal acid lipase/cholesteryl ester hydrolyase has been purified,396 and the cDNA encoding the human enzyme has been cloned and expressed.397 Accumulation of cholesterol esters apparently results from the deficiency of cholesterol ester hydrolase activity.398 Accumulations of triglycerides and other lipid products in large quantities are more difficult to explain. These biochemical abnormalities are discussed in more detail in the section on cholesterol ester storage disease. Table 69-4 shows the results of analyses of lipids in various tissues in Wolman’s disease. The triglyceride content is severalfold greater in the liver and as much as 100-fold greater in the spleen of a patient with Wolman’s disease than in comparable tissues of controls.390 The total cholesterol in liver and spleen is elevated in every case of Wolman’s disease. The majority of the increase is in the cholesterol ester fraction. Similar accumulations of triglycerides and cholesterol esters have been found in the bone marrow, thymus, and lymph nodes. Slight increases were also detected in the lungs and kidneys.394 Cholesterol and triglyceride contents of the brain were reported to be elevated in some patients;389,394 however, they were normal in others.392 That severe malabsorption is seen in Wolman’s disease but not in cholesterol ester storage disease could be explained by the differences in bile acid metabolism in these two diseases. Total serum bile acid levels were either normal or increased in patients with cholesterol ester storage disease.381 Assmann and colleagues found oxygenated steryl esters in Wolman’s disease but not in cholesterol ester
1300
Liver, wet weight (mg/g) Upper limit of normal Spleen, weight (mg/g) Upper limit of normal
Triglyceride
Total cholesterol
Free
Esterified
137 61 44 20
97 170 32 5
38 9 14 4
59 161 18 1
8 99 29 1
26 35 8 5
10 8 4 4
16 27 4 <1
storage disease. The accumulation of these oxygenated esters would suggest a defect in bile acid synthesis in Wolman’s disease.382 Boyd suggested that cholesterol linoleate may be the precursor in 7a-hydroxycholesterol formation.399 If the pathway from 7ahydroxycholesterol to bile acids were blocked, an accumulation of oxygenated steryl esters would occur. These speculations await confirmation by the study of bile acid metabolism in Wolman’s disease and the finding of such a defect in bile acid synthesis in Wolman’s disease would explain the steatorrhea.
Pathology Liver Enlargement of the liver is a constant finding and it is firm and appears yellow on gross examination. The hepatic architecture is usually distorted. The hepatic parenchymal cells show marked steatosis. Sinusoids are plugged by swollen histiocytes with foamy, vacuolated cytoplasm. Kupffer cells are distended and vacuolated. The portal areas are enlarged, with increased fibrosis that extends
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
to the periportal areas. The fibrosis may be extensive, and a classic picture of hepatic cirrhosis may be seen.387,388,394,395 Electron microscopic examination of the liver shows that parenchymal organelles are distorted by the accumulation of large osmiophilic lipid droplets, which are neutral lipids. These droplets are seen mostly in the lysosomes. The smooth endoplasmic reticulum and rough endoplasmic reticulum appear distended but are usually empty. The Kupffer cells are so distended that they almost obstruct the sinusoids.395
Spleen and Lymph Nodes The spleen is greatly enlarged, is firm, and appears mottled with yellowish flecks. Microscopic section shows that the reticular cells are transformed into large foam cells. The lymphatic follicles are atrophied and compressed. The lymph nodes are enlarged, firm, and yellowish. Microscopically, the lymph nodes are similar to the spleen. The same changes are also seen in the bone marrow and thymus.389,394
Intestines
acid hydrolase in the leukocytes of the parents and a sibling of 1 patient had half the normal activity.398 Kyriakides and associates reported similar findings in skin fibroblasts from the parents of a patient.391 A prenatal diagnosis using cultured amniotic fluid cells has been described.400
Differential Diagnosis Calcifications of the adrenal glands, invariably present in Wolman’s disease, can be seen in conditions such as Addison’s disease, adrenal hemorrhage, neuroblastoma, ganglioneuroma, adrenal cysts, pheochromocytoma, cortical carcinoma, and adrenal teratoma.387,401 However, the calcifications in Wolman’s disease characteristically outline the shape of the adrenal gland. Niemann–Pick disease may result in gastrointestinal symptoms, hepatosplenomegaly, and failure to thrive in early infancy, but adrenal calcifications are absent. The definitive diagnosis of Wolman’s disease depends on the clinical picture and assay of acid hydrolase in cultured leukocytes or skin fibroblasts, using p-nitrophenyllaurate as a substrate.402 Acid hydrolase activity that is less than 5% of normal confirms the diagnosis.
Treatment and Prognosis
The small intestine is usually thickened and has a yellowish serosal surface. The mucosa is greasy, yellowish, and granular. Microscopic sections show thick, flattened villi with numerous foamy, lipid-filled histiocytes in the lamina propria. Some of the foam cells extend into the muscularis mucosa. These changes are most evident in the proximal intestine or occasionally in the colon.389,394
Several therapeutic approaches have been tried, including TPN403 and bone marrow transplantation.404 A diet free of hydrophobic esters in which cholesterol and the essential fatty acids are bound to protein has been proposed by Wolman.405
Adrenal Glands
CHOLESTEROL ESTER STORAGE DISEASE Clinical Picture
Both glands are symmetrically enlarged, firm, and difficult to cut. The cortex on cut sections is yellowish whereas medially, the tissue is whitish. Microscopic sections of the glands show preservation of the architecture of the cortex. Many of the cells are swollen and vacuolated and contain sudanophilic lipid. Some of the foam cells are necrotic and appear as lipid cysts. Most of the calcium deposition occurs in a finely granular pattern; however, in some regions it may be condensed to form crystalline bumps.394 Extensive fibrosis is found in the inner cortical areas. The medullary cells are normal. Electron microscopic sections show that the innermost part of the adrenal cortex is necrotic, with calcification. The histiocytes are filled with large amounts of both crystalline and droplet lipid. Histochemical analysis of the lipid indicates the presence of cholesterol esters and triglycerides.394,395
Other Organs The vascular endothelium shows lipid deposition, but frank atherosclerosis is not seen. Foam cells have been observed in the intestinal tissue and in the lungs, thyroid, testes, ovaries, leptomeninges, Purkinje cells, Auerbach’s plexus, and Meissner’s plexus.
Genetics Wolman’s disease is inherited as an autosomal recessive trait. Seven of the patients reported so far have been Jews of Iraqi or Iranian origin. The other patients have been from Japan, western Europe, and North America. Young and Patrick reported that the enzyme
In 1963, Fredrickson reported the first known case of hepatic cholesterol ester storage disease. The patient was a child with hepatomegaly and hyperlipidemia.406 In 1967, Lageron and coworkers reported the case of a French adult with the same disease.407,408 In 1968, Schiff and associates reported the cases of a brother and sister with the disease. Mild cirrhosis was evident in liver biopsy specimens, along with marked increases in fat content. Four of five younger siblings were found to have hepatomegaly. Liver biopsy specimens from three of the siblings showed vacuoles in their hepatocytes similar to those seen in the hepatocytes of the original patients.381 Partin and Schubert found deposits of cholesterol esters in the lamina propria and mucosal smooth muscle and vascular pericytes409 in jejunal tissue from two of the severely affected children in the family described by Schiff. The majority of patients come to medical attention early in childhood, but 1 patient sought treatment at the age of 23. Hepatomegaly has been present in all patients reported.380,381,406–413 In 1 patient hepatomegaly was present at birth. Hepatomegaly is progressive; eventually, hepatic fibrosis develops. Splenomegaly was found in 54% of patients. Esophageal varices secondary to portal hypertension were present in 27% of the patients. One patient had recurrent episodes of abdominal pain without known cause.408,410 Two of the patients were reported to have had delays in sexual maturation.407,409
Laboratory Findings Results of liver tests are usually normal. Jaundice has not been found, except in 2 patients who may have developed hepatitis with
1301
Section XII. Inherited and Pediatric Diseases of the Liver
Pathology
Table 69-5. Lipid Concentrations in Liver Tissue of Patients with Cholesterol Ester Storage Disease
Lipids, wet weight (mg/g) Upper limit of normal
Total lipids
Triglycerides
Total cholesterol
Free
Esterified
280 244
64 36 19
121 112 4
9 11 3
187 174 1
lethal complications.398,411 Schiff and colleagues studied the serum bile acid profiles of 2 patients and those of five siblings, both parents, an uncle, and a grandfather of these patients. Total serum bile acids were markedly elevated in 1 patient and in all but three relatives. The ratio of cholic acid to chenodeoxycholic acid was decreased in duodenal bile obtained from 1 of the 2 patients.381 Plasma lipoprotein patterns show hypercholesterolemia and some hypertriglyceridemia. The majority of the patients have increased levels of low-density lipoprotein. Two patients had low levels of high-density lipoprotein.
Biochemical Characteristics Results of lipid analyses of liver tissue obtained from 3 patients with cholesterol ester storage disease are shown in Table 69-5. The major abnormality was the marked increase in cholesterol esters. The fatty acid composition of the cholesterol esters showed a predominance of oleic and linoleic acids.381,387 The stored esters contained less than the expected amount of cholesterol linoleate. Increased levels of low-density lipoproteins in serum are routinely found in cholesterol ester storage disease. The presence of atherosclerosis in 2 patients with cholesterol ester storage disease is of interest, because current speculation about the role of acid hydrolase in arterial intima predicts that accelerated atherosclerosis will be a consequence of the enzymatic deficiency. The reason for the different clinical courses of Wolman’s disease and cholesterol ester storage disease is unclear. It may be that there is a small but critical difference between the levels of residual acid hydrolase in the two groups of patients. A slightly greater deficiency of acid hydrolase in tissues has been recorded for patients with Wolman’s disease.380 Another possibility is that the two diseases are different at the levels of the enzymatic defects. For example, liver tissue from a patient with Wolman’s disease did not hydrolyze DL-hexadecanyl-1,2-dioleate, whereas liver tissue from a patient with cholesterol ester storage disease hydrolyzed this substrate at a normal rate.414 There could be isoforms of acid hydrolase with deficiency of different isoenzymes in Wolman’s disease and cholesterol ester storage disease. Alternatively, it may be that the nature of the defect in a single type of acid hydrolase differs in these disorders. Burton and co-workers reported that the intracellular acid hydrolase activity in cultured cells from Wolman’s patients was 10–20% of control hydrolytic activity, whereas cholesterol ester storage disease cells exhibited 30–45% of control activity. These differences were only found when intracellular rather than cell lysate activity of the enzyme was measured.415
1302
The pathologic changes are secondary to the intralysosomal deposition of cholesterol esters and triglycerides.
Liver Macroscopically, the liver is greatly enlarged and appears orange. On microscopic examination, the hepatic parenchymal cells and Kupffer cells show deposition of lipid droplets (Figure 69-14). In frozen sections, the lipid droplets in the hepatocytes are birefringent but not autofluorescent. Those in Kupffer cells are not birefringent but show yellow autofluorescence. The differences are thought to be due to peroxidation of the fatty acids of cholesterol esters in macrophages but not in parenchymal cells. Electron microscopic studies show the lipid deposits to be limited by a single trilaminar membrane. The deposits in the hepatocytes are electron-lucent; however, those in the Kupffer cells are interspersed with electron-dense material. In the majority of patients, dense bands of fibrous tissue extend through the liver, forming lobules of different sizes. Fibrosis may progress in some patients to give a classic picture of cirrhosis, with the development of portal hypertension and esophageal varices.411
Other Tissues
Partin and Schubert409 studied small intestinal biopsy specimens from 2 patients. The biopsy specimens had an orange tinge. In contradistinction to Tangier disease, which features storage of cholesterol esters, abnormal coloration of the colonic mucosa or the tonsils has not been found in cholesterol ester storage disease. Histologic sections of the small bowel show that the epithelial cells are normal; however, foam cells were found in the lacteal area and were especially abundant in the villus tips. Free droplets of fat were present in the extracellular spaces of the lamina propria.380 Electron microscopic studies of small intestine show foam cells in clusters beneath the basement membrane of the epithelial cells surrounding the lacteals. The cytoplasm of the lacteal endothelium is distended by numerous large osmiophilic lipid vacuoles. Smoothmuscle fiber cells, vascular pericytes, fibrocytes, and Schwann cells of nerve fibers contain lucent lipid droplets. Deposits of lipids, similar to those seen in Wolman’s disease, were found in other tissues. Although clinically there was no evidence of atherosclerosis in patients with cholesterol ester storage disease, 2 of 3 patients examined by necropsy had atherosclerosis.411
Genetics Cholesterol ester storage disease is inherited as an autosomal recessive trait.380 There is a preponderance of females among the patients reported. It is possible to detect heterozygotes by quantitation of acid hydrolase activity in leukocytes or cultured fibroblasts. Heterozygotes have 40–50% of the reported normal enzymatic activity.411 Prenatal diagnosis can be established using cultured amniotic fluid cells. Different mutations in the lysomal acid lipase can explain the level of the residual enzyme activity. Therefore, the genotype can predict the severity of the phenotype.416
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY Figure 69-14. Biopsy specimens from a patient with cholesterol ester storage disease. (A) Liver, showing dense bands of connective tissue extending through the liver, forming nodules (Wilder’s reticulum stain, ¥200). (B) The individual foam vacuoles of different sizes plus acinar clefts interpreted as cholesterol, indicated by arrow (Gomori’s trichrome, ¥500). (C) The lamina propria of the small intestine, containing foam cells, indicated by arrow, in dense nodules distorting the involved villus (Gomori’s trichrome, ¥200). (Reproduced from Beaudet AL, Ferry GD, Nichols BL, Jr, et al. Cholesterol ester storage disease: clinical, biochemical, and pathological studies. J Pediatr 1977; 90:910–914, with permission.)
A
B
C
1303
Section XII. Inherited and Pediatric Diseases of the Liver
Treatment and Prognosis
Clinical Features
Cholesterol ester storage disease is relatively benign. Death in childhood from hepatic complications has been reported.411 Two patients died at 43 years of age.407,417 Two reports suggested the utility of inhibitors of 3-hydroxy-3methylglutaryl-coenzyme A reductase as a safe and effective treatment for children with acid lipase deficiency. A significant decrease in hepatomegaly and the levels of cholesterol and triglycerides was observed in the treated patients.418,419
Progressive intrahepatic cholestasis usually manifests between the ages of 1 and 10 months.424,426 In four of the reported cases, onset of the disease occurred in the first week of life.424 Pruritus and jaundice are the earliest symptoms. In infants, pruritus may be severe enough to interfere with sleep. The jaundice is usually accompanied by dark urine and pale stools that may become totally acholic. The stools are loose, foul-smelling, and greasy because of steatorrhea.424,427 Some patients may have watery diarrhea that continues even after liver transplantion.428,429 The level of jaundice usually fluctuates. Recurrent cholestatic episodes alternate with periods of remission. The cholestatic episodes have been reported to be triggered by upper respiratory tract infections.424 After several months, remissions are less frequent. Rickets is seen in some patients.424,430 Most have growth retardation;430 developmental retardation was found in 30% of the reported patients. Bleeding episodes secondary to hypoprothrombinemia have been reported for approximately half the patients. Some patients have clubbing of the fingers.431 Xanthomatosis of the skin does not occur. Hepatomegaly has been present in all patients reported so far, but splenomegaly has been found in only half of the patients. Liver enlargement persists during remissions, and with the progression of disease, the liver becomes hard, irregular, and nodular. Extraintestinal manifestations include pancreatitis and hearing impairment.432,433 The majority of patients die between the ages of 2 and 15 years, but occasionally a patient has survived until the age of 25 years.423 Dahms reported the cases of twin brothers with Byler’s disease, both of whom developed hepatocellular carcinomas. The two brothers died at 13 and 17 years of age of liver failure.434
INBORN ERRORS OF BILE ACID METABOLISM (See Chapter 5) Inborn errors of bile acid metabolism include disorders involving molecular defects in the genes encoding canalicular transport proteins, peroxisomal disorders, and a heterogeneous group of disorders, including Alagille’s syndrome with mutations in the Jagged-1 gene and cystic fibrosis with mutations in the cystic fibrosis transconductor regulator (CFTR).
PROGRESSIVE FAMILIAL INTRAHEPATIC CHOLESTASIS Progressive familial intrahepatic cholestasis (PFIC) is a heterogeneous group of autosomal recessive disorders in which hepatocellular cholestasis often presents in the neonatal period or the first year of life, leading to death from liver failure during the childhood to adolescent periods. The clinical, biochemical, and histological features and the advances in the understanding of the canalicular membrane transport proteins in the liver have provided evidence for the heterogeneity of this clinical entity and suggested defects in bile secretion and/or bile acid metabolism without anatomic obstruction. Three types of PFIC have been described based on the recent molecular and genetic studies, which allowed identification of the molecular basis of these three types.
Progressive Intrahepatic Cholestasis Type I (Byler’s Disease) The first type is called PFIC-1, also known as Byler’s disease and originally described in 1965 by Clayton and associates, who described a syndrome of progressive intrahepatic cholestasis in an extensive study of Amish descendants from Jacob Byler, who was born in the USA in 1799. Six members of four interrelated, inbred Amish sibships were described in Clayton’s original abstract.420 The disease was characterized by onset of pruritus, jaundice, steatorrhea, and hepatosplenomegaly early in infancy. Four of the 6 patients died between 17 months and 8 years of age. Similar cases have been reported from France and Japan.421–423 Biochemically, all Clayton’s patients had conjugated hyperbilirubinemia and elevated serum alkaline phosphatase with normal serum cholesterol levels. Subsequently, Byler’s disease patients are characterized as having low serum g-glutamyltransferase concentrations, high serum bile-salt concentrations and low biliary chenodeoxycholic bile-salt concentrations. This form of progressive familial intrahepatic cholestasis has been mapped by positional cloning to chromosome 18q2122.424,425
1304
Laboratory Findings All reported patients have had elevations of total serum bilirubin. Levels of 40–50 mg/dl are not unusual during the cholestatic episodes.427 The direct fraction is usually half of the total bilirubin.421–423 The serum AST and ALT levels are elevated. Serum cholesterol is usually normal or low.421 Serum g-glutamyl transpeptidase (GGTP) levels are normal, despite the elevation of serum alkaline phosphatase levels.435 Levels of GGTP in the liver are reported to be elevated.436 Serum GGTP could be used as a marker for Byler’s disease, because normal values were found in 22 of 28 patients with the disease.437 Prothrombin time and thromboplastin time are increased secondary to malabsorption of vitamin K. Malabsorption studies reveal severe steatorrhea and excretion of fat amounting to 50–80% of intake.422 Both Tm and storage capacity for sulfobromophthalein (BSP) are markedly reduced. A parent heterozygotic for the trait is clinically normal but may have an abnormally decreased BSP Tm.421 Total serum bile acids have been elevated in the patients studied. Total bile acids in duodenal aspirates of 1 patient studied by Linarelli and associates were 0.07–0.5 mmol/l, which is below the critical micellar concentration of 2 mmol/l.438,439 The clearance of labeled cholic acid and chenodeoxycholic acid was normal at 20 minutes and slightly impaired at 60 minutes. The half-life of the labeled bile acid was prolonged. The major loss of the isotopes was in urine rather than feces. Serum lithocholic acid was reported to be high in Linarelli’s patient. Radiographic examination of the biliary
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
tree by operative cholangiography showed patency of the biliary tree in the patients studied. In the series reported from France, cholelithiasis and intrapancreatic calcification were found in some patients.423 The biochemical findings suggest a defect in the excretion of bile acids across the canalicular membrane of the liver cell. The gene for PFIC-1 is called FIC1, which encodes for a P-type ATPase that is predominantly expressed in the small intestine and the liver and likely to be important in the transfer of amino phospholipids from the outer to the inner leaflet of the plasma membrane bilayer.425 Mutations in this transporter, ATP8B1, are responsible for PFIC-1.440–442
Pathology Light microscopy of liver biopsy specimens early in the disease shows normal hepatic architecture. Hepatocellular and canalicular cholestasis with pseudoacinar transformation is commonly seen. Bile duct damage leading to ductal paucity is seen in 70% of older patients. The striking feature of all liver biopsy specimens is the marked cholestasis in the parenchymal cells and in canaliculi. As the disease progresses, the classic picture of biliary cirrhosis develops.443 Electron microscopic studies show markedly abnormal biliary duct canaliculi. The pericanalicular ectoplasm is greatly thickened, and the microvilli are swollen, fused, or blunted. The canalicular lumina are filled with coarsely particulate and amorphous granular material.443
Prognosis Death usually occurs between the ages of 17 months and 8 years from hepatic complications such as liver failure, bleeding, and malnutrition. An occasional patient has survived to 25 years of age.
Therapy Oral ursodeoxycholic acid appears effecive in all types of progressive familial intrahepatic cholestasis for improving the clinical status of some children.444 Alagille and Odievre reported rapid relief of cholestasis in 4 patients in whom cholecystojejunostomy was performed; however, the long-term benefit from such a major surgical procedure is unknown.423 Similarly, Whitington and Whitington have shown improvement in cholestasis with external biliary drainage.445 The researchers’ experience suggests heterogeneity in the response to biliary diversion. Some patients improved markedly; others showed no improvement. The latter patients had liver transplantation, and favorable experience with liver transplantation in patients with Byler’s disease has been reported.446
Progressive Intrahepatic Cholestasis Type II Patients with initial typical findings of familial progressive intrahepatic cholestasis type I, and unrelated to the Byler family, have been designated as Byler syndrome. Cases have been described in isolated populations, the Middle East, Greenland, and Sweden.447–450 Homozygosity mapping and linkage analysis in 6 consanguineous patients of Middle Eastern origin has resulted in identification of a gene location on chromosome on q24.450 Patients with type II familial progressive intrahepatic cholestasis present with severe pruritus, normal serum g-glutamyl transferase activity and cholesterol levels,
high concentration of serum primary bile acids and low biliary primary bile acid concentrations. Clinically these patients present with more severe and permanent jaundice with the onset of rapid liver failure. Liver histology shows absence of ductular proliferation with canalicular cholestasis and periportal biliary metaplasia of hepatocytes.434,443 However, the liver architecture is more severely altered with lobular and portal fibrosis and inflammation and giantcell proliferation compared to PFIC-1. More recent studies suggest that the defect may reside in the canalicular bile-salt export pump (BSEP). At least 10 BSEP mutations in PFIC-2 patients from several different populations have been determined.451 Moreover, the human BSEP gene has been mapped to 2q24 locus, which is the gene locus for the PFIC-2. The findings of mutations in BSEP are consistent with the decreased canalicular excretion of bile acids described in these patients. Indeed, patients with PFIC-2 were found to have a close correlation between BSEP gene mutations and canalicular BSEP expression, resulting in a decrease in the concentration of biliary bile acids in these deficient patients.452 The BSEP transporter is now known as ABCB11.453
Progressive Familial Intrahepatic Cholestasis Type III Patients with PFIC-3 usually present later in life with significant potential for biliary cirrhosis and liver failure at a later age. These patients can be distinguished from the other types of progressive intrahepatic cholestasis by the finding of very high serum g-glutamyl transferase activity and liver histology that show portal fibrosis with bile duct proliferation and inflammatory infiltrate in the early stages, despite the fact that their intrahepatic and extrahepatic bile ducts are patent.430 The genetic basis of PFIC-3 has been shown to be related to mutations in the human multidrug-resistant 3 (MDR-3) P-glycoprotein, which transports phospholipids into the biliary system.454,455 The finding of mutation on MDR-3 was based on the finding that analysis of bile in these patients shows very low concentration of phospholipid and the phenotype of the analogous MDR-2 knockout mouse.456 MDR-3 belongs to the family of ABC transporter and is expressed in the canalicular membrane of the hepatocytes. Immunohistochemistry revealed the lack of canalicular staining for MDR-3 in the liver tissue of patients with PFIC-3. Mutations in the MDR-3 gene have been described in patients with PFIC-3. Some of these mutations lead to a truncated MDR-3 protein, which lacks at least one ABC motif. Additional nonsense mutations and missense mutations associated with low biliary phospholipid levels have been identified in patients with PFIC-3. The liver pathology may be related to the toxic effects of bile acids on bile canaliculi and the biliary epithelium in the absence of phospholipids. Recent studies have also suggested that heterozygous state for a MDR-3 gene defect may represent a genetic predisposition in families with cholestasis, noted during pregnancy.457,458 The MDR-3 gene has been mapped to 7q21-36. One study has identified 16 different mutations in 17 different patients. These mutations include frameshifts, nonsense, and missense. Gallstones or episodes of cholestasis of pregnancy were found in patients or parents. Children with missense mutations had a less severe disease and more often a beneficial effect of ursodeoxycholic acid therapy was noted in this group of patients.456
1305
Section XII. Inherited and Pediatric Diseases of the Liver
HEREDITARY LYMPHEDEMA WITH RECURRENT CHOLESTASIS (Norwegian Cholestasis; Aagenaes’s Syndrome) In 1968, Aagenaes and associates reported the cases of 16 patients from Norway in whom hereditary recurrent intrahepatic cholestasis appeared during the neonatal period.459 In 1971, Sharp and Krivit reported two sisters of Norwegian extraction with a similar syndrome.460 Consanguinity was present in six of the seven parental couples in the series reported by Aagenaes and co-workers, suggesting an autosomal recessive inheritance. In 1974, Aagenaes described two additional families with a similar syndrome.461 All patients had early onset of intrahepatic cholestasis with elevated levels of total and direct serum bilirubin. Lymphedema of the lower extremities appeared late in childhood in all of Aagenaes’s patients. Lymphedema appeared early and at the onset of cholestatic jaundice in the patients of Sharp and Krivit.
Clinical Features All reported patients became jaundiced during the first month of life.460–462 The jaundice lasted about 1–6 years, and during this period, the patients had severe pruritus. Growth retardation with complications of malabsorption such as rickets, anemia, and bleeding tendency was evident during the period of cholestasis. When the cholestasis resolved, catch-up growth occurred, so the patients’ adult heights were normal. One or more further episodes of cholestasis have occurred in all adult patients. Lymphedema, once present, persists throughout life.
Laboratory Findings Total serum bilirubin is elevated, with the direct fraction accounting for 50–80% of the total. Especially during periods of cholestasis and in the first year of life, serum AST and ALT levels are increased.460–463 Serum alkaline phosphatase levels are always increased. Results of the BSP excretion test are abnormal, with retention of 25–40% at 45 minutes. Serum cholesterol and triglyceride levels are elevated. Lipoprotein electrophoresis shows increases in pre-beta and beta lipoproteins. Protein electrophoresis shows elevation of the a2-globulin fraction with low serum albumin. Prothrombin time and thromboplastin time are elevated secondary to malabsorption of vitamin K. Fecal fat measurements show excretion of about 30–50% of ingested fat. Fecal nitrogen excretion is normal. Lymphangiography with visualization of the deep lymphatics was attempted in only one case. The lymphatics were found to be abnormally tortuous. Injection of blue dye into the interdigital spaces in another patient resulted in a chicken-wire pattern, indicative of an absence of deep lymphatics.461
Pathology Examination of liver biopsy specimens reveals intact architecture with marked bile stasis, giant-cell transformation, and minimal hepatic cell necrosis. The portal areas show a slight increase in fibrosis, and in most instances, biliary ducts are difficult to find. In 1 patient, repeat liver biopsies at 10 years of age showed progression to cirrhosis.
1306
Biochemical Characteristics The elevated total serum bile acids and retention of BSP suggest a defect in the excretory function of the liver. However, the nature of the defect is not known. The association of lymphedema of both lower extremities and cholestasis raises the possibility of abnormal lymphatic drainage from the liver. Aagenaes and colleagues injected colloidal gold into the liver capsule of one of their patients. Instead of normal excretion by lymphatic vessels, there was puddling of the isotope in the liver, which suggests a defect in the hepatic lymphatic system.462 The close functional relationship between lymphatic and biliary drainage systems of the liver was shown experimentally in animals. After experimental obstruction of the common bile duct, bilirubin and bile acids rapidly appeared in the lymphatic system before their elevation in the serum.463–466 Conversely, an increased biliary flow with increased excretion of bilirubin and bile acids occurred after interruption of the hepatic lymphatics in cats.462 Thus, it is possible that abnormalities of the lymphatics of the liver contribute to the pathogenesis of this entity. More recently, the locus for Aagenaes syndrome has been mapped to a 6.6-cM interval on chromosome 15q.467
Prognosis Initially, the syndrome was thought to carry a favorable prognosis;460 however, a subsequent report indicated that one of the earlier reported patients had died of liver failure. As previously mentioned, in another patient, a repeat liver biopsy at 10 years of age showed evidence of cirrhosis.461
Therapy No effective therapy is available. Cholestyramine has been used to alleviate pruritus during the cholestatic episodes. Because of the cholestasis, fat-soluble vitamins and fats in the form of mediumchain triglycerides should be given.
Alagille’s Syndrome (Arteriohepatic Dysplasia) Although reports of biliary duct hypoplasia and chronic liver disease appeared in the early 1950s,468–470 it was not until 1973 that Watson and Miller described the cases of 9 patients who had neonatal liver disease and familial pulmonary dysplasia.471 A full description of the syndrome was provided by Alagille and co-workers. These investigators emphasized the characteristic facial appearance and vertebral and cardiovascular anomalies.472 The syndrome is not rare, as investigators were able to report large groups of patients.472–476 The recent identification of Jagged-1 gene mutations in patients with Alagille’s syndrome made it possible to understand the mechanism of the disease and explain the various clinical manifestations.477,478
Clinical Features The major features of this syndrome are chronic liver disease, characteristic facies, cardiovascular and vertebral anomalies, and ophthalmologic abnormalities. Minor features include central nervous system, renal, endocrine, pancreas, gut, systemic vascular system, ear, lung and larynx abnormalities. Table 69-6 summarizes the major and minor abnormalities. Chronic Liver Disease. The syndrome is associated with cholestasis, which may develop during the neonatal period and usually
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Table 69-6. Clinical Features of Alagille’s Syndrome Major features Liver Heart Eye Facies Vertebra Minor features Congenital heart disease
Systemic vascular malformations Skeletal anomalies
Eye abnormalities Renal abnormalities
Pancreas abnormalities Lung abnormalities Ear abnormalities Gut abnormalities
Paucity of intrahepatic ducts Peripheral pulmonary stenosis Posterior embryotoxon Prominent forehead, deep-set eyes, mild hyperterolism, small pointed chin Butterfly vertebrae Coarctation of the aorta, tetralogy of Fallot, ventricular and atrial septal defects, patent ductus arteriosus, truncus arteriosus, abnormal venous return, right ventricular hypoplasia Arterial hypoplasia, renal artery stenosis, carotid artery aneurysm, intracranial hemorrhage Spina bifida, short distal phalanges and metacarpal bones, clinodactyly, short distal ulna and radius Retinal pigmentation, iris strands, cataract myopia, strabismus, glaucoma Cystic disease, renal agenesis, horseshoe kidney, renal hypoplasia, mesangiolipidosis, tubular dysfunction Exocrine pancreatic dysfunction, diabetes mellitus Tracheal and bronchial stenosis Deafness, chronic otitis media Small-bowel atresia
becomes apparent in the first 2 years of life.479–481 Pruritus develops early in infancy and persists despite the disappearance of jaundice. The liver is usually enlarged, firm, smooth, and not tender. The spleen may be enlarged even in the absence of portal hypertension. The stools may be clay-colored and the urine dark-yellow. Xanthomas may be seen on the extensor surfaces of the fingers and in creases of the palms, anal folds, and popliteal and vaginal areas. These xanthomas are due to long-standing severe intrahepatic cholestasis. With the progression of the disease about the beginning of the second year of life, jaundice may subside or disappear, but cholestasis persists. The long-term outlook for the liver disease is variable, with 10–50% of the patients developing cirrhosis and portal hypertension. In one study, only 3 of 36 patients monitored by the Alagille group developed hepatic cirrhosis.423 Characteristic Faces. The face is small, with a prominent, broad forehead. The eyes are set widely and deeply (hypertelorism). The mandible is small and pointed, giving the appearance of a triangular face.479–481 The peculiar faces become more prominent with increasing age. Cardiovascular Abnormalities. All patients have cardiac murmurs as infants. Most have peripheral pulmonary stenosis, which is usually mild and does not necessitate surgery.480 Other cardiac lesions, including ventricular septal defects, coarctation of the aorta, and cyanotic heart disease, have also been reported (Table 69-6). Osseous Abnormalities. The hands show various degrees of shortening of the distal phalanges, stiffening, swelling, and limitation of
motion at the proximal interphalangeal joints. Radiographs of the spine show either frank or incomplete butterfly vertebrae and decreased interpediculate distances in the lumbar spine.473 Ocular Findings. Posterior embryotoxon (prominent Schwalbe’s ring), visible to the unaided eye by slit-lamp examination, or on gonioscopy, is the most important ocular abnormality and occurs in 56–95% of patients. In addition, retinal pigmentary changes are usually found. Nischal and co-workers reported ultrasound evidence of optic disk drusen in a large percentage of Alagille’s patients.482 Other abnormalities include microcornea keratoconus exotropia, iris hypoplasia, and abnormalities of the optic disk.483–486 Central Nervous System Findings. Gross motor delays and mental retardation were reported in a small number of patients.475 Intracranial bleeding is the most significant neurological complication. It occurs in approximately 15% of patients474,483 and in 30–50% the bleeding is fatal. It is of note that many of the patients with bleeding did not have coagulopathy, raising the issue of cerebral vascular malformation.474 A recent study supported these observations by documenting the spectrum of vascular anomalies, including intracranial aneurysms and internal carotid artery anomalies in these patients.487 Moyamoya disease (progressive arterial occlusion of the distal intracranial carotids) has been reported in patients with Alagille’s syndrome.488,489 In support of the involvement of the central nervous system is the observation that the autosomal dominant cerebral arteriopathy with subcortical infarct and leukoencephalopathy is also caused by a mutation in Notch 3.490 Pancreatic Abnormalities. Chong and his associates described pancreatic insufficiency in some patients with Alagille’s syndrome.491 Other investigators have shown that 41% of their patients had pancreatic insufficiency; some of this group also developed diabetes mellitus.492 Other Abnormalities. Growth retardation, which decreased with age, was documented in the majority of cases of children. Slight to moderate mental retardation has also been documented in some of these cases. Renal function is impaired, with hyperuricemia and decreased creatinine clearance in some patients. Hypogonadism was suspected to be present in 6 male patients reported by Alagille and colleagues. Testicular biopsy showed no abnormality in 2; in 2 others, fibrous tissue proliferation was present. In the other 2, spermatogenic cells were almost completely lacking.472 Bleeding tendency has been reported to occur spontaneously or following surgical procedures.493
Laboratory Findings Total serum bilirubin is elevated during infancy, with levels between 4 and 14 mg/dl. The direct fraction is 30–50% of the total. Serum bilirubin usually returns to normal after the second year of life.472 Some patients remain deeply jaundiced, with bilirubin levels greater than 20 mg/dl. Serum cholesterol levels may be as high as 2000 mg/dl. Serum triglyceride levels range from 500 to 1000 mg/dl. Serum alkaline phosphatase and glutamyl transferase levels are very high. Serum AST and ALT are slightly elevated. Despite the return of serum bilirubin to normal, biochemical evidence of severe cholestasis is usually found. Total serum bile acids have been markedly elevated in the patients so studied, with
1307
Section XII. Inherited and Pediatric Diseases of the Liver
increases in both cholic acid and chenodeoxycholic acid. No major unidentified chromatographic peak was present in the serum or duodenal bile of the patients who were investigated.479 Retention of BSP at 45 minutes is abnormal; however, storage capacity for BSP is normal. Prothrombin time and partial thromboplastin time are abnormal secondary to malabsorption of fats. Both values return to normal in response to intramuscular administration of vitamin K.
Pathology Exploratory laparotomy of a patient who has arteriohepatic dysplasia shows uniformly patent extrahepatic biliary ducts. In the neonatal period, liver biopsy shows intact hepatic architecture, with marked cholestasis. The hepatocytes may be swollen, with balloon degeneration and minimal giant-cell transformation. The most dramatic changes are in the portal areas, where a decrease in the number of biliary ducts is observed, with a slight increase in connective tissue. However, Ghishan and colleagues reported finding biliary duct proliferation in 2 patients with arteriohepatic dysplasia whose livers were sampled when they were 5 and 49 days old, respectively. Subsequent liver biopsies, when the patients were 2 and 27 months old, respectively,494 showed a paucity of biliary ducts in the portal areas. Biopsies of the livers of adult patients with arteriohepatic dysplasia have shown no biliary ducts in the portal areas with variable progression to biliary cirrhosis. Indeed, 3 of 26 patients followed by the Alagille group developed hepatic cirrhosis. Other investigators reported the finding of cirrhosis in as many as 50% of patients.474,475
Genetics Alagille’s syndrome is inherited as an autosomal dominant trait with reduced penetrance and variable expression when both parents are clinically normal; the percentage of sporadic cases is 45–50%. Byrne and colleagues first noted deletions of the short arm of chromosome 20.495 Since then, several investigators mapped the gene for Alagille’s syndrome to chromosome 20p12.496–500 The gene was eventually identified as the Jagged-1 gene (JAG1) by physical, genetic, and gene mapping covering the 20p12 region.501,502 Mutations in JAG1 gene have been demonstrated in 60–75% of patients with Alagille’s syndrome. Analysis of 233 cases revealed that 72% of the reported mutations lead to frameshifts that cause a premature termination codon.503 The spectrum of mutations identified is consistent with haploinsufficiency for JAG1 being a mechanism for Alagille’s syndrome. The JAG1 gene encodes a protein, which belongs to the family of Notch ligands. The Notch signaling pathway is important for control of cell fate during embryogenesis. JAG1 is expressed ubiquitously in tissues of humans, including liver biliary epithelia, heart, kidney, eye, and brain.504 The role of JAG1 in remodeling embryonic vasculature has been delineated by a null mouse model. Mice homozygous for JAG1 mutations die from bleeding during early embryogenesis. Heterozygotic mice show eye malformations.
Biochemical Characteristics The majority of the patients with Alagille’s syndrome have elevated conjugated hyperbilirubinemia. Serum bile acids and g-glutamyl
1308
transferase levels are elevated as well. Liver aminotransferase levels are modestly elevated.
Therapy Phenobarbital and cholestyramine have not been shown to be effective in the treatment of Alagille’s syndrome; however, Alagille and colleagues reported alleviation of pruritus and reductions of serum cholesterol, triglyceride, and bilirubin levels in response to very high doses of cholestyramine – 12–15 g/day.423 Antihistamines, rifampin, and naltrexone have been used to relieve pruritus.505 Some patients were subjected to biliary drainage.506 Improved nutrition via gastrostomy with supplements of fat-soluble vitamins has been utilized. Pancreatic enzyme replacement for patients with pancreatic insufficiency is indicated.
Prognosis Data from early studies suggest that the long-term prognosis is good. Three of Alagille’s 26 patients developed cirrhosis,423 but recent reports suggest that 10–50% of patients develop liver cirrhosis and portal hypertension. Hepatocellular carcinoma has been reported. Liver transplant is eventually necessary in 21–31% of patients.475,476 Hoffenberg and colleagues have estimated that 50% of patients will eventually require liver transplants.474
INBORN ERRORS OF PEROXISOME BIOGENESIS ZELLWEGER’S SYNDROME (CEREBROHEPATORENAL SYNDROME) In 1964, Bowen and associates described an autosomal recessive disease occurring in two siblings. It was characterized by severe hypotonia, growth and mental retardation, renal cortical cysts, and hepatic dysfunction. Both patients died before the age of 5 months.507 Patients with a similar clinical syndrome were described by Smith and co-workers508 and by Passarge and McAdams.509 In 1969, Opitz and colleagues described 4 new patients and identified iron deposition in the liver. Because the original 2 patients were Professor Hans Zellweger’s patients, his name was used by Opitz’s group as an eponymic designation for this condition.510 In 1973, Goldfischer and co-workers described mitochondrial abnormalities and lack of peroxisomes in electron microscopic studies of liver biopsy specimens from patients with Zellweger’s syndrome.511 In 1975, Danks and associates reported finding pipecolic acid in the urine of 4 patients.512 In 1979, Hanson and colleagues described a defect in bile acid synthesis found in 3 patients with Zellweger’s syndrome.513 Studies conducted in the 1980s confirmed that Zellweger’s syndrome belongs to a group of disorders of peroxisome biogenesis.514 The group of peroxisomal disorders now includes 17 different diseases, such as neonatal adrenoleukodystrophy, infantile Refsum’s disease, and hyperpipecolic acidemia. Affected patients have reduced or absent peroxisomes with multiple enzyme defects that are normally present within peroxisomes. The peroxisomal functions include beta-oxidation of long-chain fatty acids, ether-phospholipid biosynthesis, glyoxylate metabolism, degradation of pipecolic acid, and oxidation of phytanic acid.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Clinical Features Severe hypotonia with simian creases is common, together with camptodactyly, which may involve several fingers. There may be some ulnar deviation of hands and fingers. Partial flexion of the knees with various degrees of equinovarus deformities of the feet is often seen. All patients have high foreheads with shallow supraorbital ridges and flat facies. External ear deformities and redundant skin folds in the neck are commonly seen. Sucking and swallowing difficulties, generalized seizures, and delays in psychomotor developmental maturation are common. Patent ductus arteriosus and septal defects are present in approximately 40% of patients. Hepatomegaly is common; splenomegaly is occasionally found. Jaundice is evident in 30–50% of patients.512
Laboratory Findings Hematologic investigations have shown elevated serum iron and nearly total saturation of the serum iron-binding protein in more than half the cases reported.515 The true incidences of these abnormalities may be higher, because iron status has not been determined in many cases. Tissue concentrations of iron in the liver, spleen, kidneys, and lungs are increased. In one reported case, tissue iron in the liver was 50 mg/100 g wet weight (normal values for the same age group range from 7 to 21 mg/100 g). Ferrokinetic studies in this case demonstrated a rapid disappearance of iron and a markedly increased plasma iron turnover.515 Iron incorporation into circulating red cells was significantly impaired, with abnormal accumulation of the radiolabeled iron in the liver. Intestinal absorption of iron was normal. Desferrioxamine induced a marked increase in urinary iron excretion. However, ferrokinetic studies of a second patient done by the same investigator showed no abnormality.515 Prothrombin time and thromboplastin time are usually prolonged, secondary to hepatic dysfunction. Serum AST and ALT levels are moderately elevated. Conjugated hyperbilirubinemia may be detected. Aminoaciduria was present in four cases. Urinary excretion of pipecolic acid, a minor breakdown product of lysine, was found to be markedly increased in four infants in whom adequate qualitative tests were made.512 Hanson and colleagues showed that three infants excreted excessive 3a,7a-dihydroxy5b-cholestan-26-oic acid (DHCA), 3a,7a,12a-trihydroxy-5bcholestan-26-oic acid (THCA), and 3a,7a,12a,24gj-tetrahydroxy5b-cholestan-26-oic acid (varanic acid). These compounds are precursors of chenodeoxycholic acid and cholic acid that have undergone only partial side-chain oxidation.513 These findings were confirmed in two other cases of Zellweger’s syndrome.516 Increased accumulation of very-long-chain fatty acids in tissues, blood cells, and plasma is present in these patients secondary to defective betaoxidation of long-chain fatty acids.517 Accumulation of phytanic acid also occurs secondary to defective oxidation.518 Tissue levels of plasmalogens are reduced secondary to defective synthesis.519 These abnormalities reflect defective peroxisomal function.520 Deficient docosahexaenoic acid (DHA) in erythrocytes and plasma has been documented.521 Radiographic studies of patients with Zellweger’s syndrome reveal certain characteristic findings, mainly the presence of patches of calcification in cartilage. These have a dense cortex and a reticular central region and characteristically are present in the triradiate
cartilage in the acetabulum, the patellae, the sternum, and the scapulae.522
Pathology Liver Macroscopically, the liver appears unremarkable. Histologic sections of biopsy and autopsy specimens of the liver show changes that may vary from minimal to severe diffuse fibrosis with cirrhosis.512 The original patients were described as having biliary dysgenesis; however, recent reports show normal biliary ducts in the portal areas, with various degrees of portal fibrosis. The hepatic lobules may contain foci of liver cell necrosis and loss. Prussian blue staining for iron shows a marked increase in iron deposition, mainly in the reticuloendothelial cells.510 Some patients have minimal or no deposition of iron.512,523 There is no correlation between the extent of liver damage and iron deposition.512 Electron microscopic studies show swollen hepatocytes filled with glycogen. The mitochondria are often extremely dense and reduced in number; their cristae are twisted and irregular, with dilation of the intracristate spaces. Typical arrays of rough endoplasmic reticulum are sparse. No peroxisomes can be found in the hepatocytes.511
Kidneys Macroscopically, the surface is studded with small (<3 mmol/l in diameter) fluid-filled cysts. Microscopically, the cysts contain dysplastic glomerular and tubular elements and are lined by cuboidal or flattened epithelium. Cysts of glomerular origin and others that appear to be dilated tubular structures are also present. Ultrastructural studies demonstrate a lack of peroxisomes in the proximal tubules.510,511
Central Nervous System Severe cerebral gliosis, subependymal cyst formation, macrogyria, and polymicrogyria were noted. Myelinization was incomplete or lacking.510 Prussian blue staining shows deposition of iron. Ultrastructurally, the mitochondria of the cortical astrocytes are abnormally dense and often appear degenerate.511
Biochemical Features The mitochondria in the livers, kidneys, and brains of patients with Zellweger’s syndrome are structurally abnormal. Functionally, oxygen consumption of mitochondrial fractions prepared from the livers and brains of 2 patients was diminished by 70%. Addition of ADP to the mitochondrial fractions failed to stimulate respiration, although this response was elicited in the control mitochondria.511 In-vitro studies suggest that the formation of C24 bile acids (chenodeoxycholic acid and cholic acid) from cholesterol is defective in affected patients. It seems that mitochondrial defects are the cause of the abnormalities in bile acid synthesis (Chapter 5). Because cholic acid and chenodeoxycholic acid are present in these patients, the defect in synthesis is not a complete one. It is possible that, in Zellweger’s syndrome, chenodeoxycholic and cholic acids are synthesized via an alternate pathway requiring only microsomal enzymes.524,525
1309
Section XII. Inherited and Pediatric Diseases of the Liver
Molecular Basis of Zellweger’s Syndrome The primary defect in Zellweger’s syndrome is impaired assembly of peroxisomes.526 A rat cDNA encoding the peroxisome assembly factor-1 (PAF-1) was cloned.527 This cDNA encodes for a 35-kDa protein that restores the assembly of peroxisomes in peroxisomedeficient Chinese hamster ovary cell.528 A human cDNA has been cloned that complements the disease symptoms, including defective peroxisome assembly in fibroblasts from a patient with Zellweger’s syndrome.527 A point mutation in the cDNA of a patient with Zellweger’s syndrome resulted in premature termination of PAF-1.529 These observations were extended to show that at least 23 proteins are required for proper peroxisome assembly.530–532 Complementation analysis indicated that 12 different complementation groups are defective in patients with peroxisomal disorder. The genes involved in protein assembly of peroxisomes (peroxins) are termed PEX genes. Mutations in 10 different PEX genes have been described.532–542 Zellweger’s syndrome is caused by mutations in any of several different genes involved in peroxisome biogenesis, including PEX1, PEX2, PEX5, PEX6, PEX10, PEX12, PEX13, PEX16, and PEX19.530 The gene affected in complementation group 1 is PEX1. Approximately 65% of patients with peroxisomal biogenesis disorders harbor mutations in PEX1. A complete lack of PEX1 protein was found to be associated with Zellweger’s syndrome; however, residual amounts of PEX1 protein were found in milder phenotype (neonatal adrenoleukodystrophy and infantile Refsum’s disease). The most common mutation described is G843D. This missense mutation results in a misfolded protein.543
oxisomes seen in Zellweger’s patients. Determination of mutations in PEX genes will eventually replace the biochemical analysis of urine in these patients.
THE THCA SYNDROME In 1972, Eyssen and co-workers described two children with neonatal cholestatic liver disease and markedly elevated levels of THCA in their duodenal fluid. In normal individuals, THCA is not present in either serum or bile. The patients had other congenital abnormalities, such as frontal bossing, epicanthal folds, and simian creases.545 In 1975, Hanson and associates described two siblings with a rapidly progressive form of intrahepatic cholestasis that was associated with very high levels of THCA in bile and serum.546 Hanson’s patients differed from those described by Eyssen and colleagues in that no other congenital abnormality was present in the former group. However, all 4 patients had a paucity of bile ducts, and all died before the age of 2 years.
Clinical Features The 2 patients reported by Eyssen’s group had cholestatic jaundice at 2–3 months of age. Jaundice remitted in both between 4 and 5 months of age. The first patient died at the age of 8 months of severe malnutrition.545 The 2 patients reported by Hanson’s group had jaundice. Hepatosplenomegaly was present at birth in 1 and at 4 months of age in the other. Pruritus was not observed. Both patients failed to gain weight and died with progressive liver disease before the age of 2 years.546
Laboratory Findings
Prognosis Patients who have Zellweger’s syndrome die in the first year of life of malnutrition, liver failure, and intercurrent infections.510,512 Desferrioxamine enhances urinary excretion of iron but unfortunately does not influence the ultimate outcome.
Genetics An autosomal recessive inheritance is suggested by family studies. A prenatal diagnosis based on detection of elevated levels of a verylong-chain fatty acid, hexacoseanoic acid (C26:0), in cultured amniotic fluid cells has been described.544 The impaired oxidation of very-long-chain fatty acids is secondary to the absence of the per-
Both patients reported by Hanson and co-workers had elevated total and direct serum bilirubin levels. Serum alkaline phosphatase was markedly elevated; however, cholesterol and triglycerides were normal in 1 patient. Both patients had radiographic evidence of rickets. Results of bile acid analyses of the duodenal fluids of the patients of Eyssen and associates are shown in Table 69-7. In the 2 cases reported by Hanson, THCA amounted to 72% and 65% of total bile acids in the sera. Cholic acid was not detectable in the serum of the first patient and constituted only 6% of serum bile acids in the second. Chenodeoxycholic acid amounted to 28% and 22% of the total bile acids, respectively. Normally, THCA is not detectable in human bile or serum.
Table 69-7. Results of Duodenal Bile Acid Analysis in THCA Syndromea Relative concentration (%)
Patient 1 Patient 2 Control 1 Control 2
Age (months)
Total bile salt (mg/100 ml)
Chenodeoxycholic acid
Cholic acid
THCA
4.5 3.5 3 4
26 61 251 424
23 18 51 39
58 37 49 61
19 45 0 0
THCA, 3a,7a,12a-trihydroxy-5b-cholestan-26-oic acid. a Patients of Eyssen H, Parmentier G, Compernolle F, et al. Trihydroxycoprostanic acid in the duodenal fluid of two children with intrahepatic bile duct anomalies. Biochim Biophys Acta 1972; 273:212–221.
1310
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
Biochemical Features All 4 patients had very high levels of THCA in their sera. The metabolism of THCA was studied in 1 patient after an intravenous injection of [3H]THCA, and the cause of the increase in THCA was found to be a metabolic defect in the conversion of THCA to cholic acid.546 Because varanic acid, a metabolite of THCA, could not be identified in the serum of either of the 2 patients reported by Hanson’s group, it is reasonable to assume that the defect is due to a deficiency of the 24-hydroxylating enzyme system required to convert THCA to varanic acid.546 The toxicity of THCA has not been investigated thoroughly, although Lee and Whitehouse demonstrated that THCA is a potent uncoupler of oxidative phosphorylation.547 Thus, it is possible that the destruction of the intrahepatic biliary ducts might be due to an accumulation of THCA.
Pathology Liver biopsy specimens show a paucity of ductular structures with an increase in connective tissue in the portal areas. Subsequent liver biopsy specimens show rapid progression to cirrhosis.
Prognosis All 4 patients whose cases were reported died before the age of 2 years with hepatic cirrhosis.
Genetics The disease appears to be inherited as an autosomal recessive trait.
Therapy Cholestyramine therapy was not effective in preventing the development of hepatic cirrhosis.546
CYSTIC FIBROSIS Cystic fibrosis is the commonest cause of chronic obstructive pulmonary disease and of pancreatic insufficiency in the first three decades of life in the USA. Today, as more patients with cystic fibrosis reach reproductive age, the hepatic complications of cystic fibrosis are encountered with increasing frequency. In her first review of cystic fibrosis of the pancreas, Anderson found 3 patients with hepatic cirrhosis already reported and added 1 of her own.548 Farber provided an excellent description of an unfamiliar type of cirrhosis with gross lobulation of the liver, which he found in a few of his 87 patients with cystic fibrosis. Early histologic changes in the liver included enlargement of the portal areas with biliary duct proliferation and accumulation of eosinophilic material within the lumens of the biliary ducts. The cirrhotic change was attributed to obstruction of the biliary ductules by inspissated secretions.549 Bodian described focal cirrhotic lesions found in one-fourth of 62 patients with cystic fibrosis examined at necropsy. These findings were present in virtually all patients older than 1 year. Bodian proposed the descriptive term focal biliary cirrhosis. This lesion had been asymptomatic in all of Bodian’s patients.550 Craig and associates found 7 patients with cirrhosis of the liver among 150 patients with cystic fibrosis examined at necropsy. Microscopically, the investigators found obstruction of biliary ductules by eosinophilic amorphous
material surrounded by areas marked by fibrosis and biliary duct proliferation. The focal character of the extensive hepatic damage was emphasized.551 Blanc and di Sant’Agnese described cystic fibrosis and multilobular hepatic cirrhosis in 7 patients 4–10 years old. To establish the evolution of the hepatic lesions, Blanc and di Sant’Agnese reviewed the autopsy material of 116 patients with cystic fibrosis. Of 25 patients who had cirrhotic lesions, 16 had single or multiple lesions of focal biliary cirrhosis. The changes were more extensive in 9. A multilobular biliary cirrhosis was found in 6 of these. Younger patients had focal biliary cirrhosis; older patients had multilobular cirrhosis.552
CLINICAL FEATURES Liver Disease in Infancy Cystic fibrosis may manifest in the newborn period with obstructive jaundice.553–559 In all of 10 such infants, the onset of jaundice was before 3 weeks of age, and it occurred before 10 days of age in 8.549 Jaundice persisted from 20 days to 6 months. Meconium ileus with or without small intestinal atresia or combined with volvulus was found in approximately half of these patients. In 3 patients subjected to laparotomy, accumulation of thick, viscid bile was thought to have caused extrahepatic obstruction.557–560 Bile duct proliferation without plugging by inspissated material was detected in the livers of 5 of 9 patients with meconium ileus and intestinal atresia whose livers were examined. However, the authors do not rule out that the obstruction might have occurred in more central ducts.554 Neonatal hepatitis with marked giant-cell transformation of the hepatocytes was documented in a 2-month-old patient with cystic fibrosis and obstructive jaundice.560 Some infants may make a complete recovery, and some may die of liver failure and other complications of cystic fibrosis. The diagnosis is suspected on the basis of a history of meconium ileus.
LIVER DISEASE IN CHILDHOOD AND ADOLESCENCE Symptomatic liver disease was noted in 2.2%561,562 to 16%559 of patients with cystic fibrosis in two series. In the largest reported series, from Boston, 48 of 2500 patients with cystic fibrosis developed portal hypertension. The incidence of portal hypertension increases more than 10-fold for patients who are adolescent or older.563,564 Interestingly, several members of the same family developed cirrhosis and its complications. A similar observation was reported by Stern and colleagues. Initial symptoms of hepatic complications begin between 9 and 19 years of age. Of 693 patients with cystic fibrosis observed during a period of 18 years, 15 developed clinical hepatic disease, an incidence of 2.2%. In 13 of these 15 patients, all symptoms were related to portal hypertension, such as gastrointestinal bleeding or hypersplenism. Hepatocellular dysfunction was the principal feature in one of the remaining two cases, and massive hepatomegaly and failure to thrive were dominant in the other.561 A similar experience was reported from Switzerland. Of 204 unselected patients with cystic fibrosis, 7 were found to have hepatic cirrhosis, an incidence of 3.4%. Two of the 7 patients came to medical attention because of hematemesis. The hepatic abnormality in the others was detected because of hepatosplenomegaly
1311
Section XII. Inherited and Pediatric Diseases of the Liver
and failure to thrive.565 In two further studies, 24% of 233 adults with cystic fibrosis in the UK had abnormal results of liver tests.566 Similarly, 35% of 31 Swedish teenagers with cystic fibrosis had hepatosplenomegaly.567 To assess the evolution of liver disease in cystic fibrosis patients, Ling and co-workers prospectively followed 124 children with cystic fibrosis for 4 years. At the initial assessment 42% had hepatic biochemical abnormalities, 35% ultrasound abnormalities, and 6% had clinical abnormalities of the liver. In cross-sectional analysis, abnormal biochemistry was present in 40% of children with ultrasound and clinical abnormalities. Sixty-eight percent of the children showed ultrasound or clinical evidence of liver abnormality at some point during the 4-year follow-up. However, no association between liver disease and nutritional status was found.568 Documentation of mild hepatic involvement is often difficult, because clinical manifestations, including jaundice, may be lacking. Results of liver tests may be normal. The BSP excretion test may be of help in detecting early hepatic involvement.569 The best early indication of hepatic involvement is an elevated level of the hepatic isoenzyme of alkaline phosphatase.570,571 Total serum alkaline phosphatase activity is not reliable in the evaluation of hepatic involvement in children with cystic fibrosis because age-related normal values for alkaline phosphatase are hard to establish. Also, puberty is often delayed in patients with cystic fibrosis; hence, total alkaline phosphatase may appear to be normal in the presence of liver disease.565 It is important to emphasize that hepatic disease, including portal hypertension, may be present in patients with cystic fibrosis even when results of liver tests are normal or only slightly abnormal. In fact, the first manifestation of cystic fibrosis may be portal hypertension in a previously asymptomatic patient. Thus, children and young adults with unexplained portal hypertension or other hepatic abnormalities should be tested by pilocarpine iontophoresis and by chemical determination of the chloride level in sweat.
1312
conjugated bile acids was lower in enzyme-treated patients with cystic fibrosis than in those not currently receiving treatment. The high ratio of cholic acid to chenodeoxycholic acid is a well-known response to malabsorption of bile acids. The decrease of this ratio and the concomitant normalization of biliary lipids during treatment of patients with cystic fibrosis with pancreatic enzymes suggest that the lithogenic bile in cystic fibrosis is secondary to bile acid malabsorption.574
BIOCHEMICAL CHARACTERISTICS Cystic fibrosis is related to an abnormality in secretions of the exocrine glands. These secretions contain high concentrations of sodium chloride and defective chloride secretion is the hallmark of the disease.575–577 The biochemical characteristics of the liver disorder are not well understood. Intrahepatic obstruction by eosinophilic concentrations of mucus occurs early in the disease and has been proposed as the cause of cirrhosis. Shwachman found that cirrhosis did not develop in patients with cystic fibrosis who had pulmonary involvement without pancreatic involvement. Thus, defects in digestion and absorption appear to be prerequisites for liver disease, except possibly during the neonatal period.578 Patients with cystic fibrosis and pancreatic insufficiency have markedly elevated fecal bile acid excretion compared with age-matched controls.579 Bile acid kinetics were investigated by double-isotope dilution technique in six children with previously untreated cystic fibrosis. All six children had pancreatic insufficiency. The bile acid pool sizes were small and turned over rapidly in untreated patients. When fat excretion was reduced by therapy, the turnover rate of the pool decreased, with twofold enlargement of the pool size. These findings suggest an interruption of enterohepatic circulation.580 The pathophysiologic mechanism of the interruption of the enterohepatic circulation and its relation to hepatobiliary disease has not been defined.
Gallbladder Disease
MOLECULAR BASIS OF CYSTIC FIBROSIS
Clinical symptoms of cholelithiasis and cholecystitis have been found in some patients with cystic fibrosis. Microgallbladder has been reported to be present in approximately 15% of cases and poor or non-visualization of the gallbladder in another 25%.572 Gallbladder stones are found in as many as 8% of patients.573 The gallbladder disease in these patients may be related to occlusion of the cystic duct by precipitation of abnormal secretions and to abnormal biliary lipid composition.574 The pathogenesis of the increased incidence of gallbladder stones in cystic fibrosis was investigated by analysis of biliary lipid in 26 patients with cystic fibrosis, 7 children with cholelithiasis without cystic fibrosis, and 13 controls. For 14 patients with cystic fibrosis who had stopped taking pancreatic enzymes a week earlier, the molar percentage of lipid composition accounted for by cholesterol (mean ± SE 16.3 ± 2.9) and the saturation index (2.0 ± 0.3) were comparable to values obtained for the group with cholelithiasis but no cystic fibrosis. For 12 patients with cystic fibrosis taking pancreatic enzymes, the molar percentage of cholesterol (8.6 ± 1.7) and the saturation index (1.0 ± 0.1) did not differ from those of controls. Bile-salt analysis revealed a striking preponderance of cholic acid over chenodeoxycholic acid, and the glycine-to-taurine ratio of
The cystic fibrosis gene has been cloned and mapped to chromosome 7.581 The gene spans 250 000 base pairs and encodes a membrane protein of 1480 amino acids termed cystic fibrosis transmembrane conductance regulator.582 CFTR is believed to be a cyclic adenosine monophosphate (AMP)-regulated chloride channel.583 CFTR is predicted to have five domains: two membranespanning domains, two nucleotide-binding domains, and a unique regulatory (R) domain. The membrane-spanning domains appear to contribute to the organization of the chloride channel, because mutations of specific amino acid residues within the first membranespanning domain alter the anion selectivity of the channel. Phosphorylation of the (R) domain, generally by cyclic AMP-dependent protein kinase, is essential for opening of the channel.584 More than 750 mutations within the cystic fibrosis gene have been reported.585 The most common mutation is a 3 base pair deletion removing a phenylalanine residue at amino acid position 508 (D508). The 508 mutation accounts for approximately 70% of cystic fibrosis cases in the USA.586 There are four mechanisms by which mutations disrupt the function of CFTR (Figure 69-15).584 Class I mutations (defective protein products) result in production of little or no full-length protein, with loss of CFTR chloride channel function. These
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
CI–
MSD1 Class III. Defective regulation
MSD2 Class IV. Defective conduction
NBD2
NBD1
ATP ATP Golgi
Class II. Defective processing
PKA ATP
E.R.
Figure 69-15. Biosynthesis and function of cystic fibrosis transconductor regulator (CFTR). ER, endoplasmic reticulum; MSD, membrane-spanning domain; NBD, nuclear-binding domain. (Adapted from Welsh MJ, Smith AE. Molecular mechanisms of CFTR chloride channel dysfunction in cystic fibrosis. Cell 1993; 73:1251–1254, with permission. © by Cell Press.)
Nucleus
Genetic defect
Class I. Defective protein production
mutations produce premature termination signals because of frameshifts due to insertions, deletions, or nonsense mutations.587 Class II mutations (defective protein processing) result in failure to traffic the CFTR to the correct cellular location.588 This class includes the most common mutation (DF508). Class III mutations (defective regulation) result from mutations in the nucleotidebinding proteins.588 Class IV mutations lead to defective chloride conduction. Several mutations in the first membrane-spanning domain affect arginine residues located in putative membranespanning sequences. These mutations result in markedly reduced rates of ion flow through single open channels in inside-out membrane patches. The development of liver disease in patients with cystic fibrosis was not found to be related to specific mutations in the gene; other environmental or genetic factors may influence the development of liver disease. In this regard a recent study showed that the presence of liver cirrhosis in cystic fibrosis patients is significantly associated with the presence of mutated mannose-binding lectin variants.589 Mannose lectin binding is an important protein of the immune system and has been shown to be a modulating protein in the respiratory involvement of cystic fibrosis.
LABORATORY FINDINGS Sweat chloride levels are elevated in all patients with cystic fibrosis. Total pancreatic exocrine dysfunction is found in 80–85% of the patients. Five to 10% of patients have partial insufficiency. The remainder have normal exocrine function. Steatorrhea and azotorrhea are the major consequences of the pancreatic insufficiency. Malabsorption of fat-soluble vitamins A, D, E, and K is the result of pancreatic insufficiency and the use of antibiotics to treat the pulmonary infections also alters the bacterial flora of the intestinal tract. Vitamin K deficiency may occur in the first year of life and produce overt hemorrhage.590 Vitamin A levels are low in the serum but normal or elevated in the liver. A defect in the transport of vitamin A out of the liver is probable. Night blindness and benign increases in intracranial pressure secondary to low vitamin A levels in blood have been reported.591,592 Vitamin D is poorly absorbed; however,
overt rickets is rare. Serum levels of 25-OH vitamin D are reported to be similar to values obtained for normal controls.593 Low serum levels of 25-OH vitamin D in 4 of 17 patients have been demonstrated in one series594 and in all 21 patients in another.595 Differences in serum levels reflect different assays and the times of the year during which the patients were studied, because 85% of circulating 25-OH vitamin D is of endogenous origin.593 When patients were studied in winter and early spring, values of serum 25-OH vitamin D were low;594 however, the other studies were conducted in the summer and early fall.593,595 These observations suggest a subtle deficiency of vitamin D in patients who have cystic fibrosis and are deprived of sunlight and endogenous vitamin D production. Liver tests usually show no abnormality. Conjugated hyperbilirubinemia and elevations of serum AST, ALT, and alkaline phosphatase are noted in neonates with obstructive jaundice secondary to cystic fibrosis. Some infants with cystic fibrosis have hypoproteinemia and edema as a result of protein malabsorption, particularly infants who have been breast-fed or fed soybean-based formulas. Breast milk contains only 1.1% protein; although adequate for normal infants, it is suboptimal for those who have cystic fibrosis. Soybean-based formula contains small amounts of antitryptic activity, which may potentiate the existing malabsorption. Detecting mild liver involvement in cystic fibrosis is often difficult. The best early indication is elevated alkaline phosphatase of hepatic origin. Measurement of total serum alkaline phosphatase is not helpful due to active bone formation in children. The kinetics of BSP excretion show a decrease in hepatic removal rate and biliary secretion and an increase in hepatic-to-plasma reflux as liver disease progresses.569 In patients with cystic fibrosis and biliary cirrhosis, abnormal liver function and hematologic abnormalities or hypersplenism are usually detected.
PATHOLOGY Focal biliary cirrhosis is found in approximately 25% of patients with cystic fibrosis examined by necropsy.552 Progression to a multilobular cirrhosis occurs in about 5% of surviving children and adolescents.
1313
Section XII. Inherited and Pediatric Diseases of the Liver
Three histologic types of lesions were identified by Oppenheimer and Esterly596 in young infants with cystic fibrosis: 1. Focal biliary cirrhosis, characterized by bile duct proliferation, inspissated granular eosinophilic material filling the biliary ducts, infiltration of the liver with chronic inflammatory cells, and variable degrees of fibrosis, is most common. The lesions are focal in distribution and vary in number and severity. The incidence of focal biliary cirrhosis increases with age. 2. Excess mucus in biliary duct radicles or in the linings of epithelial cells with periportal changes may be present. The distribution of the biliary mucus is variable and focal, but it is found most often in the large intrahepatic ducts. Biliary duct proliferation and, less commonly, metaplasia of the ductal epithelium are observed. The periportal changes include edema and an inflammatory cell infiltrate. This type of lesion is commonly seen in infants younger than 1 year. 3. Periportal changes without biliary mucus may be seen on biopsy. These changes are common in infants younger than 3 months but are not observed in older children. It appears that the periportal changes are non-specific and transitory. The accumulation of biliary mucus appears to be related to cystic fibrosis, and because its frequency decreases with age, this indicates that biliary mucus is only one factor in the pathogenesis of focal biliary cirrhosis. Why such a small percentage of the patients (5%) develop multilobular biliary cirrhosis remains uncertain. Shwachman noted a familial tendency to develop hepatic complications.578 Nutritional deficiencies and intercurrent infection do not explain the progression to multilobular cirrhosis. The relationship in cystic fibrosis between the abnormalities in bile-salt metabolism and hepatobiliary disease awaits further investigation. Hemosiderosis and fatty infiltration of the liver are two other pathologic features of the liver in cystic fibrosis. The increased intestinal absorption of iron that occurs in untreated patients with pancreatic insufficiency may produce hepatic hemosiderosis. The early institution of supplemental pancreatic enzymes reduces the increased absorption of iron.597 Hepatic infiltration with fat secondary to chronic malnutrition is frequently observed with cystic fibrosis.578
GENETICS Cystic fibrosis is inherited as an autosomal recessive trait.598 The highest reported incidence of cystic fibrosis is in Caucasians, primarily those of European origin. The incidence in the Caucasian population in the USA is approximately 1 : 2000.599 The disease is less frequent in blacks and Asians.600 No good incidence figures are available for non-whites, with the exception of Hawaiians, for whom the incidence has been reported to be 1 : 90 000.601 Prenatal diagnosis can be accomplished by a number of techniques, including reverse dot blots, amplification refractory, mutation detection systems, oligonucleotide ligation assays, the invader assay, and nanochip systems.602,603
THERAPY Hepatic complications of cystic fibrosis that need treatment include those of newborns with obstructive jaundice and those of patients
1314
first seen for treatment as adolescents with multilobular biliary cirrhosis and portal hypertension. Jaundiced neonates may escape severe hepatic involvement later in their lives. Drugs that stimulate bile flow are not helpful. Flushing of the extrahepatic biliary tree with normal saline has been successful in two cases.557,559 In some cases, spontaneous relief of the jaundice has occurred without treatment.559 The hydrophilic bile acid, ursodeoxycholic acid, has been recommended by the US Cystic Fibrosis Foundation Hepatobiliary Disease Consensus Group604 for parents with established liver disease. However, the Cochrane collaboration concluded that there was insufficient evidence to justify the use of ursodeoxycholic acid.605 Vitamin supplementation with ADEK preparation appears to reduce the incidence of vitamin K deficiency.606 More recently, Nathanson and co-workers showed that ursodeoxycholic acid stimulates ATP secretion by isolated hepatocytes, thus promoting bile flow.607 The decision to perform portosystemic shunting in an older patient with portal hypertension depends largely on the pulmonary status of the patient. Patients who have severe pulmonary disease are poor candidates for surgery. The long-term survival of these patients is poor, and the complication rate for such a major operation is very high. Those patients who have had single episodes of hemorrhaging or have esophageal varices with good pulmonary function are considered candidates for the shunting procedure.563,564 Liver, lung, and heart transplantation has been performed in few patients with cystic fibrosis.
REFERENCES 1. Sachs B, Sternfeld L, Kraus G. Essential fructosuria: its pathophysiology. Am J Dis Child 1974; 63:252. 2. Chalmers RA, Pratt RTC. Idiosyncrasy to fructose. Lancet 1956; 2:340. 3. Froesch ER, Prader A, Labhart A, et al. [Hereditary fructose intolerance, a congenital metabolic disorder unknown until now.] Schweiz Med Wochenschr 1957; 87:1168–1171. 4. Baker L, Winegrad AI. Fasting hypoglycaemia and metabolic acidosis associated with deficiency of hepatic fructose-1,6diphosphatase activity. Lancet 1970; 2:13–16. 5. Henry I, Gallano P, Besmond C, et al. The structural gene for aldolase B (ALDB) maps to 9q13-32. Ann Hum Genet 1985; 49:173–180. 6. Mukai T, Yatsuki H, Arai Y, et al. Human aldolase B gene: characterization of the genomic aldolase B gene and analysis of sequences required for multiple polyadenylations. J Biochem (Tokyo) 1987; 102:1043–1051. 7. Tolan DR, Penhoet EE. Characterization of the human aldolase B gene. Mol Biol Med 1986; 3:245–264. 8. Cross NC, Tolan DR, Cox TM. Catalytic deficiency of human aldolase B in hereditary fructose intolerance caused by a common missense mutation. Cell 1988; 53:881–885. 9. Cross NC, de Franchis R, Sebastio G, et al. Molecular analysis of aldolase B genes in hereditary fructose intolerance. Lancet 1990; 335:306–309. 10. Sebastio G, de Franchis R, Strisciuglio P, et al. Aldolase B mutations in Italian families affected by hereditary fructose intolerance. J Med Genet 1991; 28:241–243. 11. Tolan DR, Brooks CC. Molecular analysis of common aldolase B alleles for hereditary fructose intolerance in North Americans. Biochem Med Metab Biol 1992; 48:19–25. 12. Rellos P, Sygusch J, Cox TM. Expression, purification, and characterization of natural mutants of human aldolase B. Role of
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
13. 14.
15. 16.
17.
18. 19.
20.
21. 22. 23. 24.
25.
26.
27.
28. 29.
30.
31.
32.
33.
34. 35.
quaternary structure in catalysis. J Biol Chem 2000; 275:1145–1151. Cornblath M, Rosenthal IM, Reisner SH, et al. Hereditary fructose intolerance. N Engl J Med 1963; 269:1271–1278. Froesch ER, Wolf HP, Baitsch H, et al. Hereditary fructose intolerance. An inborn defect of hepatic fructose-1-phosphate splitting aldolase. Am J Med 1963; 34:151–167. Swales JD, Smith ADM. Adult fructose intolerance. Q J Med 1966; 35:455. Odievre M, Gentil C, Gautier M, et al. Hereditary fructose intolerance in childhood. Diagnosis, management, and course in 55 patients. Am J Dis Child 1978; 132:605–608. Mass RE, Smith WR, Welsh JR. The association of hereditary fructose intolerance and renal tubular acidosis. Am J Med Sci 1966; 251:516. Schulte MJ, Widukind L. Fatal sorbitol infusion in a patient with fructose-sorbitol intolerance. Lancet 1977; 2:188. Morris RC, Jun, Ueki I, et al. Absence of renal fructose-1phosphate aldolase activity in hereditary fructose intolerance. Nature 1967; 214:920–921. Perheentupa J, Pitkanen E, Nikkila EA, et al. Hereditary fructose intolerance. A clinical study of four cases. Ann Paediatr Fenn 1962; 8:221–235. Shapira F, Dreyfus JC. L’aldolase hepatique dans intolerance au fructose. Rev Fr Etud Clin Biol 1967; 12:486. Hers HG, Joassin G. [Anomaly of hepatic aldolase in intolerance to fructose.] Enzymol Biol Clin (Basel) 1961; 1:4–14. Gurtler B, Leuthardt F. [Heterogeneity of aldolases.] Helv Chim Acta 1970; 53:654–658. Rutter WJ, Richards OC, Woodfin BM. Comparative studies of liver and muscle aldolase. I. Effect of carboxypeptidase on catalytic activity. J Biol Chem 1961; 236:3193–3197. Drechler ER, Boyer PD, Kowalesky AG. The catalytic activity of carboxypeptidase-degraded aldolase. J Biol Chem 1959; 234:2627. Nordmann Y, Shapira F, Dreyfus JC. A structurally modified aldolase in fructose intolerance: immunologic and kinetic evidence. Biochem Biophys Res Commun 1968; 31:884. Esposito G, Vitagliano L, Santamaria R, et al. Structural and functional analysis of aldolase B mutants related to hereditary fructose intolerance. FEBS Lett 2002; 531:152–156. Froesch ER, Prader A, Wolf HP, et al. [Hereditary fructose intolerance.] Helv Paediatr Acta 1959; 14:99–112. Froesch ER. Essential fructosuria, hereditary fructose intolerance, and fructose-1,6-diphosphatase deficiency. In: Stanbury JB, Wyngaarden JB, Fredrickson DS, eds. The metabolic basis of inherited disease, 4th edn. New York: McGraw-Hill; 1978:131. van Den Berg G, Hue L, Hers HG. Effect of administration of fructose on glycolytic action of glucagon. An investigation of the pathogeny of hereditary fructose intolerance. Biochem J 1973; 134:637. Kaufmann U, Froesch ER. Inhibition of phosphorylase-a by fructose-1-phosphate, alpha-glycerophosphate and fructose-1,6diphosphate: explanation for fructose-induced hypoglycaemia in hereditary fructose intolerance and fructose-1,6-diphosphatase deficiency. Eur J Clin Invest 1973; 3:407–413. Woods HF, Eggleston LV, Krebs HA. The cause of hepatic accumulation of fructose 1-phosphate on fructose loading. Biochem J 1970; 119:501–510. Raivio KO, Kekomaki MP, Maenpaa PH. Depletion of liver adenine nucleotides induced by D-fructose. Dose-dependence and specificity of the fructose effect. Biochem Pharmacol 1969; 18:2615–2624. Levin B, Snodgrass GJ, Oberholzer VG, et al. Fructosaemia. Observations on seven cases. Am J Med 1968; 45:826–838. Dormandy TL, Porter RJ. Familial fructose and galactose intolerance. Lancet 1961; 1:1189–1194.
36. Turner RC, Spathis GS, Nabarro JD, et al. Familial fructose and galactose intolerance. Lancet 1972; 2:872. 37. Schwartz R, Gamsu H, Mulligan PB, et al. Transient intolerance to exogenous fructose in the newborn. J Clin Invest 1964; 43:333–340. 38. Dubois R, Loeb H, Ooms HA, et al. [Study of a case of functional hypoglycemia caused by intolerance to fructose.] Helv Paediatr Acta 1961; 16:90–96. 39. Milhaud G. Technique nouvelle de mise en evidence d’erreurs congenitales du metabolism chez l’homme. Argent Bras Endocr 1964; 13:49. 40. Levin B, Oberholzer VG, Snodgrass GJ, et al. Fructosaemia. An inborn error of fructose metabolism. Arch Dis Child 1963; 38:220–230. 41. Lelong M, Alagille D, Gentil C, et al. Cirrhose hepatique et tubulopathie par absence congenitale de l’aldolase hepatique: intolerance hereditare au fructose. Bull Soc Med Hop (Paris) 1962; 113:58. 42. Gentil C, Colin J, Valette AM, et al. [Study of carbohydrate metabolism in the course of hereditary fructose intolerance. attempt at interpretation of hypoglycemia.] Rev Fr Etud Clin Biol 1964; 14:596–607. 43. Nikkila EA, Perheentupa J. Non-esterified fatty acids and fatty liver in hereditary fructose intolerance. Lancet 1962; 2:1280. 44. Morris RC. Fructose-induced disruption of renal acidification in patients with hereditary fructose intolerance. J Clin Invest 1965; 44:1076. 45. Morris RC. Evidence for an acidification defect of the proximal renal tubule in experimental and clinical renal disease. J Clin Invest 1966; 45:1048. 46. Morris RC, Jr. An experimental renal acidification defect in patients with hereditary fructose intolerance. I. Its resemblance to renal tubular acidosis. J Clin Invest 1968; 47:1389–1398. 47. Morris RC, Jr. An experimental renal acidification defect in patients with hereditary fructose intolerance. II. Its distinction from classic renal tubular acidosis; its resemblance to the renal acidification defect associated with the Fanconi syndrome of children with cystinosis. J Clin Invest 1968; 47:1648–1663. 48. Phillips MJ, Path MC, Little JA, et al. Subcellular pathology intolerance: dental status of eight patients. Br Dent J 1968; 132:597. 49. Marthaler TM, Froesch ER. Hereditary fructose intolerance. Dental status of eight patients. Br Dent J 1967; 123:597–599. 50. Raivio K, Perheentupa J, Nikkila EA. Aldolase activities in the liver in parents of patients with hereditary fructose intolerance. Clin Chim Acta 1967; 17:275–279. 51. Lau J, Tolan DR. Screening for hereditary fructose intolerance mutations by reverse dot-blot. Mol Cell Probes 1999; 13:35–40. 52. Cornblath M, Schwartz R. Carbohydrate metabolism in the neonate. Philadelphia: WB Saunders; 1989. 53. Greenberg RE, Christiansen RO. The critically ill child: hypoglycemia. Pediatrics 1970; 46:915–920. 54. Aronson SM, Perle G, Saifer A, et al. Biochemical identification of the carrier state in Tay–Sachs disease. Proc Soc Exp Biol Med 1962; 64:4. 55. Schneck L, Perle G, Volk BW. Fructose tolerance in Tay–Sachs’ disease. Pediatrics 1965; 36:272–275. 56. Volk BW, Aronson SM, Saifer A. Fructose-1 phosphate aldolase deficiency in Tay–Sachs disease. Am J Med 1964; 36:481–484. 57. Greene HL, Stifel FB, Herman RH. Hereditary fructose intolerance: treatment with pharmacologic doses of folic acid. Clin Res 1972; 20:274. 58. Melancon SB, Khachadurian AK, Nadler HL, et al. Metabolic and biochemical studies in fructose 1,6-diphosphatase deficiency. J Pediatr 1973; 82:650–657. 59. Greene HL, Stifel FB, Herman RH. “Ketotic hypoglycemia” due to hepatic fructose-1,6-diphosphatase deficiency: treatment with folic acid. Am J Dis Child 1972; 124:415–418.
1315
Section XII. Inherited and Pediatric Diseases of the Liver
60. Kikawa Y, Shin YS, Inuzuka M, et al. Diagnosis of fructose-1,6bisphosphatase deficiency using cultured lymphocyte fraction: a secure and noninvasive alternative to liver biopsy. J Inherit Metab Dis 2002; 25:41–46. 61. Kikawa Y, Inuzuka M, Jin BY, et al. Identification of genetic mutations in Japanese patients with fructose-1,6-bisphosphatase deficiency. Am J Hum Genet 1997; 61:852–861. 62. Mason HH, Turner ME. Chronic galactosemia. Am J Dis Child 1935; 50:359. 63. Komrower GM, Schwartz V, Holzel A, et al. A clinical and biochemical study of galactosemia. Arch Dis Child 1956; 31:254. 64. Isselbacher KJ, Anderson EP, Kurahashi K, et al. Congenital galactosemia, a single enzymatic block in galactose metabolism. Science 1956; 123:635–636. 65. Donnell GN, Bergren WR, Cleland RS. Galactosemia. Pediatr Clin North Am 1960; 7:315–332. 66. Andersen MW, Williams VP, Helmer GR Jr, et al. Transferasedeficiency galactosemia: evidence for the lack of a transferase protein in galactosemic red cells. Arch Biochem Biophys 1983; 222:326–331. 67. Gitzelmann R. Hereditary galactokinase deficiency, a newly recognized cause of juvenile cataracts. Pediatr Res 1967; 1:14. 68. Chacko CM, McCrone L, Nadler HL. A study of galactokinase and glucose 4-epimerase from normal and galactosemic skin fibroblasts. Biochim Biophys Acta 1972; 284:552–555. 69. Dahlqvist A, Gamstorp I, Madsen H. A patient with hereditary galactokinase deficiency. Acta Paediatr Scand 1970; 59: 669–675. 70. Gitzelmann R. Deficiency of uridine diphosphate galactose 4-epimerase in blood cells of an apparently healthy infant. Preliminary communication. Helv Paediatr Acta 1972; 27:125–130. 71. Gitzelmann R, Steinmann B, Mitchell B, et al. Uridine diphosphate galactose 4¢-epimerase deficiency. IV. Report of eight cases in three families. Helv Paediatr Acta 1977; 31:441–452. 72. Holton JB, Gillett MG, MacFaul R, et al. Galactosaemia: a new severe variant due to uridine diphosphate galactose-4-epimerase deficiency. Arch Dis Child 1981; 56:885–887. 73. Henderson MJ, Holton JB, MacFaul R. Further observations in a case of uridine diphosphate galactose-4-epimerase deficiency with a severe clinical presentation. J Inherit Metab Dis 1983; 6:17–20. 74. Garibaldi LR, Canini S, Superti-Furga A, et al. Galactosemia caused by generalized uridine diphosphate galactose-4-epimerase deficiency. J Pediatr 1983; 103:927–930. 75. Bowling FG, Fraser DK, Clague AE, et al. A case of uridine diphosphate galactose-4-epimerase deficiency detected by neonatal screening for galactosaemia. Med J Aust 1986; 144:150–151. 76. Stenstam T. Peroral and intravenous galactose tests: comparative study of their significance in different conditions. Acta Med Scand 1946; 177 (Suppl). 77. Segal S, Blair A. Some observations on the metabolism of Dgalactose in normal man. J Clin Invest 1961; 40:2016. 78. Tygstrup N, Winkler K. Galactose blood clearance as a measure of hepatic blood flow. Clin Sci (Lond) 1958; 17:1–9. 79. Colcher H, Patek AJ, Kendall FE. Galactose disappearance from the blood stream: calculation of a galactose removal constraint and its application as a test of liver function. J Clin Invest 1946; 25:768. 80. Tygstrup N. Determination of the hepatic elimination capacity (Lm) of galactose by single injection. Scand J Clin Lab Invest Suppl 1966; 18:118–125. 81. Lemaire HG, Muller-Hill B. Nucleotide sequences of the gal E gene and the gal T gene of E. coli. Nucleic Acids Res 1986; 14:7705–7711.
1316
82. Tajima M, Nogi Y, Fukasawa T. Primary structure of the Saccharomyces cerevisiae GAL7 gene. Yeast 1985; 1: 67–77. 83. Flach JE, Reichardt JK, Elsas LJ, 2nd. Sequence of a cDNA encoding human galactose-1-phosphate uridyl transferase. Mol Biol Med 1990; 7:365–369. 84. Field TL, Reznikoff WS, Frey PA. Galactose-1-phosphate uridylyltransferase: identification of histidine-164 and histidine166 as critical residues by site-directed mutagenesis. Biochemistry 1989; 28:2094–2099. 85. Reichardt JK, Packman S, Woo SL. Molecular characterization of two galactosemia mutations: correlation of mutations with highly conserved domains in galactose-1-phosphate uridyl transferase. Am J Hum Genet 1991; 49:860–867. 86. Wang BB, Xu YK, Ng WG, et al. Molecular and biochemical basis of galactosemia. Mol Genet Metab 1998; 63:263–269. 87. Reichardt JK, Levy HL, Woo SL. Molecular characterization of two galactosemia mutations and one polymorphism: implications for structure–function analysis of human galactose-1-phosphate uridyltransferase. Biochemistry 1992; 31:5430–5433. 88. Tyfield L, Reichardt J, Fridovich-Keil J, et al. Classical galactosemia and mutations at the galactose-1-phosphate uridyl transferase (GALT) gene. Hum Mutat 1999; 13:417–430. 89. Komrower GM, Lee DH. Long-term follow-up of galactosaemia. Arch Dis Child 1970; 45:367–373. 90. Koch R, Donnell GN, Fishler K, et al. Galactosemia. In: Kelley VC, ed. Practice of pediatrics. Hagerstown, MD: Harper & Row; 1979:14. 91. Segal S, Blair A, Roth H. The metabolism of galactose by patients with congenital galactosemia. Am J Med 1965; 38:62–70. 92. Segal S. Disorders of galactose metabolism. In: Stanbury JB, Wyngaarden JB, Fredrickson DS, eds. The metabolic basis of inherited disease. New York: McGraw-Hill; 1978:164. 93. Levy HL, Sepe SJ, Shih VE, et al. Sepsis due to Escherichia coli in neonates with galactosemia. N Engl J Med 1977; 297:823–825. 94. Segal S, Cuatrecasas P. The oxidation of 14C galactose by patients with congenital galactosemia. Evidence for a direct oxidative pathway. Am J Med 1968; 44:340. 95. Sidbury JB, Jr. The role of galactose-1-phosphate in the pathogenesis of galactosemia. In: Gardner LE, ed. Molecular genetics and human disease. Springfield, IL: Charles C Thomas; 1960:61. 96. Tada K. Glycogenesis and glycolysis in the liver from congenital galactosemia. Tohoku J Exp Med 1964; 82:168–171. 97. van Heyningen R. Formation of polyols by the lens of the rat with “sugar” cataracts. Nature 1959; 184:194. 98. Kinoshita JH, Dvornik D, Kraml M, et al. The effect of an aldose reductase inhibitor on the galactose-exposed rabbit lens. Biochim Biophys Acta 1968; 158:472–475. 99. Kinoshita JH, Barber GW, Merola LO, et al. Changes in the levels of free amino acids and myo-inositol in the galactoseexposed lens. Invest Ophthalmol 1969; 8:625–632. 100. Dische Z, Zelmenis G, Youlus J. Studies on protein and protein synthesis. Am J Ophthalmol 1957; 44:332–340; discussion 340–341. 101. Sippel TO. Enzymes of carbohydrate metabolism in developing galactose cataracts of rats. Invest Ophthalmol 1967; 6:59–63. 102. Korc I. Biochemical studies on cataracts in galactose fed rats. Arch Biochem 1961; 94:196. 103. Sippel TO. Energy metabolism in the lens during development of galactose cataract in rats. Invest Ophthalmol 1966; 5:576–582. 104. Kinoshita JH, Merola LO, Tung B. Changes in cation permeability in the galactose-exposed rabbit lens. Exp Eye Res 1968; 7:80–90.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
105. Heffley JD, Williams RJ. The nutritional teamwork approach: prevention and regression of cataracts in rats. Proc Natl Acad Sci USA 1974; 71:4164–4168. 106. Schwarz V. The value of galactose phosphate determinations in the treatment of galactosaemia. Arch Dis Child 1960; 35:428–432. 107. Quan-Ma R, Wells HJ, Wells WW, et al. Galactitol in the tissues of a galactosemic child. Am J Dis Child 1966; 112:477–478. 108. Quan-Ma R, Wells WW. The distribution of galactitol in tissues of rats fed galactose. Biochem Biophys Res Commun 1965; 20:486–490. 109. Keppler D, Decker K. Studies on the mechanism of galactosamine-1-phosphate and its inhibition of UDP-glucose pyrophosphorylase. Eur J Biochem 1969; 10:219–225. 110. Fox M, Thier S, Rosenberg L, et al. Impaired renal tubular function induced by sugar infusion in man. J Clin Endocrinol Metab 1964; 24:1318–1327. 111. Rosenberg LE, Weinberg AN, Segal S. The effect of high galactose diets on urinary excretion of amino acids in the rat. Biochim Biophys Acta 1961; 48:500–505. 112. Thier S, Fox M, Rosenberg L, et al. Hexose inhibition of amino acid uptake in the rat-kidney-cortex slice. Biochim Biophys Acta 1964; 93:106–115. 113. Saunders SJ, Isselbacher KJ. Inhibition of intestinal amino acid transport by hexoses. Biochim Biophys Acta 1965; 102:397–409. 114. Wells WW, Pittman TA, Wells HJ, et al. The isolation and identification of galactitol from the brains of galactosemia patients. J Biol Chem 1965; 240:1002–1004. 115. Wells HJ, Gordon M, Segal S. Galactose toxicity in the chick: oxidation of radioactive galactose. Biochim Biophys Acta 1970; 222:327–332. 116. Granett SE, Kozak LP, McIntyre JP, et al. Studies on cerebral energy metabolism during the course of galactose neurotoxicity in chicks. J Neurochem 1972; 19:1659–1670. 117. Knull HR, Lobert PF, Wells WW. Galactose neurotoxicity in chicks: effects on fast axoplasmic transport. Brain Res 1974; 79:524–527. 118. Malone JI, Wells H, Segal S. Decreased uptake of glucose by brain of the galactose toxic chick. Brain Res 1972; 43:700–704. 119. Malone JI, Wells HJ, Segal S. Galactose toxicity in the chick: hyperosmolality. Science 1971; 174:952–954. 120. Knull HR, Wells WW. Recovery from galactose-induced neurotoxicity in the chick by the administration of glucose. J Neurochem 1973; 20:415–422. 121. Woolley DW, Gommi BW. Serotonin receptors. IV. Specific deficiency of receptors in galactose poisoning and its possible relationship to the idiocy of galactosemia. Proc Natl Acad Sci USA 1964; 52:14–19. 122. Fraser IS, Russell P, Greco S, et al. Resistant ovary syndrome and premature ovarian failure in young women with galactosaemia. Clin Reprod Fertil 1986; 4:133–138. 123. Heidenreich RA, Mallee J, Rogers S, et al. Developmental and tissue-specific modulation of rat galactose-1-phosphate uridyltransferase steady state messenger RNA and specific activity levels. Pediatr Res 1993; 34:416–419. 124. Robbins SL, Cotran RS. Diseases of infancy and childhood. In: Robbins SL, Cotran RS, eds. Pathologic basis of disease, 2nd edn. Philadelphia: WB Saunders; 1979:582. 125. Smetana HF, Olen E. Hereditary galactose disease. Am J Clin Pathol 1962; 38:3–25. 126. Walker FA, Hsia DY, Slatis HM, et al. Galactosemia: a study of twenty-seven kindreds in North America. Ann Hum Genet 1962; 25:287–311. 127. Kirkman HN, Bynum E. Enzymic evidence of a galactosemic trait in parents of galactosemic children. Ann Hum Genet 1959; 23:117–126.
128. Mellman WJ, Tedesco TA, Feige P. Estimation of the gene frequency of the Duarte variant of galactose-1-phosphate uridyl transferase. Ann Hum Genet 1968; 32:1. 129. Brandt NJ. Frequency of heterozygotes for hereditary galactosemia in a normal population. Acta Genet (Basel) 1967; 17:289. 130. Tedesco TA, Miller KL, Rawnsley BE, et al. Human erythrocyte galactokinase and galactose-1-phosphate uridylyltransferase: a population survey. Am J Hum Genet 1975; 27:737–747. 131. Scriver CR. Population screening: report of a workshop. Prog Clin Biol Res 1985; 163B:89–152. 132. Kleijer WJ, Janse HC, van Diggelen OP, et al. First-trimester diagnosis of galactosaemia. Lancet 1986; 1:748. 133. Anderson EP, Kalckar HM, Kurahashi K, et al. A specific enzymatic assay for the diagnosis of congenital galactosemia. I. The consumption test. J Lab Clin Med 1957; 50:469–477. 134. Kliegman RM, Sparks JW. Perinatal galactose metabolism. J Pediatr 1985; 107:831–841. 135. Gross KC, Acosta PB. Fruits and vegetables are a source of galactose: implications in planning the diets of patients with galactosaemia. J Inherit Metab Dis 1991; 14:253–258. 136. Walter JH, Collins JE, Leonard JV. Recommendations for the management of galactosaemia. UK Galactosaemia Steering Group. Arch Dis Child 1999; 80:93–96. 137. Manis FR, Cohn LB, McBride-Chang C, et al. A longitudinal study of cognitive functioning in patients with classical galactosaemia, including a cohort treated with oral uridine. J Inherit Metab Dis 1997; 20:549–555. 138. Berry GT. The role of polyols in the pathophysiology of hypergalactosemia. Eur J Pediatr 1995; 154 (suppl 2):S53–S64. 139. Schweitzer S, Shin Y, Jakobs C, et al. Long-term outcome in 134 patients with galactosaemia. Eur J Pediatr 1993; 152:36–43. 140. Kaufman FR, Kogut MD, Donnell GN, et al. Hypergonadotropic hypogonadism in female patients with galactosemia. N Engl J Med 1981; 304:994–998. 141. Steinmann B, Gitzelmann R, Zachmann M. Hypogonadism and galactosemia. N Engl J Med 1981; 304:995. 142. Chen YT, Mattison DR, Feigenbaum L, et al. Reduction in oocyte number following prenatal exposure to a diet high in galactose. Science 1981; 214:1145–1147. 143. Chen YT, Mattison DR, Bercu BB, et al. Resistance of the male gonad to a high galactose diet. Pediatr Res 1984; 18:345–348. 144. Kaufman FR, Loro ML, Azen C, et al. Effect of hypogonadism and deficient calcium intake on bone density in patients with galactosemia. J Pediatr 1993; 123:365–370. 145. Renner C, Razeghi S, Uberall MA, et al. Hormone replacement therapy in galactosaemic twins with ovarian failure and severe osteoporosis. J Inherit Metab Dis 1999; 22:194–195. 146. Campbell S, Kulin HE. Transient thyroid binding globulin deficiency with classic galactosemia. J Pediatr 1984; 105:335–336. 147. Holtzman NA. Pitfalls of newborn screening (with special attention to hypothyroidism): when will we ever learn? Birth Defects Orig Artic Ser 1983; 19:111–120. 148. Segal S. Disorders of galactose metabolism. In: Stanbury JB, Wyngaarden JB, Fredrickson D, eds. The metabolic basis of inherited disease, 5th edn. New York: McGraw-Hill; 1983:167. 149. Sangiuolo F, Magnani M, Stambolian D, et al. Biochemical characterization of two GALK1 mutations in patients with galactokinase deficiency. Hum Mutat 2004; 23:396. 150. Gitzelmann R, Haigis E. Appearance of active UDP-galactose 4¢epimerase in cells cultured from epimerase-deficient persons. J Inherit Metab Dis 1978; 1:41. 151. von Gierke E. Hepato-nephromegalia glykogenia (Glykogenspeicherkrankheiten Leber und Nieren). Beitr Pathol Anat 1929; 82:497. 152. Pompe JC. Over idiopathische hypertrophie van het hart. Ned Tijdschr Geneeskd 1932; 76:304.
1317
Section XII. Inherited and Pediatric Diseases of the Liver
153. Cori GT, Cori CF. Glucose-6-phosphatase of the liver in glycogen storage disease. J Biol Chem 1952; 199:661–667. 154. Hug G. Glycogen storage disease. Birth Defects 1976; 12:145. 155. Howell RR. The glycogen storage diseases. In: Stanbury JB, Wyngaarden JB, Fredrickson DS, et al., eds. The metabolic basis of inherited disease, 4th edn. New York: McGraw-Hill; 1977:137. 156. Senior B, Sadeghi-Nejad A. The glycogenoses and other inherited disorders of carbohydrate metabolism. Clin Perinatol 1976; 3:79–98. 157. Tarui S, Okuno G, Ikura Y, et al. Phosphofructokinase deficiency in skeletal muscle. A new type of glycogenesis. Biochem Biophys Res Commun 1965; 19:517–523. 158. Burchell A. Molecular pathology of glucose-6-phosphatase. Faseb J 1990; 4:2978–2988. 159. Countaway JL, Waddell ID, Burchell A, et al. The phosphohydrolase component of the hepatic microsomal glucose-6-phosphatase system is a 36.5-kilodalton polypeptide. J Biol Chem 1988; 263:2673–2678. 160. Waddell ID, Zomerschoe AG, Burchell A. Isolation of a cDNA coding for the glucose-6-phosphatase stabilizing protein. Biochem Soc Trans 1990; 18:582–583. 161. Nordlie RC, Sukalski KA. Multiple forms of type 1 glucogen storage disease. Trends Biochem Sci 1986; 11:88. 162. Burchell A, Waddell ID, Stewart L, et al. Perinatal diagnosis of type 1c glycogen storage disease. J Inherit Metab Dis 1989; 12 (Suppl 2):315–317. 163. Burchell A, Jung RT, Lang CC, et al. Diagnosis of type 1a and type 1c glycogen storage diseases in adults. Lancet 1987; 1:1059–1062. 164. Shelly LL, Lei KJ, Pan CJ, et al. Isolation of the gene for murine glucose-6-phosphatase, the enzyme deficient in glycogen storage disease type 1A. J Biol Chem 1993; 268:21482–21485. 165. Brody LC, Abel KJ, Castilla LH, et al. Construction of a transcription map surrounding the BRCA1 locus of human chromosome 17. Genomics 1995; 25:238–247. 166. Lei KJ, Shelly LL, Pan CJ, et al. Mutations in the glucose-6phosphatase gene that cause glycogen storage disease type 1a. Science 1993; 262:580–583. 167. Lei KJ, Pan CJ, Shelly LL, et al. Identification of mutations in the gene for glucose-6-phosphatase, the enzyme deficient in glycogen storage disease type 1a. J Clin Invest 1994; 93:1994–1999. 168. Hemrika W, Wever R. A new model for the membrane topology of glucose-6-phosphatase: the enzyme involved in von Gierke disease. FEBS Lett 1997; 409:317–319. 169. Pan CJ, Lei KJ, Annabi B, et al. Transmembrane topology of glucose-6-phosphatase. J Biol Chem 1998; 273:6144–6148. 170. Lei KJ, Chen YT, Chen H, et al. Genetic basis of glycogen storage disease type 1a: prevalent mutations at the glucose-6phosphatase locus. Am J Hum Genet 1995; 57:766–771. 171. Parvari R, Lei KJ, Bashan N, et al. Glycogen storage disease type 1a in Israel: biochemical, clinical, and mutational studies. Am J Med Genet 1997; 72:286–290. 172. Lei KJ, Pan CJ, Liu JL, et al. Structure–function analysis of human glucose-6-phosphatase, the enzyme deficient in glycogen storage disease type 1a. J Biol Chem 1995; 270:11882–11886. 173. Kajihara S, Matsuhashi S, Yamamoto K, et al. Exon redefinition by a point mutation within exon 5 of the glucose-6-phosphatase gene is the major cause of glycogen storage disease type 1a in Japan. Am J Hum Genet 1995; 57:549–555. 174. Stroppiano M, Regis S, DiRocco M, et al. Mutations in the glucose-6-phosphatase gene of 53 Italian patients with glycogen storage disease type Ia. J Inherit Metab Dis 1999; 22:43–49. 175. Emmett M, Narins RG. Renal tranplantation in type 1 glycogenesis. Failure to improve glucose metabolism. JAMA 1978; 239:1642–1644.
1318
176. Greene HL, Slonim AE, O’Neill JA, Jr, et al. Continuous nocturnal intragastric feeding for management of type 1 glycogen-storage disease. N Engl J Med 1976; 294:423–425. 177. Roe TF, Kogut MD. Chronic effects of oral glucose alimentation and portacaval shunt in patients with glycogen storage disease type I (abstract). Pediatr Res 1976; 10:414. 178. Baker L, Mills JL. Long-term treatment of glycogen storage disease type I: clinical improvement but persistent abnormalities of lactate and triglyceride concentration (abstract). Pediatr Res 1978; 12:502. 179. Crigler JF, Jr, Folkman J. Glycogen storage disease: new approaches to therapy. Ciba Found Symp 1977; 55:331–351. 180. Coleman JE. Metabolic interrelationships between carbohydrates, lipids, and proteins. In: Bondy PK, Rosenberg LE, eds. Diseases of metabolism, 7th edn. Philadelphia: WB Saunders; 1974:107. 181. Slonim AE, Lacy WW, Terry A, et al. Nocturnal intragastric therapy in type I glycogen storage disease: effect on hormonal and amino acid metabolism. Metabolism 1979; 28:707–715. 182. Sadeghi-Nejad A, Presente E, Binkiewicz A, et al. Studies in type I glycogenesis of the liver. The genesis and disposition of lactate. J Pediatr 1974; 85:49–54. 183. Jakovcic S, Khachadurian AK, Hsia DY. The hyperlipidemia in glycogen storage disease. J Lab Clin Med 1966; 68: 769–779. 184. Howell RR, Ashton DM, Wyngaarden JB. Glucose-6phosphatase deficiency glycogen storage disease. Studies on the interrelationships of carbohydrate, lipid, and purine abnormalities. Pediatrics 1962;29:553–565. 185. Ockerman PA. Glucose, glycerol and free fatty acids in glycogen storage disease type 1. Blood levels in the fasting and non-fasting state. Effect of glucose and adrenalin administration. Clin Chim Acta 1965; 12:370–382. 186. Forget PP, Fernandes J, Begemann PH. Triglyceride clearing in glycogen storage disease. Pediatr Res 1974; 8:114–119. 187. Hulsmann WC, Eijkenboom WH, Koster JF, et al. Glucose-6phosphatase deficiency and hyperlipaemia. Clin Chim Acta 1970; 30:775–778. 188. Fine RN, Strauss J, Donnell GN. Hyperuricemia in glycogenstorage disease type 1. Am J Dis Child 1966; 112:572–576. 189. Howell RR. The interrelationship of glycogen storage disease and gout. Arthritis Rheum 1965; 8:780–785. 190. Alepa FP, Howell RR, Klinenberg JR, et al. Relationships between glycogen storage disease and tophaceous gout. Am J Med 1967; 42:58–66. 191. Jakovcic S, Sorensen LB. Studies of uric acid metabolism in glycogen storage disease associated with gouty arthritis. Arthritis Rheum 1967; 10:129–134. 192. Henderson JF, Khoo MK. Synthesis of 5-phosphoribosyl 1pyrophosphate from ribonucleosides in Ehrlich ascites tumor cells in vitro. J Biol Chem 1965; 240:2363–2366. 193. Greene ML, Seegmiller JE. Elevated erythrocyte phosphoribosylpyrophosphate in X-linked uric aciduria: importance of PRPP concentration in regulation of human purine biosynthesis. J Clin Invest 1969; 48:32a. 194. Raivio KO, Seegmiller JE. Role of glutamine in purine synthesis and in guanine nucleotide formation in normal fibroblasts and in fibroblasts deficient in hypoxanthine phosphoribosyltransferase activity. Biochim Biophys Acta 1973; 299:283–292. 195. Howell RR. Hyperuricemia in childhood. Fed Proc 1968; 27:1078–1082. 196. Brosh S, Boer P, Kupfer B, et al. De novo synthesis of purine nucleotides in human peripheral blood leukocytes. Excessive activity of the pathway in hypoxanthine-guanine phosphoribosyltransferase deficiency. J Clin Invest 1976; 58:289–297. 197. Holmes EW, McDonald JA, McCord JM, et al. Human glutamine phosphoribosylpyrophosphate amidotransferase.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
198.
199.
200. 201. 202. 203.
204. 205.
206.
207.
208. 209. 210.
211.
212. 213.
214.
215.
216.
217.
218.
219.
220.
Kinetic and regulatory properties. J Biol Chem 1973; 248:144–150. Greene HL, Wilson FA, Hefferan P, et al. ATP depletion, a possible role in the pathogenesis of hyperuricemia in glycogen storage disease type I. J Clin Invest 1978; 62:321–328. Bode JC, Zelder O, Rumpelt HJ, et al. Depletion of liver adenosine phosphates and metabolic effects of intravenous infusion of fructose or sorbitol in man and in the rat. Eur J Clin Invest 1973; 3:436–441. Roe TF, Kogut MD. The pathogenesis of hyperuricemia in glycogen storage disease, type I. Pediatr Res 1977; 11:664–669. Corby DG, Putnam CW, Greene HL. Impaired platelet function in glucose-6-phosphatase deficiency. J Pediatr 1974; 85:71–76. Cooper RA. Abnormalities of cell-membrane fluidity in the pathogenesis of disease. N Engl J Med 1977; 297:371–377. Howell RR, Stevenson RE, Ben-Menachem Y, et al. Hepatic adenomata with type 1 glycogen storage disease. JAMA 1976; 236:1481–1484. Roe TF, Kogut MD, Buckingham BA, et al. Hepatic tumors in glycogen-storage disease type I. Pediatr Res 1979; 13:931. Levine G, Mierau G, Favara BE. Hepatic glycogenosis, renal glomerular cysts and hepatocarcinoma. Am J Pathol 1976; 82:PPC-37. Maruyama Y, Kida E, Takagi M. An autopsy case of glycogen storage disease in female adults. J Osaka Med Coll 1962; 25:207. Bannasch P, Hacker HJ, Klimek F, et al. Hepatocellular glycogenosis and related pattern of enzymatic changes during hepatocarcinogenesis. Adv Enzyme Regul 1984; 22:97–121. Sidbury JB, Jr. The genetics of the glycogen storage diseases. Prog Med Genet 1965; 4:32–58. Farber M, Knuppel RA, Binkiewicz A, et al. Pregnancy and von Gierke’s disease. Obstet Gynecol 1976; 47:226–228. Havel RJ, Balasse EO, Williams HE, et al. Splanchnic metabolism in von Gierke’s disease (glycogenesis type I). Trans Assoc Am Phys 1969; 82:305–323. Tsalikian E, Simmons P, Gerich JE, et al. Glucose production and utilization in children with glycogen storage disease type I. Am J Physiol 1984; 247:E513–E519. Kalhan SC, Gilfillan C, Tserng KY, et al. Glucose production in type I glycogen storage disease. J Pediatr 1982; 101:159–160. Powell RC, Wentworth SM, Brandt IK. Endogenous glucose production in type I glycogen storage disease. Metabolism 1981; 30:443–450. Schwenk WF, Haymond MW. Optimal rate of enteral glucose administration in children with glycogen storage disease type I. N Engl J Med 1986; 314:682–685. Edelstein G, Hirshman CA. Hyperthermia and ketoacidosis during anesthesia in a child with glycogen-storage disease. Anesthesiology 1980; 52:90–92. McAdams AJ, Hug G, Bove KE. Glycogen storage disease, types I to X: criteria for morphologic diagnosis. Hum Pathol 1974; 5:463–487. Folkman J, Philippart A, Tze WJ, et al. Portacaval shunt for glycogen storage disease: value of prolonged intravenous hyperalimentation before surgery. Surgery 1972; 72: 306–314. Starzl TE, Marchioro TL, Sexton AW, et al. The effect of portacaval transposition on carbohydrate metabolism: experimental and clinical observations. Surgery 1965; 57:687–697. Riddell AG, Davies RP, Clark AD. Portacaval transposition in the treatment of glycogen-storage disease. Lancet 1966; 2:1146–1148. Burr IM, O’Neill JA, Karzon DT, et al. Comparison of the effects of total parenteral nutrition, continuous intragastric feeding, and portacaval shunt on a patient with type I glycogen storage disease. J Pediatr 1974; 85:792–795.
221. Greene HL, Slonim AE, Burr IM. Type I glycogen storage disease: a metabolic basis for advances in treatment. Adv Pediatr 1979; 26:63–92. 222. Levin S, Moses SW, Chayoth R, et al. Glycogen storage disease in Israel. A clinical, biochemical and genetic study. Isr J Med Sci 1967; 3:397–410. 223. Chen YT, Cornblath M, Sidbury JB. Cornstarch therapy in type I glycogen-storage disease. N Engl J Med 1984; 310:171–175. 224. Collins JE, Leonard JV. The dietary management of inborn errors of metabolism. Hum Nutr Appl Nutr 1985; 39:255–272. 225. Leonard JV, Dunger DB. Hypoglycaemia complicating feeding regimens for glycogen-storage disease. Lancet 1978; 2:1203–1204. 226. Greene HL, Slonim AE, Burr IM, et al. Type I glycogen storage disease: five years of management with nocturnal intragastric feeding. J Pediatr 1980; 96:590–595. 227. Michels VV, Beaudet AL, Potts VE, et al. Glycogen storage disease: long-term follow-up of nocturnal intragastric feeding. Clin Genet 1982; 21:136–140. 228. Chen YT, Coleman RA, Scheinman JI, et al. Renal disease in type I glycogen storage disease. N Engl J Med 1988; 318:7–11. 229. Baker L, Dahlem S, Goldfarb S, et al. Hyperfiltration and renal disease in glycogen storage disease, type I. Kidney Int 1989; 35:1345–1350. 230. Senior B, Loridan L. Studies of liver glycogenoses, with particular reference to the metabolism of intravenously administered glycerol. N Engl J Med 1968; 279:958–965. 231. Bialek DS, Sharp HL, Kane WJ, et al. Latency of glucose-6phosphatase in type IB glycogen storage disease (abstract). J Pediatr 1977; 91:938. 232. Narisawa K, Igarashi Y, Otomo H, et al. A new variant of glycogen storage disease type I probably due to a defect in the glucose-6-phosphate transport system. Biochem Biophys Res Commun 1978; 83:1360–1364. 233. Lange AJ, Arion WJ, Beaudet AL. Type Ib glycogen storage disease is caused by a defect in the glucose-6-phosphate translocase of the microsomal glucose-6-phosphatase system. J Biol Chem 1980; 255:8381–8384. 234. Anderson DC, Mace ML, Brinkley BR, et al. Recurrent infection in glycogenesis type Ib: abnormal neutrophil motility related to impaired redistribution of adhesion sites. J Infect Dis 1981; 143:447–459. 235. Beaudet AL, Anderson DC, Michels VV, et al. Neutropenia and impaired neutrophil migration in type IB glycogen storage disease. J Pediatr 1980; 97:906–910. 236. Heyne K, Gahr M. Phagocytotic extra-respiration: differences between cases of glycogenesis type Ia and Ib. Eur J Pediatr 1980; 133:186. 237. Seger R, Steinmann B, Tiefenauer L, et al. Short communication. Glycogenesis Ib: neutrophil microbicidal defects due to impaired hexose monophosphate shunt. Pediatr Res 1984; 18:297–299. 238. Arion WJ, Wallin BK, Lange AJ, et al. On the involvement of a glucose 6-phosphate transport system in the function of microsomal glucose 6-phosphatase. Mol Cell Biochem 1975; 6:75–83. 239. Skaug WA, Warford LL, Figueroa JM, et al. Glycogenesis type IB: possible membrane transport defect. South Med J 1981; 74:761–764. 240. Visser G, Rake JP, Fernandes J, et al. Neutropenia, neutrophil dysfunction, and inflammatory bowel disease in glycogen storage disease type Ib: results of the European Study on Glycogen Storage Disease type I. J Pediatr 2000; 137:187–191. 241. Zakim D, Edmondson DE. The role of the membrane in the regulation of activity of microsomal glucose-6-phosphatase. J Biol Chem 1982; 257:1145–1148.
1319
Section XII. Inherited and Pediatric Diseases of the Liver
242. Greene HL, Zakim D, Edmondson D. Measurement of hepatic steady state and pre-steady state kinetic of glucose-6phosphatase in a patient with type Ib glycogen storage disease. Pediatr Res 1986 (abstract). 243. Hiraiwa H, Pan CJ, Lin B, et al. Inactivation of the glucose 6phosphate transporter causes glycogen storage disease type 1b. J Biol Chem 1999; 274:5532–5536. 244. Annabi B, Hiraiwa H, Mansfield BC, et al. The gene for glycogen-storage disease type 1b maps to chromosome 11q23. Am J Hum Genet 1998; 62:400–405. 245. Chen LY, Pan CJ, Shieh JJ, et al. Structure–function analysis of the glucose-6-phosphate transporter deficient in glycogen storage disease type Ib. Hum Mol Genet 2002; 11: 3199–3207. 246. Elpeleg ON. The molecular background of glycogen metabolism disorders. J Pediatr Endocrinol Metab 1999; 12:363–379. 247. Ambruso DR, McCabe ER, Anderson D, et al. Infectious and bleeding complications in patients with glycogenesis Ib. Am J Dis Child 1985; 139:691–697. 248. Corbeel L, Hue L, Lederer B, et al. Clinical and biochemical findings before and after portacaval shunt in a girl with type Ib glycogen storage disease. Pediatr Res 1981; 15:58–61. 249. Ishiguro A, Nakahata T, Shimbo T, et al. Improvement of neutropenia and neutrophil dysfunction by granulocyte colonystimulating factor in a patient with glycogen storage disease type Ib. Eur J Pediatr 1993; 152:18–20. 250. Brown B, Brown DH. The glycogen storage diseases: types I, III, IV, V, VII, and unclassified glycogenoses. In: Whelan WJ, ed. Carbohydrate metabolism and its disorders, vol. 2. New York: Academic Press; 1968:123. 251. Bao Y, Dawson TL, Jr, Chen YT. Human glycogen debranching enzyme gene (AGL): complete structural organization and characterization of the 5¢ flanking region. Genomics 1996; 38:155–165. 252. Yang-Feng TL, Zheng K, Yu J, et al. Assignment of the human glycogen debrancher gene to chromosome 1p21. Genomics 1992; 13:931–934. 253. Yang BZ, Ding JH, Enghild JJ, et al. Molecular cloning and nucleotide sequence of cDNA encoding human muscle glycogen debranching enzyme. J Biol Chem 1992; 267:9294–9299. 254. Bao Y, Yang BZ, Dawson TL, Jr, et al. Isolation and nucleotide sequence of human liver glycogen debranching enzyme mRNA: identification of multiple tissue-specific isoforms. Gene 1997; 197:389–398. 255. Okubo M, Kanda F, Horinishi A, et al. Glycogen storage disease type IIIa: first report of a causative missense mutation (G1448R) of the glycogen debranching enzyme gene found in a homozygous patient. Hum Mutat 1999; 14:542–543. 256. Shen J, Bao Y, Liu HM, et al. Mutations in exon 3 of the glycogen debranching enzyme gene are associated with glycogen storage disease type III that is differentially expressed in liver and muscle. J Clin Invest 1996;98:352–357. 257. Shen JJ, Chen YT. Molecular characterization of glycogen storage disease type III. Curr Mol Med 2002; 2:167–175. 258. Brandt IK, DeLuca VA, Jr. Type 3 glycogenesis. A family with an unusual tissue distribution of the enzyme lesion. Am J Med 1966; 40:779–784. 259. Murase T, Ikeda H, Muro T, et al. Myopathy associated with type 3 glycogenesis. J Neurol Sci 1973; 20:287–295. 260. Van C, Huijing F. Glycogen storage disease. Biochemical and clinical data in sixteen cases. Am J Med 1965; 38:554–561. 261. Starzl TE, Putnam CW, Porter KA, et al. Portal diversion for the treatment of glycogen storage disease in humans. Ann Surg 1973; 178:525–539. 262. Alagille D, Odievre M. Inborn errors of metabolism. In: Alagille D, Odievre M, eds. Liver and biliary tract disease in children. New York: John Wiley; 1979:217.
1320
263. Ding JH, de Barsy T, Brown BI, et al. Immunoblot analyses of glycogen debranching enzyme in different subtypes of glycogen storage disease type III. J Pediatr 1990; 116:95–100. 264. Hug G, Krill CE, Jr, Perrin EV, et al. Cori’s disease (amylo-1,6glucosidase deficiency). Report of a case in a Negro child. N Engl J Med 1963; 268:113–120. 265. Hers HG. Amylo-1, 6-glucosidase activity in tissues of children with glycogen storage disease. Biochem J 1960; 76:69. 266. Van Hoof F, Hers HG. The subgroups of type 3 glycogenesis. Eur J Biochem 1967; 2:265–270. 267. Garancis JC, Panares RR, Good TA, et al. Type 3 glycogenesis. A biochemical and electron microscopic study. Lab Invest 1970; 22:468–477. 268. Slonim AE, Terry AB, Moran R, et al. Differing food consumption for nocturnal intragastric therapy in types I and III glycogen storage disease (abstract). Pediatr Res 1978; 12:512–894. 269. Borowitz SM, Greene HL. Cornstarch therapy in a patient with type III glycogen storage disease. J Pediatr Gastroenterol Nutr 1987; 6:631–634. 270. Anderson DH. Studies on glycogen disease with report of a case in which the glycogen was abnormal. In: Ajjar VA, ed. Carbohydrate metabolism. Baltimore: John Hopkins Press; 1952:28. 271. Illingworth B, Cori GT. Structure of glycogens and amylopectins. III. Normal and abnormal human glycogen. J Biol Chem 1952; 199:653–660. 272. Brown BI, Brown DH. Lack of an alpha-1,4-glucan: alpha-1,4glucan 6-glycosyl transferase in a case of type IV glycogenosis. Proc Natl Acad Sci USA 1966; 56:725–729. 273. Reed GB, Jr, Dixon JF, Neustein JB, et al. Type IV glycogenosis. Patient with absence of a branching enzyme alpha-1,4glucan:alpha-1,4-glucan 6-glycosyl transferase. Lab Invest 1968; 19:546–557. 274. Levin B, Burgess EA, Mortimer PE. Glycogen storage disease type IV, amylopectinosis. Arch Dis Child 1968; 43:548–555. 275. Motoi M, Sonobe H, Ogawa K, et al. Two autopsy cases of glycogen storage disease – cirrhotic type. Acta Pathol Jpn 1973; 23:211–223. 276. Brass K. [Histologic diagnosis of glucogenosis type IV (amylopectinosis) (author’s transl).] Z Kinderheilkd 1974; 117:187–203. 277. Ishihara T, Uchino F, Adachi H, et al. Type IV glycogenosis – a study of two cases. Acta Pathol Jpn 1975; 25:613–633. 278. Bannayan GA, Dean WJ, Howell RR. Type IV glycogen-storage disease. Light-microscopic, electron-microscopic, and enzymatic study. Am J Clin Pathol 1976; 66:702–709. 279. McMaster KR, Powers JM, Hennigar GR, Jr, et al. Nervous system involvement in type IV glycogenesis. Arch Pathol Lab Med 1979; 103:105–111. 280. Ferguson IT, Mahon M, Cumming WJ. An adult case of Andersen’s disease – type IV glycogenesis. A clinical, histochemical, ultrastructural and biochemical study. J Neurol Sci 1983; 60:337–351. 281. Holmes JM, Houghton CR, Woolf AL. A myopathy presenting in adult life with features suggestive of glycogen storage disease. J Neurol Neurosurg Psychiatry 1960; 23:302. 282. Torvik A, Dietrichson P, Svaar H, et al. Myopathy with tremor and dementia: a metabolic disorder? Case report with postmortem study. J Neurol Sci 1974; 21:181–190. 283. Pelissier JF, DeBarsy T, Bille J, et al. Polysaccharide (amylopectin-like) storage myopathy – histochemical, ultrastructural and biochemical studies. Acta Neuropathol 1970; 7 (Suppl):292. 284. Schochet SS, Jr, McCormick WF, Zellweger H. Type IV glycogenosis (amylopectinosis). Light and electron microscopic observations. Arch Pathol 1970; 90:354–363.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
285. Howell RR, Kaback MM, Brown BI. Type IV glycogen storage disease: branching enzyme deficiency in skin fibroblasts and possible heterozygote detection. J Pediatr 1971; 78:638–642. 286. Greene HL, Ghishan FK, Brown B, et al. Hypoglycemia in type IV glycogenosis: hepatic improvement in two patients with nutritional management. J Pediatr 1988; 112:55–58. 287. Greene HL, Brown BI, McClenathan DT, et al. A new variant of type IV glycogenesis: deficiency of branching enzyme activity without apparent progressive liver disease. Hepatology 1988; 8:302–306. 288. Hers HG. Glycogen storage disease. In: Levine R, Luft R, eds. Advances in metabolic disorders, vol. 1. New York: Academic Press; 1964:1. 289. Moses SW, Parvari R. The variable presentations of glycogen storage disease type IV: a review of clinical, enzymatic and molecular studies. Curr Mol Med 2002; 2:177–188. 290. Fernandes J, Huijing F. Branching enzyme-deficiency glycogenesis: studies in therapy. Arch Dis Child 1968; 43:347–352. 291. Starzl TE, Demetris AJ, Trucco M, et al. Chimerism after liver transplantation for type IV glycogen storage disease and type 1 Gaucher’s disease. N Engl J Med 1993; 328:745–749. 292. Lin EC, Knox WE. Specificity of the adaptive response to tyrosine-alpha-ketoglutarate transaminase in the rat. J Biol Chem 1958; 233:1186–1189. 293. La Du BN, Zannoni VG, Laster L, et al. The nature of the defect in tyrosine metabolism in alcaptonuria. J Biol Chem 1958; 230:251–260. 294. Shambaugh GE, 3rd, Warner DA, Beisel WR. Hormonal factors altering rhythmicity of tyrosine-alpha-ketoglutarate transaminase in rat liver. Endocrinology 1967; 81:811–818. 295. Gelehrter TD, Tomkins GM. Control of tyrosine aminotransferase synthesis in tissue culture by a factor in serum. Proc Natl Acad Sci USA 1969; 64:723–730. 296. Granner D, Chase LR, Aurbach GD, et al. Tyrosine aminotransferase: enzyme induction independent of adenosine 3¢, 5¢-monophosphate. Science 1968; 162:1018–1020. 297. Goswami MN, Knox WE. Developmental changes of phydroxyphenylpyruvate-oxidase activity in mammalian liver. Biochim Biophys Acta 1961; 50:35–40. 298. Fellman JH, Fujita TS, Roth ES. Assay, properties and tissue distribution of p-hydroxyphenylpyruvate hydroxylase. Biochim Biophys Acta 1972; 284:90–100. 299. Taniguchi K, Armstrong MD. The enzymatic formation of omicron-hydroxyphenylacetic acid. J Biol Chem 1963; 238:4091–4097. 300. Goswami MN, Knox WE. An evaluation of the role of ascorbic acid in the regulation of tyrosine metabolism. J Chron Dis 1963; 16:363–371. 301. Medes G, Berglund H, Lohmann A. An unknown reducing urinary substance in myasthenia gravis. Proc Soc Exp Biol Med 1927; 25:210. 302. Medes G. A new error of tyrosine metabolism: tyrosinosis. The intermediary metabolism of tyrosine and phenylalanine. Biochem J 1932; 26:917. 303. Mathews J, Partington MW. The plasma tyrosine levels of premature babies. Arch Dis Child 1964; 39:371–378. 304. Avery ME, Clow CL, Menkes JH, et al. Treatment of tyrosinemia of the newborn: dietary and clinical aspects. Pediatrics 1964; 39:371. 305. Levy HL, Shih VE, Madigan PM, et al. Transient tyrosinemia in full-term infants. JAMA 1969; 209:249–250. 306. Constantsas NS, Nicopoulos DA, Agathopoulos AS, et al. Properties and developmental changes of human hepatic aryl pyruvate enzymes. Biochim Biophys Acta 1969; 192:545–548. 307. Rizzardini M, Abeliuk P. Tyrosinemia and tyrosinuria in lowbirth-weight infants. A new criterion to assess maturity at birth. Am J Dis Child 1971; 121:182–185.
308. Bakker HD, Wadman SK, Van Sprang FJ, et al. Tyrosinemia and tyrosyluria in healthy prematures: time courses not vitamin Cdependent. Clin Chim Acta 1975; 61:73–90. 309. Wadman SK, van Sprang FJ, Maas JW, et al. An exceptional case of tyrosinosis. J Ment Defic Res 1968; 12:269–281. 310. Holston JL, Jr, Levy HL, Tomlin GA, et al. Tyrosinosis: a patient without liver or renal disease. Pediatrics 1971; 48:393–400. 311. Kennaway NG, Buist NRM. Metabolic studies in a patient with hepatic cytosol tyrosine aminotransferase deficiency. Pediatr Res 1971; 5:287. 312. Zaleski WA, Hill A. Tyrosinosis: a new variant. Can Med Assoc J 1973; 108:477. 313. Goldsmith LA, Kang E, Bienfang DC, et al. Tyrosinemia with plantar and palmar keratosis and keratitis. J Pediatr 1973; 83:798–805. 314. Louis WJ, Pitt DD, Davies H. Biochemical studies in a patient with “tyrosinosis.” Aust NZ J Med 1974; 4:281–286. 315. Goldsmith LA, Thorpe J, Roe CR. Hepatic enzymes of tyrosine metabolism in tyrosinemia II. J Invest Dermatol 1979; 73:530–532. 316. Zaleski WA, Hill A, Kushniruk W. Skin lesions in tyrosinosis: response to dietary treatment. Br J Dermatol 1973; 88:335–340. 317. Levine RJ, Conn HO. Tyrosine metabolism in patients with liver disease. J Clin Invest 1967; 46:2012–2020. 318. Nordlinger BM, Fulenwider JT, Ivey GL, et al. Tyrosine metabolism in cirrhosis. J Lab Clin Med 1979; 94:832–840. 319. van der Heiden C, Wauters EA, Duran M, et al. Gas chromatographic analysis of urinary tyrosine and phenylalanine metabolites in patients with gastrointestinal disorders. Clin Chim Acta 1971; 34:289–296. 320. Newberry PD, Rohrer JW, Cherian G, et al. Degree and duration of hypoxia needed to produce tyrosyluria. Lancet 1972; 1:750–751. 321. Kalliomaki JL, Lehtonen A, Seppala P. Oral tyrosine tolerance test in rheumatoid arthritis. Ann Rheum Dis 1966; 25:469–471. 322. Baber MD. A case of congenital cirrhosis of the liver with renal tubular defects akin to those in the Fanconi syndrome. Arch Dis Child 1956; 31:335–339. 323. Sakai K, Kitagawa T. An atypical case of tyrosinosis. I. Clinical and laboratory findings. Jikeikai Med J 1957; 4:1. 324. Sakai K, Kitagawa T. An atypical case of tyrosinosis. II. A research on the metabolic block. Jikeikai Med J 1957; 4:11. 325. Sakai K, Kitagawa T. An atypical case of tyrosinosis. III. The outcome of the patient: pathological and biochemical observations on the organ tissues. Jikeikai Med J 1959; 6:15. 326. Halvorsen S, Gjessing LR. Studies on tyrosinosis: 1, Effect of low-tyrosine and low-phenylalanine diet. Br Med J 1964; 5418:1171–1173. 327. Larochelle J, Mortezai A, Belanger M, et al. Experience with 37 infants with tyrosinemia. Can Med Assoc J 1967; 97:1051–1054. 328. Scriver CR, Larochelle J, Silverberg M. Hereditary tyrosinemia and tyrosyluria in a French Canadian geographic isolate. Am J Dis Child 1967; 113:41–46. 329. Gentz J, Jagenburg R, Zetterstroem R. Tyrosinemia: an inborn error of tyrosine metabolism with cirrhosis of the liver and multiple renal tubular defects (de Toni–Debre–Fanconi syndrome). J Pediatr 1965; 66:670–696. 330. Kogut MD, Shaw KN, Donnell GN. Tyrosinosis. Am J Dis Child 1967; 113:47–53. 331. Kang ES, Gerald PS. Hereditary tyrosinemia and abnormal pyrrole metabolism. J Pediatr 1970; 77:397–406. 332. Halvorsen S, Pande H, Loken AC, et al. Tyrosinosis. A study of 6 cases. Arch Dis Child 1966; 41:238–249. 333. Weinberg AG, Mize CE, Worthen HG. The occurrence of hepatoma in the chronic form of hereditary tyrosinemia. J Pediatr 1976; 88:434–438.
1321
Section XII. Inherited and Pediatric Diseases of the Liver
334. Sass-Kortsak A, Ficici S, Paunier L, et al. Secondary metabolic derangements in patients with tyrosyluria. Can Med Assoc J 1967; 97:1079–1083. 335. Scriver CR, Silverberg M, Clow CL. Hereditary tyrosinemia and tyrosyluria: clinical report of four patients. Can Med Assoc J 1967; 97:1047–1050. 336. La Du BN, Gjessing LR. Tyrosinosis and tyrosinemia. In: Stanbury JB, Wyngaarden JB, Fredrickson DS, eds. The metabolic basis of inherited disease, 4th edn. New York: McGraw-Hill; 1978:256. 337. Coward RF, Smith P. Urinary excretion of 4-hydroxyphenyllactic acids and related compounds in man, including a screening test for tyrosyluria. Biochem Med 1968; 2:216. 338. Gentz J, Johansson S, Lindblad B, et al. Exertion of deltaaminolevulinic acid in hereditary tyrosinemia. Clin Chim Acta 1969; 23:257–263. 339. Lindblad B, Lindstedt S, Steen G. On the enzymic defects in hereditary tyrosinemia. Proc Natl Acad Sci USA 1977; 74:4641–4645. 340. Melancon SB, Gagne R, Grenier A, et al. Deficiency of fumarylacetoacetase in the acute form of hereditary tyrosinemia with reference to prenatal diagnosis. In: Fisher MM, Roy CC, eds. Pediatric liver disease. New York: Plenum Press; 1983:223. 341. Berger R, van Faassen H, Smith GP. Biochemical studies on the enzymatic deficiencies in hereditary tyrosinemia. Clin Chim Acta 1983; 134:129–141. 342. Taniguchi K, Gjessing LR. Studies on tyrosinosis: 2, Activity of the transaminase, parahydroxy-phenyl-pyruvate oxidase, and homogentisic-acid oxidase. Br Med J 1965; 5440:968–969. 343. Gentz J, Lindblad B. p-hydroxyphenylpyruvate hydroxylase activity in fine-needle aspiration liver biopsies in hereditary tyrosinemia. Scand J Clin Lab Invest 1972; 29:115–126. 344. Perry TL, Hardwick DF, Hansen S, et al. Methionine induction of experimental tyrosinaemia. J Ment Defic Res 1967; 11:246–253. 345. Gaull GE, Rassin DK, Solomon GE, et al. Biochemical observations on so-called hereditary tyrosinemia. Pediatr Res 1970; 4:337–344. 346. Gaull GE, Rassin DK, Sturman JA. Significance of hypermethionaemia in acute tyrosinosis. Lancet 1968; 1:1318–1319. 347. Daniel RG, Waisman HA. The influence of excess methionine on the free amino acids of brain and liver of the weanling rat. J Neurochem 1969; 16:787–795. 348. Lindemann R, Gjessing LR, Merton B, et al. Fructosaemia – “acute-tyrosinosis.” Lancet 1969; 1:891. 349. Grant DB, Alexander FW, Seakins JW. Abnormal tyrosine metabolism in hereditary fructose intolerance. Acta Paediatr Scand 1970; 59:432–434. 350. Bakker HD, de Bree PK, van Sprang FJ, et al. Fructose-1,6diphosphatase deficiency: another enzyme defect which can present itself with the clinical features of “tyrosinosis.” Clin Chim Acta 1974; 55:41–47. 351. Berger R, Smit GP, Stoker-de Vries SA, et al. Deficiency of fumarylacetoacetase in a patient with hereditary tyrosinemia. Clin Chim Acta 1981; 114:37–44. 352. Kvittingen EA, Jellum E, Stokke O. Assay of fumarylacetoacetate fumarylhydrolase in human liver-deficient activity in a case of hereditary tyrosinemia. Clin Chim Acta 1981; 115:311–319. 353. Furukawa N, Kinugasa A, Seo T, et al. Enzyme defect in a case of tyrosinemia type I, acute form. Pediatr Res 1984; 18:463–466. 354. Kvittingen EA, Brodtkorb E. The pre- and post-natal diagnosis of tyrosinemia type I and the detection of the carrier state by assay of fumarylacetoacetase. Scand J Clin Lab Invest Suppl 1986; 184:35–40.
1322
355. Mitchell GA, Lambert M, Tanguay RM. Hypertyrosinemia. In: Scriver CR, Beaudet AL, Sly WS, et al., eds. The metabolic and molecular bases of inherited disease. New York: McGraw-Hill; 1995:1077–1106. 356. Phaneuf D, Labelle Y, Berube D, et al. Cloning and expression of the cDNA encoding human fumarylacetoacetate hydrolase, the enzyme deficient in hereditary tyrosinemia: assignment of the gene to chromosome 15. Am J Hum Genet 1991; 48:525–535. 357. Grompe M, St-Louis M, Demers SI, et al. A single mutation of the fumarylacetoacetate hydrolase gene in French Canadians with hereditary tyrosinemia type I. N Engl J Med 1994; 331:353–357. 358. Poudrier J, St-Louis M, Lettre F, et al. Frequency of the IVS12 + 5GÆA splice mutation of the fumarylacetoacetate hydrolase gene in carriers of hereditary tyrosinaemia in the French Canadian population of Saguenay-Lac-St-Jean. Prenat Diagn 1996; 16:59–64. 359. St-Louis M, Leclerc B, Laine J, et al. Identification of a stop mutation in five Finnish patients suffering from hereditary tyrosinemia type I. Hum Mol Genet 1994; 3:69–72. 360. Rootwelt H, Hoie K, Berger R, et al. Fumarylacetoacetase mutations in tyrosinaemia type I. Hum Mutat 1996; 7:239–243. 361. Rootwelt H, Berger R, Gray G, et al. Novel splice, missense, and nonsense mutations in the fumarylacetoacetase gene causing tyrosinemia type 1. Am J Hum Genet 1994; 55:653–658. 362. Prive L. Pathological findings in patients with tyrosinemia. Can Med Assoc J 1967; 97:1054–1056. 363. Partington MW, Haust MD. A patient with tyrosinemia and hypermethioninemia. Can Med Assoc J 1967; 97:1059–1067. 364. Perry TL. Tyrosinemia associated with hypermethioninemia and islet cell hyperplasia. Can Med Assoc J 1967; 97:1067–1075. 365. Bergeron P, Laberge C, Grenier A. Hereditary tyrosinemia in the province of Quebec: prevalence at birth and geographic distribution. Clin Genet 1974; 5:157–162. 366. Grenier A, Laberge C. A modified automated fluorometric method for tyrosine determination in blood spotted on paper: a mass screening procedure for tyrosinemia. Clin Chim Acta 1974; 57:71–75. 367. Grenier A, Belanger L, Laberge C. alpha1-Fetoprotein measurement in blood spotted on paper: discriminating test for hereditary tyrosinemia in neonatal mass screening. Clin Chem 1976; 22:1001–1004. 368. Kvittingen EA, Steinmann B, Gitzelmann R, et al. Prenatal diagnosis of hereditary tyrosinemia by determination of fumarylacetoacetase in cultured amniotic fluid cells. Pediatr Res 1985; 19:334–337. 369. Gentz J, Lindblad B, Lindstedt S, et al. Dietary treatment in tyrosinemia (tyrosinosis). With a note on the possible recognition of the carrier state. Am J Dis Child 1967l; 113:31–37. 370. Halvorsen S: Dietary treatment of tyrosinosis. Am J Dis Child 1967; 113:38–40. 371. Sass-Kortsak A, Ficici S, Paunier L, et al. Observations on treatment in patients with tyrosyluria. Can Med Assoc J 1967; 97:1089–1095. 372. Tada K, Wada Y, Yazaki N, et al. Dietary treatment of infantile tyrosinemia. Tohoku J Exp Med 1968; 95:337–344. 373. Bodegard G, Gentz J, Lindblad B, et al. Hereditary tyrosinemia. 3. On the differential diagnosis and the lack of effect of early dietary treatment. Acta Paediatr Scand 1969; 58:37–48. 374. Michals K, Matolon R, Wong PW. Dietary treatment of tyrosinemia type I: importance of methionine restriction. J Am Diet Assoc 1978; 73:507–514. 375. Mitchell GA, Russo PA, Dubois J, et al. Tyrosinemia. In: Suchy FJ, Sokol RJ, Balistreri WF, eds. Liver disease in children, 2nd edn. Philadelphia: Lippincott, Williams & Williams; 2001:667–685.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
376. Grompe M, Lindstedt S, al-Dhalimy M, et al. Pharmacological correction of neonatal lethal hepatic dysfunction in a murine model of hereditary tyrosinaemia type I. Nat Genet 1995; 10:453–460. 377. Starzl TE, Zitelli BJ, Shaw BW, Jr, et al. Changing concepts: liver replacement for hereditary tyrosinemia and hepatoma. J Pediatr 1985; 106:604–606. 378. Mieles LA, Esquivel CO, Van Thiel DH, et al. Liver transplantation for tyrosinemia. A review of 10 cases from the University of Pittsburgh. Dig Dis Sci 1990; 35:153–157. 379. Goldstein JL, Dana SE, Faust JR, et al. Role of lysosomal acid lipase in the metabolism of plasma low density lipoprotein. Observations in cultured fibroblasts from a patient with cholesteryl ester storage disease. J Biol Chem 1975; 250:8487–8495. 380. Sloan HR, Fredrickson DS. Enzyme deficiency in cholesteryl ester storage disease. J Clin Invest 1972; 51:1923–1926. 381. Schiff L, Schubert WK, McAdams AJ, et al. Hepatic cholesterol ester storage disease, a familial disorder. I. Clinical aspects. Am J Med 1968; 44:538–546. 382. Assmann G, Fredrickson DS, Sloan HR, et al. Accumulation of oxygenated steryl esters in Wolman’s disease. J Lipid Res 1975; 16:28–38. 383. Du H, Witte DP, Grabowski GA: Tissue and cellular specific expression of murine lysosomal acid lipase mRNA and protein. J Lipid Res 1996; 37:937–949. 384. Lohse P, Maas S, Lohse P, et al. Compound heterozygosity for a Wolman mutation is frequent among patients with cholesteryl ester storage disease. J Lipid Res 2000; 41:23–31. 385. Anderson RA, Bryson GM, Parks JS. Lysosomal acid lipase mutations that determine phenotype in Wolman and cholesterol ester storage disease. Mol Genet Metab 1999; 68:333–345. 386. Du H, Duanmu M, Witte D, et al. Targeted disruption of the mouse lysosomal acid lipase gene: long-term survival with massive cholesteryl ester and triglyceride storage. Hum Mol Genet 1998; 7:1347–1354. 387. Abramov A, Schorr S, Wolman M. Generalized xanthomatosis with calcified adrenals. AMA J Dis Child 1956; 91:282–286. 388. Wolman M, Sterk VV, Gatt S, et al. Primary familial xanthomatosis with involvement and calcification of the adrenals. Report of two more cases in siblings of a previously described infant. Pediatrics 1961; 28:742–757. 389. Crocker AC, Vawter GF, Neuhauser EB, et al. Wolman’s disease: three new patients with a recently described lipidosis. Pediatrics 1965; 35:627–640. 390. Patrick AD, Lake BD. Deficiency of an acid lipase in Wolman’s disease. Nature 1969; 222:1067–1068. 391. Kyriakides EC, Paul B, Balint JA. Lipid accumulation and acid lipase deficiency in fibroblasts from a family with Wolman’s disease, and their apparent correction in vitro. J Lab Clin Med 1972; 80:810–816. 392. Konno T, Fujii M, Watanuki T, et al. Wolman’s disease: the first case in Japan. Tohoku J Exp Med 1966; 90:375–389. 393. Eto Y, Kitagawa T. Wolman’s disease with hypolipoproteinemia and acanthocytosis: clinical and biochemical observations. J Pediatr 1970; 77:862–867. 394. Marshall WC, Ockenden BG, Fosbrooke AS, et al. Wolman’s disease. A rare lipidosis with adrenal calcification. Arch Dis Child 1969; 44:331–341. 395. Lough J, Fawcett J, Wiegensberg B. Wolman’s disease. An electron microscopic, histochemical, and biochemical study. Arch Pathol 1970; 89:103–110. 396. Sando GN, Rosenbaum LM. Human lysosomal acid lipase/cholesteryl ester hydrolase. Purification and properties of the form secreted by fibroblasts in microcarrier culture. J Biol Chem 1985; 260:15186–15193. 397. Anderson RA, Sando GN. Cloning and expression of cDNA encoding human lysosomal acid lipase/cholesteryl ester
398. 399. 400.
401. 402.
403.
404.
405. 406. 407.
408. 409.
410. 411.
412.
413.
414.
415.
416.
417. 418.
419.
420.
hydrolase. Similarities to gastric and lingual lipases. J Biol Chem 1991; 266:22479–22484. Young EP, Patrick AD. Deficiency of acid esterase activity in Wolman’s disease. Arch Dis Child 1970; 45:664–668. Boyd GS. Effect of linoleate and estrogen on cholesterol metabolism. Fed Proc 1962; 21:86–92. Coates PM, Cortner J. Acid lipase in cultured amniotic fluid cells. Implications for the prenatal diagnosis of Wolman’s disease. Pediatr Res 1978; 12:450. Meyers MA. Diseases of the adrenal glands in radiologic diagnosis. Springfield, IL: Charles C Thomas, 1963. Beaudet AL, Lipson MH, Ferry GD, et al. Acid lipase in cultured fibroblasts: cholesterol ester storage disease. J Lab Clin Med 1974; 84:54–61. Meyers WF, Hoeg JM, Demosky SJ, et al. The use of parenteral nutrition and elemental formula feeding in the treatment of Wolman’s disease. Nutr Res 1985; 5:423. Krivit W, Freese D, Chan KW, et al. Wolman’s disease: a review of treatment with bone marrow transplantation and considerations for the future. Bone Marrow Transplant 1992; 10 (Suppl 1):97–101. Wolman M. Proposed treatment for infants with Wolman disease. Pediatrics 1989; 83:1074–1075. Fredrickson DS. Newly recognized disorders of cholesterol metabolism. Ann Intern Med 1963; 58:718. Lageron A, Caroli J, Stralin H, et al. [Cholesterolic polycoria in adults. I. Clinical and histochemical study.] Presse Med 1967; 75:2785. Infante R, Polonovski J, Caroli J. [Cholesterolic polycoria in adults. II. Biochemical study.] Presse Med 1967; 75:2829–2832. Partin JC, Schubert WK. Small intestinal mucosa in cholesterol ester storage disease. A light and electron microscope study. Gastroenterology 1969; 57:542–558. Burke JA, Schubert WK. Deficient activity of hepatic acid lipase in cholesterol ester storage disease. Science 1972; 176:309–310. Beaudet AL, Ferry GD, Nichols BL, Jr, et al. Cholesterol ester storage disease: clinical, biochemical, and pathological studies. J Pediatr 1977; 90:910–914. Wolf H, Hug G, Michaelis R, et al. [Unusual congenital cholesterol ester storage in the liver.] Helv Paediatr Acta 1974; 29:105–118. Lageron A, Lichtenstein H, Bodin F, et al. [Cholesterol polycoria in adults. Clinical and histochemical aspects.] Med Chir Dig 1975; 4:9–14. Fredrickson DS, Sloan HR, Ferrans VJ, et al. Cholesteryl ester storage disease: a most unusual manifestation of deficiency of two lysosomal enzyme activities. Trans Assoc Am Phys 1972; 85:109–119. Burton BK, Remy WT, Rayman L. Cholesterol ester and triglyceride metabolism in intact fibroblasts from patients with Wolman’s disease and cholesterol ester storage disease. Pediatr Res 1984; 18:1242–1245. Pagani F, Pariyarath R, Garcia R, et al. New lysosomal acid lipase gene mutants explain the phenotype of Wolman disease and cholesteryl ester storage disease. J Lipid Res 1998; 39:1382–1388. Lageron A, Lichtenstein H, Bodin F, et al. [Letter: Cholesterol polycoria in the adult. Apropos of a new case.] Nouv Presse Med 1974; 3:1233. Jain L, Vidyasagar D. Use of simvastatin plus cholestyramine in the treatment of lysosomal acid lipase deficiency. J Pediatr 1991; 119:1008. Tarantino MD, McNamara DJ, Granstrom P, et al. Lovastatin therapy for cholesterol ester storage disease in two sisters. J Pediatr 1991; 118:131–135. Clayton RJ, Iber FL, Ruebner BH, et al. Byler’s disease: fatal familial intrahepatic cholestasis in an Amish kindred (abstract). J Pediatr 1965; 67:1025.
1323
Section XII. Inherited and Pediatric Diseases of the Liver
421. Hirooka M, Ono T. A case of familial intrahepatic cholestasis. Tohoku J Exp Med 1968; 94:293–306. 422. Clayton RJ, Iber FL, Ruebner BH, et al. Byler disease. Fatal familial intrahepatic cholestasis in an Amish kindred. Am J Dis Child 1969; 117:112–124. 423. Alagille D, Odievre M. Liver and biliary tract disease in children. New York: John Wiley; 1978. 424. Carlton VE, Knisely AS, Freimer NB. Mapping of a locus for progressive familial intrahepatic cholestasis (Byler disease) to 18q21-q22, the benign recurrent intrahepatic cholestasis region. Hum Mol Genet 1995; 4:1049–1053. 425. Bull LN, van Eijk MJ, Pawlikowska L, et al. A gene encoding a P-type ATPase mutated in two forms of hereditary cholestasis. Nat Genet 1998; 18:219–224. 426. Gray OP, Saunders RA. Familial intrahepatic cholestatic jaundice in infancy. Arch Dis Child 1966; 41:320–328. 427. Odievre M, Gautier M, Hadchouel M, et al. Severe familial intrahepatic cholestasis. Arch Dis Child 1973; 48:806–812. 428. Lykavieris P, van Mil S, Cresteil D, et al. Progressive familial intrahepatic cholestasis type 1 and extrahepatic features: no catch-up of stature growth, exacerbation of diarrhea, and appearance of liver steatosis after liver transplantation. J Hepatol 2003; 39:447–452. 429. Egawa H, Yorifuji T, Sumazaki R, et al. Intractable diarrhea after liver transplantation for Byler’s disease: successful treatment with bile adsorptive resin. Liver Transpl 2002; 8:714–716. 430. Whitington PF, Freese DK, Alonso EM, et al. Clinical and biochemical findings in progressive familial intrahepatic cholestasis. J Pediatr Gastroenterol Nutr 1994; 18:134–141. 431. Ballow M, Margolis CZ, Schachtel B, et al. Progressive familial intrahepatic cholestasis. Pediatrics 1973; 51:998–1007. 432. Bull L, Freimer N, Czubkowski P, et al. Clinical and biochemical features of FIC1 (ATP8B1) and BSEP (ABCB11) disease. Hepatology 2002; 36:310a. 433. Tygstrup N, Steig BA, Juijn JA, et al. Recurrent familial intrahepatic cholestasis in the Faeroe Islands. Phenotypic heterogeneity but genetic homogeneity. Hepatology 1999; 29:506–508. 434. Dahms BB. Hepatoma in familial cholestatic cirrhosis of childhood: its occurrence in twin brothers. Arch Pathol Lab Med 1979; 103:30–33. 435. Maggiore G, Bernard O, Riely CA, et al. Normal serum gammaglutamyl-transpeptidase activity identifies groups of infants with idiopathic cholestasis with poor prognosis. J Pediatr 1987; 111:251–252. 436. Chobert MN, Bernard O, Bulle F, et al. High hepatic gammaglutamyltransferase (gamma-GT) activity with normal serum gamma-GT in children with progressive idiopathic cholestasis. J Hepatol 1989; 8:22–25. 437. Maggiore G, Bernard O, Hadchouel M, et al. Diagnostic value of serum gamma-glutamyl transpeptidase activity in liver diseases in children. J Pediatr Gastroenterol Nutr 1991; 12:21–26. 438. Watkins JB, Perman JA. Bile acid metabolism in infants and children. Clin Gastroenterol 1977; 6:201–218. 439. Linarelli LG, Williams CN, Phillips MJ. Byler’s disease: fatal intrahepatic cholestasis. J Pediatr 1972; 81:484–492. 440. Klomp LW, Bull LN, Knisely AS, et al. A missense mutation in FIC1 is associated with Greenland familial cholestasis. Hepatology 2000; 32:1337–1341. 441. Eppens EF, van Mil SW, de Vree JM, et al. FIC1, the protein affected in two forms of hereditary cholestasis, is localized in the cholangiocyte and the canalicular membrane of the hepatocyte. J Hepatol 2001; 35:436–443. 442. Klomp LW, Vargas JC, van Mil SW, et al. Characterization of mutations in ATP8B1 associated with hereditary cholestasis. Hepatology 2004; 40:27–38.
1324
443. Alonso EM, Snover DC, Montag A, et al. Histologic pathology of the liver in progressive familial intrahepatic cholestasis. J Pediatr Gastroenterol Nutr 1994; 18:128–133. 444. Jacquemin E, Hermans D, Myara A, et al. Ursodeoxycholic acid therapy in pediatric patients with progressive familial intrahepatic cholestasis. Hepatology 1997; 25:519–523. 445. Whitington PF, Whitington GL. Partial external diversion of bile for the treatment of intractable pruritus associated with intrahepatic cholestasis. Gastroenterology 1988; 95:130–136. 446. Soubrane O, Gauthier F, DeVictor D, et al. Orthotopic liver transplantation for Byler disease. Transplantation 1990; 50:804–806. 447. Arnell H, Nemeth A, Anneren G, et al. Progressive familial intrahepatic cholestasis (PFIC): evidence for genetic heterogeneity by exclusion of linkage to chromosome 18q21q22. Hum Genet 1997; 100:378–381. 448. Bourke B, Goggin N, Walsh D, et al. Byler-like familial cholestasis in an extended kindred. Arch Dis Child 1996; 75:223–227. 449. Bull LN, Carlton VE, Stricker NL, et al. Genetic and morphological findings in progressive familial intrahepatic cholestasis (Byler disease [PFIC-1] and Byler syndrome): evidence for heterogeneity. Hepatology 1997; 26:155–164. 450. Strautnieks SS, Kagalwalla AF, Tanner MS, et al. Identification of a locus for progressive familial intrahepatic cholestasis PFIC2 on chromosome 2q24. Am J Hum Genet 1997; 61:630–633. 451. Strautnieks SS, Bull LN, Knisely AS, et al. A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis. Nat Genet 1998; 20:233–238. 452. Jansen PL, Strautnieks SS, Jacquemin E, et al. Hepatocanalicular bile salt export pump deficiency in patients with progressive familial intrahepatic cholestasis. Gastroenterology 1999; 117:1370–1379. 453. van Mil SW, van der Woerd WL, van der Brugge G, et al. Benign recurrent intrahepatic cholestasis type 2 is caused by mutations in ABCB11. Gastroenterology 2004; 127:379–384. 454. de Vree JM, Jacquemin E, Sturm E, et al. Mutations in the MDR3 gene cause progressive familial intrahepatic cholestasis. Proc Natl Acad Sci USA 1998; 95:282–287. 455. Jacquemin E, De Vree JM, Cresteil D, et al. The wide spectrum of multidrug resistance 3 deficiency: from neonatal cholestasis to cirrhosis of adulthood. Gastroenterology 2001; 120:1448–1458. 456. Smit JJ, Schinkel AH, Oude Elferink RP, et al. Homozygous disruption of the murine mdr2 P-glycoprotein gene leads to a complete absence of phospholipid from bile and to liver disease. Cell 1993; 75:451–462. 457. Jacquemin E, Cresteil D, Manouvrier S, et al. Heterozygous non-sense mutation of the MDR3 gene in familial intrahepatic cholestasis of pregnancy. Lancet 1999; 353:210–211. 458. Dixon PH, Weerasekera N, Linton KJ, et al. Heterozygous MDR3 missense mutation associated with intrahepatic cholestasis of pregnancy: evidence for a defect in protein trafficking. Hum Mol Genet 2000; 9:1209–1217. 459. Aagenaes O, van der Hagen CB, Refsum S. Hereditary recurrent intrahepatic cholestasis from birth. Arch Dis Child 1968; 43:646–657. 460. Sharp HL, Krivit W. Hereditary lymphedema and obstructive jaundice. J Pediatr 1971; 78:491–496. 461. Aagenaes O. Hereditary recurrent cholestasis with lymphoedema – two new families. Acta Paediatr Scand 1974; 63:465–471. 462. Aagenaes O, Cudeman B, Sigstad H, et al. Clinical and experimental relationships between cholestasis and abnormal hepatic lymphatics. Pediatr Res 1970; 4:377. 463. Bloom W. The role of the lymphatics in the absorption of bile pigment from the liver in early obstructive jaundice. Bull Johns Hopkins Hosp 1923; 34:316. 464. Shafiroff BG, Doubilet H, Ruggiero W. Bilirubin resorption in obstructive jaundice. Proc Soc Exp Biol Med 1939; 42:203.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
465. Cain JC, Grindlay JH, Bollman JL, et al. Lymph from liver and thoracic duct. Surg Gynecol Obstet 1947; 85:559. 466. Friedman M, Byers SO, Omoto C. Some characteristics of hepatic lymph in the intact rat. Am J Physiol 1956; 184:11–17. 467. Bull LN, Roche E, Song EJ, et al. Mapping of the locus for cholestasis-lymphedema syndrome (Aagenaes syndrome) to a 6.6-cM interval on chromosome 15q. Am J Hum Genet 2000; 67:994–999. 468. Ahrens EH, Jr, Harris RC, Mac MH. Atresia of the intrahepatic dile ducts. Pediatrics 1951; 8:628–647. 469. Mac MH, Thannhauser SJ. Congenital dysplasia of the interlobular bile ducts with extensive skin xanthomata; congenital acholangic biliary cirrhosis. Gastroenterology 1952; 21:488–506. 470. Haas L, Dobbs RH. Congenital absence of the intrahepatic bile ducts. Arch Dis Child 1958; 33:396–402. 471. Watson GH, Miller V. Arteriohepatic dysplasia: familial pulmonary arterial stenosis with neonatal liver disease. Arch Dis Child 1973; 48:459–466. 472. Alagille D, Odievre M, Gautier M, et al. Hepatic ductular hypoplasia associated with characteristic facies, vertebral malformations, retarded physical, mental, and sexual development, and cardiac murmur. J Pediatr 1975; 86:63–71. 473. Berrocal T, Gamo E, Navalon J, et al. Syndrome of Alagille: radiological and sonographic findings. A review of 37 cases. Eur Radiol 1997; 7:115–118. 474. Hoffenberg EJ, Narkewicz MR, Sondheimer JM, et al. Outcome of syndromic paucity of interlobular bile ducts (Alagille syndrome) with onset of cholestasis in infancy. J Pediatr 1995; 127:220–224. 475. Emerick KM, Rand EB, Goldmuntz E, et al. Features of Alagille syndrome in 92 patients: frequency and relation to prognosis. Hepatology 1999; 29:822–829. 476. Quiros-Tejeira RE, Ament ME, Heyman MB, et al. Variable morbidity in Alagille syndrome: a review of 43 cases. J Pediatr Gastroenterol Nutr 1999; 29:431–437. 477. Oda T, Elkahloun AG, Pike BL, et al. Mutations in the human Jagged1 gene are responsible for Alagille syndrome. Nat Genet 1997; 16:235–242. 478. Li L, Krantz ID, Deng Y, et al. Alagille syndrome is caused by mutations in human Jagged1, which encodes a ligand for Notch1. Nat Genet 1997; 16:243–251. 479. Riely CA, LaBrecque DR, Ghent C, et al. A father and son with cholestasis and peripheral pulmonic stenosis: a distinct form of intrahepatic cholestasis. J Pediatr 1978; 92:406–411. 480. Henriksen NT, Langmark F, Sorland SJ, et al. Hereditary cholestasis combined with peripheral pulmonary stenosis and other anomalies. Acta Paediatr Scand 1977; 66:7–15. 481. Greenwood RD, Rosenthal A, Crocker AC, et al. Syndrome of intrahepatic biliary dysgenesis and cardiovascular malformations. Pediatrics 1976; 58:243–247. 482. Nischal KK, Hingorani M, Bentley CR, et al. Ocular ultrasound in Alagille syndrome: a new sign. Ophthalmology 1997; 104:79–85. 483. Hingorani M, Nischal KK, Davies A, et al. Ocular abnormalities in Alagille syndrome. Ophthalmology 1999; 106:330–337. 484. Brodsky MC, Cunniff C. Ocular anomalies in the Alagille syndrome (arteriohepatic dysplasia). Ophthalmology 1993; 100:1767–1774. 485. Romanchuk KG, Judisch GF, LaBrecque DR. Ocular findings in arteriohepatic dysplasia (Alagille’s syndrome). Can J Ophthalmol 1981; 16:94–99. 486. Wells KK, Pulido JS, Judisch GF, et al. Ophthalmic features of Alagille syndrome (arteriohepatic dysplasia). J Pediatr Ophthalmol Strabismus 1993; 30:130–135. 487. Kamath BM, Spinner NB, Emerick KM, et al. Vascular anomalies in Alagille syndrome: a significant cause of morbidity and mortality. Circulation 2004; 109:1354–1358.
488. Rachmel A, Zeharia A, Neuman-Levin M, et al. Alagille syndrome associated with moyamoya disease. Am J Med Genet 1989; 33:89–91. 489. Woolfenden AR, Albers GW, Steinberg GK, et al. Moyamoya syndrome in children with Alagille syndrome: additional evidence of a vasculopathy. Pediatrics 1999; 103:505–508. 490. Joutel A, Corpechot C, Ducros A, et al. Notch3 mutations in CADASIL, a hereditary adult-onset condition causing stroke and dementia. Nature 1996; 383:707–710. 491. Chong SK, Lindridge J, Moniz C, et al. Exocrine pancreatic insufficiency in syndromic paucity of interlobular bile ducts. J Pediatr Gastroenterol Nutr 1989; 9:445–449. 492. John HA, Loomes K, Weyler R, et al. A study of pancreatic function in 17 children with Alagille syndrome. Gastroenterology 1998; 114:A885. 493. Lykavieris P, Crosnier C, Trichet C, et al. Bleeding tendency in children with Alagille syndrome. Pediatrics 2003; 111:167–170. 494. Ghishan FK, LaBrecque DR, Mitros FA, et al. The evolving nature of “infantile obstructive cholangiopathy.” J Pediatr 1980; 97:27–32. 495. Byrne JL, Harrod MJ, Friedman JM, et al. del(20p) with manifestations of arteriohepatic dysplasia. Am J Med Genet 1986; 24:673–678. 496. Spinner NB, Rand EB, Fortina P, et al. Cytologically balanced t(2;20) in a two-generation family with Alagille syndrome: cytogenetic and molecular studies. Am J Hum Genet 1994; 55:238–243. 497. Teebi AS, Murthy DS, Ismail EA, et al. Alagille syndrome with de novo del(20) (p11.2). Am J Med Genet 1992; 42:35–38. 498. Anad F, Burn J, Matthews D, et al. Alagille syndrome and deletion of 20p. J Med Genet 1990; 27:729–737. 499. Deleuze JF, Hazan J, Dhorne S, et al. Mapping of microsatellite markers in the Alagille region and screening of microdeletions by genotyping 23 patients. Eur J Hum Genet 1994; 2:185–190. 500. Desmaze C, Deleuze JF, Dutrillaux AM, et al. Screening of microdeletions of chromosome 20 in patients with Alagille syndrome. J Med Genet 1992; 29:233–235. 501. Oda T, Elkahloun AG, Meltzer PS, et al. Identification and cloning of the human homolog (JAG1) of the rat Jagged1 gene from the Alagille syndrome critical region at 20p12. Genomics 1997; 43:376–379. 502. Pollet N, Boccaccio C, Dhorne-Pollet S, et al. Construction of an integrated physical and gene map of human chromosome 20p12 providing candidate genes for Alagille syndrome. Genomics 1997; 42:489–498. 503. Spinner NB, Colliton RP, Crosnier C, et al. Jagged1 mutations in Alagille syndrome. Hum Mutat 2001; 17:18–33. 504. Louis AA, Van Eyken P, Haber BA, et al. Hepatic jagged1 expression studies. Hepatology 1999; 30:1269–1275. 505. Yerushalmi B, Sokol RJ, Narkewicz MR, et al. Use of rifampin for severe pruritus in children with chronic cholestasis. J Pediatr Gastroenterol Nutr 1999; 29:442–447. 506. Gauderer MW, Boyle JT. Cholecystoappendicostomy in a child with Alagille syndrome. J Pediatr Surg 1997; 32:166–167. 507. Bowen P, Lee CS, Zellweger H, et al. A familial syndrome of multiple congenital defects. Bull Johns Hopkins Hosp 1964; 114:402–414. 508. Smith DW, Opitz JM, Inhorn SL. A syndrome of multiple developmental defects including polycystic kidneys and intrahepatic biliary dysgenesis in 2 siblings. J Pediatr 1965; 67:617–624. 509. Passarge E, McAdams AJ. Cerebro-hepato-renal syndrome. A newly recognized hereditary disorder of multiple congenital defects, including sudanophilic leukodystrophy, cirrhosis of the liver, and polycystic kidneys. J Pediatr 1967; 71:691–702. 510. Opitz JM, ZuRhein GM, Vitale L, et al. The Zellweger syndrome (cerebro-hepato-renal syndrome). Birth Defects Orig Artic Ser 1969; 5.
1325
Section XII. Inherited and Pediatric Diseases of the Liver
511. Goldfischer S, Moore CL, Johnson AB, et al. Peroxisomal and mitochondrial defects in the cerebro-hepato-renal syndrome. Science 1973; 182:62–64. 512. Danks DM, Tippett P, Adams C, et al. Cerebro-hepato-renal syndrome of Zellweger. A report of eight cases with comments upon the incidence, the liver lesion, and a fault in pipecolic acid metabolism. J Pediatr 1975; 86:382–387. 513. Hanson RF, Szczepanik-Vanleeuwen P, Williams GC, et al. Defects of bile acid synthesis in Zellweger’s syndrome. Science 1979; 203:1107. 514. Moser HW, Bergin A, Cornblath D. Peroxisomal disorders. Biochem Cell Biol 1991; 69:463–474. 515. O’Brien RT. Iron overload: clinical and pathologic aspects in pediatrics. Semin Hematol 1977; 14:115–125. 516. Monnens L, Bakkeren J, Parmentier G, et al. Disturbances in bile acid metabolism of infants with the Zellweger (cerebrohepato-renal) syndrome. Eur J Pediatr 1980; 133:31–35. 517. McGuinness MC, Moser AB, Moser HW, et al. Peroxisomal disorders: complementation analysis using beta-oxidation of very long chain fatty acids. Biochem Biophys Res Commun 1990; 172:364–369. 518. Poll-The BT, Skjeldal OH, Stokke O, et al. Complementation analysis of peroxisomal disorders and classical Refsum. Prog Clin Biol Res 1990; 321:537–543. 519. Brul S, Westerveld A, Strijland A, et al. Genetic heterogeneity in the cerebrohepatorenal (Zellweger) syndrome and other inherited disorders with a generalized impairment of peroxisomal functions. A study using complementation analysis. J Clin Invest 1988; 81:1710–1715. 520. Wanders RJ, Schutgens RB, Barth PG, et al. Postnatal diagnosis of peroxisomal disorders: a biochemical approach. Biochimie 1993; 75:269–279. 521. Wanders RJ. Metabolic and molecular basis of peroxisomal disorders: a review. Am J Med Genet 2004; 126A:355–375. 522. Williams JP, Secrist L, Fowler GW, et al. Roentgenographic features of the cerebrohepatorenal syndrome of Zellweger. Am J Roentgenol Radium Ther Nucl Med 1972; 115:607–610. 523. Patton RG, Christie DL, Smith DW, et al. Cerebro-hepato-renal syndrome of Zellweger. Two patients with islet cell hyperplasia, hypoglycemia, and thymic anomalies, and comments on iron metabolism. Am J Dis Child 1972; 124:840–844. 524. Setoguchi T, Salen G, Tint GS, et al. A biochemical abnormality in cerebrotendinous xanthomatosis. Impairment of bile acid biosynthesis associated with incomplete degradation of the cholesterol side chain. J Clin Invest 1974; 53:1393–1401. 525. Salen G, Shefer S, Setoguchi T, et al. Bile alcohol metabolism in man. Conversion of 5beta-cholestane-3alpha, 7alpha,12alpha, 25-tetrol to cholic acid. J Clin Invest 1975; 56:226–231. 526. Santos MJ, Imanaka T, Shio H, et al. Peroxisomal membrane ghosts in Zellweger syndrome—aberrant organelle assembly. Science 1988; 239:1536–1538. 527. Tsukamoto T, Miura S, Fujiki Y. Restoration by a 35K membrane protein of peroxisome assembly in a peroxisome-deficient mammalian cell mutant. Nature 1991; 350:77–81. 528. Tsukamoto T, Yokota S, Fujiki Y. Isolation and characterization of Chinese hamster ovary cell mutants defective in assembly of peroxisomes. J Cell Biol 1990; 110:651–660. 529. Shimozawa N, Tsukamoto T, Suzuki Y, et al. A human gene responsible for Zellweger syndrome that affects peroxisome assembly. Science 1992; 255:1132–1134. 530. Muntau AC, Mayerhofer PU, Paton BC, et al. Defective peroxisome membrane synthesis due to mutations in human PEX3 causes Zellweger syndrome, complementation group G. Am J Hum Genet 2000; 67:967–975. 531. Moser HW. Genotype-phenotype correlations in disorders of peroxisome biogenesis. Mol Genet Metab 1999; 68:316–327. 532. Wanders RJ. Peroxisomal disorders: clinical, biochemical, and molecular aspects. Neurochem Res 1999; 24:565–580.
1326
533. Reuber BE, Germain-Lee E, Collins CS, et al. Mutations in PEX1 are the most common cause of peroxisome biogenesis disorders. Nat Genet 1997; 17:445–448. 534. Portsteffen H, Beyer A, Becker E, et al. Human PEX1 is mutated in complementation group 1 of the peroxisome biogenesis disorders. Nat Genet 1997; 17:449–452. 535. Braverman N, Steel G, Obie C, et al. Human PEX7 encodes the peroxisomal PTS2 receptor and is responsible for rhizomelic chondrodysplasia punctata. Nat Genet 1997; 15:369–376. 536. Motley AM, Hettema EH, Hogenhout EM, et al. Rhizomelic chondrodysplasia punctata is a peroxisomal protein targeting disease caused by a non-functional PTS2 receptor. Nat Genet 1997; 15:377–380. 537. Purdue PE, Zhang JW, Skoneczny M, et al. Rhizomelic chondrodysplasia punctata is caused by deficiency of human PEX7, a homologue of the yeast PTS2 receptor. Nat Genet 1997; 15:381–384. 538. Dodt G, Braverman N, Wong C, et al. Mutations in the PTS1 receptor gene, PXR1, define complementation group 2 of the peroxisome biogenesis disorders. Nat Genet 1995; 9:115–125. 539. Chang CC, Lee WH, Moser H, et al. Isolation of the human PEX12 gene, mutated in group 3 of the peroxisome biogenesis disorders. Nat Genet 1997; 15:385–388. 540. Okumoto K, Itoh R, Shimozawa N, et al. Mutations in PEX10 is the cause of Zellweger peroxisome deficiency syndrome of complementation group B. Hum Mol Genet 1998; 7:1399–1405. 541. South ST, Gould SJ. Peroxisome synthesis in the absence of preexisting peroxisomes. J Cell Biol 1999; 144:255–266. 542. Matsuzono Y, Kinoshita N, Tamura S, et al. Human PEX19: cDNA cloning by functional complementation, mutation analysis in a patient with Zellweger syndrome, and potential role in peroxisomal membrane assembly. Proc Natl Acad Sci USA 1999; 96:2116–2121. 543. Walter C, Gootjes J, Mooijer PA, et al. Disorders of peroxisome biogenesis due to mutations in PEX1: phenotypes and PEX1 protein levels. Am J Hum Genet 2001; 69:35–48. 544. Moser AE, Singh I, Brown FR, 3rd, et al. The cerebrohepatorenal (Zellweger) syndrome. Increased levels and impaired degradation of very-long-chain fatty acids and their use in prenatal diagnosis. N Engl J Med 1984; 310:1141–1146. 545. Eyssen H, Parmentier G, Compernolle F, et al. Trihydroxycoprostanic acid in the duodenal fluid of two children with intrahepatic bile duct anomalies. Biochim Biophys Acta 1972; 273:212–221. 546. Hanson RF, Isenberg JN, Williams GC, et al. The metabolism of 3alpha, 7alpha, 12alpha-trihydorxy-5beta-cholestan-26-oic acid in two siblings with cholestasis due to intrahepatic bile duct anomalies. An apparent inborn error of cholic acid synthesis. J Clin Invest 1975; 56:577–587. 547. Lee MJ, Whitehouse MW. Inhibition of electron transport and coupled phosphorylation in liver mitochondria by cholanic (bile) acids and their conjugates. Biochim Biophys Acta 1965; 100:317–328. 548. Anderson DH. Cystic fibrosis of the pancreas and its relation to celiac disease. Am J Dis Child 1938; 56:344. 549. Farber S. Pancreatic function and disease in early life. V. Pathologic changes associated with pancreatic insufficiency in early life. Arch Pathol 1944; 37:238. 550. Bodian M. Fibrocystic disease of the pancreas. London: Heinemann; 1952. 551. Craig JM, Gellis SS, Hsia DY. Cirrhosis of the liver in infants and children. AMA Am J Dis Child 1955; 90:299–322. 552. Blanc WA, Di Sant’Agnese PA. A distinctive type of biliary cirrhosis of the liver associated with cystic fibrosis of the pancreas; recognition through signs of portal hypertension. Pediatrics 1956; 18:387–409.
Chapter 69 INBORN ERRORS OF METABOLISM THAT LEAD TO PERMANENT LIVER INJURY
553. Gatzimos CD, Jowitt RH. Jaundice in mucoviscidosis (fibrocystic disease of pancreas); report of four cases. AMA Am J Dis Child 1955; 89:182–186. 554. Bernstein J, Vawter G, Harris GB, et al. The occurrence of intestinal atresia in newborns with meconium ileus. The pathogenesis of an acquired anomaly. Am J Dis Child 1960 99:804–818. 555. Sheir KJ, Horn RC, Jr. The pathology of liver cirrhosis in patients with cystic fibrosis of the pancreas. Can Med Assoc J 1963; 89:645. 556. Talamo RC, Hendren WH. Prolonged obstructive jaundice. Report of a case in a neonate with meconium ileus and jejunal atresia. Am J Dis Child 1968; 115:74–79. 557. Kulczycki LL. Editorial note. Quarterly annotated references to cystic fibrosis, vol. 6. New York: National Cystic Fibrosis Research Foundation; 1967:2. 558. Bachand JP. Un cas insuite de mucoviscidose: atteinte hepatique neonatale. Laval Med 1967; 38:371. 559. Valman HB, France NE, Wallis PG. Prolonged neonatal jaundice in cystic fibrosis. Arch Dis Child 1971; 46:805–809. 560. Rosenstein BJ, Oppenheimer EH. Prolonged obstructive jaundice and giant cell hepatitis in an infant with cystic fibrosis. J Pediatr 1977; 91:1022–1023. 561. Stern RC, Stevens DP, Boat TF, et al. Symptomatic hepatic disease in cystic fibrosis: incidence, course, and outcome of portal systemic hunting. Gastroenterology 1976; 70:645–649. 562. Feigelson J, Pecau Y, Cathelineau L, et al. Additional data on hepatic function tests in cystic fibrosis. Acta Paediatr Scand 1975; 64:337–344. 563. Tyson KR, Schuster SR, Shwachman H. Portal hypertension in cystic fibrosis. J Pediatr Surg 1968; 3:271–277. 564. Schuster SR, Shwachman H, Toyama WM, et al. The management of portal hypertension in cystic fibrosis. J Pediatr Surg 1977; 12:201–206. 565. Schwarz HP, Kraemer R, Thurnheer U, et al. Liver involvement in cystic fibrosis. A report of 9 cases. Helv Paediatr Acta 1978; 33:351–364. 566. Nagel RA, Westaby D, Javaid A, et al. Liver disease and bile duct abnormalities in adults with cystic fibrosis. Lancet 1989; 2:1422–1425. 567. Sinaasappel M. Hepatobiliary pathology in patients with cystic fibrosis. Acta Paediatr Scand Suppl 1989; 363:45–50; discussion 50–51. 568. Ling SC, Wilkinson JD, Hollman AS, et al. The evolution of liver disease in cystic fibrosis. Arch Dis Child 1999; 81:129–132. 569. Lebenthal E, Jacobson M, Kevy S, et al. Predictive value of BSP kinetics for early liver involvement in cystic fibrosis. Gastroenterology 1974; A-30:807. 570. Boat TF, Doershuk CF, Stern RC, et al. Serum alkaline phosphatase in cystic fibrosis. Interpretation of elevated values based on electrophoretic isoenzyme analyses. Clin Pediatr (Phila) 1974; 13:505–512. 571. Kattwinkel J, Taussig LM, Statland BE, et al. The effects of age on alkaline phosphatase and other serologic liver function tests in normal subjects and patients with cystic fibrosis. J Pediatr 1973; 82:234–242. 572. Rovsing H, Sloth K. Micro-gallbladder and biliary calculi in mucoviscidosis. Acta Radiol Diagn (Stockh) 1973; 14:588–592. 573. Warwick WT, L’Heureux P R, Sharp HL, et al. Gallstones in cystic fibrosis. Proc VII Int Cystic Fibrosis Congress 1976; 100. 574. Roy CC, Weber AM, Morin CL, et al. Abnormal biliary lipid composition in cystic fibrosis. Effect of pancreatic enzymes. N Engl J Med 1977; 297:1301–1305. 575. Quinton PM. Chloride impermeability in cystic fibrosis. Nature 1983; 301:421–422.
576. Frizzell RA, Rechkemmer G, Shoemaker RL. Altered regulation of airway epithelial cell chloride channels in cystic fibrosis. Science 1986; 233:558–560. 577. Quinton PM. Missing Cl conductance in cystic fibrosis. Am J Physiol 1986; 251:C649–C652. 578. Shwachman H. Gastrointestinal manifestations of cystic fibrosis. Pediatr Clin North Am 1975; 22:787–805. 579. Weber AM, Roy CC, Chartrand L, et al. Relationship between bile acid malabsorption and pancreatic insufficiency in cystic fibrosis. Gut 1976; 17:295–299. 580. Watkins JB, Tercyak AM, Szczepanik P, et al. Bile salt kinetics in cystic fibrosis: influence of pancreatic enzyme replacement. Gastroenterology 1977; 73:1023–1028. 581. Rommens JM, Iannuzzi MC, Kerem B, et al. Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 1989; 245:1059–1065. 582. Riordan JR, Rommens JM, Kerem B, et al. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science 1989; 245:1066–1073. 583. Rich DP, Anderson MP, Gregory RJ, et al. Expression of cystic fibrosis transmembrane conductance regulator corrects defective chloride channel regulation in cystic fibrosis airway epithelial cells. Nature 1990; 347:358–363. 584. Welsh MJ, Smith AE. Molecular mechanisms of CFTR chloride channel dysfunction in cystic fibrosis. Cell 1993; 73:1251–1254. 585. Tsui LC, Buchwald M. Biochemical and molecular genetics of cystic fibrosis. In: Harris H, Hirschorn K, eds. Advances in human genetics. New York: Plenum Press; 1991:153–266. 586. Kerem B, Rommens JM, Buchanan JA, et al. Identification of the cystic fibrosis gene: genetic analysis. Science 1989; 245:1073–1080. 587. Tsui LC. Mutations and sequence variations detected in the cystic fibrosis transmembrane conductance regulator (CFTR) gene: a report from the Cystic Fibrosis Genetic Analysis Consortium. Hum Mutat 1992; 1:197–203. 588. Cheng SH, Gregory RJ, Marshall J, et al. Defective intracellular transport and processing of CFTR is the molecular basis of most cystic fibrosis. Cell 1990; 63:827–834. 589. Gabolde M, Hubert D, Guilloud-Bataille M, et al. The mannose binding lectin gene influences the severity of chronic liver disease in cystic fibrosis. J Med Genet 2001; 38:310–311. 590. Walters TR, Koch HF. Hemorrhagic diathesis and cystic fibrosis in infancy. Am J Dis Child 1972; 124:641–642. 591. Underwood BA, Denning CR. Blood and liver concentrations of vitamins A and E in children with cystic fibrosis of the pancreas. Pediatr Res 1972; 6:26–31. 592. Keating JP, Feigin RD. Increased intracranial pressure associated with probable vitamin A deficiency in cystic fibrosis. Pediatrics 1970; 46:41–46. 593. Weisman Y, Reiter E, Stern RC, et al. Serum concentrations of 25-hydroxyvitamin D and 24,25-dihydroxyvitamin D in patients with cystic fibrosis. J Pediatr 1979; 95:416–418. 594. Hubbard VS, Farrell PM, di Sant’Agnese PA. 25Hydroxycholecalciferol levels in patients with cystic fibrosis. J Pediatr 1979; 94:84–86. 595. Hahn TJ, Squires AE, Halstead LR, et al. Reduced serum 25hydroxyvitamin D concentration and disordered mineral metabolism in patients with cystic fibrosis. J Pediatr 1979; 94:38–42. 596. Oppenheimer EH, Esterly JR. Hepatic changes in young infants with cystic fibrosis: possible relation to focal biliary cirrhosis. J Pediatr 1975; 86:683–689. 597. Caplan A, Gross S. Hematologic and serologic studies in cystic fibrosis. J Pediatr 1968; 73:540–547. 598. Danks DM, Allan J, Anderson CM. A genetic study of fibrocystic disease of the pancreas. Ann Hum Genet 1965; 28:323.
1327
Section XII. Inherited and Pediatric Diseases of the Liver
599. Merritt AD, Hanna BL, Todd CW, et al. The incidence and mode of inheritance of cystic fibrosis. J Lab Clin Med 1962; 60:998. 600. Oppenheimer EH, Esterly JR. Cystic fibrosis in non-Caucasian patients. Pediatrics 1968; 42:547–548. 601. Wright SW, Morton NE. Genetic studies on cystic fibrosis in Hawaii. Am J Hum Genet 1968; 20:157–169. 602. Richards CS, Grody WW. Prenatal screening for cystic fibrosis: past, present and future. Expert Rev Mol Diagn 2004; 4:49–62. 603. Richards CS, Haddow JE. Prenatal screening for cystic fibrosis. Clin Lab Med 2003; 23:503–530, x–xi. 604. Sokol RJ, Durie PR. Recommendations for management of liver and biliary tract disease in cystic fibrosis. Cystic Fibrosis
1328
Foundation Hepatobiliary Disease Consensus Group. J Pediatr Gastroenterol Nutr 1999; 28 (Suppl 1):S1–S13. 605. Cheng K, Ashby D, Smyth R. Ursodeoxycholic acid for cystic fibrosis-related liver disease. Cochrane Database Syst Rev 2000; 2:CD000222. 606. Wilson DC, Rashid M, Durie PR, et al. Treatment of vitamin K deficiency in cystic fibrosis: effectiveness of a daily fat-soluble vitamin combination. J Pediatr 2001; 138:851–855. 607. Nathanson MH, Burgstahler AD, Masyuk A, et al. Stimulation of ATP secretion in the liver by therapeutic bile acids. Biochem J 2001; 358:1–5.
Section XII. Inherited and Pediatric Diseases of the Liver
70
FIBROCYSTIC DISEASES OF THE LIVER R. Brian Doctor, S. Russell Nash, Matthew T. Nichols, and Gregory T. Everson
Abbreviations ADPKD autosomal dominant polycystic kidney disease AP-1 activating protein ARPKD autosomal recessive polycystic kidney disease CA19-9 carbohydrate antigen 19-9 CEA carcinoembryonic antigen CT computed tomography CXCR2 human chemokine receptor 2 ENA-78 epithelial neutrophil activating peptide
ER ERCP LDL IL-8 MRCP PCLD PC-1 PC-2
endoplasmic reticulum endoscopic retrograde cholangiopancreatography low-density lipoprotein interleukin-8 magnetic resonance cholangiopancreatography polycystic liver disease polycystin-1 polycystin-2
INTRODUCTION Fibrocystic liver diseases constitute a group of congenitally acquired conditions that target bile ducts and surrounding portal tracts within the liver and biliary tree. These diseases include autosomal dominant polycystic kidney disease (ADPKD), polycystic liver disease (PCLD), congenital hepatic fibrosis and autosomal recessive polycystic kidney disease (ARPKD), Caroli’s disease, choledochal cysts, and solitary hepatic cysts (Table 70-1). While these diseases are distinct from one another, they do share common features, including proliferation of biliary ductular epithelium, biliary ectasia, cyst formation, and periductular fibrosis. The last decade has witnessed enormous advances in understanding the genetic, molecular, and cellular events that underlie fibrocystic liver diseases. These advances will be highlighted in the first section of the chapter, entitled Biology of fibrocystic liver diseases. In the second section, Histopathology of fibrocystic liver diseases, we present the gross anatomical and histopathologic features of these disorders. In the third section, Clinical manifestations and treatments of fibrocystic liver diseases, the clinical features and therapies for each form of fibrocystic liver disease are described. It is hoped that recent advances that have been made in the laboratory will translate in the near future into improved medical therapies in patients afflicted with fibrocystic liver disease.
PKCa PRKCSH PTC siRNA STAT TIPS VEGF
protein kinase Ca protein kinase C substrate 80K-H gene percutaneous transhepatic cholangiography small interfering RNA signal transducer and activator of transcription transjugular intrahepatic portosystemic shunt vascular endothelial growth factor
BIOLOGY AND PATHOBIOLOGY OF FIBROCYSTIC LIVER DISEASES DUCTAL PLATE MALFORMATION HYPOTHESIS Formation of the biliary tree begins during the first trimester of fetal life, when precursor cells lying in contact with the mesenchymal tissue of the portal tracts differentiate to a glandular morphology and give rise to the ductal plate.1,2 This intermediary biliary structure consists of a double-layered tube of biliary epithelial cells surrounding the periphery of the future portal tracts, and its formation proceeds from the central portion of the liver toward progressively smaller and more peripheral branches of the biliary tree (Figure 701). During the beginning of the second trimester of fetal life, remodeling of the ductal plate normally begins with one portion, which is destined to become the functional bile duct, becoming embedded within the connective tissue of the portal tract while other sections of the circumferential ductal plate gradually degenerate and disappear. This process is completed following birth, and a discontinuous vestige of the ductal plate can be identified by cytokeratin staining in newborns.3 Histopathological examination of livers from patients with fibrocystic liver diseases commonly shows abnormal biliary structures that are reminiscent of the ductal plate stage of fetal development.
1329
Section XII. Inherited and Pediatric Diseases of the Liver
Table 70-1. The Fibrocystic Diseases of the Liver
Affected gene
Affected protein Anatomical features
ADPKD
PCLD
PKD1 (85–90%) PKD2 (10–15%) Chromosome 16 Chromosome 4 p13.3–p13.12; q21–q23; 60 230 distinct mutations have mutations in been described PKD1 have been described Polycystin-1 Polycystin-2 (~460 kDa) (110 kDa) Development of multiple large cysts in kidneys and liver
Protein kinase C substrate 80K-H (PRKCSH) Chromosome 19p13
SEC63
Hepatocystin SEC63p (59 kDa) (63 kDa) Development of multiple, large cysts in liver only
Hepatic histopathologic features
Macrocystic disease; cysts lined by simple flattened to columnar epithelium; immunohistochemical profile similar to biliary epithelium
Macrocystic disease; cysts lined by simple flattened to columnar epithelium; immunohistochemical profile similar to biliary epithelium
Clinical presentation of liver cysts
Hepatomegaly and abdominal pain; Incidence of hepatic cysts increases with age and gender; disease related to PKD2 mutations typically has later onset and associated with greater life expectancy
Similar to the hepatic presentation of ADPKD, PCLD typically presents with hepatomegaly and abdominal pain
Radiographic findings
CT, US, and MRI scanning shows multiple, large non-communicating cysts within the hepatic and renal parenchyma
CT, US, and MRI scanning shows multiple, large non-communicating cysts withing the hepatic parenchyma only
Treatment options
Radiographic cyst aspiration and sclerosis; surgical fenestration; liver resection; liver transplantation
Radiographic cyst aspiration and sclerosis; surgical fenestration; liver resection; liver transplantation
Congenital hepatic fibrosis
Choledochal cysts
Caroli’s disease
Solitary hepatic cyst
Given its association with ARPKD, thought to be related to mutations in PKHD1 (chromosome 6p) Possibly fibrocystin (447 kDa) Extensive fibrosis and malformations of the interlobular bile ducts
Unknown
Associated with mutations in PKHD1 and possibly PKD1, or some other site
Unknown
Unknown
Possibly fibrocystin
Unknown
Cystic dilations or diverticula of the extrahepatic bile ducts
Cystic dilation of the intrahepatic bile ducts that communicate with the extrahepatic biliary tree Dilatation of large bile ducts with marked periductal inflammation; bridges, and soft-tissue protrusions into dilated ducts Typically presents with recurrent cholangitis or complications of portal hypertension
Intrahepatic cyst formation that does not communicate with the biliary tree
Ductal plate malformations of the interlobular bile ducts, associated with extensive fibrosis
Biliary or intestinalized epithelium; often with extensive denudation; inflammation and reactive epithelial changes Portal hypertension, Chronic recurrent intermittent cholangitis; abdominal pain, typically jaundice, and diagnosed early in recurrent childhood; cholangitis. incidence of Can be 1 : 20 000– congenital or 1 : 40 000 acquired Large multilobulated Cholangiography Cholangiography liver with rare cysts shows cystic shows nondilation of bile obstructing duct without cystic dilations overt obstruction. that Dilations can also communicate be seen on CT, with the biliary MRCP, or EUS tree Management focused Given increased Adequate biliary on treatment of risk of drainage is complications of cholangiocarcinoma, mainstay; often cholangitis and surgical resection is requires portal indicated lobectomy or hypertension liver transplant
Simple epithelium; rounded contour
Typically asymptomatic and discovered incidentally, although right upper quadrant pain can occur when cysts exceed 5 cm in diameter
Ultrasound can distinguish simple hepatic cysts from other cystic lesions
Conservative management usually approapriate, if symptomatic radiographic aspiration and sclerosis
ADPKD, autosomal dominant polycystic kidney disease; PCLD, polycystic liver disease; CT, computed tomography; US, ultrasound; MRI, magnetic resonance imaging; MRCP, magnetic resonance cholangiopancreatography; EUS, endoscopic ultrasound; UDTs, ••
The recognition of this similarity was first noted by Jorgensen, who termed the lesion ductal plate malformation.4 Isolated lesions are often referred to as biliary microhamartoma or von Meyenburg complexes, and may be observed in normal populations. Desmet has expanded on the observations of Jorgensen to hypothesize that many cystic liver diseases represent malformations of biliary development.2,5 Although studies have not yet completely validated this hypothesis, the concept is useful to explain the similarities and overlap in the histological findings in patients with various fibrocystic diseases. Immunohistochemical characterization of the epithelia within ductal plate malformations in a variety of fibrocystic diseases demonstrates similarity to the phenotype of normal embryonic ductal plates after 20 weeks of gestational age.6 Anomalous ductal plate morphology is also observed in the hepatic conditions associ-
1330
ated with renal polycystic disease and a number of other genetic syndromes, including Meckel syndrome.5,7
GENETICS OF FIBROCYSTIC LIVER DISEASES The adult, autosomal dominant forms of fibrocystic liver diseases are typically not expressed until after puberty. Adult forms, which include the cystic liver disease associated with ADPKD and PCLD, are phenotypically autosomal dominant. Juvenile, autosomal recessive forms of fibrocystic liver disease are expressed early in life and may even be detected in utero. PKD1 and PKD2 are the genes linked to ADPKD. Despite its phenotypic expression, ADPKD is likely a molecular recessive disease with an initial germline mutation in one copy of the linked gene and a second loss-of-function somatic mutation that occurs in
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER Figure 70-1. (A, B) Morphology of portal tracts during development. Evidence of residual ductal plate structures can be seen in fetal liver. Biliary structures are highlighted by staining with cytokeratin 7. (Hematoxylin & eosin 200¥; anticytokeratin 7 immunostain 100¥.)
A
B
individual duct epithelial cells over time (Figure 70-2). The loss of a functional gene product initiates a focal proliferation and cyst formation from the affected ductular epithelial cell. This hypothesis is supported in human ADPKD where cysts emerge independently along the nephron or duct8 and mutational analysis of the cystic epithelial cells demonstrates that independent cysts are monoclonally derived.9 Animal models further support the hypothesis. Homozygous PKD1 or PKD2 null mice die before or shortly after birth. In contrast, heterozygous animals with a recombinationsensitive second allele, which allows development of spontaneous null mutations, results in formation of focal renal and liver cysts in the post-puberty period, similar to human disease.10,11 The genetic details of PCLD have begun to emerge but at this time it is unclear
whether or not PCLD is genetically an autosomal recessive or dominant disease.
PRIMARY CILIA CONTRIBUTE TO CYSTOGENESIS A general concept emerging from studies of different forms of cystic disease is that genetic mutations translate into dysfunctional proteins that alter key molecular and cellular events to disrupt normal cell growth and differentiation. This loss of normal function would then initiate cystogenesis. Detailed molecular and immunohistological studies have localized a number of cyst-linked proteins, at least in part, to the primary cilium, highlighting the sentinel importance of abnormalities in function of the primary cilium in cystogenesis (Table 70-2). The primary cilium is a lone, non-motile microtubule-
1331
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 70-2. “Two-hit”hypothesis of cyst formation in autosomal dominant polycystic kidney disease, a molecular recessive disease. The germline mutation predisposes biliary epithelium to cystic transformation, but a second somatic mutation is required to initiate cystogenesis within a single cell. Clonal expansion of this single biliary epithelial cell stimulates budding and subsequent detachment from the biliary ductule. Cyst expansion and growth ensue under regulation of a number of factors defined within the text.
Germline mutation
Somatic mutation
Clonal expansion
Detachment
Growth & expansion
Table 70-2. Cyst-linked proteins that localize to ciliary structures Gene
Protein
Location
Associated human disease
PKD1
Polycystin-1
Cilium
PKD2 PKHD1
Polycystin-2 Fibrocystin
Cilium Cilium
NPHP1 NPHP2 NPHP3 NPHP4 BBS4
Nephrocystin-1 Inversin Nephrocystin-3 Nephroretinin BBS4
Cilium Cilium Cilium Cilium Basal body
BBS5 BBS8 OFD1
BBS5 BBS8
Basal body Basal body Basal body
Autosomal dominant polycystic kidney disease (ADPKD) ADPKD Autosomal recessive polycystic kidney disease (ARPKD) Nephronophthisis (NPHP) NPHP NPHP NPHP Bardet–Biedl syndrome (BBS) BBS BBS Oral–facial–digital syndrome type 1 (OFD1)
cys1 TgN737Rpw kif3a
Cystin Polaris Kinesin-II 3a
Cilium Cilium Cilium
a a a
a
Not yet associated with a known human cystic disease.
based structure found on the luminal surface of a number of cell types, including intrahepatic bile duct epithelial cells (Figure 70-3). Normally, ciliary bending in response to fluid flow across the epithelial surface triggers Ca2+ influx into the interior of the epithelial cell.12 This process is highly regulated by a diverse set of proteins which, if dysfunctional, would alter normal function and physiology. Interestingly, polycystin-1 (PC-1) and polycystin-2 (PC-2), the proteins linked to ADPKD and fibrocystin, the protein linked to ARPKD, congenital hepatic fibrosis, and Caroli’s disease localize to the primary cilium of biliary epithelial cells. In rat bile duct epithelia, blockade of fibrocystin synthesis by small interfering RNA (siRNA) reduces ciliary length. PCK rats, an animal model of ARPKD, have abnormally short, bulbous primary cilia and the biliary
1332
tree displays ductular dilation and sites of cyst formation.13,14 These studies and others indicate that mutations affecting diverse ciliary proteins, and presumably their functions, are associated with cystogenesis.
GENES AND PROTEINS OF FIBROCYSTIC DISEASES The last decade has witnessed the discovery and characterization of the genes and proteins responsible for different forms of cystic diseases. The following section will describe the genes, proteins, and functions linked to human ADPKD, PCLD, congenital hepatic fibrosis and ARPKD, and Caroli’s disease. ADPKD will be emphasized since it is understood in the greatest detail.
AUTOSOMAL DOMINANT POLYCYSTIC KIDNEY DISEASE The most common form of autosomal dominant PCLD coexists with renal cystic disease (ADPKD) and is linked to mutations in either PKD1 or PKD2.15–17 Mutations in PKD1 account for ~85% of mutations in ADPKD families; mutations in PKD2 account for ~15% of the disease.18 Some cases are unlinked to mutations in either PKD1 or PKD2, stimulating a search for a third gene, PKD3. The phenotypic characteristics stemming from mutations in PKD1 and PKD2 are quite similar but patients with mutations in PKD2 have a later onset of disease and approximately 16 years of increased life expectancy compared to patients with mutations in PKD1.19
PKD1 and Polycystin-1 In 1957, Dalgaard demonstrated autosomal dominant inheritance in over 90% of cases of polycystic renal disease.20 In 1985, linkage techniques localized the first gene for ADPKD, PKD1, to the short arm of chromosome 16 p13.3-p13.12.21 The PKD1 gene was subsequently cloned, sequenced, and the resultant protein characterized.22 PKD1 encodes a 14.1-kb message that translates into a 4304-amino-acid protein, PC-1.23,24 Additional copies of exons 1–34
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER
A
B
Figure 70-3. Primary cilia extend from the apical surface of cholangiocytes, detect flow of fluid over the surface of the cell, and transduce flow information into the cell interior. (A) Scanning electron micrograph of primary cilia extending from the apical surface of rat intrahepatic bile duct epithelial cells (cholangiocytes). (B) Transmission electron micrograph showing the ultrastructure at the base of a primary cilium, including the basal body at the base of the primary cilium and the primary cilium extending into the lumen. Inset shows a cross-section of a primary cilium and the ring of microtubules that form the core of the primary cilium. Images reprinted from Masynk et al.13
lie adjacent to the active PKD1 locus. These duplicated copies are probably non-functional and fail to express protein but their presence has hampered the development of molecular genetic testing. There are over 230 distinct mutations of the PKD1 gene which are evenly dispersed without evidence for clustering.25 The majority are missense or nonsense mutations, but splicing mutations and gene rearrangements have also been reported. Approximately 60% of all mutations introduce premature stop codons that result in truncated proteins. Specific mutations and the positioning of mutations have been linked to intracranial aneurysms and more severe polycystic disease within individual ADPKD families.26,27 For example, mutations in the 5¢ end of PKD1 are predictive of a more rapid development and greater severity of end-stage renal disease.28 Effect of mutational type or position on hepatic cystic disease has not been specifically analyzed. PKD1 encodes for PC-1, a 460-kDa integral membrane protein with a large extracellular NH2-terminal domain, 11 putative transmembrane domains, and a comparatively small intracellular COOHterminal domain. PC-1 is predicted to transduce extracellular signals from the apical surface to the cell interior. The NH2-terminus of the protein contains a number of domains that are consistent with initiating signaling through protein–protein and protein–carbohydrate interactions. Constituting two-thirds of the protein, the extracellular NH2-terminus contains a region of leucine-rich repeats, a segment with C-type lectin characteristics, another with lowdensity lipoprotein (LDL)-like features, 16 immunoglobulin G-like PKD repeats and an REJ domain.29–31 The PKD repeat domains permit direct PC-1–PC-1 interaction. In other proteins, the REJ
domain moderates fluxes in ion-channel complexes. Its presence supports the hypothesis that PC-1 serves, in part, to regulate Ca2+ signaling. Specific protein–protein interactions of the intracellular COOH-terminal domain transduce the extracellular cues into intracellular signals. These interactions include binding with heterotrimeric G-proteins, JAK2 kinase and the COOH terminus of PC-2.32
PKD2 and Polycystin-2 In 1993, a second genetic locus linked to ADPKD was established on chromosome 4 q21-q2333 and 3 years later the PKD2 gene was identified, sequenced, and cloned.34 PKD2 produces a 5.3-kb message that codes for the 968 amino acid PC-2 protein. There are over 60 identified mutations of the PKD2 gene and, like the mutations of PKD1, the mutations of PKD2 are evenly dispersed throughout the gene without clustering at any particular position.35 Relationships of the mutations of PKD2 to phenotypic expression of polycystic disease are under study but no clearcut relationships have been reproducibly defined.36 PC-2 is a 110-kDa integral membrane protein with intracellular NH2 and COOH tails. The transmembrane domains share marked homology with voltage-gated cation channels and the COOH terminus has an EF-hand domain, a motif often expressed in voltagegated calcium channels. PC-2 likely forms a homotetramer and biophysical studies have confirmed that PC-2 functions as a cation channel.37–40 The PC-2-dependent Ca2+ transients may be further amplified by PC-2, forming a heterotetramer with the transient receptor potential channel, a calcium-activated calcium channel.41
1333
Section XII. Inherited and Pediatric Diseases of the Liver
Polycystin-1 and -2 Form a Mechanosensory Complex
PC-1–PC-2 Complexes Moderate Transcription Factor Activities
Genetic linkage of mutations in PKD1 and PKD2 to phenotypically similar forms of ADPKD and demonstration of PC-1 and PC-2 interactions through coiled-coil domains in their respective COOH tails stimulated investigation for common pathways of cystogenesis. Defining where and how the PC-1–PC-2 complex functioned within cells became a priority. Initially, PC-1 was localized to sites of cell–cell and cell–matrix contact while PC-2 was found in different regions of epithelial cells, including lateral membranes and the endoplasmic reticulum (ER). Subsequent studies demonstrated codistribution of PC-1 and PC-2 within the primary cilium of epithelial cells, which led to investigation of their roles in transducing information about flow of fluid at the epithelial surface into the cell.42,43 In cultured embryonic renal epithelial cells, flow of fluid across the cell bends the primary cilium and triggers calcium influx into the cell. This normal response is blocked by pretreatment of embryonic renal epithelial cells with channel-blocking antibodies directed against PC-2, or genetic mutations that impair PC-1 synthesis (pkd1(–/–) embryonic kidneys that lack PC-1). In these cases, fluid flow across the epithelial surface fails to elicit an increase in Ca2+i.44 This co-dependency of PC-1 and PC-2 for transducing flow information into an intracellular Ca2+i signal was confirmed in vitro by studies demonstrating PC-1 modified the ion-channel gating of PC2. In addition to flow-dependent Ca2+i signaling, the PC-1–PC-2 complex is integrated into additional signaling pathways. For example, PC-1 constitutively activates heterotrimeric Ga proteins to initiate downstream effects45,46 and PC-2 antagonizes this constitutive activity.46 Figure 70-4 demonstrates the proposed interactions of these proteins at the ciliary surface membrane.
Loss of functional PC-1 or PC-2 is predicted to misdirect cell differentiation and proliferation. Cystic epithelial cells from patients with ADPKD overexpress proto-oncogenes and growth factor receptors, suggesting a role for PC-1 and PC-2 in nuclear regulation. Consistent with this observation is the recent finding that PC-1 and PC-2 independently but coordinately mediate the activity of the nuclear activating protein (AP-1) nuclear transcription factor complex.47,48 In this regulation, PC-1 works in a protein kinase Ca (PKCa)-dependent manner, PC-2 works in a PKCe-dependent manner, and the co-expression of PC-1 with PC-2 enhances PC-2mediated activation of AP-1. Furthermore, in cells overexpressing PC-1, the carboxy-terminus of PC-1 can be cleaved from the intact protein, translocate into the nucleus, and directly activate AP-1.49 Co-expression of PC-2 blunts the effect of cleaving of the carboxyterminus of PC-1, suggesting that PC-2 modulates this pathway by buffering the concentration of the PC-1 carboxy-terminus available for nuclear translocation and signaling. Nuclear regulation by PC-1 and PC-2 is not restricted to AP-1. PC-1 also activates the signal transducer and activator of transcription (STAT) through the protein JAK2.50 The activation of PC-1-bound JAK2 requires PC-2, once again highlighting the interrelationship of PC-1 and PC-2. Activated JAK2 is then capable of activating cytoplasmic STAT allowing activated STAT to enter the nucleus and arrest the cell cycle progression in G0/G1. Loss of either PC-1 or PC-2 would be predicted to allow cells to re-enter the cell cycle and promote cell replication (Figure 70-5). Many of the details and interrelationships of these disparate pathways for regulating signaling, transcription, and growth require additional investigation. It is important to recognize, however, the emerging evidence for PC-1/PC-2 in the regulation or coordination of cell proliferation and differentiation. These pathways can now be investigated to determine which pathways, when disrupted, give rise to fibrocystic liver disease.
Ca2+
IgG
N
REJ
PKD
PC2
PC1
G
PC-1 and PC-2 interact via their COOH-terminal tails Activated PC-1 induces PC-2 Ca2+ channel activity Figure 70-4. Genes and proteins of autosomal dominant polycystic kidney disease: two genes, two proteins, one common pathway? Interaction of polycystin-1 (PC-1) and polycystin-2 (PC-2) at the surface membrane of primary cilium may regulate calcium flux and trigger intracellular events (see text for description). PKD, polycystic kidney disease repeat domain; REJ domain, IgG, immunoglobulin G.
1334
Contributing Events to ADPKD Liver Cyst Disease In ADPKD, the clinical course of the disease is markedly heterogeneous, even among identical twins. While part of this heterogeneity is likely due to differences in somatic mutations in PKD1 and PKD2, additional modifier genes and other factors likely promote the growth of the liver cysts. Modifier genes, genes that are not directly linked to a genetic disease but influence the disease phenotype, influence a number of genetic diseases. There are a number of strong candidate modifier genes for ADPKD and significant efforts have been made directly to identify and characterize these modifier genes. In the case of ADPKD liver cyst disease, estrogen levels, luminal fluid secretion, and the accumulation of cytokines and growth factors all likely influence cyst expansion. ADPKD liver cyst severity is much greater in women than men. Studies showing a positive correlation between liver cyst volumes and numbers of pregnancies, estrogen-based birth control and postmenopausal estrogen therapy indicate that estrogen promotes liver cyst growth and is responsible for the sexual dimorphism observed in ADPKD liver cyst disease. At the cellular level, cholangiocytes express a
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER
Flow
Gα STAT-P Ca2+
PKCα
Figure 70-5. Characterization of putative polycystin1 (PC-1)/PC-2 signaling pathways from the primary cilium. A number of signaling pathways linked to PC1/PC-2 have been initially characterized. Intracellular sites for this signal transduction complex include the primary cilium, where bending of the primary cilium induces PC-1/PC-2-dependent Ca2+ signaling. In addition, heterotrimeric G protein signaling, activation of nuclear activating factor-1 (AP-1) transcription factors and signal transducer and activator of transcription (STAT) phosphorylation and nuclear translocation have all been implicated in PC-1/PC-2-directed signal transduction. Loss of these functions is predicted to impair growth and differentiation and permit cystogenesis.
PKCε PC1 tail
AP-1
STAT-P
and b estrogen receptors and estrogen can induce proliferation in cholangiocytes.51 Unlike the juvenile fibrocystic forms of liver cyst diseases, in ADPKD, and likely PCLD, the proliferating epithelial cells form nascent cysts that detach from the original duct to form enclosed, autonomous structures. Expansion of the autonomous cysts is likely promoted by luminal fluid secretion, epithelial proliferation, and angiogenesis. Normal intrahepatic bile duct cells respond to specific gut hormones, such as secretin, to secrete a bicarbonate-rich fluid. Human hepatic cysts retain this regulated secretory capacity, generate a positive intraluminal cyst pressure under basal conditions, and have increased rates of fluid secretion following intravenous administration of secretin.52,53 Increasing intraluminal pressure in cell culture models of epithelial cysts increases rates of epithelial cell proliferation.54,55 Accordingly, increased epithelial stretching that follows from the regulated secretion into enclosed cysts will may initiate proliferation of the lining epithelial cells and cyst growth. One factor that is released from epithelial cells during stretching is interleukin-8 (IL-8). Analysis of liver cyst fluid from human ADPKD subjects showed that levels of IL-8, along with epithelial neutrophil activating peptide (ENA-78), vascular endothelial growth factor (VEGF), and IL-6, are specifically elevated to physiologically relevant levels.56 Human chemokine receptor 2 (CXCR2), a receptor for both IL-8 and ENA-78, is capable of driving cell proliferation and is localized to the apical domain of human liver cyst
epithelial cells. Thus an autocrine/paracrine loop is established where these factors are secreted and accumulate within the cyst fluid and then can bind and signal cell proliferation of the lining epithelium. The secretion of these factors is not unidirectional. In a cellular model of ADPKD, VEGF is preferentially secreted across the basolateral membrane.56 Basolateral secretion of potent angiogenic factors like VEGF would position these factors to initiate angiogenesis. In total, the secretory and growth responses of an individual’s cystic epithelium to hormones, cytokines, and growth factors are likely to have significant effects on the growth rates of their cysts and contribute to the heterogeneity in cystic disease severity in patients with similar genetic backgrounds.
POLYCYSTIC LIVER DISEASE In contrast to the concurrent renal and liver cystic disease characteristic of ADPKD, PCLD gives rise to PCLD with no discernible renal manifestations. This phenotypic distinction is paralleled by disparate genetic linkages. To date, PCLD has been definitively linked to two genes, PRKCSH and Sec63. Early analysis suggests these two genes only partially account for the PCLD cases, indicating there is at least one additional locus linked to PCLD.
PRKCSH and Hepatocystin The first demonstration that PCLD was genetically distinct from PDK1 and PKD2 was from phenotypic and genetic study of a family
1335
Section XII. Inherited and Pediatric Diseases of the Liver
affected by PCLD without renal cystic disease in three generations.57 PCLD was initially linked to mutations in protein kinase C substrate 80K-H gene (PRKCSH) on chromosome 19p13.213.1.58–60 PRKCSH encodes for a 527-amino-acid protein of 59 kDa, termed hepatocystin, that is expressed in a number of different tissues.61 The protein contains a membrane translocation signal sequence for ER at its amino end, an LDLa domain, two EF-hand domains, a glutamic acid-rich region, and an ER retrieval sequence at its carboxyl end. There have been several reported functions for hepatocystin but the weight of evidence indicates a primary role as the non-catalytic subunit of the glucosidase II protein complex.62,63 Glucosidase II is localized in the ER where it modifies protein glycosylation and contributes to post-translational processing of newly synthesized glycoproteins. Mutations that truncate or alter mRNA splicing of hepatocystin have been reported in PCLD families. Initial studies suggest there may be a correlation between the site of the mutation and the severity of the disease but this awaits confirmation in studies with a larger population base.
Sec63 and Sec63p The functional association of PCLD with the ER and protein processing was bolstered by the linkage of a second gene, Sec63, to PCLD. In humans, the Sec63 gene is on chromosome 6q21, and mutations are distributed throughout the gene.64,65 The protein product of the Sec63 gene is designated as Sec63p, an integral membrane protein within the ER that functions as a component of the protein translocation complex. This proteinprocessing step is upstream of the glycosylation modification step that is performed by glycosidase II. Interestingly, despite the apparent liver specificity of PCLD that is initiated by Sec63 mutations, Sec63 is broadly expressed in a number of tissues, including the kidney.
Comparative Mechanisms of PCLD and ADPKD The developmental characteristics of liver cysts in ADPKD and PCLD are markedly similar. In both diseases, liver cysts develop after puberty, clinical manifestations appear in the fourth decade of life, and disease expression is sexually dimorphic, with women having a greater degree of cyst expansion. The striking difference between these two diseases is that ADPKD affects kidneys, vasculature, and liver while expression of PCLD is limited to the liver. As described above, the proteins linked to ADPKD form a mechanosensory signal transduction complex within the primary cilia. In contrast, proteins linked to PCLD participate in the translocation, processing, and quality control of ER proteins. One hypothesis, to explain the divergent phenotypes, suggests that hepatocystin and Sec63p are essential for the ER processing of PC-1 or PC-2 in bile duct epithelial cells but are alternatively compensated for in renal epithelial cells. Understanding the physiologic intersecting pathways influenced by the mutated genes responsible for ADPKD and PCLD will likely reveal the pivotal steps involved in liver cystogenesis.
AUTOSOMAL RECESSIVE POLYCYSTIC KIDNEY DISEASE Genetic and molecular events responsible for ARPKD are also unfolding. ARPKD is far less common than ADPKD, with an inci-
1336
dence of 1 in 20 000 live births, but its high degree of morbidity and mortality makes ARPKD an important cause of pediatric disease.66,67
PKHD1 and Fibrocystin Unlike the autosomal dominant forms of polycystic diseases, ARPKD is linked to a single gene, PKHD1 (polycystic kidney and hepatic disease 1). PKHD1 is a large, alternatively spliced gene located on chromosome 6p12. Over 100 mutations in the PKHD1 gene have been cataloged. The majority of affected individuals are compound heterozygotes. The presence of two truncation mutations is correlated with a severe phenotype and high mortality rates in the perinatal and neonatal period. PKHD1 encodes for a 4074-aminoacid, 447-kDa integral membrane protein termed fibrocystin. The function of fibrocystin is currently undetermined but the multiple immunoglobulin-like motifs in its extracellular domain suggest that fibrocystin serves as a surface receptor. As noted above, fibrocystin localizes to the primary cilium in rat intrahepatic bile duct epithelial cells.13 Interestingly, primary cilia in rat bile duct epithelial cells that lack fibrocystin are short and bulbous in stature. Further, tomography of the biliary tree showed that loss of fibrocystin is paralleled by marked distortions of the biliary tree, including ductular dilatation and focal budding.13 It is hypothesized that errant cell signaling from the primary cilia occurs in cells without functional fibrocystin and this errant signaling, in turn, induces the ductular dilation and fibrosis that are observed in ARPKD.
HISTOPATHOLOGY OF FIBROCYSTIC LIVER DISEASES AUTOSOMAL DOMINANT POLYCYSTIC LIVER DISEASE The histopathologic features of liver cysts in ADPKD and PCLD are indistinguishable. Gross anatomical examination shows varying degree of enlargement due to the expansion of multiple fluid-filled cysts (Figure 70-6). Involvement can be either diffuse or segmental within the liver, although diffuse disease is more common. In segmental disease, the left lobe is preferentially affected. Cysts appear to cluster along the track of major vascular/portal structures within the liver, but mature cysts do not communicate with the biliary tree.68 The presence of cysts is commonly associated with the presence of biliary microhamartomas, and these structures have been proposed as cyst precursors.69 The lining of the cysts is composed of a simple epithelium of cuboidal to columnar-type cells that overlie a dense band of fibrous connective tissue (Figures 70-7 and 70-8). Inflammation is generally not found around cysts except in the cases where they show evidence of infection with predominantly acute inflammatory cells. The cytological features of the cyst-lining cells resemble the epithelia of interlobular bile ducts, and immunohistochemically both structures express a particular subset of cytokeratin proteins (socalled biliary types 7 and 19).70 Cyst-lining cells show only weak reaction with antibodies, recognizing tumor markers such as carbohydrate antigen 19-9 (CA19-9) and carcinoembryonic antigen (CEA).71 MUC1 is an oncofetal mucin antigen that is expressed in the epithelial cell lining of many types of hepatic cysts including
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER Figure 70-6. Autosomal dominant polycystic liver disease. Macroscopic view of segmental polycystic disease. Less involved hepatic parenchyma is present to the right side of the slide, where isolated cystic structures are seen in proximity to large portal tracts.
those of ADPKD. Interestingly, only interlobular bile ducts in patients with ADPKD showed abnormal MUC1 expression in contrast to interlobular bile ducts in normal livers or livers with solitary cysts.72 Development of cholangiocarcinoma, while rare, is a recognized complication of ADPKD that may arise following a dysplasia–carcinoma sequence in the epithelium of peribiliary cysts.73
CONGENITAL HEPATIC FIBROSIS AND AUTOSOMAL RECESSIVE POLYCYSTIC KIDNEY DISEASE Congenital hepatic fibrosis (Figures 70-9 and 70-10) demonstrates one of two histopathological broad patterns, either focal or diffuse.2,5 In focal disease, there is enlargement of individual portal tracts associated with the presence of many abnormal biliary structures suggestive of maldevelopment of fetal ductal plates, but without bridging fibrosis of the type seen in the diffuse form. The diffuse form of the disease shows similar abnormal biliary structures; however, these are embedded within prominent irregular bands of fibrosis that connect portal tracts to other portal tracts and occasionally to hepatic venules. Inflammation is generally non-existent to minimal. As yet, prospective studies have not shown that specific histological patterns impact patient prognosis. Although many patients present in early childhood with portal hypertension, patients without coincident renal disease may remain asymptomatic, or present with portal hypertension as adults.74,75 Some cases demonstrate decreased numbers or decreased size of portal vein branches that may represent a presinusoidal contribution to the portal hypertension in these patients.68 Typically, cystic lesions are rarely identified in patients with congenital hepatic fibrosis by gross inspection of the liver after explantation or at the time of postmortem examination. However, some patients with congenital
hepatic fibrosis also manifest Caroli’s disease.2,5,68 In these cases, biliary cystic dilatations may be prominent.
CAROLI’S DISEASE This disease is defined by the presence of congenitally dilated intrahepatic bile ducts in the absence of obstruction, and the disease may be superimposed on congenital hepatic fibrosis (ARPKD) or polycystic disease (ADPKD)76,77 (Figures 70-11 and 70-12). Several studies have correlated radiological and pathological findings in patients with Caroli’s disease or syndrome.78,79 By macroscopic examination, bridges and protrusions of soft tissue are seen within the lumens of the large, ectatic bile ducts. Brown pigmented stones, composed of inspissated bile, may also be present within the ducts. Upon microscopic examination, some protrusions contain hepatic arteries and portal vein branches, suggesting that the biliary dilatation is related to maldevelopment of fetal ductal plates at the level of the central biliary branches. Inflammation is usually prominent around the areas showing cystic change. In some cases, small peripheral portal tracts may show relatively normal morphology without abnormal ductal plate elements, although periductal fibrosis suggestive of secondary sclerosing cholangitis can be seen. In other cases, ductal plate malformation occurs at all levels of the biliary tree and microhamartomatous structures can be seen in peripheral portal tracts.
CHOLEDOCHAL CYSTS These lesions represent one end of a spectrum that includes Caroli’s disease, as considerable overlap in histological findings exists. Classification is usually based on the radiographic appearance. Choledochal cysts typically show extensive denudation of the epithelial lining80 (Figure 70-13). When the lining is present, there is most commonly a simple epithelium with reactive cytological and architectural features. Whereas some authors report a predominance of
1337
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 70-7. Autosomal dominant polycystic liver. (A) Epithelial variation within cysts ranges from flattened cells (left vertical surface) to cuboidal (top horizontal surface) to columnar (lower right side). (Hematoxylin & eosin 100¥.) (B) Biliary microhamartomas are often associated with autosomal dominant polycystic liver disease. This view shows enlarging cystic spaces within a microhamartoma that contains bile pigment. (Hematoxylin & eosin 100¥.)
A
B
1338
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER Figure 70-8. Autosomal dominant polycystic liver. Electron microscopy demonstrated the thickened fibrotic matrix underlying cystic epithelial cells. Arrowheads point to lining epithelial cells. Hashed circles highlight areas of deuse collagen bundles. Asterisk denotes a potential fibroblast embedded within the dense matrix material.
Figure 70-9. Congenital hepatic fibrosis. Macroscopic view of an asymptomatic adult patient shows yellow-white patches of mild fibrosis (arrowheads) scattered through the liver. (Courtesy of Dr. Atul Bhan, Massachusetts General Hospital, Boston, MA, USA.)
1339
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 70-10. Congenital hepatic fibrosis. Fibrosis ranges from focal portal tract enlargement (Masson’s trichrome 50¥) to severe fibrosis in a 57year-old patient with massive hepatomegaly (Masson’s trichrome 100¥).
Figure 70-11. Caroli’s disease. Ectatic bile ducts in the hilum of the liver show papillary protrusions into duct lumen. (Hematoxylin & eosin 100¥.)
1340
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER Figure 70-12. Caroli’s disease. Other ducts show marked periductal chronic inflammation. (Hematoxylin & eosin 100¥.) Peripheral bile ducts do not show evidence of ductal plate malformation, although this case shows periductal fibrosis indicating secondary biliary sclerosis. (Hematoxylin & eosin 50¥.)
Figure 70-13. Choledochal cyst. Low-power examination of the lining of a choledochal cyst shows peribiliary glands within the cyst wall (arrow) with epithelial denudation (right side) adjacent to an island of preserved epithelium (left side). (Hematoxylin & eosin 50¥.)
intestinal-type epithelium in younger patients with evolution to a non-descript columnar or biliary epithelium in older patients,81 others suggest that the likelihood of finding intestinalized epithelium increases with patient age.82 Subepithelial inflammation is usually present and of variable density, with a mixture of inflammatory cell types represented. Inflammatory infiltration is probably related to epithelial disruption and bacterial overgrowth associated with bile stasis that may also predispose cysts to metaplastic and dysplastic changes.83 The risk of carcinoma is increased 5–35-fold
in patients with choledochal cysts, and risk appears to increase with patient age.84
SIMPLE (NON-NEOPLASTIC, NON-PARASITIC, NON-COMMUNICATING) HEPATIC CYSTS This common lesion (affecting roughly 1–2% of the general population) is generally referred to as solitary hepatic cyst, although as many as 40% of patients may have multiple lesions.85,86 The
1341
Section XII. Inherited and Pediatric Diseases of the Liver
size of these lesions is generally small, although giant lesions as large as 27 cm in diameter have been reported.87 Histologically, the epithelial lining of these hepatic cysts is a single layer of low cuboidal to columnar epithelium. Epithelial cells may show mucinous features but without goblet cell formation. Immunohistochemical features of cytokeratin and mucin expression are similar to cysts arising in ADPKD.71,72 Cyst-lining cells show only weak immunohistochemical expression of tumor markers such as CA19-9 and CEA despite elevated levels of these markers in cyst fluid.71,88
CLINICAL MANIFESTATIONS AND TREATMENT OF FIBROCYSTIC LIVER DISEASES POLYCYSTIC LIVER DISEASE Natural History The natural history of autosomal dominant forms of PCLD is strikingly similar, regardless of etiologic mutation (PKD1, PKD2, PRKSCH, Sec63). The prevalence and number of hepatic cysts in patients with ADPKD increase with increasing age (Figure 70-14), female gender, severity of renal cystic disease, and severity of renal dysfunction. By age 60 nearly 80% of ADPKD patients have hepatic cysts.69,89–91 Men and women have equal lifetime risk to develop hepatic cysts but women experience greater numbers and larger sizes of hepatic cysts. Severe hepatic cystic disease correlates with both pregnancy and use of exogenous female steroid hormones. One longitudinal study of anovulatory women with ADPKD treated with hormone replacement suggested that estrogens selectively increase severity of hepatic cystic disease (Figure 70-15).92 Female tendency to develop massive hepatic cystic disease is also characteristic of isolated PCLD.93–95
100
15
60 40 20 0 1st
2nd
3rd
4th
5th
6th
>6th
N=16
28
80
137
80
45
42
Age (decade) Figure 70-14. Polycystic liver. Prevalence of hepatic and renal cysts in autosomal dominant polycystic kidney disease (ADPKD). The frequency of renal and hepatic cysts is displayed by age in at-risk members of kindreds known to be affected by ADPKD. Cysts were detected by real-time ultrasonography. The population at risk included 239 patients with ADPKD and 189 unaffected family members. Hepatic cysts are rarely detected before puberty, but by the fifth decade of life approximately 80% of patients with renal cysts have liver cysts.
Change from baseline (%)
Percentage
The molecular diagnostic approaches to ADPKD have advanced considerably with the availability of direct gene sequencing. Using commercially available methods, pathologic mutations are detected in 44–76% of families with mutations of PKD1 and 75% of families with mutations of PKD2.64,65,96–99 Molecular genetic testing is relatively new and consensus guidelines are lacking. Ultrasonography is a reasonable screening tool in adults. In young (age < 30 years), presymptomatic individuals at risk of ADPKD, molecular genetic testing may offer a number of advantages over other approaches. In this group ultrasonography may lack sensitivity and linkage analysis is impractical. The identification of PKD1 or PKD2 mutations could affect family planning or choice of future diagnostic studies. Discovery of ADPKD mutations may also encourage regular blood pressure monitoring and screening for associated conditions such as cerebral aneurysm or mitral valve prolapse. Genetic testing may also have a role in the evaluation of young family members being considered as living donors for renal transplantation to another family member with renal failure from ADPKD. Clarifying the mutation status as negative in such potential donors would reduce future risks to both the donor and the recipient. Clinical genetic testing for PCLD is also available in Europe and includes genetic sequencing of PRKCSH. Clinical testing for SEC63 mutations will likely be available soon. Because sensitivity of ultrasonography in PCLD is undefined, use of PRKCSH testing (and ultimately SEC63 testing also) to identify presymptomatic patients at risk of the disease may have even more relevance than in ADPKD. While genetic testing results may not immediately alter the management of a PCLD-affected patient, the ability to screen other asymptomatic at-risk family members (and potential transplant donors) for mutations is important for patient care. As patients are frequently concerned about the risks to their offspring, formal genetic counseling is recommended even when the mutation status is unknown. An algorithm for the use of genetic tests is given in Figure 70-16.100,101
Renal Hepatic
80
1342
Molecular Diagnostics
Control Estrogen
10 5 0 -5 -10 Total liver
Liver cyst
Liver Parenchyma
Total kidney
Figure 70-15. Polycystic liver. Postmenopausal estrogen selectively increases liver volume. Percentage of volume change in liver and kidney following 1 year of postmenopausal estrogen treatment is shown. Estrogen-treated patients are depicted by black bars and non-treated controls as hatched bars. Estrogen treatment increases hepatic volume and hepatic cystic disease but does not influence renal cystic disease.
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER
Patient with PLD
Are renal cysts present?
Yes
No
AD-PKD likely
PCLD likely
Patient evaluation • Assess renal and hepatic function • Treat hypertension • MRI for cerebral aneurysms if suggested by symptoms or family history of aneurysmal disease
Patient evaluation • Assess hepatic function • Possible discontinuation of estrogen therapy • Cyst aspiration/sclerosis/fenestration as indicated
Genetic evaluation • Pedigree interpretation • PKD1 / PKD2 genetic testing if clinically indicated (reproductive decision making, screening of younger relatives, transplant evaluation of potential donor(s)
Genetic evaluation • Pedigree interpretation • PRKCSH / SEC63 genetic testing if clinically indicated (reproductive decision making, screening of younger relatives, transplant evaluation of potential donor(s)
Family screening • Renal ultrasound screening of at-risk relatives
New diagnosis of AD-PKD or PCLD
Family screening • Hepatic ultrasound screening of at-risk relatives
Figure 70-16. Polycystic liver. Algorithm for evaluation of patients with known or suspected polycystic liver. ADPKD, autosomal dominant polycystic kidney disease; PCLD, polycystic liver disease; MRI, magnetic resonance imaging.
Clinical Features Patients with small (<2 cm) or few hepatic cysts tend to be clinically asymptomatic. In contrast, patients who develop massive hepatic cystic disease, based upon a definition of a total liver cyst:parenchymal volume ratio >1, become symptomatic with abdominal pain or discomfort, early postprandial fullness, or shortness of breath (Table 70-3).90,92 Consequences of progressive ADPKD include renal failure, requirement for hemodialysis, and renal transplantation. Dialysis, in particular, is thought to increase risk for hepatic complications. A single center reported that 21% of polycystic patients on dialysis experienced hepatic complications, mainly hepatic cyst infection, hemorrhage, or cyst carcinoma,102 a finding that was not confirmed in a subsequent report by a different center.103 Typically, liver parenchymal volume is preserved despite extensive hepatic cystic disease and extraordinary distortion of hepatic architecture.104 The only blood test abnormality is modest elevation of g-glutamyltransferase, which correlates with hepatic cyst burden. The diagnosis of polycystic liver is readily established by its characteristic appearance on computed tomography (CT) scan (Figures 70-17 and 70-18). Quantitative tests of liver function indicate
impairment of metabolic capacity and increased portosystemic shunting in patients with massive polycystic disease, in the absence of any obvious clinical manifestations.90 Rarely, a patient with PCLD will experience hepatic decompensation and variceal hemorrhage, ascites, or encephalopathy. The most common, clinically relevant complications arising in hepatic cysts are intracystic hemorrhage,105 infection, or post-traumatic rupture (Table 70-4). Cyst adenocarcinoma, biliary obstruction, Budd–Chiari syndrome,106,107 or hepatic failure are rarely reported. Associated conditions include mitral valve prolapse, diverticulosis, inguinal hernia, and cerebral aneurysm.108
THERAPY Medical treatments There are no effective medical therapies for PCLD. Somatostatin analogue, which blocks secretin effects, failed to demonstrate any significant effect on hepatic cyst growth or size.109 As noted above, hepatic cystic disease, but not renal cystic disease, worsens under the influence of female gender, pregnancy, or use of exogenous female steroid hormones.92 These observations suggest that women
1343
Section XII. Inherited and Pediatric Diseases of the Liver
Table 70-3. Gastrointestinal symptoms in patients with polycystic liver disease PLD versus control
PLDmin versus PLDmass
Esophageal Belching Dysphagia Odynophagia Regurgitation Heartburn Hematemesis
0.055 NS NS NS NS NS
NS NS NS NS NS NS
Upper gastrointestinal tract Postprandial fullness Bloating Pain Melena Nausea Vomiting
0.010 0.0005 0.003 NS NS NS
0.015 0.014 0.0005 NS NS NS
Lower gastrointestinal symptoms Diarrhea NS Nocturnal diarrhea NS Constipation 0.079
NS NS NS
Liver-specific symptoms Jaundice Bilirubinuria Acholic stools Pruritus
NS NS NS NS
NS NS NS NS
PLD, patients with polycystic liver disease; PLDmin, patients with PLD with liver cyst:parenchymal volume < 1; PLDmass, patients with PLD with liver cyst:parenchymal volume ≥ 1; NS, not significant. Significance of differences in frequency of symptoms was assessed by chi-squared test.
with polycystic liver should avoid estrogen replacement therapy. However, proof that avoidance of estrogen effectively alters the course of hepatic cystic disease is lacking. The clinician must individualize hormonal replacement therapy in polycystic patients by weighing the potentially deleterious effect on hepatic cystic disease against other potential benefits.
Radiologic cyst aspiration and sclerosis Symptomatic patients with one or few dominant cysts may be considered for cyst aspiration and sclerosis. Most patients with polycystic disease have too many cysts or cysts are of insufficient size to warrant this approach. Cyst sclerotherapy requires ultrasonographic or CT-guided percutaneous puncture of the targeted cyst and placement of an intracystic drainage catheter. Success in obliterating individual cysts in polycystic patients is approximately 70–90%.110,111
Cyst fenestration Cyst fenestration is a common surgical treatment in the management of symptomatic massive hepatic cystic disease.112–115 Two approaches have been used: open laparotomy and, more recently, laparoscopy. Several series of open laparotomy, encompassing large numbers of patients, indicate that this approach results in satisfactory resolution of symptoms. However, open laparotomy is associated with prolonged hospitalization, morbidity (bleeding, infection, bile leak, ascites) of major abdominal surgery (0–50%), and even mortality (< 1%). Because of its less invasive nature, laparoscopic
Figure 70-17. Polycystic liver. Autosomal dominant polycystic kidney disease with liver cysts. Computed tomography demonstrates numerous cysts in liver and both kidneys. This patient also demonstrates a large umbilical hernia which may complicate massive polycystic disease.
1344
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER
cyst fenestration is gaining increasing acceptance as an alternative surgical technique. Advantages of laparoscopic surgery include less morbidity, reduced hospital stay, and the potential for outpatient surgical management. One recent review,115 totaling 40 cases, indicated that symptoms recurred in about half, necessitating repeat laparoscopic cyst fenestration. Although there were no deaths, sur-
gical conversion to open cyst fenestration occurred in 10% of cases and 35% of patients suffered complications (Table 70-5).
Liver Resection One center reported their experience with partial liver resection in the management of 31 patients with highly symptomatic, massive
Figure 70-18. Polycystic liver. Polycystic liver with minimal renal involvment in patient from autosomal dominant polycystic kidney disease cohort. Computed tomography demonstrates numerous hepatic cysts and a few cysts in kidneys.
Table 70-4. Complications of polycystic liver disease Classification
Specific type
Diagnostic tests
Treatment
Infection
MRI, indium111 WBC scan
Hemorrhage
CT or MRI
Carcinoma
CT or MRI Aspiration cytology
Fluoroquinolones Drainage Pain control Drainage (rare) Surgery
Biliary obstruction
ERCP
Stent placement Cyst decompression
Venous obstruction Hepatic
Hepatic venography
Cyst decompression Resection Transplantation Cyst decompression Resection Transplantation
Arising within cyst
Compression by cyst
Portal
MRI/MRA Mesenteric angiography
Portal hypertension
Endoscopy (varices) US/CT/MRI
Hepatic failure
Exceedingly rare Look for other cause
Hepatic dysfunction Band ligation Cyst decompression Resection Transplantation Transplantation
MRI, magnetic resonance imaging; WBC, white blood cell; CT, computed tomography; ERCP, endoscopic retrograde cholangiopancreatography; MRA, magnetic resonance angiography; US, ultrasound.
1345
Section XII. Inherited and Pediatric Diseases of the Liver
100
Table 70-5. Outcome after laparoscopic cyst decompression112–115 Series 1
Series 2
Series 3
Series 4
Totals
62%
57%
33%
72%
58%
0
29%
11%
9%
10%
46% 13
57% 7
33% 9
10% 11
35% 40
Liver+kidney
80 Survival (%)
Recurrent symptoms Conversion to open Complications Number of patients
Liver only
60
P=NS
40 20 0
hepatic polycystic disease.116 Their ages ranged from 34 to 69, the gender ratio (M:F) was 3:28, and renal function varied from normal to dialysis dependence. Nearly all patients experienced significant relief from symptoms and long-term sustained reduction in symptoms was common (>95%). However, over 50% experienced significant perioperative morbidity and there was one perioperative death (due to rupture of intracranial aneurysm). Most surgeons reserve hepatic resection for those cases that are refractory to cyst decompression.
Liver Transplantation Polycystic liver patients with symptoms refractory to other treatments or symptomatic hepatic cystic disease with end-stage renal cystic disease may be considered for liver or combined liver–kidney transplantation.117–122 Rarer indications for hepatic transplantation include variceal hemorrhage, ascites, obstruction of hepatic venous outflow (Budd–Chiari-equivalent), or biliary tract obstruction by extensive cystic disease not amenable to other interventions. One-, 3-, and 5-year survivals for patients undergoing isolated hepatic transplantation (n = 128) are 78.1%, 71.7%, and 68.7%, respectively (www.ustransplant.org; from US Scientific Registry of Transplant Recipients with the aid of Dawn Zinsser, Director of Analytic Support, February 2004). One-, 3-, and 5-year survivals for patients undergoing combined liver–kidney transplantation (n = 78) are 79.5%, 75.5%, and 75.5%, respectively (Figure 70-19). The genetic basis of PCLD warrants caution with respect to living donor liver transplantation.
CONGENITAL HEPATIC FIBROSIS Characteristics Congenital hepatic fibrosis is a rare, inherited, autosomal-recessive disorder that is most often associated with ARPKD. Other clinical associations of congenital hepatic fibrosis include renal dysplasia, nephronophthisis, Meckel–Gruber syndrome, Ivemark syndrome, Jeune syndrome, vaginal atresia, and tuberous sclerosis. Congenital hepatic fibrosis can coexist with other fibrocystic liver diseases such as Caroli’s disease and choledochal cyst.121–123
Clinical Features Congenital hepatic fibrosis presents in three clinical forms: (1) portal hypertension; (2) recurrent cholangitis; and (3) asymptomatic or latent disease.125 The first two forms are usually diagnosed in early childhood in patients who present with variceal hemorrhage or unexplained biliary sepsis. In contrast, some patients will be
1346
0
2
4
6
Years post-transplant
Figure 70-19. Polycystic liver. Patient survival after liver transplantation for polycystic liver disease. Results for combined liver–kidney and isolated liver transplantation are shown. Immediate survival is adversely affected by hemorrhage related to extensive dissection, infection, and complications related to renal dysfunction. Beyond 6 months post-transplantation, survival curves are flat. The difference at 5 years is not statistically significant between the two types of transplant.
detected later, during their adult years, when they are evaluated for unexplained hepatomegaly or portal hypertension. Rarely, patients present with evidence of both portal hypertension and cholestasis, the latter due to either associated biliary anomalies (Caroli’s disease) or to intrinsic destructive cholangiopathy. In general, hepatic function is well preserved, despite portal hypertension or bouts of cholangitis, although some patients experience progressive hepatic failure in long-term follow-up.
Treatment The first-line treatment of variceal hemorrhage is endoscopic variceal eradication (either sclerotherapy or ligation treatment), followed by institution of b-adrenergic blockade. In most cases, varices may be successfully obliterated by the endoscopic approach, thereby controlling this potentially life-threatening complication. Surgical shunts or transjugular intrahepatic portosystemic shunt (TIPS) are reserved for patients who fail endoscopic therapy, bleed from gastric varices, or who have portal hypertensive gastropathy. Occasionally patients will experience progressive hepatic fibrosis and hepatic dysfunction after long-standing portosystemic shunt surgery, and development of this complication may necessitate consideration for liver transplantation. In patients with cholangitis, radiologic imaging (ultrasonography, biliary radioscintigraphy, CT, or magnetic resonance imaging) may be required to determine whether the patient with congenital hepatic fibrosis has concomitant biliary cystic disease. If the latter is present, the treatment of cholangitis is centered on provision of adequate biliary drainage, relief of obstruction (papillotomy with stone extraction or stricture dilatation), and control of infection with antibiotics. In the absence of biliary cystic disease or cholangiocarcinoma (6% of cases of congenital hepatic fibrosis), cholestasis may be related to idiopathic inflammatory destructive cholangiopathy and respond to ursodeoxycholic acid therapy. Indications for hepatic transplantation include variceal hemorrhage or
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER
hemorrhage from portal hypertensive gastropathy that is not responsive to endoscopic treatment or amenable to portosystemic shunt surgery or TIPS; recurrent cholangitis that is not amenable to medical, endoscopic, radiologic, or surgical therapy; and hepatic failure (development of coagulopathy, biochemical deterioration, ascites, or portosystemic encephalopathy).
CAROLI’S DISEASE Characteristics In 1958, Caroli described a syndrome of congenital ductal plate malformations of intrahepatic bile ducts characterized by segmental cystic dilatation of the intrahepatic ducts, increased incidence of biliary lithiasis, cholangitis and liver abscesses, absence of cirrhosis and portal hypertension, and association of renal cystic disease.76 Subsequent to Caroli’s initial reports two distinct forms of the disease have been recognized: (1) the simple type, which is associated with medullary sponge kidney in 60–80% of cases; and (2) the periportal fibrosis variant which is associated with congenital hepatic fibrosis, cirrhosis, portal hypertension, and esophageal varices.126
Clinical Features The most common presenting symptoms of Caroli’s disease are recurrent episodes of fever, chills, and abdominal pain due to cholangitis, with peak incidence in early adult life. Males and females are equally affected, in contrast to the female predominance of massive PCLD, simple hepatic cysts, and choledochal cysts. More than 80% present with symptoms before the age of 30 years. Rarely, the disease presents later in life, with evidence of portal hypertension and its complications, most commonly bleeding esophageal varices. The risk of development of cholangiocarcinoma in Caroli’s disease is about 7%.1 Biliary lithiasis is found in one-third and predisposes patients to recurrent episodes of cholangitis due to obstruction and ascending infections. Occasional patients also experience multiple liver abscesses. The molecular pathogenesis of Caroli’s disease is not fully understood. Caroli’s disease has been associated with mutations in PKHD1, the gene responsible for ARPKD.127 In rarer instances it has been described in the setting of PKD1 mutations, one of the two genes implicated in ADPKD.128 Other chromosomal mutations may be involved in inherited forms of Caroli’s disease not associated with renal cystic disease.129
Diagnosis Caroli’s disease is typically discovered by imaging modalities performed during evaluation of biliary obstruction or cholangitis.130 These studies typically demonstrate bile duct ectasia and nonobstructive saccular or fusiform dilatation of the intrahepatic bile ducts. Communication of the intrahepatic cysts with the biliary tree and an otherwise normal common bile duct is the key feature of the diagnosis and is usually confirmed by ultrasonography, scintigraphy, CT scan after biliary contrast, magnetic resonance cholangiopancreatography (MRCP), endoscopic retrograde cholangiopancreatography (ERCP), or percutaneous transhepatic cholangiography (PTC). PTC and ERCP provide detailed examination of the biliary tree and may aid in therapy (Figure 70-20). Magnetic resonance cholangiography with a dynamic contrast-enhanced study is an excellent
screening tool for Caroli’s disease and allows for the diagnosis to be made without invasive imaging of the biliary tree.131
Treatment Adequate biliary drainage is the primary approach in the management of Caroli’s disease. Endoscopic therapy with ERCP is effective in removing sludge or stones from the common bile duct but is of limited utility in providing adequate drainage of intrahepatic cysts. In contrast, PTC is more effective in draining these cysts and avoids recurrent episodes of cholangitis, especially if patients comply with periodic flushing and changing of drainage catheters. Rarely the cystic disease is confined to one hepatic lobe, and, in this circumstance, hepatic lobectomy may be curative. Although some have advocated hepaticojejunostomy after partial hepatectomy as primary therapy, the long-term efficacy of this procedure is uncertain, and the extensive surgery could compromise the outcome from hepatic transplantation. There are three indications for hepatic transplantation in Caroli’s disease: (1) hepatic decompensation; (2) recurrent cholangitis that is unresponsive to endoscopic or radiologic interventions; and (3) development of focal adenocarcinoma.132
CHOLEDOCHAL CYST Characteristics Choledochal cysts are usually diagnosed in childhood but may remain silent and undetected until adulthood in up to 50% of patients.133–135 Choledochal cysts are cystic dilatations that may occur throughout the macroscopic intra- and extrahepatic biliary tree. Although the term choledochal cyst has been used for any cystic dilatation of the biliary tree, isolated choledochal cysts are usually restricted to the common hepatic or bile duct. Despite the uncommon occurrence of choledochal cysts, there are hundreds of reports in the literature encompassing over 3000 cases. Choledochal cyst is a rare condition in the western hemisphere, but it is relatively more common among Japanese and other oriental populations. Several classifications of choledochal cysts have been proposed but the most commonly cited in the medical literature is by Todani et al.136 In this system, the most common is type I cyst, representing more than 80% of cases. The pattern of inheritance is unclear.
Clinical Features The most common clinical presentation of choledochal cyst is a relatively young patient (child or adolescent) with pain, mass in the right upper quadrant or epigastrium, and jaundice. In one series of 740 cases jaundice was the most common and consistent presenting feature. In infants, jaundice is often the only sign and the disorder may be difficult to distinguish from biliary atresia. The majority of the patients have been diagnosed before age 30 years and the male-to-female ratio in most series is about 1:4. Reported complications include spontaneous and traumatic rupture; rupture and increased levels of secondary bile acids may contribute to cyst metaplasia and carcinoma. The tumors may originate in different parts of the pancreatobiliary system, including liver, gallbladder, intrahepatic ducts, pancreatic ducts, and pancreas.
Diagnosis The diagnosis of a choledochal cyst should be suspected when a patient presents with recurrent abdominal pain, jaundice, raised
1347
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 70-20. Caroli’s disease. Cholangiogram demonstrating ectatic biliary ducts and saccular dilatations of both intrahepatic and extrahepatic ducts.
serum amylase, and cystic mass on imaging. Initial imaging of the biliary tree by ultrasonography or radioscintigraphy is usually diagnostic. Confirmation and anatomic definition may require CT, ERCP, or PTC (Figures 70-21 and 70-22). Patients with extrahepatic choledochal cysts have an increased incidence of anomalous pancreaticobiliary junction, which requires ERCP when planning for excision of the cyst. In recent years, MRCP has shown to be equivalent to ERCP in detecting and defining not only choledochal cysts but also the presence of an anomalous union of the pancreatic and bile ducts. MRCP can also define the length of extrahepatic bile duct involved by the cyst, which is important when planning a surgical approach. Particularly in the pediatric population MRCP offers an attractive non-invasive diagnostic alternative with the lack of ionizing radiation. Endoscopic ultrasonography has also been a usef imaging method for patients with suspected anomalous pancreaticobiliary junction. Prenatal ultrasonography can detect choledochal cysts in utero, which may help antenatal counseling since early neonatal cyst excision and duct revision may be required.
Treatment It is generally agreed that choledochal cysts require surgical treatment. The preferred surgical treatment is complete cyst excision with Roux en Y hepaticojejunostomy.133–138 This eliminates any opportunity for stasis, infection, stone formation, and possible
1348
development of cholangiocarcinoma. The procedure provides excellent long-term results with low morbidity and mortality, but lifelong follow-up may be necessary to avoid potential problems, such as biliary cirrhosis. Internal cyst drainage procedures (cystoduodenostomy, cystojejunostomy) have often been unsatisfactory with a complication rate as high as 50%, and this procedure may interfere with future transplantation.
SOLITARY HEPATIC CYST Characteristics Solitary hepatic cysts are relatively common, usually asymptomatic, and most often discovered incidentally during the evaluation of abdominal symptoms or disorders (Figure 70-23).139 Exact prevalence of solitary hepatic cysts for the US population is unknown, but the female-to-male gender ratio is approximately 4:1. In a European study of 26 000 patients undergoing diagnostic ultrasonography, the prevalence of solitary hepatic cyst was 2.8%.140 A Taiwanese study used ultrasonography in a large-scale screening program for simple hepatic cysts to explore age- and gender-specific prevalence.141 A total of 3600 subjects underwent screening ultrasonography, and 156 simple hepatic cysts in 132 study subjects were detected. The overall prevalence was 3.6%. Prevalence increased with age, ranging from 0.83% below age 40 to 7.81% in subjects over 60 years of age. The sizes of 219 hepatic cysts in 167 hospitalized
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER Figure 70-21. Choledochal cyst. Computed tomography demonstrated cystic dilation of common hepatic duct near the gallbladder fossae.
Figure 70-22. Choledochal cyst. Endoscopic retrograde cholangiopancreatography demonstrated Todani type I choledochal cyst.
1349
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 70-23. Solitary hepatic cyst. Computed tomography demonstrating large solitary cyst in left lobe of liver.
patients were measured; 53% had diameters between 1 and 3 cm, and only 7% were larger than 5 cm. Cysts occurred more commonly in the right lobe and were twice as prevalent in women. All of these cysts were asymptomatic and none of the patients suffered clinical consequences.
Treatment Asymptomatic solitary hepatic cysts are best managed conservatively. The preferred treatment of symptomatic cysts is ultrasoundor CT-guided percutaneous cyst aspiration followed by sclerotherapy.139,142–145 This approach is more than 90% effective in controlling symptoms and ablating the cyst cavity. The recurrence rate after successful ablation is only 5–15%. If the radiologically guided, percutaneous approach is ineffective or unavailable, treatment may include either laparoscopic or open surgical cyst fenestration. The laparoscopic approach is increasingly utilized for anatomically accessible cysts and greater than 90% efficacy is reported.
REFERENCES 1. Nakanuma Y, Hoso M, Sanzen T, Sasaki M. Microstructure and development of the normal and pathologic biliary tract in humans including blood supply. Microsc Res Tech 1997; 38:512. 2. Desmet VJ. Congenital disease of intrahepatic bile ducts: variations on the theme ductal plate malformation. Hepatology 1992; 16:1069–1083. 3. Shah KD, Gerber MA. Development of the intrahepatic bile ducts in humans. Arch Pathol Lab Med 1989; 113:1135–1138. 4. Jorgensen MJ. The ductal plate malformation. A study of the intrahepatic bile duct lesion in infantile polycystic disease and congenital hepatic fibrosis. Acta Pathol Microbiol Scand 1977; 257 (suppl):1–88. 5. Desmet VJ. What is congenital hepatic fibrosis? Histopathology 1992; 20:465–477.
1350
6. Awasthi A, Das A, Srinivasan R, Joshi K. Morphological and immunohistochemical analysis of ductal plate malformation: correlation with fetal liver. Histopathology 2004; 45:260– 267. 7. Sergi C, Adam S, Kahl P, Otto HF. Study of the malformation of ductal plate of the liver in Meckel syndrome and review of other syndromes presenting with this anomaly. Pediatr Dev Pathol 2000; 3:568–583. 8. Baert L. Hereditary polycystic kidney disease (adult form): a microdissection study of two cases at an early stage of the disease. Kidney Int 1978; 13:519–525. 9. Qian F, Watnick TJ, Onuchic LF, Germino GG. The molecular basis of focal cyst formation in human ADPKD type I. Cell 1996; 87:979–987. 10. Wu G, D’Agati V, Cai Y, et al. Somatic inactivation of Pkd2 results in polycystic kidney disease. Cell 1998; 93:177–188. 11. Colgin LM, Hackmann AFM, Emond MJ, Monnat RJJ. The unexpected landscape of in vivo somatic mutation in a human epithelial cell lineage. Proc Natl Acad Sci USA 2002; 99:1437–1442. 12. Praetorius HA, Spring KR. The renal cell primary cilium functions as a flow sensor. Curr Opin Nephrol Hyperten 2003; 12:517–520. 13. Masyuk TV, Huang BQ, Ward CJ, et al. Defects in cholangiocyte fibrocystin expression and ciliary structure in the PCK rat. Gastroenterology 2003; 125:1303–1310. 14. Masyuk TV, Huang BQ, Masyuk AI, et al. Biliary dysgenesis in the PCK rat, an orthologous model of autosomal recessive polycystic kidney disease. Am J Pathol 2004; 165:1719– 1730. 15. Gabow PA, Johnson AM, Kaehny WD, et al. Risk factors for the development of hepatic cysts in autosomal dominant polycystic kidney disease. Hepatology 1990; 11:1033–1037. 16. Everson GT. Hepatic cysts in autosomal dominant polycystic kidney disease. Am J Kidney Dis 1993; 22:520–525. 17. Everson G. Hepatic cysts in ADPKD. Mayo Clin Proc 1990; 65:1020–1025. 18. Fick GM, Gabow PA. Hereditary and acquired cystic disease of the kidney. Kidney Int 1994; 46:951–964.
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER
19. Hateboer N, van Dijk MA, Bogdanova M. Comparison of phenotypes of PKD types 1 and 2. European PKD1-PKD2 study group. Lancet 1999; 353:103–107. 20. Dalgaard OZ. Bilateral polycystic disease of the kidneys; a follow-up of 284 patients and their families. Dan Med Bull 1957; 4:128–133. 21. Reeders ST, Breuning MH, Davies KE, et al. A highly polymorphic DNA marker linked to adult polycystic kidney disease on chromosome 16. Nature 1985; 317:542–544. 22. Ward CJ, Turley H, Ong ACM, et al. Polycystin, the polycystic kidney disease 1 protein, is expressed by epithelial cells in fetal, adult, and polycystic kidney. Proc Natl Acad Sci USA 1996; 93:1524–1528. 23. Consortium IP. Polycystic kidney disease: the complete structure of the PKD1 gene and its protein. Cell 1995; 81:289–298. 24. Hughes J, Ward CJ, Peral B, et al. The polycystic kidney disease I (PKD1) gene encodes a novel protein with multiple cell recognition domains. Nat Genet 1995; 10:151–159. 25. Rossetti S, Strmecki L, Gamble V, et al. Mutation analysis of the entire PKD1 gene: genetic and diagnostic implications. Am J Hum Genet 2001; 68:46–63. 26. Watnick TJ, Phakdeekitcharoen B, Johnson A, et al. Mutation detection of PKD1 identifies a novel mutation common to three families with aneurysms and/or very-early-onset disease. Am J Hum Genet 1999; 65:1561–1571. 27. Rossetti S, Chauveau D, Kubly V, et al. Association of mutation position in polycystic kidney disease 1 (PKD1) gene and development of a vascular phenotype. Lancet 2003; 361:2196–2201. 28. Rossetti S, Burton S, Strmecki L, et al. The position of the polycystic kidney disease 1 (PKD1) gene mutation correlates with the severity of renal disease. J Am Soc Nephrol 2002; 13:1230–1237. 29. Weston BS, Malhas AN, Price RG. Structure–function relationships of the extracellular domain of the ADPKDassociated protein, polycystin-1. FEBS Lett 2003; 538:8–13. 30. Malhas AN, Abuknesha RA, Price RG. Interaction of the leucine-rich repeats of polycystin-1 with extracellular matrix proteins: possible role in cell proliferation. J Am Soc Nephrol 2002; 13:19–26. 31. Ibraghimov-Beskrovnaya O, Bukanov NO, Donohue LC, et al. Strong homophilic interactions of the Ig-like domains of polycystin-1, the protein product of an ADPKD gene, PKD1. Hum Mol Genet 2000; 9:1641–1649. 32. Qian F, Germino FJ, Cai Y, et al. PKD1 interacts with PKD2 through a probable coiled-coil domain. Nat Genet 1997; 16:179–183. 33. Gabow PA. Autosomal dominant polycystic kidney disease. N Engl J Med 1993; 329:332–342. 34. Mochizuki T, Wu G, Hayashi T, et al. PKD2, a gene for polycystic kidney disease that encodes an integral membrane protein. Science 1996; 272:1339–1342. 35. Deltas CC. Mutations of the human polycystic kidney disease 2 (PKD2) gene. Hum Mut 2001; 18:13–24. 36. Hateboer N, Veldhuisen B, Peters D, et al. Location of mutations within the PKD2 gene influences clinical outcomes. Kidney Int 2000; 57:1444–1451. 37. Gonzales-Perrett S, Kim K, Ibarra C, et al. Polycystin-2, the protein mutated in ADPKD, is a Ca2+-permeable nonselective cation channel. Proc Natl Acad Sci USA 2001; 98:1182–1187. 38. Gonzales-Perrett S, Battelli M, Kim K, et al. Voltage dependence and pH regulation of human polycystin-2-mediated cation channel activity. J Biol Chem 2002; 277:24959–24966. 39. Vassilev PM, Guo L, Chen X, et al. Polycystin-2 is a novel cation channel implicated in defective intracellular Ca2+ homeostasis in polycystic kidney disease. Biochem Biophys Res Commun 2001; 282:341–350.
40. Koulen P, Cai Y, Geng L, et al. Polycystin-2 is an intracellular calcium release channel. Nat Cell Biol 2002; 4:191–197. 41. Cantiello HF. Regulation of calcium signaling by polycystin-2. Am J Physiol 2004; 286:F1012–F1029. 42. Pazour GJ, Witman GB. The vertebrate primary cilium is a sensory organelle. Curr Opin Cell Biol 2003; 15:105–110. 43. Yoder BK, Hou X, Guay-Woodford LM. The polycystic kidney disease proteins, polycystin-1 and polycystin-2, polaris and cystin are co-localized in renal cilia. J Am Soc Nephrol 2002; 13:2508–2516. 44. Nauli SM, Alenghat FJ, Luo Y, et al. Polycystins 1 and 2 mediate mechanosensation in the primary cilium of kidney cells. Nature Genet 2003; 33:129–137. 45. Parnella SC, Magenheimera BS, Masera RL, et al. The polycystic kidney disease-1 protein, polycystin-1, binds and activates heterotrimeric G-proteins in vitro. Biochem Biophys Res Commun 1998; 251:625–631. 46. Delmas P, Nomura H, Li X, et al. Constitutive activation of Gproteins by polycystin-1 is antagonized by polycystin-2. J Biol Chem 2002; 277:11276–11283. 47. Arnould T, Kim E, Tsiokas L, et al. The polycystic kidney disease 1 gene product mediates protein kinase C-dependent and c-Jun N-terminal kinase-dependent activation of the transcription factor AP-1. J Biol Chem 1998; 273:6013–6018. 48. Arnould T, Sellin L, Benzing T, et al. Cellular activation triggered by the autosomal dominant polycystic kidney disease gene product PKD2. Mol Cell Biol 1999; 19:3423–3434. 49. Chauvet V, Tian X, Husson H, et al. Mechanical stimuli induce cleavage and nuclear translocation of the polycystin-1 C terminus. J Clin Invest 2004; 114:1433–1443. 50. Bhunia AK, Piontek K, Boletta A, et al. PKD1 Induces p21waf1 and regulation of the cell cycle via direct activation of the JAKSTAT signaling pathway in a process requiring PKD2. Cell 2002; 109:157–168. 51. Alvaro D, Alpini G, Onori P, et al. Alpha and beta estrogen receptors and the biliary tree. Mol Cell Endocrinol 2002; 193:105–108. 52. Everson G, Emmett M, Brown W, et al. Functional similarities of hepatic and biliary epithelium: studies of fluid constituents and in vivo secretion in response to secretin. Hepatology 1990; 11:557–565. 53. Perrone RD, Grubman SA, Rogers LC, et al. Continuous epithelial cell lines from ADPKD liver cysts exhibit characteristics of intrahepatic biliary epithelium. Am J Physiol 1995; 269:G335–G345. 54. Vandenburgh HH. Mechanical forces and their second messengers in stimulating cell growth in vitro. Am J Physiol Regul Integr Comp Physiol 1992; 262:R350–R355. 55. Tanner GA, McQuillan PF, Maxwell MR, et al. An in vitro test of the cell stretch-proliferation hypothesis of renal cyst enlargement. J Am Soc Nephrol 1995; 6:1230–1241. 56. Nichols MT, Gidey E, Matzakos T, et al. Secretion of cytokines and growth factors into ADPKD liver cyst fluid. Hepatology 2004; 40:836–846. 57. Pirson Y, Lannoy N, Peters D, et al. Isolated polycystic liver disease as a distinct genetic disease, unlinked to polycystic kidney disease 1 and polycystic kidney disease 2. Hepatology 1996; 23:249–252. 58. Reynolds DM, Falk CT, Li A, et al. Identification of a locus for autosomal dominant polycystic liver disease, on chromosome 19p13.2-13.1. Am J Hum Genet 2000; 67:1598–1604. 59. Li A, Davila S, Furu L, et al. Mutations in PRKCSH cause isolated autosomal dominant polycystic liver disease. Am J Hum Genet 2003; 72:691–703. 60. Drenth JPH, te Morsche RHM, Smink R, et al. Germline mutations in PRKCSH are associated with autosomal dominant polycystic liver disease. Nat Genetic 2003; 33:345–347.
1351
Section XII. Inherited and Pediatric Diseases of the Liver
61. Li A, Davila S, Furu L, et al. Mutations in PRKCSH cause isolated autosomal dominant polycystic liver disease. Am J Hum Genet 2003; 72:691–703. 62. Drenth JPH, te Morsche RHM, Smink R, et al. Germline mutations in PRKCSH are associated with autosomal dominant polycystic liver disease. Nat Genetic 2003; 33:345–347. 63. Drenth JPH, Tahvanainen E, te Morsche RHM, et al. Abnormal hepatocystin caused by truncating PRKCSH mutations leads to autosomal dominant polycystic liver disease. Hepatology 2004; 39:924–931. 64. Davila S, Furu L, Gharavi AG, et al. Mutations in SEC63 cause autosomal dominant polycystic liver disease. Nat Genet 2004; 36:575–577. 65. Tahvanainen P, Tahvanainen E, Reijonen H, et al. Polycystic liver disease is genetically heterogeneous: clinical and linkage studies in eight Finnish families. J Hepatol 2003; 38:39–43. 66. Zerres K, Volpel MC, Weiss H. Cystic kidneys. Genetics, pathologic anatomy, clinical picture, and prenatal diagnosis. Human Genet 1984; 68:104–135. 67. Zerres K, Mücher G, Becker J, et al. Prenatal diagnosis of autosomal recessive polycystic kidney disease (ARPKD): molecular genetics, clinical experience, and fetal morphology. Am J Med Genet 1998; 76:137–144. 68. Summerfield JA, Nagafuchi Y, Sherlock S, et al. Hepatobiliary fibropolycystic diseases. A clinical and histological review of 51 patients. J Hepatol 1986; 2:141–156. 69. Gabow P, Johnson A, Kaehney W, et al. Risk factors of the development of hepatic cysts in autosomal dominant polycystic kidney disease. Hepatology 1990; 11:1033–1037. 70. Perrone RD, Grubman SA, Rogers LC, et al. Continuous epithelial cell lines from ADPKD liver cysts exhibit characteristics of intrahepatic biliary epithelium. Am J Physiol 1995; 269 (Gastrointest Liver Physiol 32):G335–G345. 71. Terada T, Nakanuma Y, Ohta T, et al. Mucin-histochemical and immunohistochemical profiles of epithelial cells of several types of hepatic cysts. Virchows Arch 1991; A419:499–504. 72. Sasaki M, Nakanuma Y. Abnormal expression of MUC1 apomucin and mature MUC1 mucin in biliary epithelial cells in various cystic liver diseases. Hepatology 1996; 24:539–543. 73. Sasaki M, Katayanagi K, Watanabe K, et al. Intrahepatic cholangiocarcinoma arising in autosomal dominant polycystic kidney disease. Virchows Arch 2002; A441:98–100. 74. Averback P. Congenital hepatic fibrosis: asymptomatic adults without renal anomaly. Arch Pathol Lab Med 1977; 101:260–261. 75. Zeitoun D, Brancatelli G, Colombat M, et al. Congenital hepatic fibrosis: CT findings in 18 adults. Radiology 2004; 231:109–116. 76. Caroli J. Disease of the intrahepatic biliary tree. Clin Gastroenterol 1972; 2:147–161. 77. Nakanuma Y, Terada T, Ohta G, et al. Caroli’s disease in congenital hepatic fibrosis and infantile polycystic disease. Liver 1982; 2:346–354. 78. Marchal GJ, Desmet VJ, Proesmans MD, et al. Caroli disease: high frequency US and pathological findings. Radiology 1986; 158:507–511. 79. Levy, AD, Rohrmann CA Jr, Murakata LA, Lonergan GJ. Caroli’s disease: radiologic spectrum with pathologic correlation. Am J Radiol 2002; 179:1053–1057. 80. Alonso-Lej F, Rever WB, Pessagno DJ. Congenital choledochal cyst with a report of two and an analysis of 94 cases. Surg Gynecol Obstet 1959; 108:1–30. 81. Schimpl G, Sauer H, Goriupp U, Becker H. Choledochocele: importance of histological evaluation. J Pediatr Surg 1993; 28:1562–1565. 82. Komi N, Tamura T, Tsuge S, et al. Relation of patient age to premalignant alterations in choledochal cyst epithelium: histochemical and immunohistochemical studies. J Pediatr Surg 1986; 21:430– 433.
1352
83. Reveille RM, Van Stiegmann G, Everson GT. Increased secondary bile acids in a choledochal cyst. Gastroenterology 1990; 99:525–527. 84. Voyles CR, Smadja C, Shands C, Blumgart LH. Carcinoma in choledochal cysts: age related incidence. Arch Surg 1983; 118:986–988. 85. Buffet C, Hagege H. Non-parasitic cysts of the liver. Presse Med 1993; 22:1823–1826. 86. Nardello O, Muggianu M, Cabras V, et al. Dysplastic cysts of the liver: our experience. Minerva Chir 2004; 59:351–362. 87. Eriguchi N, Aoyagi S, Tamae T, et al. Treatments of non-parasitic giant hepatic cysts. Kurume Med J 2001; 48:193–195. 88. Iwase K, Takenaka H, Oshima S, et al. Determination of tumor marker levels in cystic fluid of benign liver cysts. Dig Dis Sci 1992; 37:1648–1654. 89. Harris RA, Gray DW, Britton BJ, et al. Hepatic cystic disease in an adult polycystic kidney disease transplant population. Aust NZ J Surg 1996; 66:166–168. 90. Everson GT. Hepatic cysts in autosomal dominant polycystic kidney disease. Am J Kidney Dis 1993; 22:520–525. 91. Everson G. Hepatic cysts in ADPKD. Mayo Clin Proc 1990; 65:1020 –1025. 92. Shrestha R, Mckinley C, Russ P, et al. Postmenopausal estrogen therapy selectively stimulates hepatic enlargement in women with autosomal dominant polycystic kidney disease. Hepatology 1997; 26:1282–1286. 93. Pirson Y, Lannoy N, Peters D, et al. Isolated polycystic liver disease as a distinct genetic disease, unlinked to polycystic kidney disease 1 and polycystic kidney disease 2. Hepatology 1996; 23:249–252. 94. Shrestha R, Mckinley C, Russ P, et al. Postmenopausal estrogen therapy selectively stimulates hepatic enlargement in women with autosomal dominant polycystic kidney disease. Hepatology 1997; 26:1282–1286. 95. Qian Q, Li A, King BF, et al. Clinical profile of autosomal dominant polycystic liver disease. Hepatology 2003; 37:164–171. 96. Rossetti S, Strmecki L, Gamble V, et al. Mutation analysis of the entire PKD1 gene: genetic and diagnostic implications. Am J Hum Genet 2001; 68:46–63. 97. Rossetti S, Burton S, Strmecki L, et al. The position of the polycystic kidney disease 1 (PKD1) gene mutation correlates with the severity of renal disease. J Am Soc Nephrol 2002; 13:1230–1237. 98. Deltas CC. Mutations of the human polycystic kidney disease 2 (PKD2) gene. Hum Mut 2001; 18:13 99. Thomas R, McConnel R, Whittacker J, et al. Identification of mutations in the repeated part of the ADPKD type 1 gene, PKD1, by long-range PCR. Am J Hum Genet 1999; 65: 39–49. 100. Everson GT, Taylor MR, Doctor RB. Polycystic liver disease. Hepatology 2004; 40:774–782. 101. Everson GT, Taylor MR. Management of polycystic liver disease. Curr Gastroenterol Rep 2005; 7:19–25. 102. Chauveau D, Fakhouri F, Grunfeld J-P. Liver involvement in autosomal-dominant polycystic kidney disease: therapeutic dilemma. J Am Soc Nephrol 2000; 11:1767–1775. 103. Perrone RD, Ruthazer R, Terrin NC. Survival after end-stage renal disease in autosomal dominant polycystic kidney disease: contribution of extrarenal complications to mortality. Am J Kidney Dis 2001; 38:777–784. 104. Everson GT, Scherzinger A, Berger-Leff N, et al. Polycystic liver disease: quantitation of parenchymal and cyst volumes from computed tomography images and clinical correlates of hepatic cysts. Hepatology 1988; 8:1627–1634. 105. Telenti A, Torres VE, Gross JB, et al. Hepatic cyst infection in autosomal dominant polycystic kidney disease. Mayo Clin Proc 1990; 65:933–942.
Chapter 70 FIBROCYSTIC DISEASES OF THE LIVER
106. Uddin W, Ramage JK, Portmann B, et al. Hepatic venous outflow obstruction in patients with polycystic liver disease: pathogenesis and treatment. Gut 1995; 36:142–145. 107. Torres VE, Rastogi S, King BF, et al. Hepatic venous outflow obstruction in autosomal dominant polycystic kidney disease. J Am Soc Nephrol 1994; 5:1186–1192. 108. Chapman AB, Rubinstein D, Hughes R, et al. Intracranial aneurysms in autosomal dominant polycystic kidney disease. N Engl J Med 1992; 327:916–920. 109. Chauveau D, Martinez F, Grunfeld JP. Evaluation of octreotide in massive polycystic liver disease. 12th International Congress of Nephrology Conference Program. Jerusalem, Israel: 1993:487A. 110. van Sonnenberg E, Wroblicka JT, D’Agostino HB, et al. Symptomatic hepatic cysts: percutaneous drainage and sclerosis. Radiology 1994; 190:387–392. 111. Tikkakoski T, Makela JT, Leinonen S, et al. Treatment of symptomatic congenital hepatic cysts with single-session percutaneous drainage and ethanol sclerosis: technique and outcome. J Vasc Intervent Radiol 1996; 7:235–239. 112. Gigot JF, Jadoul P, Que F, et al. Adult polycystic liver disease: is fenestration the most adequate operation for long-term management? Ann Surg 1997; 225:286–294. 113. Kabbej M, Sauvanet A, Chauveau D, et al. Laparoscopic fenestration in polycystic liver disease. Br J Surg 1996; 83:1697–1701. 114. Morino M, De Guili M, Festa V, Garrone C. Laparoscopic management of symptomatic nonparasitic cysts of the liver. Indications and results. Ann Surg 1994; 219:157–164. 115. Robinson TN, Stiegmann GV, Everson GT. Laparoscopic palliation of polycystic liver disease. Surg Endosc 2005; 19:130–132. 116. Que F, Nagorney DM, Gross JBJ, Torres VE. Liver resection and cyst fenestration in the treatment of severe polycystic liver disease. Gastroenterology 1995; 108:487–494. 117. Klupp J, Bechstein WO, Lobeck H, Neuhaus P. Orthotopic liver transplantation in therapy of advanced polycystic liver disease. Chirurg 1996; 67:515–521. 118. Washburn WK, Johnson LB, Lewis WD, Jenkins RL. Liver transplantation for adult polycystic liver disease. Liver Transpl Surg 1996; 2:17–22. 119. Lang H, Woellwarth JV, Oldhafer KJ, et al. Liver transplantation in patients with polycystic liver disease. Transplant Proc 1997; 29:2832–2833. 120. Swenson K, Seu P, Kinkhabwala M, et al. Liver transplantation for adult polycystic disease. Hepatology 1998; 28:412–415. 121. Pirenne J, Aerts R, Yoong K, et al. Liver transplantation for polycystic liver disease. Liver Transpl 2001; 7:238–245. 122. Gustafsson BI, Friman S, Mjornstedt L, et al. Liver transplantation for polycystic liver disease – indications and outcomes. Transplant Proc 2003; 35:813–814. 123. Onuchic LF, Furu L, Nagasawa Y, et al. PKHD1, the polycystic kidney and hepatic disease 1 gene, encodes a novel large protein containing multiple immunoglobulin-like plexin-transcriptionfactor domains and parallel beta-helix 1 repeats. Am J Hum Genet 2002; 70:1305–1317. 124. Desmet VJ. Ludwig symposium on biliary disorders – part I. Pathogenesis of ductal plate abnormalities. Mayo Clin Proc 1998; 73:80–89. 125. De Vos M, Barbier F, Cuvelier C. Congenital hepatic fibrosis. J Hepatol 1988; 6:222–228.
126. Taylor AC, Palmer KR. Caroli’s disease. Eur J Gastroenterol Hepatol 1998; 10:105–108. 127. Sgro M, Rossetti S, Barozzino T, et al. Caroli’s disease: prenatal diagnosis, postnatal outcome and genetic analysis. Ultrasound Obstet Gynecol 2004; 23:73–76. 128. Torra R, Badenas C, Darnell A, et al. Autosomal dominant polycystic kidney disease with anticipation and Caroli’s disease associated with a PKD1 mutation. Kidney Int 1997; 52:33–38. 129. Parada LA, Hallen M, Hagerstrand I, et al. Clonal chromosomal abnormalities in congenital bile duct dilatation (Caroli’s disease). Gut 1999; 45:780–782. 130. Miller WJ, Sechtin AG, Campbell WL, Pieters PC. Imaging findings in Caroli’s disease. Am J Roentgenol 1995; 165:333–337. 131. Guy F, Cognet F, Dranssart M, et al. Caroli’s disease: magnetic resonance imaging features. Eur Radiol 2002; 12:2730–2736. 132. Waechter FL, Sampaio JA, Pinto RD, et al. The role of liver transplantation in patients with Caroli’s disease. HepatoGastroenterology 2001; 48:672–674. 133. Jordan PH, Goss JA Jr, Rosenberg WR, Woods KL. Some considerations for management of choledochal cysts. Am J Surg 2004; 187:790–795. 134. Visser BC, Suh I, Way LW, Kang S-M. Congenital choledochal cysts in adults. Arch Surg 2004; 139:855–862. 135. Lee KF, Lai ECH, Lai PBS. Adult choledochal cyst. Asian J Surg 2005; 28:29–33. 136. Todani T, Watanabe Y, Narusue M, et al. Congenital bile duct cysts: classification, operative procedures and review of 37 cases including cancer arising from choledochal cyst. Am J Surg 1977; 134:263–269. 137. Rha SY, Stovroff MC, Glick PL, et al. Choledochal cysts: a ten year experience. Am Surg 1996; 62:30 –34. 138. Plata-Munoz JJ, Mercado MA, Chan C, et al. Complete resection of choledochal cyst with Roux-en-Y derivation vs cystenterostomy as standard treatment of cystic disease of the biliary tract in the adult patient. Hepatogastroenterology 2005; 52:13–16. 139. Cowles RA, Mulholland MW. Solitary hepatic cysts. J Am Coll Surg 2000; 191:311–321. 140. Caremani M, Vincenti A, Benci A, et al. Ecographic epidemiology of non-parasitic hepatic cysts. J Clin Ultrasound 1993; 21:115–118. 141. Huang JF, Chen SC, Lu SN, et al. Prevalence and size of simple hepatic cysts in Taiwan: community- and hospital-based sonographic surveys. Kao-Hsiung i Hsueh Ko Hsueh Tsa Chih [Kaohsiung J Med Sci] 1995; 11:564–567. 142. Tikkakoski T, Makela JT, Leinonen S, et al. Treatment of symptomatic congenital hepatic cysts with single-session percutaneous drainage and ethanol sclerosis: technique and outcome. J Vasc Intervent Radiol 1996; 7:235–239. 143. Pozniczek M, Wysocki A, Bobrzynski A, et al. Sclerosant therapy as first-line treatment for solitary liver cysts. Dig Surg 2004; 21:452–454. 144. Ammori BJ, Jenkins BL, Lim PC, et al. Surgical strategy for cystic diseases of the liver in a western hepatobiliary center. World J Surg 2002; 26:462–469. 145. Tagaya N, Nemoto T, Kubota K. Long-term results of laparoscopic unroofing of symptomatic solitary nonparasitic hepatic cysts. Surg Laparosc Endosc Percutan Tech 2003; 13:76–79.
1353
Section XII. Inherited and Pediatric Diseases of the Liver
71
PEDIATRIC CHOLESTATIC SYNDROMES Diana M. Flynn and Deirdre A. Kelly Abbreviations a1AT a1-antitrypsin deficiency ABC ATP-binding cassette AGS alagille syndrome ASBT apical sodium-dependent bile acid transporter ATP adenosine triphosphate ATPase adenosine triphosphatase BCAA branched-chain amino acids BEC biliary epithelial cells BRIC benign recurrent intrahepatic cholestasis CMV cytomegalovirus
EHBA ERCP FIC1 HDN HLA ICAM-1 JAG1 MARS MCT MDR3
extrahepatic biliary atresia endoscopic retrograde cholangiopancreatography familial intrahepatic cholestasis-1 hemorrhagic disease of the newborn human leukocyte antigen intercellular adhesion molecule jagged1 gene molecular absorbent recirculating system medium-chain triglycerides multidrug resistance 3 gene
INTRODUCTION Neonatal jaundice is common, and is transient in most normal infants. In the majority it is related to the immaturity of glucuronosyl transferase or to breast-feeding.1 Conjugated hyperbilirubinemia, on the other hand, is always significant, and due to one of several underlying disorders. The most important of these is extrahepatic biliary atresia (EHBA), the leading cause of morbidity and mortality of all the childhood liver diseases.2 However, there are several other important causes of cholestasis presenting in neonates that also benefit from early diagnosis. It is therefore important to distinguish between jaundice due to normal variation, and that due to disease. All infants with jaundice that occurs or persists after the age of 14 days require investigation, so that cholestasis can be recognized and appropriate investigations and management instigated promptly where necessary. This is especially pertinent for EHBA, where early surgical treatment is vital. Recent advances have been made in our understanding of the etiology and pathogenesis of many causes of cholestasis. This, together with improved investigative techniques and management strategies, has improved not only the accuracy of diagnosis, but also the outcome for many children. This is of major importance as there are a number of cholestatic syndromes that present in infancy which lead to significant childhood illness, and in some cases end-stage liver disease. These can be divided into disorders of primarily extrahepatic or intrahepatic nature (Table 71-1). Extrahepatic disorders include EHBA, choledochal cyst, spontaneous perforation of the bile duct, bile or mucous plugs, or indeed a mass compressing the extrahepatic biliary tree. Intrahepatic ones include bile duct paucity syndromes, persistent familial intrahepatic cholestases, inborn errors of bile metabolism, anatomical disorders such as Caroli’s disease, and certain metabolic disorders, including a1-antitrypsin deficiency
MMR MRP2 NCAM NSC PCR PFIC PN PTC PUFA RT VDRL
measles, mumps, and rubella multidrug resistance-associated protein neural cell adhesion molecule neonatal sclerosing cholangitis polymerase chain reaction progressive familial intrahepatic cholestasis parenteral nutrition percutaneous transhepatic cholangiography polyunsaturated fatty acid reverse transcriptase venereal Disease Research Laboratory
(a1AT) and cystic fibrosis. The most common intrahepatic disorders come under the heading of neonatal hepatitis, for which many causes, often infectious, have now been elucidated, although idiopathic hepatitis remains a significant subgroup. In this chapter the commoner causes of cholestatic syndromes will be discussed.
CLINICAL PRESENTATIONS Cholestasis presenting in early infancy may be due to one of several causes. Diagnosis is usually not possible on clinical grounds alone, although the history and examination are important in differentiating between major types of disorder (Table 71-1). The most pressing need is to distinguish EHBA from other causes of cholestasis, most often from neonatal hepatitis. In EHBA the baby is most often born at term with normal birth weight, whereas the child with neonatal hepatitis or a1AT is often of low birth weight, and may be premature. Jaundice may be present from soon after birth and certainly by 4 weeks of age, associated with the development of completely acholic (clay-colored) stools. In some babies associated situs inversus, cardiac or renal abnormalities may also alert the clinician to the possibility of EHBA, whereas the abnormal facies of Alagille syndrome (AGS) may be present in both infant and parents. A family history of cardiac or renal problems points to AGS as opposed to EHBA. Other dysmorphic features are suggestive of neonatal hepatitis. Timing of the onset of jaundice, whether present from birth or late onset, is important. Intermittent jaundice is seen in familial intrahepatic cholestasis and choledochal cysts. A detailed antenatal and birth history revealing an infectious episode or fever during pregnancy is sometimes seen in neonatal hepatitis. A family history, including consanguinity, previous neonatal or in utero deaths, is more indicative of a possible metabolic dis-
1355
Section XII. Inherited and Pediatric Diseases of the Liver
Table 71-1. Causes of infantile cholestatic syndromes I. A. B. C. D. E. F. G. H. I. J. K. L. M. N.
Extrahepatic abnormalities Extrahepatic biliary atresia (EHBA) Choledochal cyst Caroli disease Neonatal sclerosing cholangitis Spontaneous perforation of bile duct Bile plug syndrome/inspissated bile syndrome Agenesis of the extrahepatic bile ducts Bile duct stenosis Duplication of the biliary tree Bile duct tumors Agenesis of the extrahepatic bile ducts Gallstones Bile duct tumors Extramural compression of the common bile duct
II. Intrahepatic disorders A. Neonatal hepatitis syndrome a. Intrauterine infection i. Toxoplasmosis ii. Rubella iii. Cytomegalovirus iv. Herpes simplex v. Varicella zoster vi. Syphilis vii. Parvovirus B19 viii. Paramyxovirus ix. Coxsackie virus x. Reovirus type 3 xi. Coxsackievirus xii. Echovirus xiii. Adenovirus xiv. Human immunodeficiency virus (HIV) xv. Human herpesvirus 6 (HHV) xvi. Hepatitis B virus xvii. Listeriosis xviii. Tuberculosis b. Endocrine i. Hypopituitarism ii. Hypothyroidism
order. Cardiac or renal problems in family members may indicate AGS. Examination often reveals pale stools and dark urine. Completely acholic stools indicate EHBA, although occasionally those with AGS present in a very similar way and careful investigation is required to distinguish the two to avoid unnecessary laparotomy. Hepatosplenomegaly is seen in late presentations of EHBA, but earlier in neonatal hepatitis, especially where there is an infectious etiology. An intra-abdominal mass may be present if there is a choledochal cyst.
INVESTIGATIONS As the presentation for many of these disorders is similar, investigation needs to exclude both serious and common causes of cholestasis (Table 71-1).
1356
B.
C.
D.
E.
.
c. Chromosomal i. Trisomy 18 ii. Trisomy 13 iii. Trisomy 21 iv. Cat-eye syndrome d. Idiopathic Bile duct paucity a. Alagille syndrome b. Non-syndromic paucity of bile ducts c. Hereditary cholestasis with lymphedema (Aagenaes) Persistent familial intrahepatic cholestasis (PFIC) a. PFIC1 b. BRIC c. PFIC2 d. PFIC3 Inborn errors of bile acid metabolism a. 3b-D5-C27-hydroxysteroid oxoreductase deficiency b. D4-3-Oxosteroid-5b-reductase deficiency c. 3b-Hydroxy-D5-steroid dehydrogenase isomerase deficiency d. Zellweger’s syndrome (cerebrohepatorenal syndrome) and other peroxisomal disorders e. Microfilament dysfunction Metabolic disorders 1. Disorders of lipid metabolism a. Wolman’s disease b. Niemann–Pick C disease c. Gaucher’s disease 2. Disorders of carbohydrate metabolism a. Galactosemia b. Fructosemia c. Glycogenosis III/IV 4. Metabolic disease (uncharacterized defect) a. a1-Antitrypsin deficiency (a1AT) b. Cystic fibrosis c. Familial erythrophagocytic lymphohistiocytosis F. Intestinal failure-associated liver disease
EXTRAHEPATIC DISORDERS EXTRAHEPATIC BILIARY ATRESIA Incidence The incidence of EHBA was reported in 1963 to be 1:25 000. In recent reviews the incidence was reported to be 1:20 000 in metropolitan France,3 1:16 700 in England,4 1:3500 live births in French Polynesia,3 and between 1 in 2500 and 1in 8000 in Soweto.5 EHBA occurs more frequently in girls than in boys.6
Etiology EHBA is defined as an idiopathic, localized, complete obliteration or discontinuity of the hepatic or common bile ducts at any point in the extrahepatic biliary tract from the porta hepatis to the duodenum, resulting in complete obstruction to bile flow.6 The earliest
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
description of EHBA, made by Professor John Burns in 1817, was of a disease process caused by “an incurable state of the biliary apparatus” leading to a “deficiency of bile, jaundice and acholic stools.” By the end of the 19th century Thomson concluded, in a review of 49 patients with EHBA, that it represented a progressive inflammatory lesion of bile ducts, of unknown cause. Alternative ideas at this time regarding etiology included descending cholangitis or a developmental anomaly. Other suggestions have included congenital syphilis, fetal peritonitis, and erythroblastosis fetalis. The etiology remains obscure. It is unlikely to be an inherited disorder, except perhaps in a very small minority of cases.6 Human leukocyte antigen (HLA)-identical twins are seen who are discordant for EHBA, and recurrence within families is exceedingly rare.7 A single study reported an increase in HLA-B12 in biliary atresia three times that of controls in a European population, but these findings have not been replicated. Associations with HLACw4/78 and, in Japan, with A33, B44, and DR69 have been reported. However, two further studies did not detect any HLA association. The most recent study reported a significant increase in HLA-B8 and DR3 in 18 Egyptian children with EHBA.10 This is of particular interest, as both of these HLA subtypes are associated with primary sclerosing cholangitis and inflammatory bowel disease. Environmental toxins have produced a similar picture to EHBA in animals, for example in sheep, but no evidence for this has been found in humans. The subgroup of patients who present with associated extrahepatic abnormalities may represent a distinct etiopathogenesis from the isolated biliary atresia group. Other rare associations with EHBA have also been noted. Several studies looking for the presence of viral agents have been carried out. Viruses that may be etiologically implicated in EHBA include reovirus and rotavirus C infection, and have been suggested to provoke an immune reaction with an inflammatory response. Reovirus RNA was first suggested, as infection in weanling mice resulted in a similar picture to EHBA, and lesions persisted after infectious virus or viral antigens were no longer detected.11 Reovirus has been detected in 55% of snap-frozen liver or bile duct remnants in patients with biliary atresia compared to 12% of controls and 21% of patients with other hepatobiliary disorders using nested polymerase chain reaction (PCR),12 although a recent study has refuted this, examining hepatobiliary as well as stool samples using nested reverse transcriptase (RT)-PCR from 26 patients with EHBA.13 Rotavirus C is also of the reoviridae family. RT-PCR for rotavirus C in hepatobiliary tissues showed positivity in 50% of EHBA and none of the control livers14 and a mouse model of EHBA induced by rotavirus infection has also been developed,15 further supporting the idea that viral infection plays a role in the etiopathogenesis of this condition. However, studies by other groups have found no association with either reovirus or rotavirus infection.13 Recent studies have also implicated cytomegalovirus (CMV) in the pathogenesis of EHBA.16 Other viruses investigated include human papillomavirus and human herpes virus 6, also with conflicting results.17 Overall, despite considerable efforts to isolate viruses, there is as yet no overwhelming evidence to support any one viral agent as causal for EHBA. About 20% of patients with EHBA have associated anomalies or malformations.18 These are divided into three groups: one (29%) with anomalies within the laterality sequence, including polysplenia
or asplenia, cardiovascular defects, abdominal situs inversus, intestinal malrotation, and anomalies of the portal vein and hepatic artery; the second (59%) with single or dual anomalies involving the cardiac, gastrointestinal, and urinary systems not following a recognizable pattern; and a third, smaller group (12%), with intestinal malrotation and some similarity to the laterality sequence group. This last group may represent a more limited phenotypical result of faulty situs determination.18 Davenport et al.19 studied the splenic malformation syndrome, analyzing the case records of 308 infants with EHBA for extrahepatic anomalies. Twenty-three (7.5%) had polysplenia, four had other splenic malformations (two with double spleen and two with asplenia). The presence of other anomalies such as situs inversus and portal vein anomalies suggested that they formed part of a larger association, for which they proposed the term biliary atresia splenic malformation syndrome, as a distinct subgroup of patients with EHBA, tending to have a worse prognosis, and requiring special attention to vascular anatomy at transplantation. Other reported extrahepatic anomalies in EHBA include intestinal malrotation with partial abdominal heterotaxia and craniofacial anomalies; esophageal, duodenal, and pancreatic atresia; anorectal and esophageal atresia; and Kabuki syndrome. EHBA with abnormalities of laterality might prove a suitable candidate for a major gene mutation, but data so far are limited. A mutation in the inversin (inv) gene in the mouse leads to anomalous development of the hepatobiliary system.20 The human inv gene has recently been mapped, but no consistent mutations were found in 64 patients with heterotaxia, and specifically there were no mutations in seven patients with EHBA and various congenital laterality defects.21 Mutations in the CFC1 gene encoding for the human CRYPTIC protein have also been examined. The precise function of CRYPTIC is unknown, but it appears to be a cofactor in the Nodal pathway, determining left–right axis development. Bamford et al.22 studied 144 patients with laterality defects. Of nine patients with heterozygous mutations in the CRYPTIC gene, one had EHBA and polysplenia syndrome. Jacquemin et al.23 further identified two brothers with laterality deects, one with EHBA, who had heterozygous gene mutations in CFC1. However, although these findings are interesting, they do not represent the major underlying cause for EHBA, and teratogenic, infectious, and polygenic multifactorial causes may play a more significant role in EHBA associated with “non-syndromic” organ system anomalies. Morphological abnormalities, due to defective development of the biliary tree, have also been suggested to play a role in the development of EHBA. The occurrence of ductal plate malformation of the intrahepatic bile ducts has been reported in up to 38% of patients with EHBA.24 Early severe forms of EHBA suggest an antenatal onset of the disease process. Developmental anomalies of the intrahepatic bile ducts in EHBA include a report of EHBA in association with Caroli’s disease and the combination of common hepatic duct stricture (possibly a forme fruste of EHBA) and histologic congenital hepatic fibrosis. It has been proposed that EHBA may be caused by failure of remodeling of the ductal plate at the liver hilum, with the persistence of fetal bile ducts poorly supported by mesenchyme.25 Bile leakage from these abnormal ducts may trigger an inflammatory reaction, with subsequent obliteration of the biliary tree. Hepatic innervation is also abnormal in EHBA: neural cell
1357
Section XII. Inherited and Pediatric Diseases of the Liver
adhesion molecule (NCAM)-positive and S100-positive nerve fibers were found to be increased near branches of the hepatic artery and portal vein, whereas no nerve fibers were observed around bile ducts and periportal ductules that themselves became NCAM-positive.26 The possibility of cell signaling proteins and transcription factors involved in bile duct development being altered in EHBA has also begun to be explored. Jagged1, a ligand for the Notch signaling pathway, involved in cell fate decisions and regulating cellular differentation and proliferation, is known to be mutated in AGS. Recent studies suggest it may also play a role as a modifying factor in EHBA.27,28 This may be via its effect on Hes1. Hes1 is a basic helix-loop-helix protein encoded by the gene Hes1, whose expression is controlled by the Notch pathway. Hes1 is expressed in the extrahepatic biliary epithelium throughout development. A recent study has demonstrated that Hes1-deficient mice have gallbladder agenesis and severe hypoplasia of extrahepatic bile ducts.29 Another basic helix-loop-helix protein regulated by Notch is Hex, which is important in the regulation of liver development in the zebrafish. Murine knockouts of hnf6, a transcription factor also important in early liver development, cause a ductal plate malformation associated with abnormalities of the hepatic artery, highlighting a link in vascular and biliary development.30 There is very close proximity of newly formed bile ducts in utero to hepatic artery branches, and it has been suggested that a vasculopathy may be the primary lesion in patients with EHBA, related to insufficient vascularization of the biliary tree. In support of this, experimental atresia of bile ducts can be induced in fetal rabbits and sheep by ligation of the hepatic artery or its branches. Definite proof for an ischemic origin in human EHBA, however, is lacking. In several cholestatic liver diseases there is marked formation of abnormal ductular reactive cells. In EHBA this is especially prominent with an increase in cells similar to those seen in the formation of the ductal plate in utero.6 Cytokeratin studies have been performed to determine the nature of bile duct formation, and compared to patterns of cytokeratin expression in EHBA. These have demonstrated that ductular reactive cells in EHBA have an immature phenotype, and are flattened and smaller than mature biliary epithelial cells (BEC), staining not only for cytokeratins 8, 18, and 19, but also for cytokeratin 7, which is normally only present on developing BEC from 20 weeks’ gestation until shortly after birth.31 Ductular reactive cells occurring in cholestasis also display certain neuroendocrine features, including the presence of the NCAM, not seen on normal BEC, but also associated with a more primitive phenotype.32 This further highlights the possibility of abnormalities in ductal plate formation in EHBA. Progression of disease in EHBA seems to involve an immunologically mediated mechanism of recruitment and activation of various immune factors, including T-lymphocyte and macrophage activation, which may have a role in continued inflammation. Both CD4+ lymphocytes and CD56+ natural killer cells are increased in the liver of patients with EHBA.33,34 Increased macrophage infiltration may also be associated with poorer outcome.33 Other abnormalities seen in EHBA include up-regulation of intercellular adhesion molecule (ICAM-1) on ductular reactive cells with aberrant expression of the major histocompatibility complex type I antigens.34 Using gene array technology, Bezerra et al.35 have proposed a Th1 response in EHBA, suggested by up-regulation of genes
1358
encoding products involved in lymphocyte differentiation and several that regulate the Th1 reponse, compared with biopsies from patients with intrahepatic neonatal cholestasis. The effects of immunosuppresion given to patients with EHBA are being evaluated as a further possible medical treatment for these patients. In conclusion, EHBA probably represents a common phenotype induced by diverse triggering and pathogenetic mechanisms. These are presumably multifactorial in most instances. The resulting obliteration of variable segments of the biliary tree leads to cholestasis, which involves retention of potentially toxic hydrophobic bile salts and proliferation of reactive ductules producing several cytokines.6 Retention of chenodeoxycholic acid (or other toxic bile acids) induces hepatocyte apoptosis and necrosis. Bile acids appear to impact on mitochondrial function (with altered oxidative metabolism and release of oxygen free radicals). It is proposed that the sequence of mitochondrial injury, oxidant stress, adenosine triphosphate depletion, increased cytosolic free calcium, and activation of degradative hydrolases leads to bile salt-induced hepatocellular injury.6 Hepatocytes in turn may release additional factors that stimulate fibrosis. The fibrotic process is enhanced by profibrogenic cytokines released from proliferating ductules. The evolving processes of parenchymal injury and regeneration in a fibrosing environment finally result in secondary biliary cirrhosis.
Pathogenesis Two theories of disease progression in EHBA have been proposed. It has been considered a congenital anomaly due to failure of recanalization of the bile duct, thought to be occluded by proliferated epithelial cells early in fetal life. Alternatively, it is thought to be due to a progressive destruction of developed extrahepatic and even intrahepatic bile ducts by an inflammatory process of unspecified nature.6 The latter is now taken to be the most likely. This has led to a change in emphasis to EHBA being an acquired disorder. Landing proposed the concept of infantile obstructive cholangiopathy based on the histological similarity of the liver lesions in patients with neonatal hepatitis, EHBA, and choledochal cyst.36 This suggests that EHBA may be the result of a cholangiopathic process that starts in postnatal life in most cases, due to an inflammatory process that destroys liver parenchymal cells and bile duct epithelium, resulting in obliteration of bile duct lumina. This concept has had great impact in the subsequent literature on the subject, but is still a hypothesis. Its main merit is in stimulating thought about EHBA as an acquired progressive and destructive process. However, when considering EHBA as a postnatally acquired disorder, the subgroup of patients with associated malformations and the minority of patients where there is a familial aspect must be considered. Thus it may be that there are two types of EHBA, a fetal–embryonic form and a perinatal or postnatal acquired form of the disease6. The fetal–embryonic form is characterized clinically by the early onset of jaundice without a jaundice-free interval, with physiological jaundice in the newborn changing into a pathological, conjugated hyperbilirubinuria in the second or third week of life. This has been taken as evidence that the atresia of bile ducts commences in the embryonic or fetal period in this subgroup of patients. These patients have a poorer prognosis.6 The more common perinatal or postnatal form of EHBA is characterized clinically by a jaundice-free period after birth. This often
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
lasts only a few days, but in some cases it may be as long as several weeks. Bile duct remnants may be demonstrated in this group as segments with epithelium-lined lumina or epithelial clusters. Other malformations are not seen. At present, the distinction between two clinical forms of EHBA is generally accepted.6
Pathology of the Extrahepatic Biliary Tree Macroscopic. There are several anatomical variants of EHBA. These range from total absence of the system at one extreme to complete normality at the other, with presence or absence, stenosis or patency, of all or any parts in between. Kasai proposed a classification of the numerous observed anatomic variants, with three main types37 (Figure 71-1), distinguishing between correctable forms (where only part of the extrahepatic tree is involved and anastamosis of the remaining bile ducts is possible) and uncorrectable. Type I corresponds to atresia of the common bile duct, whereas the more proximal extrahepatic bile duct segments are intact. Type IIa represents obliteration of the common hepatic duct with or without atresia of its main branches; the common bile duct, the cystic duct, and the gallbladder are normal. Reanastomosis in this type of EHBA is performed through cystically dilated ducts near the porta hepatis. Type IIb also belongs to the correctable
f
e
g c
d b
A
B
a
C
D
Figure 71-1. The three main anatomical types of extrahepatic biliary atresia (EHBA). (A) Type I. The common bile duct is partially or completely occluded or is reduced to a fibrous cord, with the more proximal extrahepatic bile duct segments left intact. (B) Type IIa. There is obliteration of the common hepatic duct. There may also be atresia of its main branches. The common bile duct and cystic duct are patent, and the gallbladder is intact. Bile ducts at the porta hepatis are dilated. (C) Type IIb. There is obliteration of common bile duct and hepatic and cystic ducts. The gallbladder is not involved. The bile ducts at the porta hepatis are dilated. (D) Type III. There is obliteration of common, hepatic, and cystic ducts. There are no anastomosable ducts at the porta hepatis. A prehilar fibrous cone is present. Types I, IIa, and IIb represent approximately 10% of cases of EHBA. The majority are type III. a, duodenum; b, common bile duct; c, common hepatic duct; d, cystic duct; e, gallbladder; f, liver; g, cystically dilated ducts at the porta hepatis. (Modified from Schweizer P, Müller G. Cholestase-syndrome im Neugeborenen und Säuglingsalter. Stuttgart: Hippokrates Verlag; 1984.)
category, although all main branches of the extrahepatic system of bile ducts (common duct, hepatic ducts, and cystic ducts) are obliterated or missing. Reanastomosis with the duodenum is possible through cystically dilated ducts in the hilum, although the value of this type of corrective surgery has been questioned. Unfortunately, the correctable types I, IIa, and IIb represent less than 10% of all patients with EHBA, with 90% or more of patients belonging to type III, which represents non-correctable biliary atresia, i.e., the lack of or atresia of common, hepatic, and cystic ducts. There are no cystically dilated hilar ducts that can be used for anastomosis in the type III patients. Often the presence of a peculiar prehilar fibrous cone is seen. The gallbladder is involved in the atretic process in about 80% of patients. Microscopic. The diagnosis of EHBA is confirmed by identifying complete fibrous obliteration of at least part of the extrahepatic biliary tree. Usually only a segment is reduced to a fibrous cord (Figure 71-2A). The remaining parts reveal remnants of lumina and inflammatory changes (Figure 71-2B). The variable histological appearances have been categorized into three or four types38 depending on the presence or absence of a lumen lined by biliary epithelium, and the presence or absence of inflammatory infiltration (Figures 71-2A–C, 71-2A–D). In a proportion of specimens the lumen is filled either with cellular debris and macrophages containing bile or with a biliary concrement. In other instances the lumen appears free but narrowed by surrounding inflammatory tissue. These variable histologic appearances reflect a dynamic process of progressive inflammatory destruction of the extrahepatic bile ducts, with the degree of fibrous obliteration increasing with the increasing age of the patient, in parallel with increasing fibrosis in the liver. The earliest stage corresponds to periductal inflammation with necrosis and sloughing of the epithelial lining, followed by progressive periductal fibrosis and narrowing of the lumen. The end stage is a complete fibrous scar of a destroyed epithelium-lined tube that remains identifiable as a fibrous cord in which the collagen texture is denser than the surrounding connective tissue The most advanced stage of complete obliteration is frequently seen at the distal end of the common hepatic duct,38 whereas the more proximal segments and the prehepatic fibrous cone often show the early stages of inflammatory destruction. This has led to the hypothesis that the lesion represents an ascending progressive inflammatory process.38
Clinical Features The clinical presentation of EHBA is that of an unremitting, progressive jaundice. The patient is usually born at term, with a normal birth weight. Jaundice is typically present from shortly after birth, often continuous with physiological jaundice. There may be some variability in intensity; however, jaundice can be readily identified in affected infants by 4 weeks of age. Yellow or dark urine with increasingly pale stools, which eventually become acholic, is noted, although initially there may be variation in stool color, which may be confusing. Atypical presentation may be seen. Prematurity has been reported as a more frequent feature in a recent epidemiological study from New York.39 It is important to differentiate EHBA from other casuses of neonatal cholestasis in order to prevent unnecessary surgery (Table
1359
Section XII. Inherited and Pediatric Diseases of the Liver
A
B
C
D
Figure 71-2. Extrahepatic biliary atresia. The four histological types of fibrous remnant of the extrahepatic bile duct seen in extrahepatic biliary atresia. (A) No remaining extrahepatic bile duct. Instead, the bile duct is replaced by dense fibrous connective tissue. (Hematoxylin & eosin ¥64.) (B) A cluster of small epithelium-lined tubes lies embedded in connective tissue, characterized by inflammatory cell infiltration and beginnings of fibrosis. Note degenerative changes (e.g., nuclear pyknosis) (arrow) in part of the lining epithelial cells. (Hematoxylin & eosin ¥128.) (C) Most of the lining bile duct epithelium is destroyed; some segments remain (arrow). A lumen can still be recognized, surrounded by granulation-like tissue rich in fibroblasts and inflammatory cells. (Hematoxylin & eosin ¥160.) (D) The epithelial lining of the duct has virtually disappeared. A slit-like lumen is still discernible, surrounded by concentric fibrosis. Collapse of the lumen and further fibrosis would lead to the formation of a fibrous cord. (Hematoxylin & eosin ¥40.)
71-2). Neonatal hepatitis, a1AT, and AGS in particular often show similar presentations to EHBA. There are, however, features that may distinguish them from EHBA (Table 71-1). Examination findings in EHBA include hepatomegaly with a firm liver. Splenomegaly is a late sign. Failure to thrive despite adequate feeding is due to high nutritional requirements and fat malabsorption. Around 30% have associated cardiovascular anomalies (ventricular or atrial septal defects). Polysplenia syndrome, including preduodenal portal vein, situs inversus, absence of inferior vena cava, and malrotation are also well recognized. Vitamin K-responsive coagulopathy is more common in breast-fed infants who did not receive vitamin K at birth. Ascites and pruritus are late complications indicating progression to cirrhosis. Late presentation is still a problem, due to neonatal jaundice being wrongly ascribed to physiological or breast milk jaundice. To try and reduce this there have been several awareness campaigns, including
1360
the yellow-baby alert in the UK, and pilot schemes using color charts to detect acholic stools.
Specific Investigations The aim of investigations is to establish atresia of extrahepatic ducts and at the same time to exclude alternative diagnoses (Tables 72-1 and 72-2). Prompt investigation is vital so that if possible surgery can be carried out before the age of 8 weeks.
Liver Function 1. Serum conjugated bilirubin at presentation ranges from 40 to 200 mmol/l (normal range <15 mmol/l). 2. Serum aminotransferases are always abnormal: concentrations of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) are typically in the range of 80–200 U/l (normal range <50 U/l).
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
Table 71-2. Differentiating features of neonatal cholestatic liver disorders Presenting features
Differential diagnosis
Intrauterine growth retardation, prematurity
a1-antitrypsin deficiency Neonatal hepatitis Alagille syndrome PFIC Choledochal cyst “Syndromic” EHBA Alagille syndrome Alagille syndrome “Syndromic” EHBA Hypopituitarism Neonatal hepatitis Hypopituitarism Metabolic liver disease Alagille syndrome Hypopituitarism Neonatal hepatitis of viral or chromosomal etiology
Intermittent jaundice Cardiovascular abnormalities Renal abnormalities Microcephaly Hypoglycemia Abnormal facies
PFIC, persistent familial intrahepatic cholestasis; EHBA, extrahepatic biliary atresia.
3. Serum alkaline phosphatase is usually elevated to above 1000 U/l (normal range 150–700 U/l) due to biliary damage or rickets. 4. g-Glutamyltranspeptidase is usually elevated (¥10 normal). 5. Serum albumin is usually normal. 6. Cholesterol may be elevated but triglycerides are normal. 7. Prothrombin time is normal, although 5–10% of cases present with vitamin K-responsive coagulopathy. 8. Blood glucose is usually normal. Ultrasound scan of abdomen. This is performed after a 4-h fast, and may not demonstrate a gallbladder or only a contracted gallbladder. Rarely, a dilated extrahepatic biliary tree is seen, consistent with distal, correctable atresia; dilated intrahepatic bile ducts are uncommonly found. Polysplenia and situs inversus and any renal abnormalities can be seen, as well as abnormal vascular anatomy consistent with the polysplenia syndrome. Recently, newer ultrasound techniques have been developed and two novel findings have been reported that may be of more diagnostic significance. Choi et al.40 reported the triangular cord sign as being specific for EHBA. This is a unique triangular or tubular echogenic density representing the fibrous cone of bile duct remnant at the porta hepatis. Overall results show 98% specificity and 80% sensitivity. A further study using a higher frequency of transducer (13 MHz as opposed to 7 MHz) identified abnormalities in gallbladder shape, wall thickness, and morphology, that are characteristic of EHBA.41 Hepatobiliary scanning, using DISIDA or TBIDA. Hepatobiliary scintigraphy excludes the diagnosis of EHBA when biliary excretion of isotope into the intestine is demonstrated. If it fails to demonstrate passage of the radiolabelled substance into the intestinal tract over a 24-hour period then EHBA is likely. However, although about 95% sensitive, results are not 100% specific and patients with cystic fibrosis, severe neonatal hepatitis, and bile duct paucity syndromes may fail to excrete isotope into the intestine. Phenobarbital induc-
tion, with at least 3 days’ pretreatment (5 mg/kg per day for 3–5 days) is now used to promote excretion of isotope and thus increase the specificity of the test. Percutaneous liver biopsy. This is essential and has high diagnostic specificity. Features of bile duct obstruction, with bile ductular proliferation, bile plugs in small bile ducts, and portal tract edema are usually obvious, along with variable fibrosis and some giant-cell transformation (Figure 71-3). The earlier the liver biopsy is performed, the more difficult it may be to interpret because of the overlap with giant-cell hepatitis. In recent years the diagnostic accuracy of liver biopsy is up to 90–95% in biopsies of sufficient size and comprising five to seven portal tracts.6 Classic EHBA. The histological features represent a dynamic process and change in the course of the disease (Figure 71-3). The timing of these stages is only approximate. Early on (from about 1 to 4 weeks), non-specific features of bilirubinostasis predominate. Granules of bile pigment are seen in hepatocytes, with intercellular bile plugs. A few hepatocytes may show degeneration and necrosis, but no inflammatory infiltration is present. From about 4–7 weeks the portal tracts show characteristic changes, with rounding of the portal tract, edema, dilation of lymph vessels, and ductular proliferation (Figure 71-3B). The increase in ductules occurs typically at the periphery of the portal tracts, socalled marginal bile duct proliferation, and is associated with an inflammatory infiltrate of both lymphocytes and polymorphonuclear leukocytes, and destruction of bile ducts. Periportal fibrosis is noted. This stage is considered diagnostic of EHBA and explains the recommendation by some to postpone diagnostic liver biopsy until the sixth week of life. Later biopsies (after 7–8 weeks) show progressive portal and periportal fibosis, together with extension of the ductular reaction, but a decrease in inflammatory infiltration, and in the number of intact bile ducts (Figure 71-3C). Periportal fibrosis eventually leads to biliary fibrosis and subsequent biliary cirrhosis (Figure 71-3D), characterized by nodular regeneration of the parenchyma and perinodular septal fibrosis. Severe early form of EHBA. The basic lesion of the intrahepatic bile ducts in the severe early form of EHBA corresponds to a lack of resorption and remodeling of the fetal ductal plate, termed ductal plate malformation. Bile ducts here differ in appearance from classical EHBA, being hyperplastic. Ductal plate malformation is the basic lesion of several disorders, including congenital hepatic fibrosis, infantile polycystic disease, von Meyenburg complexes, Ivemark’s syndrome, Meckel’s syndrome, Caroli’s disease, and some rare other disorders. However, it may also be associated with EHBA (Figure 71-4). This histological feature is combined with the typical features of classical EHBA. Duodenal intubation for bilirubin content. This is possible using either 24-h collection and visual examination of duodenal fluid, or by examining bile flow under direct endoscopic visualization following intravenous injection of cholecystokinin. A positive result excludes EHBA. A negative result is not diagnostic because intrahepatic cholestasis may also cause complete lack of bile flow into the intestine. This test is not now used routinely.
1361
Section XII. Inherited and Pediatric Diseases of the Liver
A
B
D
C
Figure 71-3. Histology of the portal tract in normal liver and in extrahepatic biliary atresia. (A) Normal adult liver. Bile ducts are present in the portal tract together with small ductules at the portal–parenchymal interphase, as stained with immunoperoxidase polyclonal anticytokeratin. (Immunoperoxidase stain with polyclonal rabbit antikeratin, wide-spectrum screening (Dako Corporation, Santa Barbara, CA). Counterstained with hematoxylin ¥400.) (B) Extrahepatic biliary atresia in an infant aged 43 days. The portal tract shows mild mononuclear cell infiltration. The bile duct contains bile concrements (arrow). Immediately surrounding the bile duct are ductular reactive cells at any early stage (arrowheads). (Hematoxylin & eosin ¥400.) (C) Extrahepatic biliary atresia in an infant aged 8 weeks. Portal tract with two bile ducts (BD) showing swelling, vacuolization, attenuation, and sloughing of their lining epithelial cells. There is a mild inflammatory infiltrate. Ductular reactive cells are demonstrable at the margins of the portal tract (arrows). (Hematoxylin & eosin ¥400.) (D) Extrahepatic biliary atresia in an infant aged 10 months showing biliary cirrhosis. There is extensive fibrosis and an increase in ductular reactive cells between the nodular parenchymal areas (P). (Shikata’s orcein stain ¥160.)
Cholangiography. This is performed to identify patency of bile ducts and is necessary if the diagnosis is uncertain. Usually an intraoperative cholangiogram is performed so that, if EHBA is diagnosed, it is possible to proceed to hepatoportoenterostomy. Endoscopic retrograde cholangiopancreatography (ERCP) and magnetic resonance cholangiography are possible alternatives. Percutaneous transhepatic cholangiography (PTC) is not usually helpful as intrahepatic bile ducts are rarely dilated. ERCP can be performed successfully and without complication in patients with EHBA following the development of instruments designed specifically for pediatric use. Considerable skill and expertise are required on the part of the endoscopist, as the diagnostic result depends on the failure to visualize the biliary tree. However, some experienced centers consider that ERCP does not add con-
1362
siderable value to the evaluation of infants with EHBA as the procedure is costly, requires considerable endoscopic skill, and is not without potential complications. False positives may be seen in children with AGS. Magnetic resonance cholangiopancreatography was thought to have a role in diagnosis, but more recent evidence indicates that differentiation from severe intrahepatic cholestasis from EHBA may be difficult due to low bile flow.
Treatment and Prognosis Optimal therapy for EHBA remains the Kasai procedure, or portoenterostomy, which should be carried out as soon as possible after diagnosis. General management strategies must be put in place as soon a child is known to have a cholestatic liver disease, including
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
B
A
Figure 71-4. Histology of the portal tract in extrahepatic biliary atresia with ductal plate malformation. (A) Extrahepatic biliary atresia showing portal tract in an infant aged 43 days. Bile ducts are seen in a ductal plate configuration. Some segments of the ductal plate show epithelial irregularities and involution. A mild ductular reaction occurs at the interphase between the portal tract and parenchyma. Note prominent arteries and poorly developed portal vein in the portal tract. (Hematoxylin & eosin ¥160.) (B) Extrahepatic biliary atresia in an infant aged 30 days. There is disturbed liver architecture with bridging between large portal areas. Portal tracts contain several bile ducts. Some of these appear in a ductal plate configuration (arrows). This degree of fibrosis at the age of 30 days represents early, severe extrahepatic biliary atresia. Note the resemblance to congenital hepatic fibrosis. (Hematoxylin & eosin ¥64.) (C) Extrahepatic biliary atresis in an infant aged 48 days. Enlarged portal tract with moderately dense inflammatory infiltration. Two concentric rings of bile ducts in ductal plate configuration can be recognized (1, 2). Ductular reactive cells occur at the portal–parenchymal interface (arrows). (Hematoxylin & eosin ¥160.) C
optimizing nutrition and fat-soluble vitamin intake (see below, General management of cholestatic liver disease). This is of particular importance in EHBA, as resting energy expenditure is increased and long-term outcome is improved with optimal nutrition. Appropriate counseling for parents is vital and psychological support is necessary for many families (see below, General management of cholestatic liver disease).
Surgical Management Historical perspective. Surgical treatment of EHBA was first discussed by Holmes in 1916, with successful surgery first reported by Ladd in 1928. However, only limited success was reported between 1927 and 1970. The progress in understanding the varying froms of neonatal cholestasis led to a variety of approaches over the next few decades, as it appeared that surgery could be harmful in infants with neonatal hepatitis. It was therefore recommended that surgery be postponed until the patient was 4 months old as the diagnosis of EHBA versus neonatal hepatitis could not be established in most infants during the early months of life. This was obviously not of benefit to those with EHBA. It was not until the 1950s that a reliable surgical procedure was developed in Japan. In 1957 Dr. Morio Kasai performed exploratory surgery on an infant and found no patent extrahepatic ducts. He made a shallow exploration into the porta hepatis, just anterior to the portal vein, and found a small amount of
bile seepage. A segment of small bowel was anastomosed to the liver capsule around this opening into the liver. A good flow of bile developed subsequently, and the patient remained well for more than 17 years. An account of this surgical approach – hepatoportoenterostomy, or the Kasai operation – was published originally in Japanese in 1959, but not in the English literature until 1968.42 This met with skepticism initially, and these two factors combined led to a delay in the wider implementation of this surgical technique. It is now recognized that hepatoportoenterostomy and its subsequent modifications are the treatment of choice for EHBA. The work of Kasai demonstrated that intrahepatic bile ducts were patent from the interlobular ducts of the liver to the porta hepatis in nearly all patients during the first 2 or 3 months after birth. Interlobular bile ducts appeared to be destroyed rapidly and their number decreased progressively after 2 months of age. This explained why surgery was most effective for patients younger than 10 weeks of age, which made early and rapid investigation mandatory. Increased experience worldwide with hepatoportoenterostomy has revealed that the intrahepatic component of duct inflammation and destruction is of great importance in determining the prognosis, even in patients with good bile flow soon after surgery. The rationale of hepatoportoenterostomy lies in establishing continuity between the bile ducts near the porta hepatis and the intestinal lumen. Theoretically, successful restoration of bile flow is expected in patients with patent biliary structures near the porta
1363
Section XII. Inherited and Pediatric Diseases of the Liver
hepatis. Postoperative bile flow would be expected to be greatest in patients with biliary structures having the largest diameters or in patients with the greatest numbers of such structures, or both. Investigations by Kasai indicated that a satisfactory postoperative flow of bile is only obtained when the diameter of the patent ducts near the porta hepatis is at least 200 mm.37 Subsequent studies have suggested that duct lumina of 100-mm diameter are sufficient for ensuring adequate bile flow. Some authors recommend intraoperative confirmation by frozen section of the presence of microscopically patent ducts at the level of anastomosis in portoenterostomy. However, there is not always a correlation between the presence of patent ducts and postoperative flow of bile in these studies. Furthermore, others have reported no correlation between postoperative drainage of bile and the number or size of bile ducts at the porta hepatis. Gautier and Eliot38 found, for example, that six of 13 patients with no demonstrable bile ducts at the porta hepatis had an adequate bile flow postoperatively. Although refinement in recognition and measurement of bile ducts draining at the porta hepatis might help to assure a better predictive value of operative intervention, this still does not explain the occurrence of adequate bile flow in patients without recognizable epithelial structures. These puzzling cases raise the question of how bile is drained when there is no duct continuity and point to alternative pathways like lymphatic drainage of bile constituents. The concept of lymphatic drainage forms the basis for some modifications of the Kasai procedure (omentopexy).
Surgical Prognostic Factors Prognosis of EHBA is dependent on several factors. The age at diagnosis and subsequent surgery are critical determinants of both shortterm and long-term outcome. If portoenterostomy is performed within the first 60 days of life by an experienced surgeon, then bile drainage should occur in at least 70–80% of patients.43 If performed between 60 and 90 days of age, 40–50% show bile drainage, and after 90 days only up to 25% will drain bile.3 However, age at operation is not the only factor. Two recent studies have examined the extremes of age for portoenterostomy. Volpert et al.44 found that outcome was worse, in terms of the need for transplantation, if surgery was performed before 30 days. This may be due to a different phenotype presenting earlier, as these infants did not have a jaundice-free interval. At the other end of the spectrum, there is no strict upper age limit for portoenterostomy because the time of onset of disease, its rate of progression, and its severity vary.45 In fact, concerns regarding performance of a late portoenterostomy, after 100 days, have recently been challenged by Davenport et al.46 Of a total of 422 patients diagnosed with EHBA, 35 were 100 days old or older, 34 of whom underwent surgery. Of the 35, 5- and 10year survival with native liver was 45% and 40% respectively.45 Thus it appears, with improving surgical techniques, better pre- and postoperative care, portoenterostomy after 100 days must still be considered unless there are signs of decompensating liver disease. However, overall the chance for long-term survival with native liver decreases with increasing age at surgery. Other prognostic factors include the anatomical pattern of atresia, with complete extrahepatic atresia associated with worse outcome, presence of polysplenia, and experience of the center performing
1364
the portoenterostomy.43 A study in the UK and Ireland found that greater experience in performing the portoenterostomy is associated with improved survival with native liver, when more than five such procedures were performed annually.4,46 Other important factors to consider when discussing prognosis with families includes the degree of fibrosis. Where disease is advanced at diagnosis, portoenterostomy is less likely to be successful.
Postoperative Care Protocols for ongoing management of EHBA postoperatively vary, and in the USA there is no standardized protocol. Many advocate the use of steroids immediately postoperatively and in some centers pulsed corticosteroids are given to manage refractory cholangitis. Ursodeoxycholic acid is often used to stimulate bile flow,47 and prophylactic antibiotics given to protect against ascending cholangitis.48 The protocol used in our unit includes 5 days of intravenous gentamicin and amoxicillin, followed by prophylactic antibiotics orally to prevent cholangitis. This includes cycles of cefalexin, trimethoprim, and amoxicillin each for 12 weeks. The initial postoperative period is also covered by intravenous methylprednisolone in decreasing dose for 10 days, then oral prednsiolone (5 mg/day) for 1 week. Phenobarbital and ursodeoxycholic acid are used to promote bile excretion and decrease pruritus.
Complications Ascending cholangitis. Cholangitis is the most important complication and has a significant bearing on long-term outcome and potential need for early liver transplantation. It is very common in children with EHBA post-portoenterostomy, occurring in 40–60% of patients. It must be treated early, as infection may encourage progressive biliary cirrhosis and lead to early development of liver failure. Organisms are mainly enteric ones, although fungal infection may occur. The use of prophylactic antibiotics decreases the incidence of recurrence48 and varying regimes have been used. Much consideration has been given to corticosteroid use postportoenterostomy, in order theoretically to reduce ongoing bile duct injury and fibrosis. However, there are as yet no published randomized controlled trials of effectiveness. Many centers have used short-term (1–2-week) courses postoperatively, and one study found an 8–10-week course post-portoenterostomy improved outcome, when compared to historical controls. Pruritus and nutrition. Treatment for this is as detailed below. Intrahepatic biliary cysts (IBC). The development of IBC following portoenterostomy was first reported in 1980 and is seen in up to 21% of patients.49 It is thought to be due to progressive inflammation and cirrhosis in the intrahepatic lobular spaces, leading to intrahepatic biliary obstruction and cyst formation.49 Treatment with antibiotics may not be effective, percutaneous transhepatic cholagiography with drainage is not always effective, and patients may require liver transplantation.49 Cirrhosis and portal hypertension. This develops in the majority of patients. Progressive splenomegaly, decrease in size of the liver, with a firm texture and variceal bleeding are all early signs. These may eventually progress to decompensated liver disease, with ascites, hypoalbuminemia, encephalopathy, or hepatopulmonary
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
syndrome. Management of these is as for any other cause of portal hypertension or liver decompensation. The presence of variceal bleeding alone, without a rise in bilirubin, is not an indication for urgent transplantation50,51 unless there is a failure of medical and endoscopic therapy.
Medium-term Outcome In the medium term, prevention of cholangitis is the main aim. Recurrent cholangitis leads to progression of fibrosis and earlier decompensation. Thus to avoid transplantation, use of prophylactic antibiotics in high-risk children and prompt treatment of any suspected episodes of cholangitis are vital. It is important to maintain good nutrition to ensure adequate growth and development. All infants need monitoring to detect signs of decompensating liver disease and the development of portal hypertension.46
Long-term Outcome
90.5%) after transplantation (P = 0.0001, log rank test). More recently a pediatric end-stage liver disease (PELD) score has been developed using data accumulated by the Studies for Pediatric Liver Transplantation Research Group.56 This uses five variables to predict 3-month survival of children listed for liver transplantation: serum bilirubin, international normalized ratio, albumin, growth failure (height or weight z-score <–2) and age less than 1 year. This may assist in early referral and listing but does not appear to affect outcome. Overall survival for elective biliary atresia is up to 95% at 1 year and 89% at 5 years (UK Transplant, personal communication).
CHOLEDOCHAL CYST Choledochal cysts were first described in 1959. They are a group of congenital malformations of the pancreatobiliary system, involving intrahepatic and/or extrahepatic bile ducts. They are frequently accompanied by pancreatobiliary malunion.57 They have been classified into six types.57 They are unusual in western countries, but common in Asia. There is a female predominance (female:male 5:1).
A successful portoenterostomy is defined as achieving bile drainage, as documented by normalization of serum bilirubin at 3 months postoperatively. Although at least two-thirds of patients have successful outcome of surgery, a cure is not possible for the majority. Long-term results after 10 and 20 years of follow-up43,52 indicate that only a minority appear to be cured following hepatic portoenterostomy. Even those with normal liver function tests at 10 years post-portoenterostomy have fibrosis or cirrhosis on liver biopsy.53 In approximately one-third of patients, bile flow after portoenterostomy is inadequate and progressive fibrosis and cirrhosis develop, leading to death by the age of 2 years without liver transplantation. In other children, cirrhosis develops at a slower rate, despite bile drainage, and transplantation is required later in childhood. In France, a 10-year study of 472 patients with EHBA found that actuarial survival with native liver was 29% at 10 years, with 5-year survival after liver transplant of 71%.
Choledochal cysts may present at any age. The classical presentation is with a triad of symptoms consisting of jaundice, abdominal mass and pain, but this is unusual in the neonatal period. Most affected infants have jaundice, abdominal mass or distension, and acholic stools,58 and differentiation from biliary atresia or choledocholithiasis is important. Later presentation is associated with biliary colic, acute cholangitis, or gallstone pancreatitis. The diagnosis is made by identifying the choledochal cyst on ultrasound examination of the liver. Diagnosis can also be made antenatally in the fetus by prenatal sonography.58,59 Diagnosis is confirmed by cholangiography, either percutaneous or endoscopic. Hepatobiliary scanning has limited utility for diagnosis. Liver function tests are compatible with biliary obstruction.
Indications for Transplantation
Treatment and Outcome
EHBA is the commonest indication for liver transplantation in childhood, with two-thirds of patients requiring transplantation following portoenterostomy. Technical advances, including the use of “cut-down” livers and “split” livers, and living related donation have addressed issues of donor availability and extended transplantation to infants. The age of the child is no longer a contraindication to transplantation, and 1-year survival rates following transplantion are approaching 90% for all ages. The indications for liver transplantation in EHBA are decompensating liver function, intractable portal hypertension with recurrent variceal bleeding or severe malnutrition with fat soluble vitamin deficiency, and metabolic bone disease resulting in pathological fractures. Elective transplantation is associated with better short-term survival compared with transplantation for acute liver failure.54 A number of risk factors have been established by Rodeck et al., one or more of which indicates the need for urgent transplantation: bilirubin >340 mmol/l, albumin <33 g/l, and standard deviation score for weight less than –2.2.55 Comparing the post-transplantation survival in these groups, there is a statistically significant difference at 1 year (57% versus 90.5%) and 4 years (49% versus
Treatment is aimed at surgical removal. Excision of the cyst with hepaticoenterostomy offers the best outcome.60 Complications are less with early surgical intervention, with specific modifications of the hepaticojejunostomy.61 Surgery should be performed promptly in infants diagnosed prenatally who have conjugated hyperbilirubinemia. If the infant remains free of jaundice, elective surgical resection of the choledochal cyst may be postponed until the infant is 1 month old, but should not be greatly delayed. Although approximately 50% of infants with prenatally identified bile duct dilatation have hepatic fibrosis, and a few have cirrhosis, most of these infants do well. In older patients, where cirrhosis is advanced, transplantation may be necessary. A minority of infants may have correctable biliary atresia, and close follow-up is warranted. Pancreatitis may also occur. Longer-term, there is a risk of malignancy, but this is low in childhood (less than 1% under 10 years of age), but rises with age, such that it is greater than 10% by the third decade.62
Clinical Features and Diagnosis
CAROLI DISEASE Caroli disease (also known as type V choledochal cyst) is characterized by congenital segmental saccular dilatation of the intra-
1365
Section XII. Inherited and Pediatric Diseases of the Liver
hepatic bile ducts, without hepatic fibrosis or portal hypertension. It more commonly affects boys, and is often associated with autosomal recessive polycystic kidney disease.63 Caroli disease is rarely evident in infancy. The main presenting feature is jaundice due to acute cholangitis, which may be complicated by intrahepatic calculi and hepatic abscess formation.63 Some newborn infants with severe autosomal recessive polycystic kidney disease have extensive cystic bile duct changes, but renal insufficiency dominates the clinical picture. Ultrasound of the liver is often adequate for diagnosing Caroli disease; cholangiography is confirmatory. Outcome is related to the severity of renal disease and/or the development of cholangitis, intrahepatic gallstones, or portal hypertension. Caroli syndrome is considered to be present if there is congenital hepatic fibrosis, which presents in infancy with hepatomegaly and either autosomal recessive polycystic kidney disease or systemic hypertension. Jaundice and abnormal serum aminotransferases are uncommon. Treatment depends on the clinical features. If disease is localized to one hepatic lobe then hepatectomy relieves symptoms and possibly also the risk of malignancy. Outcome is variable and depends on the progression of hepatic and renal disease as above. Malignancy may complicate Caroli’s disease in approximately 7% of cases.63
ness at this site, but a viral infection has also been suggested as a cause for this disease. Some cases are associated with distal obstruction resulting from sludge or EHBA. Clinical signs of the disease are commonly noted between 1 week and 2 months after birth, although instances of later onset have been reported. A single case was diagnosed before birth.68 The clinical presentation may be one of an intra-abdominal catastrophe. Usually, however, fluctuating jaundice, acholic stools, and dark urine develop slowly. Symptoms may be so mild that the development of an inguinal or umbilical hernia secondary to ascites may be the first sign of illness. In some cases general symptoms of weight loss, irritability, and vomiting may be noted before jaundice. On clinical examination there is abdominal distention from bile ascites and sometimes bile staining of the umbilicus or scrotum caused by bile tracking along patent hernial sacs. Signs of peritonitis and pyrexia are usually lacking. Laboratory investigation reveals a moderate rise in serum levels of bilirubin and normal hepatic transaminases, which differentiates this condition from EHBA. Definitive preoperative diagnosis can be made by abdominal paracentesis, yielding fluid with a high concentration of bilirubin. Intravenous cholangiography demonstrates a leak in the extrahepatic bile ducts. Surgical treatment depends on the findings at time of operation.
NEONATAL SCLEROSING CHOLANGITIS Neonatal sclerosing cholangitis (NSC) was first reported in 1987 with a few subsequent reports.64,65 It is characterized by irregular narrowing of extrahepatic or intrahepatic bile ducts. The etiology of this condition is unknown but may have a genetic basis.65 Currently the true nature of NSC remains uncertain. It has been associated, in one case, with non-specific autoantibodies.66 NSC is distinguished from childhood primary sclerosing cholangitis by the presentation in early infancy with conjugated hyperbilirubinemia which then resolves within 3–6 months.64 Recurrent hyperbilirubinemia develops 1–2 years later or in mid-childhood (8–10 years old). This is distinct from childhood primary sclerosing cholangitis presenting in infancy, where there is no early cholestatic jaundice. Other features include hepatosplenomegaly, biliary cirrhosis, and portal hypertension. Investigations indicate obstructive biliary disease with elevated serum alkaline phosphatase and g-glutamyltranspeptidase. Endoscopic or percutaneous cholangiography demonstrates beaded irregularity of medium to large intrahepatic bile ducts in all patients and in extrahepatic ducts in 80%. Liver histology shows portal fibrosis with ductal proliferation developing into biliary cirrhosis. Surgical treatment with Kasai portoenterostomy is not indicated and nutritional and supportive management is required. The disease is progressive, requiring liver transplantation at some stage.
SPONTANEOUS PERFORATION OF THE COMMON BILE DUCT Spontaneous perforation of the common bile duct is a highly specific clinical entity in infancy and should always be considered in an infant in whom jaundice develops after an anicteric period of good health. It is a rare condition: some 60 cases have been reported since 1932.67 The cause is unknown. In the majority of patients the perforation occurs at the union of the cystic and common ducts. This suggests a developmental weak-
1366
BILE PLUG SYNDROME AND INSPISSATED BILE SYNDROME The terms bile plug syndrome and inspissated bile syndrome are often used interchangeably, although Bernstein et al. stated explicitly that obstruction of the common duct by a plug of secretions and bile is to be differentiated from the so-called inspissated bile syndrome.69 The majority of infants have massive hemolytic-induced jaundice caused by rhesus and ABO blood group incompatibility. The histological feature of bile plugs in the parenchyma is the result rather than the cause of cholestasis and is caused by bilirubin overload of the liver and by dehydration. Hemolytic disease of the newborn caused by blood group incompatibility is now partially preventable using maternal anti-D treatment, and can be treated by exchange transfusion and so has become more infrequent. Clinically, bile plug syndrome cannot be differentiated from EHBA. Ultrasonography is diagnostic and confirmed at exploratory laparotomy. The inspissated material is removed by irrigation of the biliary tree or use of a mucolytic agent. On occasion, inspissation may proceed to the production of stones, necessitating manual removal of the calculi, or may resolve spontaneously. Choledochal cyst formation may develop after inspissated bile syndrome.
BILE DUCT STENOSIS Localized narrowing of the distal part of the common bile duct is a rare cause of extrahepatic obstruction in children beyond the neonatal age. Fewer than 15 cases have been reported. In three such cases the proximal segments of the duct were dilated and contained small biliary concrements that had accumulated near the narrowed part of the bile duct. The condition was cured by simple choledochotomy and cleaning of the duct. Two cases of post-traumatic stricture of the common bile duct have been reported, possibly caused by previous blunt abdominal
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
trauma, and of two patients with partial stenosis of the confluence of the hepatic ducts without evidence of antecedent trauma.70. Further abnormalities have recently been reported in association with bile duct stenosis, including ductal plate malformation, eosinophilia, and pancreatitis.71
malignant tumors of lymph nodes near the porta hepatis (Hodgkin’s and non-Hodgkin’s lymphomas) and benign lesions such as chronic pancreatitis and post-traumatic cyst of the pancreas.70 Both benign and malignant tumors of the pancreas or duodenum may cause obstructive jaundice in infants (duodenal fibrosarcoma and carcinoma and hemangioendothelioma of the pancreas).
DUPLICATION OF THE BILIARY TREE Duplication of the biliary tree is a condition of exceptional rarity.70 The etiology is identical to that of intestinal duplication. Clinical symptoms include abdominal pain and recurring intestinal obstruction with or without cholestasis. A hard mass of variable size is palpated in the right hypochondrium. The diagnosis is made during surgical intervention for presumed cholangitis or choledochal cyst.
AGENESIS OF THE EXTRAHEPATIC BILE DUCTS Schwartz et al. described five neonates with obstructive jaundice in whom exploration revealed absence of the proximal extrahepatic biliary ducts (four cases) or total absence of the extrahepatic ducts and gallbladder (one case).72 Jaundice was diagnosed from birth to 3 weeks of age. Surgery revealed absence of bile duct remnants, absence of inflammation, and a fibrous mass at the porta hepatis. Liver biopsy specimens showed histologic evidence of cholestasis, minimal bile duct proliferation and fibrosis, and nearly complete absence of inflammation. The authors conclude that this group of patients represents true agenesis of the extrahepatic bile ducts rather than EHBA and that liver transplantation is the primary mode of treatment in this rare entity.
GALLSTONES Cholelithiasis is reported with increasing frequency in infants and neonates, because of the widespread use of ultrasonography. The pigmentary nature of cholelithiasis has been established in most cases, but the pathogenesis of stone formation remains unclear.73. Gallstone formation can occur in the fetus, premature infants treated with parenteral nutrition (PN) and furosemide. Intrahepatic bile stone formation has been ascribed to Ascaris lumbricoides infestation. Other possible causes include chronic hemolytic disease,70 mucoviscidosis, bile duct malformations, and septicemia.70 In contrast with adult gallstone disease, there is no female predominance. Recent reviews of the clinical features, diagnostic procedures, and therapy have been published.73
BILE DUCT TUMORS Tumors of the extrahepatic bile ducts are extremely rare in children. The tumors correspond to embryonal rhabdomyosarcoma and exceptionally to liposarcoma.74,75 Clinical symptoms are those of complete cholestasis with progressive onset; an abdominal mass may be palpable. Operative cholangiography reveals the obstruction, which is often near the ampullary region.70 Treatment consists of surgical excision. The prognosis is poor.
EXTRAMURAL COMPRESSION OF THE COMMON BILE DUCT Occasionally, extrahepatic cholestasis is secondary to extrinsic compression of the common bile duct. A case has been reported of bile duct obstruction by peritoneal bands, and cure was affected by simple division of the constricting bands. Other causes include
INTRAHEPATIC DISORDERS NEONATAL HEPATITIS SYNDROME Etiology and Pathogenesis Neonatal hepatitis was first described in 1952 by Craig and Landing, who reported a “form of hepatitis in the neonatal period simulating biliary atresia.” The disease presents clinically as obstructive jaundice indistinguishable from EHBA, and histologically is characterized by parenchymal giant-cell transformation. The biliary tree is patent on abdominal exploration or at postmortem. The histological picture was initially interpreted as “active hepatitis having the characteristics of viral hepatitis” but a viral etiology remained unproven. Neonatal hepatitis or intrahepatic cholestasis is almost as frequent as extrahepatic bile duct abnormalities. The incidence of neonatal hepatitis in south-eastern England was found to be higher, at 1:2500 live births.76 Failure to identify a specific viral infection in the majority of cases and the clinical similarity to EHBA have led to terms such as neonatal hepatitis syndrome and infantile obstructive cholangiopathy. There are now many causes of prolonged, conjugated neonatal hyperbilirubinuria (Table 71-1). The early detection of specific infectious diseases, toxins, and disorders of amino acid, lipid, and carbohydrate metabolism is important because in many instances cholestasis attributable to these factors is treatable and reversible. A similar clinical picture of obstructive jaundice may be seen in homozygous a1AT and in cystic fibrosis. Endocrine disorders (hypopituitarism and hypothyroidism) may also be associated with cholestasis. There remains a subgroup with unclear etiology, and for whom the term idiopathic neonatal hepatitis remains. In neonatal hepatitis, the disease process is thought to focus on the hepatocyte and not, or only secondarily, on intrahepatic and extrahepatic segments of the biliary tree. Intralobular cholestasis may be due to infectious, metabolic, toxic, and unknown causes. Neonatal hepatitis syndrome thus represents a continuously evolving spectrum of problems. It is important to distinguish neonatal hepatitis from EHBA, and to avoid an exploratory laparotomy which may have a harmful effect on the prognosis of patients with neonatal hepatitis. In a follow-up study from France it was noted that progression to cirrhosis was more likely in patients who had undergone surgery, but this may also have been related to advanced liver disease at the time of surgery. Poor prognosis has been observed in patients with neonatal hepatitis mimicking EHBA, that is, severe forms with complete cholestasis.
INTRAUTERINE INFECTION This is the largest subgroup of disorders resulting in giant-cell hepatitis. Most are due to in utero viral infections (Table 71-1).
1367
Section XII. Inherited and Pediatric Diseases of the Liver
TOXOPLASMOSIS, RUBELLA, CYTOMEGALOVIRUS, HERPES SIMPLEX (TORCH) INFECTIONS Congenital infections grouped under the acronym TORCH often have very similar clinical features: hepatosplenomegaly, jaundice, pneumonitis, petechial or purpuric rash, and a tendency to prematurity or poor intrauterine growth. A presentation with fulminant hepatic failure in the newborn period is common with herpes simplex infection. Whenever possible, direct identification of viral infection or measurement of specific immunoglobulin M antibodies should be sought for rapid diagnosis; relying on conventional TORCH titers is less preferable.
Toxoplasmosis Congenital toxoplasmosis is comparatively rare. Infection is acquired through maternal ingestion of raw or undercooked meat or of food, especially fruit or vegetables, contaminated with infected cat feces. Rates of transmission to fetus via transplacental infection are up to 40%77 and vary with gestational age at time of infection. Maternal infection in the third trimester is more likely to cause fetal infection than that earlier in pregnancy. Neonatal hepatitis is an important feature but may be less obvious than central nervous system involvement with chorioretinitis (with large pigmented scars), hydrocephaly, or microcephaly. Intracranial calcification is usually prominent, leading to convulsions, nystagmus, and evidence of increased intracranial pressure. Liver biopsy may demonstrate a nonspecific hepatitis or portal fibrosis with biliary ductule proliferation. Spiramycin or pyrimethamine therapy may prevent progression of ocular, central nervous system, and liver disease77. Prognosis depends on the extent of neurological or optic disease.
Rubella Congenital infection with rubella was first identified in 1941 by Gregg. The virus is now rare because of immunization in the western world. However, with recent controversies over the measles, mumps, and rubella (MMR) vaccine, uptake has fallen and congenital infection could conceivably become a problem. Transplacental transmission of the virus occurs during the first trimester of pregnancy, with severity increasing the earlier during pregnancy the infection occurs. Risk is low after 17 weeks’ gestation. Teratogenicity leads to intrauterine growth retardation, anemia/thrombocytopenia, congenital heart disease (patent ductus arteriosus or pulmonary artery stenosis), cataracts, chorioretinitis (“salt and pepper” appearance), mental retardation, and sensorineural deafness. Hepatosplenomegaly is usual. Liver histology shows typical giant-cell hepatitis. The disease may be self-limited or progress to cirrhosis.
Cytomegalovirus CMV is the commonest cause of congenital infection, affecting 0.15–2% of newborns, almost 90% of whom are asymptomatic at birth.78 It is thought to be more likely in association with maternal primary infection in the second and third trimester, but recent studies examining severity of congenital CMV after recurrent infection have demonstrated equivalent severity.78 Those with evident disease may have intrauterine growth retardation or be premature.
1368
Fetal ascites may occur but CMV rarely causes acute liver failure in the newborn. The commonest presenting features include jaundice, petechiae, and hepatosplenomegaly. Clinical findings include petechial rash, hepatosplenomegaly, and jaundice in 60–80%. CMV infection often affects the central nervous system, producing microcephaly, intracranial calcification, and chorioretinitis; progressive sensorineural deafness or cerebral palsy may develop later in childhood. Liver biopsy demonstrates giant-cell hepatitis; the classical inclusion bodies are rarely seen in neonatal infection. In a study of liver tissue in infants with neonatal hepatitis or EHBA, Chang et al. found CMV DNA in 23 of 50 infants with neonatal hepatitis by PCR, but in only two of 26 with EHBA, and in none of control specimens.79 Although differentiation from biliary atresia is usually easy, CMV may be associated with EHBA. In one report of fraternal twins, both had congenital CMV infection: one had hepatitis only and the other presented with late-pattern EHBA. In addition, 25% of infants with EHBA have been found to have CMV infection and were referred later than those without CMV infection.80 CMV is a candidate virus for causing late-presentation EHBA as it can infect bile duct epithelial cells directly and increase expression of major histocompatibility class II antigens.81 Infants with congenital CMV infection and persisting conjugated hyperbilirubinemia should have EHBA excluded. Conclusive diagnosis requires CMV to be cultured from the infant within the first 4 weeks of life. Serological studies and clinical features provide support for the presence of CMV infection but do not distinguish congenital from early postnatal infection. Treatment with ganciclovir has been suggested; this appears to improve the outcome with regard to hearing loss. However, neutropenia was a significant side effect in two-thirds of those receiving intravenous ganciclovir for 6 weeks and no report was made regarding other parameters. Ganciclovir may also improve congenital CMV-associated liver disease. In most children CMV hepatitis is mild and resolves completely. A few children develop hepatic fibrosis or non-cirrhotic portal hypertension. Intrahepatic calcification has been reported. Cirrhosis with chronic cholestasis necessitated liver transplantation in one child. Persisting neurodevelopmental abnormalities become the main problem in the majority of patients.78
Herpes Simplex Virus Both type1 and type 2 viruses can lead to perinatal infection, causing a severe multisystem disorder with encephalitis, severe hepatitis, or acute liver failure. Type 2 virus shed from genital herpes at birth is the commonest route of transmission, with greatest risk if maternal infection involves the cervix. A true congenital infection from in utero viral infection is less common, but may be associated with extensive hepatic calcifications.82 Liver biopsy shows areas of necrosis with viral inclusions in intact hepatocytes; however, profound coagulopathy may preclude biopsy. Scrapings from vesicular skin lesions reveal herpes simplex virus, but these typical herpetic skin, mouth, or eye lesions may not be present in neonates. Antiviral treatment with aciclovir should be administered to avert the otherwise high mortality, although outcome is more severe if there is disseminated disease or encephalitis.
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
OTHER VIRAL INFECTIONS Varicella Infection with varicella may occur following maternal infection. Infection in early pregnancy is associated with complications in 0.4–2%,83 but is more likely if infection occurs within 14 days of delivery. Complications are more likely if infection is prenatal, but disease is mild in term infants after 10 days of age. Early presentation or protracted disease in an infant of any gestational age may lead to a fatal outcome. There is jaundice, and multisystem involvement, especially pneumonia, usually but not always associated with skin lesions. In fatal cases hepatic parenchymal involvement can be demonstrated. Prompt treatment with zoster immune globulin in infants born to mothers with chickenpox and treatment of infection with acilovir may ameliorate symptoms.
Syphilis Congenital syphilis is now rare in the developed world, but is rising in Eastern Europe and the former Soviet Union, and is common in sub-Saharan Africa. It causes a multisystem illness, which may include intrauterine growth retardation and subsequent failure to thrive, severe anemia and thrombocytopenia, nephrotic syndrome, periosteal reaction and metaphyseal dystrophy, nasal discharge, skin rash with symmetrical superficial desquamation of the hands and feet, diffuse lymphadenopathy, and hepatosplenomegaly. Infants may be born with hydrops fetalis or jaundice may develop within 24 h of birth or following treatment. Cholestasis has been reported in premature infants. Central nervous system involvement occurs in up to 30% of infants. Liver histology in untreated congenital syphilis may reveal numerous treponemes in hepatic tissue, but after treatment with penicillin, giant-cell hepatitis without detectable treponemes is the usual finding. Diagnosis involves serological testing, including the Venereal Disease Research Laboratory (VDRL) test and confirmatory testing for specific antitreponemal antibodies. Radiographs of long bones may show typical bony abnormalities in the first 24 h of life and aid rapid diagnosis.
OTHER VIRUSES ASSOCIATED WITH NEONATAL HEPATITIS There are several other viruses that may cause neonatal hepatitis syndrome (Table 71-1). These include parvovirus B19, which may lead to hydrops fetalis and death in utero, and paramyxovirus, which may cause a severe syncytial giant-cell hepatitis, progressing to cirrhosis, and usually requiring liver transplantation.
BACTERIAL INFECTIONS ASSOCIATED WITH NEONATAL HEPATITIS Both generalized sepsis and localized infections may lead to cholestasis in neonates. Galactosemia must be excluded in all jaundiced infants with Gram-negative septicemia, as infection is a common presenting feature. More unusual infections that may result in cholestasis include congenital listeriosis, where infants may have hepatospenomegaly and occasionally jaundice, with meningitis; and tuberculosis. Heptosplenomegaly is common in congenital tuberculosis, but jaundice is rare, and indicates severe disease.
Endocrine Hypopituitarism. Pituitary–adrenal dysfunction is associated with neonatal hepatitis syndrome in 30–70% of patients.84 The cause of the hypopituitarism is variable and may be due to hypothalamic dysfunction or deficiency of anterior and/or posterior pituitary function. Adrenal insensitivity to adrenocorticotropin has also been described. Clinical features include conjugated hyperbilirubinemia, hypoglycemia, and septo-optic dysplasia. The hypoglycemia commences perinatally and is usually symptomatic and persistent. Septo-optic dysplasia is a neuro-optical malformation with a defect in ventral midline development (absence of the septum pellucidum or corpus callosum) and hypoplasia of one or both optic nerves in association with hypopituitarism. There may also be midline facial abnormalities, nystagmus, and boys may demonstrate microgenitalia.85 The diagnosis is confirmed by detecting an extremely low random or 9.00 a.m. cortisol in association with a low thyroid-stimulating hormone and thyroxine.85 Liver biopsy usually reveals typical giantcell hepatitis, but severe cholestasis may be present with dilated bile canaliculi and hepatocellular necrosis. There may be delayed excretion on radioisotope scanning. Treatment with hormone replacement is essential, including thyroxine, corticosteroids, and growth hormone where appropriate. Progression of the disease to cirrhosis and portal hypertension has been reported where there was delay in hormone replacement.85 There is also a report of association of septo-optic dysplasia with congenital hepatic fibrosis.86 Hypothyroidism, usually associated with unconjugated hyperbilirubinemia, may be associated with the neonatal hepatitis syndrome and should be excluded in every patient.
CHROMOSOMAL DISORDERS Trisomy 18 has been reported as being associated with both giantcell hepatitis and EHBA. Other cytogenetic abnormalities, including trisomy 13, deletion of the short arm of chromosome 18 and 49 XXXXY, have also been reported in association with EHBA. Trisomy 21 has also been reported in association with neonatal cholestasis and EHBA but this is rare. Severe hepatic fibrosis associated with transient myeloproliferative disorder has also been reported with trisomy 21, raising the possibility that hepatic fibrogenesis may be due to high concentrations of growth factors derived from megakaryocytes. Cat-eye syndrome is a rare developmental disorder in humans, associated with the presence of three or four copies of a segment of chromosome 22q11.2, usually in the form of a bisatellite, isodicentric supernumerary chromosome.87 It is characterized by a variety of congenital defects, including ocular coloboma, anal atresia, preauricular tags/pits, heart and kidney defects, dysmorphic facial features, and mental retardation. It is also reported to be associated with EHBA.
IDIOPATHIC NEONATAL HEPATITIS In up to 25% of infants presenting with conjugated hyperbilirubinemia before 3 months of age, no etiology is found. Infants with idiopathic neonatal hepatitis are more likely to be premature or small for gestational age than those with EHBA,88 which may reflect an as yet unknown genetic disorder, or intrauterine infection. There
1369
Section XII. Inherited and Pediatric Diseases of the Liver
is a subset of patients, up to 15% of cases, where more than one child in a single family is affected. In preterm babies some will have cholestasis due to immaturity of the biliary tree. They may be prone to hypoglycemia and have a functionally immature gastrointestinal tract, resulting in difficulties with feeding. It is important to differentiate this condition from other known causes of neonatal hepatitis and, in particular, EHBA.
Clinical Features Conjugated hyperbilirubinemia may present at any time after birth. If detected in the first 24 h of life infection is usually the cause. Most causes of the neonatal hepatitis syndrome have a similar presentation: jaundice, which may not be obvious at first, together with dark urine and pale yellow stools. Abnormal stool color, though suggestive of liver disease, is neither a specific nor reliable feature. Infants may be small for gestational age, especially those with AGS, metabolic liver disease, and intrauterine infection. Failure to thrive or poor feeding may also be a feature. Dysmorphic features should be sought and are present in trisomy 18, trisomy 21, AGS, Zellweger syndrome, and in certain congenital infections. Hypoglycemia may suggest metabolic liver disease, hypopituitarism, or severe liver disease. Hepatomegaly and splenomegaly may be present. Ascites is rarely evident except in metabolic liver disease. Cardiac murmurs or neurological abnormalities are associated with specific congenital syndromes. Bleeding from vitamin K deficiency or thrombocytopenia may occur. Distinguishing neonatal hepatitis from EHBA may be difficult. Hepatomegaly is often obvious in the first or second week in neonatal hepatitis, whereas liver enlargement usually becomes detectable in the third or fourth week of life in EHBA. In EHBA, conjugated hyperbilirubinemia rises more progressively than in neonatal hepa-
titis. Hepatic transaminases are not helpful in discrimination. In the first 3 months of life neonatal hepatitis may cause higher elevations of AST and ALT, whereas g-glutamyltranspeptidase levels are more abnormal in cases of EHBA.89 Different scoring systems based on clinical and laboratory data have been developed in attempts to differentiate better EHBA from neonatal hepatitis. According to Alagille,90 clinical features and laboratory data allow a differentiation between EHBA and neonatal hepatitis in 83% of cases before the age of 3 months. The following features occurred more frequently in infants with neonatal hepatitis than in those with EHBA: male gender (66% versus 45%), low birth weight (mean 2680 g versus 3230 g), other congenital anomalies (32% versus 17%), onset of jaundice (mean 23 days versus 11 days of age), onset of acholic stools (mean 30 days versus 16 days), white stools during the first 10 days after admission (26% versus 79%), and enlarged liver with a firm or hard consistency (53% versus 87%).90
Histology The histological abnormalities in idiopathic neonatal cholestasis are more prominent in the parenchyma than in the portal tracts. There is more extensive parenchymal giant-cell transformation than in EHBA; giant cells are larger, appear more hydropic or ballooned, and often show degenerative features (Figure 71-5). Necrosis of parenchymal giant cells may be associated with infiltration by neutrophils; this may require differentiation from so-called surgical necrosis in surgically resected wedge biopsy specimens. There is also a peculiar large type of cell with strongly eosinophilic cytoplasm and a bizarrely shaped, homogeneously staining nucleus. It is difficult to differentiate these cells from megakaryocytes, and their nature remains unclear. Foci of extramedullary hematopoiesis are found
Figure 71-5. Liver histology in neonatal hepatitis in an infant aged 47 days. There is expansion of the portal tract with prominent giant cells. (Hematoxylin & eosin ¥160.)
1370
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
more often in neonatal hepatitis than in EHBA. Hemosiderin deposition in parenchymal cells and Kupffer cells is more conspicuous in neonatal hepatitis than in EHBA. Intralobular inflammation and Kupffer cell hyperplasia are usually more striking in neonatal hepatitis, whereas portal and periportal inflammation predominates in EHBA. Cholestasis is variable but may be marked, with pigmented granules present in parenchymal cells and Kupffer cells, with intercellular bile plugs. Ductular proliferation at the periphery of portal tracts may be observed in some cases, but not to the extent found in EHBA. Intralobular fibrosis is common in neonatal hepatitis, in contrast to the periportal predominance of fibrosis in EHBA. The reason for formation of parenchymal giant cells is unknown. It has been suggested they represent a morphologic marker of infantile obstructive cholangiopathy.36 However, parenchymal giant cells occur in virtually all liver disorders of early infancy and may represent a non-specific reaction to various types of injury. It is more agespecific than disease-specific. However, parenchymal giant cells have also been observed in adults, in viral hepatitis, and drug-induced liver disease. Parenchymal giant cells may be the result of faulty development of hepatocytes with aplasia of intercellular canaliculi, possibly on a hereditary basis. Alternatively, these multinucleated cells may arise by nuclear multiplication not followed by cell division, resulting in multinucleated plasmodial cells. They may represent syncytial masses formed by fusion of several mononuclear hepatocytes. The spontaneous disappearance of giant cells is similarly unresolved. Serial biopsy specimens from infants with giantcell transformation suggest that giant cells have a limited lifespan of only around up to 6 months. Their disappearance parallels recovery from cholestasis. The bile ducts in idiopathic neonatal hepatitis appear normal. A few infants with severe inflammation may have small bile duct paucity.
Management and Treatment The specific management of neonatal hepatitis is dependent on the cause (see above). Nevertheless, all require nutritional support and promotion of bile excretion, as in EHBA (see below, General management of cholestatic liver disease). Follow-up is indicated until liver disease resolves.
Prognosis The outcome of neonatal hepatitis is highly variable, dependent on the etiology, but is usually good. It is better in sporadic cases without known associated factors, whereas the prognosis is poor in familial cases. Danks et al. reported that in sporadic cases approximately 60% recovered, 10% had persisting inflammation or fibrosis, 2% developed cirrhosis, and 30% died. In familial cases the outcome was worse: only 30% recovered, chronic liver disease with cirrhosis developed in 10%, and 60% died.91 Other predictors of poor prognosis include prolonged severe jaundice (beyond 6 months of age), acholic stools, persistent hepatomegaly, and severe inflammation on biopsy. Peak bilirubin level is not necessarily predictive of outcome, and the prognostic importance of ductopenia has not been rigorously investigated. Septic complications may lead to decompensation. Overall mortality is 13–25%. The long-term outlook for infants surviving the first year of life with little evidence of chronic liver disease is very good.
BILE DUCT PAUCITY ALAGILLE SYNDROME AGS, known as arteriohepatic dysplasia, is a complex inherited disorder, with cholestatic liver disease with paucity of intralobular bile ducts.92,93 It was originally noted by Daniel Alagille in 1969 and was formally described in 1975,92 as a subgroup of children with cholestatic liver disease with idiopathic bile duct paucity and complex congenital heart disease, most often peripheral pulmonary artery stenosis. The incidence of AGS is at least 1 in 70 000, although this may be an underestimate, reflecting only those with disease severe enough to be recognized clinically. Inheritance is autosomal dominant with variable penetrance. The phenotype has considerable variability, and can present from the neonatal period through to adult life depending on the severity of disease. It affects the sexes equally.
Etiology and Pathogenesis The underlying cause for AGS has only recently been discovered. It was known that some patients have deletions of chromosome 20p11-12, suggesting that the gene for AGS lies in that region. In 1997 the gene for Jagged1 was mapped to chromosome 20p12. It was tested as a candidate gene for AGS and coding mutations in Jagged1 were shown to be present in four families with AGS.94 A number of frameshift and splice donor mutations were also found at this time in the gene for Jagged1 in individuals with AGS.95 Further studies in greater numbers of patients with AGS have confirmed that there are a considerable number of different mutations in the Jagged1 gene (JAG1) identified in up to 70% of patients with AGS. Very few are total gene deletions (6%), a few are missense mutations (12%), and the majority (82%) are protein-truncating mutations.96,97 There is a high frequency of sporadic cases, up to 70%.98 Jagged1 is a ligand for the Notch receptor, part of a highly conserved and fundamental signaling pathway in the normal development of many organs. In humans there are at least five ligands for the Notch receptor. These are designated Jagged 1 and 2 and Delta 1, 3, and 4. There are also four known members of the Notch receptor family in humans, Notch 1 to Notch 4, known to be essential for normal development in other organs, for example in hematopoiesis and formation of the pancreas, as well as blood vessel formation. Both ligand and receptor are transmembrane, containing epidermal growth factor-like motifs. Mutations in JAG1 are thought to lead to protein haploinsufficiency, and depleted numbers of biliary epithelial cells.99 Alternatively, there could be a dominant negative effect of a resulting truncated Jagged1 protein produced from genetic mutations in JAG1 such that the abnormal protein antagonizes the activity of remaining wild-type protein, so inhibiting the normal function of the protein.98 However, as not all patients have these mutations, these theories do not fully explain the clinical picture. In mouse models, knockouts of JAG1 die in utero from vascular defects, with heterozygous mutations leading to very limited ocular defects.100 When mice are doubly heterozygous for a JAG1 null allele and a Notch2 hypomorphic allele, more clinical features of AGS manifest, with bile duct, cardiac, ocular, and renal abnor-
1371
Section XII. Inherited and Pediatric Diseases of the Liver
malities.101 The Notch2 gene may act as a genetic modifier to interact with a JAG1 mutation, possibly as a digenic disorder, but most likely to be in the form of polymorphisms in Notch2, which may influence the severity of AGS.101 While the role of Notch signaling in biliary development is not known, immunohistochemical studies indicate that Jagged1 is present in human fetal liver on portal vein,102,103 the ductal plate,28,104,105 and on BEC in normal pediatric28,105 and adult liver. Notch receptor expresson has been described in fetal,28 pediatric,28 and adult liver. There appears to be up-regulation of Notch 2 and 3 in AGS,31 although genetic studies to determine the role of any Notch mutations in AGS are outstanding. Overall there is poor genotype–phenotype correlation with JAG1 mutations in AGS. Even in families where there is a single mutation, the expression of the disorder ranges from mild to severe.94 Discordant phenotype has also been reported in monozygotic twins. One study has demonstrated that, in four sporadic cases, where JAG1 mutations occurred in the domain of the gene encoding for the binding region of Jagged1 to Notch receptor (the DSL region) leading to total deletion of the DSL domain, outcome was poor, with progressive liver failure requiring transplantation at an early age.106 There is also a single family with isolated heart disease in 11 individuals, where there is a mutation in exon 6 of JAG1. This is a missense mutation in a region necessary for ligand–receptor interaction.107
Clinical Features AGS was originally defined as paucity of intralobular bile ducts in association with at least three out of five other main clinical features: (1) cholestasis; (2) cardiac disease with pulmonary artery stenosis; (3) skeletal abnormalities seen as abnormally shaped butterfly vertebrae; (4) ocular abnormalities with posterior embryotoxon; and (5) characteristic facies.92,93 Since then, a wide range of further abnormalities have also been noted in this group of patients, including renal abnormalities, pancreatic insufficiency, growth and developmental delay, and an increased risk of intracranial hemorrhage.108,109 In the majority, AGS is fairly benign. Most patients with clinically important AGS have conjugated hyperbilirubinemia in the neonatal period.108 The liver disease in AGS most commonly presents in the neonatal period, with cholestatic jaundice in a similar manner to EHBA, or later in childhood with pruritus as a predominant feature. Earlier disease presentation correlates with poorer outcome.110 The cholestasis may be sufficiently severe to produce acholic stools and dark urine, and may be confused with EHBA. In the past some children with AGS have mistakenly undergone portoenterostomy. Chronic cholestasis with pruritus and fat malabsorption is occasionally exacerbated by exocrine pancreatic insufficiency. The other major features are as follows: Cardiac disease. Cardiac lesions are present in more than 95% of patients with AGS. They are mainly right-sided, the most common being peripheral pulmonary artery stenosis, and severe hypoplasia of the pulmonary artery branches. Other abnormalities include pulmonary aresia, tetralogy of Fallot, pulmonary valve stenosis, aortic stenosis, ventricular septal defect, atrial septal defect, truncus atreriosus, and anomalous pulmonary venous return. The severity of
1372
cardiac disease varies between patients and careful assessment is required, particularly if liver transplantation is contemplated. Skeletal abnormalities, including butterfly vertebrae due to failure of fusion on the anterior arch of the vertebral body, are found in the thoracic spine in approximately two-thirds of patients.111 Multiple vertebral anomalies may be seen in individual patients, including a decrease in the interpedicular distance in the lumbar spine, spina bifida occulta, and fusion of adjacent vertebrae.111 Absent 12th rib, shortened distal ulna and radius, shortened phalanges, and fifthfinger clinodactyly may also occur. Ocular abnormalities. These may be very diverse.112 The most common is posterior embryotoxon, an abnormal prominence of Schwalbe’s line (junction of Descemet’s membrane with the uvea at the angle of the anterior chamber), seen in up to 95% of patients.112 It requires slit-lamp examination to detect it. It is not pathognomonic for AGS as it occurs in 8–15% of the normal population. Optic disc drusen, calcific deposits in the extracellular space of the optic nerve head, are also common in AGS and are not seen in other cholestatic conditions. They are detected by ocular ultrasound examination. Abnormal retinal pigmentation without evidence of functional retinal degeneration may occur. Strabismus, iris abnormalities, ectopic pupil, microcornea, and hypotrophic optic disks with or without abnormal retinal vessels have been reported. Characteristic facies. Patients typically have a broad forehead, deep-set eyes, mild hypertelorism, straight nose, and a small pointed chin. The facies may not be evident in the first months of life, and the classic childhood appearance differs from the adult form where the face becomes elongated.113 This is a fairly “soft” feature and even experienced pediatricians may have difficulty in detecting them accurately.113 In addition failure to thrive in association with intrauterine retardation is common, with severe malnutrition present in up to half of patients. This may be part of the syndrome or secondary to fat malabsorption or gastroesophageal reflux. Minor features. A number of other features may be present in children with AGS. Renal disease may be present, including renal tubular acidosis, nephrolithiasis, or structural abnormalities such as small kidneys or congenital single kidney, or renal cystic disease.114 Histological examination may reveal a membranous nephropathy or lipid accumulation in the kidney (mesangiolipidosis). Renal artery stenosis has also been documented, and therefore monitoring children with AGS for hypertension is important. Non-cardiac vascular anomalies have been reported in 9% of patients with AGS.115 These include decreased intrahepatic portal vein radicals, coarctation of the aorta, aortic aneurysm, intracranial vessel abnormalities leading to cerebrovascular accident, and moyamoya disease.116 The most significant is intracranial bleeding, occurring in up to 15% of patients.108 In 30–50% bleeding is fatal. Hypercholesterolemia is common. Associated xanthomata can be disfiguring. Other documented associations include delayed puberty or hypogonadism; abnormal cry or voice; mental retardation, learning difficulties, or antisocial behaviour; hypothyroidism and pancreatic insufficiency; recurrent otitis media; recurrent chest infections, perhaps secondary to gastrointestinal reflux and aspiration pneumo-
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
nia; supernumery digital flexion creases; and craniosynostosis. Neurological abnormalities, such as peripheral neuropathy, may be related to vitamin E deficiency from severe chronic cholestasis.
Diagnosis The diagnosis of AGS is based on the above characteristic clinical features together with characteristic liver histology. Liver function tests. In neonates there is a conjugated hyperbilirubinemia, which may improve with age. AST and alkaline phosphatase concentrations are usually elevated by around 10 times normal values. g-Glutamyltranspeptidase concentration can be elevated 3–20 times above normal. Serum cholesterol and triglyceride may be raised to values three times or more the upper limit of normal. Serum albumin and prothrombin time are usually normal unless there is decompensated liver disease. Abdominal ultrasound may be normal or show a small contracted gallbladder. Radioisotope scanning may show delayed or no excretion if intrahepatic cholestasis is severe. This has led to confusion with EHBA in the past and even to portoenterostomy, especially if the liver biopsy does not show classical bile duct paucity. Liver biopsy. Histology shows that atresia of the intrahepatic bile ducts is usually not complete but consists of a reduced ratio in the number of interlobular bile ducts to the number of portal tracts (Figure 71-6). A sufficient biopsy specimen is therefore required to evaluate the number of bile ducts effectively. Alagille et al. have recommended a surgical liver biopsy in order to obtain sufficient numbers of portal tracts,92 although this is not a universal opinion. For optimal evaluation between 6 and 20 portal tracts should be present. There is hypoplasia of intrahepatic bile ducts if the ratio of interlobular bile ducts to the number of portal areas is less than 0.5
(normally 0.9–1.8). It may be lower in premature infants. The portal areas without bile ducts usually appear hypoplastic as a whole. The total number of portal tracts per square millimeter of tissue is also reduced. Diagnostic confusion can arise when there is both EHBA and hypoplasia of interlobular bile ducts, and care must be taken in differentiating the two conditions. This is especially important as interlobular bile duct paucity is not always present on the initial biopsy in AGS, particularly in young infants. During the first few weeks of life, interlobular bile ducts can be demonstrated in the portal areas of liver biopsy specimens from patients with AGS, but their number gradually diminishes over time,117 perhaps due to destructive inflammation. In some patients there is an almost complete lack of interlobular ducts during the first 4 weeks of life, whereas in others paucity develops later.119. Ductular proliferation may also be seen during the phase of inflammatory destruction. Hence paucity of interlobular bile ducts may be unreliable in the earlier stages of AGS.117 Bile ducts are infiltrated with mononuclear leukocytes, and the epithelial lining cells of the bile ducts show desquamation and necrobiosis. Gradually, the lumen is obliterated by connective tissue, with fading of the inflammatory component. Periportal or centrilobular fibrosis is usually absent in infancy but progressive disease with biliary cirrhosis develops in 15–20% of patients. If there are problems with confirming the diagnosis histologically, an exploratory laparotomy may be necessary to confirm patency of the extrahepatic ducts, although the presence of extrahepatic manifestations (characteristic facies, vertebral abnormalities, heart murmur, and posterior embryotoxon) is helpful. Genetic testing. This is not yet widespread, and is currently most often used as a research tool, investigating the types and number of
Figure 71-6. Liver histology in Alagille syndrome in an infant aged 69 days. There are prominent arteries but no bile ducts visible. (Hematoxylin & eosin ¥160.)
1373
Section XII. Inherited and Pediatric Diseases of the Liver
JAG1 mutations present in a given population. However, in some families it may be important. Kamath et al.118 have shown that in 34 patients with AGS and known JAG1 mutations, when 53 family members were studied, 11 (21%) of mutation-positive relatives had features diagnostic of AGS, and another 17 (32%) had mild features suggestive of AGS. This obviously has important implications for genetic counseling and certainly implies that a detailed history and examination is necessary to exclude AGS in first-degree relatives of newly diagnosed patients with AGS.
Management Management depends on the severity of associated extrahepatic disease and cholestasis. Cholestasis requires supportive management (see below). Cardiac anomalies may require corrective surgery, with balloon dilatation or surgical correction of pulmonary valve or pulmonary artery stenosis. Detailed discussion of the varying management strategies is outwith the scope of this book. Close liaison with an experienced cardiologist may be necessary, especially where transplantation is being considered. Outcome from transplantation in general, even with elevated right ventricular pressure, appears to be good. Nutrition and growth. Nutritional support is vital. Children with AGS often have poor growth.109 Rovner et al. recently demonstrated that in a group of 26 prepubertal children with AGS, more than half were less than the 5th percentile for height and/or weight and 20% had inadequate calorie and fat-soluble vitamin intake contributing to this despite being recommended a high-calorie, high-fat diet by their physicians.119 In particular poor calcium intake correlated with lower height–age z-scores. This highlights the ongoing need to optimize nutritional intake. Feeding via a nasogastric tube may be necessary to try and obtain optimal growth. Additional fat-soluble vitamins are required. Hypercholesterolemia is related to severity of hyperbilirubinemia and is difficult to treat but may partially respond to a modified-fat diet and bile salt resins in some children, although not all respond. In severe cases associated with disabling pruritus, partial external biliary drainage may lead to resolution of xanthomata and decrease cholesterol levels significantly. The use of statins is not recommended in young children and it is not yet known whether they will effectively decrease any cholesterol-related cardiac complications in older patients. Pruritus is often far more severe than in other pediatric cholestatic disease, and is ongoing in up to 80% of patients.110 Renal disease, including renal tubular acidosis, nephrolithiasis, and structural abnormalities, requires specific management as indicated. Liver transplantation. The indications for liver transplantation include the development of cirrhosis and chronic liver failure, marked deterioration in quality of life, such as intense refractory pruritus, or malnutrition with recurrent bone fractures despite optimal nutrition.110 Occasionally early liver transplantation is indicated prior to the development of severe pulmonary hypertension. Children with AGS are prone to bleeding episodes without definite abnormalities of coagulation. Special caution must be exercised with respect to head trauma.
1374
Prognosis Most estimates put overall mortality at 20–30%, due to cardiac disease, intercurrent infection, or progressive liver disease,108,110 the outcome depending on disease severity. The largest study of outcome showed that in 163 children with AGS and liver involvement, 44 (33%) required liver transplantation, with presentation with neonatal cholestasis more likely to result in a worse outcome.110 Actuarial survival rates with native liver were 51% and 38% at 10 and 20 years respectively, and overall survival rates were 68% and 62% respectively. Liver transplantation should be reserved for patients with hepatic failure, intolerable pruritus unresponsive to medical treatment, and severe growth failure. Catch-up growth after transplantation may occur, but not in all cases.
NON-SYNDROMIC PAUCITY OF BILE DUCTS Paucity of bile ducts is seen in a number of conditions other than AGS. These disorders fall into the broad categories of infection, genetic (with chromosomal abnormalities), and metabolic diseases. When idiopathic neonatal hepatitis is clinically severe, bile duct paucity may also be present. Among congenital infections CMV is the most important cause. Chromosomal abnormalities associated with duct paucity include trisomy 18 and 21. Metabolic disorders associated with duct paucity in the infant are diverse and include a1AT deficiency (which usually indicates more severe liver disease and a poor prognosis), progressive familial intrahepatic cholestasis (PFIC), and, rarely, cystic fibrosis or Zellweger syndrome. Duct paucity may also develop in late stages of EHBA following a Kasai portoenterostomy or in primary sclerosing cholangitis. Where no specific associated condition can be found, then isolated non-syndromic bile duct paucity can be diagnosed. These children are supposed to have a less favorable outlook than children with AGS, with persistent severe cholestasis and progressive liver damage. The relationship of childhood non-syndromic duct paucity to idiopathic adult ductopenia, which has recently been described and may be familial, remains uncertain.
HEREDITARY CHOLESTASIS WITH LYMPHEDEMA (AAGENAES) Aagenaes syndrome is a very rare disorder with cholestasis and lower-limb edema. It was initially reported in a Norwegian kindred but has also been reported in children of Norwegian descent and in other ethnic groups. A neonatal hepatitis evolves to a chronic cholestatic condition with pruritus and fat-soluble vitamin deficiencies requiring treatment. While the initial cholestasis resolves in early childhood, recurrent bouts of cholestasis, similar to benign recurrent cholestasis, and lymphedema become a prominent problem in adulthood. Chronic liver disease with portal hypertension has not been reported. Abnormal development of hepatic lymphatics has been postulated as part of the pathogenesis of this condition. The lymphatic disorder may be due to abnormal development of hepatic lymphatics and can present later than the cholestasis with localized lower-limb lymphedema despite normal serum albumin, or hemangioma(s) and/or lymphangioma(s).
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
The genetic basis of this familial cholestatic disorder remains unknown, but the genetic locus has been mapped to chromosome 15q.
PROGRESSIVE FAMILIAL INTRAHEPATIC CHOLESTASIS PFIC is a group of autosomal recessive disorders with characteristic clinical, biochemical, and histological features. Much recent progress has been made in the understanding of the genetic basis of some of these disorders, as well as the reclassification into specific subgroups with inborn errors of bile acid biosynthesis and bile acid transport. The control of bile salt metabolism is complex. Bile formation at the level of the canalicular membrane is driven by adenosine triphosphate (ATP)-dependent transporters. These include the ATPbinding cassette (ABC) proteins MDR1 and MDR3, and the multidrug resistance-associated protein (MRP2). The main bile-salt export pump is BSEP, mediating mainly secretion of conjugated cholic acids and other hydrophobic bile acids. It has been proposed that familial intrahepatic cholestasis-1 (FIC1) protein may mediate transport of the more hydrophilic bile acids. Other genes involved in bile salt metabolism include the apical sodium-dependent bile acid transporter (ASBT) and the farsenoid-X-receptor (FXR), a bile acid-activated transcription factor that mediates transcriptional repression of genes important in bile acid and cholesterol homeostasis. This section will deal with the three types of PFIC associated with specific gene defects in one of the members of the ABC superfamily: (1) PFIC1 (known originally as Byler disease) and benign recurrent intrahepatic cholestasis (BRIC), caused by mutations in the FIC1 gene; (2) PFIC2, now renamed as BSEP deficiency due to mutations in the BSEP gene and previously known as Byler syndrome; and (3) PFIC3, with abnormalities in the MDR3 gene.
PFIC1 GENETICS PFIC1 was first described in the Amish population. It was originally known as Byler disease, as the patients were all descendants of Jacob and Nancy Byler.120 Recently PFIC1 was discovered to be due to mutations of the coding sequence of the FIC1 gene (ATP8B1). This is located on chromosome 18q21. Roughly 50% of all mutations described to date are missense mutations, with point mutations often affecting conserved amino acid residues. The FIC1 gene encodes the FIC1 protein,121 which is expressed in many tissues, including the liver, pancreas, and small intestine.121 In the liver, FIC1 is localized to the canalicular membrane of hepatocytes and cholangiocytes. FIC1 is a P-type adenosine triphosphatase (ATPase). These are membrane-bound enzymes that transport a variety of substances across the lipid bilayer using ATP hydrolysis. They are evolutionarily well conserved and are of at least three types: FIC1 is a P4 Ptype APTase. The function of FIC1 is incompletely understood, but it may act as a bile transporter. Abnormalities in FIC1 may therefore act to decrease biliary secretion of bile acids. This would explain the low biliary bile acid concentration seen in PFIC1.
When FIC1 is defective, transport of bile acids in the liver and the gut is impaired, resulting in cholestasis and watery diarrhea. A mouse model for the homozygous form of PFIC1, the Atp8b1G308V/G308V mutant mouse, has been developed. These mice have unimpaired bile secretion and no liver damage, until exogenous bile salts are administered. After bile-salt feeding, they demonstrate serum bile salt accumulation, hepatic injury, and expansion of the systemic bile-salt pool.122 This occurs in the absence of any impairment in canalicular bile secretion. It is therefore possible that mutations in FIC1 may also allow increased bile acid uptake by the gut. FIC1 is more highly expressed in the small intestine than the liver. It has recently been suggested that there is also aberrant regulation of ASBT in the gut. Chen et al. found increased ASBT mRNA in children with PFIC1. This was associated with down-regulation of FXR.123 FIC1 appears to activate FXR gene product, possibly by post-translational modifications, leading to nuclear translocation of FXR. Loss of FIC1 is associated with decreased nuclear transport of FXR. Decreased FXR may lead to changes in bile acid transporter expression as bile acids are the endogenous ligand for FXR. This in turn could lead to decreased activation of BSEP. In addition, the increased ASBT could lead to aberrant enhanced ileal uptake of bile salts, further contributing to cholestasis.123 Mutations in the ATP8B1 gene also give rise to BRIC and Greenland familial cholestasis.
CLINICAL FEATURES PFIC1 often presents with recurrent episodes of intrahepatic cholestasis without anatomical obstruction in the first 6 months of life, leading to permanent cholestasis, intractable pruritus, fibrosis, and eventually to liver failure in the first decade.124 Other features include short stature, diarrhea, pancreatitis, and occasionally hearing loss. Diarrhea persists after liver transplantation, presumably because of the defective transporter in the ileum. In some cases it may become intractable when the bile salt secretion is restored. Short stature may also not be improved by liver transplantation and this should be explained to families.
INVESTIGATION Investigation of familial intrahepatic cholestasis is as described above and can differentiate between PFIC3 and either PFIC1 or PFIC2 (Table 71-3). Specific immunohistochemistry may also differentiate between the variants but is not widely available. Genetic testing is the only reliable method of distinguishing between PFIC1 and 2 and may become more widespread in the future. In all types of familial intrahepatic cholestasis serum bile salt concentrations are high and cholesterol is normal. An ultrasound should also be performed to exclude any anatomical obstruction.
Liver Biopsy in PFIC1 Patients from the original Byler kindred were found to have minimal giant-cell transformation, without any bile duct paucity or ductular reaction, with bland intracanalicular cholestasis.125 Initial biopsy often shows only mild anormalities, with unremarkable canalicular cholestasis with only minimal hepatocyte degeneration, bile duct
1375
Section XII. Inherited and Pediatric Diseases of the Liver
Table 71-3. Investigation of the cholestatic infant 1. Blood Liver function tests Electrolytes Metabolic investigations
Vitamin levels Hematology Serology
Total and unconjugated bilirubin, AST, ALT, albumin, total protein, ALP, GGT Sodium, potassium, urea, creatinine, calcium, phosphate, magnesium, bicarbonate, glucose Fasting bloods for cortisol, glucose, lactate, 3-hydroxybutyrate, free fatty acids and amino acids; cholesterol and triglycerides Galactosemia and tyrosinemia screen; thyroid function tests, parathyroid hormone; chitotriosidase; immunoreactive trypsin; a1-antitrypsin level and phenotype Vitamins A and E Full blood count with differential and reticulocyte count; clotting screen including fibrinogen; blood group and Coombs test Toxoplasma antibodies, CMV IgM, adenovirus antibody and consider herpes simplex PCR, hepatitis B/C serology, syphilis
2. Urine General Clinical chemistry Virology Microbiology
Urine pH, glucose, ketones, and protein. Protein-to-creatinine ratio if protein is present Amino acids and organic acids; reducing substances; bile acids (if possible); tubular reabsorption of phosphate CMV culture Urine microscopy and culture
Specific metabolic tests
3. Stool Sample should be inspected for pigment 4. Radiology Abdominal ultrasound (after a 4-h fast) TIBIDA scan if acholic stools 5. Liver biopsy for histology, electron microscopy and snap-frozen and stored if suspicion of metabolic disorder Investigation
PFIC1
BRIC
PFIC2
GGT Transaminases Liver biopsy histology
Low/normal <10 ¥ normal Minimal giant-cell transformation; intracanalicular cholestasis; no ductular proliferation minimal inflammation; fibrosis late
Low/normal <3 ¥ normal Normal or cholestasis; no ductular proliferation; minimal inflammation; fibrosis rare
Low/normal <10 ¥ normal Giant-cell transformation; intracanalicular cholestasis; no ductular proliferation; moderate inflammation; fibrosis
Liver biopsy Electron microscopy
Coarsely Granular bile within the canaliculus
—
Morphous/filamentous bile within the canaliculus
Immunohistochemistry Genetics
Mutations in FIC1 on chromosome 18q21
Mutations in FIC1 on chromosome 18q21
Absent canalicular BSEP staining Mutations in BSEP on chromosome 2q4
PFIC3 High (>10 ¥ normal) 5 ¥ normal Giant-cell transformation; intracanalicular cholestasis; ductular proliferation; moderate inflammation; marked fibrosis —
Absent canalicular MDR3 staining Mutations in MDR3 on chromosome 7q21
AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; GGT, g-glutamyltransferase; CMV, cytomegalovirus; IgM, immunoglobulin M; PCR, polymerase chain reaction; TIBIDA, trimethylbromoiminodiacetic acid; BSEP, bile-salt export pump.
proliferation, or fibrosis. Later the liver architecture may be distorted, with giant-cell transformation and some inflammation. Fibrosis and subsequent cirrhosis develop with increasing age, without ductular proliferation. On electron microscopy there is a paucity of canalicular microvilli, and a thickened canalicular network of microfilaments and coarse granular bile called Byler bile.
term amelioration of pruritus with this treatment, and even catchup growth. Histological changes may also show some improvement. Liver transplantation is successful for PFIC1. It is not, however, a perfect treatment. There may not be any catch-up growth, liver steatosis has been described, and there may be worsening of diarrhea with poor quality of life.126 Pancreatits may occur for the first time following transplantation.
MANAGEMENT Treatment is unsatisfactory. Cholestyramine has been used in the past, but with little success. Ursodeoxycholic acid may decrease cholestasis, but its effect on disease progression is unknown. Rifampicin may be more effective. Partial biliary diversion may be effective in some patients, resulting in a significant decrease in symptoms. If performed early enough it may even delay or interrupt hepatic injury. There may be long-
1376
BRIC GENETICS BRIC is also due to mutations in the ATP8B1 gene. One missense mutation is common in BRIC patients in western Europe, with considerable variation in phenotype. Other mutations are rare. However, there is evidence that there may be locus heterogeneity
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
in BRIC as families have been described where it did not map to chrolmosome 18q21. The alternative loci are as yet undefined.
PRESENTING FEATURES BRIC was first described in 1959. It can present with the first episode of cholestasis at any age, from 2 months old to middle age. Patients tend to present with pruritus, although a subgroup of up to 15% may not have pruritus and sometimes, but not always, jaundice. Attacks are associated with pregnancy and the oral contraceptive pill. The number of attacks and frequency also vary considerably, from annually to less than one attack per decade. Each attack may last weeks to months and be severely debilitating.127 During attacks there is severe jaundice, pruritus, steatorrhea and weight loss. In between attacks patients are completely asymptomatic. The frequency of episodes may decrease with age. There is a small subgroup that may eventually develop progressive disease similar to PFIC1, suggesting that there is a continuum rather than strict separation of the two disorders.128
INVESTIGATION See Table 71-3.
logical improvement.130. In extreme cases of BRIC liver transplantation for severe pruritus may be necessary.
PFIC2 GENETICS PFIC2 was discovered when patients thought to have PFIC1 did not have mutations on chromosome 18. PFIC2 is due to mutations in the BSEP or ABCB11 gene on chromosome 2q24.131 This gene encodes the canalicular BSEP, a P-glycoprotein belinging to the ABC transporter superfamily.132 This protein is liver-specific and is located in the canalicular domain of the hepatocyte plasma membrane. There are many mutations in the BSEP gene that give rise to cholestasis132,133 and heterogeneous abnormalities in BSEP protein.134 Two mutations are common; E297G and D482G, found in 25 and 16 families of European descent, respectively. Affected individuals may be homozygotes or compund heterozygotes, implying that most patients have sporadic disease. A defective BSEP is expressed which leads to impaired bile-salt secretion, accumulation of bile salts in hepatocytes, and subsequent hepatocellular injury, apoptosis, and/or necrosis. The D482G mutation has recently been found to produce a functional BSEP, which is highly unstable and temperature-sensitive.
Liver Biopsy This may be entirely normal. Alternatively it may show signs of cholestasis with minimal or no inflammation. There is usually no progression to chronic liver disease. Fibrosis is rare.
MANAGEMENT Ursodeoxycholic acid in BRIC patients does not prevent attacks, although it may reduce duration.129 Partial external biliary diversion led to clinical improvement in three patients, together with histo-
CLINICAL FEATURES It is not always easy to differentiate between PFIC1 and PFIC2 on clinical presentation. The presentation and progression of PFIC2 may be more severe than PFIC1. In PFIC2 jaundice is usually continuous from the outset, and the disease is associated with nonspecific giant-cell hepatitis. Progression to decompensated liver disease is often rapid, requiring liver transplantation in the first few years of life. Other clinical features are also similar to PFIC1, with
Table 71-4. Investigation and management of nutritional deficiencies in cholestatic liver disease Nutritional deficit
Investigation
Management
Protein
Plasma proteins (albumin) BCAA/AAA ratio Protein stores: muscle mass; TBN Triceps skinfold thickness Body composition EFA deficiency Plasma lipid profile Calorie intake Energy expenditure Anthropometry: body cell mass (TBK); DEXA Vitamin D: plasma 25-OH-D; skeletal X-rays; DEXA Vitamin K: prothrombin time Vitamin E and A: plasma levels
Ensure adequate protein (3–4 g/kg per day) BCAA-enriched protein (32%) Albumin infusion if serum albumin <25 g/l Change fat intake to 50 : 50 MCT/LCT Provide saturated fats high in EFA ? Supplement DHA Increased calorie intake to 130–150% EAR Overnight enteral feeding Continuous enteral feeding Light exposure Vitamin D1a (50 ng/kg) Vitamin K (2.5–5 mg/day) Vitamin E (50–400 IU/day) (as TPGS) Vitamin A (5000–10 000 IU/day) Supplement as required
Fat
Energy/carbohydrate
Fat-soluble vitamins
Water-soluble vitamins Trace elements
Specific levels Full blood count Specific levels Cardiac evaluation
Supplement as required
BCAA, branched-chain amino acids; AAA, aromatic amino acids; TBN, total body nitrogen; EFA, essential fatty acids; MCT, medium-chain triglycerides; LCT, long-chain triglycerides; DHA, docosahexanoic acid; TBK, total body potassium; DEXA, dual-energy X-ray absorptiometry; TPGS, tocopherol polyethyleneglycol-1000 succinate
1377
Section XII. Inherited and Pediatric Diseases of the Liver
severe pruritus, growth failure, and cholelithiasis.133 Diarrhea is not a prominent feature.
INVESTIGATION The results of investigation in this disorder are similar to PFIC1 (Table 71-3).
available, in order to monitor patients more effectively and determine the most appropriate treatment. Genetic defects in MRD3 are also responsible for intrahepatic cholestasis of pregnancy and probably also in cholesterol gallstone formation.
PRESENTING FEATURES Liver Biopsy The biopsy findings may be similar to PFIC1. However, the liver architecture is more disturbed. The initial biopsy may show nonspecific giant-cell hepatitis, similar to idiopathic neonatal hepatitis. Canalicular cholestasis is demonstrable but with no real ductular proliferation. There is periportal metaplasia of hepatocytes. This progresses to show more inflammatory activity, giant-cell transformation, and lobular and portal fibrosis than PFIC1. In the early stages bile duct loss occurs and later hepatocellular carcinoma can develop. The bile in PFIC2 is amorphous or filamentous on electron microscopy. In the majority of patients immunohistochemical staining for canalicular BSEP is negative.
Presentation is usually with jaundice, pale stools, hepatosplenomegaly, or pruritus. However short stature and pruritus may be less noticeable than in PFIC1 or PFIC2. Age at presentation varies from infant onset to adulthood. In a recent review of 31 patients, clinical features of cholestasis were noted within the first year of life in 12, although only in two presented in the neonatal period.136 Older patients are more likely to present with signs of portal hypertension, with variceal bleeding. Progression is usual, with chronic cholestasis and development of liver failure.
INVESTIGATION Investigative features of PFIC3 are very different from PFIC1 and 2 (Table 71-3).
Liver Biopsy MANAGEMENT Patients with PFIC2 do not respond to ursodeoxycholic acid and may have an increase in symptoms or a rise in bile acids without any increase in biliary bile acid secretion. Partial external biliary diversion may be useful, and in the majority produces a significant improvement in symptoms. If performed early enough it may even delay or interrupt hepatic injury. There may be long-term amelioration of pruritus with this treatment, and even catch-up growth. Histological changes may also show some improvement. Liver transplantation is successful in this group of patients.133 There has, however, been a report of recurrence after living related transplant.
PFIC3 GENETICS Patients with PFIC3 have a defect in the multidrug resistance 3 gene (MDR3) on chromosome 7q21.135 MDR3 is a class III multidrug resistance P-glycoprotein. It is an ABC transporter, and acts as phospholipid translocator involved in biliary phospholipid (phosphotidylcholine) excretion, expressed mainly in the hepatocyte canaliular membrane.136 MDR3 was initially suspected to be responsible for PFIC3 as the histological features are similar to mice with a homozygous disruption to mdr2 (the murine equivalent of MDR3), the mdr2–/– mouse. Several different types of mutation have been discovered: It appears that those with missense mutations may have a milder form, more likely to respond to ursodeoxycholic acid and less likely to require liver transplantation. Those where the resulting protein is truncated are more likely to require early transplantation.136 This may be of use in the future when genetic testing is more widely
1378
There is portal fibrosis with a mixed inflammatory infiltrate, together with giant-cell transformation of hepatocytes. Inflammatory changes are demonstrable from the early stages despite the patency of intra- and extrahepatic bile ducts and a normal cholangiogram. Cholestasis may be present in the lobules and there is extensive ductular proliferation in the portal tract.135,136 Portal and periportal fibrosis are often present. In older patients there is more extensive fibrosis and biliary cirrhosis. Genetic testing is not available.
MANAGEMENT Ursodeoxycholic acid is useful in around half of affected patients, with either normalization or improvement in liver function tests.136 Children with a missense mutation in MDR3 appear to have less severe disease than those with a mutation leading to a truncated protein. This seems to have a later onset and is more likely to respond to ursodeoxycholic acid. This may be due to residual transport activity of MDR3 where there is a missense mutation. Where this fails liver transplantation may be necessary. In a study of 31 patients, liver transplantation was performed in 18 patients for liver failure, persistent choelstatic jaundice, or severe portal hypertension at a mean age of 7.5 years.136 In the future it may be possible to obtain a genetic diagnosis, and where there is a missense mutation to commence ursodeoxycholic acid, and where there is truncation of protein to consider whether gene therapy is an option.
OVERALL OUTCOME Many patients have not yet undergone genetic testing to determine accurately whether they have PFIC1 or PFIC2 and thus it may be difficult to determine a difference in outcome for these two disorders.
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
INBORN ERRORS OF BILE ACID SYNTHESIS Defects in bile acid synthesis may resemble PFIC2 and have only been distinguishable with recent genetic diagnostic advances. They have been termed PFIC4 by some authors. A defect in 3b-D5-C27-hydrosteroid oxoreductase has been described as a cause of giant-cell hepatitis. Other deficiencies leading to neonatal hepatitis and cholestasis are of D4-3-oxosteroid5b reductase and 3b-hydroxy-D5-steroid dehydrogenase isomerase. In these disorders toxic intermediates form. These lead to intrahepatic cholestasis and a clinical picture similar to neonatal hepatitis. They may have a normal or raised gglutamyltransferase and low or elevated serum total bile acids. Other genetic forms of cholestasis also exist which may be due to disorders of hepatic transport. These include Zellweger syndrome and microfilament dysfunction.
METABOLIC DISORDERS PRESENTING WITH CHOLESTASIS A number of metabolic disorders may present with a cholestatic picture in the neonatal period (Table 71-1). Of the disorders of lipid metabolism, Niemann–Pick type C in particular must be excluded if there is hepatosplenomegaly. Gaucher’s and Wolman’s should also be considered. Of the carbohydrate metabolism disorders, galactosemia must be considered, especially where there is septicemia, as, although jaundice is most often unconjugated, cholestasis may occur. Galactosefree milk should be used until the diagnosis is excluded. Rarer disorders of carbohydrate metabolism include fructosemia and glycogenosis III/IV. Other rare disorders include lymphohistiocytosis. The more common disorders to consider, however, include a1-AT (see below) and cystic fibrosis. Both of these must be excluded as they may coexist with other conditions.
a1-ANTITRYPSIN DEFICIENCY For a detailed description of this disorder, see Chapter 68. However, it is worth mentioning in the context of infant cholestasis.
Presentation a1AT presents as a conjugated hyperbilirubinemia in infancy in 10% of PiZZ patients. It is the commonest genetic cause of liver disease in children and the most frequent genetic diagnosis leading to liver transplantation. The most common presentation is with jaundice following on from physiological jaundice. Jaundice may, however, commence at any time within the first 4 months of life and usually lasts around 3 months, but occasionally up to 1 year. Stools may be pale, and urine dark. Most of these babies are small for gestational age and may show poor weight gain. Hepatomegaly is usually present with splenomegaly in around 50%. There may be pruritus.
Differentiation from EHBA is important as histology may be similar. Features in favor of a1AT over EHBA include low birth weight and stools that are incompletely acholic (Table 71-1). In approximately 5% of those presenting in infancy, a1AT tragically presents as late hemorrhagic disease of the newborn (HDN). Presentation is with bleeding between 2 and 6 weeks of age. Minor bleeds may have been overlooked initially and the child may suffer an intracranial bleed, with its long-term consequences. Vitamin K is traditionally administered at birth to try and prevent this. This is particularly important for breast-fed babies, as there is little vitamin K in breast milk, and any fat malabsorption secondary to cholestasis may leave them vulnerable to clotting abnormalities. Since the controversy surrounding the use of intramuscular vitamin K, oral vitamin K regimes have been introduced. These vary in dosage, frequency, and type of vitamin K preparation administered. In many units four doses of oral vitamin K are given (2 mg at birth, and at 2, 4, and 6 weeks of age) but late HDN is still seen, with an incidence of two in 30 breast-fed infants with a1AT.137 In Denmark, vitamin K is given orally on a weekly basis until children are 3 months old, if they are mainly breast-fed (2 mg at birth, then 1 mg weekly). This has resulted in no cases of HDN due to vitamin K deficiency from 1992 to 2000.138 The vitamin K deficiency seen in cholestasis is rapidly correctable with parenteral vitamin K. The reasons for very early presentation in only a subgroup of patients with a1AT are not clear. It has been suggested that there may be an association with intrauterine infection, as these babies tend to be small for gestational age. However this has not been substantiated. As mentioned above, breast-fed babies are more likely to suffer the effects of deranged clotting due to vitamin K deficiency and may present earlier than their bottle-fed counterparts. An alternative theory is that genetic differences in rate of polymer degradation occur and may account for the variability in presentation, with those presenting early and developing more significant liver dysfunction having far slower degradation rates of the abnormal Z phenotype.137 Patients with a1AT who develop neonatal cholestasis are more likely to have more severe abnormalities than those who develop liver dysfunction later in childhood. Although some recover completely, others develop cirrhosis, and up to a third develop end-stage liver disease in childhood.139 Progressive liver disease in childhood is associated with more prolonged infantile jaundice (more than 6 weeks), more severe derangement of liver transaminases, and fibrosis or even cirrhosis on the initial liver biopsy. Both clinical presentation and liver histology may mimic EHBA, with marked ductular reaction in the portal tracts.
Investigation Liver function tests. There is a conjugated hyperbilirubinemia, which may improve with age. AST and alkaline phosphatase concentrations are usually elevated up to around 10 times normal values. g-Glutamyltranspeptidase concentration can be elevated up to five times normal. a1AT serum level. The serum level in patients with PiZZ phenotype is often reduced to <0.6 g/l (normal range 0.8–1.8 g/l). However, as a1AT is an acute-phase reactant and therefore may be
1379
Section XII. Inherited and Pediatric Diseases of the Liver
artificially elevated in liver inflammation, these levels cannot be relied upon and phenotype must be obtained in all patients with cholestasis.137 a1AT protease inhibitor (Pi) phenotype. This is assessed by isoelectric focusing on polyacrylamide gels. The normal phenotype is PiMM and the commonest homozygote form leading to a1AT is PiZZ. Other forms may also result in liver disease, including the PiSZ phenotype. Note that it is essential that the test be performed in an experienced laboratory. CMV infection may cause a spurious Z band. Genotype. This is only available in reference laboratories and is not a commonly used diagnostic tool. There are PCR primers available for M, Z, and S alleles. Liver biopsy. In early infancy there is an acute hepatitis of varying severity. This may resemble idiopathic neonatal hepatitis, but giant cells are rarely prominent. Liver histology may also mimic EHBA, with marked ductular reaction in the portal tracts. Fatty infiltration may be seen around portal tracts. There is hepatocellular necrosis and inflammatory cell infiltrate. Fibrosis may be present, with or without portal bridging. Cirrhosis has been described as early as 8 weeks of age. There may be periodic acid–Schiff-positive diastase-resistant granules in hepatocytes. These are 2–20 nm in diameter and correspond to amorphous material within the endoplasmic reticulum, seen on electron microscopy. However, they may not be prominent in early biopsies, and only become marked after 3 months of age. With increasing age, the cholestasis, inflammation, and hepatocellular necrosis begin to settle. By 1–2 years old inflammation becomes limited to expanded portal tracts and adjacent hepatocytes. Where cholestatic features and fibrosis have been prominent cirrhosis may develop. However, if there is little early fibrosis, or if there is paucity of interlobular bile ducts on the intial biopsy, then cirrhosis is less likely to develop.
Management Management of neonates presenting with a1AT includes the general management of cholestatic liver disease (see below). Nutritional support and fat-soluble vitamin supplementation are important. Close follow-up throughout childhood at a specialist center is mandatory, in order to detect signs of progressive liver disease and the possible need for transplantation. It is important to counsel the family as this is a recessive disorder. All siblings should be screened and parents informed of the risks for future pregnancies. Indications for transplant. Overall, a1AT is the second commonest indication for liver transplantation in childhood. Francavilla et al. documented that, of 26 children with end-stage liver disease due to a1AT, 21 had neonatal onset of disease, presenting at a median age of 2.1 months.139 Eighteen of the 21 presenting as neonates had jaundice for more than 6 weeks. However, some children with early onset of liver disease are stable with only slow progression of liver disease. Thus transplantation should be avoided unless there is evidence of liver decompensation, hence the need for close follow-up.
1380
INTESTINAL FAILURE-ASSOCIATED LIVER DISEASE (IFALD) For a detailed description of this disorder, see Chapter 57. IFALD is more common in infants than in adults or older children and is the commonest indication worldwide for combined liver and intestinal transplantation. The survival of infants with intestinal failure has improved greatly in recent years due to improvements in both neonatal intensive care and in PN.140 Parenteral solutions are now safer to use, while improvements in catheter design and placement techniques have reduced septic complications. Hepatobiliary dysfunction remains a significant life-threatening complication and is an important indication for combined liver and small-bowel transplantation. The underlying causes of intestinal failure in infants include anatomical abnormalities, surgical excision of the bowel due to complications of prematurity, gastroschisis, intestinal volvulus, or dysmotility due to primary or secondary intestinal pseudo-obstruction. The spectrum of IFALD differs in adults and children. Cholestasis occurs in 40–60% of infants, steatosis in 40–55% of adults, while biliary sludge and cholelithiasis occur in both adults and children. The etiology is multifactorial and the incidence varies according to the population.
ETIOLOGY OF CHOLESTASIS In infants, the development of cholestasis is associated with prematurity, sepsis, and possibly lack of enteral feeding.141 In adults, it is related to the length of the bowel (<50 cm) and the concentration of lipid infusions. Many studies have noted the close relationship between the development of IFALD, prematurity, and low birth weight.141,142 As many infants requiring PN are likely to be premature with low birth weight, it is difficult to ascertain whether these are independent risk factors or not. However, when Beale et al. reviewed 62 premature infants on PN, the overall incidence of cholestasis was 23% but infants receiving therapy for more than 60 days had an incidence of 80%, increasing to 90% in those treated for more than 3 months.142 The incidence of cholestasis was 50% in infants with birth weight <1000 g but fell to 7% if birth weight was >1500 g. This has been confirmed by Beath et al.,141 who found the highest incidence in infants less than 34 weeks’ gestation and who weighed less than 2 kg. The increased incidence of IFALD in premature babies suggests that the development of disease may be related to immaturity of the neonatal liver. It is known that in premature infants the total bile-salt pool is reduced. There is both diminished hepatic uptake and synthesis of bile salts and a reduced enterohepatic circulation as compared to full-term infants or adults. It is possible that other essential components of bile secretion such as glutathione may be reduced in the newborn as hepatic glutathione depletion has been demonstrated in young animals on PN. Sulfation, an important step in the solubilization of toxic bile salts such as lithocholic acid, is also deficient in the fetus and neonate. It is therefore likely that the liver and biliary system of the premature infant is more susceptible to toxic damage of any kind. IFALD is more common in neonates who have recurrent episodes of sepsis, whether this is related to central-line infections or bacte-
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
rial translocation from bacterial overgrowth.141 Bacterial overgrowth from intestinal stasis has been related to the development of IFALD, possibly due to the combination of the reduction in bile flow, the production of secondary bile salts, and sepsis from bacterial translocation. Hepatic failure also appears closely related to septic episodes and/or peritonitis. How early bacterial or fungal infection develops is also important. Sondheimer et al. reviewed 42 patients with intestinal resection in the neonatal period, who subsequently became dependent on PN.143 Cholestasis developed in 67%, and 25% of these proceeded to liver failure. The overall number of septic episodes was similar in all patients, even those without cholestasis, but development of liver failure was associated with significantly younger age at first infection. Inability to establish enteral feeding is common in children requiring PN and is one of the important indications for intravenous feeding. Nevertheless, IFALD is more likely to develop in those children who are unable to tolerate any enteral feeding compared to those with partial enteral feeding, although not all studies concur on this.141 Experimental studies have shown that short-term fasting has several metabolic and endocrine consequences on intestinal and liver function. Levels of gastrointestinal hormones in patients on TPN are reduced, which may lead not only to intestinal stasis but also to reduced gallbladder contractility. Intestinal stasis may lead to bacterial overgrowth, bacterial translocation, and sepsis, which may increase cholestasis,141 and to the production of lithocholic acid, which has been shown to be toxic to the liver. The reduction in cholecystokinin release may influence gallbladder size and contractility and cause the development of biliary sludge. Fasting may reduce the size of the bile-salt pool and bile formation, compounding the difficulties with gallbladder contractility and the formation of sludge. Components of PN itself may be detrimental to the liver. It has long been suggested that IFALD may be associated with bacterial or chemical toxins in PN solutions. Both animal and clinical studies have suggested a direct effect of amino acid infusions on the hepatocyte canalicular membrane in the production of cholestasis, but this is unlikely to be a significant problem with modern solutions. The hepatotoxicity may alternatively relate to a deficiency of an essential amino acid (e.g., tyrosine or cysteine). In older infants and adults cysteine and taurine are synthesized from methionine but this production is diminished in premature infants. Not only is taurine one of the main bile acid conjugates in the neonate, but it has been shown to increase bile flow and protect against lithocholate toxicity. Despite this, the benefits of taurine supplementation are unproven. One recent suggestion is that choline deficiency may exacerbate hepatic steatosis in adults and children. In a pilot study the addition of 2 g of choline reduced steatosis in adults, as demonstrated by normalization of hepatic transaminases and computed tomography scans.144 There are also potentially toxic components in PN. Historically, degradation of tryptophan in feeding solutions contaminated by sodium bisulfate was thought to produce cholestatic metabolites, but this is not relevant to modern solutions. Although aluminum toxicity is well recognized in PN and may lead to bone disease, there is no evidence that it is implicated in IFALD. Chromium toxicity has been reported in animals. Both serum and urine chromium levels are higher in children on long-term PN compared to control values, although there is no correlation with liver disease.
Manganese toxicity, on the other hand, is important in the development of IFALD. A number of studies have reported the effects of manganese toxicity in children on long-term PN. Fell et al. studied 57 children receiving long-term PN, including the multitrace element solutions of Pedel or Addamel.145 Forty-five children (79%) had whole-blood manganese concentrations above the reference range. Children with impaired liver function had the highest manganese levels and there was a significant correlation between wholeblood manganese levels, AST (r = 0.63, P = <0.001), and total plasma bilirubin (r = 0.64, P = <0.001). Eleven children had both hypermagnasemia and cholestasis and four of these died. In the seven survivors whole-blood manganese declined when manganese supplements were reduced or withdrawn145 and there was a resolution of both brain and liver disease.146 As manganese is excreted in bile, the toxic effect may be secondary to cholestasis and thus monitoring of manganese levels is important in patients with IFALD cholestasis. There is now some evidence that lipid emulsions induce cholestasis as well as steatosis. It has been accepted that excess lipid calories may lead to hepatic steatosis, hyperlipidemia, and thrombocytopenia and that close monitoring of triglyceride levels is necessary to monitor lipemia, particularly in neonates with hepatic dysfunction. More recently, Colomb et al. correlated episodes of cholestasis in 23 infants with alteration in lipid concentration,147 while Cavicchi et al. demonstrated that cholestasis in adults was related to the use of >1 g/kg of lipid. It has been suggested that the mechanism may be due to a direct effect of lipid on hepatocytes, accumulation of phytosterols, or production of inflammatory cytokines. Acalculous cholecystitis, biliary sludge, gallbladder distention, and gallstones have all been reported in adults and children on long-term PN.140 The incidence of biliary sludge increases with duration of PN, from 6% at 3 weeks to 100% at 6–13 weeks,148 as does the incidence of gallstone formation, particularly in children who have had ileal resection or disease. The increase in gallbladder size noted in parenterally fed infants when compared to enterally fed infants is related to the reduction in cholecystokinin production and other gut hormones. Gallbladder stasis may be prevented by administration of cholecystokinin or by stimulating endogenous release of cholecystokinin through pulsed infusions of large volumes of amino acids or by small amounts of enteral nutrition.
Clinical Features The earliest clinical sign of cholestasis is a rise in conjugated bilirubin, particularly during episodes of intercurrent sepsis.141 Persistent elevation of serum bilirubin (>200 mmol/l) has an adverse prognosis.141,149 In 22 children evaluated for combined liver and smallbowel transplantation, a raised plasma bilirubin concentration (>200 mmol/l) predicted death from liver failure within 6 months in the 11 children who subsequently died of liver failure. Even in children without obvious cholestasis hepatic dysfunction with portal fibrosis and splenomegaly may be present. In 37 children referred for small-bowel and liver transplantation, splenomegaly was a constant feature in 75% of this group irrespective of cholestasis. Despite extensive hepatic fibrosis and splenomegaly, esophageal varices are infrequent in these children.149
1381
Section XII. Inherited and Pediatric Diseases of the Liver
Figure 71-7. Liver histology in intestinal failureassociated liver disease. (A) Cholestasis in an infant aged 79 days. There are prominent bile plugs with a mild inflammatory infiltrate. There is no fibrosis. (Hematoxylin & eosin ¥160.) (B) Advanced intestinal failure-associated liver disease in an infant aged 5 months. There is prominent fibrosis with parenchymal nodular formation. There is some inflammatory cell infiltrate and prominent bile plugs in the parenchyma. (Hematoxylin & eosin ¥64.)
A
B
Investigation Liver function tests. There is a rise in conjugated bilirubin, particularly during episodes of intercurrent sepsis, associated with an increase in alkaline phosphatase, amino transferases, and gglutamyltranspeptidase. Liver biopsy. The histopathological changes of IFALD include centrilobular cholestasis, portal inflammation, and necrosis with or without fatty infiltration (Figure 71-7). As liver disease advances portal fibrosis, pericellular fibrosis, and bile ductular proliferation
1382
are seen, together with pigmented Kupffer cells and eventually cirrhosis. Cholestasis is not always present. Biliary cirrhosis is a late development which may be associated with death within 6 months.149
Management IFALD is potentially reversible if the PN can be discontinued before the development of severe fibrosis or cirrhosis. In many children and adults this is not possible and prevention or treatment of IFALD
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
includes a number of approaches. The most important strategy is the prevention of sepsis, especially early in life.143 There is a marked difference between the incidence and the age of development of liver disease in children whose central-line catheter care is managed by units with nutritional care teams with experience in PN than those without.143 Strict catheter care and reduction of central line infections are important preventive mechanisms in the development of IFALD. The introduction of some enteral feeding will encourage normal biliary dynamics, decrease gallbladder size, improve bile flow, and reduce intestinal stasis and bacterial overgrowth, and may decrease episodes of bacterial translocation and sepsis. If enteral feeding is impossble, reducing the duration of daily PN or using cyclic infusions may be helpful. If cholestasis is severe, restriction of lipid intake and control of manganese and copper levels are important. In order to improve bile flow and reduce the formation of biliary sludge oral ursodeoxycholic acid may be advantageous. Spagnuolo et al. reported biochemical resolution in seven children on long-term PN treated with ursodeoxycholic acid,150 but another study in infants given ursodeoxycholic acid prophylactically found no benefit.151 Other strategies to prevent bacterial overgrowth include the addition of fiber to enteral feeds if tolerated, as this reduces bacterial translocation. Addition of glutamine to PN solutions may be beneficial as glutamine reverses the inhibition of mitochondrial metabolism observed in endotoxemia, while also improving intestinal adaptation. Saccharomyces boulardii, a non-pathogenic yeast, may also have a trophic effect on the gut and reduce bacterial translocation. It is possible that the presence of liver dysfunction and portal hypertension may prevent adequate intestinal adaptation. Weber and Keller demonstrated that survival and time to the development of feeding tolerance were related to severity of liver dysfunction,152 while a number of authors have drawn attention to the improvement in feeding intolerance after isolated liver transplantation in children with severe IFALD.153 If prevention of liver failure is not possible then the only remaining option may be small-bowel transplantation. Over 500 smallbowel operations have been carried out worldwide, with a 5-year survival of over 50%.154 Therefore this must be considered as a therapeutic option in children with intestinal failure. Current results suggest that isolated small-bowel transplantation has more favorable long-term results than combined small-bowel and liver transplantation, and this has important implications for children with intestinal failure, but the benefits of isolated liver transplantation for children with severe liver disease and potentially salvageable intestines should be considered.
GENERAL MANAGEMENT OF CHOLESTATIC LIVER DISEASE Whatever the underlying diagnosis, there are a number of general measures that should be taken in the management of infantile cholestases.
NUTRITIONAL SUPPORT Malnutrition is present in 50–80% of children with chronic liver disease. The pathophysiology is complex and multifactorial (Table 71-5). Infants with chronic cholestatic liver disease are particularly vulnerable to the effects of malnutrition because of their high energy and growth requirements. Anorexia is common in children with chronic liver disease who often take less than the recommended requirements, or less than is appropriate for their energy consumption because of increased energy expenditure. Energy requirements may be increased up to 140%. Mechanisms implicated include portosystemic shunting and ascites, abnormal intermediate metabolism, and the energy demands of specific complications such as sepsis and variceal hemorrhage. Fat malnutrition develops first with loss of fat stores. Protein malnutrition is a late development and is associated with a reduction in muscle bulk, stunting, and significant motor developmental delay. In time, children with significant malnutrition will have impaired growth and psychosocial development,155 and thus malnutrition is not only an important indication for liver transplantation, but also one of the most important prognostic factors for survival after liver transplantation.155
NUTRITIONAL ASSESSMENT Accurate nutritional assessment is essential for the management of children with cholestatic liver disease. It is important to start with a comprehensive clinical history, feeding history, and a careful physical examination. Serial anthropometric examination is critical and may identify early malnutrition. Standard weight and height ratios are of little value in children with liver disease because of misinterpretation due to fluid overload, ascites, and visceromegaly. Many researchers, using sophisticated methods such as whole-body
Table 71-5. Pathophysiology of malnutrition in liver disease Abnormality
Etiology
Decreased calorie intake
Anorexia Fat malabsorption Unpalatable feeds Use of bile-salt resins Portal hypertension Energy expenditure Calorie requirements Abnormal nitrogen metabolism Negative protein balance Reduction in glycogen stores GH/IGF-1 Insulin resistance
OTHER DISORDERS
Increased metabolic needs
There are several other disorders presenting with cholestasis in infancy that need to be remembered and differentiated from the above conditions (Table 71-1). Among these are metabolic conditions, including Niemann-Pick type C and cystic fibrosis. A full investigation of infants with cholestasis must therefore exclude such rarer conditions (Table 71-3).
Inappropriate substrate utilization
Hormonal dysregulation
GH, growth hormone, IGF-1, insulin growth factor-1.
1383
Section XII. Inherited and Pediatric Diseases of the Liver
potassium and dual-energy X-ray absorptiometry scanning, have demonstrated that body weight underestimated the incidence of malnutrition by 50% in both adults and children and, for this reason, simple methods of measuring body composition such as body impedence analysis are not reliable in liver disease in children. Thus, the assessment of malnutrition should be performed using a number of parameters such as triceps or subscapular skinfolds, mid-arm circumference and arm muscle measurements (mid-arm muscle area). Triceps skinfold and mid-arm circumference are useful indicators of body fat and protein and serial recording demonstrates early loss of fat stores before weight and height changes become obvious. Although linear growth is a sensitive parameter, it is a late sign of growth failure in infancy as stunting (or negative height velocity) may not be apparent until 1 year of age. Growth data are best expressed as standard deviation scores (or z-scores) related to the median value for the child’s age and sex in which the z-score of 0 = 50th percentile. This is particularly useful in comparisons between centers and for evaluation of new feeds. Biochemical evaluation of vitamin deficiency is a useful adjunct to nutritional assessment (Table 71-3). Plasma vitamin A levels can be measured but may not reflect hepatic stores. Plasma b-carotene or the ratio of plasma retinol to retinol-binding protein may be helpful. Vitamin E deficiency is monitored by serum vitamin E levels or the ratio of vitamin E to total lipid. Vitamin D deficiency is evaluated by measuring serum levels of calcium, phosphate, and alkaline phosphatase, while the diagnosis of rickets is confirmed by X-ray examination of wrist or knee. In some centers, it is possible to measure 25-hydroxy-vitamin D levels. Vitamin K deficiency is identified by measuring coagulation times and monitoring the response to parenteral vitamin K.
INDICATIONS FOR NUTRITIONAL THERAPY The aim of nutritional therapy is to prevent or treat malnutrition by providing adequate calories for energy, and sufficient nitrogen for protein synthesis, to restore plasma amino acid imbalance, to prevent vitamin and trace element deficiency, and to achieve normal growth and activity. The need for nutritional support is often underestimated, particularly in infants with liver disease, who may have an increased appetite in the first few months of life.
STRATEGIES FOR NUTRITIONAL SUPPORT (Table 71-4) Increased Energy Intake As the resting energy requirements are increased, it is important to increase the energy intake to 140–200% of estimated average requirements. This can be achieved by using concentrated formulas (the formula is concentrated 13–15%, which increases the kilocalories from 67 to 80 kcal/100 ml) or by supplementing milk feeds with extra carbohydrate and fat to produce a feed with an energy density of 4.18 kJ/ml (1 kcal/ml or more). As such a feed may have a high osmolality (500–800 mmol/l), it should be introduced gradually to establish intestinal tolerance. Calorie supplementation added to drinks may be effective for older children. If there is no response to an increase in energy intake alone, nocturnal enteral feeding by nasogastric tube may be required. An alternative is placing a gastrostomy tube if nasogastric tubes are poorly tolerated,
1384
but this should be avoided in children with severe portal hypertension because of the development of stomal varices.
Medium-Chain Triglycerides (MCT) Hydrolyzed protein infant formulas which contain 50% MCT will maximize fat absorption and improve steatorrhea. Formulas with more than 80% MCT may lead to essential fatty acid deficiency. In older children, MCT oil may be added to meals and should be balanced by fats with high polyunsaturated fatty acid (PUFA) content.
Long-Chain Polyunsaturated Fatty Acids It is important to ensure adequate intake of PUFA. The minimal intake of linoleic acid recommended for infants is 1–2% of energy in a ratio of linoleic to linolenic acid of 5 : 15.1. The diet can be supplemented by the addition of soy bean or rape seed oil or dietary products like egg yolk (which is rich in essential amino acids) or fish oil (which is rich in docosahexaenoic acid). Alternatively infants may be given conventional PUFA-supplemented formula feeds which are commercially available.
Structured Lipids Recently, chemically defined structured lipids have been developed to increase absorption of both medium- and long-chain fat and essential fatty acids. These lipids combine pure MCTs with longchain triglycerides, resulting in a triglyceride which contains combinations of short-, medium-, and long-chain fatty acids on a single glycerol backbone which should be absorbed like MCT. To date, clinical studies in adults have evaluated structured lipids in postoperative patients receiving PN, and demonstrated that they were safe and effective when compared to PUFA emulsion. Although clinical studies in childhood of these modified lipids are currently in the preliminary stages, animal studies in rats have demonstrated improved fat absorption in and reversal of essential fatty acid deficiency in vitro in caco-2 cells.156 If successful, there might be considerable benefit for cholestatic infants.
Carbohydate Carbohydrate is a major source of energy and particularly useful for increasing calorie intake. It can be given as a monomer, short-chain polymer, or starch but complex carbohydrates such as maltodextrin or glucose polymer restrict the osmolality of the feed while maintaining a high energy density >1 kcal/ml, allowing fluid restriction if necessary while providing up to 20 g/kg per day of carbohydrate. In infants, glucose polymers are best added to milk feeds while in older children they may be provided as supplemental drinks.
Protein Historical advice, which restricted protein in advanced end-stage liver disease, is now considered inappropriate in both children and adults. Children with end-stage liver disease require minimal protein intake of around 2–3 g/kg per day but will tolerate up to 4 g/kg per day without developing encephalopathy or a significant increase in plasma amino abnormalities. Severe protein restriction below 2 g/kg per day may be required for acute severe encephalopathy but should be avoided in the long term as it may lead to endogenous muscle protein consumption. There is no necessity to use semielemental
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
diets or protein hydrolysates as there is no evidence of protein malabsorption. In view of the abnormal ratio of essential amino acids to branchedchain amino acids (BCAA), there has been considerable interest in the use of modified amino acid formulations designed to improve this imbalance. BCAA-enriched formulas may have significant nutritional benefit in children. A study which compared a feed containing 32% BCAA to a standard feed demonstrated improved lean body mass in children awaiting liver transplantation but no improvement in amino acid levels.157 In another study, the effect of a modified amino acid feed containing 50% BCAA was compared to an isonitrogenous formula containing 22% BCAA in infants with liver disease by measuring whole-body protein turnover. The BCAAsupplemented feed improved protein retention when compared with the standard formula by suppressing endogenous protein catabolism and normalizing the plasma amino acid profile. Formulas rich in BCAA complete with MCTs and vitamin and mineral supplements have been developed for use in infants (Generaid or Generaid Plus), while oral supplements rich in BCAA are available for older children, but both are particularly unpalatable. Fat-soluble vitamins are required in all children with prolonged or cholestatic liver disease. Most children will be maintained adequately on oral fat-soluble vitamin administration but monthly intramuscular administration is occasionally required for those children with severe cholestasis.
Growth Hormone Therapy In view of the disruption of the growth hormone–insulin growth factor-1 axis, it is tempting to prescribe growth hormone therapy
for children with significant growth failure, but this has not proven beneficial to date.
PARENTERAL NUTRITION PN should only be considered in children with chronic liver disease if they cannot be enterally fed because of feed intolerance or secondary to complications such as recurrent variceal bleeding or abdominal sepsis. Standard amino acid and lipid solutions are well tolerated in stable patients and lipids may be particularly beneficial in order to achieve adequate calorie intake. If encephalopathy develops the amino acid content of the feed could be reduced to 1–2 g/kg per day, while lipid administration requires careful monitoring in children with severe liver dysfunction, hepatic encephalopathy, and sepsis. In general the use of PN in children with established liver disease is for short-term purposes only and thus they are unlikely to develop biliary sludge and gallstones from the prescription of PN alone, although prescription of ursodeoxycholic acid 15–20 mg/kg may be of value.
MANAGEMENT OF PRURITUS Over recent years many advances have been made in the treatment of intractable pruritus. In many units it is standard practice to commence all cholestatic children on ursodeoxycholic acid, adding in other agents as required either singly or in combination (Figure 718). Cholestyramine is very effective but unfortunately is unpalatable to many children, leading to more common use of other agents. The use of a pruritus score (1–10) may help in objective assessment and management.
Figure 71-8. Therapy for pruritus. First line therapy Ursodeoxycholic acid: 5–7 mg/kg tds (maximum 45 mg/kg/day) Rifampicin: 3–10 mg/kg daily Cholestyramine: under 6 years 2 g/day; over 6 years 4 g/day in divided doses Phenobarbitone: 3–5 mg/kg daily
Central action Ondansetron: aged < 12 years 2–4 mg bd; > 12 years 4–8 mg bd Naltrioxone: 6–20 mg/day Trimeprazine: aged 6–12 months 250 g/kg qds; aged 1–2 years 2.5 mg qds; aged 2–12 years 5 mg qds; aged 12–18 years 10 mg tds MARS therapy (see text)
The above can be used singly or in combination. However, it is recommended that single agents be tried in the order given above, in order to assess effectiveness. Cholestyramine may be poorly tolerated; pheobarbitone should be administered at night to avoid daytime drowsiness.
1385
Section XII. Inherited and Pediatric Diseases of the Liver
BILIARY DIVERSION Partial external biliary diversion may be a very effective tool in the treatment of cholestasis. This procedure involves diversion of bile from the gallbladder through a loop of jejunum connecting the dome of the gallbladder to the skin of the abdomen. It therefore interrupts the enterohepatic circulation of bile salts. This can be a useful long-term treatment to relieve the symptoms of cholestasis, and allow normalization of liver function tests.130,158 A new alternative treatment for intractable pruritus is the molecular absorbent recirculating system (MARS) therapy. This is a form of extracorporeal albumin dialysis. It was originally developed for patients in acute liver failure. The resulting decrease in bilirubin levels led it to be tried in patients with severe pruritus. It appears to reduce pruritus successfully for 6–12 months, but more research in this area is needed to determine frequency and duration of therapy, and possible longer-term benefits and side effects.
Cholangitis Where surgical management is necessary, especially in EHBA, then cholangitis is a risk. Prophylaxis regimes vary in duration of rotating antibiotics and overall duration of prophylaxis. Current regimes involve rotating prophylactic cephalosporins, trimethoprim, and amoxicillin. Aggressive management is vital, with intravenous antibiotics for at least 2 weeks where cholangitis is proven and where there is a high index of suspicion. If symptoms do not settle completely then changing to second-line antibiotics is essential.
Psychological Support This is important, especially in families where the diagnosis is of a lifelong, life-threatening, and/or inherited disorder. A multidisciplinary team providing accurate and timely information and support is invaluable as part of the ongoing management of the child. This includes dietetic, social, and psychological input, as well as that of a dedicated nurse specialist to explain the diagnosis, ongoing treatment, and outcome likelihoods.
Liver Transplantation As discussed above, for many of these conditions liver transplantation may be the only option for some children, where the disease is life-threatening or severely impairs the quality of life. Survival is now greatly improved, with up to 95% 1-year survival and 80–90% 5-year survival in some cases. It is important not to leave transplantation too late, as outcome is better when the procedure is elective, rather than for acute liver failure.
SUMMARY There are many causes of neonatal cholestasis, and early diagnosis is beneficial for outcome, particularly in EHBA. With modern molecular and genetic developments, the diagnosis of many conditions is now possible. There have been significant advances in treatment, especially in liver transplantation and intensive care, with improved nutrition and antipruritic agents which have dramatically changed the outcome for children with persistent cholestasis. Improvements are likely to continue with the development of new therapies like MARS, further understanding of the genetic and
1386
developmental components of many disorders, and increasing experience of many centers in managing rare conditions. Many children with neonatal cholestasis now have the prospect of a childhood spent achieving normal educational and recreational goals, and survival with minimal morbidity into adulthood.
REFERENCES 1. Kelly DA, Stanton A. Jaundice in babies: implications for community screening for biliary atresia. Br Med J 1995; 310:1172–1173. 2. Chardot C. Treatment for biliary atresia in 2003. J Pediatr Gastroenterol Nutr 2003; 37:407–408. 3. Chardot C, Carton M, Spire-Bendelac N, et al. Epidemiology of biliary atresia in France: a national study 1986–96. J Hepatol 1999; 31:1006–1013. 4. McKiernan PJ, Baker AJ, Kelly DA. The frequency and outcome of biliary atresia in the UK and Ireland. Lancet 2000; 355:25–29. 5. Zuckerman M, Hajinicolaou C. Incidence and outcome of biliary atresia in black infants in Soweto (South Africa): review of cases from 1993–1996. J Pediatr Gastroenterol Nutr 1998; 26:587 (abstract). 6. Balistreri WF, Grand R, Hoofnagle JH, et al. Biliary atresia: current concepts and research directions. Summary of a symposium. Hepatology 1996; 23:1682–1692. 7. Hyams JS, Glaser JH, Leichtner AM, Morecki R. Discordance for biliary atresia in two sets of monozygotic twins. J Pediatr 1985; 107:420–422. 8. Broome U, Nemeth A, Hultcrantz R, Scheynius A. Different expression of HLA-DR and ICAM-1 in livers from patients with biliary atresia and Byler’s disease. J Hepatol 1997; 26:857–862. 9. Kobayashi H, Puri P, O’Briain DS, et al. Hepatic overexpression of MHC class II antigens and macrophage-associated antigens (CD68) in patients with biliary atresia of poor prognosis. J Pediatr Surg 1997; 32:590–593. 10. Kader HH, El-Ayyouti M, Hawas S, et al. HLA in Egyptian children with biliary atresia. J Pediatr 2002; 141:432–433. 11. Bangaru B, Morecki R, Glaser JH, et al. Comparative studies of biliary atresia in the human newborn and reovirus-induced cholangitis in weanling mice. Lab Invest 1980; 43:456–462. 12. Tyler KL, Sokol RJ, Oberhaus SM, et al. Detection of reovirus RNA in hepatobiliary tissues from patients with extrahepatic biliary atresia and choledochal cysts. Hepatology 1998; 27:1475–1482. 13. Saito T, Shinozaki K, Matsunaga T, et al. Lack of evidence for reovirus infection in tissues from patients with biliary atresia and congenital dilatation of the bile duct. J Hepatol 2004; 40:203–211. 14. Riepenhoff-Talty M, Gouvea V, Evans MJ, et al. Detection of group C rotavirus in infants with extrahepatic biliary atresia. J Infect Dis 1996; 174:8–15. 15. Petersen C, Kuske M, Bruns E, et al. Progress in developing animal models for biliary atresia. Eur J Pediatr Surg 1998; 8:137–141. 16. Domiati-Saad R, Dawson DB, Margraf LR, et al. Cytomegalovirus and human herpesvirus 6, but not human papillomavirus, are present in neonatal giant cell hepatitis and extrahepatic biliary atresia. Pediatr Dev Pathol 2000; 3:367–373. 17. Drut R, Drut RM, Gomez MA, et al. Presence of human papillomavirus in extrahepatic biliary atresia. J Pediatr Gastroenterol Nutr 1998; 27:530–535. 18. Carmi R, Magee CA, Neill CA, Karrer FM. Extrahepatic biliary atresia and associated anomalies: etiologic heterogeneity suggested by distinctive patterns of associations. Am J Med Genet 1993; 45:683–693.
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
19. Davenport M, Savage M, Mowat AP, Howard ER. Biliary atresia splenic malformation syndrome: an etiologic and prognostic subgroup. Surgery 1993; 113:662–668. 20. Mazziotti MV, Willis LK, Heuckeroth RO, et al. Anomalous development of the hepatobiliary system in the Inv mouse. Hepatology 1999; 30:372–378. 21. Schon P, Tsuchiya K, Lenoir D, et al. Identification, genomic organization, chromosomal mapping and mutation analysis of the human INV gene, the ortholog of a murine gene implicated in left-right axis development and biliary atresia. Hum Genet 2002; 110:157–165. 22. Bamford RN, Roessler E, Burdine RD, et al. Loss-of-function mutations in the EGF-CFC gene CFC1 are associated with human left–right laterality defects. Nat Genet 2000; 26:365–369. 23. Jacquemin E, Cresteil D, Raynaud N, Hadchouel M. CFCI gene mutation and biliary atresia with polysplenia syndrome. J Pediatr Gastroenterol Nutr 2002; 34:326–327. 24. Low Y, Vijayan V, Tan CE. The prognostic value of ductal plate malformation and other histologic parameters in biliary atresia: an immunohistochemical study. J Pediatr 2001; 139:320–322. 25. Vijayan V, Tan CE. Development of the human intrahepatic biliary system. Ann Acad Med Singapore 1999; 28:105–108. 26. Libbrecht L, Cassiman D, Desmet V, Roskams T. Expression of neural cell adhesion molecule in human liver development and in congenital and acquired liver diseases. Histochem Cell Biol 2001; 116:233–239. 27. Kohsaka T, Yuan ZR, Guo SX, et al. The significance of human Jagged 1 mutations detected in severe cases of extrahepatic biliary atresia. Hepatology 2002; 36:904–912. 28. Flynn DM, Nijjar S, Hubscher SG, et al. Notch receptor expression in human fetal and pediatric liver: role in bile duct development and disease. J Pathol 2004; 204:55–64. 29. Sumazaki R, Shiojiri N, Isoyama S, et al. Conversion of biliary system to pancreatic tissue in Hes1-deficient mice. Nat Genet 2004; 36:83–87. 30. Clotman F, Lannoy VJ, Reber M, et al. The onecut transcription factor HNF6 is required for normal development of the biliary tract. Development 2002; 129:1819–1828. 31. Cocjin J, Rosenthal P, Buslon V, et al. Bile ductule formation in fetal, neonatal, and infant livers compared with extrahepatic biliary atresia. Hepatology 1996; 24:568–574. 32. Fabris L, Strazzabosco M, Crosby HA, et al. Characterization and isolation of ductular cells coexpressing neural cell adhesion molecule and Bcl-2 from primary cholangiopathies and ductal plate malformations. Am J Pathol 2000; 156:1599–1612. 33. Davenport M, Gonde C, Redkar R, et al. Immunohistochemistry of the liver and biliary tree in extrahepatic biliary atresia. J Pediatr Surg 2001; 36:1017–1025. 34. Dillon PW, Belchis D, Minnick K, Tracy T. Differential expression of the major histocompatibility antigens and ICAM-1 on bile duct epithelial cells in biliary atresia. Tohoku J Exp Med 1997; 181:33–40. 35. Bezerra JA, Tiao G, Ryckman FC, et al. Genetic induction of proinflammatory immunity in children with biliary atresia. Lancet 2002; 360:1653–1659. 36. Landing BH. Editorial: Changing approach to neonatal hepatitis and biliary atresia. Pediatrics 1974; 53:647–648. 37. Kasai M. Treatment of biliary atresia with special reference to hepatic porto-enterostomy and its modifications. Prog Pediatr Surg 1974; 6:5–52. 38. Gautier M, Eliot N. Extrahepatic biliary atresia. Morphological study of 98 biliary remnants. Arch Pathol Lab Med 1981; 105:397–402. 39. Caton AR, Druschel CM, McNutt LA. The epidemiology of extrahepatic biliary atresia in New York State, 1983–98. Pediatr Perinat Epidemiol 2004; 18:97–105.
40. Choi SO, Park WH, Lee HJ, Woo SK. “Triangular cord”: a sonographic finding applicable in the diagnosis of biliary atresia. J Pediatr Surg 1996; 31:363–366. 41. Farrant P, Meire HB, Mieli-Vergani G. Improved diagnosis of extraheptic biliary atresia by high frequency ultrasound of the gallbladder. Br J Radiol 2001; 74:952–954. 42. Kasai M, Kimura S, Asakura S, et al. Surgical treatment of biliary atresia. J Pediatr Surg 1968; 3:665. 43. Chardot C, Carton M, Spire-Bendelac N, et al. Prognosis of biliary atresia in the era of liver transplantation: French national study from 1986 to 1996. Hepatology 1999; 30:606–611. 44. Volpert D, White F, Finegold MJ, et al. Outcome of early hepatic portoenterostomy for biliary atresia. J Pediatr Gastroenterol Nutr 2001; 32:265–269. 45. Davenport M, Puricelli V, Farrant P, et al. The outcome of the older (> or =100 days) infant with biliary atresia. J Pediatr Surg 2004; 39:575–581. 46. Davenport M, de Ville D, Stringer MD, et al. Seamless management of biliary atresia in England and Wales (1999–2002). Lancet 2004; 363:1354–1357. 47. Balistreri WF. Bile acid therapy in pediatric hepatobiliary disease: the role of ursodeoxycholic acid. J Pediatr Gastroenterol Nutr 1997; 24:573–589. 48. Bu LN, Chen HL, Chang CJ, et al. Prophylactic oral antibiotics in prevention of recurrent cholangitis after the Kasai portoenterostomy. J Pediatr Surg 2003; 38:590–593. 49. Takahashi A, Tsuchida Y, Suzuki N, et al. Incidence of intrahepatic biliary cysts in biliary atresia after hepatic portoenterostomy and associated histopathologic findings in the liver and porta hepatis at diagnosis. J Pediatr Surg 1999; 34:1364–1368. 50. Miga D, Sokol RJ, MacKenzie T, et al. Survival after first esophageal variceal hemorrhage in patients with biliary atresia. J Pediatr 2001; 139:291–296. 51. van Heurn LW, Saing H, Tam PK. Portoenterostomy for biliary atresia: long-term survival and prognosis after esophageal variceal bleeding. J Pediatr Surg 2004; 39:6–9. 52. Howard ER, Davenport M. The treatment of biliary atresia in Europe 1969–1995. Tohoku J Exp Med 1997; 181:75–83. 53. Hadzic N, Davenport M, Tizzard S, et al. Long-term survival following Kasai portoenterostomy: is chronic liver disease inevitable? J Pediatr Gastroenterol Nutr 2003; 37:430– 433. 54. Kelly DA. Current results and evolving indications for liver transplantation in children. J Pediatr Gastroenterol Nutr 1998; 27:214–221. 55. Rodeck B, Melter M, Kardorff R, et al. Liver transplantation in children with chronic end stage liver disease: factors influencing survival after transplantation. Transplantation 1996; 62:1071–1076. 56. McDiarmid SV, Anand R, Lindblad AS. Development of a pediatric end-stage liver disease score to predict poor outcome in children awaiting liver transplantation. Transplantation 2002; 74:173–181. 57. Todani T, Watanabe Y, Toki A, Morotomi Y. Classification of congenital biliary cystic disease: special reference to type Ic and IVA cysts with primary ductal stricture. J Hepatobiliary Pancreat Surg 2003; 10:340–344. 58. Stringer MD, Dhawan A, Davenport M, et al. Choledochal cysts: lessons from a 20 year experience. Arch Dis Child 1995; 73:528–531. 59. Bancroft JD, Bucuvalas JC, Ryckman FC, et al. Antenatal diagnosis of choledochal cyst. J Pediatr Gastroenterol Nutr 1994; 18:142–145. 60. Yamataka A, Ohshiro K, Okada Y, et al. Complications after cyst excision with hepaticoenterostomy for choledochal cysts and their surgical management in children versus adults. J Pediatr Surg 1997; 32:1097–1102.
1387
Section XII. Inherited and Pediatric Diseases of the Liver
61. Yamataka A, Kobayashi H, Shimotakahara A, et al. Recommendations for preventing complications related to Roux-en-Y hepatico-jejunostomy performed during excision of choledochal cyst in children. J Pediatr Surg 2003; 38:1830–1832. 62. Benjamin IS. Biliary cystic disease: the risk of cancer. J Hepatobiliary Pancreat Surg 2004; 10:335–339. 63. Taylor AC, Palmer KR. Caroli’s disease. Eur J Gastroenterol Hepatol 1998; 10:105–108. 64. Amedee-Manesme O, Bernard O, Brunelle F, et al. Sclerosing cholangitis with neonatal onset. J Pediatr 1987; 111:225–229. 65. Baker AJ, Portmann B, Westaby D, et al. Neonatal sclerosing cholangitis in two siblings: a category of progressive intrahepatic cholestasis. J Pediatr Gastroenterol Nutr 1993; 17:317–322. 66. Bar MM, Hadas-Halperin I, Fisher D, et al. Neonatal sclerosing cholangitis associated with autoimmune phenomena. J Pediatr Gastroenterol Nutr 2000; 30:332–334. 67. Smethurst FA, Carty H. Case report: spontaneous perforation of the common bile duct in infancy. Br J Radiol 1993; 66: 556–557. 68. Chilukuri S, Bonet V, Cobb M. Antenatal spontaneous perforation of the extrahepatic biliary tree. Am J Obstet Gynecol 1990; 163:1201–1202. 69. Bernstein J, Braylan R, Brough AJ. Bile-plug syndrome: a correctable cause of obstructive jaundice in infants. Pediatrics 1969; 43:273–276. 70. Alagille D, Odièvre M. Liver and biliary tract disease in children. New York: John Wiley; 1979. 71. Waguet J, Benamara MA, Chardot C, Pariente D. Common bile duct stenosis with hypereosinophilia and pancreatitis in childhood: report of 2 cases. Pediatr Radiol 2000; 30:383–385. 72. Schwartz MZ, Hall RJ, Reubner B, et al. Agenesis of the extrahepatic bile ducts: report of five cases. J Pediatr Surg 1990; 25:805–807. 73. Debray D, Pariente D, Gauthier F, et al. Cholelithiasis in infancy: a study of 40 cases. J Pediatr 1993; 122:385–391. 74. von der O, Spaar HJ, Lieber T, et al. Embryonal rhabdomyosarcoma of the common bile duct. Eur J Pediatr Surg 1991; 1:161–165. 75. Soares, Landell GA, Peres LC, et al. Liposarcoma of hepatic hilum in childhood: report of a case and review of the literature. Med Pediatr Oncol 1989; 17:239–243. 76. Dick MC, Mowat AP. Hepatitis syndrome in infancy—an epidemiological survey with 10 year follow up. Arch Dis Child 1985; 60:512–516. 77. Martin S. Congenital toxoplasmosis. Neonatal Netw 2001; 20:23–30. 78. Leung AK, Sauve RS, Davies HD. Congenital cytomegalovirus infection. J Natl Med Assoc 2003; 95:213–218. 79. Chang MH, Huang HH, Huang ES, et al. Polymerase chain reaction to detect human cytomegalovirus in livers of infants with neonatal hepatitis. Gastroenterology 1992; 103:1022–1025. 80. Tarr PI, Haas JE, Christie DL. Biliary atresia, cytomegalovirus, and age at referral. Pediatrics 1996; 97:828–831. 81. Arnold JC, Portmann BC, O’Grady JG, et al. Cytomegalovirus infection persists in the liver graft in the vanishing bile duct syndrome. Hepatology 1992; 16:285–292. 82. Chalhub EG, Baenziger J, Feigen RD, et al. Congenital herpes simplex type II infection with extensive hepatic calcification, bone lesions and cataracts: complete postmortem examination. Dev Med Child Neurol 1977; 19:527–534. 83. Harger JH, Ernest JM, Thurnau GR, et al. Frequency of congenital varicella syndrome in a prospective cohort of 347 pregnant women. Obstet Gynecol 2002; 100:260–265. 84. Ellaway CJ, Silinik M, Cowell CT, et al. Cholestatic jaundice and congenital hypopituitarism. J Pediatr Child Health 1995; 31:51–53.
1388
85. Spray CH, Mckiernan P, Waldron KE, et al. Investigation and outcome of neonatal hepatitis in infants with hypopituitarism. Acta Pediatr 2000; 89:951–954. 86. Minami K, Izumi G, Yanagawa T, et al. Septo-optic dysplasia with congenital hepatic fibrosis. Pediatr Neurol 2003; 29:157–159. 87. McDermid HE, Duncan AM, Brasch KR, et al. Characterization of the supernumerary chromosome in cat eye syndrome. Science 1986; 232:646–648. 88. Mowat AP, Psacharopoulos HT, Williams R. Extrahepatic biliary atresia versus neonatal hepatitis. Review of 137 prospectively investigated infants. Arch Dis Child 1976; 51:763–770. 89. Maggiore G, Bernard O, Hadchouel M, et al. Diagnostic value of serum gamma-glutamyl transpeptidase activity in liver diseases in children. J Pediatr Gastroenterol Nutr 1991; 12:21–26. 90. Alagille D. Cholestasis in the first three months of life. Prog Liver Dis 1979; 6:471–485. 91. Danks DM, Campbell PE, Smith AL, Rogers J. Prognosis of babies with neonatal hepatitis. Arch Dis Child 1977; 52:368–372. 92. Alagille D, Odievre M, Gautier M, Dommergues JP. Hepatic ductular hypoplasia associated with characteristic facies, vertebral malformations, retarded physical, mental, and sexual development, and cardiac murmur. J Pediatr 1975; 86:63–71. 93. Alagille D, Estrada A, Hadchouel M, et al. Syndromic paucity of interlobular bile ducts (Alagille syndrome or arteriohepatic dysplasia): review of 80 cases. J Pediatr 1987; 110:195–200. 94. Li L, Krantz ID, Deng Y, et al. Alagille syndrome is caused by mutations in human Jagged1, which encodes a ligand for Notch1. Nat Genet 1997; 16:243–251. 95. Oda T, Elkahloun AG, Pike BL, et al. Mutations in the human Jagged1 gene are responsible for Alagille syndrome. Nat Genet 1997; 16:235–242. 96. Colliton RP, Bason L, Lu FM, et al. Mutation analysis of Jagged1 (JAG1) in Alagille syndrome patients. Hum Mutat 2001; 17:151–152. 97. Crosnier C, Driancourt C, Raynaud N, et al. Fifteen novel mutations in the JAGGED1 gene of patients with Alagille syndrome. Hum Mutat 2001; 17:72–73. 98. Crosnier C, Driancourt C, Raynaud N, et al. Mutations in JAGGED1 gene are predominantly sporadic in Alagille syndrome. Gastroenterology 1999; 116:1141–1148. 99. Loomes KM, Underkoffler LA, Morabito J, et al. The expression of Jagged1 in the developing mammalian heart correlates with cardiovascular disease in Alagille syndrome. Hum Mol Genet 1999; 8:2443–2449. 100. Xue Y, Gao X, Lindsell CE, et al. Embryonic lethality and vascular defects in mice lacking the Notch ligand Jagged1. Hum Mol Genet 1999; 8:723–730. 101. McCright B, Lozier J, Gridley T. A mouse model of Alagille syndrome: Notch2 as a genetic modifier of Jag1 haploinsufficiency. Development 2002; 129:1075–1082. 102. Crosnier C, Attie-Bitach T, Encha-Razavi F, et al. JAGGED1 gene expression during human embryogenesis elucidates the wide phenotypic spectrum of Alagille syndrome. Hepatology 2000; 32:574–581. 103. Jones EA, Clement-Jones M, Wilson DI. JAGGED1 expression in human embryos: correlation with the Alagille syndrome phenotype. J Med Genet 2000; 37:663–668. 104. Loomes KM, Taichman DB, Glover CL, et al. Characterization of Notch receptor expression in the developing mammalian heart and liver. Am J Med Genet 2002; 112:181–189. 105. Louis AA, Van Eyken P, Haber BA, et al. Hepatic jagged1 expression studies. Hepatology 1999; 30:1269–1275. 106. Yuan ZR, Okaniwa M, Nagata I, et al. The DSL domain in mutant JAG1 ligand is essential for the severity of the liver defect in Alagille syndrome. Clin Genet 2001; 59:330–337.
Chapter 71 PEDIATRIC CHOLESTATIC SYNDROMES
107. Eldadah ZA, Hamosh A, Biery NJ, et al. Familial tetralogy of Fallot caused by mutation in the jagged1 gene. Hum Mol Genet 2001; 10:163–169. 108. Emerick KM, Rand EB, Goldmuntz E, et al. Features of Alagille syndrome in 92 patients: frequency and relation to prognosis. Hepatology 1999; 29:822–829. 109. Wasserman D, Zemel BS, Mulberg AE, et al. Growth, nutritional status, body composition, and energy expenditure in prepubertal children with Alagille syndrome. J Pediatr 1999; 134:172–177. 110. Lykavieris P, Hadchouel M, Chardot C, Bernard O. Outcome of liver disease in children with Alagille syndrome: a study of 163 patients. Gut 2001; 49:431–435. 111. Sanderson E, Newman V, Haigh SF, et al. Vertebral anomalies in children with Alagille syndrome: an analysis of 50 consecutive patients. Pediatr Radiol 2002; 32:114–119. 112. Hingorani M, Nischal KK, Davies A, et al. Ocular abnormalities in Alagille syndrome. Ophthalmology 1999; 106:330–337. 113. Kamath BM, Loomes KM, Oakey RJ, et al. Facial features in Alagille syndrome: specific or cholestasis facies? Am J Med Genet 2002; 112:163–170. 114. Martin SR, Garel L, Alvarez F. Alagille’s syndrome associated with cystic renal disease. Arch Dis Child 1996; 74:232–23. 115. Berard E, Sarles J, Triolo V, et al. Renovascular hypertension and vascular anomalies in Alagille syndrome. Pediatr Nephrol 1998; 12:121–124. 116. Kamath BM, Spinner NB, Emerick KM, et al. Vascular anomalies in Alagille syndrome: a significant cause of morbidity and mortality. Circulation 2004; 109:1354–1358. 117. Deutsch GH, Sokol RJ, Stathos TH, Knisely AS. Proliferation to paucity: evolution of bile duct abnormalities in a case of Alagille syndrome. Pediatr Dev Pathol 2001; 4:559–563. 118. Kamath BM, Bason L, Piccoli DA, et al. Consequences of JAG1 mutations. J Med Genet 2003; 40:891–895. 119. Rovner AJ, Schall JI, Jawad AF, et al. Rethinking growth failure in Alagille syndrome: the role of dietary intake and steatorrhea. J Pediatr Gastroenterol Nutr 2002; 35:495–502. 120. Clayton RJ, Iber FL, Ruebner BH, McKusick VA. Byler disease. Fatal familial intrahepatic cholestasis in an Amish kindred. Am J Dis Child 1969; 117:112–124. 121. Bull LN, van Eijk MJ, Pawlikowska L, et al. A gene encoding a P-type ATPase mutated in two forms of hereditary cholestasis. Nat Genet 1998; 18:219–224. 122. Pawlikowska L, Groen A, Eppens EF, et al. A mouse genetic model for familial cholestasis caused by ATP8B1 mutations reveals perturbed bile salt homeostasis but no impairment in bile secretion. Hum Mol Genet 2004; 13:881–892. 123. Chen F, Ananthanarayanan M, Emre S, et al. Progressive familial intrahepatic cholestasis, type 1, is associated with decreased farnesoid X receptor activity. Gastroenterology 2004; 126:756–764. 124. Jacquemin E, Hadchouel M. Genetic basis of progressive familial intrahepatic cholestasis. J Hepatol 1999; 31:377–381. 125. Knisely AS. Progressive familial intrahepatic cholestasis: a personal perspective. Pediatr Dev Pathol 2000; 3:113–125. 126. Lykavieris P, van Mil S, Cresteil D, et al. Progressive familial intrahepatic cholestasis type 1 and extrahepatic features: no catch-up of stature growth, exacerbation of diarrhea, and appearance of liver steatosis after liver transplantation. J Hepatol 2003; 39:447–452. 127. Luketic VA, Shiffman ML. Benign recurrent intrahepatic cholestasis. Clin Liver Dis 2004; 8:133–149, vii. 128. van Ooteghem NA, Klomp LW, van Berge-Henegouwen GP, Houwen RH. Benign recurrent intrahepatic cholestasis progressing to progressive familial intrahepatic cholestasis: low GGT cholestasis is a clinical continuum. J Hepatol 2002; 36:439–443.
129. Nicolas I, Pons JA, Vargas A, et al. [Ursodeoxycholic acid treatment shortens the course of cholestasis in two patients with benign recurrent intrahepatic cholestasis.] Gastroenterol Hepatol 2003; 26:421–423. 130. Kurbegov AC, Setchell KD, Haas JE, et al. Biliary diversion for progressive familial intrahepatic cholestasis: improved liver morphology and bile acid profile. Gastroenterology 2003; 125:1227–1234. 131. Strautnieks SS, Kagalwalla AF, Tanner MS, et al. Identification of a locus for progressive familial intrahepatic cholestasis PFIC2 on chromosome 2q24. Am J Hum Genet 1997; 61:630–633. 132. Strautnieks SS, Bull LN, Knisely AS, et al. A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis. Nat Genet 1998; 20:233–238. 133. Thompson R, Strautnieks S. BSEP: function and role in progressive familial intrahepatic cholestasis. Semin Liver Dis 2001; 21:545–550. 134. Wang L, Soroka CJ, Boyer JL. The role of bile salt export pump mutations in progressive familial intrahepatic cholestasis type II. J Clin Invest 2002; 110:965–972. 135. de Vree JM, Jacquemin E, Sturm E, et al. Mutations in the MDR3 gene cause progressive familial intrahepatic cholestasis. Proc Natl Acad Sci USA 1998; 95:282–287. 136. Jacquemin E. Role of multidrug resistance 3 deficiency in pediatric and adult liver disease: one gene for three diseases. Semin Liver Dis 2001; 21:551–562. 137. Primhak RA, Tanner MS. Alpha-1 antitrypsin deficiency. Arch Dis Child 2001; 85:2–5. 138. Hansen KN, Minousis M, Ebbesen F. Weekly oral vitamin K prophylaxis in Denmark. Acta Pediatr 2003; 92:802–805. 139. Francavilla R, Castellaneta SP, Hadzic N, et al. Prognosis of alpha-1-antitrypsin deficiency-related liver disease in the era of pediatric liver transplantation. J Hepatol 2000; 32:986– 992. 140. Van Gossum A, Vahedi K, Abdel M, et al. Clinical, social and rehabilitation status of long-term home parenteral nutrition patients: results of a European multicentre survey. Clin Nutr 2001; 20:205–210. 141. Beath SV, Davies P, Papadopoulou A, et al. Parenteral nutritionrelated cholestasis in postsurgical neonates: multivariate analysis of risk factors. J Pediatr Surg 1996; 31:604–606. 142. Beale EF, Nelson RM, Bucciarelli RL, et al. Intrahepatic cholestasis associated with parenteral nutrition in premature infants. Pediatrics 1979; 64:342–347. 143. Sondheimer JM, Asturias E, Cadnapaphornchai M. Infection and cholestasis in neonates with intestinal resection and long-term parenteral nutrition. J Pediatr Gastroenterol Nutr 1998; 27:131–137. 144. Buchman AL, Ament ME, Sohel M, et al. Choline deficiency causes reversible hepatic abnormalities in patients receiving parenteral nutrition: proof of a human choline requirement: a placebo-controlled trial. JPEN J Parenter Enteral Nutr 2001; 25:260–268. 145. Fell JM, Reynolds AP, Meadows N, et al. Manganese toxicity in children receiving long-term parenteral nutrition. Lancet 1996; 347:1218–1221. 146. Kafritsa Y, Fell J, Long S, et al. Long-term outcome of brain manganese deposition in patients on home parenteral nutrition. Arch Dis Child 1998; 79:263–265. 147. Colomb V, Jobert-Giraud A, Lacaille F, et al. Role of lipid emulsions in cholestasis associated with long-term parenteral nutrition in children. JPEN J Parenter Enteral Nutr 2000; 24:345–350. 148. Messing B, Bories C, Kunstlinger F, Bernier JJ. Does total parenteral nutrition induce gallbladder sludge formation and lithiasis? Gastroenterology 1983; 84:1012–1019. 149. Beath SV, Needham SJ, Kelly DA, et al. Clinical features and prognosis of children assessed for isolated small bowel or
1389
Section XII. Inherited and Pediatric Diseases of the Liver
150.
151.
152.
153.
1390
combined small bowel and liver transplantation. J Pediatr Surg 1997; 32:459–461. Spagnuolo MI, Iorio R, Vegnente A, Guarino A. Ursodeoxycholic acid for treatment of cholestasis in children on long-term total parenteral nutrition: a pilot study. Gastroenterology 1996; 111:716–719. Heubi JE, Wiechmann DA, Creutzinger V, et al. Tauroursodeoxycholic acid (TUDCA) in the prevention of total parenteral nutrition-associated liver disease. J Pediatr 2002; 141:237–242. Weber TR, Keller MS. Adverse effects of liver dysfunction and portal hypertension on intestinal adaptation in short bowel syndrome in children. Am J Surg 2002; 184:582– 586. Muiesan P, Dhawan A, Novelli M, et al. Isolated liver transplant and sequential small bowel transplantation for intestinal failure and related liver disease in children. Transplantation 2000; 69:2323–2326.
154. Sudan DL, Kaufman SS, Shaw BWJ, et al. Isolated intestinal transplantation for intestinal failure. Am J Gastroenterol 2000; 95:1506–1515. 155. Wayman KL, Cox KL, Esquivel CO. Neurodevelopmental outcome of young children with extrahepatic biliary atresia after liver transplantation. J Pediatr 1997; 131:894. 156. Sraarup EM, Hoy CE. Structured lipids improve fat absorption in normal and malabsorbing rats. J Nutr 2000; 130:2802– 2808. 157. Chin SE, Shepherd RW, Thomas BJ, et al. Nutritional support in children with end-stage liver disease: a randomised crossover trial of a branched-chain amino acid supplement. Am J Clin Nutr 1992; 56:158–163. 158. Neimark E, Shneider B. Novel surgical and pharmacological approaches to chronic cholestasis in children: partial external biliary diversion for intractable pruritus and xanthomas in Alagille syndrome. J Pediatr Gastroenterol Nutr 2003; 36:296–297.
Section XII. Inherited and Pediatric Diseases of the Liver
72
THE PORPHYRIAS Karl E. Anderson Abbreviations ADH antidiuretic hormone ADP d-aminolevulinic acid dehydratase porphyria AIP acute intermittent porphyria ALA d-aminolevulinic acid ALAD d-aminolevulinic acid dehydratase ALAS ALA synthase CEP congenital erythropoietic porphyria CPO coproporphyrinogen oxidase CRIM cross-reactive immunologic material EDTA ethylenediaminetetraacetic acid
EPP ER Fe2+ GABA GnRH HCP HEP HIV HMB HPLC
erythropoietic protoporphyria endoplasmic reticulum ferrous iron g-aminobutyric acid gonadotropin-releasing hormone hereditary coproporphyria hepatoerythropoietic porphyria human immunodeficiency virus hydroxymethylbilane high-pressure liquid chromatography
INTRODUCTION The eight human porphyrias are metabolic diseases due to deficiencies of specific enzymes of the heme biosynthetic pathway. Most are classified as hepatic (Table 72-1), because excess production and initial accumulation of intermediates occur in liver, and these are often accompanied by liver damage and increased risk for hepatocellular carcinoma. One of the two erythropoietic porphyrias can lead to severe hepatic disease requiring liver transplantation. The cDNA and genomic sequences and chromosomal locations are established for the genes of the human heme biosynthetic pathway enzymes, and multiple disease-related mutations have been found for each porphyria. A deficiency of uroporphyrinogen decarboxylase (UROD, the fifth enzyme in the pathway) can develop in the liver even in the absence of a mutation for this enzyme, and lead to porphyria cutanea tarda (PCT). However, an inherited deficiency of UROD is a predisposing factor in some cases (then termed familial PCT). All but two of the inherited porphyrias display autosomal dominant inheritance. Homozygous and heterozygous deficiencies of UROD can lead to different types of porphyria (familial PCT and hepatoerythropoietic porphyria (HEP), respectively, Table 72-1). Rare homozygous cases of the porphyrias that are autosomal dominant have been described, and may be referred to as homozygousdominant cases. Dual porphyria results from deficiencies of more than one heme biosynthetic pathway enzyme. Homozygousdominant and dual porphyria cases may have both hepatic and erythropoietic features. Severity of the hepatic porphyrias is greatly influenced by factors that alter rates of synthesis of heme pathway intermediates in the liver (in acute porphyrias) or contribute to oxidative stress in liver (in PCT). Severity of erythropoietic porphyrias generally correlates with the degree of inherited enzyme deficiency. Heme biosynthetic pathway enzymes may be deficient in some other medical conditions and toxic exposures that resemble por-
HRM PBG PBGD PCT PPO TCDD UROCoS UROD URO-OX VP
heme-regulatory motif porphobilinogen PBG deaminase porphyria cutanea tarda protoporphyrinogen oxidase 2,3,7,8-tetrachlorodibenzo-p-dioxin uroporphyrinogen III cosynthase uroporphyrinogen decarboxylase uroporphyrinogen oxidase variegate porphyria
phyrias clinically. For example, in hereditary tyrosinemia type 1, an endogenous chemical accumulates and inhibits the second enzyme of the pathway (d-aminolevulinic acid dehydratase (ALAD)). Lead and several other environmental chemicals also inhibit ALAD and cause porphyria-like manifestations. Polycyclic halogenated hydrocarbons can lead to inhibition of UROD. Porphyrin levels are commonly non-specifically increased in chronic liver diseases and other conditions without deficiencies of heme pathway enzymes.
HEME SYNTHESIS IN LIVER AND OTHER TISSUES The eight enzymes in the pathway for heme synthesis are believed to be active in all tissues, as heme is a vital substance for a variety of hemoproteins (e.g., hemoglobin, myoglobin, respiratory cytochromes, and cytochrome P450 enzymes (CYPs)). Intermediates in this important pathway (Figure 72-1) are the porphyrin precursors d-aminolevulinic acid (ALA, also known as 5-aminolevulinic acid) and porphobilinogen (PBG) and porphyrins (mostly in their reduced forms, known as porphyrinogens). These normally do not accumulate in significant amounts or have important physiological functions. A deficiency of each enzyme of the pathway, with the exception of the first, is associated with a type of porphyria (Figure 72-2). The erythroid-specific form of the first enzyme, ALA synthase (ALAS), termed ALAS2, is deficient in X-linked sideroblastic anemia, which is due to mutations of the ALAS2 gene found on chromosome Xp11.2. Disease-related mutations have not been described for the ubiquitous form of this enzyme, termed ALAS1, which is found in all tissues including liver. The ALAS1 gene is located on chromosome 3p21.1.1 Approximately 85% of daily heme synthesis in humans occurs in erythroid cells to provide heme for hemoglobin. Most of the rest occurs in the liver primarily for synthesis of CYPs, which are espe-
1391
Section XII. Inherited and Pediatric Diseases of the Liver
Table 72-1. Human porphyrias, patterns of inheritance, and classification. These classifications are based on the major tissue sites of overproduction of heme pathway intermediates (hepatic versus erythropoietic) or the type of major symptoms (acute neurovisceral versus cutaneous) Classifications Disease
Autosomal inheritance
d-Aminolevulinic acid dehydratase porphyria (ADP) Acute intermittent porphyria (AIP) Congenital erythropoietic porphyria (CEP) Porphyria cutanea tarda (PCT)
Hepatic
d-Aminolevulinic acid dehydratase (ALAD) Porphobilinogen deaminase (PBGD) Uroporphyrinogen III cosynthase (UROCoS) Uroporphyrinogen decarboxylase (UROD) ≤
Recessive Dominant Recessive Dominanta
Hepatoerythropoietic porphyria (HEP) Hereditary coproporphyria (HCP)
Deficient enzyme
Recessive
Erythropoietic
Acute
?X
X
X
X X
Cutaneous
X
X
X
X
X
X
Dominant
Coproporphyrinogen oxidase (CPO)
X
X
X
Variegate porphyria (VP)
Dominant
X
X
X
Erythropoietic protoporphyria (EPP)
Dominant
Protoporphyrinogen oxidase (PPO) Ferrochelatase (FECH)
X
X
a
PCT is primarily due to inhibition or inactivation of hepatic UROD. Autosomal dominant inheritance of a partial deficiency of UROD is a predisposing factor in cases defined as familial (type 2) PCT.
Glycine + succinyl CoA ␦-Aminolevulinic acid synthase
COOH
Feedback repression
N
N Fe
N
N
␦-Aminolevulinic acid H2N
HOOC
␦-Aminolevulinic acid dehydratase
Ferrochelatase
COOH
Heme Fe2+
NH N N HN
Protoporphyrin IX
COOH HOOC
Porphobilinogen
HOOC
H2N
N H
Protoporphyrinogen oxidase
MITOCHONDRION
Porphobilinogen deaminase COOH HOOC
HOOC
COOH HOOC
COOH HOOC
NH HN NH HN
COOH
Hydroxymethylbilane HO
N H
N H
COOH
N H
HOOC
(nonenzymatic)
Uroporphyrinogen III cosynthase
Uroporphyrinogen I
Uroporphyrinogen decarboxylase
HOOC
NH HN NH HN
COOH COOH
COOH
Coproporphyrinogen oxidase
CYTOSOL
COOH COOH
HOOC
Protoporphyrinogen IX
N H
COOH
Uroporphyrinogen decarboxylase
NH HN NH HN
COOH
Coproporphyrinogen I HOOC
COOH
Uroporphyrinogen III
HOOC
COOH
Coproporphyrinogen III
Figure 72-1. Chemical structures of intermediates of the heme biosynthetic pathway. Also shown is the mitochondrial or cytosolic locations of the eight enzymes of this pathway. The first enzyme, d-aminolevulinic acid synthase, is rate-limiting in the liver and forms the first intermediate of the pathway that is exclusively committed to heme synthesis. The pathway is regulated in liver by the end-product, heme, mainly by feedback repression (dashed arrow). Newly synthesized heme is utilized for synthesis of hemoproteins such as hemoglobin (in bone marrow) and cytochrome P450 enzymes (especially in liver). CoA, coenzyme A;
1392
Chapter 72 THE PORPHYRIAS
Intermediates
Enzymes
Diseases
␦-Aminolevulinic acid synthase
Sideroblastic anemia (X-linked)
␦-Aminolevulinic acid dehydratase
ALAD porphyria
Porphobilinogen deaminase
Acute intermittent porphyria
Uroporphyrinogen III cosynthase
Congenital erythropoietic porphyria
Uroporphyrinogen decarboxylase
Porphyria cutanea tarda
Coproporphyrinogen oxidase
Hereditary coproporphyria
Protoporphyrinogen oxidase
Variegate porphyria
Ferrochelatase
Erythropoietic protoporphyria
Glycine + succinyl CoA
␦-Aminolevulinic acid
Porphobilinogen
Hydroxymethylbilane (nonenzymatic)
Uroporphyrinogen I
Coproporphyrinogen I
Uroporphyrinogen III
Coproporphyrinogen III
Protoporphyrinogen IX
Fe2+
Heme
Figure 72-2. The sequence of intermediates and enzymes of the heme biosynthetic pathway, and the major diseases associated with deficiencies of these enzymes. CoA, coenzyme A; ALA-D, d-aminolevulinic acid dehydratase.
cially abundant in liver endoplasmic reticulum (ER, accounting for the reddish color of liver microsomal preparations). Mitochondrial respiratory cytochromes turn over much more slowly than hepatic CYPs. Thus, up to 65% of heme synthesized in liver is utilized for the formation of CYPs found in the ER, 8% for the formation of cytochrome b5 (also localized in ER), about 15% for the synthesis of catalase (localized in peroxisomes), and only 6% for the formation of mitochondrial cytochromes.2 Regulation of heme synthesis differs in these major heme-forming tissues. Heme biosynthesis in the liver is primary controlled by ALAS1. This enzyme’s rate-limiting features include its relatively low Vmax value (compared with most other enzymes in the pathway), inducibility when the requirement for hemoprotein (especially CYPs) synthesis increases,3 short half-life, and great sensitivity to repression by cellular heme (at concentrations below 10–6 mol/l).4,5 Hepatic synthesis of ALAS1 is regulated by a “free” heme pool that can be augmented by newly synthesized heme, heme released from hemoproteins, and heme entering hepatocytes from the circulation during, for example, intravenous heme administration, leading to repression of ALAS1. Heme represses the synthesis of hepatic ALAS1 at transcriptional and translational steps, and also its transfer into mitochondria. A conserved hemeregulatory motif (HRM) in the presequence of ALAS precursors is involved in hemin inhibition of transport, especially of ALAS1 into mitochondria.6,7 Therefore, when the regulatory heme pool becomes depleted in hepatocytes (which may occur when more
heme is required for synthesis of CYPs), the synthesis of ALAS1 increases. While heme can directly inhibit the activity of ALAS1, this occurs at concentrations considerably higher than are attained by heme generated in mitochondria.8 Because the affinity of ALAS1 for glycine is low, the intracellular glycine concentration also influences the rate of ALA formation. Cytosolic heme-binding proteins in liver and intestinal cells may facilitate heme uptake as well as the efflux of newly synthesized heme from mitochondria. In the erythron, novel regulatory mechanisms allow for the production of the very large amounts of heme needed for hemoglobin synthesis. The response to stimuli for heme synthesis occurs during cell differentiation, leading to an increase in cell numbers. Also unlike the liver, the stimulation of heme synthesis in erythroid cells is not only accompanied by increases in ALAS (especially ALAS2), but also by sequential induction of other heme biosynthetic enzymes, with heme having a stimulatory role in hemoglobin formation.1 Separate erythroid-specific and non-erythroid or “housekeeping” transcripts are known for the first four enzymes in the pathway. For ALAS, these separate forms are encoded by genes on different chromosomes (see above), whereas for each of the other three, erythroid and non-erythroid transcripts are transcribed from the same gene. Heme also regulates the rate of its synthesis in erythroid cells by controlling the transport of iron into reticulocytes.9 Control of heme biosynthesis in tissues other than the liver and erythroid cells may be different but has been less studied.1
1393
Section XII. Inherited and Pediatric Diseases of the Liver
HEME PATHWAY INTERMEDIATES – CHARACTERISTICS AND ROUTES OF EXCRETION Under normal conditions, the intermediates of the heme biosynthetic pathway are efficiently utilized for heme synthesis, and are only excreted in small amounts. Some undergo chemical modifications before excretion. ALA and PBG are colorless and nonfluorescent compounds that are largely excreted unchanged in urine. However, PBG may degrade to colored products such as the brownish pigment called porphobilin, and may spontaneously polymerize to uroporphyrins. Porphyrins are red in color and display bright-red fluorescence when exposed to long-wavelength ultraviolet light. Porphyrinogens, which are colorless and non-fluorescent, are the reduced form of porphyrins and when they accumulate are readily autoxidized to the corresponding porphyrins outside the cell. Porphyrins and porphyrinogens with many carboxyl groups are water-soluble and readily excreted in urine. Those with fewer carboxyl groups are less water-soluble and are excreted primarily in bile and feces. Therefore, in the various porphyrias excess ALA, PBG, uroporphyrin (octacarboxyl porphyrin), and hepta-, hexa-, and pentacarboxyl porphyrins are excreted mostly in the urine, coproporphyrin (tetracarboxyl porphyrin) is excreted partly in urine and partly in bile, and harderoporphyrin (tricarboxyl porphyrin) and protoporphyrin (dicarboxyl porphyrin) are excreted almost entirely in bile and feces. Coproporphyrin I is more readily excreted in bile than is coproporphyrin III, which may explain the increase in the ratio of these isomers that occurs when hepatobiliary function is impaired.10
CLASSIFICATION AND DIAGNOSIS OF HUMAN PORPHYRIAS An understanding of the classification of the porphyrias is important for diagnosis of these conditions (Table 72-1). Porphyrias have been classified as hepatic or erythropoietic, according to the source of excess production of porphyrin precursors and porphyrins. The four acute porphyrias are associated with neurologic symptoms and elevated plasma and urinary concentrations of one or both of the porphyrin precursors ALA and PBG. In the cutaneous porphyrias porphyrins are transported in blood from the liver or bone marrow to the skin where they cause cutaneous photosensitivity. Patients
with dual porphyria have deficiencies of more than one heme pathway enzyme. Acute intermittent porphyria (AIP), PCT, and erythropoietic protoporphyria (EPP) are the three commonest porphyrias, and are likely to be encountered by any physician. It is important to be aware of the striking differences in their clinical presentations, precipitating factors, tests for diagnosis, and effective therapies (Table 72-2). Although these three porphyrias share no common features, AIP and PCT can be considered prototypic, in that there are some important similarities with other less common porphyrias.
LABORATORY TESTING FOR PORPHYRIAS First-line tests that are both sensitive and specific should be relied upon for diagnosis, and if positive, more comprehensive testing should follow to establish the type of porphyria. Laboratory reports that were the basis for a past diagnosis must be examined, and if these were inadequate, further testing should be considered. Many porphyria-related tests are available to practicing physicians,11 but their overuse is expensive and can delay diagnosis. Overuse of tests can also lead to misdiagnoses in patients who do not have these disorders, especially in patients with compatible (but non-specific) symptoms and minimal increases in urinary porphyrins or porphyrin precursors that may have no diagnostic significance. Abnormal porphyrin results can occur in disorders other than porphyrias, such as chronic liver disease.10
FIRST-LINE TESTING Different screening tests are appropriate for acute and cutaneous porphyrias.
Acute Porphyries Urinary porphyrin precursors (ALA and PBG) should be measured in patients with neurovisceral symptoms, such as abdominal pain, when initial clinical evaluation does not suggest another cause. Because urinary PBG is always increased during acute attacks of AIP, hereditary coproporphyria (HCP), and variegate porphyria (VP), and is not substantially increased in any other medical conditions, this will detect the great majority of patients with acute porphyrias, and a normal level will almost always correctly exclude the diagnosis.12
Table 72-2. The three most common human porphyrias and the major features of these disorders that are distinctly different from each other Presenting symptoms
Exacerbating factors
Most important screening tests
Treatment
Neurologic
Drugs (mostly P450-inducers); progesterone; dietary restriction Iron; alcohol; smoking, estrogens; hepatitis C; HIV, halogenated hydrocarbons
Urinary porphobilinogen
Heme; glucose
Plasma (or urine) porphyrins
Phlebotomy; low-dose hydroxychloroquine or chloroquine b-Carotene
Acute intermittent porphyria Porphyria cutanea tarda
Skin-blistering and fragility (chronic)
Erythropoietic protoporphyria
Skin pain and swelling (mostly acute)
HIV, human immunodeficiency virus.
1394
Erythrocyte (or plasma) porphyrins
Chapter 72 THE PORPHYRIAS
Rapid testing for urinary PBG should be available in-house at all major medical facilities. Spot (single void) urine specimens are generally preferred, and are informative because substantial increases are expected during acute attacks of porphyria, and a 24-h collection can unnecessarily delay diagnosis. The same urine specimen should be saved for quantitative determination of ALA and PBG, in order to confirm the qualitative PBG result, and also detect patients with ALAD porphyria (ADP). It is also useful to measure total urinary porphyrins, because urinary porphyrins may remain increased longer than porphyrin precursors in HCP and VP. But it must be kept in mind that urinary porphyrin increases are often nonspecific.12 Most tests for PBG, a colorless pyrrole, rely on formation of a violet pigment with Ehrlich’s reagent (p-dimethylaminobenzaldehyde). PBG must be separated from other urinary substances, principally urobilinogen, that also react with Ehrlich’s aldehyde. A reliable quantitative method for both ALA and PBG, using small anion and cation exchange columns to separate interfering substances before adding Ehrlich’s reagent, has been available for many years.13 ALA is reacted to form a pyrrole, which is then also measured using Ehrlich’s reagent.13 The Trace PBG kit (Trace America/Trace Diagnostics, Louisville, Colorado) is a reliable qualitative test based on this method.14 The Watson–Schwartz15 and Hoesch tests,16 which involved initial addition of Ehrlich’s reagent to urine, are considered obsolete.
Blistering Cutaneous Porphyries Total plasma porphyrins are always increased in patients with active, blistering skin lesions due to porphyria, most commonly PCT. A simple and direct fluorometric method first described by PohFitzpatrick is preferred.17–20 The porphyrins in plasma in VP are mostly covalently linked to plasma proteins and are readily detected by this method but may not be detected by high-pressure liquid chromatography (HPLC).21 The normal range for plasma porphyrins is higher in patients with end-stage renal disease than in normals.22
Non-blistering Cutaneous Porphyria EPP is unique among the porphyrias, because it presents with nonblistering photosensitivity that is experienced shortly after sunlight exposure. A total plasma porphyrin determination will detect most patients with this condition, but an erythrocyte protoporphyrin determination is more sensitive. Unfortunately, this measurement is
subject to non-specific increases in many conditions other than protoporphyria.
SECOND-LINE TESTING The three porphyrias that cause increases in PBG can be differentiated from each other by measuring erythrocyte PBG deaminase (PBGD), urinary porphyrins, fecal porphyrins, and plasma porphyrins, as detailed in Table 72-3. Urinary porphyrins can be measured in the same spot sample used for initial screening.12 Measurements of porphyrins in urine, feces, and plasma will also differentiate the various porphyrias that cause blistering skin lesions. Further diagnostic details are covered in the sections on each type of porphyria.
LABORATORY TESTING OF RELATIVES AND PATIENTS WITH SUBCLINICAL PORPHYRIA Latent acute porphyrias can be difficult to detect in relatives who have never had symptoms, as porphyrin precursors and porphyrins may remain normal. It is also difficult to rule out porphyria when symptoms were present months or years previously, because porphyrin precursors and porphyrins can decrease over time and eventually even normalize. In these instances, more extensive testing and consultation with a specialist laboratory and physician may be needed. The diagnosis of porphyria should be firmly established in a propositus or index case, and the laboratory results from that individual obtained to guide the choice of tests for other family members. Retesting the propositus or another family member with confirmed porphyria may be necessary. Identification of a diseasecausing mutation in an index case greatly facilitates the detection of additional gene carriers.
d-AMINOLEVULINIC ACID DEHYDRATASE PORPHYRIA History, Definition, and Prevalence This is the most recently described human porphyria, and is undoubtedly underrecognized. It is sometimes termed Doss porphyria after the investigator who described the first two cases.23 The
Table 72-3. Results of second-line laboratory tests that differentiate the three acute porphyrias that cause substantial increases in porphobilinogen Disease
Erythrocyte porphobilinogen deaminase
Urine porphyrins
Fecal porphyrins
Plasma porphyrins
Acute intermittent porphyria Hereditary coproporphyria Variegate porphyria
Decreased by ~50% (in ~90% of cases) Normal
Markedly increased, mostly uroporphyrin Markedly increased, mostly coproporphyrin Markedly increased, mostly coproporphyrin
Normal or slightly increased Markedly increased, mostly coproporphyrina Markedly increased, mostly coproporphyrina and protoporphyrin
Normal or slightly increased Usually normal
Normal
Markedly increased, characteristic fluorescence peak18
a
Mostly coproporphyrin III.449,450
1395
Section XII. Inherited and Pediatric Diseases of the Liver
term plumboporphyria emphasizes the similarity of this condition to lead poisoning, but incorrectly implies that it is due to lead exposure. The underlying deficiency of ALAD is inherited as an autosomal recessive trait. Only six cases have been confirmed by mutation analysis,23–29 and a few additional cases have been reported without such confirmation.30–32 The prevalence of heterozygous ALAD deficiency was estimated to be less than 1% (2 or 139 individuals) in Germany, where three cases were recognized,33 and ~2% among 880 individuals in Sweden.25
Clinical Manifestations Most well-documented cases have been adolescent males with symptoms resembling other acute porphyrias, including abdominal pain and neuropathy.23,29 Precipitating factors, such as exposure to harmful drugs, have not been evident in most cases. The first two German cases had initial acute attacks and then remained well during 20 years of follow-up.34 A third German case was a 17-yearold male who presented with a 2-year history of abdominal pain and polyneuropathy.28 A 14-year-old male case recently documented in the USA also presented with typical acute attacks.29 A Swedish infant had more severe disease, including failure to thrive.25 A 63year-old man in Belgium developed an acute motor polyneuropathy concurrently with a myeloproliferative disorder. He was heterozygous for ALAD deficiency and porphyria was thought to have resulted from expansion of an erythrocytic clone deficient in ALAD.35,36 Cases not documented by molecular studies include Chilean siblings (26 and 28 years of age) who presented with symptoms typical of acute porphyria and ALAD activity less than 4% of normal,30 a 69-year-old Japanese woman with manifestations that included hyponatremia and the syndrome of inappropriate antidiuretic hormone (ADH) secretion,32 and a lead worker with bilateral wrist drop and persistently deficient erythrocyte ALAD activity who, in the absence of toxic lead levels, was considered to have this condition.31 Heterozygotes have been asymptomatic, but may have enhanced susceptibility to lead,33 and to the toxic effects of other chemicals such as iron, trichloroethylene, and styrene that can adversely affect the normal enzyme.37
Etiology and Pathogenesis An inherited deficiency of ALAD is the underlying cause of this form of porphyria. Human ALAD is a zinc-containing enzyme consisting of eight identical 31-kDa subunits, and catalyzes the condensation of two molecules of ALA to form a pyrrole, PBG (Figure 72-1).38 The octamer has four active sites and four catalytic zincs bound by cysteine ligands, and four additional zincs. Each active site contains two substrate-binding sites.39 The tertiary structure of the yeast enzyme has been described at a resolution of 0.23 nm.40 ALAD is identical to the 240-kDa proteasome inhibitor (CF-2).41 Its dual role in both the adenosine triphosphate/ubiquitin-dependent pathway and the heme biosynthetic pathway may be an example of “gene-sharing” and may explain the unexpected abundance of this enzyme in the cell, as compared to other heme biosynthetic pathway enzymes. Inhibition of erythrocyte ALAD activity is a sensitive index of lead exposure.42 Lead and other heavy metals can displace the zinc atoms of ALAD, and indeed ALAD is the principal lead-binding
1396
protein in erythrocytes.43 A common ALAD polymorphism, K59N, termed ALAD2, is detected in ~20% of caucasians.44 The polymorphic enzyme retains normal ALAD activity but may bind zinc more effectively, and in epidemiologic studies was associated with higher lead levels.45,46 It is not known whether this polymorphism alters lead toxicity, but it may affect lead distribution to target organs such as the brain. The enzyme is also subject to inhibition by other exogenous and endogenous chemicals.27 The human ALAD gene is found on chromosome 9 (9q34).47 Two promoter regions and two alternative non-coding exons, 1A and 1B, provide housekeeping and erythroid-specific transcripts that both include 12 coding exons.48 The housekeeping transcript consists of exon 1A fused to coding exons 2–12, and the erythroid-specific transcript exon 1B fused to exons 2–12. Because translation begins in exon 2, both transcripts encode the same amino acid sequence, but differences in the upstream promoter regions for exons 1A and 1B allow for expression of housekeeping and erythroid-specific transcripts.48 All d-aminolevulinic acid dehydratase porphyria (ADP) cases described to date inherited a different ALAD mutation from each unrelated parent. Eleven ALAD mutations, mostly point mutations, have been identified.28,29,49,50 As expected, some expressed enzyme product with partial activity, such that heme synthesis was partially preserved. The amount of residual enzyme activity may predict the phenotypic severity of this disease.49 Immunochemical studies in three cases demonstrated non-functional enzyme protein that crossreacted with anti-ALAD antibodies.51,52 Late-onset cases associated with a myeloproliferative disorder may be heterozygous or have a somatic mutation, with expansion of an affected clone of erythroid cells. Like other acute porphyrias, ADP is often classified as hepatic, although the site of overproduction of ALA is not established. The patient with severe, early-onset disease underwent liver transplantation, without significant clinical or biochemical improvement,53 which might suggest that the excess intermediates did not originate in the liver. Excess urinary coproporphyrin III may originate from metabolism of ALA to porphyrinogens in a tissue other than the site of ALA overproduction. Studies in normal subjects have shown that administration of large doses of ALA leads to substantial coproporphyrinuria.54 The increased erythrocyte protoporphyrin may, as in all other homozygous porphyrias, be explained by accumulation of earlier pathway intermediates in bone marrow erythroid cells during hemoglobin synthesis, followed by their transformation to protoporphyrin after hemoglobin synthesis is complete.55 As in other acute porphyrias, the pathogenesis of the neurologic symptoms is poorly understood (see section on AIP, below).
Laboratory Evaluation and Diagnosis Characteristic laboratory findings are markedly increased urinary ALA and coproporphyrin III and increased erythrocyte zinc protoporphyrin. Urinary PBG is normal or slightly increased. Erythrocyte ALAD activity is markedly reduced, and is not restored by the in vitro addition of sulfhydryl reagents such as dithiothreitol, which helps distinguish this disease from lead poisoning. Both parents should have approximately half-normal activity of this enzyme and normal urinary ALA.
Chapter 72 THE PORPHYRIAS
Confirmation of the diagnosis by mutation analysis is essential. Lead, styrene, and succinylacetone (which accumulates in hereditary tyrosinemia type 1 and is structurally similar to ALA) inhibit ALAD, causing increased urinary excretion of ALA and clinical manifestations that resemble those of acute porphyria.27,56 Lead poisoning is excluded by finding a normal blood lead level and showing that ALAD activity is not restored by dithiothreitol. Idiopathic acquired ALAD deficiency has been reported.57
Treatment Limited experience indicates that glucose is not very effective but may be tried for mild symptoms. Hemin therapy was apparently effective in the several cases of adolescent males with acute attacks, as manifested by decreases in urinary or serum levels of ALA as well as clinical improvement.28,29,34 In one, weekly infusions to prevent attacks were beneficial.28 Hemin was not effective either biochemically or clinically in the Swedish child with severe disease,25 and produced a biochemical response but no clinical improvement in the late-onset case in Belgium, who had a peripheral neuropathy but no acute attacks.35 Hemin is also effective in treating porphyria-like symptoms associated with hereditary tyrosinemia,58 and can significantly reduce urinary ALA and coproporphyrin in lead poisoning.59 Avoidance of drugs that are harmful in other acute porphyrias is prudent; however it is not known if this is helpful.35 Liver transplantation, which was not effective in the child with severe disease,53 has not been attempted in patients with more typical presentations.
ACUTE INTERMITTENT PORPHYRIA History, Definition, and Prevalence AIP is the most common of the acute porphyrias in most countries, and is considered the prototypic example of these disorders. It results from a deficiency of the housekeeping form of PBGD. It has also been termed pyrroloporphyria, Swedish porphyria, and intermittent acute porphyria. Acute porphyria was first described by Stokvis in 1889 in an elderly woman in Holland who had taken the sedative drug Sulfonal. The patient excreted red urine and later died. Remarkably, exacerbation by a drug was part of the original description, as Stokvis showed that porphyrinuria could be induced in rabbits with this drug.60,61 In 1931, Sachs noted that the urine of a patient with acute porphyria contained a substance other than urobilinogen that reacted with Ehrlich’s aldehyde to form a red pigment. In 1937 Waldenström reported that this material, which he termed PBG, occurred in the urine of some relatives of patients with acute porphyria, and suggested that the disease was transmitted as an autosomal dominant trait and was often latent.61 AIP occurs in all races. Prevalence in most countries is commonly estimated to be ~5 per 100 000, but such estimates are not precise. AIP is more common than HCP or VP in most locations. Prevalence has been estimated to be 7.7 per 100 000 in Sweden (including latent cases with normal porphyrin precursors).62 The prevalence is very high in northern Sweden (approximately 60–100 per 100 000), where almost all cases have the same PBGD mutation.63 In Finland the combined prevalence of AIP and VP was calculated to be 3.4 per 100 000.64 A high prevalence (210 per 100 000) was reported in a chronic psychiatric population in the USA based on erythrocyte PBGD determinations,65 whereas a study in Mexico found a similar
prevalence in psychiatric patients and controls.66 Population screening by erythrocyte PBGD activity or DNA analysis revealed a prevalence of ~200 heterozygotes per 100 000 in Finland,67 and 1 in about 1675 (60 in 100 000) in France.68 These findings suggest that undetected gene carriers are common. Suggestions that a patient described by Hippocrates69 and that notable figures in history, including King George III, Mary Queen of Scots, and Vincent van Gogh,70 had acute porphyria, perhaps in the case of George III exacerbated by arsenic,71 lack laboratory confirmation and are highly speculative. DNA studies of exhumed tissues of two descendants of George III have not established a diagnosis of AIP or VP in the British royal family.71 The suggestion that AIP explains the vampire legends is unfounded and portrays the disease unfavorably.72 Proposals that patients with poorly understood conditions such as multiple chemical sensitivity syndrome actually have acute porphyria are also poorly founded.73–75
Clinical Manifestations Neurovisceral manifestations of acute porphyrias are due to effects on the nervous system. Adverse effects on the liver and kidney sometimes occur as well. The disease can be disabling but is only occasionally fatal. Because the disease so often remains latent, there is often no family history of porphyria. Symptoms may appear any time after puberty but often not until the third or fourth decade of life, and more commonly in women than men. But in some women attacks first appear after menopause. The disease is rarely manifest in children. In very rare cases of homozygous AIP severe neurological manifestations began in childhood, and acute attacks were less prominent.76–78 Abdominal pain is the most common symptom and has been reported in 85–95% of cases.79–81 Tachycardia is the most common physical sign, occurring in up to 80% of acute attacks.81 Abdominal pain is usually severe, steady, and poorly localized, but may be cramping. It is usually accompanied by signs of ileus, including abdominal distention and decreased bowel sounds. Other manifestations include nausea, vomiting, and constipation. However, increased bowel sounds and diarrhea may occur. Bladder dysfunction may cause hesitancy and dysuria. Because the abdominal manifestations are neurologic rather than inflammatory, there is little or no tenderness, fever, or leukocytosis. Other manifestations include hypertension, mental symptoms, limb, head, neck, or chest pain, muscle weakness, and sensory loss. Pain in the extremities is often described as muscle or bone pain, and may be a manifestation of early peripheral neuropathy. Tachycardia, hypertension, restlessness, course or fine tremors, and excess sweating may be due to sympathetic overactivity and increased catecholamines.1 Peripheral neuropathy in the acute porphyrias is primarily motor and appears to result from axonal degeneration rather than demyelinization.82–84 Paresis does not occur in all patients with attacks of porphyria, even when abdominal symptoms are severe. Weakness most commonly begins in the proximal muscles during an attack and more often in the arms than in the legs. Rarely, a significant degree of peripheral neuropathy develops acutely in this and other acute porphyrias when there is little or no abdominal pain.85 Motor weakness is often symmetrical, but may be asymmetrical or even strikingly focal. Tendon reflexes may be little affected or hyperactive in early stages but are usually decreased or absent when
1397
Section XII. Inherited and Pediatric Diseases of the Liver
neuropathy is advanced. Cranial nerves, most commonly the 10th and seventh, can be affected. Rarely, involvement of the optic nerves or occipital lobes may produce blindness. There may be sensory involvement with areas of paresthesia, dysesthesia, and loss of sensation. Chronic neuropathy may occur, but is less severe than with acute attacks.86 The course of the neurological manifestations is highly variable. Muscle weakness can progress to respiratory and bulbar paralysis and death. But this seldom occurs unless the disease is not recognized, harmful drugs not discontinued, and appropriate treatment not instituted. Sudden death may occur, presumably due to cardiac arrhythmia.82,87 Attacks may resolve quite rapidly, with abdominal pain disappearing within a few hours and paresis within a few days. Even advanced motor neuropathy is potentially reversible. Motor function can improve for several years after a severe attack but may leave some residual weakness.80 Involvement of the central nervous system is common. Seizures may occur as a neurologic manifestation of the acute attack itself, or result from hyponatremia (see below) or a concurrent condition. Mental symptoms such as anxiety, insomnia, depression, disorientation, hallucinations, and paranoia can be especially severe during acute attacks, suggesting a primary mental disorder or hysteria, and patients may become violent and difficult to control. Chronic depression and other mental symptoms sometimes occur, but whether these are caused by AIP is often difficult to determine. Hyponatremia is common during acute attacks and is sometimes due to inappropriate ADH secretion; autopsy studies suggest that this can result from damage to the supraoptic nuclei of the hypothalamus.88 However, unexplained reductions in total blood and red blood cell volumes occur in some patients89 and increased ADH secretion in this setting can be considered an appropriate physiologic response, and hypothalamic changes may reflect a hyperfunctional state rather than a degenerative process.87 Some have suggested that salt depletion from gastrointestinal loss and poor intake, and excess renal sodium loss are more frequently the cause of hyponatremia in AIP.87,90,91 A possible nephrotoxic effect of ALA may explain renal tubular sodium loss and impaired renal function in some patients.90,92 Other electrolyte abnormalities may include hypomagnesemia and hypercalcemia—the latter as a result of prolonged immobilization.93 Previously the mortality rate for patients with acute attacks was as high as 80%. However, the morbidity and mortality of these conditions have markedly improved in the past several decades. For example, 74% of patients with AIP or VP in Finland reported in 1992 that they lead normal lives. During several years of follow-up fewer than one-third of patients had recurrences of attacks and only 6% of gene carriers who had never had attacks developed symptoms. In patients who presented with acute symptoms, recurrent attacks were most likely within the subsequent 1–3 years.94 The improved outlook in acute porphyrias may be attributed in part to earlier detection, less use of drugs such as barbiturates and sulfonamides in medical practice, and better treatment of acute attacks. However, some patients have recurrent attacks, chronic pain, and other symptoms even after avoiding known exacerbating factors. AIP is commonly associated with mild abnormalities in liver function.95 More advanced liver disease and hepatocellular carcinoma
1398
sometimes complicate any of the acute porphyrias, including asymptomatic individuals who have increased porphyrins or porphyrin precursors.96–100 Hepatocellular carcinoma is not related to specific PBGD mutations.101,102 The incidence or mortality of this neoplasm may be 60–70-fold greater in acute porphyrias than in age- and sexmatched controls.96,100,103,104 A prospective study of 650 patients with acute porphyrias in France found seven cases of hepatocellular carcinoma over a period of 7 years, versus an expected occurrence of 0.2 cases.101,102 Only a minority of patients have had increases in serum a-fetoprotein. Therefore, current recommendations are that patients with acute porphyrias, especially over age 50, be screened at least annually by ultrasound or an alternative imaging method to detect hepatocellular carcinoma at an early stage.105 AIP may also predispose to chronic hypertension and impaired renal function.94,106,107 In some cases this has progressed to severe renal failure and successful renal transplantation.108,109 Some patients with AIP have increased serum thyroxin levels due to increased thyroxin-binding globulin. Occasionally true hyperthyroidism and porphyria occur together.110 Hypercholesterolemia and elevated low-density lipoprotein cholesterol appear to be less common in this disorder than previously thought,111 and may be accompanied by an increase in high-density lipoprotein cholesterol and apo-A1.112
Etiology and Pathogenesis An inherited deficiency of PBGD is the underlying cause of AIP. This enzyme (also known as hydroxymethylbilane (HMB) synthase, and formerly as uroporphyrinogen I synthase) catalyzes the deamination and head-to-tail condensation of four PBG molecules to form the linear tetrapyrrole HMB (also known as preuroporphyrinogen, Figure 72-2). HMB can rapidly cyclize in the presence of uroporphyrinogen III cosynthase (UROCoS) to uroporphyrinogen III, or non-enzymatically and less rapidly to uroporphyrinogen I. A unique feature of PBGD is its dipyrromethane cofactor, which binds the pyrrole intermediates at the catalytic site until six pyrroles (including the dipyrrole cofactor) are assembled in a linear fashion, after which the tetrapyrrole HMB is released.113 The apodeaminase generates the dipyrrole cofactor to form the holo-deaminase, and this occurs more readily from HMB than from PBG.114 Indeed, high concentrations of PBG can inhibit formation of the holo-deaminase.115 A single PBGD gene on chromosome 11 (11q24.1Æq24.2) contains 10 kb and 15 exons and accounts for two tissue-specific isoenzymes.116–118 Erythroid and non-erythroid (housekeeping) forms of the enzyme, resulting from alternative splicing of exons 1 and 2, are both monomeric and differ only slightly in molecular weight (~38 and 40 kDa, respectively).117 The erythroid-specific isoenzyme is encoded by exons 2 through 15, and the erythroid promoter, which functions only in erythroid cells, is found immediately upstream from exon 2. The non-erythroid enzyme is encoded by exons 1 and 3 through 15, and its promoter is immediately upstream from exon 1. Cis-acting sequences bind the erythroid-specific DNA-binding factors NF-E1 and NF-E2, and are necessary for expression of the erythroid promoter.119 Because the housekeeping promoter functions in all cell types, both enzyme forms can be expressed in erythroid cells.120
Chapter 72 THE PORPHYRIAS
More than 200 PBGD mutations have been identified, including missense, nonsense, and splicing mutations, as well as insertions and deletions.50,121,122 Most have been reported in only one or a few families, and they occur in many population groups, including blacks.123 A few are more common in Sweden (W198X), Holland (R116W), Argentina (G116R), Nova Scotia (R173W), and Switzerland (W283X), and are examples of the founder effect.63,124–126 Chester porphyria, which was initially described as a variant condition in a single, very large family in England,127 was recently found to be due to a PBGD mutation. De novo mutations may be found in approximately 3% of cases.128 The nature of the underlying mutation in AIP explains little, if any, of the marked variation in clinical expression of the disease. However, the crystal structure of Escherichia coli PBGD, which has 35% homology and more than 70% similarity to the human enzyme, provides insight into structure–function relationships of the human enzyme and allows prediction of effects of known human mutations.1,129 For example, mutations R167Q and R173Q involve highly conserved arginines, with R167Q predicted to impair substrate binding. Many exon 10 mutations are positioned at conserved amino acids in the active site, while many exon 12 mutations alter the ahelix connecting domains 1 and 3. Immunochemical methods can distinguish three classes of mutations.121,130 Type I is most common, and leads to approximately halfnormal PBGD activity and enzyme protein in all tissues.50,121 Type II mutations, which cause the uncommon, non-erythroid variant of AIP (~5% of unrelated patients), leads to absence of the housekeeping isozyme whereas the erythroid-specific isozyme is normal. These mutations may, for example, alter the 5¢ splice donor site of intron 1, the exon 1 initiation of translation codon,131–134 the ubiquitous housekeeping promoter, or cause a frameshift in exon 3 that introduces a stop codon into the mRNA for the housekeeping isoform only.135 Type III mutations are CRIM-positive (excess crossreactive immunologic material (CRIM) relative to enzyme activity) and include mutations that decrease PBGD activity, but do not substantially decrease enzyme stability. Retrospective molecular analysis of lymphoblastoid cell lines from a child with homozygous AIP showed that she inherited a different CRIM-positive mutation from each parent.78
Pathogenesis of Acute Attacks In gene carriers who have never developed symptoms, porphyrin precursor excretion is usually normal, and presumably hepatic ALAS1 activity is not increased. Therefore, half-normal hepatic PBGD activity in these individuals is sufficient to avoid any accumulation of PBG. Most of the additional factors known to be necessary for clinical expression of AIP, including many drugs and steroid hormones, have the capacity to induce the synthesis of hepatic ALAS1 and CYPs, thereby increasing hepatic production of ALA, PBG, and other heme pathway intermediates. Half-normal PBGD activity can then become limiting for heme synthesis, and heme-mediated repression of ALAS is also impaired. The notion that hepatic PBGD remains constant at ~50% of normal activity during exacerbations and remission of AIP, as in erythrocytes, is not proven, and an early report suggested that the enzyme activity is considerably less than half-normal in liver during an acute attack.136 A suggested mechanism of further impairment of
hepatic PBGD once the disease becomes activated is that excess PBG may interfere with assembly of the dipyrromethane cofactor for this enzyme.115 Additional genetic factors that are not currently known are also likely to contribute, particularly in individuals who remain susceptible to repeated attacks even after avoidance of known precipitants.
Endocrine Factors Many observations suggest that steroid hormones are among the most important factors that contribute to clinical expression of acute porphyrias. For example, AIP gene carriers are rarely symptomatic before puberty and porphyrin precursor excretion almost always remains normal in childhood. The disease is expressed clinically in some individuals after adult levels of steroid hormones are achieved. Clinical expression is more common in women than in men, suggesting an important role for female hormones. Cyclic, premenstrual attacks of the disease occur in some women, are probably due to endogenous progesterone, and can be prevented by the administration of gonadotropin-releasing hormone (GnRH) analogues.137 Acute porphyrias are sometimes exacerbated by exogenous steroids, including oral contraceptive preparations. Changes in steroid hormone metabolism, such as a deficiency of hepatic steroid 5a-reductase activity, in some patients with AIP may predispose to the excess production of steroid hormone metabolites that are inducers of hepatic ALAS1.138 Pregnancy is usually well tolerated. In a large case series (76 women with AIP or VP with a total of 176 deliveries), there were no porphyric symptoms in 92% of pregnancies.94 Worsening symptoms during pregnancy are sometimes due to harmful drugs or reduced caloric intake. Metoclopramide, a contraindicated drug, has been associated with exacerbation of the disease when used to treat hyperemesis gravidarum.139,140 However, some women experience continuing attacks during pregnancy even when other harmful factors are avoided. Although progesterone levels are considerably increased during pregnancy, and this steroid hormone and its metabolites are potent inducers of hepatic ALAS1,3,5 beneficial metabolic changes may favor less induction of ALAS1.141 Other endocrine conditions, including diabetes mellitus, are not known to precipitate attacks of porphyria. In fact, the onset of diabetes mellitus has been observed to decrease the frequency of attacks and lower porphyrin precursor levels, possibly in relation to high circulating glucose levels.142
Drugs Many patients do well after the diagnosis is established and harmful drugs are avoided. The major drugs known or strongly suspected to be harmful in the acute porphyrias, as well as those known to be safe, are listed in Table 72-4. Updated and more extensive listings are available from the European Porphyria Initiative (www. porphyria-europe.com/) and the American Porphyria Foundation (www.porphyriafoundation.com/). Safe and established drugs should be chosen for treating concurrent diseases whenever possible. However, information regarding safety is lacking for many drugs, especially for those recently introduced, and sometimes opinions are conflicting. Within some classes of drugs, some are reported as safe and others unsafe. For example, some calcium-channel antagonists induce ALAS1 and CYPs in
1399
Section XII. Inherited and Pediatric Diseases of the Liver
Table 72-4. Drugs regarded as unsafe and safe in acute porphyrias Unsafe
Safe
Barbituratesa Sulfonamide antibioticsa Meprobamatea (also mebutamate,a tybutamatea) Carisoprodola Glutethimidea Methyprylon Ethchlorvynola Mephenytoin Phenytoina Succinimides Carbamazepinea Clonazepam Primidonea Valproic acida Pyrazolones (aminopyrine, antipyrine) Griseofulvina Ergots Metoclopramidea Rifampina Pyrazinamidea Diclofenaca Progesterone and synthetic progestinsa Danazola Alcohol Angiotensin-converting enzyme (ACE) inhibitors (especially enalapril) Calcium-channel blockers (especially nifedipine) Ketoconazole Rifampin
Narcotic analgesics Aspirin Phenothiazines Penicillin and derivatives Streptomycin Glucocorticoids Bromides Insulin Atropine Cimetidine Ranitidineb Acetaminophen (paracetamol) Acetazolamide Allopurinol Amiloride Bethanidine Bumetanide Cimetidine Coumarins Fluoxetine Gabapentin Gentamicin Guanethidine Ofloxacin Propranolol Succinylcholine Tetracycline
This partial listing does not include all available information about drug safety in acute porphyrias. Other sources should be consulted for drugs not listed here. a Porphyria is listed as a contraindication, warning, precaution, or adverse effect in US labeling for these drugs. Estrogens are also listed as harmful in porphyria, but have been implicated as harmful in acute porphyrias based mostly on experience with estrogen–progestin combinations. While estrogens can exacerbate porphyria cutanea tarda, there is little evidence that they are harmful in the acute porphyrias. b Porphyria is listed as a precaution in US labeling for this drug. However, this drug is regarded as safe by other sources.
rats143,144 and/or chick embryo liver cells,144–146 and were implicated in some attacks of porphyria.147 Of these agents, amlodipine was safely administered in one patient148 but not another.149 Angiotensinconverting enzyme inhibitors also differ in their capacity to induce porphyrin accumulation in cultured liver cells, with enalapril being the most potent.144 Thiazide diuretics have been generally regarded as safe, but hydrochlorothiazide147 and even all thiazides150 have been listed as unsafe. There is disagreement regarding safety of several antibiotics, such as chloramphenicol, cephalosporins, erythromycin, and vancomycin.150 The antidepressant nefazodone is a potent inducer in cultured liver cells.151 Major surgery can be carried out safely in patients with acute porphyria if harmful factors, such as barbiturates, are avoided.152 Anesthetic agents have been studied in animal models and their use in patients with porphyria reviewed.153,154 Halothane has been recommended as an inhalation agent and propofol and midazolam appear suitable as intravenous induction agents.154 Immunosuppressive regimens have been employed in patients who required renal
1400
transplantation.108,109 Antineoplastic drugs have been safely administered to patients with acute porphyrias and cancer.155 Most drugs that exacerbate porphyria have the capacity to induce hepatic ALAS1, which is closely associated with induction of CYPs and increased demand for hepatic heme synthesis.3 Some drugs, such as griseofulvin, can increase heme turnover by promoting the destruction of specific CYPs and form inhibitors of ferrochelatase (FECH, the final enzyme in the pathway), such as N-methyl protoporphyrin. Sulfonamide antibiotics are harmful but are apparently not inducers of hepatic heme synthesis.156 The suggestion that they inhibit PBGD has been questioned.157 Ethanol and other alcohols found in beverages are inducers of ALAS and some CYPs.158,159 Clinical observations suggest that attacks following ingestion of a harmful drug are partly due to the additive effects of other predisposing factors, such as other drugs, endogenous hormones, nutritional factors, and smoking. For example, exposure to drugs and other precipitating factors poses a lower risk in patients who have had no recent symptoms of acute porphyria than in those with recent and frequent symptoms.94 This was indicated by a large retrospective study of risk from anesthetic use, in which barbiturates or other inducing drugs were frequently detrimental in patients who already had porphyric symptoms but seldom exacerbated latent disease.153
Nutritional Factors Reduced dietary intake, often in an effort to lose weight, commonly contributes to attacks, and may be underrecognized, because obtaining an accurate dietary history is difficult. Starvation in animals enhances, whereas glucose or protein can repress, the inducing effect of chemicals on ALAS1 and PBG excretion.160,161 Starvation also induces heme oxygenase,162 which may deplete regulatory hepatic heme pools and contribute to ALAS1 induction. Reductions in caloric and carbohydrate intakes increased urinary excretion of ALA and PBG and precipitated symptoms of acute porphyria during metabolic ward studies,163,164 and carbohydrate loading is sometimes effective in treating mild attacks.12 The carbohydrate effect on ALAS may be secondary at least in part to reduced amounts (and presumably synthesis) of hepatic CYPs.165 Effects of borderline iron deficiency, which is common in young women and might impair hepatic heme synthesis, have not been studied in acute porphyrias. But it is reasonable to assure adequate status of iron as well as all other essential nutrients. More importantly, it was shown recently that hepatic ALAS1 is regulated by the proxisome proliferator-activated receptor g coactivator 1a (PGC-1a). Induction of ALAS1 by fasting is lost in liver specific PGC-1a knockout mice. Therefore, PGC-1a is an important link between nutritional status and acute porphyrias.165a
Other Factors Cigarette smoke contains chemicals such as polycyclic aromatic hydrocarbons that induce hepatic CYPs and heme synthesis. A survey of 144 patients with AIP suggested an association between cigarette smoking and repeated attacks of porphyria.166 Attacks may be provoked by intercurrent infections and other illnesses and by major surgery, perhaps by inducing metabolic stress, impairing nutrition, and increasing production of steroid hormones that induce hepatic ALAS1.167
Chapter 72 THE PORPHYRIAS
Neurological Mechanisms The mechanism of neural damage in AIP and related disorders remains poorly understood.1 Formation of heme and important hemoproteins might be compromised in nervous tissue in the acute porphyrias. However, direct evidence for this hypothesis is lacking.168 Vasospasm, perhaps resulting from decreased production of nitrous oxide by the hemoprotein enzyme nitrous oxide synthase, has been suggested to cause cerebral manifestations.169,170 and was suggested as a cause of bowel gangrene in a 31-year-old woman who died with AIP.171 Increased hepatic ALAS1 in this condition may result in depletion of pyridoxine, the cofacter for this enzyme. Neuropathic manifestations of pyridoxine deficiency and acute porphyria have some similarities, and evidence suggesting low pyridoxine status in some acute porphyria patients has been reported.172,173 Another hypothesis is that heme deficiency in liver might predispose to unsaturation of hepatic tryptophan pyrrolase and lead to altered tryptophan delivery to nervous tissue.174,175 The most favored mechanism at present is that ALA or PBG originating from the liver, especially during acute attacks, is neurotoxic. Increased ALA and similar neurologic manifestations in AIP, HCP, VP, ADP, plumbism, and hereditary tyrosinemia type 1 favors a neuropathic role for this precursor or perhaps a derivative. ALA is taken up by most tissues more readily than is PBG, although PBG appears to cross the blood–brain barrier more readily.176 ALA can enter cells readily and then be converted to porphyrins, which in turn may have toxic potential.177 ALA is also structurally similar to g-aminobutyric acid (GABA) and can interact with GABA receptors.178,179 Mice with PBGD deficiency induced by gene targeting display impaired motor function, ataxia, increased levels of ALA in plasma and brain, and decreased heme saturation of liver tryptophan pyrrolase.168,180 Induction of CYPs is impaired in these animals and corrected by heme.181 However, motor neuropathy can develop in these mice even with normal or only slightly increased ALA in plasma and urine, which may support a primary role for heme deficiency in porphyric neuropathy.182 Recently, an allogeneic liver transplant in a woman with severe AIP promptly normalized porphyrin precursor excretion and eliminated recurrent attacks.183 This single case experience helps support the role of the liver and overproduction of heme precursors in causing the neurological manifestations of acute porphyria.
Laboratory Evaluation and Diagnosis Because signs and symptoms are often non-specific in AIP, a high index of suspicion and specific laboratory testing is required for diagnosis. A markedly increased urinary PBG, which can be determined rapidly by a commercial kit,14 indicates very specifically that a patient has one of the three most common acute porphyrias.12 Measuring PBG in serum is preferred when there is coexistent severe renal disease. During acute attacks of AIP, PBG excretion is generally 50–200 mg/day (normal range 0~4 mg/day). Excretion of ALA is usually about half that of PBG (expressed as mg/day). If monitored, ALA and PBG are often increased for some time before an attack, increase further during the attack, and then decrease with recovery – but usually not to normal except after prolonged latency. Dramatic but usually transient decreases occur after intravenous hemin.
ALA and PBG are colorless. Reddish urine in AIP is due to markedly increased porphyrins, and especially uroporphyrins, which can form non-enzymatically from PBG. Urinary porphyrin increases in AIP are less specific diagnostically, but are predominantly type III, which suggests that they are formed enzymatically.184-186 This might occur if excess ALA produced in liver enters cells in other tissues and is there converted to porphyrins via the heme biosynthetic pathway.177 Brownish discoloration of the urine in AIP is attributed to porphobilin, a degradation product of PBG, and dipyrrylmethenes. Total fecal porphyrins and plasma porphyrins are normal or slightly increased in AIP; fecal porphyrins are considerably increased in HCP, and both are increased in VP. Erythrocyte protoporphyrin concentrations are often somewhat increased in patients with clinically expressed AIP, as reported in VP.187 The normal range for erythrocyte PBGD activity is wide and overlaps with the range for AIP. Therefore, while this enzyme activity is approximately half-normal in most (70–80%) patients with AIP,188 this measurement is not definitive for confirming or excluding the diagnosis. As noted above, there are PBGD gene mutations that cause the enzyme to be deficient only in non-erythroid tissues. Moreover, the erythrocyte enzyme is highly age-dependent,189 such that an increase in younger cells in the circulation can raise the activity into the normal range in AIP patients with a concurrent condition such as anemia or hepatic disease.190,191 Because a low PBGD or a disease-causing mutation fails to distinguish between latent and active disease, a clinical diagnosis of AIP as a cause for symptoms continues to rest on the finding of increased PBG in urine or serum. Erythrocyte PBGD activity may be falsely low if processing, storage, or transport of the sample was compromised. Assays that assess multiple heme biosynthetic pathway enzymes simultaneously, as offered by some commercial laboratories, are less reliable than assays that utilize specific substrates. A report of low activities of both erythrocyte ALAD and PBGD, for example, strongly suggests an unreliable result. Confirming a low enzyme value with a specialty laboratory that uses a specific, published method is advisable. Measuring erythrocyte PBGD can be very useful for identifying asymptomatic carriers in family members, if it is already known that this enzyme activity is low in the propositus or another relative. This is not useful in infants less than 4 months of age, because the enzyme can be physiologically increased in erythrocytes. The enzyme deficiency can be detected in other cell types, such as cultured fibroblasts or lymphoblasts, but mutation analysis is now the preferred alternative. Knowledge of the precise mutation in a family enables reliable identification of other gene carriers by DNA testing. Alternatively, restriction fragment length polymorphism studies in informative families may be useful. DNA analysis is more reliable than measuring erythrocyte PBGD activity, as 5–15% of individuals can be misclassified using the enzymatic assay.122,192,193 PBGD deficiency was documented in a fetus by measuring the enzyme activity in amniotic fluid cells,194 and identification of a known mutation in these cells is now feasible using DNA methods.
Treatment of Acute Attacks General and supportive measures. Hospitalization facilitates treatment of severe pain, nausea, and vomiting and the administration of intravenous glucose and hemin. Drugs that may exacerbate
1401
Section XII. Inherited and Pediatric Diseases of the Liver
porphyrias should be discontinued whenever possible. The treatment plan should include monitoring of nutritional status, identification of the precipitating causes of the attack, and observation for neurologic complications and electrolyte imbalances. The vital capacity should be checked, and if impaired the patient should be placed in intensive care.12 Narcotic analgesics are usually required for pain, and chlorpromazine or another phenothiazine for nausea, vomiting, anxiety, and restlessness. Although AIP is listed as a specific treatment indication for chlorpromazine in US labeling, at a suggested dosage of 25–50 mg by mouth (or 25 mg i.m.) t.i.d. or q.i.d. for several weeks, with maintenance therapy required for some patients, considerably smaller dosages are effective for acute attacks, and maintenance therapy is rarely indicated. Phenothiazines do not specifically address the underlying pathophysiology or reduce the overproduction of porphyrin precursors in this disease. Chloral hydrate is commonly recommended for insomnia. Low doses of short-acting benzodiazepines are probably safe if a minor tranquilizer is required. b-Adrenergic blocking agents may be useful during acute attacks to control tachycardia and hypertension, and were considered by some to hasten recovery.195,196 This supposition was based mostly on studies in animal models and cultured hepatocytes. But both D- and 197 L-propranolol reduced experimentally induced porphyria, indicating that non-specific membrane effects rather than b-adrenergic blockade account for its effects. These drugs may be hazardous in patients with hypovolemia and incipient cardiac failure, because in this situation increased catecholamine secretion may be an important compensatory mechanism.198 Carbohydrate loading. Effects of carbohydrate to repress hepatic ALAS1 and reduce porphyrin precursor excretion are weak, compared to those of hemin, and only mild attacks (e.g., mild pain, no paresis or hyponatremia) are treated with carbohydrate loading.12 Glucose polymer solutions are sometimes used, but are not tolerated by patients with ileus, nausea, and vomiting. At least 300 g of intravenous glucose has been recommended for patients hospitalized with attacks of porphyria.91 Amounts up to 500 g daily may be more effective.93 Glucose is usually administered intravenously as a 10% solution, but the large volume may favor the development of hyponatremia. A more complete nutritional regimen, including intravenous vitamins, lipids, and amino acids, is considered, especially if oral or enteral feeding is not possible. Parenteral and enteral nutrition regimens have not been studied in porphyria. Because increasing dietary fat may increase porphyrin precursor excretion in AIP,163 and highprotein diets can induce hepatic CYPs in animals and humans,165 amounts of fat and protein in parenteral solutions for AIP patients should probably not greatly exceed basic requirements. Intravenous hemin. Hemin is considered a specific form of therapy because its beneficial biochemical effects in acute porphyrias are much greater than those of glucose.176,199 After intravenous administration, hemin binds to hemopexin and albumin in plasma, is taken up primarily in hepatocytes, and then enters the regulatory heme pool that represses the synthesis of hepatic ALAS1. This leads to a dramatic reduction in porphyrin precursor excretion. In the past it was often recommended that therapy with hemin be started only after there is no response to intravenous glucose
1402
administered for several days. However, because numerous uncontrolled clinical studies suggest a favorable biochemical and clinical response to hemin (reviewed by Anderson et al.12), most acute attacks of porphyria should be treated promptly with intravenous hemin, without an initial trial of intravenous glucose.12,200 The only randomized, double-blind, placebo-controlled trial of hemin for acute attacks of porphyria enrolled 12 patients and found striking decreases in urinary PBG and trends in clinical benefit, but was underpowered, and hemin therapy was delayed for 2 days.201 This experience contrasts with a larger uncontrolled study that enrolled 22 patients who had 51 acute attacks, in which heme arginate was initiated within 24 h of admission in 37 attacks (73%). All patients responded, including two with paresis, and hospitalization was less than 7 days in 90% of cases.200 Clinical response may be delayed or incomplete when there is advanced neurologic damage, as may occur when treatment is started late. Subacute or chronic symptoms are unlikely to respond.199,202 Hemin is not effective when given by mouth, subcutaneously, or intramuscularly,199 because it is quickly broken down to bilirubin and iron by heme oxygenase and does not reach the liver intact. Hemin preparations. (Hemin is the generic name for all heme preparations used for intravenous administration, including hematin and heme arginate. Hemin is also a chemical term that refers to the oxidized (ferric) form of heme (iron protoporphyrin IX), and is usually isolated as hemin chloride. In alkaline solution, the chloride is replaced by the hydroxyl ion, forming hydroxyheme, or hematin.) Hemin is insoluble at neutral pH and can be prepared as hematin (hydroxyheme) for intravenous infusion at pH 8 or higher. A lyophilized hematin preparation is available in the USA for treatment of porphyria (Panhematin®, Ovation), and was the first drug approved under the Orphan Drug Act. Hematin is unstable in solution, and should be infused promptly.203,204 Degradation products are responsible for phlebitis at the site of infusion and a transient anticoagulant effect.204,205 Phlebitis and loss of venous access are common with repeated courses of hematin. Other side effects are uncommon, and include fever, aching, malaise, hemolysis, anaphylaxis, and circulatory collapse.206,207 Excessive doses of hematin can cause acute renal tubular damage associated with excretion of heme in urine.208 Heme albumin and heme arginate are much more stable than hematin and are seldom associated with phlebitis and coagulopathy.209 Lyophilized hematin can be stabilized as heme albumin by solubilizing it for infusion in 30% human albumin rather than saline.210 This is recommended, especially if hematin is infused into a peripheral vein, but adds intravascular volume-expanding effects and increases the cost of treatment.12 Heme arginate (a stable preparation of heme and arginine) is available in Europe and South Africa but not the USA.209,211 When infused, the heme and arginine moieties dissociate and the heme becomes bound to circulating hemopexin and albumin. Therefore, hematin, heme albumin, and heme arginate do not differ in regard to distribution in vivo or efficacy, although a more stable product is likely to have some efficacy advantage. Dosages and indications for heme therapy. The standard regiment of hemin for treatment of acute porphyric attacks is 3–4 mg/kg daily
Chapter 72 THE PORPHYRIAS
for 4 days.12,200,209 Lower doses have less effect on porphyrin precursor excretion and probably less clinical benefit. An investigational approach is to combine heme therapy with an inhibitor of heme breakdown, such as tin protoporphyrin or tin mesoporphyrin, to prolong the efficacy of the administered heme.212,213 Another potential agent, zinc mesoporphyrin has not yet been studied in humans.214 Treatment with hemin should only be instituted after a diagnosis of acute porphyria has been initially confirmed by a marked increase in urinary PBG (determined most rapidly using a kit)12,14 in the presence of consistent clinical findings. The diagnosis becomes more difficult during and for at least several days after treatment with hemin, because porphyrin precursor levels fall rapidly. It is not essential to confirm an increase in PBG with every recurrent attack; if AIP was previously well-documented biochemically, the recurrence is similar to previous attacks and other causes of the symptoms are excluded clinically. Clearance of drugs that are metabolized by hepatic CYPs is reduced in some patients with acute porphyrias215 and rapidly restored after intravenous hemin.216,217 Hemin also increases drug oxidation rates in normal subjects, and may stabilize or enhance the synthesis of CYP apoenzymes.210,217 PBGD-deficient mice manifest impaired gene transcription and decreased induction of CYP2A5 by phenobarbital, which is restored by exogenous hemin.181 Enzyme replacement. Recombinant human erythrocyte PBGD is currently being studied for the treatment of AIP. Circulating levels of PBG decreased markedly when this drug was administered intravenously to asymptomatic patients, and there were no adverse effects.218 Studies of efficacy in patients with acute attacks are in progress. Early studies of gene therapy using fibroblasts from patients and mice with PBGD deficiency and whole mice with adenoviral or non-viral delivery of plasmids encoding the normal enzyme have provided evidence for correction of the metabolic defect.219,220 Concomitant conditions. Seizures during acute attacks may not recur, especially if they were due to hyponatremia, and therefore not require prolonged treatment. Anticonvulsant drugs are problematic, since almost all have at least some potential for exacerbating acute porphyrias. Clonazepam may be less harmful than phenytoin or barbiturates.221,222 Bromides are safe but seldom preferred. Studies of newer anticonvulsants in chick embryo liver cells suggest that felbamate, lamotrigine, and tiagabine are inducers of heme synthesis and are likely to be harmful in acute porphyrias, whereas gabapentin and vigabatrin are not.223 Clinical experience indicates that gabapentin is safe. Control of hypertension may help prevent chronic renal impairment. AIP patients with end-stage renal disease have undergone successful renal transplantation.108,109 As noted above, liver transplantation was effective in one patient with severe AIP.183 However, further evidence of efficacy is needed before this can be recommended. Other therapies. Cimetidine, a well-known inhibitor of hepatic CYPs, can prevent experimental forms of porphyria induced by agents such as allylisopropylacetamide, which undergoes activation by these enzymes. Although sometimes recommended for human
acute porphyrias,224 the proposed mechanisms are not relevant to inherited porphyrias. 3,5,3¢-Triiodothyronine can partially correct a 5a-reductase defect in patients with AIP, but this approach can lead to symptoms of hyperthyroidism, and clinical benefit has not been reported.225 Treatment with glucocorticoids to suppress adrenal androgen production is probably not beneficial.167 Increased levels of plasma and urinary zinc have been reported during exacerbations of porphyria, possibly due to increased levels of porphyrins, which can complex zinc.226 However, neither ethylenediaminetetraacetic acid (EDTA) treatment to chelate zinc nor administration of zinc has been convincingly beneficial.226–228 EDTA and other chelators such as desferrioxamine can compromise hepatic heme synthesis by binding iron, and have been found to worsen experimental hepatic porphyrias.3 Vitamin B6 status has been of concern because pyridoxal 5-phosphate is a coenzyme for ALAS.173 However, there is no convincing evidence that administration of vitamin B6 or other micronutrients, in the absence of a nutritional deficiency, is commonly beneficial in acute porphyrias. Sorbents have been shown to bind porphyrin precursors in vitro but have been little studied in vivo.229 Antioxidants were not found to be clinically effective.230 Other therapies, such as adenosine monophosphate, vitamin E, folic acid, hemodialysis, hemoperfusion, and plasmapheresis, are not generally regarded as effective.1
Prevention of Acute Attacks Identification of multiple inciting factors is important for all AIP patients and especially for patients who continue to suffer repeated attacks. Drugs for concurrent medical conditions should be reviewed, and potentially harmful agents discontinued whenever possible. Consultation with a dietitian may be useful because dietary indiscretions may not be readily apparent. A well-balanced diet somewhat high in carbohydrate (60–70% of total calories) with sufficient calories to maintain weight should be followed. There is little evidence that additional dietary carbohydrate helps further in preventing attacks. Iron deficiency, if present, should be corrected. Patients who wish to lose excess weight should do so gradually and when they are clinically stable. GnRH analogues can be dramatically effective for preventing repeated attacks that are confined to the luteal phase of the menstrual cycle,137,231,232 but are less effective in patients with attacks partially associated with the cycle. Baseline and continuing gynecological evaluation and bone density measurements are important, especially if treatment is continued for longer than several months. Low-dose estradiol, preferably by the transdermal route, or a biphosphonate may be added to control adverse effects, especially bone loss. Alternatively, changing to a low-dose oral contraceptive can be considered after attacks have been prevented for several months. GnRH analogue treatment may not be needed for longer than 1–3 years, because cyclic attacks do not occur throughout the reproductive period of life in women with porphyria. Oophorectomy is not an acceptable option when prevention of cyclic attacks is the only indication. Exogenous estrogens, progestins, and androgens given alone or in combination have prevented cyclic attacks in some women.91,94,233,234 However, such steroids or their metabolites may induce hepatic ALAS1 and sometimes worsen the disease. GnRH analogues are
1403
Section XII. Inherited and Pediatric Diseases of the Liver
generally preferable for initial use for prevention of cyclic attacks because they are peptides without detrimental effects on heme metabolism. Synthetic steroids with an ethynyl substituent can cause a mechanism-based destruction of hepatic CYPs and should probably be avoided in patients with acute porphyria.235 Danazol is especially contraindicated.236,237 Hemin administered once or twice weekly can prevent frequent, non-cyclic attacks of porphyria in some patients.238 Weekly hemin may also prevent cyclic attacks,239 but a GnRH analogue is generally more practical.
CONGENITAL ERYTHROPOIETIC PORPHYRIA History, Definition, and Prevalence Congenital erythropoietic porphyria (CEP), also termed Günther disease, is a very rare autosomal recessive disease due to a deficiency of UROCoS. About 130 cases had been reported as of 1997,240 with no apparent predominance by race or sex. Mathias Petry, one of the original cases described by Günther, was both a laboratory aid and a source of study material for Hans Fischer, who developed a standard nomenclature for porphyrins. Petry had severe skin lesions and mutilation and died at age 34.61 CEP also occurs in cattle, pigs, cats, and the fox squirrel, with much milder disease manifestations in spite of marked porphyrin overproduction. In fact, all fox squirrels are affected and have no apparent disability.
Clinical Manifestations CEP can cause fetal loss, or present as intrauterine hemolytic anemia and non-immune hydrops fetalis.241 But in most cases, reddish urine or pink staining of diapers by urine or meconium is observed shortly after birth. Marked cutaneous photosensitivity may first appear after phototherapy for hyperbilirubinemia.242 The severe blistering cutaneous features on sun-exposed areas such as the face and hands have been termed hydroa aestivale because they are more severe in summer, when sunlight exposure is greater. Skin friability, hypertrichosis, scarring, thickening, and areas of hypo- and hyperpigmentation are also common. These manifestations are similar to those seen in PCT, but are usually more severe. Recurrent bullae and vesicles, which often become infected, can lead to scarring and distressing deformities, loss of facial features, and damage to the cornea, ears, nails, and fingers. Porphyrins are deposited in dentine and bone during development. Thus, the teeth are characteristically reddish-brown in normal light—an appearance termed erythrodontia – and display reddish fluorescence under longwave ultraviolet light. Bone demineralization243 may be due in part to expansion of the hyperplastic bone marrow. Hemolysis and splenomegaly are features of most, if not all, cases. Bone marrow compensation may be adequate, but severely affected cases are transfusion-dependent. Splenomegaly may contribute to the anemia and cause leukopenia and thrombocytopenia. The latter is sometimes associated with significant bleeding.244,245 There are no neuropathic features in this disease, and no apparent sensitivity to drugs, hormones, and diet. The liver is not directly affected, but may be damaged by iron overload and hepatitis acquired from blood transfusions. In some cases symptoms first appeared in adult life, usually with milder disease that more closely mimics PCT, and without erythrodontia.244–247 At least some late-onset cases were associated with
1404
myeloproliferative disorders, suggesting that expansion of a clone of cells carrying a UROCoS somatic mutation was the underling cause.36
Etiology and Pathogenesis CEP is due to the markedly deficient activity of UROCoS, which is inherited as an autosomal recessive trait.247,248 The enzyme substrate HMB accumulates in bone marrow erythroid cells and spontaneously forms uroporphyrinogen I, which accumulates in its oxidized form, uroporphyrin I. Excess porphyrins accumulate in the bone marrow during hemoglobin synthesis, and are then found in circulating erythrocytes, plasma, and urine. UROCoS catalyzes inversion of pyrrole ring D (the pyrrole ring shown on the right end of the HMB molecule in Figure 72-1) and rapid cyclization of HMB to form uroporphyrinogen III. HMB can cyclize non-enzymatically and less rapidly to form uroporphyrinogen I. PBGD and UROCoS may exist as an enzyme complex, possibly in association with the other two cytosolic enzymes of the pathway.249 This enzyme is also correctly known as uroporphyrinogen synthase, but is still commonly referred to as the cosynthase to avoid confusion with uroporphyrinogen I synthase, the obsolete term for PBGD. The human enzyme purified from human erythrocytes is a monomer with a subunit molecular weight of about 30 kDa. The erythroid and hepatic forms of the enzyme appear to be identical.249 The crystal structure has been determined at 0.185 nm resolution.250 The gene for the enzyme is found on chromosome 10q25.3Æq26.3, is ~34 kb in length, and contains 10 exons.251,252 Alternative promoters generate housekeeping and erythroid transcripts. The housekeeping transcript contains exon 1 (untranslated) fused to exons 2B through 10, while the erythroid transcript contains exon 2A (untranslated), also fused to exons 2B through 10. The housekeeping promoter is upstream of exon 1, whereas the erythroid-specific proximal promoter is upstream of exon 2A and contains erythroid transcription factor-binding sites, including GATA1 and NF-E2. A variety of cosynthase mutations have been identified in patients with this disease (for reviews, see1,253,254), including missense and nonsense mutations, large and small deletions and insertions, splicing defects, and intronic branch point mutations. Most patients were heteroallelic, having inherited a different mutation from each parent. Most mutations have been detected in only one or a few affected families, but a common mutation, C73R, is at a mutational hotspot that is important for enzyme activity, and was found in about 33% of alleles.255 Expression studies of mutations found in CEP have enabled their relative residual activities to be estimated and genotype–phenotype comparisons to be made.256 For example, expression of the C73R allele in E. coli resulted in less than 1% of normal enzyme activity, which is consistent with the severe phenotype (non-immune hydrops fetalis and/or transfusion dependency from birth) in C73R/C73R homozygotes.241 Patients heteroallelic for C73R and another mutation expressing little residual activity also manifested a severe or moderately severe phenotype, whereas those heteroallelic for mutations that expressed more residual activity had milder disease.257 At least four mutations have been identified in the erythroid-specific promoter.258
Chapter 72 THE PORPHYRIAS
UROCoS activity is markedly deficient in all tissues studied in CEP, but is sufficient to produce enough uroporphyrinogen III for heme formation. Increased heme production in the bone marrow is necessary to compensate for hemolytic anemia in CEP, and occurs at the expense of accumulation of HMB. The excess HMB cyclizes non-enzymatically to form uroporphyrinogen I, which is a substrate for UROD but not coproporphyrinogen oxidase (CPO, Figure 722). Therefore, uroporphyrinogen I, coproporphyrinogen I, and the intermediate porphyrinogens accumulate in erythroid cells and are oxidized to the corresponding porphyrins. Erythroid hyperplasia and ineffective erythropoiesis (i.e., substantial destruction of porphyrin-laden erythroid cells leading to heme breakdown in the marrow) are characteristic of CEP. Intravascular hemolysis and increased splenic uptake of circulating erythrocytes result from excess porphyrins, which cause cell damage, perhaps by a phototoxic mechanism. Erythroid hypoplasia sometimes results from deficiencies of nutrients such as folic acid or acute infections.
Laboratory Evaluation and Diagnosis The diagnosis of CEP should be documented by full characterization of porphyrin patterns and identification of the underlying mutation(s). A myeloproliferative disorder should be suspected in late-onset cases. Urinary porphyrin excretion, which can be has high as 50–100 mg daily, consists mostly of uroporphyrin I and coproporphyrin I. ALA and PBG are normal. Fecal porphyrins are markedly increased, with a predominance of coproporphyrin I.242 Circulating erythrocytes in most reported cases of CEP have contained large amounts of uroporphyrin I, and lesser but still excessive amounts of coproporphyrin I. But the porphyrin pattern in erythrocytes can be influenced by rates of erythropoiesis and erythroid maturation. Protoporphyrin is the predominant porphyrin in erythrocytes in bovine CEP, but blood-letting to stimulate erythropoiesis leads to increases in erythrocyte uroporphyrin and coproporphyrin.259,260 Similar observations were made in a patient with CEP and a predominance of protoporphyrin in erythrocytes.261,262 Uroporphyrin I and coproporphyrin I are increased in bone marrow, spleen, and plasma, and to a lesser extent in liver.263,264 CEP is readily distinguished from EPP, in which urine porphyrins are normal and the predominant erythrocyte porphyrin is protoporphyrin. Urinary and fecal porphyrin patterns in HEP resemble PCT. Increased erythrocyte protoporphyrin helps distinguish HEP, as well as very rare homozygous cases of VP and HCP, from CEP. Diagnosis of CEP as the cause of non-immune hydrops or hemolytic anemia in utero can enable intrauterine transfusion and prevent harmful effects of photodynamic therapy for hyperbilirubinemia.242 Prenatal diagnosis is feasible by finding red-brown discoloration and increased porphyrins in amniotic fluid, measuring UROCoS activity in cultured amniotic fluid cells, or identifying UROCoS gene mutations in chorionic villi, cultured amniotic cells, or fetal tissues.241
Treatment Protection of the skin from sunlight, trauma, and infection is highly important in CEP. Sunscreen lotions and b-carotene are sometimes beneficial. Bacterial infections should be treated promptly to avoid cellulitis, bacteremia, scarring, and mutilation. Blood transfusions
sufficient to suppress erythropoiesis significantly can greatly reduce porphyrin levels and photosensitivity.243 Concurrent desferrioxamine administration can reduce iron overload. Hydroxyurea, which can suppress erythropoiesis, was of some additional benefit in one patient.265 Splenectomy may substantially reduce transfusion requirements in some patients.243 Oral charcoal may increase fecal loss of porphyrins,266 but may contribute little in more severe cases.267 Intravenous hemin may be somewhat effective,268 but has not been extensively studied, and seems unlikely to provide longterm benefit. Chloroquine has not been beneficial.269–271 Bone marrow or stem cell transplantation has resulted in marked reduction in porphyrin levels and photosensitivity and long-term survival.272–274 Human UROCoS cDNA has been subcloned into retroviral vectors, which have been used to transduce fibroblasts and lymphoblasts from patients with CEP, resulting in significant levels of enzyme expression. Transduction of hematopoietic progenitor cells and early erythroid cells was also achieved.275–279
PORPHYRIA CUTANEA TARDA History, Definition, and Prevalence PCT, the most common and readily treated human porphyria, is readily recognized by blistering and crusted skin lesions on the backs of the hands. The disease is due to an acquired deficiency of UROD in the liver, although genetic factors sometimes contribute. PCT itself appears to cause liver damage and is also strongly associated with hepatitis C and alcohol use, as well as other susceptibility factors. PCT has also been termed symptomatic porphyria, PCT symptomatica, and idiosyncratic porphyria. Prevalences in the USA and Czechoslovakia have been estimated at about 1 in 25 000 and 1 in 5000, respectively, and the annual incidence in the UK was estimated at 2–5 per million.280 PCT was reported to be prevalent in the Bantus of South Africa in association with iron overload.281 The disease is generally more common in males, possible due to greater alcohol intake, and in women is commonly associated with estrogen use.
Clinical Manifestations Chronic blistering lesions develop most commonly on the dorsa of the hands, and also on the forearms, face, ears, neck, legs, and feet, especially in summer. Fluid-filled vesicles (Figure 72-3) commonly rupture, leading to crusted lesions and denuded areas that heal slowly and may become infected. The sun-exposed skin becomes friable, and minor trauma may precede the formation of bullae or may cause denudation of skin. Small white plaques, termed milia, may precede or follow vesicle formation. Facial hypertrichosis and hyperpigmentation sometimes present even in the absence of vesicles. Thickening, scarring, and calcification of affected areas of skin is sometimes striking and has been termed pseudoscleroderma because it can mimic the cutaneous changes of systemic sclerosis. The skin lesions in PCT are indistinguishable from those in VP and HCP, and resemble those of CEP and HEP but are usually less severe. Skin histopathology includes subepidermal blistering and deposition of periodic acid–Schiff-positive material around blood vessels and fine fibrillar material at the dermoepithelial junction, which may relate to excessive skin fragility. Immunoglobulin G, other immunoglobulins, and complement are also deposited at the
1405
Section XII. Inherited and Pediatric Diseases of the Liver
matosus and other immunological disorders than would have been expected by chance; the basis of this association is not known. A massive outbreak of a syndrome resembling PCT occurred in eastern Turkey in the 1950s after wheat intended for planting, and previously treated with hexachlorobenzene as a fungicide, was consumed by an impoverished population. Hexachlorobenzene was shown subsequently in animals to induce a deficiency of hepatic UROD and patterns of excess porphyrin accumulation and excretion similar to human PCT.289,290 Cases and small outbreaks of PCT have also been reported after exposure to other chemicals, including di- and trichlorophenols and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD, dioxin),291,292 which can also induce hepatic UROD deficiency. In most such cases porphyria improved when the exposure was stopped. However, there have been some apparent instances of delayed onset many years after chemical exposure.293
Etiology and Pathogenesis Figure 72-3. Fluid-filled blister on the dorsum of a finger in a patient with porphyria cutanea tarda.
dermoepithelial junction and around dermal blood vessels. These histologic changes are found in other cutaneous porphyrias and are not diagnostic. Although few patients have advanced liver disease at initial presentation, PCT is almost always associated with non-specific liver abnormalities, especially increased transaminases and g-glutamyltranspeptidase, even in the absence of heavy alcohol intake or hepatitis C. Distorted lobular architecture and cirrhosis are more common in older patients with more long-standing disease and at autopsy.282,283 Liver histopathology is mostly non-specific and includes necrosis, inflammation, increased iron, and increased fat. Red fluorescence of fresh liver tissue, unfixed sections, or even fixed sections if processed properly, is evident on exposure to long-wave ultraviolet light.284 Needle-like inclusions that are fluorescent and birefringent are seen microscopically. By electron microscopy, these inclusions are located in lysosomes; there are also paracrystalline inclusions in mitochondria. Contact of sections with water can cause porphyrins to redistribute artifactually to nuclei.285 Hemosiderosis is usually mild or moderate, but with concurrent familial hemochromatosis may be severe. Risk of developing hepatocellular carcinoma is increased; reported incidences have ranged from 4% to 47%, and was highest in an autopsy series.282,283 These tumors seldom contain large amounts of porphyrins. Mild or moderate erythrocytosis in some patients is not well understood; smoking and consequent chronic lung disease may contribute. An earlier onset of symptoms may be noted in patients with genetic predisposing factors, such as an inherited partial deficiency of UROD or the C282Y/C282Y HFE genotype.286 PCT may be unusually severe and intractable in patients with end-stage renal disease because lack of urinary porphyrin excretion leads to much higher concentrations of porphyrins in plasma and the excess porphyrins are poorly dialyzable.287,288 Because PCT is not a rare disorder it is not surprising that it sometimes occurs with other conditions such as diabetes mellitus. It occurs more frequently in patients with systemic lupus erythe-
1406
UROD, the enzyme that is deficient in PCT and HEP, catalyzes the sequential, clockwise removal of the four carboxyl groups from the acetic acid side chains of uroporphyrinogen III (an octacarboxyl porphyrinogen) to form the four methyl groups of coproporphyrinogen III (a tetracarboxyl porphyrinogen, Figure 72-2). Uroporphyrinogen III as compared with other uroporphyrinogen isomers is the preferred substrate for the enzyme. UROD purified from human erythrocytes is a monomer (42 kDa), but analysis of the crystallized recombinant human enzyme indicates that the enzyme is a dimer with two active site clefts adjacent to each other.294 Unlike most decarboxylases, the enzyme does not require coenzymes or metal cofactors. The UROD gene contains 10 exons, has only one promoter, and is found on chromosome 1p34.295 Two initiation of transcription sites are used with the same frequencies and the gene is transcribed as a single mRNA in all tissues.296 In PCT, hepatic UROD is substantially reduced in a tissuespecific manner, probably to <20% of normal activity.280,297 Enzyme protein in liver, when measured immunochemically, remains at its genetically determined level, suggesting that the enzyme is inhibited.298 Figure 72-4 shows an accepted schema for formation of a specific inhibitor of UROD in the liver in PCT.280,297,299 Iron is essential to this process, and is thought to participate not by directly inhibiting UROD but by promoting the formation of reactive oxygen species, which in turn oxidize uroporphyrinogen to uroporphyrin and to an uncharacterized, non-porphyrin product acting as a UROD inhibitor. CYPs such as CYP1A2 contribute to this process. Induction of hepatic ALAS1 is not a prominent feature in PCT. Although slight to moderate increases in this enzyme have been reported, there have been differences in methodology, and more specific assays for the enzyme have provided results that are closer to normal.280 Alcohol, iron, and estrogens are not potent inducers of ALAS1 in animals, although they may accentuate the effects of other inducers. Alcohol may increase this enzyme slightly in patients with PCT.300 Exacerbation by drugs that induce hepatic ALAS1 and CYPs is much less commonly observed in PCT than in acute porphyrias. Cutaneous manifestations result from porphyrins that circulate from the liver to the skin. Sunlight exposure and generation of reactive oxygen species are followed by complement activation and lysosomal damage.301,302
Chapter 72 THE PORPHYRIAS
Experimental Models
Glycine + succinyl CoA ALAS ␦-Aminolevulinic acid CYP1A2 Fe Uroporphyrinogen
Uroporphyrin
Specific inhibitor UROD
Coproporphyrinogen
Fe2+ Heme Figure 72-4. Formation of a specific inhibitor of uroporphyrinogen decarboxylase in the liver in porphyria cutanea tarda. CoA, coenzyme A; ALAS, daminolevulinic acid synthase; UROD, uroporphyrinogen decarboxylase.
Classification and Enzymatic Defects A current classification distinguishes types 1, 2, and 3 PCT,280,303 with a substantial hepatic UROD deficiency and very similar clinical features in all three. The majority of patients (~80%) are type 1 (sporadic), with normal UROD activity in non-hepatic tissues such as erythrocytes and cultured skin fibroblasts, no demonstrable mutations of the UROD gene or its upstream promoter region,304 and unaffected relatives. Mutations in uncharacterized promoter regions or at other loci are still possible. Although earlier enzyme kinetic data suggested that PCT might be associated with an intrinsically abnormal erythrocyte UROD with altered chemical properties,305,306 evidence for more than one form of the enzyme was not found in other studies.304 In familial (type 2) PCT, a partial (~50%) deficiency of UROD in all tissues (e.g., erythrocytes and cultured skin fibroblasts) due to a UROD mutation is inherited as an autosomal dominant trait. Both UROD activity and immunoreactivity are reduced by approximately 50% in erythrocytes, suggesting that in most instances the mutant allele does not express detectable enzyme protein.298 Because penetrance is low, the trait is seldom associated with overt PCT. The hepatic enzyme activity becomes much lower than 50% when type 2 PCT is active, whereas enzyme protein measured immunologically remains about half-normal.307 UROD mutations identified in type 2 PCT include missense, nonsense, and splice site mutations, several small and large deletions, and small insertions,1,50,280 with only a few identified in more than one family. The crystal structure of UROD suggests that a few of these mutations may be located near the active site cleft whereas most appear to involve regions that have important structural roles.294 Type 3 is rare, and describes the occurrence of PCT in more than one family member, but with normal erythrocyte UROD activity.280,303,308 Because UROD gene mutations or another genetic basis have not been reported in type 3, it is not definitively distinguished from type 1.
Experimental models that involve the administration of halogenated hydrocarbon, loading with iron and ALA, and UROD knockouts closely resemble human PCT and provide insights into the mechanisms for generation of a specific inhibitor of hepatic UROD. Marked decreases in activity of hepatic UROD seem necessary for the development of uroporphyria in rodents, as in human PCT. Hexachlorobenzene and other halogenated hydrocarbons are metabolized to products that bind to cellular macromolecules, and the time course of biochemical changes is complex, and can be greatly influenced by the method of inclusion of the chemical in the diet.309 Induction of CYPs occurs early and precedes the accumulation of porphyrins, inhibition of UROD, and induction of ALAS1. ALAS1 is induced after a deficiency of UROD develops and appears to be a secondary event. The physicochemical properties of hepatic UROD are altered and its activity markedly decreased, but without a decrease in enzyme immunoreactivity.310 Administration of ALA to rodents treated with iron and cyclic hydrocarbons greatly accelerates the development of uroporphyria.280,311,312 In the SWR strain of mice, long-term administration of ALA alone leads to decreased UROD and uroporphyria.312 Such observations indicate that the UROD inhibitor in PCT is derived from a heme pathway intermediate.280,311 Hepatic UROD deficiency in mice is influenced by levels of hepatic iron stores, and also by the inherited inducibility of certain CYPs. Several susceptibility loci for this chemically induced porphyria have been identified in mice by chromosomal linkage analysis, but the nature of these genes is not yet known.297 An important role for CYPs in PCT was evident by studies in rodents and hepatocyte cultures using chemicals that cause or enhance experimental uroporphyria and are inducers of the CYP1A subfamily of these enzymes. CYP1A2 is constitutive and inducible in rodents and can catalyze the oxidation of uroporphyrinogen to uroporphyrin. Measured in vitro as uroporphyrinogen oxidase (URO-OX) activity, this reaction is enhanced by iron, may involve formation of hydroxyl radicals by the flavoprotein enzyme NADPHcytochrome P450 reductase, and, as noted in Figure 72-4, leads to formation of an inhibitor of UROD.280,297,313–315 URO-OX activity is greater for CYP1A2 than for other rodent CYPs, and this enzyme seems essential for the development of uroporphyria in rodents. CYP1A2 knockout mice, in contrast to controls, do not develop uroporphyria when treated with ALA and iron,316 and these mice are also protected from TCDD-induced uroporphyria and liver damage.317 Studies with acetone, a CYP2E1 inducer, in wild-type mice, CYP2E1(–/–) knockouts and CYP1A2(–/–) knockouts suggests that this enzyme can also contribute.318 Hexachlorobenzeneinduced porphyria in rodents is accentuated by estrogens and attenuated by the antioxidant ascorbic acid.319,320 As discussed below, ascorbic acid can prevent uroporphyria in laboratory models.321
Susceptibility Factors in Human Porphyria Cutanea Tarda A multiplicity of susceptibility factors contributes to the development of PCT. These coexist in the individual patient,322 and have been studied, mostly individually, in laboratory models.
1407
Section XII. Inherited and Pediatric Diseases of the Liver
Iron and HFE mutations. A normal or increased amount of iron in the liver appears essential for developing PCT. Serum ferritin levels are usually in the upper part of the normal range or moderately increased, and may relate to the degree of transaminase elevation.323 Liver histology commonly shows increased iron staining.324–326 Phlebotomy to decrease hepatic iron is effective in treating PCT, and iron supplementation can lead to relapse.327,328 Studies of iron absorption and kinetics have been consistent with normal or increased iron stores and absorption.325,326 Prevalence of the C282Y mutation, which is the major cause of hemochromatosis in caucasians, is increased in both sporadic (type 1) and familial (type 2) PCT, and ~10% of patients may be C282Y homozygotes.286,322,329–332 In southern Europe, where the C282Y is less prevalent, the H63D mutation is more commonly associated.333–336 Murine models with disruption of one UROD allele (UROD(+/–)) and either one or two disrupted HFE alleles (HFE(+/–) or HFE(–/–)) provide insight into the roles of these mutations.299,337 UROD(+/–) mice injected with iron-dextran and given drinking water containing ALA for 21 days accumulated porphyrins and manifested hepatic UROD activity reduced to 20% of normal. UROD(+/–)/HFE(–/–) genotype mice developed porphyria by 14 weeks of age even without ALA supplementation, and UROD activity was reduced to 14% of normal. Thus, iron overload alone was sufficient in these mice to reduce UROD activity to ratelimiting levels.299,337 CYP1A2 and other CYPs were not increased.337 Hepatitis C. The prevalence of hepatitis C virus (HCV) infection in PCT varies considerably by geographic location, and in the USA has ranged from 56% to 74%,322,332,338,339 which is similar to earlier reports from southern Europe.340–342 Weaker associations with this infection in northern Europe (<20%)343–345 suggest that HCV and probably PCT are less common in such areas, or that risk factors other than HCV, such as excess alcohol use, are more prevalent. HCV-positive patients with PCT are more commonly males, use alcohol, and smoke cigarettes when compared to patients who are HCV-negative.322 Mechanisms underlying the development of PCT in patients with hepatitis C are not established. HCV in humans and chlorinated aromatic hydrocarbons may release storage iron in hepatocytes in a form that enhances the formation of reactive oxygen species.280,346 Slight increases in hepatic coproporphyrin concentrations are common in hepatitis C, but this is related to hepatic injury and not inhibition of UROD.347,348 HIV. Many reports suggest an association between human immunodeficiency virus (HIV) infection and PCT,349–353 although this is less common than hepatitis C, and the mechanism is not known. It is not surprising that other susceptibility factors often coexist with HIV in PCT.351 For example, in a series of 39 patients, six were HIVinfected, and while four had two or more additional susceptibility factors, HCV was present in only two.322 Alcohol. The long-recognized association between alcohol use and PCT is incompletely understood. Hepatic UROD activity was somewhat reduced by chronic alcohol intake in rats.354 Alcohol may generate active oxygen species and contribute to oxidative damage, mitochondrial injury, depletion of reduced glutathione and other antioxidant defenses, and increased production of endotoxin and activation of Kupffer cells.355,356 Alcohol may also increase iron absorption in normal subjects and PCT patients.280 Ethanol effects
1408
on hepatic iron metabolism may be influenced by genetic factors. For example, it increases hepatic iron content and decreases UROD activity only in HFE(–/–) but not wild-type mice.357 Smoking and cytochrome P450 enzymes. Smoking is commonly associated with alcohol use in PCT.322 Although not extensively studied as a risk factor, it may increase susceptibility for PCT by inducing hepatic CYPs and oxidative stress.322,358 Hepatic CYP levels have been reported to be increased in human PCT.359 These enzymes are thought to be important in oxidizing uroporphyrinogen and generating a UROD inhibitor (Figure 72-4). CYP1A2 in particular accounts for URO-OX activity and contributes to uroporphyria in experimental models. URO-OX activity is considerably lower for microsomes from humans than for rodents, and human CYP1A2 is less active than the rodent isoforms.360 Nonetheless, mouse hepatoma cells expressing either mouse or human CYP1A2 can support uroporphyrin accumulation after administration of ALA, which indicates a potential role for CYP1A2 in the human disease.361 Although assessment of CYP1A2 by caffeine metabolism in PCT patients showed no significant difference compared to controls, even when smokers and non-smokers were analyzed separately,362 constitutive levels of CYP1A2 may still support the development of PCT, and increased activity due to smoking or genetic influences may contribute further in some PCT patients. In addition, other CYPs may be more active than CYP1A2 in uroporphyrinogen oxidation in humans, including CYP2E1 and CYP3A4, which are inducible by alcohol and smoking.360 CYP2E1 also produces more reactive oxygen species than do other CYPs.363 Genetic polymorphisms of CYP1A2 and 1A1 have been implicated in human PCT. A common CYP1A2 intron 1 polymorphism (164AÆC; CYP1A2*1F) reduces enzyme activity, especially in smokers. An increased frequency of the inducible A/A genotype was found in Danish patients with PCT; frequencies of the A/A genotype were 70% in type I, 77% in type II, and 47% in controls.364 Spanish investigators reported similar findings.365 Prevalence of a common m4 CYP1A1 polymorphism that might indicate a higher susceptibility to provoking chemicals was increased in familial PCT in Germany.366 CYP2E1 and other polymorphisms deserve study in this condition. Antioxidants. Ascorbic acid deficiency can contribute to uroporphyria in laboratory models and possibly human PCT.321,367 For example, in an ascorbic acid-requiring strain of rats, massive accumulation of uroporphyrin occurred when exogenous ALA was provided, CYP1A2 was chemically induced, and the animals were ascorbate-deficient. Ascorbate in amounts needed to achieve normal hepatic concentrations prevented uroporphyrin accumulation.368 Plasma ascorbate levels were substantially reduced in 84% of patients with PCT.367 Although ascorbate repletion was not found to influence porphyrin excretion in this disease,369 an immediate response might not be expected given that the amounts of porphyrins that have accumulated in the liver are very large. Low levels of serum carotenoids further suggest that oxidant stress in hepatocytes may play an important role in PCT.370 Estrogens. Use of estrogens in oral contraceptive combinations and for postmenopausal estrogen replacement is commonly associated with PCT (type 1 or 2) in women.322,324,371 In the past, some men
Chapter 72 THE PORPHYRIAS
treated with estrogens for prostate cancer developed the disease. PCT can develop during pregnancy, although it is not established that the risk is increased. Endogenous estrogens may contribute in cirrhotic patients and in Klinefelter syndrome.372 Estrogens may have an additive effect in animals treated with hexachlorobenzene320,373,374 and can generate reactive oxygen species in some experimental systems.375
3 COOH COOH
2
2 1
3 4 A B NH HN NH HN D C 5 8 6 7
HOOC HOOC
1
HOOC
Laboratory Evaluation and Diagnosis
COOH
COOH
Uroporphyrinogen III (Octacarboxyl porphyrinogen III)
Pathogenesis of Porphyrin Patterns Patterns of excess porphyrins are remarkably complex in PCT. Hepatic UROD deficiency leads to accumulation of uroporphyrinogen (isomers I and III), the intermediate substrates hepta-, hexa-, and pentacarboxyl porphyrinogens, and isocoproporphyrinogen. These porphyrinogens (reduced porphyrins) undergo non-enzymatic oxidation to the corresponding porphyrins (uro-, hepta-, hexa-, and pentacarboxyl porphyrins, and isocoproporphyrins), accumulate in large amounts in the liver, and then appear in plasma and urine. Accumulation of isocoproporphyrins occurs via a normally minor pathway, as explained in Figure 72-5. Uroporphyrin and heptacarboxyl porphyrin predominate in urine, with lesser amounts of coproporphyrin and penta- and hexacarboxyl porphyrin. Although urinary porphyrins are increased to a greater extent than fecal porphyrins (relative to normal values), the total amount of porphyrins excreted in feces exceeds that in urine, and total excretion of type III isomers (including isocoproporphyrins, which are mostly derived from the type III series) exceeds that of type I isomers.376 Excess uroporphyrin in PCT is predominantly isomer I; hepta- and hexacarboxyl porphyrin are mostly isomer III; and pentacarboxyl porphyrin and coproporphyrin are approximately equal mixtures of isomers I and III.377,378 Uroporphyrinogen III is the preferred substrate for UROD, and this may favor accumulation of the type I isomer when the enzyme is deficient. Inhibition of UROCoS by ferrous iron379 or by porphyrins and porphyrinogens380 that accumulate in the liver may also favor increased formation of uroporphyrinogen I from HMB. Hepatic porphyrin accumulation is a striking feature of PCT. Porphyrins accumulate mostly as the oxidized porphyrins rather than porphyrinogens, as indicated by the immediate red fluorescence observed in liver tissue. Elder376 has calculated that total porphyrin excretion in urine and feces in PCT is about 6 mmol/day, and if porphyrins accumulate at a similar rate it would take months to reach the porphyrin levels found in patients’ livers (0.22–2.0 mmol/g wet weight). In presymptomatic phases of PCT a smaller proportion of uroporphyrin and heptacarboxyl porphyrin may be excreted, suggesting that in early stages these porphyrins are being preferentially retained in the liver, and later become the predominant porphyrins excreted. Moreover, excess porphyrin excretion in this disease is only a small fraction of the total amount of uroporphyrinogen III formed in the liver.376 Therefore, only a very small increase in synthesis of heme pathway intermediates is required to account for the excess porphyrins excreted. This would explain why there is little or no increase in hepatic ALAS1, and why studies with [14C]ALA suggest that heme is synthesized at a normal rate in PCT.381
COOH
Uroporphyrinogen decarboxylase Heptacarboxyl porphyrinogen III Hexacarboxyl porphyrinogen III
COOH CH H3C
3
N N HH HH N N
H3C
COOH COOH
4 HOOC
COOH
Pentacarboxyl porphyrinogen III Coproporphyrinogen oxidase
Uroporphyrinogen decarboxylase COOH
CH3 H3C
COOH
NH HN NH HN
COOH
H C 3 HOOC
CH3 H C 3
COOH
Autooxidation Intestinal bacterial enzymes
CH
H3C
4
3
HOOC
Isocoproporphyrinogen
COOH
NH HN NH HN
COOH
Coproporphyrinogen III (Tetracarboxyl porphyrinogen III) Uroporphyrinogen decarboxylase
Coproporphyrinogen oxidase CH
Isocoproporphyrins
3
H C 3
N
N HH NH HN
CH
H3C
Coproporphyrinogen oxidase
COOH
3
HOOC
COOH
Harderoporphyrinogen (Tricarboxyl porphyrinogen) Protoporphyrinogen IX
Heme
Figure 72-5. Metabolism of uroporphyrinogen III and the formation of isocoproporphyrins in porphyria cutanea tarda, due to a deficiency of uroporphyrinogen decarboxylase (UROD). The order of decarboxylation of the acetic acid groups occurs in the order indicated by the encircled numbers, beginning with the 8 position on ring D. Dehydroisocoproporphyrinogen is formed via a minor pathway, whereby pentacarboxyl porphyrinogen III undergoes metabolism by coproporphyrinogen oxidase such that the acetic acid group at the 2 position (open arrow) becomes a vinyl group. This pathway becomes significant when hepatic UROD is deficient because coproporphyrinogen III (the preferred substrate for coproporphyrinogen oxidase) is less available, and transformation of dehydroisocoproporphyrinogen to harderoporphyrinogen is impaired. Dehydroisocoproporphyrinogen becomes oxidized to isocoproporphyrin and is excreted in bile, after which bacterial enzymes probably act on the vinyl group (at the 2 position) to give isocoproporphyrin and de-ethylisocoproporphyrin.
Plasma porphyrins are always increased in clinically manifest PCT, and a total plasma porphyrin determination is most useful for
1409
Section XII. Inherited and Pediatric Diseases of the Liver
screening. A normal value rules out PCT and other porphyrias that produce blistering skin lesions. If increased, it is useful to determine the plasma fluorescence emission maximum at neutral pH, because a maximum near 619 nm is characteristic of PCT (as well as CEP and HCP) and, most importantly, excludes VP, which has a distinctly different fluorescence maximum.17,18 Increased urinary porphyrins, with a predominance of uroporphyrin and heptacarboxyl porphyrin, are confirmatory. Urine porphyrins are less useful for initial screening because non-specific increases occur in liver disease and other medical conditions. Urinary ALA may be increased slightly, and PBG is normal. Familial (type 2) and sporadic (type 1) PCT can be distinguished by finding decreased erythrocyte UROD activity (in type 2), or more reliably by finding a disease-related UROD mutation.382 Type 3 is distinguished from type 1 only by the occurrence of PCT in a relative. Biochemical findings in HEP are similar to those in PCT, but with an additional marked increase in erythrocyte zinc protoporphyrin. Pseudoporphyria (also known as pseudo-PCT) presents with photocutaneous lesions that closely resemble PCT, but without significant increases in plasma porphyrins. Like PCT, it can occur in patients with or without end-stage renal disease. Sometimes a photosensitizing drug such as a non-steroidal anti-inflammatory agent is implicated.383
Treatment The diagnosis of PCT must be firmly established, because pseudoporphyria, VP, HCP, and even mild cases of CEP can produce similar cutaneous lesions but are unresponsive to specific treatment. Treatment can usually be started after excluding VP by a plasma porphyrin determination (with analysis of the fluorescence spectrum at neutral pH) while urine and fecal studies are still pending.17,18 Liver imaging and a serum a-fetoprotein determination are advisable to exclude complicating hepatocellular carcinoma and to serve as a baseline for follow-up. Because susceptibility factors influence treatment, patients should be evaluated for alcohol use, smoking, HCV and HIV infections, estrogen use (in women) and HFE mutations,322 and should cease exposures to exogenous agents that have contributed. Drugs that are harmful to patients with acute porphyrias are seldom reported to contribute to PCT, but should be avoided initially as a precaution. Finding low erythrocyte UROD activity or a UROD mutation identifies those with an underlying genetic predisposition, which is useful for genetic counseling but does not alter treatment. Improvement without phlebotomy or low-dose hydroxychloroquine is unpredictable or slow.384 Phlebotomy is considered standard therapy, and if completed correctly almost always achieves a full remission,385 even in children with PCT. When proposed by Ippen in 1961, the aim was to decrease the commonly associated mild or moderately increased hemoglobin, stimulate erythropoiesis, and perhaps channel excess heme pathway intermediates to hemoglobin synthesis.386 However, the intermediates that accumulate in the liver in PCT are oxidized porphyrins, and only reduced porphyrinogens could re-enter the heme biosynthetic pathway in the marrow. The current rationale is that phlebotomy reduces body iron stores and liver iron content, and alleviates the iron-mediated oxidative stress that causes decreased hepatic UROD activity.
1410
About 450 ml of blood can be removed at 1–2-week intervals. In one series an average of 5.4 phlebotomies was required for remission.324 Patients with coexisting hemochromatosis and marked increases in ferritin levels may require many more phlebotomies. Hemoglobin or hematocrit levels should be followed to prevent symptomatic anemia. Hemoglobin should not fall below 10–11 g/dl, and probably should be maintained at a higher level in elderly patients by removing blood less frequently. Treatment is guided by plasma (or serum) ferritin and porphyrin levels.387,388 Treatment is stopped when a target serum ferritin near the lower limit of normal (~15 ng/ml) is achieved. Plasma porphyrin concentrations continue to fall from pretreatment levels (generally 10–25 mg/dl) to near the upper limit of normal (~1 mg/dl) usually after several more weeks (Figure 72-6). This is followed by gradual clearing of skin lesions, sometimes including pseudoscleroderma.385 Liver function abnormalities can also improve.389 Hepatic siderosis, needle-like inclusions, and red fluorescence improve or disappear, but other histological abnormalities may not.282,390 After a remission is obtained, continued phlebotomies are generally not needed, even if ferritin levels later return to normal. However, it is advisable to follow porphyrin levels and reinstitute phlebotomies if porphyrin levels begin to rise. Infusions of desferrioxamine, an iron chelator, may be an alternative treatment approach when phlebotomy is contraindicated.391,392 A low-dose regimen of chloroquine or hydroxychloroquine is also effective.280,393–398 This is most appropriate when phlebotomy is contraindicated or poorly tolerated, but is preferred at some centers. Normal doses of these 4-aminoquinoline antimalarials exacerbate photosensitivity in PCT, markedly increase urinary and plasma porphyrins, induce fever, malaise, nausea, and increase serum transaminases, other liver function tests and ferritin, and even unmask previously unrecognized PCT.399 While these adverse effects are followed by complete remission,400 they are largely avoided by a lowdose regimen (chloroquine 125 mg or hydroxychloroquine 100 mg— one half of a normal tablet—twice weekly) until plasma or urine porphyrins are normalized.396–398 However, some patients may require later treatment with larger doses, or phlebotomy.400 There is at least some risk of retinopathy,401 which may be lower with hydroxychloroquine. Prospective comparative treatment trials are lacking, and in a recent retrospective study low-dose chloroquine was not effective in patients homozygous for the C282Y mutation of the HFE gene,402 which suggests that the degree of excess hepatic iron may influence response to this treatment. Other antimalarials, including 8-aminoquinolines, are not effective. A highly specific mechanism of action of 4-aminoquinolines is indicated by efficacy in PCT but not other porphyrias271 and by studies showing that chloroquine does not mobilize all forms of excess porphyrins from liver and other tissues.403 Chloroquine concentrates in liver, particularly in lysosomes and mitochondria, and may form complexes with and mobilize porphyrins.404,405 Mobilization of hepatic iron may be more important.396,406 Although chloroquine is reported not to reduce hepatic siderosis acutely,407 iron excretion may increase in some patients,396,401 and a low-dose regimen may decrease hepatic hemosiderin.408 Although PCT may improve after treatment of coexisting hepatitis C, in most cases PCT should be treated first and hepatitis C later.409 PCT is usually more symptomatic and can be treated more
Chapter 72 THE PORPHYRIAS Figure 72-6. Effect of phlebotomy in porphyria cutanea tarda. Serum ferritin and total plasma porphyrin concentrations fell during a course of phlebotomies (each arrow indicates removal of 450 ml whole blood) and the appearance of new skin lesions ceased.
14 Porphyria cutanea tarda: Effect of phlebotomy
400
12
Ferritin (ng/ml serum)
300
Porphyrin (g/dl plasma)
10 8
250 200
6
150 4
Porphyrin (g/dl plasma)
Ferritin (ng/ml serum)
350
100 2
50
70 80 90 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0 19 0 20 0 21 0
50 60
30 40
0 0 10 20
–1 0
0 Time (days)
quickly and effectively. Moreover, treatment of PCT may be precluded during interferon-ribavirin treatment by complications such as anemia, and treatment of hepatitis C may be more effective after iron reduction. PCT patients may be resistant to treatment of hepatitis C with interferon-alpha410 or pegylated interferon and ribavirin.411 For example, sustained virological response was obtained with interferon in 27.3% of 44 patients with hepatitis C alone and only 4.5% of 22 patients who had been previously treated for PCT (P < 0.05).410 Prospective studies regarding these observations are needed. Phlebotomy is often contraindicated by anemia when PCT is associated with end-stage renal disease. Erythropoietin administration can correct anemia, mobilize iron, and support phlebotomy in many cases.287,412,413 Until a remission is achieved, high-flux hemodialysis may remove porphyrins from plasma and provide some benefit.414 Response may also occur after renal transplantation,415 due in part to resumption of endogenous erythropoietin production.
HEPATOERYTHROPOIETIC PORPHYRIA History, Definition, and Prevalence
with an inherited deficiency of UROD was associated with dyserythropoiesis.423
Etiology and Pathogenesis The biochemical findings in HEP are due to a profound deficiency of UROD, and resemble those in PCT. In addition, erythrocyte zinc protoporphyrin is substantially increased. This may be explained by accumulation of porphyrinogens in bone marrow erythroid cells during hemoglobin synthesis, followed by their transformation to protoporphyrin after hemoglobin synthesis is complete.55 A variety of UROD mutations have been identified in this disease.50,280 In contrast to most familial PCT mutations, many are CRIM-positive.421,424 Some encode residual UROD activity, as expected, since heme synthesis is essential for life, and the homozygous UROD knockout mouse is a fetal lethal.299 At least one genotype may be associated with predominant excretion of pentacarboxyl porphyrin.425
Laboratory Evaluation and Diagnosis
HEP is the homozygous form of familial (type 2) PCT,55,416,417 and resembles CEP clinically, but with a distinguishable pattern of excess porphyrins, which mostly originate from the liver. Twenty cases were known as of 1994,418 with no particular racial predominance.
HEP can be distinguished from CEP by predominant accumulation and excretion of uroporphyrin, heptacarboxyl porphyrin and isocoproporphyrins, and increased erythrocyte zinc protoporphyrin. In EPP skin manifestations are generally milder, blistering is unusual, increased free rather than zinc protoporphyrin predominates in erythrocytes, and urinary porphyrins are normal.
Clinical Manifestations
Treatment
HEP usually presents in infancy or childhood with blistering skin lesions, hypertrichosis, scarring, and red urine. Sclerodermoid skin changes may be prominent.419 Unusually mild cases have been described.420,421 Concurrent conditions have contributed in some cases. For example, HEP became manifest in a child at age 2 with development of hepatitis A and improved dramatically with its resolution.422 A severe case of erythropoietic porphyria associated
Therapeutic considerations for CEP, especially avoiding sunlight, are applicable in HEP. Oral charcoal was helpful in a severe case associated with dyserythropoiesis.426 Phlebotomy has shown little or no benefit.419 Correction of porphyria in transduced lymphoblastoid cells from patients with this disease was demonstrated after retrovirus-mediated gene transfer.427 Therefore, gene therapy may be applicable in the future.
1411
Section XII. Inherited and Pediatric Diseases of the Liver
HEREDITARY COPROPORPHYRIA History, Definition, and Prevalence HCP is an autosomal dominant hepatic porphyria that results from a partial deficiency of CPO.428 The disease was named by Berger and Goldberg in 1955, although similar case descriptions appeared earlier.429 The prevalence of HCP has not been carefully estimated, but has been reported mostly in the UK, Europe, and North America and is less common than AIP and VP. There is no obvious racial predominance. Harderoporphyria is a biochemically distinguishable variant of HCP.430 Homozygous HCP is rare and presents during childhood.430–433
Clinical Manifestations Symptoms and exacerbating influences are identical to those of AIP except for the occasional occurrence of cutaneous lesions, which resemble PCT.432 The disease is latent before puberty, and symptoms are more common in adult women than men. Although HCP is generally somewhat milder than AIP, severe motor neuropathy and respiratory paralysis can occur, occasionally even in the absence of abdominal symptoms.85 HCP is exacerbated by many of the same factors that cause attacks in AIP, including barbiturates and other drugs, endogenous or exogenous steroid hormones including oral contraceptive steroids, during the luteal phase of the menstrual cycle,432,434 and with fasting. Increased production of 17-oxosteroids and reduced androgen metabolism by the 5a-reductive pathway have been noted in some HCP patients.435 Hepatitis and other superimposed liver diseases may increase porphyrin retention and photosensitivity.428 The risk of hepatocellular carcinoma is increased, as in other acute porphyrias.101 Clinical features of homozygous HCP begin in early childhood. These are generally distinct from those seen in heterozygotes, and may include jaundice, hemolytic anemia, hepatosplenomegaly, and skin photosensitivity.430–433
Etiology and Pathogenesis CPO, the enzyme that is deficient in HCP, is localized in the intermembrane space, where it catalyzes the oxidative decarboxylation of two of the four propionic acid groups of coproporphyrinogen III (on ring A followed by ring B) to form the two vinyl groups at position 2 and 4 of protoporphyrinogen IX (Figure 72-1). This reaction requires molecular oxygen. Harderoporphyrinogen, a tricarboxyl porphyrinogen, is an intermediate in the two-step decarboxylation. (Harderoporphyrin, the corresponding porphyrin, was first isolated from the rodent harderian gland.) A single active site on the enzyme is believed to carry out both decarboxylations.436 Decarboxylation occurs first and more rapidly at the 2 position, and most of the harderoporphyrinogen formed is not released before being further decarboxylated to protoporphyrinogen IX.430 Coproporphyrinogen I is not a substrate for this enzyme and therefore cannot be metabolized to heme. Human CPO expressed in E. coli appears to be a nearly globular homodimer composed of ~39-kDa subunits.437,438 The enzyme contains no metals or prosthetic groups such as flavins or thiol groups in the active site.437,438 The mRNA is probably the same in erythroid and non-erythroid cells. The human CPO gene is on chromosome
1412
3q12.1439 and contains 7 exons, with a single promoter containing elements for housekeeping and erythroid-specific expression.440 Two polyadenylation signals may also play a role in tissue-specific expression.441 In HCP, a partial (~50%) deficiency in CPO activity has been documented in cultured fibroblasts, circulating lymphocytes, and buffy coat preparations,428,432 with more profound deficiencies in homozygous cases.430,431 A variety of CPO gene mutations have been described.50,441,442 For example, molecular studies of 17 unrelated patients in the UK revealed 10 novel and four previously reported CPO mutations, a predominance of missense mutations, and no genotype–phenotype correlations. Mutations known or predicted to cause homozygous disease or harderoporphyria can produce typical HCP in adults when present as single copies.442 Several mutations involve highly conserved amino acids.443 A homozygous patient with related parents was homozygous for a mutation that was shown by E. coli expression studies to encode an enzyme with decreased activity and heat stability.444 Mutations in harderoporphyria may impair substrate binding.433,441,442,445 Harderoporphyrinogen accumulates because the propionate side chain in the 2 position is decarboxylated first and at a faster rate than the side chain in the 4 position, and with these mutations the enzyme prematurely releases harderoporphyrinogen. Although CPO activity has been estimated to be 30-fold greater than that of PBGD, coproporphyrinogen III concentration in liver is probably less than the Km for the enzyme, and therefore it is likely that substrate concentration is a determinant of the reaction rate. Furthermore, coproporphyrinogen appears to be lost more readily from the liver cell than, for example, uroporphyrinogen, and its loss may increase when heme synthesis is stimulated.428 Increased ALA and PBG may be explained by induction of ALAS1 and by the normally relatively low activity of PBGD in liver.428 Limited observations suggest that hepatic ALAS1 is increased during acute attacks, but is normal when the disease is latent and porphyrin precursor excretion is normal.446,447 Coproporphyrin and coproporphyrinogen are readily transported into bile, as well as excreted in urine, and do not appear to accumulate in liver in this disease.448
Laboratory Evaluation and Diagnosis The diagnosis of HCP is readily established in patients with clinically manifest disease by marked increases in coproporphyrin III in urine and feces accompanied by increases in urinary porphyrin precursors.449 Fecal porphyrins are mostly coproporphyrin (isomer III), whereas in VP coproporphyrin III and protoporphyrin are often increased approximately equally. Plasma porphyrins are usually normal in HCP, and increased in VP. Urinary ALA, PBG, and uroporphyrin are increased during acute attacks, but revert to normal more readily than in AIP. The ratio of fecal coproporphyrin III to coproporphyrin I is especially sensitive for detecting latent heterozygotes (especially adults).450 Porphyrin increases may be more severe in homozygous HCP and accompanied by substantial increases in erythrocyte zinc protoporphyrin. Harderoporphyria is characterized by a marked increase in fecal excretion of harderoporphyrin (tricarboxyl porphyrin) as well as coproporphyrin.430 Assays for CPO, a mitochondrial enzyme, require cells such as lymphocytes,11,450 and are not widely available. There is no published
Chapter 72 THE PORPHYRIAS
evidence that the enzyme can be assayed in erythrocytes for diagnosis of HCP, although such an assay has been offered commercially. The diagnosis of HCP is commonly made incorrectly, because urinary coproporphyrin can be increased in many conditions other than porphyria, such as liver disease,10 and minimal increases may not be clinically significant. Increases in urinary or fecal coproporphyrin during drug therapy do not necessarily indicate the presence of a “coproporphyria-like syndrome.”451
Treatment Acute attacks of HCP are treated as in AIP, which includes identifying and then avoiding precipitating factors. Intravenous hemin is useful, as in attacks of the other acute porphyrias.12,452 Cholestyramine may be of some value for photosensitivity occurring with liver dysfunction.453 Phlebotomy and chloroquine are not effective. GnRH analogues can be effective for the prevention of cyclic attacks.232
VARIEGATE PORPHYRIA History, Definition, and Prevalence VP is an autosomal dominant type of hepatic porphyria, and results from a deficiency of protoporphyrinogen oxidase (PPO). This porphyria was first described in 1937, and is especially common in South Africa, where it was described in 1945.61,454,455 The disease is termed variegate because it can present with neurologic manifestations, cutaneous photosensitivity, or both. It has also been known as porphyria variegata, protocoproporphyria, and South African genetic porphyria. Some cases of VP were described as having porphyria cutanea tarda hereditaria before this condition and PCT were clearly distinguished. The obsolete term “mixed porphyria” referred to either HCP or VP. The high prevalence of VP in South African whites (approximately 3 of 1000) is among the best-known examples of genetic drift or “founder effect.” Most cases have been traced to a man or his wife who emigrated from Holland to South Africa and were married in 1688.455 In most countries this porphyria is less commonly recognized than AIP. However, in Finland and perhaps Taiwan VP is about as common as AIP.187,456 In Finland the prevalence is about 1.3 per 100 000.187 VP has been termed the “royal malady” based on speculation that some British royalty may have had acute porphyria (see section on AIP, above).71 Rare cases of homozygous VP are documented.454
Clinical Manifestations Symptoms almost always develop after puberty, but often not until the third decade of life. The onset may occur late in life.457 Neurovisceral symptoms are identical to other types of acute porphyria. Hyponatremia with evidence of sodium depletion or inappropriate ADH secretion is common during acute attacks.92 Skin manifestations, including increased fragility, vesicles, bullae, hyperpigmentation, and hypertrichosis of sun-exposed areas, are more common than in HCP, may occur apart from the neurovisceral symptoms, and are usually of longer duration. At least in South Africa, patients with VP have milder and less frequent attacks than do patients with AIP, and these occur equally in males and females.454,458 In the past this porphyria most commonly presented with neurovisceral symptoms.92 Apparently due to early diagnosis and coun-
seling of asymptomatic patients, acute attacks are now much less common, and skin manifestations are more frequently the initial presentation. For example, in a recent report from the UK and France on 108 patients, 59% presented only with skin lesions, 20% only with an acute neurovisceral attack, and 21% had both symptoms.459 Improved outcomes have been noted in Finland, also presumably due to earlier and improved treatment of acute attacks.94 Like other acute porphyrias, VP may be complicated by hepatocellular carcinoma.101,460,461 The same drug, steroid, and nutritional factors that exacerbate AIP can exacerbate VP.92 Restriction of calories can be detrimental, particularly near the time of menses.462 Oral contraceptives may precipitate cutaneous manifestations of VP. Homozygous patients have inherited a PPO mutation from each parent and have markedly reduced enzyme activity. Severe photosensitivity, neurologic symptoms, convulsions, developmental disturbances, and sometimes growth retardation begin in infancy or childhood.454 Of interest, these patients do not have acute attacks.
Etiology and Pathogenesis An inherited deficiency of PPO is the underlying cause of VP. This enzyme is found in the inner mitochondrial membrane, where it catalyzes the oxidation of protoporphyrinogen IX to protoporphyrin IX by the removal of six hydrogen atoms (Figure 72-1). Non-enzymatic oxidation of protoporphyrinogen to protoporphyrin occurs readily in vitro under aerobic conditions, or under conditions where protoporphyrinogen accumulates first in mitochondria and is exported into the cytosol. The enzyme is highly specific for protoporphyrinogen IX, but the murine enzyme is inhibited by certain tetrapyrroles, such as heme, biliverdin, and bilirubin.454,463 Inhibition of the human enzyme by bilirubin may account for decreased PPO activity in Gilbert’s disease.464 Certain herbicides inhibit PPO and cause protoporphyrin accumulation and phototoxicity in plants.465 These compounds also inhibit the mammalian enzyme. The human PPO gene on chromosome 1q22-q23466,467 encodes a 477-amino-acid polypeptide of ~50.8 kDa.466,468,469 There are 1 non-coding and 12 coding exons and a single PPO transcript of 1.8 kb in a variety of tissues.468,470,471 Recently studied putative transcriptional element binding sequences may allow for erythroidspecific expression.472 The enzyme exists as a homodimer.470 The human enzyme requires molecular oxygen and contains one noncovalently bound FAD per dimer.470,471 Although the enzyme has been localized to the cytosolic side of the inner mitochondrial membrane, the gene does not encode sequences that suggest mitochondrial targeting.469 The crystal structure of PPO from tobacco reveals a loosely associated dimer, and the active site architecture is consistent with specific substrate binding and the unusual six-electron oxidation.473 Membrane-binding domains can be docked on to human FECH, the next enzyme in the pathway, which is embedded in the opposite side of the membrane. This may explain uncoupling of the pathway in VP.473 Approximately half-normal PPO activity, which is found in cultured skin fibroblasts, peripheral leukocytes and lymphocytes, cultured lymphocytes, and liver from patients with VP, is transmitted as an autosomal dominant trait.474 The enzyme is more markedly
1413
Section XII. Inherited and Pediatric Diseases of the Liver
deficient in rare cases of homozygous VP, with approximately halfnormal enzyme activity in parents. A variety of PPO gene mutations have been reported in VP.50,454,459 The missense mutation R59W is prevalent in South Africa, where it may be present in 20 000 individuals.475 Other less common mutations occur in South Africa among both caucasians and blacks with this disease.476 In 108 unrelated English and French families, 60 novel and six previously reported mutations were identified; 47 mutations were each detected in single, unrelated families, and 14 in two or three families. Several accounted for 26% of the mutations, and two were recurrent and did not originate from a common founder. No genotype–phenotype correlations were identified.459 Multiple mutations were also found among VP patients in Chile, where one common mutation represented a hotspot mutation and another was a founder mutation.477,478 Mutations in homozygous cases of VP are often different from those found in heterozygous disease, and are more likely to encode enzyme proteins with residual activity.454 Most heterozygotes for PPO deficiency (~75%) remain asymptomatic. Attacks can develop in the presence of additional influences such as drugs, steroids, and nutritional alterations, many of which increase hepatic ALAS1. Increased ALA and PBG during acute attacks may be explained, as in HCP, by induction of ALAS1 and by the normally relatively low activity of PBGD in liver.428 Furthermore, PBGD is inhibited by protoporphyrinogen (but not protoporphyrin), which may explain a modest reduction of this enzyme in VP.479 When protoporphyrinogen IX accumulates in patients with VP, it undergoes auto-oxidation to protoporphyrin IX. A close functional association between PPO in the inner mitochondrial membrane and CPO in the intermembrane space may relate to the excess excretion of both protoporphyrin and coproporphyrin in this disease. Liver porphyrin content is not increased. Accumulation of protoporphyrinogen may lead to formation of porphyrin–peptide conjugates, which may be present in plasma.
Laboratory Evaluation and Diagnosis Urinary ALA, PBG, and uroporphyrin are increased during acute attacks but may be normal or only slightly increased during remission. Increases of porphyrins in plasma, protoporphyrin in feces, and coproporphyrin III in urine and feces are more persistent between attacks and are useful for distinguishing VP from other acute porphyrias. The increased plasma porphyrins in VP consist in part of a dicarboxyl porphyrin tightly bound to plasma proteins. As a result, the neutral fluorescence emission spectrum of plasma porphyrins at neutral pH in VP is characteristic and can rapidly and reliably distinguish VP from other types of porphyria. The emission maximum occurs at or near 626 nm in VP, 619 nm in PCT, CEP, HCP, and (sometimes) AIP, and 634 nm in EPP.17–19 This fluorometric method is more effective than examination of fecal porphyrins for detecting asymptomatic VP.480,481 Fecal porphyrin analysis is also very useful, but timed fecal collections are usually not feasible or meaningful, and results can be confounded by dietary and other substances, such as porphyrin-containing yeast tablets.482 Increased biliary porphyrins may be useful in detecting asymptomatic VP,483 but bile samples are more difficult to obtain. PPO assays are sensitive for identifying asymptomatic carriers in affected families, but require cells that contain mitochondria, such as lymphocytes, and are not widely
1414
available.484 Erythrocyte zinc protoporphyrin levels are markedly increased in homozygous VP, and may be moderately increased in heterozygous cases.187 An increase in fecal “pseudo-pentacarboxyl porphyrin,” a dicarboxyl porphyrin derived from protoporphyrin, is sometimes noted by HPLC in VP and is diagnostically useful.454 The X porphyrin fraction (containing a heterogeneous group of porphyrin–peptide conjugates that are ether-acetic acid-insoluble and extracted from feces by urea-Triton) is more increased in this disease than in other porphyrias, but is not measured by common analytic methods.485 Fecal levels of meso- and deuteroporphyrins derived in the gut from protoporphyrin are also increased.
Treatment Treatment measures that are effective for attacks of AIP, including identifying and avoiding exacerbating factors, are effective in VP.12,458 Hemin is beneficial for acute attacks, but not for cutaneous symptoms.486 Clothing, gloves, a broad-brimmed hat, and opaque sunscreen preparations to protect the skin from sunlight are useful. Increasing the amount of skin pigment by exposure to short-wavelength ultraviolet light, which does not excite porphyrins, may provide some protection. Phlebotomy and chloroquine are not effective. Other therapies, such as propranolol, D-penicillamine, hemodialysis, alkalization of urine, and b-carotene are of little or no benefit. Surprisingly, oral activated charcoal was reported to increase porphyrin levels and worsen skin manifestations.487 Improved treatment, earlier diagnosis, and detection of latent cases have greatly improved the outlook for South African patients with this disease.92,458 Cyclic acute attacks in women can be prevented with a GnRH analogue, as in AIP. A diagnosis of VP or any other acute porphyria should not lead to difficulty obtaining insurance, because the prognosis is usually good once the diagnosis is established. Because the disorder usually remains latent, abortion should seldom be considered.455
ERYTHROPOIETIC PROTOPORPHYRIA History, Definition, and Prevalence EPP is due to a partial deficiency of FECH, the last enzyme in the heme biosynthetic pathway. In most lineages EPP is an autosomal dominant condition, with considerable variation in penetrance. This is the third most common porphyria, but because the cutaneous features are usually relatively mild and urinary porphyrins are not increased, it was not clearly described until 1961.488 EPP is described mostly in caucasians, but occurs in other races as well, including blacks.489 This disease has also been termed erythrohepatic protoporphyria and protoporphyria. Autosomal recessive EPP occurs in cattle, and two murine models have been developed.490,491
Clinical Manifestations Cutaneous photosensitivity, which begins in childhood, and is generally worse in the spring and summer, is manifested by itching, painful erythema, and swelling within minutes of sun exposure, and may resemble angioneurotic edema. Petechiae and purpuric lesions sometimes occur. Vesicles and bullae are absent or sparse; in one series they were reported in only 10% of cases.492 Skin lichenification, leathery pseudovesicles, and nail changes can be pronounced
Chapter 72 THE PORPHYRIAS
and labial grooving is sometimes noted.493 Pigment changes and severe scarring are unusual. Notably absent are deformities of facial features and digits, increased skin fragility, hirsutism, and fluorescent teeth. Neurological symptoms are also absent, except in some patients with severe hepatic complications.494 The skin of patients with EPP is maximally sensitive to light in the 400-nm range, which corresponds to the so-called Soret band (the narrow peak absorption maximum that is characteristic for protoporphyrin and other porphyrins).488 When porphyrins absorb light they enter an excited energy state and then release energy as fluorescence and by the formation of singlet oxygen and other oxygen species that can produce tissue damage. This may be accompanied by lipid peroxidation, oxidation of amino acids, and cross-linking of proteins in cell membranes.495–497 Light-induced damage to capillary endothelial cells in the upper dermis has been demonstrated immediately after light exposure in this disease. Repeated acute damage and repair in the basement membrane area lead to thickening of the vessel walls and perivascular deposits from accumulation of serum components. Photoactivation of the complement system and release of histamine, kinins, and chemotactic factors may mediate skin damage.497 Histologic changes are predominantly in the upper dermis and may include deposition of amorphous material containing immunoglobulin, complement components, glycoproteins, acid glycosaminoglycans, and lipids around blood vessels.497,498 Amorphous deposits are more prominent than in VP and PCT.499 Photosensitivity and the level of protoporphyrin in erythrocytes, plasma, and feces remain remarkably stable for many years in most patients.500 There are no known precipitating factors, such as those associated with the hepatic porphyrias, and little or no anemia. Somewhat lower levels of erythrocyte protoporphyrin and increased sunlight tolerance during pregnancy, and mild hypertriglyceridemia in some patients, are unexplained.501–503 Although liver function and liver protoporphyrin content usually remain normal in EPP, some patients develop distinctive hepatobiliary complications. Biliary stones that contain protoporphyrin are sometimes symptomatic and require cholecystectomy.504 Liver disease occurs in 1–2% of EPP patients and is often preceded by increasing levels of erythrocyte and plasma protoporphyrin, abnormal liver function tests, marked deposition of protoporphyrin in liver cells and bile canaliculi, and increased photosensitivity. This may be chronic or progress rapidly to death from liver failure,505 and is sometimes the major presenting feature of EPP.506 Upper abdominal pain may suggest biliary obstruction, and unnecessary laparotomy to exclude this possibility can be detrimental.505 Concurrent viral hepatitis, alcohol intake, iron deficiency, fasting, or oral contraceptive steroids, which may impair liver function or the metabolism of protoporphyrin to heme, have appeared to contribute in some patients.507,508 An enterohepatic circulation of protoporphyrin may favor its retention in the liver, especially when liver function begins to be impaired. Liver transplantation or other surgery in patients with protoporphyric liver disease can be complicated by marked photosensitivity with extensive burns of the skin and peritoneum and photodamage of circulating erythrocytes due to artificial lights, such as operatingroom lights.509 End-stage protoporphyric liver disease may even be accompanied by a severe motor neuropathy that resembles that seen in the acute porphyrias.494
Etiology and Pathogenesis FECH, the enzyme that is deficient in EPP, is also known as heme synthetase or protoheme ferrolyase. It catalyzes the insertion of ferrous iron (Fe2+) into protoporphyrin IX, which is the final step in heme synthesis (Figure 72-2). FECH is specific for the reduced form of iron, but can utilize other metals such as Zn2+ and Co2+ and other dicarboxyl porphyrins. It is sensitive to inhibition by lead and other metals, and zinc content of mitochondrial membranes may influence its activity.510 N-methyl protoporphyrin and certain other alkylated porphyrins are potent FECH inhibitors. These are derived from heme during mechanism-based “suicidal” reactions catalyzed by CYPs, and can cause marked protoporphyrin accumulation in laboratory animals.156 These N-alkyl porphyrins can be detected using human recombinant FECH.511 The human enzyme is a dimer, and the homodimer crystal structure revealed a [2Fe-2S] cluster,512 which may have a structural role, possibly bridging homodimers.513,514 FECH may be associated with complex I of the mitochondrial electron transport chain, and ferrous ion may be produced upon NADH oxidation.515 The human FECH gene is found on chromosome 18q21.3, and a pseudogene on chromosome 3.516 The functional gene has a single promoter sequence and contains 11 exons. FECH, like most other mitochondrial proteins, is synthesized in the cytoplasm as a precursor that is larger (423 amino acids, including a leader sequence of 54 amino acids) than the mature enzyme (369 amino acids), and after translocation and processing the enzyme is found in the mitochondrial inner membrane where its active site faces the mitochondrial matrix. Two mRNAs of ~1.6 and ~2.5 kb were found by Northern blot analysis, due to the use of two alternative polyadenylation signals.517 The larger transcript was found to be more abundant in murine erythroid cells.518 Such evidence for erythroid-specific mechanisms for regulation of FECH518 suggests a rate-limiting role for this enzyme, which is consistent with protoporphyrin accumulation almost entirely in erythroid cells in EPP. A variety of FECH mutations in EPP include missense, nonsense, and splicing mutations as well as small and large deletions and an insertion.50 Splicing mutations that resulted in exon skipping were predominant, perhaps because many mutations were detected in reverse transcriptase polymerase chain reaction products in which deletions are easily recognizable.50,519 But in one study, genomic DNA from 29 unrelated Swiss and French patients with protoporphyria was systematically screened by denaturing gradient gel electrophoresis and direct sequencing, and only four splice site mutations were identified.520 Among 14 Swiss families in this study, four common mutations accounted for 86% of those identified. Three of six apparently unrelated protoporphyria patients in Northern Ireland had the same mutation.521 An analysis of haplotypes flanking the FECH gene showed that some mutations shared by unrelated American families with European forebears were probably hotspot mutations, whereas others with shared FECH mutations and haplotypes represented wide dispersion of ancestral mutant alleles.522 Symptomatic EPP patients have FECH activity as low as 15–25% of normal, which is much less than the 50% expected for a strictly autosomal dominant condition.520 It was established recently that in most families such low enzyme activities in EPP patients is explained by the combined presence of a disabling FECH mutation and a
1415
Section XII. Inherited and Pediatric Diseases of the Liver
common polymorphism affecting the other FECH allele.523,524 This intronic single nucleotide polymorphism, IVS3-48T/C, leads to an aberrantly spliced mRNA that is degraded by a nonsense-mediated decay mechanism, which decreases the steady-state level of FECH mRNA.523 The polymorphism by itself does not cause disease, even when homozygous. The IVS3-48T/C polymorphism is common in Caucasians and East Asians, but is rare in Africans. In a few families, two FECH mutations and a pattern of autosomal recessive inheritance has been found.519,525,526 EPP with autosomal recessive inheritance occurs in cattle and in mice.490,491 Therefore, it is now established that inheritance of two alleles associated with reduced FECH activity, one of which is most commonly a low expression allele that produces a structurally normal enzyme, is necessary for development of protoporphyria. Genetic background modulates the disease in mice, suggesting that modifying genes could influence the severity of EPP.490 Other molecular explanations for the 10–25% of normal FECH activity have been explored and may sometimes apply. Because the enzyme is a dimer, it might be possible for a defective subunit to interact with a normal subunit and act in a “dominant-negative” fashion to render the dimer non-functional.527,528 If only dimers composed of two normal subunits are functional, the predicted enzyme activity would then be ~25% of normal. Some heterodimers containing normal and certain mutated subunits may initially retain normal activity but be unstable.529 Patients with protoporphyric liver failure, which represents a severe phenotype, have “null mutations”—splice site mutations, insertions, deletions, or nonsense mutations, which result in a truncated enzyme protein, rather than missense mutations, which may retain some enzyme activity,520,530,531 as well as the IVS3-48T/C polymorphism in the non-mutant allele.524,532 Patients with two mutant alleles and autosomal recessive inheritance may also develop liver disease.519 Reticulocytes account for almost all the fluorescence in the bone marrow in EPP, and are thought to be the primary source of the excess protoporphyrin.533–535 Circulating erythrocytes are a limited source, because these cells have ceased making heme and hemoglobin. Most of the excess erythrocyte protoporphyrin is found in a small percentage of cells, and the rate of protoporphyrin leakage from these cells is proportional to their protoporphyrin content.536 The liver has been considered a possible source,533–535,537,538 but quantitation of its contribution relative to that of the erythron has not been possible.537 Ferrochelatase is deficient in all tissues in EPP, including bone marrow, liver, cultured fibroblasts, and blood leukocytes.491,533,539,540 Recent tissue transplantation studies in mice with EPP suggest that skin photosensitivity and liver damage only occur when FECH is deficient in these tissues.541 Liver damage in EPP is attributed to excess protoporphyrin, which is insoluble and can form crystalline structures and impair mitochondrial functions in liver cells, and decrease hepatic bile formation and flow.505,542 Hepatocellular uptake of protoporphyrin is facilitated by fatty acid-binding protein (Z protein).543 DNA microarray analysis of explanted liver of patients who underwent transplantation showed significant changes in expression of several genes involved in wound-healing, organic anion transport, and oxidative stress.524 The bone marrow is probably the major source of pro-
1416
toporphyrin even in EPP patients with hepatic failure. In one patient, for example, splenectomy to ameliorate hemolysis and reduce protoporphyrin production led to improved liver function.544 Late-onset EPP can occur in patients with myelodysplastic syndromes and expansion of a clone of hematopoietic cells with deletion of one FECH allele.545,546 One patient with a myeloproliferative disorder later developed unusually severe EPP due to clonal expansion of an erythropoietic cell line with a FECH deletion and the IVS3-48C/T polymorphism causing low expression of the remaining FECH allele, and died of liver disease.547
Laboratory Evaluation and Diagnosis A diagnosis of EPP is primarily confirmed by finding a substantially elevated concentration of erythrocyte protoporphyrin, which is predominantly free and not complexed with zinc. Increased free protoporphyrin in EPP indicates that formation of zinc protoporphyrin is dependent on FECH activity in vivo. Protoporphyrin is also increased in bone marrow, plasma, bile, and feces. Urinary porphyrins and porphyrin precursors are normal. Erythrocyte zinc protoporphyrin concentration is increased in many conditions other than EPP, such as some homozygous porphyrias, iron deficiency, lead poisoning, anemia of chronic disease,548 hemolytic conditions,189 and many other erythrocytic disorders. Many assays for erythrocyte protoporphyrin or free erythrocyte protoporphyrin measure both zinc and free protoporphyrin, and assays that specifically measure true metal-free erythrocyte protoporphyrin are less widely available. Therefore, reports of increased erythrocyte protoporphyrin must be interpreted with care. Protoporphyrin in erythrocytes of EPP patients declines relatively rapidly with erythrocyte aging. Free protoporphyrin is lost from erythrocytes more rapidly than zinc protoporphyrin, which may persist in the red cell as long as it circulates.534,535 Moreover, ultraviolet light may cause free protoporphyrin to photodamage its hemoglobinbinding site and thus be released from the red cell even without disruption of the cell membrane, and then bind to albumin in plasma.549,550 The plasma porphyrin concentration is almost always at least somewhat increased in EPP. However, the increase may be less than in other cutaneous porphyrias, in part because plasma porphyrins in EPP are particularly subject to photodegradation during sample processing unless great care is taken to avoid exposure of the sample to sunlight or fluorescent lighting.551 There is little evidence for impaired erythropoiesis or abnormal iron metabolism in EPP.325,492,533 In most uncomplicated cases, hemolysis is absent or very mild. However, mild anemia with hypochromia and microcytosis or mild anemia with reticulocytosis is sometimes noted,132,492,533, 552 and depletion of iron stores may be relatively common, even in the absence of anemia.325 On the other hand, iron accumulation in erythroblasts and ring sideroblasts have been noted in bone marrow, suggesting that FECH deficiency impairs erythroid heme synthesis in some patients.553 Life-threatening hepatic complications of EPP may be preceded, usually in the short term, by increased photosensitivity and by increasing erythrocyte and plasma protoporphyrin levels and abnormal liver function tests. Liver biopsy reveals marked deposition of protoporphyrin in liver cells and bile canaliculi.505,554,555
Chapter 72 THE PORPHYRIAS
Assays for FECH require cells containing mitochondria and are not widely available. DNA studies are increasingly important, with the aim of identifying a disabling FECH mutation and the IVS348T/C polymorphism in most EPP patients.
tion of hematopoietic stem cells with erythroid-specific expression of the therapeutic FECH gene, leading to a progressive increase of normal erythrocytes and correction of photosensitivity.573 This suggests a potential future role for human gene therapy in protoporphyria.
Treatment Oral administration of b-carotene, which was first reported to be useful for treating EPP in 1970,556 leads to clinical improvement and greater tolerance of graded exposures to long-wave ultraviolet light, as substantiated in large series of patients.557,558 Improved tolerance to sunlight is usually maximal 1–3 months after initiation of treatment. Doses of 120–180 mg daily in adults are usually required to maintain serum carotene levels in the recommended range of 600–800 mg/dl, but doses up to 300 mg/day may be needed. Suntanning resulting from better tolerance of sunlight may lead to further protection. Solatene (Tishcon) is the recommended product and was developed specifically for treating this disease. No side effects other than a mild, dose-related skin discoloration due to carotenemia have been noted.557 The mechanism of action may involve quenching of singlet oxygen or free radicals. Oral cysteine may also quench excited oxygen species and increase tolerance to sunlight in EPP.559 Darkening of the skin with topical dihydroxyacetone and lawsone (napthoquinone) may partially block exposure of the dermis to light and be of some benefit in EPP.492,560 Narrow-band ultraviolet B phototherapy to increase skin melanin content may also improve symptoms.561 Cholestyramine, which may interrupt the enterohepatic circulation of protoporphyrin and promote its fecal excretion, has been reported to reduce liver protoporphyrin and improve cutaneous symptoms in some patients.505,562 Oral bile acid supplementation was beneficial in some animal models,563 and ursodeoxycholic acid may be of some value in the early stages of protoporphyric liver disease.564 Splenectomy may be beneficial when EPP is complicated by hemolysis and splenomegaly. Caloric restriction and drugs or hormone preparations that impair hepatic excretory function should be avoided, and iron deficiency should be corrected if present.565,566 Hepatic complications of EPP may resolve spontaneously, especially if another reversible cause of liver dysfunction, such as viral hepatitis or alcohol, is contributing.507,508 Otherwise, treatment must be individualized and results are unpredictable. Cholestyramine and other porphyrin absorbents such as activated charcoal should be considered in this situation. Other therapeutic options include ursodeoxycholic acid,564 red blood cell transfusions, exchange transfusion, plasma exchange, and intravenous hematin to suppress erythroid and hepatic protoporphyrin production,567 as well as liver transplantation.554 Neuropathy resembling that seen in acute porphyrias has developed in some EPP patients with liver disease after transfusion568 or liver transplantation,494,569 and is sometimes reversible.570 Bone marrow transplantation was beneficial in a murine model of protoporphyria in which homozygotes have severe FECH deficiency and develop liver disease.571 An EPP patient who underwent bone marrow transplantation for acute myelogenous leukemia experienced remission of protoporphyria with marked decreases in porphyrin levels.572 Gene therapy strategies have also been studied in the murine model. Most recently, a dual gene therapy approach allowed selec-
DUAL PORPHYRIA Patients with porphyria and deficiencies of more than one enzyme of the heme biosynthetic pathway are classified as having dual porphyria. For example, kindreds with individuals having both VP and familial PCT have been described.574 Patients with deficiencies of both PBGD and UROD may develop symptoms of AIP, PCT, or both.575 An infant with severe porphyria was found to have inherited CPO deficiency from one parent and UROCoS deficiency from both parents.576 Coexistence of UROCoS and UROD deficiencies were described in a patient with an erythropoietic porphyria.577 A family with deficiencies of both PBGD and CPO has also been described.578 Dual enzyme defects were recently confirmed at the molecular level in a man with acute porphyria, an unusual pattern of excess porphyrin precursors and porphyrins, and heterozygous mutations of both ALAD and CPO. 579
PORPHYRIA DUE TO TUMORS Porphyria is very rarely a neoplastic condition. Hepatocellular tumors have been reported on a few occasions to contain excess porphyrins; the porphyrin excretion patterns in these cases have not been uniform, and heme biosynthetic pathway enzymes were not measured.580 Hepatocellular carcinomas complicating PCT and acute hepatic porphyrias have usually not contained large amounts of porphyrins. Erythropoietic porphyrias can develop late in life in the presence of myelodysplastic or myeloproliferative syndromes, probably due to clonal expansion of erythroid cells containing a specific enzyme deficiency.36,547
Acknowledgments Preparation of this chapter was supported in part by grants from the US Food and Drug Administration Office of Orphan Product Development (FD-R-001459), the American Porphyria Foundation, and the National Center for Research Resources, National Institutes of Health (MO1 RR-00073).
REFERENCES 1. Anderson KE, Sassa S, Bishop DF, Desnick RJ. Disorders of heme biosynthesis: X-linked sideroblastic anemias and the porphyrias. In: Scriver CR, Beaudet AL, Sly WS, et al., eds. The metabolic and molecular basis of inherited disease, 8th edn, vol. II. New York: McGraw-Hill; 2001:2991–3062. 2. Sassa S, Kappas A. Genetic, metabolic, and biochemical aspects of the porphyrias. In: Harris H, Hirschhorn K, eds. Advances in human genetics. New York: Plenum Publication; 1981:121–231. 3. Anderson KE, Freddara U, Kappas A. Induction of hepatic cytochrome P-450 by natural steroids: relationships to the induction of d-aminolevulinate synthase and porphyrin accumulation in the avian embryo. Arch Biochem Biophys 1982; 217:597–608. 4. Granick S, Sinclair P, Sassa S, Grieninger G. Effects by heme, insulin, and serum albumin on heme and protein synthesis in
1417
Section XII. Inherited and Pediatric Diseases of the Liver
5.
6.
7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18. 19.
20.
21.
22.
23.
1418
chick embryo liver cells cultured in a chemically defined medium, and a spectrofluorometric assay for porphyrin composition. J Biol Chem 1975; 250:9215–9225. Sassa S, Bradlow HL, Kappas A. Steroid induction of daminolevulinic acid synthase and porphyrins in liver. Structure–activity studies on the permissive effects of hormones on the induction process. J Biol Chem 1979; 254:10011–10020. Lathrop JT, Timko MP. Regulation by heme of mitochondrial protein transport through a conserved amino acid motif. Science 1993; 259:522–525. Munakata H, Sun JY, Yoshida K, Nakatani T, Honda E, Hayakawa S, Furuyama K, Hayashi N: Role of the heme regulatory motif in the heme-mediated inhibition of mitochondrial import of 5-aminolevulinate synthase. J Biochem (Tokyo) 2004;136:233-8. Wolfson SJ, Bartczak A, Bloomer JR. Effect of endogenous heme generation of d-aminolevulinic acid synthase activity in rat liver mitochondria. J Biol Chem 1979; 254:3543–3546. Ponka P, Schulman HM. Regulation of heme biosynthesis: distinct regulatory features in erythroid cells. Stem Cells 1993; 11:24–35. Gibson PR, Grant J, Cronin V, et al. Effect of hepatobiliary disease, chronic hepatitis C and hepatitis B virus infections and interferon-alpha on porphyrin profiles in plasma, urine and faeces. J Gastroenterol Hepatol 2000; 15:192–201. Elder GH, Smith SG, Jane Smyth S. Laboratory investigation of the porphyrias. Ann Clin Biochem 1990; 27:395–412. Anderson KE, Bloomer JR, Bonkovsky HL, et al. Recommendations for the diagnosis and treatment of the acute porphyrias. Ann Intern Med 2005; 142:439–450. Mauzerall D, Granick S. The occurrence and determination of d-aminolevulinic acid and porphobilinogen in urine. J Biol Chem 1956; 219:435–446. Deacon AC, Peters TJ. Identification of acute porphyria: evaluation of a commercial screening test for urinary porphobilinogen. Ann Clin Biochem 1998; 35:726–732. Watson CJ, Taddeini L, Bossenmaier I. Present status of the Ehrlich aldehyde reaction for urinary porphobilinogen. JAMA 1964; 190:501–504. Lamon J, With TK, Redeker AG. The Hoesch test: bedside screening for urinary porphobilinogen in patients with suspected porphyria. Clin Chem 1974; 11:1438–1440. Poh-Fitzpatrick MB, Lamola AA. Direct spectrophotometry of diluted erythrocytes and plasma: a rapid diagnostic method in primary and secondary porphyrinemias. J Lab Clin Med 1976; 87:362–370. Poh-Fitzpatrick MB. A plasma porphyrin fluorescence marker for variegate porphyria. Arch Dermatol 1980; 116:543–547. Longas MO, Poh-Fitzpatrick MB. A tightly bound protein–porphyrin complex isolated from the plasma of a patient with variegate porphyria. Clin Chim Acta 1982; 118:219–228. De Salamanca RE, Sepulveda P, Moran MJ, et al. Clinical utility of fluorometric scanning of plasma porphyrins for the diagnosis and typing of porphyrias. Clin Exp Dermatol 1993; 18:128–130. Blake D, Ratnaike S. Plasma – all you need for the diagnosis of porphyria with current dermatological symptoms. In: Blake D, McManus J, Ratnaike S, et al., eds. Porphyrins, porphyrias and photodynamic therapy (conference proceedings). Melbourne, Australia: 1993. Poh-Fitzpatrick MB, Sosin AE, Bemis J. Porphyrin levels in plasma and erythrocytes of chronic hemodialysis patients. J Am Acad Dermatol 1982; 7:100–104. Doss M, von Tiepermann R, Schneider J, Schmid H. New type of hepatic porphyria with porphobilinogen synthase defect and intermittent acute clinical manifestation. Klin Wochenschr 1979; 57:1123–1127.
24. Doss M, Von Tiepermann R, Schneider J. Acute hepatic porphyria syndrome with porphobilinogen synthase defect. Int J Biochem 1980; 12:823–829. 25. Thunell S. Aminolaevulinate dehydratase porphyria in infancy: a clinical and biochemical study. J Clin Chem Clin Biochem 1987; 25:5–14. 26. Hassoun A, Verstraeten L, Mercelis R, Martin J-J. Biochemical diagnosis of an hereditary aminolaevulinic dehydratase deficiency in a 63-year-old man. J Clin Chem Clin Biochem 1989; 27:781–786. 27. Sassa S. ALAD porphyria. Semin Liver Dis 1998; 18:95–101. 28. Doss MO, Stauch T, Gross U, et al. The third case of Doss porphyria (delta-amino-levulinic acid dehydratase deficiency) in Germany. J Inherit Metab Dis 2004; 27:529–536. 29. Akagi R, Kato N, Inoue R, et al. d-Aminolevulinate dehydratase (ALAD) porphyria: the first case in North America with two novel ALAD mutations. Moi Genet Metab, in press. 30. Wolff C, Piderit F, Armas-Merino R. Deficiency of porphobilinogen synthase associated with acute crisis. Diagnosis of the first two cases in Chile by laboratory methods. Eur J Clin Chem Clin Biochem 1991; 29:313–315. 31. Dyer J, Garrick DP, Inglis A, Pye IF. Plumboporphyria (ALAD deficiency) in a lead worker – a scenario for potential diagnostic confusion. Br J Ind Med 1993; 50:1119–1121. 32. Muraoka A, Suehiro I, Fujii M, Murakami K. deltaAminolevulinic acid dehydratase deficiency porphyria (ADP) with syndrome of inappropriate secretion of antidiuretic hormone (SIADH) in a 69-year-old woman. Kobe J Med Sci 1995; 41:23–31. 33. Doss M, Laubenthal F, Stoeppler M. Lead poisoning in inherited d-aminolevulinic acid dehydratase deficiency. Int Arch Occup Environ Health 1984; 54:55–63. 34. Gross U, Sassa S, Jacob K, et al. 5-Aminolevulinic acid dehydratase deficiency porphyria: a twenty-year clinical and biochemical follow-up. Clin Chem 1998; 44:1892–1896. 35. Mercelis R, Hassoun A, Verstraeten L, et al. Porphyric neuropathy and hereditary d-aminolevulinic acid dehydratase deficiency in an adult. J Neurol Sci 1990; 95:39–47. 36. Sassa S, Akagi R, Nishitani C, et al. Late-onset porphyrias: what are they? Cell Mol Biol (Noisy-le-Grand) 2002; 48:97–101. 37. Sassa S, Fujita H, Kappas A. Genetic and chemical influences on heme biosynthesis. In: Kotyk A, Skoda J, Paces V, Kostka V, eds. Highlights of modern biochemistry. Utrecht: VSP International Science; 1989:329–337. 38. Anderson PM, Desnick RJ. Purification and properties of daminolevulinate dehydrase from human erythrocytes. J Biol Chem 1979; 254:6924–6930. 39. Jaffe EK, Abrams WR, Kaempfen HX, Harris KA Jr. 5Chlorolevulinate modification of porphobilinogen synthase identifies a potential role for the catalytic zinc. Biochem 1992; 31:2113–2123. 40. Erskine PT, Senior N, Awan S, et al. X-ray structure of 5aminolaevulinate dehydratase, a hybrid aldolase. Nature Struct Biol 1997; 4:1025–1031. 41. Guo GG, Gu M, Etlinger JD. 240-kDa proteasome inhibitor (CF-2) is identical to delta-aminolevulinic acid dehydratase. J Biol Chem 1994; 269:12399–12402. 42. Granick JL, Sassa S, Granick S, et al. Studies in lead poisoning. II. Correlations between the ratio of activated and inactivated daminolevulinic acid dehydratase of whole blood and the blood lead level. Biochem Med 1973; 8:149–159. 43. Bergdahl IA, Grubb A, Schutz A, et al. Lead binding to deltaaminolevulinic acid dehydratase (ALAD) in human erythrocytes. Pharmacol Toxicol 1997; 81:153–158. 44. Wetmur JG. Influence of the common human d-aminolevulinate dehydratase polymorphism on lead body burden. Environ Health Perspect 1994; 102 (suppl 3):215–219.
Chapter 72 THE PORPHYRIAS
45. Fleming DE, Chettle DR, Wetmur JG, et al. Effect of the deltaaminolevulinate dehydratase polymorphism on the accumulation of lead in bone and blood in lead smelter workers. Environ Res 1998; 77:49–61. 46. Alexander BH, Checkoway H, Costa-Mallen P, et al. Interaction of blood lead and delta-aminolevulinic acid dehydratase genotype on markers of heme synthesis and sperm production in lead smelter workers. Environ Health Perspect 1998; 106:213–216. 47. Potluri VR, Astrin KH, Wetmur JG, et al. Human daminolevulinate dehydratase: chromosomal localization to 9q34 by in situ hybridization. Hum Genet 1987; 76: 236–239. 48. Kaya AH, Plewinska M, Wong DM, et al. Human daminolevulinate dehydratase (ALAD) gene: structure and alternative splicing of the erythroid and housekeeping mRNAs. Genomics 1994; 19:242–248. 49. Maruno M, Furuyama K, Akagi R, et al. Highly heterogeneous nature of delta-aminolevulinate dehydratase (ALAD) deficiencies in ALAD porphyria. Blood 2001; 97:2972–2978. 50. Stenson PD, Ball EV, Mort M, et al. Human gene mutation database (HGMD): 2003 update. Hum Mutat 2003; 21:577–581. 51. De Verneuil H, Doss M, Brusco N, et al. Hereditary hepatic porphyria with delta-aminolevulinate dehydrase deficiency: immunologic characterization of the non-catalytic enzyme. Hum Genet 1985; 69:174–177. 52. Fujita H, Sassa S, Lundgren J, et al. Enzymatic defect in a child with hereditary hepatic porphyria due to homozygous daminolevulinic acid dehydratase deficiency: immunochemical studies. Pediatrics 1987; 80:880–885. 53. Thunell S, Henrichson A, Floderus Y, et al. Liver transplantation in a boy with acute porphyria due to aminolaevulinate dehydratase deficiency. Eur J Clin Chem Clin Biochem 1992; 30:599–606. 54. Shimizu Y, Ida S, Naruto H, Urata G. Excretion of porphyrins in urine and bile after the administration of delta-aminolevulinic acid. J Lab Clin Med 1978; 92:795–802. 55. Elder GH, Smith SG, Herrero C, et al. Hepatoerythropoietic porphyria: a new uroporphyrinogen decarboxylase defect or homozygous porphyria cutanea tarda? Lancet 1981; 1:916–919. 56. Anderson KE, Fischbein A, Kestenbaum D, et al. Plumbism from airborne lead in a firing range. An unusual exposure to a toxic heavy metal. Am J Med 1977; 63:306–312. 57. Akagi R, Prchal JT, Eberhart CE, Sassa S. An acquired acute hepatic porphyria: a novel type of d-aminolevulinate dehydratase inhibition. Clin Chim Acta 1992; 212:79–84. 58. Rank JM, Pascual-Leone A, Payne W, et al. Hematin therapy for the neurological crisis of tyrosinemia. J Pediatr 1991; 118:136–139. 59. Lamon JM, Frykholm BC, Tschudy DP. Hematin administration to an adult with lead intoxication. Blood 1979; 53:1007–1011. 60. Wilson JP. Stokvis and acute porphyria. Mol Aspects Med 1990; 11:6. 61. Moore MR, McColl KEL, Rimington C, Goldberg A. Disorders of porphyrin metabolism. New York: Plenum; 1987. 62. Wetterberg L. A neuropsychiatric and genetical investigation of acute intermittent porphyria. Norstedts: Svenska Bokfölaget, Scandinavian University Books; 1967. 63. Lee J-S, Anvret M. Identification of the most common mutation within the porphobilinogen deaminase gene in Swedish patients with acute intermittent porphyria. Proc Natl Acad Sci USA 1991; 88:10912–10915. 64. Mustajoki P, Koskelo P. Hereditary hepatic porphyrias in Finland. Acta Med Scand 1976; 200:171–178. 65. Tishler P, Woodward B, O’Connor J, et al. High prevalence of intermittent acute porphyria in a psychiatric patient population. Am J Psychiatry 1985; 142:1430–1436.
66. Jara-Prado A, Yescas P, Sanchez FJ, et al. Prevalence of acute intermittent porphyria in a Mexican psychiatric population. Arch Med Res 2000; 31:404–408. 67. Mustajoki P, Kauppinen R, Lannfelt L, et al. Frequency of low erythrocyte porphobilinogen deaminase activity in Finland. J Intern Med 1992; 231:389–395. 68. Nordmann Y, Puy H, Da Silva V, et al. Acute intermittent porphyria: prevalence of mutations in the porphobilinogen deaminase gene in blood donors in France. J Intern Med 1997; 242:213–217. 69. Rimington C. Was Hippocrates the first to describe a case of acute porphyria? Int J Biochem 1993; 25:1351–1352. 70. Loftus LS, Arnold WN. Vincent van Gogh’s illness: acute intermittent porphyria. Br Med J 1991; 303:1589–1591. 71. Cox TM, Jack N, Lofthouse S, et al. King George III and porphyria: an elemental hypothesis and investigation. Lancet 2005; 366:332–335. 72. Winkler M, Anderson KE. Vampires, porphyria and the media: the medicalization of a myth. Perspect Biol Med 1990; 33:598–611. 73. Daniell WE, Stockbridge HL, Labbe RF, et al. Environmental chemical exposures and disturbances of heme synthesis. Environ Health Perspect 1997; 105 (suppl 1):37–53. 74. Hahn M, Bonkovsky HL. Multiple chemical sensitivity syndrome and porphyria. A note of caution and concern. Arch Intern Med 1997; 157:281–285. 75. McDonagh AF, Bissell DM. Porphyria and porphyrinology – the past fifteen years. Semin Liver Dis 1998; 18:3–15. 76. Beukeveld GJJ, Wolthers BG, Nordmann Y, et al. A retrospective study of a patient with homozygous form of acute intermittent porphyria. J Inherit Metab Dis 1990; 13:673–683. 77. Solis C, Martinez-Bermejo A, Naidich TP, et al. Acute intermittent porphyria: studies of the severe homozygous dominant disease provides insights into the neurologic attacks in acute porphyrias. Arch Neurol 2004; 61:1764–1770. 78. Picat C, Delfau MH, De Rooij FWM, et al. Identification of the mutations in the parents of a patient with a putative compound heterozygosity for acute intermittent porphyria. J Inherit Metab Dis 1990; 13:684–686. 79. Waldenström J. The porphyrias as inborn errors of metabolism. Am J Med 1957; 22:758–773. 80. Goldberg A. Acute intermittent porphyria. A study of 50 cases. Q J Med 1959; 28:183–209. 81. Stein JA, Tschudy DP. Acute intermittent porphyria: a clinical and biochemical study of 46 patients. Medicine 1970; 49:1–16. 82. Ridley A. Porphyric neuropathy. In: Dyck PJ, Thomas PK, Lambert EH, Bunge R, eds. Peripheral neuropathy, vol. 2. Philadelphia: WB Saunders; 1984:1704–1716. 83. Cavanagh JB, Mellick RS. On the nature of the peripheral nerve lesions associated with acute intermittent porphyria. J Neurol Neurosurg Psychiatry 1965; 28:320–327. 84. Sweeney VP, Pathak MA, Asbury AK. Acute intermittent porphyria: increased ALA-synthetase activity during an acute attack. Brain 1970; 93:369–380. 85. Barohn RJ, Sanchez JE, Anderson KE. Acute peripheral neuropathy due to hereditary coproporphyria. Muscle Nerve 1994; 17:793–799. 86. Wikberg A, Andersson C, Lithner F. Signs of neuropathy in the lower legs and feet of patients with acute intermittent porphyria. J Intern Med 2000; 248:27–32. 87. Stein JA, Curl FD, Valsamis M, Tschudy DP. Abnormal iron and water metabolism in acute intermittent porphyria with new morphologic findings. Am J Med 1972; 53:784–789. 88. Perlroth MG, Tschudy DP, Marver HS, et al. Acute intermittent porphyria: new morphologic and biochemical findings. Am J Med 1966; 41:149–162.
1419
Section XII. Inherited and Pediatric Diseases of the Liver
89. Bloomer JR, Berk PD, Bonkowsky HL, et al. Blood volume and bilirubin production in acute intermittent porphyria. N Engl J Med 1971; 284:17–20. 90. Eales L, Dowdle EB, Sweeney GD. The electrolyte disorder of the acute porphyric attack and the possible role of deltaaminolaevulic acid. S Afr J Lab Clin Med 1971; 17 (special issue):89–97. 91. Tschudy DP, Lamon JM. Porphyrin metabolism and the porphyrias. In: Bondy PK, Rosenberg LE, eds. Duncan’s diseases of metabolism. Philadelphia: WB Saunders; 1980:939–1007. 92. Eales L. Porphyria as seen in Cape Town. A survey of 250 patients and some recent studies. S Afr J Lab Clin Med 1963; 9:151. 93. Tschudy DP, Valsamis M, Magnussen CR. Acute intermittent porphyria: clinical and selected research aspects. Ann Intern Med 1975; 83:851–864. 94. Kauppinen R, Mustajoki P. Prognosis of acute porphyria: occurrence of acute attacks, precipitating factors, and associated diseases. Medicine 1992; 71:1–13. 95. Ostrowski J, Kostrzewska E, Michalak T, et al. Abnormalities in liver function and morphology and impaired aminopyrine metabolism in hereditary hepatic porphyrias. Gastroenterology 1983; 85:1131–1137. 96. Lithner F, Wetterberg L. Hepatocellular carcinoma in patients with acute intermittent porphyria. Acta Med Scand 1984; 215:271–274. 97. Hardell L, Bengtsson NO, Jonsson U, et al. Aetiological aspects on primary liver cancer with special regard to alcohol, organic solvents and acute intermittent porphyria – an epidemiological investigation. Br J Cancer 1984; 50:389–397. 98. Bengtsson NO, Hardell L. Porphyrias, porphyrins and hepatocellular cancer. Br J Cancer 1986; 54:115–117. 99. Gubler JG, Bargetzi MJ, Meyer UA. Primary liver carcinoma in two sisters with acute intermittent porphyria. Am J Med 1990; 89:540–541. 100. Kauppinen R, Mustajoki P. Acute hepatic porphyria and hepatocellular carcinoma. Br J Cancer 1987; 57:117–120. 101. Andant C, Puy H, Faivre J, Deybach JC. Acute hepatic porphyrias and primary liver cancer (letter). N Engl J Med 1998; 338:1853–1854. 102. Andant C, Puy H, Bogard C, et al. Hepatocellular carcinoma in patients with acute hepatic porphyria: frequency of occurrence and related factors. J Hepatol 2000; 32:933–939. 103. Andersson C, Bjersing L, Lithner F. The epidemiology of hepatocellular carcinoma in patients with acute intermittent porphyria. J Intern Med 1996; 240:195–201. 104. Linet MS, Gridley G, Nyren O, et al. Primary liver cancer, other malignancies, and mortality risks following porphyria: a cohort study in Denmark and Sweden. Am J Epidemiol 1999; 149:1010–1015. 105. Sardh E, Harper P, Andersson D. Clinical, biochemical and pathohistological aspects in 12 patients with acute porphyria and hepatocellular carcinoma. In: Meissner P, ed. Porphyrins and porphyrias 2005. Cape Town: 2005:39. 106. Church SE, McColl KE, Moore MR, Youngs GR. Hypertension and renal impairment as complications of acute porphyria. Nephrol Dial Transplant 1992; 7:986–990. 107. Andersson C, Wikberg A, Stegmayr B, Lithner F. Renal symptomatology in patients with acute intermittent porphyria. A population-based study. J Intern Med 2000; 248:319–325. 108. Nunez DJ, Williams PF, Herrick AL, et al. Renal transplantation for chronic renal failure in acute porphyria. Nephrol Dial Transplant 1987; 2:271–274. 109. Barone GW, Gurley BJ, Anderson KE, et al. The tolerability of newer immunosuppressive medications in a patient with acute intermittent porphyria. J Clin Pharmacol 2000; 41:113–115.
1420
110. Hollander CS, Scott RL, Tschudy DP, et al. Increased protein bound iodine and thyroxine binding globulin in acute intermittent porphyria. N Engl J Med 1967; 277:995–1000. 111. Mustajoki P, Nikkila EA. Serum lipoproteins in asymptomatic acute porphyria: no evidence for hyperbetalipoproteinemia. Metabolism 1984; 33:266–269. 112. Fernandez-Miranda C, De La Calle M, Larumbe S, et al. Lipoprotein abnormalities in patients with asymptomatic acute porphyria. Clin Chim Acta 2000; 294:37–43. 113. Jordan PM. The biosynthesis of 5-aminolevulinic acid and its transformation into coproporphyrinogen in animals and bacteria. In: Dailey HA, ed. Biosynthesis of heme and chlorophylls. New York: McGraw-Hill; 1990:55–121. 114. Awan SJ, Siligardi G, Shoolingin-Jordan PM, Warren MJ. Reconstitution of the holoenzyme form of Escherichia coli porphobilinogen deaminase from apoenzyme with porphobilinogen and preuroporphyrinogen: a study using circular dichroism spectroscopy. Biochemistry 1997; 36:9273–9282. 115. Shoolingin-Jordan P, McNeill L, Al Dbass A, et al. Structural aspects of human PBG deaminase in relation to acute intermittent porphyria. In: Meissner P, ed. Porphyrins and porphyrias 2005. Cape Town: 2005:40. 116. Raich N, Romeo PH, Dubart A, et al. Molecular cloning and complete primary sequence of human erythrocyte porphobilinogen deaminase. Nucleic Acids Res 1986; 14:5955–5968. 117. Grandchamp B, De Verneuil H, Beaumont C, et al. Tissue specific expression of porphobilinogen deaminase. Two isoenzymes from a single gene. Eur J Biochem 1987; 162:105–110. 118. Namba H, Narahara K, Tsuji K, et al. Assignment of human porphobilinogen deaminase to 11q24.1Æq24.2 by in situ hybridization and gene dosage studies. Cytogenet Cell Genet 1991; 57:105–108. 119. Mignotte V, Eleout JF, Raich N, Romeo PH. Cis- and transacting elements involved in the regulation of the erythroid promoter of the human porphobilinogen deaminase gene. Proc Natl Acad Sci USA 1989; 86:6548–6452. 120. Raich N, Mignotte V, Dubart A, et al. Regulated expression of the overlapping ubiquitous and erythroid transcription units of the human porphobilinogen deaminase (PBG-D) gene introduced into non-erythroid and erythroid cells. J Biol Chem 1989; 264:10186–10192. 121. Astrin KH, Desnick RJ. Molecular basis of acute intermittent porphyria: mutations and polymorphisms in the human hydroxymethylbilane synthase gene. Hum Mutat 1994; 4:243–252. 122. Grandchamp B. Acute intermittent porphyria. Semin Liver Dis 1998; 18:17–24. 123. Robreau-Fraolini AM, Puy H, Aquaron C, et al. Porphobilinogen deaminase gene in African and Afro-Caribbean ethnic groups: mutations causing acute intermittent porphyria and specific intragenic polymorphisms. Hum Genet 2000; 107:150–159. 124. Gu XF, De Rooij F, Lee JS, et al. High prevalence of a point mutation in the porphobilinogen deaminase gene in Dutch patients with acute intermittent porphyria. Hum Genet 1993; 91:128–130. 125. Greene-Davis ST, Neumann PE, Mann OE, et al. Detection of a R173W mutation in the porphobilinogen deaminase gene in the Nova Scotian “foreign Protestant” population with acute intermittent porphyria: a founder effect. Clin Biochem 1997; 30:607–612. 126. Schneider-Yin X, Bogard C, Rufenacht UB, et al. Identification of a prevalent nonsense mutation (W283X) and two novel mutations in the porphobilinogen deaminase gene of Swiss patients with acute intermittent porphyria. Hum Hered 2000; 50:247–250.
Chapter 72 THE PORPHYRIAS
127. Norton B, Lanyon WG, Moore MR, et al. Evidence for involvement of a second genetic locus on chromosome 11q in porphyrin metabolism. Hum Genet 1993; 91:576–578. 128. Whatley SD, Roberts AG, Elder GH. De-novo mutation and sporadic presentation of acute intermittent porphyria. Lancet 1995; 346:1007–1008. 129. Brownlie PD, Lambert R, Louie GV, et al. The threedimensional structures of mutants of porphobilinogen deaminase: toward an understanding of the structural basis of acute intermittent porphyria. Protein Sci 1994; 3: 1644–1650. 130. Desnick RJ, Ostasiewicz LT, Tishler PA, Mustajoki P. Acute intermittent porphyria: characterization of a novel mutation in the structural gene for porphobilinogen deaminase. J Clin Invest 1985; 76:865–874. 131. Grandchamp B, Picat C, De Rooij FWM, et al. Molecular analysis of acute intermittent porphyria in a Finnish family with normal erythrocyte porphobilinogen deaminase. Eur J Clin Invest 1989; 19:415–418. 132. Mathews-Roth MM. Anemia in erythropoietic protoporphyria (letter). JAMA 1974; 230:824. 133. Puy H, Gross U, Deybach JC, et al. Exon 1 donor splice site mutations in the porphobilinogen deaminase gene in the nonerythroid variant form of acute intermittent porphyria. Hum Genet 1998; 103:570–575. 134. Mustajoki S, Pihlaja H, Ahola H, et al. Three splicing defects, an insertion, and two missense mutations responsible for acute intermittent porphyria. Hum Genet 1998; 102:541–548. 135. Whatley SD, Roberts AG, Llewellyn DH, et al. Non-erythroid form of acute intermittent porphyria caused by promoter and frameshift mutations distant from the coding sequence of exon 1 of the HMBS gene. Hum Genet 2000; 107:243–248. 136. Strand LJ, Felsher BF, Redeker AG, Marver HS. Heme biosynthesis in intermittent acute porphyria: decreased hepatic conversion of porphobilinogen to porphyrins and increased deltaaminolevulinic acid synthetase activity. Proc Natl Acad Sci USA 1970; 67:1315–1320. 137. Anderson KE, Spitz IM, Bardin CW, Kappas A. A GnRH analogue prevents cyclical attacks of porphyria. Arch Intern Med 1990; 150:1469–1474. 138. Anderson KE, Bradlow HL, Sassa S, Kappas A. Studies in porphyria. VIII. Relationship of the 5a-reductive metabolism of steroid hormones to clinical expression of the genetic defect in acute intermittent porphyria. Am J Med 1979; 66:644–650. 139. Milo R, Neuman M, Klein C, et al. Acute intermittent porphyria in pregnancy. Obstet Gynecol 1989; 73:450–452. 140. Shenhav S, Gemer O, Sassoon E, Segal S. Acute intermittent porphyria precipitated by hyperemesis and metoclopramide treatment in pregnancy. Acta Obstet Gynecol Scand 1997; 76:484–485. 141. Kappas A, Sassa S, Anderson KE. The porphyrias. In: Stanbury JB, Wyngaarden JB, Fredrickson DS, et al., eds. The metabolic basis of inherited disease. New York: McGraw-Hill; 1983:1301–1384. 142. Andersson C, Bylesjo I, Lithner F. Effects of diabetes mellitus on patients with acute intermittent porphyria. J Intern Med 1999; 245:193–197. 143. Koleva M, Stoytchev T. On the enzyme-inducing action of calcium antagonists. Arch Toxicol 1993; 67:294–296. 144. Lambrecht RW, Gildemeister OS, Williams A, et al. Effects of selected antihypertensives and analgesics on hepatic porphyrin accumulation: implications for clinical porphyria. Biochem Pharmacol 1999; 58:887–896. 145. Schoenfeld N, Aelion J, Beigel Y, et al. The porphyrogenic effects of calcium channel blocking drugs. Clin Sci 1985; 69:581–586. 146. Marks GS, Goldman DR, McCluskey SA, et al. The effects of dihydropyridine calcium antagonists on heme biosynthesis in
147.
148.
149.
150. 151.
152.
153. 154. 155. 156.
157.
158. 159.
160.
161.
162.
163.
164. 165. 166.
167.
168.
chick embryo liver cell culture. Can J Physiol Pharmacol 1986; 64:438–443. Moore MR, Hift RJ. Drugs in the acute porphyries – toxicogenetic diseases. Cell Mol Biol (Noisy-le-grand) 1997; 43:89–94. Gorchein A. Drug treatment of hypertension in acute intermittent porphyria: doxazosin and amlodipine (letter). Br J Clin Pharmacol 1997; 43:339–340. Kepple A, Cernek PK. Amlodipine-induced acute intermittent porphyria exacerbation (letter). Ann Pharmacother 1997; 31:253. Gorchein A. Drug treatment in acute porphyria. Br J Clin Pharmacol 1997; 44:427–434. Lambrecht RW, Gildemeister OS, Pepe JA, et al. Effects of antidepressants and benzodiazepine-type anxiolytic agents on hepatic porphyrin accumulation in primary cultures of chick embryo liver cells. J Pharmacol Exp Ther 1999; 291:1150– 1155. Dover SB, Plenderleith L, Moore MR, McColl KE. Safety of general anaesthesia and surgery in acute hepatic porphyria. Gut 1994; 35:1112–1115. Mustajoki P, Heinonen J. General anesthesia in “inducible” porphyrias. Anesthesiology 1980; 53:15–20. Harrison GG, Meissner PN, Hift RJ. Anaesthesia for the porphyric patient. Anaesthesia 1993; 48:417–421. Samuels B, Bezwoda WR, Derman DP, Goss G. Chemotherapy in porphyria. S Afr Med J 1984; 65:924–926. Marks GS, McCluskey SA, Machie JE, et al. Disruption of hepatic heme biosynthesis after interaction of xenobiotics with cytochrome P-450. FASEB J 1988; 2:2774–2783. Tishler PV. The effect of therapeutic drugs and other pharmacologic agents on activity of porphobilinogen deaminase, the enzyme that is deficient in intermittent acute porphyria. Life Sci 1999; 65:207–214. Thunell S, Floderus Y, Henrichson A, et al. Alcoholic beverages in acute porphyria. J Stud Alcohol 1992; 53:272–276. Louis CA, Sinclair JF, Wood SG, et al. Synergistic induction of cytochrome-P450 by ethanol and isopentanol in cultures of chick embryo and rat hepatocytes. Toxicol Appl Pharmacol 1993; 118:169–176. Tschudy DP. The influence of hormonal and nutritional factors on the regulation of liver heme biosynthesis. In: De Matteis F, Aldridge WN, eds. Heme and hemoproteins – handbook of experimental pharmacology, vol. 44. Berlin: Springer Verlag; 1978:255–271. Tschudy DP, Welland FH, Collins A, Hunter GJ. The effect of carbohydrate feeding on the induction of d-aminolevulinic acid synthetase. Metabolism 1964; 13:396–406. Thaler MM, Dawber NH. Stimulation of bilirubin formation in liver of newborn rats by fasting and glucagon. Gastroenterology 1977; 72:312–315. Welland FH, Hellman ES, Gaddis EM, et al. Factors affecting the excretion of porphyrin precursors by patients with acute intermittent porphyria. 1. The effects of diet. Metabolism 1964; 13:232–250. Felsher BF, Redeker AG. Acute intermittent porphyria: effect of diet and griseofulvin. Medicine 1967; 46:213–217. Anderson KE, Kappas A. Dietary regulation of cytochrome P450. Annu Rev Nutr 1991; 11:141–167. Lip GYH, McColl KEL, Goldberg A, Moore MR. Smoking and recurrent attacks of acute intermittent porphyria. Br Med J 1991; 302:507. Paxton JW, Moore MR, Beattie AD, Goldberg A. 17-Oxosteroid conjugates in plasma and urine of patients with acute intermittent porphyria. Clin Sci Mol Med 1974; 46:207–222. Meyer UA, Schuurmans MM, Lindberg RLP. Acute porphyrias: pathogenesis of neurological manifestations. Semin Liver Dis 1998; 18:43–52.
1421
Section XII. Inherited and Pediatric Diseases of the Liver
169. Kupferschmidt H, Bont A, Schnorf H, et al. Transient cortical blindness and bioccipital brain lesions in two patients with acute intermittent porphyria. Ann Intern Med 1995; 123:598–600. 170. Sze G. Cortical brain lesions in acute intermittent porphyria (letter; comment). Ann Intern Med 1996; 125:422–423. 171. Lithner F. Could attacks of abdominal pain in cases of acute intermittent porphyria be due to intestinal angina? J Intern Med 2000; 247:407–409. 172. Cavanagh JB, Ridley AR. The nature of the neuropathy complicating acute intermittent porphyria. Lancet 1967; 2:1023–1024. 173. Hamfelt A, Wetterberg L. Pyridoxal phosphate in acute intermittent porphyria. Ann NY Acad Sci 1969; 166:361–364. 174. Litman DA, Correia MA. Elevated brain tryptophan and enhanced 5-hydroxytryptamine turnover in acute hepatic heme deficiency: clinical implications. J Pharmacol Exp Ther 1985; 232:337–345. 175. Puy H, Deybach JC, Baudry P, et al. Decreased nocturnal plasma melatonin levels in patients with recurrent acute intermittent porphyria attacks. Life Sci 1993; 53:621–627. 176. Bonkowsky HL, Tschudy DP, Collins A, et al. Repression of the overproduction of porphyrin precursors in acute intermittent porphyria by intravenous infusions of hematin. Proc Natl Acad Sci USA 1971; 8:2725–2729. 177. Anderson KE, Drummond GS, Freddara U, et al. Porphyrogenic effects and induction of heme oxygenase in vivo by daminolevulinic acid. Biochim Biophys Acta 1981; 676:289–299. 178. Müller WE, Snyder SH: d-Aminolevulinic acid: influences on synaptic GABA receptor binding may explain CNS symptoms of porphyria. Ann Neurol 1977; 2:340–342. 179. Brennan MJW, Cantrill RC. d-Aminolaevulinic acid is a potent agonist for GABA autoreceptors. Nature 1979; 280:514–515. 180. Lindberg RL, Porcher C, Grandchamp B, et al. Porphobilinogen deaminase deficiency in mice causes a neuropathy resembling that of human hepatic porphyria. Nat Genet 1996; 12:195– 199. 181. Jover R, Hoffmann F, Scheffler-Koch V, Lindberg RL. Limited heme synthesis in porphobilinogen deaminase-deficient mice impairs transcriptional activation of specific cytochrome P450 genes by phenobarbital. Eur J Biochem 2000; 267:7128–7137. 182. Lindberg RL, Martini R, Baumgartner M, et al. Motor neuropathy in porphobilinogen deaminase-deficient mice imitates the peripheral neuropathy of human acute porphyria. J Clin Invest 1999; 103:1127–1134. 183. Soonawalla ZF, Orug T, Badminton MN, et al. Liver transplantation as a cure for acute intermittent porphyria. Lancet 2004; 363:705–706. 184. Doss M, Schermuly E. Urinary porphyrin excretion pattern and isomer distribution of I and III in human porphyrin disorders. In: Doss M, ed. Porphyrins in human diseases. Basel: S Karger; 1976:189–204. 185. Doss M, Schermuly E, Look D, et al. Studies on the metabolism of d-aminolevulinic acid and porphobilinogen in liver biopsies from patients with acute intermittent porphyria. In: Doss M, ed. Porphyrins in human diseases. Basel: S Karger; 1976:205–216. 186. Minder EI. Coproporphyrin isomers in acute-intermittent porphyria. Scand J Clin Lab Invest 1993; 53:87–90. 187. Mustajoki P. Variegate porphyria. Twelve years’ experience in Finland. Q J Med 1980; 44:191–203. 188. Kauppinen R. Prognosis of acute porphyrias and molecular genetics of acute intermittent porphyria in Finland (thesis). University of Helskinki; 1992. 189. Anderson KE, Sassa S, Peterson CM, Kappas A. Increased erythrocyte uroporphyrinogen-I-synthetase, d-aminolevulinic acid dehydratase and protoporphyrin in hemolytic anemias. Am J Med 1977; 63:359–364.
1422
190. Blum M, Koehl C, Abecassis J. Variations in erythrocyte uroporphyrinogen I synthetase activity in non porphyrias. Clin Chim Acta 1978; 87:119–125. 191. Kostrzewska E, Gregor A. Increased activity of porphobilinogen deaminase in erythrocytes during attacks of acute intermittent porphyria. Ann Clin Res 1986; 18:195–198. 192. Kauppinen R, Mustajoki S, Pihlaja H, et al. Acute intermittent porphyria in Finland: 19 mutations in the porphobilinogen deaminase gene. Hum Mol Genet 1995; 4:215–222. 193. Puy H, Deybach JC, Lamoril J, et al. Molecular epidemiology and diagnosis of PBG deaminase gene defects in acute intermittent porphyria. Am J Hum Genet 1997; 60:1373– 1383. 194. Sassa S, Solish G, Levere RD, Kappas A. Studies in porphyria. IV. Expression of the gene defect of acute intermittent porphyria in cultured human skin fibroblasts and amniotic cells: prenatal diagnosis of the porphyric trait. J Exp Med 1975; 142:722–731. 195. Atsmon A, Blum I, Fischl J. Treatment of an acute attack of porphyria variegata with propranolol. S Afr Med J 1972; 46:311–314. 196. Beattie AD, Moore MR, Goldberg A, Ward RL. Acute intermittent porphyria: response of tachycardia and hypertension to propranolol. Br Med J 1973; 3:257–260. 197. Schoenfeld N, Epstein O, Atsmon A. The effect of betaadrenergic blocking agents on experimental porphyria induced by 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC) in vivo and in vitro. Biochim Biophys Acta 1976; 444:286–293. 198. Bonkowsky HL, Tschudy DP. Hazard of propranolol in treatment of acute porphyria (letter). Br Med J 1974; 4:47–48. 199. Lamon JM, Frykholm BC, Hess RA, Tschudy DP. Hematin therapy for acute porphyria. Medicine 1979; 58:252–269. 200. Mustajoki P, Nordmann Y. Early administration of heme arginate for acute porphyric attacks. Arch Intern Med 1993; 153:2004–2008. 201. Herrick AL, McColl KEL, Moore MR, et al. Controlled trial of haem arginate in acute hepatic porphyria. Lancet 1989; 1:1295–1297. 202. Bissell DM. Treatment of acute hepatic porphyria with hematin. J Hepatol 1988; 6:1–7. 203. Goetsch CA, Bissell DM. Instability of hematin used in the treatment of acute hepatic porphyria. N Engl J Med 1986; 315:235–238. 204. Jones RL. Hematin-derived anticoagulant. Generation in vitro and in vivo. J Exp Med 1986; 163:724–439. 205. Green D, Reynolds N, Klein J, et al. The inactivation of hemostatic factors by hematin. J Lab Clin Med 1983; 102:361–369. 206. Khanderia U. Circulatory collapse associated with hemin therapy for acute intermittent porphyria. Clin Pharmacy 1986; 5:690–692. 207. Daimon M, Susa S, Igarashi M, et al. Administration of heme arginate, but not hematin, caused anaphylactic shock. Am J Med 2001; 110:240. 208. Jeelani Dhar G, Bossenmaier I, Cardinal R, et al. Transitory renal failure following rapid administration of a relatively large amount of hematin in a patient with acute intermittent porphyria in clinical remission. Acta Med Scand 1978; 203:437–443. 209. Tenhunen R, Mustajoki P. Acute porphyria: treatment with heme. Semin Liver Dis 1998; 18:53–55. 210. Bonkovsky HL, Healey BS, Lourie AN, Gerron GG. Intravenous heme-albumin in acute intermittent porphyria: evidence for repletion of hepatic hemoproteins and regulatory heme pools. Am J Gastroenterol 1991; 86:1050–1056. 211. Mustajoki P, Tenhunen R, Tokola O, Gothani G. Haem arginate in the treatment of acute hepatic porphyrias. Br Med J 1986; 293:538–539.
Chapter 72 THE PORPHYRIAS
212. Dover SB, Graham A, Fitzsimons E, et al. Tin protoporphyrin prolongs the biochemical remission produced by heme arginate in acute hepatic porphyria. Gastroenterology 1993; 105:500–506. 213. Anderson KE, Lee C, Egger NG, et al. Prolongation of the biochemical effects and clearance of heme by tin mesoporphyrin in acute porphyria. J Invest Med 2004; 52:S383 (abstract). 214. Russo SM, Pepe JA, Cable EE, et al. Repression of ALA synthase by heme and zinc-mesoporphyrin in a chick embryo liver cell culture model of acute porphyria. Eur J Clin Invest 1994; 24:406–415. 215. Anderson KE, Alvares AP, Sassa S, Kappas A. Studies in porphyria. V. Drug oxidation rates in hereditary hepatic porphyria. Clin Pharmacol Ther 1976; 19:47–54. 216. Herrick A, McColl KEL, McLellan A, et al. Effect of haem arginate therapy on porphyrin metabolism and mixed function oxygenase activity in acute hepatic porphyria. Lancet 1987; 2:1178–1179. 217. Mustajoki P, Himberg JJ, Tokola O, Tenhunen R. Rapid normalization of antipyrine oxidation by heme in variegate porphyria. Clin Pharmacol Ther 1992; 51:320–324. 218. Sardh E, Rejkjaer L, Harper P, Andersson DEH. First clinical trial of I.V. rhPBGD in healthy subjects with and without diagnosed manifest acute intermittent porphyria (AIP). Physiol Res 2003; 52:23S (abstract). 219. Johansson A, Nowak G, Moller C, et al. Adenoviral-mediated expression of porphobilinogen deaminase in liver restores the metabolic defect in a mouse model of acute intermittent porphyria. Mol Ther 2004; 10:337–343. 220. Johansson A, Nowak G, Moller C, Harper P. Non-viral delivery of the porphobilinogen deaminase cDNA into a mouse model of acute intermittent porphyria. Mol Genet Metab 2004; 82:20–26. 221. Bonkowsky HL, Sinclair PR, Emery S, Sinclair JF. Seizure management in acute hepatic porphyria: risks of valproate and clonazepam. Neurology 1980; 30:588–592. 222. Larson AW, Wasserstrom WR, Felsher BF, Shih JC. Posttraumatic epilepsy and acute intermittent porphyria: effects of phenytoin, carbamazepine, and clonazepam. Neurology 1978; 28:824–828. 223. Hahn M, Gildemeister OS, Krauss GL, et al. Effects of new anticonvulsant medications on porphyrin synthesis in cultured liver cells: potential implications for patients with acute porphyria. Neurology 1997; 49:97–106. 224. Siepmann M, Stolzel U, Sieg I, et al. [Cimetidine in treatment of acute intermittent porphyria.] Z Gastroenterol 1993; 31:246–249. 225. Kappas A, Bradlow HL. Enhancement of 5a-steroid hormone metabolism in patients with acute intermittent porphyria following treatment with triiodothyronine. In: Doss M, ed. Porphyrins and human diseases. Basel: S Karger; 1976: 274–276. 226. Olsson RA, Ticktin HE. Zinc metabolism in acute intermittent porphyria. J Lab Clin Med 1962; 60:48–52. 227. Peters HA, Cripps DJ, Reese HH. Porphyria: types of etiology and treatment. Int Rev Neurobiol 1974; 16:301–355. 228. Roman W, Oon R, West RF, Reid DP. Zinc sulphate in acute porphyria. Lancet 1967; 2:716. 229. Winston SH, Tishler PV. Sorbent therapy of the porphyrias. V. Adsorption of the porphyrin precursors delta-aminolevulinic acid and porphobilinogen by sorbents in vitro. Methods Find Exp Clin Pharmacol 1986; 8:233–237. 230. Thunell S, Andersson D, Harper P, et al. Effects of administration of antioxidants in acute intermittent porphyria. Eur J Clin Chem Clin Biochem 1997; 35:427–433. 231. De Block CE, Leeuw IH, Gaal LF. Premenstrual attacks of acute intermittent porphyria: hormonal and metabolic aspects – a case report. Eur J Endocrinol 1999; 141:50–54.
232. Yamamori I, Asai M, Tanaka F, et al. Prevention of premenstrual exacerbation of hereditary coproporphyria by gonadotropinreleasing hormone analogue. Intern Med 1999; 38:365–368. 233. Perlroth MG, Marver HS, Tschudy DP. Oral contraceptive agents and the management of acute intermittent porphyria. JAMA 1965; 194:1037–1042. 234. Schley G, Anlauf M, Bock KD. Orale Kontrazeptiva zur Prophylaxe akuter Schübe der intermittierenden Porphyrie. Dtsch Med Wschr 1976; 101:1901–1907. 235. White INH, Muller-Eberhard U. Decreased liver cytochrome P-450 in rats caused by norethindrone or ethynyloestradiol. Biochem J 1977; 166:57–64. 236. Lamon JM, Frykholm BC, Herrera W, Tschudy DP. Danazol administration to females with menses-associated exacerbations of acute intermittent porphyria. J Clin Endocrinol Metab 1979; 48:123–126. 237. Hughes MJ, Rifkind AB. Danazol, a new steroidal inducer of daminolevulinic acid synthetase. J Clin Endocrinol Metab 1981; 52:549–552. 238. Anderson KE, Egger NG, Goeger DE. Heme arginate for prevention of acute porphyric attacks. Acta Haematologica 1997; 98 (suppl 1):120 (abstract). 239. Lamon JM, Frykholm BC, Bennett M, Tschudy DP. Prevention of acute porphyric attacks by intravenous haematin. Lancet 1978; 2:492–494. 240. Fritsch C, Bolsen K, Ruzicka T, Goerz G. Congenital erythropoietic porphyria. J Am Acad Dermatol 1997; 36:594–610. 241. Daikha-Dahmane F, Dommergues M, Narcy F, et al. Congenital erythropoietic porphyria: prenatal diagnosis and autopsy findings in two sibling fetuses. Pediatr Dev Pathol 2001; 4:180–184. 242. Verstraeten L, Van Regemorter N, Pardou A, et al. Biochemical diagnosis of a fatal case of Gunther’s disease in a newborn with hydrops-fetalis. Eur J Clin Chem Clin Biochem 1993; 31:121–128. 243. Piomelli S, Poh-Fitzpatrick MB, Seaman C, et al. Complete suppression of the symptoms of congenital erythropoietic porphyria by long-term treatment with high-level transfusions. N Engl J Med 1986; 314:1029–1031. 244. Pain RW, Welch FW, Woodroffe AJ, et al. Erythropoietic uroporphyria of Gunther first presenting at 58 years with positive family studies. Br Med J 1975; 3:621–623. 245. Weston MJ, Nicholson DC, Lim CK, et al. Congenital erythropoietic uroporphyria (Gunther’s disease) presenting in a middle aged man. Int J Biochem 1978; 9:921–926. 246. Kramer S, Viljoen E, Meyer AM, Metz J. The anemia of erythropoietic porphyria with the first description of the disease in an elderly patient. Br J Haematol 1965; 11:666–675. 247. Deybach JC, De Verneuil H, Phung N, et al. Congenital erythropoietic porphyria (Günther’s disease): enzymatic studies on two cases of late onset. J Lab Clin Med 1981; 97:551–558. 248. Romeo G, Levin EY. Uroporphyrinogen III cosynthetase in human congenital erythropoietic porphyria. Proc Natl Acad Sci USA 1969; 63:856–863. 249. Tsai SF, Bishop DF, Desnick RJ. Human uroporphyrinogen III synthase: molecular cloning, nucleotide sequence and expression of a full-length cDNA. Proc Natl Acad Sci USA 1988; 85:7049–7053. 250. Mathews MA, Schubert HL, Whitby FG, et al. Crystal structure of human uroporphyrinogen III synthase. Embo J 2001; 20:5832–5839. 251. Astrin KH, Warner CA, Yoo HW, et al. Regional assignment of the human uroporphyrinogen III synthase (UROS) gene to chromosome 10q25.2Æq26.3. Hum Genet 1991; 87:18–22. 252. Aizencang G, Solis C, Bishop DF, et al. Human uroporphyrinogen-III synthase: genomic organization, alternative promoters, and erythroid-specific expression. Genomics 2000; 70:223–231.
1423
Section XII. Inherited and Pediatric Diseases of the Liver
253. Fontanellas A, Bensidhoum M, Enriquez de Salamanca R, et al. A systematic analysis of the mutations of the uroporphyrinogen III synthase gene in congenital erythropoietic porphyria. Eur J Hum Genet 1996; 4:274–282. 254. Desnick RJ, Glass IA, Xu W, et al. Molecular genetics of congenital erythropoietic porphyria. Semin Liver Dis 1998; 18:77–84. 255. Frank J, Wang X, Lam HM, et al. C73R is a hotspot mutation in the uroporphyrinogen III synthase gene in congenital erythropoietic porphyria. Ann Hum Genet 1998; 62:225–230. 256. Xu W, Warner CA, Desnick RJ. Congenital erythropoietic porphyria: identification and expression of 10 mutations in the uroporphyrinogen III synthase gene. J Clin Invest 1995; 95:905–912. 257. Warner CA, Poh-Fitzpatrick MB, Zaider EF, et al. Congenital erythropoietic porphyria: a mild variant with low uroporphyrin I levels due to a missense mutation (A66V) encoding residual uroporphyrinogen III synthase activity. Arch Dermatol 1992; 128:1243–1248. 258. Solis C, Aizencang GI, Astrin KH, et al. Uroporphyrinogen III synthase erythroid promoter mutations in adjacent GATA1 and CP2 elements cause congenital erythropoietic porphyria. J Clin Invest 2001; 107:753–762. 259. Watson CJ, Bossenmaier I, Cardinal R. Formation of porphyrin isomers from porphobilinogen by various hemolysates of red cells from bovine and human subjects with erythropoietic (Uro)porphyria. Z Klin Chem Klin Biochem 1969; 7:119–129. 260. Runge W, Watson CJ. The effect of bleeding on the proportion of red fluorescing forms among the total normoblasts of bovine porphyric bone marrow. Blood 1969; 33:119–123. 261. Haining RG, Cowger ML, Shurtleff DB, Labbe RF. Congenital erythropoietic porphyria. I. Case report, special studies and therapy. Am J Med 1968; 45:624–636. 262. Haining RG, Cowger ML, Labbe RF, Finch CA. Congenital erythropoietic porphyria. II. The effects of induced polycythemia. Blood 1970; 36:297–309. 263. Schmid R, Schwartz S, Watson CJ. Porphyrin content of bone marrow and liver in the various forms of porphyria. Arch Intern Med 1954; 93:167–190. 264. Rosenthal IM, Lipton EL, Asrow G. Effect of splenectomy on porphyria erythropoietica. Pediatrics 1955; 15:663–675. 265. Guarini L, Piomelli S, Poh-Fitzpatrick MB. Hydroxyurea in congenital erythropoietic porphyria (letter). N Engl J Med 1994; 330:1091–1092. 266. Tishler PV, Winston SH. Rapid improvement in the chemical pathology of congenital erythropoietic porphyria with treatment with superactivated charcoal. Methods Find Exp Clin Pharmacol 1990; 12:645–648. 267. Minder EI, Schneider-Yin X, Möll F. Lack of effect of oral charcoal in congenital erythropoietic porphyria (letter). N Engl J Med 1994; 330:1092–1094. 268. Watson CJ, Bossenmaier I, Cardinal R, Petryka ZJ. Repression by hematin of porphyrin biosynthesis in erythrocyte precursors in congenital erythropoietic porphyria. Proc Natl Acad Sci USA 1974; 71:278–282. 269. Varadi S. Haematological aspects in a case of erythropoietic porphyria. Br J Haematol 1958; 4:270–280. 270. Moore MR, Thompson GG, Goldberg A. The biosynthesis of haem in congenital (erythropoietic) porphyria. Int J Biochem 1978; 9:933–938. 271. Pimstone NR, Mukerji SK, Saicheur T, Lee J. Utility of chloroquine therapy in two patients with congenital erythropoietic porphyria (CEP). In: Blake D, McManus J, Ratnaike S, et al., eds. Porphyrins, porphyrias and photodynamic therapy (conference proceedings). Melbourne, Australia: 1993. 272. Tezcan I, Xu W, Gurgey A, et al. Congenital erythropoietic porphyria successfully treated by allogeneic bone marrow transplantation. Blood 1998; 92:4053–4058.
1424
273. Shaw PH, Mancini AJ, McConnell JP, et al. Treatment of congenital erythropoietic porphyria in children by allogeneic stem cell transplantation: a case report and review of the literature. Bone Marrow Transplant 2001; 27:101–105. 274. Dupuis-Girod S, Akkari V, Ged C, et al. Successful matchunrelated donor bone marrow transplantation for congenital erythropoietic porphyria (Gunther disease). Eur J Pediatr 2005; 164:104–107. 275. Moreau-Gaudry F, Barbot C, Mazurier F, et al. Correction of the enzyme deficit of bone marrow cells in congenital erythropoietic porphyria by retroviral gene transfer. Hematol Cell Ther 1996; 38:217–220. 276. Mazurier F, Moreau-Gaudry F, Salesse S, et al. Gene transfer of the uroporphyrinogen III synthase cDNA into haematopoietic progenitor cells in view of a future gene therapy in congenital erythropoietic porphyria. J Inherit Metab Dis 1997; 20:247–257. 277. Kauppinen R, Glass IA, Aizencang G, et al. Congenital erythropoietic porphyria: prolonged high-level expression and correction of the heme biosynthetic defect by retroviralmediated gene transfer into porphyric and erythroid cells. Mol Genet Metab 1998; 65:10–17. 278. Mazurier F, Geronimi F, Lamrissi-Garcia I, et al. Correction of deficient cd34(+) cells from peripheral blood after mobilization in a patient with congenital erythropoietic porphyria. Mol Ther 2001; 3:411–417. 279. Geronimi F, Richard E, Lamrissi-Garcia I, et al. Lentivirusmediated gene transfer of uroporphyrinogen III synthase fully corrects the porphyric phenotype in human cells. J Mol Med 2003; 81:310–320. 280. Elder GH. Porphyria cutanea tarda and related disorders. In: Kadish KM, Smith K, Guilard R, eds. Porphyrin handbook, part II, vol. 14. San Diego: Academic Press; 2003:67–92. 281. Barnes HD. Porphyria in the Bantu races on the Witwaterrand. S Afr Med J 1955; 29:781–784. 282. Cortés JM, Oliva H, Paradinas FJ, Hernandez-Guío C. The pathology of the liver in porphyria cutanea tarda. Histopathology 1980; 4:471–485. 283. Kordac V. Frequency of occurrence of hepatocellular carcinoma in patients with porphyria cutanea tarda in long-term followup. Neoplasma 1972; 19:135–139. 284. James KR, Cortes JM, Paradinas FJ. Demonstration of intracytoplasmic needle-like inclusions in hepatocytes of patients with porphyria cutanea tarda. J Clin Pathol 1980; 33:899–900. 285. Enerback L, Lundvall O. Properties and distribution of liver fluorescence in porphyria cutenua tarda (PCT). Virchows Arch A Pathol Pathol Anat 1970; 350:293–302. 286. Brady JJ, Jackson HA, Roberts AG, et al. Co-inheritance of mutations in the uroporphyrinogen decarboxylase and hemochromatosis genes accelerates the onset of porphyria cutanea tarda. J Invest Dermatol 2000; 115:868–874. 287. Anderson KE, Goeger DE, Carson RW, et al. Erythropoietin for the treatment of porphyria cutanea tarda in a patient on longterm hemodialysis. N Engl J Med 1990; 322:315–317. 288. Hebert AA, Farmer KL, Poh-Fitzpatrick MB. Peritoneal dialysis does not reduce serum porphyrin levels in porphyria cutanea tarda. Nephron 1992; 60:240. 289. Taljaard JJF, Shanley BC, Deppe WM, Joubert SM. Porphyrin metabolism in experimental hepatic siderosis in the rat. II. Combined effect of iron overload and hexachlorobenzene. Br J Haematol 1972; 23:513–519. 290. Taljaard JJF, Shanley BC, Deppe WM, Joubert SM. Porphyrin metabolism in experimental hepatic siderosis in the rat. III. Effect of iron overload and hexachlorobenzene on the liver haem biosynthesis. Br J Haematol 1972; 23:587–593. 291. Lynch RE, Lee GR, Kushner JP. Porphyria cutanea tarda associated with disinfectant misuse. Arch Intern Med 1975; 135:549–552.
Chapter 72 THE PORPHYRIAS
292. Doss M, Sauer H, Von Tiepermann R, Colombi AM. Development of chronic hepatic porphyria (porphyria cutanea tarda) with inherited uroporphyrinogen decarboxylase deficiency under exposure to dioxin. Int J Biochem 1984; 16:369–373. 293. McConnell R, Anderson K, Russell W, et al. Angiosarcoma, porphyria cutanea tarda, and probable chloracne in a worker exposed to waste oil contaminated with 2,3,7,8tetrachlorodibenzo-p-dioxin. Br J Ind Med 1993; 50:699–703. 294. Whitby FG, Phillips JD, Kushner JP, Hill CP. Crystal structure of human uroporphyrinogen decarboxylase. Embo J 1998; 17:2463–2471. 295. Dubart A, Mattei MG, Raich N, et al. Assignment of human uroporphyrinogen decarboxylase (URO-D) to the p34 band of chromosome 1. Hum Genet 1986; 73:277–279. 296. Roméo PH, Raich N, Dubart A, et al. Molecular cloning and nucleotide sequence of a complete human uroporphyrinogen decarboxylase cDNA. J Biol Chem 1986; 261:9825–9831. 297. Smith A. Porphyria caused by chlorinated AH receptor ligands and associated mechanisms of liver injury and cancer. In: Kadish KM, Smith K, Guilard R, eds. Porphyrin handbook, part II, vol. 14. San Diego: Academic Press; 2003:169–210. 298. Elder GH, Sheppard DM, Tovey JA, Urquhart AJ. Immunoreactive uroporphyrinogen decarboxylase in porphyria cutanea tarda. Lancet 1983; 1:1301–1304. 299. Phillips JD, Jackson LK, Bunting M, et al. A mouse model of familial porphyria cutanea tarda. Proc Natl Acad Sci USA 2001; 98:259–264. 300. Shanley BC, Zail SS, Joubert SM. Effect of ethanol on liver daminolaevulinate synthetase activity and urinary porphyrin excretion in symptomatic porphyria. Br J Haematol 1969; 17:389–396. 301. Pigatto PD, Polenghi MM, Altomare GF, et al. Complement cleavage products in the phototoxic reaction of porphyria cutanea tarda. Br J Dermatol 1986; 114:567–573. 302. Nomura N, Lim HW, Levin JL, Sassa S. Effect of soluble complement receptor type 1 on porphyrin-induced phototoxicity in guinea pigs. J Photochem Photobiol B 1998; 42:28–31. 303. Held JL, Sassa S, Kappas A, Harber LC. Erythrocyte uroporphyrinogen decarboxylase activity in porphyria cutanea tarda: a study of 40 consecutive patients. J Invest Dermatol 1989; 93:332–334. 304. Garey JR, Franklin KF, Brown DA, et al. Analysis of uroporphyrinogen decarboxylase complementary DNAs in sporadic porphyria cutanea tarda. Gastroenterology 1993; 105:165–169. 305. Mukerji SK, Pimstone NR. Reduced substrate affinity for human erythrocyte uroporphyrinogen decarboxylase constitutes the inherent biochemical defect in porphyria cutanea tarda. Biochem Biophys Res Commun 1985; 127:517–525. 306. Mukerji SK, Pimstone NR. Evidence of two uroporphyrinogen decarboxylase isoenzymes in human erythrocytes. Biochem Biophys Res Commun 1987; 146:1196–1203. 307. Elder GH, Urquhart AJ, de Salamanca RE, et al. Immunoreactive uroporphyrinogen decarboxylase in the liver in porphyria cutanea tarda. Lancet 1985; 2:229–233. 308. Roberts AG, Elder GH, Newcombe RG, et al. Heterogeneity of familial porphyria cutanea tarda. J Med Genet 1988; 25:669–676. 309. Kennedy SW, Wigfield DC. Dose–response relationships in hexachlorobenzene-induced porphyria. Biochem Pharmacol 1990; 40:1381–1388. 310. Chaufan G, del Carmen Rios de Molina M, San Martin de Viale LC. How does hexachlorobenzene treatment affect liver uroporphyrinogen decarboxylase? Int J Biochem Cell Biol 2001; 33:621–630. 311. Sweeney GD. Porphyria cutanea tarda, or the uroporphyrinogen decarboxylase deficiency diseases. Clin Biochem 1986; 19:3– 15.
312. Constantin D, Francis JE, Akhtar RA, et al. Uroporphyria induced by 5-aminolaevulinic acid alone in Ahrd SWR mice. Biochem Pharmacol 1996; 52:1407–1413. 313. De Matteis F. Role of iron in the hydrogen peroxide-dependent oxidation of hexahydroporphyrins (porphyrinogens): a possible mechanism for the exacerbation by iron of hepatic uroporphyria. Mol Pharmacol 1988; 33:463–469. 314. Jacobs JM, Sinclair PR, Lambrecht RW, Sinclair JF. Effects of iron-EDTA on uroporphyrinogen oxidation by liver microsomes. FEBS Lett 1989; 250:349–352. 315. Lambrecht RW, Sinclair PR, Gorman N, Sinclair JF. Uroporphyrinogen oxidation catalyzed by reconstituted cytochrome P450IA2. Arch Biochem Biophys 1992; 294:504–510. 316. Sinclair PR, Gorman N, Dalton T, et al. Uroporphyria produced in mice by iron and 5-aminolaevulinic acid does not occur in Cyp1a2(-/-) null mutant mice. Biochem J 1998; 330:149–153. 317. Smith AG, Clothier B, Carthew P, et al. Protection of the Cyp1A2(-/-) null mouse against uroporphyria and hepatic injury following exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol Appl Pharmacol 2001; 173:89–98. 318. Sinclair PR, Bement WJ, Lambrecht RW, et al. Uroporphyria caused by acetone and 5-aminolevulinic acid in iron-loaded mice. Biochem Pharmacol 1989; 38:4341–4344. 319. Legault N, Sabik H, Cooper SF, Charbonneau M. Effect of estradiol on the induction of porphyria by hexachlorobenzene in the rat. Biochem Pharmacol 1997; 54:19–25. 320. Anderson KE, Vadhanavikit S, Goeger DE. Estrogen and antioxidants in hexachlorobenzene-induced hepatic porphyria in rats. Hepatology 1995; 22:378A (abstract). 321. Sinclair PR, Gorman N, Walton HS, et al. Ascorbic acid inhibition of cytochrome P450-catalyzed uroporphyrin accumulation. Arch Biochem Biophys 1993; 304:464–470. 322. Egger NG, Goeger DE, Payne DA, et al. Porphyria cutanea tarda: multiplicity of risk factors including HFE mutations, hepatitis C, and inherited uroporphyrinogen decarboxylase deficiency. Dig Dis Sci 2002; 47:419–426. 323. D’Alessandro Gandolfo L, Griso D, Macri A, et al. Iron and porphyria cutanea tarda. Cell Mol Biol (Noisy-le-grand) 1997; 43:75–79. 324. Grossman ME, Bickers DR, Poh-Fitzpatrick MB, et al. Porphyria cutanea tarda. Clinical features and laboratory findings in 40 patients. Am J Med 1979; 67:277–286. 325. Turnbull A, Baker H, Vernon-Roberts B, Magnus IA. Iron metabolism in porphyria cutanea tarda and in erythropoietic protoporphyria. Q J Med 1973; 42:341–355. 326. Lundvall O, Weinfeld A, Lundin P. Iron storage in porphyria cutanea tarda. Acta Med Scand 1970; 188:37–53. 327. Lundvall O. The effect of replenishment of iron stores after phlebotomy therapy in porphyria cutanea tarda. Acta Med Scand 1971; 189:51–63. 328. Felsher BF, Jones ML, Redeker AG. Iron and hepatic uroporphyrin synthesis. Relation in porphyria cutanea tarda. JAMA 1973; 226:663–665. 329. Roberts AG, Whatley SD, Morgan RR, et al. Increased frequency of haemochromatosis Cys282Tyr mutation in sporadic porphyria cutanea tarda. Lancet 1997; 349:321–323. 330. Jackson H, Whatley SD, Morgan RR, et al. The haemochromatosis C282Y mutation in sporadic and familial porphyria cutanea tarda. Acta Haematol 1997; 98 S1:114 (abstract). 331. Stuart KA, Busfield F, Jazwinska EC, et al. The C282Y mutation in the haemochromatosis gene (HFE) and hepatitis C virus infection are independent cofactors for porphyria cutanea tarda in Australian patients. J Hepatol 1998; 28:404–409. 332. Bonkovsky HL, Poh-Fitzpatrick M, Pimstone N, et al. Porphyria cutanea tarda, hepatitis C, and HFE gene mutations in North America. Hepatology 1998; 27:1661–1669.
1425
Section XII. Inherited and Pediatric Diseases of the Liver
333. Fargion S, Fracanzani AL, Romano R, et al. Genetic hemochromatosis in Italian patients with porphyria cutanea tarda: possible explanation for iron overload. J Hepatol 1996; 24:564–569. 334. Sampietro M, Piperno A, Lupica L, et al. High prevalence of the His63Asp HFE mutation in Italian patients with porphyria cutanea tarda. Hepatology 1998; 27:181–184. 335. Enriquez de Salamanca R, Morales P, Castro MJ, et al. The most frequent HFE allele linked to porphyria cutanea tarda in Mediterraneans is His63Asp (letter). Hepatology 1999; 30:819–820. 336. Dereure O, Aguilar-Martinez P, Bessis D, et al. HFE mutations and transferrin receptor polymorphism analysis in porphyria cutanea tarda: a prospective study of 36 cases from southern France. Br J Dermatol 2001; 144:533–539. 337. Sinclair PR, Gorman N, Walton HS, et al. Uroporphyria in HFE mutant mice given 5-aminolevulinate: a new model of Femediated porphyria cutanea tarda. Hepatology 2001; 33:406–412. 338. Bulaj ZJ, Phillips JD, Ajioka RS, et al. Hemochromatosis genes and other factors contributing to the pathogenesis of porphyria cutanea tarda. Blood 2000; 95:1565–1571. 339. Chuang TY, Brashear R, Lewis C. Porphyria cutanea tarda and hepatitis C virus: a case-control study and meta-analysis of the literature. J Am Acad Dermatol 1999; 41:31–36. 340. Fargion S, Piperno A, Cappellini MD, et al. Hepatitis-C virus and porphyria cutanea tarda – evidence of a strong association. Hepatology 1992; 16:1322–1326. 341. DeCastro M, Sanchez J, Herrera JF, et al. Hepatitis-C virus antibodies and liver disease in patients with porphyria cutanea tarda. Hepatology 1993; 17:551–557. 342. Navas S, Bosch O, Castillo I, et al. Porphyria cutanea tarda and hepatitis C and B viruses infection: a retrospective study. Hepatology 1995; 21:279–284. 343. Stolzel U, Kostler E, Koszka C, et al. Low prevalence of hepatitis C virus infection in porphyria cutanea tarda in Germany. Hepatology 1995; 21:1500–1503. 344. Siersema PD, ten Kate FJ, Mulder PG, Wilson JH. Hepatocellular carcinoma in porphyria cutanea tarda: frequency and factors related to its occurrence. Liver 1992; 12:56–61. 345. Lamoril J, Andant C, Bogard C, et al. Epidemiology of hepatitis C and G in sporadic and familial porphyria cutanea tarda. Hepatology 1998; 27:848–852. 346. Van Gelder W, Siersema PD, Voogd A, et al. The effect of desferrioxamine on iron metabolism and lipid peroxidation in hepatocytes of C57BL/10 mice in experimental uroporphyria. Biochem Pharmacol 1993; 46:221–228. 347. Brudieux E, de Ledinghen V, Moran MJ, et al. Hepatic porphyrin concentration and uroporphyrinogen decarboxylase activity in hepatitis C virus infection. J Viral Hepat 2001; 8:41–47. 348. Moran MJ, Fontanellas A, Brudieux E, et al. Hepatic uroporphyrinogen decarboxylase activity in porphyria cutanea tarda patients: the influence of virus C infection. Hepatology 1998; 27:584–589. 349. Wissel PS, Sordillo P, Anderson KE, et al. Porphyria cutanea tarda associated with the acquired immune deficiency syndrome. Am J Hematol 1987; 25:107–113. 350. Boisseau AM, Couzigou P, Forestier JF, et al. Porphyria cutanea tarda associated with human immunodeficiency virus infection. A study of four cases and review of the literature. Dermatologica 1991; 182:155–159. 351. McAlister F, McClean K, Hamilton PG, Houston S. Human immunodeficiency virus infection and porphyria cutanea tarda: coexistence of risk factors or causative association? Clin Infect Dis 1995; 20:348–351.
1426
352. Drobacheff C, Derancourt C, Van Landuyt H, et al. Porphyria cutanea tarda associated with human immunodeficiency virus infection. Eur J Dermatol 1998; 8:492–496. 353. Hift RJ, Corrigall AV, Hancock V, et al. Porphyria cutanea tarda: the etiological importance of mutations in the HFE gene and viral infection is population-dependent. Cell Mol Biol (Noisy-legrand) 2002; 48:853–859. 354. Doss M, Tiepermann RV, Stutz G, Teschke R. Uroporphyrinogen decarboxylase inhibition in rat liver after alcohol ingestion. IRCS Med Sci Pharmacol 1980; 8:562. 355. Cederbaum AI. Introduction – serial review: alcohol, oxidative stress and cell injury. Free Radic Biol Med 2001; 31:1524– 1526. 356. Cederbaum AI, Wu D, Mari M, Bai J. CYP2E1-dependent toxicity and oxidative stress in HepG2 cells. Free Radic Biol Med 2001; 31:1539–1543. 357. Sinclair PR, Gorman N, Trask HW, et al. Uroporphyria caused by ethanol in Hfe(-/-) mice as a model for porphyria cutanea tarda. Hepatology 2003; 37:351–358. 358. Gorman N, Ross KL, Walton HS, et al. Uroporphyria in mice: thresholds for hepatic CYP1A2 and iron. Hepatology 2002; 35:912–921. 359. Blekkenhorst GH, Pimstone NR, Webber BL, Eales L. Hepatic haem metabolism in porphyria cutanea tarda (PCT): enzymatic studies and their relation to liver ultrastructure. Ann Clin Res 1976; 8:108–121. 360. Sinclair PR, Gorman N, Tsyrlov IB, et al. Uroporphyrinogen oxidation catalyzed by human cytochromes P450. Drug Metab Dispos 1998; 26:1019–1025. 361. Nichols RC, Cooper S, Trask HW et al. Uroporphyrin accumulation in hepatoma cells expressing human or mouse CYP1A2: relation to the role of CYP1A2 in human porphyria cutanea tarda. Biochem Pharmacol 2003; 65:545–550. 362. Bulaj ZJ, Franklin MR, Phillips JD, et al. Transdermal estrogen replacement therapy in postmenopausal women previously treated for porphyria cutanea tarda. J Lab Clin Med 2000; 136:482–488. 363. Ekstrom G, Ingelman-Sundberg M. Rat liver microsomal NADPH-supported oxidase activity and lipid peroxidation dependent on ethanol-inducible cytochrome P-450 (P-450IIE1). Biochem Pharmacol 1989; 38:1313–1319. 364. Christiansen L, Bygum A, Jensen A, et al. Association between CYP1A2 polymorphism and susceptibility to porphyria cutanea tarda. Hum Genet 2000; 107:612–614. 365. Martinez-Fresno M, Henriques-Gil N, Garrido M-C, et al. Frequency of intron 1 C/A polymorphism of the CYP1A2 in Spanish patient with porphyria cutanea tarda. In: Meissner P, ed. Porphyrins and porphyrias 2005. Cape Town: 2005:32. 366. Gardlo K, Selimovic D, Bolsen K, et al. Cytochrome p450A1 polymorphisms in a Caucasian population with porphyria cutanea tarda. Exp Dermatol 2003; 12:843–848. 367. Sinclair PR, Gorman G, Shedlofsky SI, et al. Ascorbic acid deficiency in porphyria cutanea tarda. J Lab Clin Med 1997; 130:197–201. 368. Sinclair PR, Gorman N, Sinclair JF, et al. Ascorbic acid inhibits chemically induced uroporphyria in ascorbate-requiring rats. Hepatology 1995; 22:565–572. 369. Percy VA, Naidoo D, Joubert SM, Pegoraro RJ. Ascorbate status of patients with porphyria cutanea tarda symptomatica and its effect on porphyrin metabolism. S Afr J Med Sci 1975; 40:185–196. 370. Rocchi E, Stella AM, Cassanelli M, et al. Liposoluble vitamins and naturally occurring carotenoids in porphyria cutanea tarda. Eur J Clin Invest 1995; 25:510–514. 371. Sixel-Dietrich F, Doss M. Hereditary uroporphyrinogendecarboxylase deficiency predisposing porphyria cutanea tarda
Chapter 72 THE PORPHYRIAS
372.
373.
374.
375.
376. 377.
378.
379.
380. 381. 382.
383.
384.
385.
386. 387.
388.
389.
390.
391.
392.
(chronic hepatic porphyria) in females after oral contraceptive medication. Arch Dermatol Res 1985; 278:13–16. Saeed-uz-Zafar M, Gronewald WR, Bluhm GB. Co-existent Klinefelter’s syndrome, acquired cutaneous hepatic porphyria and systemic lupus erythematosus. Henry Ford Hosp Med J 1970; 18:227–232. Rizzardini M, Smith AG. Sex differences in the metabolism of hexachlorobenzene by rats and the development of porphyria in females. Biochem Pharmacol 1982; 31:3543–3548. Smith AG, Francis JE, Green JA, et al. Sex-linked hepatic uroporphyria and the induction of cytochromes P450IA in rats caused by hexachlorobenzene and polyhalogenated biphenyls. Biochem Pharmacol 1990; 40:2059–2068. Liehr JG. Vitamin C reduces the incidence and severity of renal tumors induced by estradiol or diethylstilbestrol. Am J Clin Nutr 1991; 54:S1256–S1260. Elder GH. Porphyrin metabolism in porphyria cutanea tarda. Semin Hematol 1977; 14:227–242. Moore MR, Thompson GG, Allen BR, et al. Plasma porphyrin concentrations in porphyria cutanea tarda. Clin Sci Mol Med 1973; 45:711–714. Moran MJ, Fontanellas A, Santos JL, Enriquez de Salamanca R. Correlation between levels of free and protein-bound plasma porphyrin and urinary porphyrins in porphyria cutanea tarda. Int J Biochem Cell Biol 1995; 27:585–588. Kushner JP, Lee GR, Nacht S. The role of iron in the pathogenesis of porphyria cutanea tarda: an in vitro model. J Clin Invest 1972; 51:3044–3051. Levin EY. Enzymatic properties of uroporphyrinogen III cosynthetase. Biochemistry 1971; 10:4669–4675. Dowdle E, Mustard P, Spong N, Eales L. The metabolism of [5-14C] d-aminolevulinic acid in normal and porphyric human subjects. Clin Sci 1968; 34:233–251. Aarsand AK, Brun A, Boman H, Sandberg S. Differentiation between sporadic and familial porphyria cutanea tarda (abstract). Physiol Res 2003; 52:1S. Poh-Fitzpatrick M. Porphyria, pseudoporphyria, pseudopseudoporphyria . . . ? Arch Dermatol 1986; 122:403–404. Topi GC, Amantea A, Griso D. Recovery from porphyria cutanea tarda with no specific therapy other than avoidance of hepatic toxins. Br J Dermatol 1984; 3:75–82. Ramsay CA, Magnus IA, Turnbull A, Barker H. The treatment of porphyria cutanea tarda by venesection. Q J Med 1974; 43:1–24. Ippen H. Treatment of porphyria cutanea tarda by phlebotomy. Semin Hematol 1977; 14:253–259. Rocchi E, Gibertini P, Cassanelli M, et al. Serum ferritin in the assessment of liver iron overload and iron removal therapy in porphyria cutanea tarda. J Lab Clin Med 1986; 107:36–42. Ratnaike S, Blake D, Campbell D, et al. Plasma ferritin levels as a guide to the treatment of porphyria cutanea tarda by venesection. Aust J Dermatol 1988; 29:3–7. Adjarov D, Ivanov E. Clinical value of serum g-glutamyl transferase estimation in porphyria cutanea tarda. Br J Dermatol 1980; 102:541–543. Chlumsky J, Malina L, Chlumská A. The effect of venesection therapy on liver tissue in porphyria cutanea tarda. Acta HepatoGastroenterol 1973; 20:124–130. Rocchi E, Cassanelli M, Ventura E. High weekly intravenous doses of desferrioxamine in porphyria cutanea tarda. Br J Dermatol 1987; 117:393–396. Rocchi E, Gibertini P, Cassanelli M, et al. Iron removal therapy in porphyria cutanea tarda: phlebotomy versus slow subcutaneous desferrioxamine infusion. Br J Dermatol 1986; 114:621–629.
393. Timonen K, Niemi KM, Mustajoki P. Skin morphology in porphyria cutanea tarda does not improve despite clinical remission. Clin Exp Dermatol 1991; 16:355–358. 394. Bruce AJ, Ahmed I. Childhood-onset porphyria cutanea tarda: successful therapy with low-dose hydroxychloroquine (Plaquenil). J Am Acad Dermatol 1998; 38:810–814. 395. Freesemann A, Frank M, Sieg I, Doss MO. Treatment of porphyria cutanea tarda by the effect of chloroquine on the liver. Skin Pharmacol 1995; 8:156–161. 396. Taljaard JJF, Shanley BC, Stewart-Wynne EG, et al. Studies on low dose chloroquine therapy and the action of chloroquine in symptomatic porphyria. Br J Dermatol 1972; 87:261–269. 397. Kordac V, Semradova M. Treatment of porphyria cutanea tarda with chloroquine. Br J Dermatol 1974; 90:95–100. 398. Ashton RE, Hawk JLM, Magnus IA. Low-dose oral chloroquine in the treatment of porphyria cutanea tarda. Br J Dermatol 1984; 3:609–613. 399. Thornsvard MAJCT, Guider BA, Kimball DB. An unusual reaction to chloroquine-primaquine. JAMA 1976; 235:1719–1720. 400. Sweeney GD, Jones KG. Porphyria cutanea tarda: clinical and laboratory features. Can Med Assn J 1979; 120:803–807. 401. Malkinson FD, Levitt L. Hydroxychloroquine treatment of porphyria cutanea tarda. Arch Dermatol 1980; 116: 1147–1150. 402. Stolzel U, Kostler E, Schuppan D, et al. Hemochromatosis (HFE) gene mutations and response to chloroquine in porphyria cutanea tarda. Arch Dermatol 2003; 139:309–313. 403. Egger NG, Goeger DE, Anderson KE. Effects of chloroquine in hematoporphyrin-treated animals. Chemico-Biological Interact 1996; 102:69–78. 404. Cohen SN, Phifer KO, Yielding KL. Complex formation between chloroquine and ferrihaemic acid in vitro, and its effect on the antimalarial action of chloroquine. Nature 1964; 202:805–806. 405. Scholnick PL, Epstein J, Marver HS. The molecular basis of the action of chloroquine in porphyria cutanea tarda. J Invest Dermatol 1973; 61:226–232. 406. Vizethum W, Dahlmann D, Bolsen K, Goerz G. Influence of chloroquine (Resochin) on hexachlorobenzene (HCB) induced porphyria of the rat. Arch Dermatol Res 1979; 264:125. 407. Felsher BF, Redeker AG. Effect of chloroquine on hepatic uroporphyrin metabolism in patients with porphyria cutanea tarda. Medicine 1966; 45:575–583. 408. Chlumska A, Chlumsky J, Malina L. Liver changes in porphyria cutanea tarda patients treated with chloroquine. Br J Dermatol 1980; 102:261–266. 409. Anderson KE. Approaches to treatment and prevention of human porphyries. In: Kadish KM, Smith K, Guilard R, eds. Porphyrin handbook, part II, vol. 14. San Diego: Academic Press; 2003:247–284. 410. Fernandez I, Castellano G, de Salamanca RE, et al. Porphyria cutanea tarda as a predictor of poor response to interferon alfa therapy in chronic hepatitis C. Scand J Gastroenterol 2003; 38:314–319. 411. Rossini A, Contessi GB, Leali C, et al. Efficacy of iron depletion and antiviral therapy in patients with porphyria cutanea tarda (PCT) and hepatitis C virus (HCV) chronic infection. Hepatology 2004;40 (suppl 1):320A (abstract). 412. Yaqoob M, Smyth J, Ahmad R, et al. Haemodialysis-related porphyria cutanea tarda and treatment by recombinant human erythropoietin. Nephron 1992; 60:428–431. 413. Shieh S, Cohen JL, Lim HW. Management of porphyria cutanea tarda in the setting of chronic renal failure: a case report and review. J Am Acad Dermatol 2000; 42:645–652. 414. Carson RW, Dunnigan EJ, DuBose TDJ, et al. Removal of plasma porphyrins with high-flux hemodialysis in porphyria
1427
Section XII. Inherited and Pediatric Diseases of the Liver
415.
416.
417.
418.
419.
420.
421.
422. 423.
424.
425.
426.
427.
428. 429. 430.
431. 432.
433.
1428
cutanea tarda associated with end-stage renal disease. J Am Soc Nephrol 1992; 2:1445–1450. Stevens BR, Fleischer AB, Piering F, Crosby DL. Porphyria cutanea tarda in the setting of renal failure: response to renal transplantation. Arch Dermatol 1993; 129:337–339. Roberts AG, Elder GH, De Salamanca RE, et al. A mutation (G281E) of the human uroporphyrinogen decarboxylase gene causes both hepatoerythropoietic porphyria and overt familial porphyria cutanea tarda: biochemical and genetic studies on Spanish patients. J Invest Dermatol 1995; 104:500–502. Moran-Jimenez MJ, Ged C, et al. Uroporphyrinogen decarboxylase: complete human gene sequence and molecular study of three families with hepatoerythropoietic porphyria. Am J Hum Genet 1996; 58:712–721. Meguro K, Fujita H, Ishida N, et al. Molecular defects of uroporphyrinogen decarboxylase in a patient with mild hepatoerythropoietic porphyria. J Invest Dermatol 1994; 102:681–685. Koszo F, Elder GH, Roberts A, Simon N. Uroporphyrinogen decarboxylase deficiency in hepatoerythropoietic porphyria: further evidence for genetic heterogeneity. Br J Dermatol 1990; 122:365–370. Fujita H, Sassa S, Toback AC, Kappas A. Immunochemical study of uroporphyrinogen decarboxylase in a patient with mild hepatoerythropoietic porphyria. J Clin Invest 1987; 79:1533–1537. De Verneuil H, Beaumont C, Deybach J-C, et al. Enzymatic and immunological studies of uroporphyrinogen decarboxylase in familial porphyria cutanea tarda and hepatoerythropoietic porphyria. Am J Hum Genet 1984; 36:613–622. Hift RJ, Meissner PN, Todd G. Hepatoerythropoietic porphyria precipitated by viral hepatitis. Gut 1993; 34:1632–1634. Kushner JP, Pimstone NR, Kjeldsberg CR, et al. Congenital erythropoietic porphyria, diminished activity of uroporphyrinogen decarboxylase and dyserythropoiesis. Blood 1982; 59:725–737. Toback A, Sassa S, Poh-Fitzpatrick MB, et al. Hepatoerythropoietic porphyria: clinical, biochemical, and enzymatic studies in a three-generation family lineage. N Engl J Med 1987; 316:645–650. Armstrong DK, Sharpe PC, Chambers CR, et al. Hepatoerythropoietic porphyria: a missense mutation in the UROD gene is associated with mild disease and an unusual porphyrin excretion pattern. Br J Dermatol 2004; 151:920–923. Pimstone NR, Gandhi SN, Mukerji SK. Therapeutic efficacy of oral charcoal in congenital erythropoietic porphyria. N Engl J Med 1987; 316:390–393. Fontanellas A, Mazurier F, Moreau-Gaudry F, et al. Correction of uroporphyrinogen decarboxylase deficiency (hepatoerythropoietic porphyria) in Epstein–Barr virustransformed B-cell lines by retrovirus-mediated gene transfer: fluorescence-based selection of transduced cells. Blood 1999; 94:465–474. Elder GH, Evans JO, Thomas N, et al. The primary enzyme defect in hereditary coproporphyria. Lancet 1976; 2:1217–1219. Berger H, Goldberg A. Hereditary coproporphyria. Br Med J 1955; 2:85–88. Nordmann Y, Grandchamp B, De Verneuil H, et al. Harderoporphyria: a variant hereditary coproporphyria. J Clin Invest 1983; 72:1139–1149. Grandchamp B, Phung N, Nordmann Y. Homozygous case of hereditary coproporphyria. Lancet 1977; 2:1348–1349. Brodie MJ, Thompson GG, Moore MR, et al. Hereditary coproporphyria. Demonstration of the abnormalities in haem biosynthesis in peripheral blood. Q J Med 1977; 46:229–241. Lamoril J, Puy H, Gouya L, et al. Neonatal hemolytic anemia due to inherited harderoporphyria: clinical characteristics and molecular basis. Blood 1998; 91:1453–1457.
434. Andrews J, Erdjument H, Nicholson DC. Hereditary coproporphyria: incidence in a large English family. J Med Genet 1984; 21:341–349. 435. Paxton JW, Moore MR, Beattie AD, Goldberg A. Urinary excretion of 17-oxosteroids in hereditary coproporphyria. Clin Sci Mol Med 1975; 49:441–444. 436. Lash TD, Hall T, Mani UN, Jones MA. Normal and abnormal heme biosynthesis. 3.(1) synthesis and metabolism of tripropionate analogues of coproporphyrinogen-III: novel probes for the active site of coproporphyrinogen oxidase. J Org Chem 2001; 66:3753–3759. 437. Medlock AE, Dailey HA. Human coproporphyrinogen oxidase is not a metalloprotein. J Biol Chem 1996; 271:32507–32510. 438. Martasek P, Camadro JM, Raman CS, et al. Human coproporphyrinogen oxidase. Biochemical characterization of recombinant normal and R231W mutated enzymes expressed in E. coli as soluble, catalytically active homodimers. Cell Mol Biol (Noisy-le-grand) 1997; 43:47–58. 439. Cacheux V, Martasek P, Fougerousse F, et al. Localization of the human coproporphyrinogen oxidase gene to chromosome band 3q12. Hum Genet 1994; 94:557–559. 440. Takahashi S, Taketani S, Akasaka JE et al. Differential regulation of coproporphyrinogen oxidase gene between erythroid and nonerythroid cells. Blood 1998; 92:3436–3444. 441. Martasek P. Hereditary coproporphyria. Semin Liver Dis 1998; 18:25–32. 442. Lamoril J, Puy H, Whatley SD, et al. Characterization of mutations in the CPO gene in British patients demonstrates absence of genotype–phenotype correlation and identifies relationship between hereditary coproporphyria and harderoporphyria. Am J Hum Genet 2001; 68:1130–1138. 443. Rosipal R, Lamoril J, Puy H, et al. Systematic analysis of coproporphyrinogen oxidase gene defects in hereditary coproporphyria and mutation update. Hum Mutat 1999; 13:44–53. 444. Martasek P, Nordmann Y, Grandchamp B. Homozygous hereditary coproporphyria caused by an arginine to tryptophane substitution in coproporphyrinogen oxidase and common intragenic polymorphisms. Hum Mol Genet 1994; 3:477–480. 445. Lamoril J, Martasek P, Deybach JC, et al. A molecular defect in coproporphyrinogen oxidase gene causing harderoporphyria, a variant form of hereditary coproporphyria. Hum Mol Genet 1995; 4:275–278. 446. Sasaki H, Kaneko K, Tseneyama H. Activities of daminolevulinic acid synthetase in the liver and bone marrow of hepatic coproporphyria (hereditary coproporphyria). Acta Med Biol (Niigata) 1969; 17:97–99. 447. McIntyre N, Pearson AJG, Allan DJ, et al. Hepatic daminolaevulinic acid synthetase in an attack of hereditary coproporphyria and during remission. Lancet 1971; 1:560–564. 448. Haeger-Aronsen B, Stathers G, Swahn G. Hereditary coproporphyria: study of a Swedish family. Ann Intern Med 1968; 69:221–227. 449. Kuhnel A, Gross U, Jacob K, Doss MO. Studies on coproporphyrin isomers in urine and feces in the porphyrias. Clin Chim Acta 1999; 282:45–58. 450. Blake D, McManus J, Cronin V, Ratnaike S. Fecal coproporphyrin isomers in hereditary coproporphyria. Clin Chem 1992; 38:96–100. 451. Moder KG. A coproporphyria-like syndrome induced by glipizide. Mayo Clin Proc 1991; 66:312–316. 452. Manning DJ, Gray TA. Haem arginate in acute hereditary coproporphyria. Arch Dis Child 1991; 66:730–731. 453. Hunter JAA, Khan SA, Hope E, et al. Hereditary coproporphyria. Photosensitivity, jaundice and neuropsychiatric manifestations associated with pregnancy. Br J Dermatol 1971; 84:301–310.
Chapter 72 THE PORPHYRIAS
454. Meissner P, Hift RJ, Corrigall A. Variegate porphyria. In: Kadish KM, Smith K, Guilard R, eds. Porphyrin handbook, part II, vol. 14. San Diego: Academic Press; 2003:93–120. 455. Dean G. The porphyrias: a study of inheritance and environment. London: Pitman Medical; 1971. 456. Tu J-B, Blackwell RQ, Feng Y-S. Clinical and biochemical studies of hereditary hepatic porphyria in Chinese subjects in Taiwan. Metabolism 1971; 20:629–641. 457. Grabczynska SA, McGregor JM, Hawk JL. Late onset variegate porphyria. Clin Exp Dermatol 1996; 21:353–356. 458. Hift RJ, Meissner PN. An analysis of 112 acute porphyric attacks in Cape Town, South Africa: evidence that acute intermittent porphyria and variegate porphyria differ in susceptibility and severity. Medicine (Baltimore) 2005; 84:48–60. 459. Whatley SD, Puy H, Morgan RR, et al. Variegate porphyria in western Europe: identification of PPOX gene mutations in 104 families, extent of allelic heterogeneity, and absence of correlation between phenotype and type of mutation. Am J Hum Genet 1999; 65:984–994. 460. Germanaud J, Luthier F, Causse X, et al. A case of association between hepatocellular carcinoma and porphyria variegata. Scand J Gastroenterol 1994; 29:671–672. 461. Tidman MJ, Higgins EM, Elder GH, MacDonald DM. Variegate porphyria associated with hepatocellular carcinoma. Br J Dermatol 1989; 121:503–505. 462. Quiroz-Kendall E, Wilson FA, King LE Jr. Acute variegate porphyria following a Scarsdale gourmet diet. J Am Acad Dermatol 1983; 8:46–49. 463. Ferriera GC, Dailey HA. Mouse protoporphyrinogen oxidase: kinetic parameters and demonstration of inhibition by bilirubin. Biochem J 1988; 250:597–603. 464. McColl KEL, Thompson GG, El Omar E, et al. Porphyrin metabolism and haem biosynthesis in Gilbert’s syndrome. Gut 1987; 28:125–130. 465. Duke SO, Lydon J, Becerril JS, et al. Protoporphyrinogen oxidase-inhibiting herbicides. Weed Sci 1991; 39:465–473. 466. Taketani S, Inazawa J, Abe T, et al. The human protoporphyrinogen oxidase gene (PPOX): organization and location to chromosome 1. Genomics 1995; 29:698–703. 467. Roberts AG, Whatley SD, Daniels J, et al. Partial characterization and assignment of the gene for protoporphyrinogen oxidase and variegate porphyria to human chromosome 1q23. Hum Mol Genet 1995; 4:2387–2390. 468. Nishimura K, Taketani S, Inokuchi H. Cloning of a human cDNA for protoporphyrinogen oxidase by complementation in vivo of a hemG mutant of Escherichia coli. J Biol Chem 1995; 270:8076–8080. 469. Puy H, Robreau AM, Rosipal R, et al. Protoporphyrinogen oxidase: complete genomic sequence and polymorphisms in the human gene. Biochem Biophys Res Commun 1996; 226:226–230. 470. Dailey TA, Dailey HA. Human protoporphyrinogen oxidase: expression, purification, and characterization of the cloned enzyme. Protein Sci 1996; 5:98–105. 471. Dailey TA, Dailey HA, Meissner P, Prasad AR. Cloning, sequence, and expression of mouse protoporphyrinogen oxidase. Arch Biochem Biophys 1995; 324:379–384. 472. Dailey TA, McManus JF, Dailey HA. Characterization of the mouse protoporphyrinogen oxidase gene. Cell Mol Biol (Noisyle-grand) 2002; 48:61–69. 473. Koch M, Breithaupt C, Kiefersauer R, et al. Crystal structure of protoporphyrinogen IX oxidase: a key enzyme in haem and chlorophyll biosynthesis. Embo J 2004; 23:1720–1728. 474. Brenner DA, Bloomer JR. The enzymatic defect in variegate porphyria: studies with human cultured skin fibroblasts. N Engl J Med 1980; 302:765–769.
475. Meissner PN, Dailey TA, Hift RJ, et al. A R59W mutation in human protoporphyrinogen oxidase results in decreased enzyme activity and is prevalent in South Africans with variegate porphyria. Nat Genet 1996; 13:95–97. 476. Corrigall AV, Hift RJ, Davids LM, et al. Identification of the first variegate porphyria mutation in an indigenous black South African and further evidence for heterogeneity in variegate porphyria. Mol Genet Metab 2001; 73:91–96. 477. Frank J, Aita VM, Ahmad W, et al. Identification of a founder mutation in the protoporphyrinogen oxidase gene in variegate porphyria patients from Chile. Hum Hered 2001; 51:160–168. 478. Frank J, Jugert FK, Merk HF, et al. A spectrum of novel mutations in the protoporphyrinogen oxidase gene in 13 families with variegate porphyria. J Invest Dermatol 2001; 116:821–823. 479. Meissner PN, Day RS, Moore MR, et al. Protoporphyrinogen oxidase and porphobilinogen deaminase in variegate porphyria. Eur J Clin Invest 1986; 16:257–261. 480. Long C, Smyth SJ, Woolf J, et al. Detection of latent variegate porphyria by fluorescence emission spectroscopy of plasma. Br J Dermatol 1993; 129:9–13. 481. Hift RJ, Davidson BP, van der Hooft C, et al. Plasma fluorescence scanning and fecal porphyrin analysis for the diagnosis of variegate porphyria: precise determination of sensitivity and specificity with detection of protoporphyrinogen oxidase mutations as a reference standard. Clin Chem 2004; 50:915–923. 482. Lim CK, Rideout JM, Peters TJ. Pseudoporphyria associated with consumption of brewers’ yeast. Br Med J 1984; 288:1640–1642. 483. Logan GM, Weimer MK, Ellefson M, et al. Bile porphyrin analysis in the evaluation of variegate porphyria. N Engl J Med 1991; 324:1408–1411. 484. Da Silva V, Simonin S, Deybach JC, et al. Variegate porphyria: diagnostic value of fluorometric scanning of plasma porphyrins. Clin Chim Acta 1995; 238:163–168. 485. Elder GH, Magnus IA, Handa F, Doyle M. Faecal “X porphyrin” in the hepatic porphyrias. Enzyme 1974; 17:29–38. 486. Timonen K, Mustajoki P, Tenhunen R, Lauharanta J. Effects of haem arginate on variegate porphyria. Br J Dermatol 1990; 123:381–387. 487. Hift RJ, Todd G, Meissner PN, Kirsch RE. Administration of oral activated charcoal in variegate porphyria results in a paradoxical clinical and biochemical deterioration. Br J Dermatol 2003; 149:1266–1269. 488. Magnus IA, Jarrett A, Prankert TAJ, Rimington C. Erythropoietic protoporphyria: a new porphyria syndrome with solar urticaria due to protoporphyrinaemia. Lancet 1961; 2:448–451. 489. Poh-Fitzpatrick MB. Erythropoietic porphyrias: current mechanistic, diagnostic, and therapeutic considerations. Semin Hematol 1977; 14:211–219. 490. Abitbol M, Bernex F, Puy H, et al. A mouse model provides evidence that genetic background modulates anemia and liver injury in erythropoietic protoporphyria. Am J Physiol Gastrointest Liver Physiol 2005; 288:G1208–G1216. 491. Straka JG, Hill HD, Krikava JM, et al. Immunochemical studies of ferrochelatase protein: characterization of the normal and mutant protein in bovine and human protoporphyria. Am J Hum Genet 1991; 48:72–78. 492. De Leo VA, Poh-Fitzpatrick M, Mathews-Roth M, Harber LC. Erythropoietic protoporphyria. 10 years experience. Am J Med 1976; 60:8–22. 493. Schmidt H, Snitker G, Thomsen K, Lintrup J. Erythropoietic protoporphyria. A clinical study based on 29 cases in 14 families. Arch Dermatol 1974; 110:58–64. 494. Rank JF, Carithers R, Bloomer J. Evidence for neurological dysfunction in end-stage protoporphyric liver disease. Hepatol 1993; 18:1404–1409.
1429
Section XII. Inherited and Pediatric Diseases of the Liver
495. Sandberg S, Romslo I. Porphyrin-induced photodamage at the cellular and the subcellular level as related to the solubility of the porphyrin. Clin Chim Acta 1981; 109:193–201. 496. Goldstein BD, Harber LC. Erythropoietic protoporphyria: lipid peroxidation and red cell membrane damage associated with photohemolysis. J Clin Invest 1972; 51:892–902. 497. Lim HW. Pathophysiology of cutaneous lesions in porphyrias. Semin Hematol 1989; 26:114–119. 498. Gigli I, Schothorst AA, Soter NA, Pathak MA. Erythropoietic protoporphyria: photoactivation of the complement system. J Clin Invest 1980; 66:517–522. 499. Timonen K, Kariniemi AL, Niemi KM, et al. Vascular changes in erythropoietic protoporphyria: histopathologic and immunohistochemical study. J Am Acad Dermatol 2000; 43:489–497. 500. Poh-Fitzpatrick MB. Human protoporphyria: long-term variance in protoporphyrin balance in RBC and fecal distribution compartments in 24 patients. J Invest Dermatol 1990;94:567. 501. Poh-Fitzpatrick MB. Human protoporphyria: reduced cutaneous photosensitivity and lower erythrocyte porphyrin levels during pregnancy. J Am Acad Dermatol 1997;36(1):40–3. 502. Bewley AP, Keefe M, White JE. Erythropoietic protoporphyria improving during pregnancy. Br J Dermatol 1998; 139: 145–147. 503. Poh-Fitzpatrick MB, Palmer RH. Elevated plasma triglyceride levels are associated with human protoporphyria. J Lab Clin Med 1984; 104:257–263. 504. Doss MO, Frank M. Hepatobiliary implications and complications in protoporphyria, a 20-year study. Clin Biochem 1989; 22:223–229. 505. Bloomer JR. The liver in protoporphyria. Hepatology 1988; 8:402–407. 506. Singer JA, Plaut AG, Kaplan MM. Hepatic failure and death from erythropoietic protoporphyria. Gastroenterology 1978; 74:588–591. 507. Bonkovsky HL, Schned AR. Fatal liver failure in protoporphyria: synergism between ethanol excess and the genetic defect. Gastroenterology 1986; 90:191–201. 508. Poh-Fitzpatrick MB, Whitlock RT, Lefkowitch JH. Changes in protoporphyrin distribution dynamics during liver failure and recovery in a patient with protoporphyria and Epstein–Barr viral hepatitis. Am J Med 1986; 80:943–950. 509. Key NS, Rank JM, Freese D, et al. Hemolytic anemia in protoporphyria: possible precipitating role of liver failure and photic stress. Am J Hemat 1992; 39:202–207. 510. Camadro JM, Ibraham NG, Levere RD. Kinetic studies of human liver ferrochelatase. Role of endogenous metals. J Biol Chem 1984; 259:5678–5682. 511. Gamble JT, Wong SG, Dailey HA, Marks GS. Use of recombinant human ferrochelatase as a sensitive bioassay for N-alkylprotoporphyrin IX formed after interaction of porphyrinogenic xenobiotics with rat liver microsomes. Can J Physiol Pharmacol 2000; 78:578–581. 512. Burden AE, Wu C, Dailey TA, et al. Human ferrochelatase: crystallization, characterization of the [2Fe-2S] cluster and determination that the enzyme is a homodimer. Biochim Biophys Acta 1999; 1435:191–197. 513. Dailey HA, Finnegan MG, Johnson MK. Human ferrochelatase is an iron sulfur protein. Biochemistry 1994; 33:403–407. 514. Dailey HA, Sellers VM, Dailey TA. Mammalian ferrochelatase: expression and characterization of normal and two human protoporphyric ferrochelatases. J Biol Chem 1994; 269:390–395. 515. Taketani S, Tanaka-Yoshioka A, Masaki R, et al. Association of ferrochelatase with complex I in bovine heart mitochondria. Biochim Biophys Acta 1986; 883:277–283. 516. Whitcombe DM, Albertson DG, Cox TM. Molecular analysis of functional and nonfunctional genes for human ferrochelatase:
1430
517.
518.
519.
520.
521.
522.
523.
524.
525.
526. 527.
528.
529.
530.
531.
532.
533.
534.
535.
isolation and characterization of a FECH pseudogene and its sublocalization on chromosome 3. Genomics 1994; 20:482–486. Nakahashi Y, Taketani S, Okuda M, et al. Molecular cloning and sequence analysis of cDNA encoding human ferrochelatase. Biochem Biophys Res Commun 1990; 173:748–755. Chan RYY, Schulman HM, Ponka P. Expression of ferrochelatase mRNA in erythroid and non-erythroid cells. Biochem J 1993; 292:343–349. Cox TM. Protoporphyria. In: Kadish KM, Smith K, Guilard R, eds. Porphyrin handbook, part II, vol. 14. San Diego: Academic Press; 2003:121–149. Rufenacht UB, Gouya L, Schneider-Yin X, et al. Systematic analysis of molecular defects in the ferrochelatase gene from patients with erythropoietic protoporphyria. Am J Hum Genet 1998; 62:1341–1352. Todd DJ, Hughes AE, Ennis KT, et al. Identification of a single base pair deletion (40 del G) in exon 1 of the ferrochelatase gene in patients with erythropoietic protoporphyria. Hum Mol Genet 1993; 2:1495–1496. Wang X, Yang L, Kurtz L, et al. Haplotype analysis of families with erythropoietic protoporphyria and novel mutations of the ferrochelatase gene. J Invest Dermatol 1999; 113:87–92. Gouya L, Puy H, Robreau AM, et al. The penetrance of dominant erythropoietic protoporphyria is modulated by expression of wildtype FECH. Nat Genet 2002; 30:27–28. Bloomer J, Wang Y, Singhal A, Risheg H. Molecular studies of liver disease in erythropoietic protoporphyria. J Clin Gastroenterol 2005; 39:S167–S175. Lamoril J, Boulechfar S, De Verneuil H, et al. Human erythropoietic protoporphyria: two point mutations in the ferrochelatase gene. Biochem Biophys Res Commun 1991; 181:594–599. Went LN, Klasen EC. Genetic aspects of erythropoietic protoporphyria. Ann Hum Genet 1984; 48:105–117. Brenner DA, Didier JM, Frasier F, et al. A molecular defect in human protoporphyria. Am J Hum Genet 1992; 50:1203–1210. Najahi-Missaoui W, Dailey HA. Production and characterization of erythropoietic protoporphyric heterodimeric ferrochelatases. Blood 2005; 106:1098–1104. Ohgari Y, Sawamoto M, Yamamoto M, et al. Ferrochelatase consisting of wild-type and mutated subunits from patients with a dominant-inherited disease, erythropoietic protoporphyria, is an active but unstable dimer. Hum Mol Genet 2005; 14:327–334. Bloomer J, Bruzzone C, Zhu L, et al. Molecular defects in ferrochelatase in patients with protoporphyria requiring liver transplantation. J Clin Invest 1998; 102:107–114. Minder EI, Gouya L, Schneider-Yin X, Deybach JC. A genotype–phenotype correlation between null-allele mutations in the ferrochelatase gene and liver complication in patients with erythropoietic protoporphyria. Cell Mol Biol (Noisy-le-grand) 2002; 48:91–96. Risheg H, Chen FP, Bloomer JR. Genotypic determinants of phenotype in North American patients with erythropoietic protoporphyria. Mol Genet Metab 2003; 80:196–206. Bottomley SS, Tanaka M, Everett MA. Diminished erythroid ferrochelatase activity in protoporphyria. J Lab Clin Med 1975; 86:126–131. Clark KGA, Nicholson DC. Erythrocyte protoporphyrin and iron uptake in erythropoietic protoporphyria. Clin Sci 1971; 41:363–370. Piomelli S, Lamola AA, Poh-Fitzpatrick MB, et al. Erythropoietic protoporphyria and Pb intoxication: the molecular basis for difference in cutaneous photosensitivity. I. Different rates of disappearance of protoporphyrin from the erythrocytes, both in vivo and in vitro. J Clin Invest 1975; 56:1519–1527.
Chapter 72 THE PORPHYRIAS
536. Sassaroli M, Dacosta R, Vaananen H, et al. Distribution of erythrocyte free porphyrin content in erythropoietic protoporphyria. J Lab Clin Med 1992; 120:614–623. 537. Nicholson DC, Cowger ML, Kalivas J, et al. Isotopic studies of the erythropoietic and hepatic components of congenital porphyria and “erythropoietic” protoporphyria. Clin Sci 1973; 44:135–150. 538. Scholnick P, Marver HS, Schmid R. Erythropoietic protoporphyria: evidence for multiple sites of excess protoporphyrin formation. J Clin Invest 1971; 50: 203–207. 539. Bonkowsky HL, Bloomer JR, Ebert PS, Mahoney MJ. Heme synthetase deficiency in human protoporphyria. Demonstration of the defect in liver and cultured skin fibroblasts. J Clin Invest 1975; 56:1139–1148. 540. Sassa S, Zalar GL, Poh-Fitzpatrick MB, et al. Studies in porphyria: functional evidence for a partial deficiency of ferrochelatase activity in mitogen-stimulated lymphocytes from patients with erythropoietic protoporphyria. J Clin Invest 1982; 69:809–815. 541. Pawliuk R, Tighe R, Wise RJ, et al. Prevention of murine erythropoietic protoporphyria-associated skin photosensitivity and liver disease by dermal and hepatic ferrochelatase. J Invest Dermatol 2005; 124:256–262. 542. Berenson MM, Kimura R, Samowitz W, Bjorkman D. Protoporphyrin overload in unrestrained rats: biochemical and histopathologic characterization of a new model of protoporphyric hepatopathy. Int J Exp Pathol 1992; 73:665– 673. 543. Knobler E, Poh-Fitzpatrick MB, Kravetz D, et al. Interaction of hemopexin, albumin and liver fatty acid-binding protein with protoporphyrin. Hepatology 1989; 10:995–997. 544. Porter FS, Lowe BA. Congenital erythropoietic protoporphyria. I. Case reports, clinical studies and porphyrin analyses in two brothers. Blood 1963; 22:521–531. 545. Shirota T, Yamamoto H, Hayashi S, et al. Myelodysplastic syndrome terminating in erythropoietic protoporphyria after 15 years of aplastic anemia. Int J Hematol 2000; 72:44–47. 546. Aplin C, Whatley SD, Thompson P, et al. Late-onset erythropoietic porphyria caused by a chromosome 18q deletion in erythroid cells. J Invest Dermatol 2001; 117:1647–1649. 547. Goodwin RG, Kell WJ, Laidler P, et al. Photosensitivity and acute liver injury in myeloproliferative disorder secondary to late-onset protoporphyria caused by deletion of a ferrochelatase gene in hematopoietic cells. Blood 2006; 107:60–62. 548. Hastka J, Lasserre JJ, Schwarzbeck A, et al. Zinc protoporphyrin in anemia of chronic disorders. Blood 1993; 81:1200–1204. 549. Sandberg S, Brun A. Light-induced protoporphyrin release from erythrocytes in erythropoietic protoporphyria. J Clin Invest 1982; 70:693–698. 550. Sandberg S, Talstad I, Hovding G, Bjelland N. Light-induced release of protoporphyrin, but not of zinc protoporphyrin, from erythrocytes in a patient with greatly elevated erythrocyte protoporphyrin. Blood 1983; 62:846–851. 551. Poh-Fitzpatrick MB, DeLeo VA. Rates of plasma porphyrin disappearance in fluorescent vs. red incandescent light exposure. J Invest Dermatol 1977; 69:510–512. 552. Suurmond D. Some aspects of erythropoietic protoporphyria in the Netherlands. Dermatologica 1969; 138:303–311. 553. Rademakers LHPM, Koningsberger JC, Sorber CWJ, et al. Accumulation of iron in erythroblasts of patients with erythropoietic protoporphyria. Eur J Clin Invest 1993; 23:130–138. 554. Morton KO, Schneider F, Weimer MK, et al. Hepatic and bile porphyrins in patients with protoporphyria and liver failure. Gastroenterology 1988; 94:1488–1492.
555. Poh-Fitzpatrick MB. Protoporphyrin metabolic balance in human protoporphyria. Gastroenterology 1985; 88:1239– 1242. 556. Mathews-Roth MM, Pathak MA, Fitzpatrick TB, et al. Betacarotene as a photoprotective agent in erythropoietic protoporphyria. N Engl J Med 1970; 282:1231–1234. 557. Mathews-Roth MM, Pathak MA, Fitzpatrick TB, et al. Beta carotene therapy for erythropoietic protoporphyria and other photosensitivity diseases. Arch Dermatol 1977; 113:1229–1332. 558. Thomsen K, Schmidt H, Fischer A. Beta-carotene in erythropoietic protoporphyria: 5 years’ experience. Dermatologica 1979; 159:82–86. 559. Mathews-Roth MM, Rosner B. Long-term treatment of erythropoietic protoporphyria with cysteine. Photodermatol Photoimmunol Photomed 2002; 18:307–309. 560. Fusaro RM, Runge WJ. Erythropoietic protoporphyria. IV. Protection from sunlight. Br Med J 1970; 1:730–731. 561. Warren LJ, George S. Erythropoietic protoporphyria treated with narrow-band (TL-01) UVB phototherapy. Aust J Dermatol 1998; 39:179–182. 562. Kniffen JC. Protoporphyrin removal in intrahepatic porphyrastasis. Gastroenterology 1970; 58:1027. 563. Van Hattum J, Baart de la Faille H, Van den Berg JWO, et al. Chenodeoxycholic acid therapy in erythrohepatic protoporphyria. J Hepatol 1986; 3:407–412. 564. Gross U, Frank M, Doss MO. Hepatic complications of erythropoietic protoporphyria. Photodermatol Photoimmunol Photomed 1998; 14:52–57. 565. Gordeuk VR, Brittenham GM, Hawkins CW, et al. Iron therapy for hepatic dysfunction in erythropoietic protoporphyria. Ann Intern Med 1986; 105:27–31. 566. Mercurio MG, Prince G, Weber FL, et al. Terminal hepatic failure in erythropoietic protoporphyria. J Am Acad Dermatol 1993; 29:829–833. 567. Van Wijk HJ, Van Hattum J, Delafaille HB, et al. Blood exchange and transfusion therapy for acute cholestasis in protoporphyria. Dig Dis Sci 1988; 33:1621–1625. 568. Todd DJ, Callender ME, Mayne EE, et al. Erythropoietic protoporphyria, transfusion therapy and liver disease. Br J Dermatol 1992; 127:534–537. 569. Nordmann Y. Erythropoietic protoporphyria and hepatic complications. J Hepatol 1992; 16:4–6. 570. Muley SA, Midani HA, Rank JM, et al. Neuropathy in erythropoietic protoporphyrias. Neurology 1998; 51: 262–265. 571. Fontanellas A, Mazurier F, Landry M, et al. Reversion of hepatobiliary alterations by bone marrow transplantation in a murine model of erythropoietic protoporphyria. Hepatology 2000; 32:73–81. 572. Poh-Fitzpatrick MB, Wang X, Anderson KE, et al. Erythropoietic protoporphyria: altered phenotype after bone marrow transplantation for myelogenous leukemia in a patient heteroallelic for ferrochelatase gene mutations. J Am Acad Dermatol 2002; 46:861–866. 573. Richard E, Robert E, Cario-Andre M, et al. Hematopoietic stem cell gene therapy of murine protoporphyria by methylguanineDNA-methyltransferase-mediated in vivo drug selection. Gene Ther 2004; 11:1638–1647. 574. Day RS, Eales L, Meissner D. Coexistent variegate porphyria and porphyria cutanea tarda. N Engl J Med 1982; 30:36–41. 575. Doss M. Dual porphyria in double heterozygotes with porphobilinogen deaminase and uroporphyrinogen decarboxylase deficiencies. Clin Genet 1989; 35:146–151. 576. Nordmann Y, Amram D, Deybach JC, et al. Coexistent hereditary coproporphyria and congenital erythropoietic porphyria (Günther disease). J Inherit Metab Dis 1990; 13:687–691.
1431
Section XII. Inherited and Pediatric Diseases of the Liver
577. Freesemann AG, Hofweber K, Doss MO. Coexistence of deficiencies of uroporphyrinogen III synthase and decarboxylase in a patient with congenital erythropoietic porphyria and in his family. Eur J Clin Chem Clin Biochem 1997; 35:35–9. 578. Gregor A, Kostrzewska E, Tarczynska-Nosal S, Stachurska H. Coexistence of hereditary coproporphyria with acute intermittent porphyria. Ann Med 1994; 26:125–127.
1432
579. Akagi R, Inoue R, Muranaka S, et al. Dual gene defects involving d-aminolaevulinate dehydratase and coproporhyrinogen oxidase in a porphyria patient. Br J Hematol 2006; 132:237–243. 580. Poh-Fitzpatrick MB. Is porphyria cutanea tarda a paraneoplastic disorder? Clin Dermatol 1993; 11:119–124.
Section XII: Inherited and Pediatric Diseases of the Liver
73
PEDIATRIC VIRAL HEPATITIS Scott A. Elisofon and Maureen M. Jonas Abbreviations AAP american academy of pediatrics ACIP advisory committee on immunization practices ALT alanine aminotransferase CDC centers for disease control CMV cytomegalovirus EBV Epstein–Barr virus ELU ELISA-linked units HAI histologic activity index
HAV HBeAg HBsAg HBV HCC HCV HDV HEV HHV-6
hepatitis A virus hepatitis B e antigen hepatitis B surface antigen hepatitis B virus hepatocellular carcinoma hepatitis C virus hepatitis D virus hepatitis E virus human herpes virus-6
INTRODUCTION Viral hepatitis is a systemic infection with predominant involvement in the liver. The most common hepatotropic viruses seen in the pediatric population are hepatitis A, hepatitis B, and hepatitis C. Other less common hepatotropic viruses in children are hepatitis D and hepatitis E. These viruses can lead to a wide array of clinical entities, including asymptomatic infection, fulminant hepatitis, and chronic hepatitis. It is important to recognize that the epidemiology, natural history, clinical features, and treatment options are different in children compared with adults, and differ with age groups in children. This recognition has led to prophylaxis regimens and newer treatment protocols for children and adolescents. In addition to the hepatotropic viruses, other viruses contribute to the viral hepatitis seen in children. These viruses are more age-specific, including herpes simplex virus, enteroviruses, and adenovirus in infants, as well as parvovirus B19, Epstein–Barr virus (EBV), and others in children and adolescents. Beside viral etiologies, the differential diagnosis for acute hepatitis in children is different from that in adults, and should be considered during evaluation (Table 73-1).
HEPATITIS A IN CHILDREN INTRODUCTION Infection due to hepatitis A virus (HAV) continues to be a widely reported disease in the USA, but numbers are continuing to decrease with the development and implementation of vaccination for high-risk individuals. This is important because of the greater risk of serious disease in adults older than 50 years and patients with underlying liver disease. Although children are responsible for many cases and much of the transmission of hepatitis A, they are less ill and use fewer medical-care dollars for their care. Seven percent of children under 15 years of age are hospitalized, in comparison to 27% of adults older than 45 years.1
HIV IFN-a IG IgM LKM1 PCR SPLIT VCA
human immunodeficiency virus interferon-a immunoglobulin immunoglobulin M liver–kidney microsomal antibody type 1 polymerase chain reaction studies of pediatric liver transplantation viral capsid antigen
EPIDEMIOLOGY As in adults, HAV is primarily acquired by the fecal–oral route. The virus is found in the stool and blood of an infected individual for up to 2–3 weeks before clinical symptoms, and it can persist in stool from 1 to 2 weeks after symptom onset. HAV may be transmitted by personal contact or water/food ingestion. Transfusion-acquired HAV from viremic donors is very rare because this phase of the infection is very brief. Of 685 children younger than 15 years with acute hepatitis A infection in 2001, 28.6% acquired HAV from international travel, 19% from household or sexual contacts, 6.7% from daycare, and 2.9% from contact with a daycare attendee or worker. Other sources of infection were unknown (34.9%), contact with hepatitis A patient (5%), food- or water-borne outbreak (2.2%), homosexual activity (0.4%), and intravenous drug use (0.3%).2 Hepatitis A is reported in all age groups, but the highest incidence is in children 5–14 years of age, at a reported rate of 15–20 per 100 000 population.3 Infections in children under 5 years are less commonly reported, at a rate of 10 per 100 000 population. Children under 15 years account for approximately one-third of all reported cases each year.4 These numbers are most likely underestimates, as many young children have asymptomatic infections or show only non-specific clinical features. In the USA the incidence of hepatitis A varies with race, socioeconomic status, and location. American Indians and Alaskan natives have the highest rates of infection, followed by Hispanics, non-Hispanic Caucasians, non-Hispanic blacks, and Asians. In addition, there are 11 states (Alaska, Arizona, California, Idaho, Nevada, New Mexico, Oklahoma, Oregon, South Dakota, Utah, and Washington) with high rates of hepatitis A infection, defined as more than 20 cases per 100 000 population. These differences are most likely caused by close living conditions, poor socioeconomic status, and exposure to persons from endemic areas. In these areas, 30–40% of children under age 5 years have been exposed to HAV and most individuals are infected before adulthood.
1433
Section XII. Inherited and Pediatric Diseases of the Liver
Table 73-1. Differential Diagnosis of Acute Hepatitis in Infants and Children Neonate aged <6 months
Childhood >6 months
Viral Hepatitis B Herpes simplex virus Adenovirus Enteroviruses Human herpes virus-6 Metabolic Neonatal hemochromatosis Tyrosinemia Galactosemia Fatty acid oxidation defects Mitochondrial disorders Other metabolic disorders
Viral Hepatitis A Hepatitis B Hepatitis C Hepatitis E Epstein–Barr virus Parvovirus B19 Other viruses Hemophagocytic syndrome Medications Toxins/herbs Autoimmune hepatitis Type I Type II Metabolic Fatty acid oxidation defects Mitochondrial disorders Other metabolic disorders
PATHOGENESIS The pathogenesis of HAV is still not completely known, although it is thought to be immune-mediated hepatic injury. This hypothesis has been supported by active replication of HAV in cell cultures without cell death,5 lack of complement-dependent antibodymediated cytolytic activity in sera,6 and evidence of HAV-specific cytotoxic T lymphocytes in the liver.7 Pathogenesis is probably the same in children as in adults, although children typically have milder disease with this infection.
CLINICAL FEATURES Infections in infants and young children may be completely asymptomatic, or children may have gastroenteritis symptoms. Jaundice is rare in this age group, which makes diagnosing acute HAV infection very difficult. Older children and adolescents, like adults, may have a prodrome of fever, headache, and malaise for several days. These symptoms are followed by jaundice, abdominal pain, nausea, vomiting, and anorexia. In acute cases of hepatitis A reported to the Centers for Disease Control (CDC) in 2001, 75% of children aged 5–15 were jaundiced.2 As with adults, hyperbilirubinemia will most likely normalize within 4 weeks. On physical exam, older children may be dehydrated, jaundiced, and have a mildly enlarged, tender liver. Splenomegaly is a rare finding. Aminotransferases are usually elevated at the time of jaundice and can range from 20 to 100 times the upper limit of normal. They usually improve significantly within 2–3 weeks. Atypical manifestations of HAV seem to be less common in children. Cholestatic hepatitis, with jaundice lasting more than 12 weeks and significant pruritus, and biphasic or relapsing hepatitis are uncommon in children, but have been reported.8,9 Extrahepatic manifestations such as cutaneous vasculitis,10 arthritis, and cryoglobulinemia have been reported but are also rare in children.11 Pancreatitis has been reported during acute HAV infection.12,13
1434
DIAGNOSIS The diagnosis of HAV infection in children is made as in adults, with detection of immunoglobulin M (IgM) to HAV (IgM anti-HAV). This antibody is present in serum approximately 5–10 days before the onset of symptoms and can remain for up to 6 months. Immunoglobulin G to HAV (IgG anti-HAV) appears early and will remain for the individual’s lifetime, conferring long-term immunity.
DIFFERENTIAL DIAGNOSIS In infants and children with acute HAV, clinical symptoms of viral gastroenteritis are common. Differential diagnosis includes viral infections such as rotavirus or other enteroviruses, and bacterial or parasitic enteritis. Aminotransferases and bilirubin are not usually measured in children with these symptoms. In older children with acute hepatitis symptoms, other viral etiologies, such as EBV, acute hepatitis B, and acute hepatitis C, must be considered. Other possible infections include herpes simplex virus and cytomegalovirus (CMV) in immunocompromised patients, as well as non-A–E hepatitis, which frequently leads to fulminant hepatic failure. Etiologies such as parvovirus B19, autoimmune hepatitis, and medication/toxin hepatotoxicity must be considered in the appropriate clinical setting.
PROGNOSIS AND NATURAL HISTORY Acute HAV infection resolves spontaneously and fully in most cases. Few children have the atypical manifestations, as described above. The most significant complication is fulminant hepatic failure causing death or requiring transplantation. The frequency of fulminant hepatic failure from acute hepatitis A is quite low, estimated to be between 0.1 and 0.4%.2 The risk is higher in children under 5 years of age14 and patients with hepatitis C infection.15 In 2001, there were 3 reported deaths in the USA due to acute hepatitis A infection: one was a child under 5 years of age.2 According to the Studies of Pediatric Liver Transplantation (SPLIT) registry that collects data from 37 North American pediatric liver transplant centers, only 2 (0.1%) of the 1378 pediatric transplants between 1995 and June 2003 were for fulminant hepatitis A.16
TREATMENT Treatment of acute hepatitis A infection is purely supportive. For young children with vomiting or diarrhea, close monitoring of hydration is important. Any evidence of altered mental status or bleeding must be followed to monitor for development of fulminant hepatic failure.
IMMUNOPROPHYLAXIS Prophylaxis for hepatitis A can be given via two methods: vaccine or immunoglobulin (IG). This decision depends on multiple factors. For travel to endemic areas, children under 2 years requiring preexposure prophylaxis should receive immune globulin intramuscularly (0.02 ml/kg), as vaccine is not yet approved for this age group. If the child is at least 2 years old, vaccination is appropriate if there are more than 2–4 weeks before travel. For postexposure prophylaxis in unvaccinated children, IG should be given within 2 weeks of exposure – 0.02 ml/kg intramuscularly.17 Testing for anti-HAV is not needed for these children prior to the IG.
Chapter 73 PEDIATRIC VIRAL HEPATITIS
Effective vaccines for HAV infection have been available since 1995, yet there continues to be discussion regarding which children should be immunized. As mentioned previously, the HAV vaccine is approved for use in children 2 years of age and older. Children aged 2–17 should receive two doses (an initial dose and one 6–12 months later) of either Havrix (SmithKline Beecham Biologicals) 720 ELISA-linked units (ELU) or Vaqta (Merck) 25 units. In 1999, after reviewing data from 1987 to 1997, the CDC suggested routine vaccination for children living in areas of the USA that had HAV incidence rates of at least 20 per 100 000 (Alaska, Arizona, California, Idaho, Nevada, New Mexico, Oklahoma, Oregon, South Dakota, Utah, and Washington) and consideration of vaccination in states (Arkansas, Colorado, Missouri, Montana, Texas, Wyoming) and counties with more than 10 per 100 000 during that time period.17 Since these recommendations have been implemented, the incidence in these states is comparable to that in states in the northeast. The number of reported acute hepatitis A infections in 2001 was lower than ever before, with the greatest decline in children aged 5–14 years.2 Despite the recommendations of the Advisory Committee on Immunization Practices (ACIP), many pediatricians and hepatologists support a strategy of universal hepatitis A vaccination, with cost–benefit analyses supporting this policy.18,19 Other recommendations for vaccination in children include gay or bisexual adolescent men, adolescents using illegal drugs, and patients with clotting disorders.20 Vaccination should be considered for children attending daycare centers with recurrent outbreaks or children living in custodial care institutions. In addition to healthy children and adolescents, those with chronic liver disease or pre/post liver transplantation should be vaccinated. Recent studies have demonstrated good immunogenicity and safety with HAV vaccination in children with chronic liver disease.21,22
DAILY ACTIVITIES Children, like adults, should be excluded from activities, daycare, or school for at least 1 week after the onset of clinical symptoms. For those in close contact personal hygiene is important to prevent secondary cases.
CONCLUSIONS Hepatitis A is an acute hepatitis that is frequently asymptomatic or well tolerated in the pediatric population. Although most children recover from this illness, vaccinations can prevent sequelae such as fulminant hepatitis or hepatitis in adult contacts. Vaccination programs have already been implemented in certain areas of the USA, and universal vaccination programs are being considered.
HEPATITIS B Hepatitis B virus (HBV) infection is a worldwide health issue for all populations, but children have different epidemiology, modes of transmission, clinical features, and natural progression of the disease. Because of these differences, monitoring and treatment decisions are not the same for children and adults.
HBV include densely populated areas of Africa and Asia. Vertical transmission from mother to infant, either in utero or around the time of delivery, is a dominant mode of HBV acquisition in these areas. Vertical transmission accounts for 40–50% of the transmission in Asia, and chronic HBV infections in these endemic populations occur before 2 years of age.23 The risk of a newborn becoming infected by a hepatitis B surface antigen (HBsAg)-positive, hepatitis B e antigen (HBeAg)-positive mother is 85–90%.24 Vertical transmission is important to recognize, as it carries the highest risk of chronic HBV and the lowest risk of acute symptomatic hepatitis. Vertical transmission is an important issue to address in hepatitis B prevention, since it has been shown that the likelihood of chronic infection is inversely proportional to the age at acquisition.25 Horizontal transmission, from child to child, is the main mode of transmission for children in the USA, since perinatal transmission is prevented with neonatal IG administration and vaccination. Children at risk for horizontal transmission include those living in households with chronically infected individuals and children living in communities in which HBV is highly endemic. Adolescents at risk for infection are those who engage in high-risk behavior, including unprotected sexual activity and intravenous drug use. The primary children at risk for hepatitis B infection in this country are immigrant or adopted children from HBV-endemic regions of the world. Data collected by the CDC over the last 20 years have shown the declining incidence of hepatitis B infection in the USA, secondary to routine screening of pregnant women, wide use of the HBV vaccines, and changing practices among intravenous drug users. Children are a small proportion of the cases in the USA, with 10% occurring in patients aged 11–19, and 8% occurring between ages 0 and 10.26 Before 1982, when the first vaccine became widely available, at least 20 000 US children were infected annually.27 From 1986 to 2000, the rate of acute hepatitis B infection in children aged 1–9 years declined by more than 80%, from 0.9 to 0.13 per 100 000 children aged 1–9 years.28 Between 1990 and 2002, acute hepatitis B had declined by 89% among young people aged 0–19 years.29 These data demonstrate a decline in acute HBV cases in childhood, but do not reflect new pediatric cases due to immigration of alreadyinfected children, and chronic asymptomatic infections that are due to perinatal or early childhood infections in high-risk groups.
PATHOGENESIS The primary difference in pathogenesis between adults and children is reflected in the percentage of children who become chronically infected. Most infants who acquire the virus perinatally have some sort of immune tolerance that permits chronic infection with high rates of viral replication and minimal immune response. These children usually have minimal hepatic inflammation or mild hepatitis. Immune tolerance may be induced by transplacental exposure to HBeAg in utero. Absence of T-cell responses to HBcAg has been demonstrated in children born to HBsAg-positive, HBeAg-positive mothers, but an active response of T cells to HBcAg has been shown in infants with acute hepatitis B born to HBeAg-negative mothers.30
CLINICAL FEATURES EPIDEMIOLOGY The prevalence of HBV and its modes of transmission differ significantly in populations around the world. High-prevalence areas of
There is a wide spectrum of clinical manifestations of acute and chronic hepatitis B infection in children. Manifestations are different for neonates compared to older children.
1435
Section XII. Inherited and Pediatric Diseases of the Liver
Vertically infected neonates are usually asymptomatic, with normal aminotransferases. However, fulminant hepatitis B, albeit rare in the USA, has been reported and is more common in endemic areas such as Taiwan, and in infants born to HBeAg-negative mothers.31–33 Children with fulminant hepatitis B present with jaundice and abnormal aminotransferases by 2–3 months of age. The mortality of this condition is very high.34 Young children and adolescents may also have asymptomatic infections, with the diagnosis discovered through routine laboratory tests. When symptoms do occur, they may include a prodrome consisting of fatigue, malaise, nausea, low-grade fever, or a serum sickness-like reaction. Within a week or two of the prodrome, children may develop an icteric hepatitis. During this icteric phase, children may experience nausea, vomiting, or pruritus. Physical exam findings may include tender hepatomegaly or splenomegaly. In addition to constitutional symptoms, children with acute HBV infections will occasionally exhibit extrahepatic manifestations. Most of these symptoms are attributed to immune complexes. As stated earlier, children may exhibit a serum sickness-like illness, consisting of arthralgia or arthritis, urticaria or angioedema, and a maculopapular rash. These symptoms usually improve with the onset of jaundice. Young children will rarely develop Gianotti–Crosti syndrome, which includes papular acrodermatitis of the face, extremities, and trunk, with lymphadenopathy. This may be the only clinical sign of hepatitis B infection, but can be seen with other viral infections as well.35 Chronic hepatitis B infection in children is usually asymptomatic. The children usually grow well and are clinically healthy. Children with chronic hepatitis B infection occasionally develop membranoproliferative glomerulonephritis or nephrotic syndrome,36–38 but other extrahepatic manifestations are rare. HBV-associated glomerular disease has been shown to improve with antiviral therapy.39,40 Chronic HBV infection in children may not be manifest until symptoms of cirrhosis or hepatocellular carcinoma (HCC) develop.
DIAGNOSIS The diagnosis of acute HBV infection in children is made as it is in adults, by detection of HBsAg, and IgM antibody to hepatitis B core (anti-HBc). Chronic HBV infection is defined by the presence of HBsAg in serum for more than 6 months. Chronic infections are further characterized as with or without active viral replication, depending on the detection of HBeAg or HBV DNA in serum (Chapter 31). Most chronic HBV infections in children are associated with high levels of viral replication, leading to very high levels of viremia. ALT values may either be normal, reflecting lack of significant immune response, or elevated, indicating hepatocellular injury.
DIFFERENTIAL DIAGNOSIS Acute Hepatitis Acute hepatitis B in neonates is rare, but is still occasionally seen. Neonatal hepatitis is more commonly seen with herpes simplex virus,41 echoviruses,42 and adenoviruses43 (Table 73-1). These neonates are usually quite ill, with coagulopathy and other evidence of fulminant hepatic failure. The differential diagnosis of acute hepatitis in toddlers, children, and adolescents includes infection with
1436
hepatitis A, hepatitis C, EBV, and parvovirus B1944 as well as noninfectious disorders (Table 73-1). Acute hepatitis due to superinfection of children with chronic HBV infection by hepatitis D virus (HDV) is uncommon.
Chronic Hepatitis In addition to chronic HBV, chronic aminotransferase elevations in childhood may be due to hepatitis C, a1-antitrypsin deficiency, tyrosinemia, cystic fibrosis, or other metabolic disorders such as amino acid or carbohydrate metabolism defects or fatty acid oxidation defects (Table 73-2). Other causes of chronic hepatitis in children include Wilson’s disease, autoimmune hepatitis, and medication hepatotoxicity.
ASSOCIATED CONDITIONS In addition to the clinical features discussed earlier, co-infection with HDV, human immunodeficiency virus (HIV), or hepatitis C virus (HCV) is rare, but does occur. A 1985 report from Italy, an endemic region for HDV, indicated that 13 of 102 (12.7%) HBVinfected Italian children had HDV co-infection.45 However, HDV is seen with decreasing frequency, even in endemic areas. Recent data regarding prevalence of HDV infection in children are not available.
DISEASE COMPLICATIONS PERTINENT TO INFECTION ACQUIRED IN CHILDHOOD As discussed earlier, extrahepatic manifestations in children are uncommon, with glomerular disease being the only condition reported.36–38 One center in Turkey reported 14 children with HBVrelated glomerulonephritis over 20 years.38 Of greatest concern in infants with perinatally acquired disease is severe hepatitis with fulminant hepatic failure. Since implementation of universal HBV vaccination programs, the incidence of ful-
Table 73-2. Differential Diagnosis of Chronic Hepatitis in Children Viral Hepatitis B Hepatitis C Hepatitis D Autoimmune hepatitis Type I Type II Non-alcoholic fatty liver disease Primary sclerosing cholangitis Hepatotoxicity due to medications or toxins Alpha1-antitrypsin deficiency Wilson’s disease Hereditary hemochromatosis Celiac disease Cystic fibrosis Metabolic disorders Fatty acid oxidation defects Mitochondrial disorders Tyrosinemia Carbohydrate metabolism Amino acid metabolism Bile acid synthesis disorders Others
Chapter 73 PEDIATRIC VIRAL HEPATITIS
minant hepatitis B has significantly decreased in children. In Taiwan in 1998, the infant mortality rate from HBV infection had declined to 1.71 per 100 000.46 In the USA during 2001, there were zero deaths from acute hepatitis B in children under 15 years.2 It has been shown in the USA, Taiwan, and Japan that HBV-associated fulminant hepatic failure is more common in infants born to HBeAgnegative or anti-HBe-positive mothers.32,33 The mortality rate from HBV fulminant hepatic failure is as high as 61–77%.34
PROGNOSIS AND NATURAL HISTORY Age at infection is the most important factor in determining the risk of chronic hepatitis B infection in children.25 As stated previously, close to 90% of infants infected during the perinatal period will develop chronic infection. This is in comparison to the development of chronic infection in 25–50% of children infected between ages 1 and 5 years, and 6–10% of older children.47 In Asia, infants with perinatally acquired HBV will have high levels of HBV DNA and continue to be HBeAg-positive into late childhood. These children usually have minimal liver disease and normal alanine aminotransferase (ALT).48 Spontaneous seroconversion from HBeAg to anti-HBe occurs in less than 2% per year in children under 3 years, and 4–5% per year in children over 3 years.49 In contrast to the Asian population, long-term studies have been conducted in Italy and Spain showing the outcomes of chronic hepatitis B in Caucasian children. One series of 76 Italian children followed longitudinally for 1–12 years (mean 5 years) showed that HBeAg-positive children seroconverted to anti-HBe and lost HBV DNA at a mean annual rate of 16%.50 This represents an overall 70% rate of seroconversion and loss of HBV DNA. Those patients who lost HBeAg had higher ALT values, indicating more active liver disease. Five of the 76 children cleared HBsAg.50 Another outcome study conducted in Italy and Spain included 185 children with chronic hepatitis B and no other underlying chronic disorders. Ninety-one percent of these children were HBeAg-positive.51 The mean age of these children was 5 years, and only 14% were known to have contracted hepatitis B perinatally. Five children (3%) had cirrhosis at entrance in the study. In followup that averaged 13 years, more than 80% of the HBeAg-positive children developed anti-HBe seroconversion and normal ALT before adulthood. Of the children who presented originally with anti-HBe, 88% achieved sustained normal ALT.51 The differences in the rate of early seroconversion are most likely due to the age of infection of the Italian and Spanish children. Later infection is associated with a significantly better immune response. Liver histology in HBeAg-positive children usually shows mild inflammation and fibrosis. This inflammation is usually milder than in adults.52 Cirrhosis is an uncommon finding, but if present, usually develops early and is often noted on the initial biopsy.51 Of greatest concern for long-standing hepatitis B infection is the risk of HCC and its poor prognosis, with less than 30% 5-year survival.53 This is a rare cancer in children without HBV or an inborn error of metabolism known to be associated with HCC. Pediatric HCC associated with chronic HBV has been described in both Asian and western populations.54,55 HCC is usually thought to occur after decades of chronic infection, although studies have shown a significant incidence of childhood HCC in HBV-endemic areas.56 Before hepatitis B vaccine was available, the annual incidence of HCC was 0.7 per 100 000 chil-
dren aged 6–14 in Taiwan, and 0.05 for Caucasians and 0.02 for blacks per 100 000 children under 15 years in the USA.56,57 Subsequent studies from Taiwan have demonstrated a significant decrease in HCC incidence: from 0.70 per 100 000 children from 1981 to 1986, 0.57 from 1986 to 1990, and 0.36 from 1990 to 1994.56 A recent report describes 426 Taiwanese children with chronic hepatitis B infection who were prospectively followed for 5.1–27.2 years (median, 14.9 years). During 6250 person-years, 2 boys developed HCC, with an incidence of 32 per 100 000 person-years. The diagnosis was made in these patients at age 11 and 14 years. Of note, both children had early HBeAg seroconversion (< 3 years of age).58 Early seroconversion or the presence of anti-HBe in patients with HCC has been shown in other studies in both Asian and western populations. Many of these children have had rapid development of cirrhosis.59 These data suggest that early HBeAg seroconversion, especially in the presence of early cirrhosis, is an important risk factor for HCC in childhood. No specific recommendations for HCC screening in children have been made, but many practitioners will obtain annual a-fetoprotein levels and, beginning in later childhood or adolescence, periodic hepatic ultrasounds in children with chronic HBV.
TREATMENT Acute hepatitis B has no proven treatment, and care is purely supportive. There has been no proven role for antiviral medications during acute or fulminant infection in children. As discussed above, children have different immune tolerance and different rates and types of progressive liver disease. Because of this, studies in adults cannot be extrapolated to infants and children. The treating physician must take into consideration the patient’s age, age at acquisition, and level of immune reactivity. General principles of management of these patients are listed in Table 73-3. Once aminotransferases are documented to remain greater than 1.5–2 times the upper limit of normal for at least 3 months, treatment can be considered. Treatment candidates should have evidence for active HBV replication, such as positive HBeAg or HBV DNA
Table 73-3. Management of Chronic Hepatitis B in Children 1. Prove chronic infection by at least two HBsAg-positive samples 6 months apart or HBsAg-positive with anti-HBc (not IgM) 2. Measure ALT every 6 months in children older than 2 years 3. Yearly HBeAg and anti-HBe in patients with normal ALT to detect spontaneous seroconversion 4. Liver biopsy, and consider treatment for children older than 2 years with ALT >1.5–2 ¥ upper limit of normal for more than 3 months and no evidence of seroconversion 5. Regular physical exams for evidence of chronic liver disease 6. Immunize all household contacts 7. Immunize all hepatitis B patients with hepatitis A vaccine 8. Yearly serum alpha-fetoprotein and liver ultrasounda HBsAg, hepatitis B surface antigen; anti-HBc, antibody to hepatitis B core; IgM, immunoglobulin M; ALT, alanine aminotransferase; HBeAg, hepatitis Be antigen; anti-HBe, antibody to hepatitis Be antigen. a No data to support optimal age to begin liver ultrasound surveillance and optimal interval between ultrasounds. Reproduced from Broderick A, Jonas MM. Management of hepatitis B in children. In: Tran TT, Martin P, eds. Clinics in liver disease. Philadelphia, PA: WB Saunders; 2004: 387–401.155
1437
Section XII. Inherited and Pediatric Diseases of the Liver
of at least 4 log (measured by quantitative polymerase chain reaction (PCR) assays) detected in serum. It is crucial to ensure that elevated aminotransferases are not indicative of active seroconversion, since in this case treatment is not indicated or necessary. This is done by following the ALT values and HBeAg for at least 4–6 months before initiating therapy. Prior to any treatment plan, a liver biopsy should be considered to confirm that hepatitis B is the etiology of the hepatitis, and to quantify the degree of inflammation and fibrosis. Currently, the two medications approved by the US Food and Drug Administration for the treatment of chronic hepatitis B in children 2–18 years of age are interferon-a (IFN-a) and lamivudine. Interferon is the only medication approved for use in European children at this time.
Interferon-a IFN-a has been evaluated in separate populations as well as in large, multinational studies. It is tolerated relatively well in children, with a good safety profile. Data from western countries show loss of HBV DNA or HBeAg seroconversion to be 20–58% in IFN-treated children compared to 8–17% in controls. These data differed from that from Chinese populations, with responses ranging from 3 to 17% HBeAg seroconversion or loss of HBV DNA.60–62 Originally, these differences were thought to be secondary to the early age of infection in the Chinese population as well as genetic and immunologic factors. More recently, these differences are attributed to the inclusion of large numbers of children with normal or near-normal ALT values in these studies. A multinational randomized controlled trial included 144 children with chronic hepatitis B and ALT values averaging greater than twice the upper limit of normal. In comparison to no treatment, 26% of children who received IFN-a at 6 MU/m2 thrice weekly for 24 weeks lost HBeAg and HBV DNA compared to 11% of controls. HBsAg became negative in 10% of treated patients and 1% of controls. In contrast to previous studies, there was no significant difference in response rates between Asians (22%) and non-Asians (26%). Response rates appeared to be higher in patients under 13 years, females, and those with low levels of HBV DNA.63 Factors that have been shown to be associated with response to IFN-a in children are listed in Table 73-4. Early side effects of IFN-a in children include flu-like symptoms of fever, myalgia, headache, arthralgia, and anorexia. Bone marrow suppression, specifically neutropenia, has been shown to occur as frequently as 39% during the first month of treatment. Some children require dose adjustments, but most do not require discontinuation.64 Children may have personality changes, irritability, and temper tantrums.63,64 Weight gain and height velocity may be compromised in children during therapy.65,66 Data from a multinational study indicated that children undergoing treatment have temporary delays in both growth and weight gain in comparison to control children with chronic hepatitis B. Height velocity and weight gain improved in the 6 months after treatment, becoming similar to those of control patients with chronic hepatitis B.65
Lamivudine Lamivudine (2¢,3¢dideoxycytosine) is an orally administered nucleoside analog, making it an attractive option in children. Pharmaco-
1438
Table 73-4. Variables Associated with Virologic Response to Interferona63,156–160 or Lamivudine68,69 in Children with Chronic Hepatitis B Interferon-a Associated with higher likelihood of response ALT ≥ twice the upper limit of normal Female gender Low level of hepatitis B virus DNA Age less than 13 years, if ALT ≥ 2 ¥ upper limit of normal Active inflammation on liver biopsya No association with likelihood of response Ethnicity Body surface area Lamivudine Associated with higher likelihood of response Elevated baseline ALT High baseline histologic activity index score No association with likelihood of response Baseline hepatitis B virus DNA level Race/ethnicity Age Gender Previous interferon therapy Baseline weight Baseline body mass index ALT, alanine aminotransferase. a Shown to be associated with higher response in adults, and presumed to be similar in children.
kinetics and safety of this drug were originally studied in 53 European children with chronic hepatitis B infection.67 Proper dosing was determined to be 3 mg/kg up to the adult maximum of 100 mg/day. A randomized, double-blind, placebo-controlled multicenter trial of lamivudine for 52 weeks was reported in 2002.68 In this study, 191 children aged 2–17 years were assigned to receive lamivudine and 97 to placebo. All children had been HBsAg-positive for at least 6 months, were positive for HBeAg, anti-HBe-negative, and had detectable HBV DNA and ALT greater than 1.3 times the upper limit of normal but less than 500 IU/l. Twenty-three percent of children who received lamivudine became HBeAg-negative and had undetectable HBV DNA after 52 weeks, in comparison to 13% of those who had received placebo. Sustained normalization of ALT level was achieved in 55% of patients treated with lamivudine and 12% of children treated with placebo. Eighty-two percent of virologic responses were sustained for 6 months after treatment. Logistic regression analyses showed that higher ALT values and higher histologic activity index (HAI) scores were associated with greater likelihood of response.69 In addition, there was no difference in virologic response between patients who had previously failed interferon and those who did not receive interferon. Virologic response was not influenced by age, gender, or ethnic origin (Table 73-4). In this study, 31 of 166 (19%) treated children developed the mutation conferring lamivudine resistance. Of these, only one had HBeAg seroconversion at 52 weeks. Preliminary results of prolonged treatment with lamivudine in children indicate that additional virologic responses can be achieved, and that seroconversions are durable in 88% of the treated children.70 The frequency of viral resistance continues to increase, as in adults. The safety profile of lamivudine appears to be quite good in children, and it is well tolerated in liquid or tablet form. Serious
Chapter 73 PEDIATRIC VIRAL HEPATITIS
side effects were not reported. In comparison to treatment with IFN-a, decreased height velocity and weight loss were not observed.
greater than 10 mIU/ml 12 years later, and that additional booster doses were not required.75
DAILY ACTIVITIES Other Treatment Options Adefovir dipivoxil is another option for adults with chronic hepatitis B infection. Currently, adefovir is not approved for use in children younger than 18 years. Peginterferon, entecavir, and emtricitabine have not been studied in the pediatric population.
IMMUNOPROPHYLAXIS Prevention of hepatitis B has been a major area of progress over the last 20 years. Vaccine recommendations regarding infants, children, and adolescents have evolved since the 1991 ACIP of the CDC made five major recommendations: 1. prevention of perinatal transmission by screening all pregnant women for HBsAg; 2. immunizing and implementing HBIG and HBV vaccine within 12 hours to babies born to HBsAg-seropositive mothers; 3. vaccine administration within 12 hours of birth to infants of untested mothers; 4. universal immunization for all babies born to HBsAg-negative mothers; 5. vaccination of high-risk adolescents and adults. Since that time, the recommendations have been broadened, with the most recent ACIP in 1999 suggesting vaccination of all children aged 0–18 years.71 Currently, the recommendation is to start the HBV vaccine series at birth, while the infant is in the nursery. In 1999, there was some concern regarding mercury exposure from thimerosal, a preservative used in the vaccine. By 2000, this compound was eliminated from HBV vaccines, but not all pediatricians have resumed HBV vaccination at birth for all infants.72 In addition, during the period of postponement of HBV vaccination at birth, one study demonstrated a sixfold increase in the number of hospitals not vaccinating all highrisk infants.73 Since most of these recommendations have been implemented, significant progress has been made towards decreasing transmission, acute infection, and chronic hepatitis in the pediatric population. After introduction of universal hepatitis B vaccination in Taiwan, fulminant neonatal hepatitis B mortality decreased from 5.36 to 1.71 per 100 000.46 In 1994, the chronic HBV prevalence rate was zero in Alaskan native children under 10 years of age who had been vaccinated at birth, in comparison to 16% of those aged 11–30 years who had not.74 The current recommendation is three intramuscular doses, starting in infancy. If administered according to the recommended schedule, the vaccine will promote a protective response in 95% of infants, children, and adolescents. For those children born to hepatitis Binfected mothers, anti-HBs and HBsAg should be checked after 12 months to determine the success of the HBV vaccines and HBIG. Anti-HBs should also be checked in patients co-infected with HIV, those with other immunodeficiencies, hemodialysis patients, and those living with chronically infected individuals. It is not necessary in other healthy children. Studies from Hong Kong have shown that more than 80% of children immunized with 3 doses had anti-HBs
Chronically infected children should not be allowed to share toothbrushes, razors, or any other items that may share blood. Children with chronic HBV infection should be allowed to participate in all activities without restrictions. Universal precautions are recommended for all children in schools, sports, or daycare situations. Although HBV has been detected in breast milk, mothers who are HBV-infected may continue to breast-feed if the infants are appropriately immunized.47
CONCLUSIONS Hepatitis B infection acquired during childhood is similar in some respects to that acquired by adults, but there are important differences in epidemiology, clinical features, natural history, and treatment. Chronic hepatitis B infection in children confers risks of cirrhosis and HCC. Other important considerations are the timing of optimal treatment, which, according to recent studies, appears to be when there is evidence of active hepatitis, as determined by abnormal ALT values and high HAI scores.63,68,69 Currently, in the USA, both IFN-a and lamivudine are available for children aged 2–17 years with chronic hepatitis B. Interferon is the only medication approved in Europe. However, prevention remains a primary goal, since safe and effective vaccines are available.
HEPATITIS C INTRODUCTION Children are only a small proportion of the HCV-infected population, but there are a significant number of children with chronic HCV. Chronic infection may not present with many clinical findings during childhood, but long-term infection can lead to significant morbidity and mortality, such as cirrhosis and HCC, later in life. Understanding that HCV in children has different modes of acquisition, complications, and natural history will influence management and treatment decisions.
EPIDEMIOLOGY The seroprevalence of antibody to HCV in the USA is approximately 1.8%, which is close to 4 million people. Data from the CDC have shown the seroprevalence to be 0.2% for children 6–11 years, and 0.4% for those aged 12–19 years.76 Other studies in the USA have shown prevalence in urban adolescents to be 0.1% in Boston and Cambridge, Massachusetts,77 and 0.1% in children under 12 years of age in a large cohort from Baltimore, Maryland.78 Older studies from around the world have reported seroprevalence rates of 0% in Japan,79 Taiwan,80 and Egypt,81 0.4% in Italy,82,83 0.9% in Saudi Arabia,84 and up to 14.5% in Cameroon.85 Exposures to hepatitis C such as intravenous drug use, needlesticks, and sexual contact are more common in adults than children. Prior to 1992, the primary route of HCV transmission in children was from transfusion of blood or blood products, including antihemophilic factor, factor IX concentrates, and immune globulin. The risk of infection appeared to correlate with the amount of trans-
1439
Section XII. Inherited and Pediatric Diseases of the Liver
fusion exposure. After 1992 and universal testing of blood products, vertical transmission has become the leading source of infection for children.86 Vertical transmission is much less common than that for hepatitis B, averaging about 5% in most studies.87–92 The likelihood of HCV transmission to newborns increases to 14% (range 5–36%) if mothers are infected with both HCV and HIV.89,92–95 Most of the HCV–HIV-co-infected women have higher levels of HCV viremia, which contributes to transmission.87,96 Infected household contacts may pose a small risk to children, although this has not been clearly documented. Seroprevalence rates in household contacts have ranged from 0 to 6.5%. These studies have shown an increasing risk with age and/or duration of exposure for sexual and non-sexual contacts.97,98 The mechanism of nonsexual, non-perinatal infection in children is not known. Child-tochild transmission is extremely rare, and the American Academy of Pediatrics (AAP) has advised no need to restrict school or daycare attendance. This includes full contact sports and other routine activities.99
PATHOGENESIS Acute hepatitis C is often not clinically recognized, so limited information is available. The mechanism for hepatocellular injury is thought to be both direct cytopathic effect and immune-mediated injury. Children have a propensity to develop chronic infection, although perhaps less frequently than adults.100,101 One hypothesis regarding the mechanism for viral persistence implicates a weak HCV-specific immune response that may occur through insufficient induction of a primary response or inability to maintain the response.102 Another hypothesis involves viral evasion of the immune responses, secondary to development of quasispecies, mutations, and different genotypes. At this point, immune responsiveness to HCV has not been compared between adults and children.
CLINICAL FEATURES The incubation period of hepatitis C infection averages 6–7 weeks, with a range of 2 weeks to 6 months. Most children are asymptomatic. However, when symptomatic, the infection is usually mild, with only about 25% of patients developing jaundice, and most having normal to mild elevation of aminotransferases.103–106 Children may rarely have associated anorexia, malaise, or abdominal pain with acute hepatitis. Occasionally, more severe acute hepatitis is seen.107
DIAGNOSIS Since hepatitis C is uncommon in children, most pediatricians are unsure of which children should be tested for this infection. Universal screening is not cost-effective or useful in children. The AAP Committee on Infectious Disease has made recommendations for the risk factors that should prompt testing for HCV (Table 73-5). The diagnosis of HCV infection is the same for children and adults. The first test is anti-HCV antibody. If this is positive, most clinicians will look for HCV RNA by a PCR-based assay to distinguish active infection from previous exposure to the virus or a falsepositive antibody. Care must be taken in patients with suspected acute hepatitis C, as anti-HCV may not appear until 5–6 weeks after symptoms of acute hepatitis. In this situation, testing for HCV RNA may be required for diagnosis. In immunocompromised children, or
1440
Table 73-5. Indications for Testing for Hepatitis C in Children and Adolescents Infants born to hepatitis C virus-infected mothers Receipt of blood or blood products before 1992 Hemodialysis History of injection drug use Receipt of intravenous immunoglobulin April 1993 to March 1994 Acute or chronic hepatitis with negative workup for hepatitis B and other disorders International adoptee
those receiving chemotherapy, anti-HCV may not develop, and testing directly for HCV RNA may be necessary. Infants born to anti-HCV-positive mothers will most likely have anti-HCV at birth, due to transplacental transfer of maternal antibody. The duration of this anti-HCV in the infant’s serum in the absence of true infection depends on the titer, but is usually less than 12–15 months. To evaluate for perinatal viral transmission early in infancy, PCR has been used. Several studies have confirmed the best time for testing of HCV RNA to be the third or fourth month of life.108 However, even in some of these infants with HCV, the infection will resolve spontaneously in the first 2 years of life. Children and adolescents with positive anti-HCV should have a qualitative PCR for HCV RNA to confirm infection. Once infection is confirmed, HCV genotype may be ascertained. Quantitative HCV RNA testing is reserved for assessment of response to treatment.
DIFFERENTIAL DIAGNOSIS Acute Hepatitis Acute hepatitis C in the pediatric population is uncommon, but some children do have mild to moderate elevation of aminotransferases in infancy and childhood. The differential diagnosis is similar to that of hepatitis B and other viral infections (Table 73-1). Histologic features of the liver are usually suggestive of a viral process, but occasionally may be confused with those of autoimmune hepatitis.
Chronic Hepatitis As discussed earlier, children may have chronic hepatitis secondary to a1-antitrypsin deficiency, hepatitis B, tyrosinemia, cystic fibrosis, or other metabolic disorders, as well as Wilson’s disease, autoimmune hepatitis, and medication or toxin-induced injury (Table 732). Of note, liver biopsy specimens from children and adolescents with chronic hepatitis C may demonstrate steatosis suggestive of metabolic disorders such as non-alcoholic fatty liver disease, fatty acid oxidation defects, or Wilson’s disease.
ASSOCIATED CONDITIONS Extrahepatic manifestations in adults, including cryoglobulinemia,109 vasculitis,110 and membranoproliferative glomerulonephritis,111 are not reported in children. Children may be co-infected with HBV or HIV, especially if infected by perinatal transmission. Non-organ-specific autoantibodies are commonly reported in children with hepatitis C, most commonly liver–kidney microsomal antibody type 1 (LKM1).112,113 This antibody, usually associated with type 2 autoimmune hepatitis, has been demonstrated in approximately 10% of HCV-infected children in European studies. Anti-
Chapter 73 PEDIATRIC VIRAL HEPATITIS
nuclear antibody and anti-smooth muscle antibodies have also been identified in this patient population. These autoantibodies are of concern, as a multicenter study indicated that LKM1-positive children with hepatitis C had higher fibrosis scores on liver biopsies, more severe liver disease, poorer response to interferon and, in some cases, required corticosteroid therapy for presumed autoimmune hepatitis.114
PROGNOSIS AND NATURAL HISTORY PERTINENT TO INFECTION ACQUIRED IN CHILDHOOD Although more data are being collected regarding the natural history of HCV infection in children, the ultimate prognosis and natural history are not fully understood. Studies have shown differences in the frequency of progression to chronicity and the importance of age at infection as a predictor of chronic viremia. The first long-term studies in children involved mostly transfusion-acquired infections. A large study of more than 400 children who had blood transfusions during cardiac surgery prior to 1991 demonstrated a 14.6% acquisition rate of hepatitis C as evidenced by anti-HCV. Of these infected patients, 55% developed chronic hepatitis, and were PCR-positive after a mean follow-up of 17 years (range 12–27 years).100 Most of these patients did not have clinical or biochemical signs of liver disease and 1 of 17 patients who underwent liver biopsy had cirrhosis. In the USA, a large cohort of HCV-infected cancer survivors who had been transfused prior to 1992 has been followed at St. Jude’s Children’s Research Hospital in Memphis, Tennessee. At screening in 1995, 81% of anti-HCV-seropositive patients were PCR-positive, with a median time from diagnosis of malignancy of 15.9 years. ALT abnormalities were seen in 30% of these PCR-positive patients and their biopsies showed mild to moderate fibrosis in 64%, and cirrhosis in 13%. A previous report from this cohort had described 1 patient who died of liver failure 9 years after onset of HCV, and 2 others who died of HCC after 25 and 27 years.115,116 Since the screening of blood products, perinatal transmission has been the focus of pediatric HCV natural history and prognosis studies. Vertical transmission is associated with a high incidence of viremia and abnormal aminotransferases during the first 12 months. Of 70 prospectively followed infants in five European centers from 1990 to 1999, 93% had evidence of abnormal ALT during the first 12 months, with only 19% clearing HCV RNA with normal ALT by 30 months of age.117 Clearance of viremia was independent of sex and maternal HIV co-infection. Peak ALT greater than five times normal during the first 18 months and genotype 3 were more common in the patients in whom viremia resolved spontaneously. The largest pediatric study to date describes a cohort of 200 HCV-infected children in Europe.118 The majority had genotype 1b, 45% from vertical transmission and 39% from transfusion. Fifteen percent of these patients had normal ALT, and none had jaundice or extrahepatic manifestations. After follow-up of 1–17.5 years (mean 6.2), only 6% achieved sustained virologic clearance and normalization of ALT. Liver biopsies were performed in 118 of these patients at various times during follow-up; the majority (76%) had mild hepatitis and low fibrosis scores. One patient (1%) had cirrhosis and 1 (1%) had severe hepatitis. Greater degrees of fibrosis were seen in children older than 15 years, suggesting long-term effects of chronic HCV infection.
The histopathology of the liver in children with chronic hepatitis C has been studied in three large series. All have shown that findings such as sinusoidal lymphocytosis, steatosis, portal lymphoid aggregates, and bile duct epithelial damage are present, and occur in relatively the same frequency in children as in adults.119–121 The first large study from Japan evaluated 109 liver specimens. The majority of these patients were infected via transfusion, biopsies were from patients with a mean duration of infection of 2.6 years, and genotypes were unknown. Ninety-seven percent of the specimens had no fibrosis or periportal fibrosis with intact architecture. Three percent had septal fibrosis with architectural distortion, and no patients had cirrhosis.120 In contrast, a study from the USA demonstrated portal fibrosis in 78% of 40 liver biopsies.119 Fibrosis was mild in 26%, moderate in 22%, and severe in 22%, with cirrhosis in 8%. Two of the children with cirrhosis were young adolescents who had acquired HCV perinatally. In comparison to the Japanese study, the mean duration of infection prior to biopsy was longer (6.8 years). Genotype 1a was present in 60% of this population, and 1b in 32%. A third series was a collection of 80 children from Italy and Spain, mostly infected with genotype 1, with a mean duration of infection of 3.5 years.121 Necroinflammatory scores were low in this population, and the frequency and severity of bile duct damage and lymphoid follicles increased with patient age. Fibrosis was seen in 72.5% of the biopsies, and increased with age and duration of disease, as in the American series. Only one child had cirrhosis. These three series have shown that, although necrosis and inflammation are usually mild, fibrosis is seen and worsens with age. Another complicating factor in children is the coexistence of autoantibodies, such as LKM1, as previously discussed.113,114 Although they include only small numbers of patients, these studies have shown higher fibrosis scores and more significant hepatic flares during interferon therapy. Of the 3 anti-LKM-positive patients in a European study who received interferon, treatment needed to be discontinued in all secondary to persistent viremia and ALT abnormalities.114 Potential long-term sequelae of chronic HCV infection in children are cirrhosis and HCC. HCC secondary to hepatitis C is extremely rare during childhood, and there have been only a few case reports.122–124 Liver transplantation for complications of chronic hepatitis C during childhood is uncommon. According to the SPLIT registry that collects data from 37 North American pediatric liver transplant centers, chronic hepatitis C with cirrhosis or “subacute hepatitis C” was the reason for transplant in 13 of 1378 children (1%) from 1995 through June 2003.16 Despite some complications that arise in childhood, HCV infection is benign during the first two decades in most cases. Severe liver disease and decompensated cirrhosis are rare during childhood. The greatest concern to the pediatric hepatologist regarding children with chronic HCV is the development of complications during adulthood.
TREATMENT Until recently, no therapies for hepatitis C were approved for use in children in the USA. In 2003 the Food and Drug Administration approved the combination of interferon and ribavirin for the treatment of chronic HCV in children aged 3–17 years. However, decid-
1441
Section XII. Inherited and Pediatric Diseases of the Liver
Table 73-6. Special Considerations Regarding Hepatitis C Therapy in Children
Table 73-7. Variables Associated with Sustained Virologic Response to Interferon-a and Ribavirin in Children with Chronic Hepatitis C127
Differences in natural history Mode of acquisition: perinatal Shorter duration of infection Fewer co-infections and comorbid diseases Longer anticipated life expectancy Differences in liver disease Milder grades of necroinflammation Less frequent severe fibrosis or cirrhosis Differences in response to interferon-a with or without ribavirin Higher frequency of response Lower frequency of relapse Fewer drug discontinuations because of side effects Unknown long-term side effects No cost–benefit data Preliminary data about peginterferon in children
Association
No association
£12 years of age Genotype 2 and 3 If genotype 1, hepatitis C virus RNA £2 million copies/ml Receiving at least 80% of the medication doses and completing ≥38 weeks of treatment
Baseline ALT Gender Mode of acquisition
Reproduced from Jonas MM. Treatment of chronic hepatitis C in pediatric patients. In: Keeffe EB, ed. Clinics in liver disease. Philadelphia, PA: WB Saunders; 1999:4:855–867,161 with permission.
ing to treat a child with hepatitis C may be difficult, as physicians must consider the epidemiology, natural history, and treatment efficacy in children. Safety profiles and appropriate monitoring must be recognized before treatment, since some children have comorbidities that may complicate therapy (Table 73-6). Interferon monotherapy was studied in children in small clinical trials, but it has not been tested in a large multicenter trial. Sustained virologic response rates were significantly better than in large trials in adults. A systematic analysis of multiple studies included 270 treated patients and 37 control subjects. The overall sustained virologic response in the treated children was 35%. There was a difference in response by genotype: 26% for genotype 1 and 70% for the other genotypes.125 A report in abstract form described a cohort of 61 patients, aged 5–11 years, treated with 3 MU per meter squared of body surface area of IFN-a2b thrice weekly in combination with ribavirin at 8, 12, or 15 mg/kg of body weight daily.126 Data from this cohort demonstrated pharmacokinetics similar to those in adults. The overall sustained virological response rate was 38%. Subsequently, results from a larger multicenter trial of combination therapy were presented in abstract form.127 Seventy children aged 3–17 years with chronic hepatitis C were given IFN-a2b, 3 MU per meter squared of body surface area thrice weekly, and 15 mg/kg of ribavirin daily for 24 weeks. Therapy was continued for an additional 24 weeks if they demonstrated undetectable HCV RNA or greater than 2 log decrease in viremia. Most patients were infected via vertical transmission (61%) and were infected with genotype 1 (74%). Overall, 49% had a sustained virologic response (HCV RNA < 100 copies/ml) 24 weeks after completion of therapy. Factors associated with likelihood of sustained virologic response were age less than 12 years, genotypes 2 and 3, and fewer than 2 million copies/ml of hepatitis C RNA in those children with genotype 1 infection. Variables not associated with sustained virologic response were ethnicity, gender, mode of acquisition, duration of infection, baseline ALT, and ribavirin preparation (tablet versus liquid) (Table 73-7). The side effects of combination therapy were similar to those seen in adults,
1442
Estimated duration of infection
Ethnicity ALT, alanine aminotransferase.
but less severe. Adverse events led to dosage modifications in 20% and discontinuation in 7% of the treated children. Peginterferon-a2a has been studied in a small group of children to evaluate safety and pharmacokinetics.128 Research is under way to evaluate the safety and efficacy of peginterferon-a2a and ribavirin in children and adolescents with chronic hepatitis C.
DAILY ACTIVITIES Chronically infected children should not be allowed to share toothbrushes, razors, or any other items that may be contaminated with blood. Teenagers, if sexually active, must also be advised regarding the unlikely – but not impossible – risk of sexual transmission to their partners. Mothers who are HCV-infected may continue to breast-feed according to recommendations of the CDC, but should probably abstain if their nipples are cracked or bleeding.99 Universal precautions are recommended for all children in schools, sports, or daycare situations, regardless of HCV status. Children with chronic hepatitis C should not be restricted in activities or school settings.
CONCLUSIONS Hepatitis C infection in children continues to be a health care burden despite its changing epidemiology from transfusion acquired to vertical transmission. At this point, there are no proven methods to prevent perinatal transmission, so chronic HCV will continue to be seen by pediatricians. Further research is needed to delineate the factors that predict progression of disease and fibrosis in children. Randomized, prospective trials are required to determine optimal type and timing of the treatment.
HEPATITIS IN CHILDREN CAUSED BY OTHER VIRAL AGENTS As in adults, hepatitis A, hepatitis B, and hepatitis C are the predominant hepatitis viruses in children. Other hepatotropic viruses and other viral agents should be considered in the correct clinical settings and age groups.
Hepatitis D HDV is an uncommon infection in children. HDV is not usually transmitted perinatally,129 as most mothers with HDV are usually anti-HBe-positive, and less infectious. It appears that most cases are due to horizontal spread. The most common regions for HDV infec-
Chapter 73 PEDIATRIC VIRAL HEPATITIS
tion in children are southern Italy, parts of Eastern Europe, South America, Africa, and the Middle East. HDV infection is rare in the Far East, which is important in the pediatric population, since it will be uncommon in children adopted from or emigrated from Asia, but may occasionally be encountered in children from parts of Europe and the Middle East. The clinical course appears to be similar to that in adults, in that HBV-infected children with HDV have more advanced liver disease.129 Treatment of HDV infection in children with IFN-a has some short-term effects on HDV RNA, but did not appear to cause significant long-term virologic response or histologic change in several trials in Europe.130,131 Universal immunization of infants for HBV should decrease the risk of HDV infection.
Hepatitis E Hepatitis E virus (HEV) is a rare cause of hepatitis in children in the USA and is only slightly more common in the adult population. It is seen in individuals who have recently traveled to endemic areas, which include parts of India, areas of central and South-East Asia, north-west China, and parts of Africa. Multiple studies of sporadic acute hepatitis in parts of Africa have shown HEV to be a significant contributor.132,133 There have also been reports that children in endemic areas who are co-infected with HAV can have a more severe course than with either infection alone.134 Diagnosis is made by HEV IgM antibody from serum. There is no current therapy, and most children recover uneventfully.
Enteroviruses The non-polio enteroviruses (coxsackieviruses, echoviruses, and enteroviruses) are those most commonly associated with severe hepatitis in neonates.42,43 Transmission may be prenatal, intrapartum, or perinatal. Mothers of these infants usually have a history of a viral syndrome, with fever or diarrhea approximately 1 week prior to delivery. Most affected infants have a mild course, but severe neonatal hepatitis can occur within the first week, with jaundice, coagulopathy, and ascites. Echovirus 11 appears to be the most common, but coxsackie B virus has been implicated as well.42,135 Diagnosis can now be made by PCR in serum or urine, although this test is only available in research laboratories.136 There has been a report of clinical improvement in infants with severe hepatitis using pleconaril, an antienteroviral therapy inhibiting viral uncoating and blocking viral attachment to host cell receptors.135
Adenovirus Adenovirus is a commonly acquired infection, usually in the respiratory tract. Neonatal hepatitis due to adenovirus usually presents within the first week of life. This may be associated with overwhelming, potentially fatal, infection. Clinical findings are hepatomegaly, hepatitis, and bleeding secondary to thrombocytopenia and coagulopathy. Other symptoms include lethargy, pneumonia, and fever or hypothermia.43 Adenovirus may cause hepatitis in immunocompromised individuals as well.139
Cytomegalovirus Congenital CMV infection is often asymptomatic, but a minority of infected newborns will develop hepatitis, hepatosplenomegaly, conjugated hyperbilirubinemia, and/or thrombocytopenia. The diagnosis of CMV infection may be made by liver biopsy, urine CMV culture, or IgM antibody to CMV. CMV does not cause chronic infection in immunocompetent children.140 Acute CMV hepatitis is usually not a disorder of healthy children or adolescents, but is a common disease in immunocompromised hosts.141,142
Parvovirus B19 Parvovirus B19 is the virus associated with erythema infectiosum (fifth disease). Typical clinical features are fever, a lacy maculopapular rash, and erythematous “slapped” cheeks. Occasionally, children will develop acute hepatitis and, rarely, fulminant hepatitis. It has been suggested that fulminant hepatitis due to parvovirus B19 has a better prognosis than that due to other viral infections.143 Associated with fulminant hepatic failure, some children develop aplastic anemia.144 Diagnosis of acute infection is confirmed by parvovirus B19 IgM antibody or PCR.
Epstein–Barr Virus Clinical symptoms of infectious mononucleosis due to EBV infection include fever, pharyngitis, lymphadenopathy, splenomegaly, and hepatitis with jaundice.145 Abnormal aminotransferases are present in up to 80% of patients, although a small percentage develop jaundice.146 Fulminant hepatitis and EBV-associated hemophagocytic syndrome are rare complications.147 Diagnosis may be made with a monospot, but there are frequent false-negative results in children younger than 4 years, and false-positives in older children with other viral infections or medical conditions. Serum IgM to the EB viral capsid antigen (VCA) is both more sensitive and specific, and has usually developed by the time of clinical presentation.
Herpes Simplex Viruses Herpes simplex virus may cause severe hepatitis in the neonatal period.41 It is usually in conjunction with disseminated infection, and presents in the first week of life. Associated clinical symptoms include the typical cutaneous vesicular lesions, hepatosplenomegaly, fever, and coagulopathy. The treatment for these children is acyclovir, which must be instituted promptly. Occasionally, liver transplantation is necessary, with continued acyclovir after transplantation.137 In addition to infection in neonates, herpes simplex virus may cause hepatitis in immunocompromised children and adolescents, such as bone marrow and solid organ transplant recipients.138
Human Herpes Virus-6 Human herpes virus-6 (HHV-6) is the etiologic agent of roseola infantum in children. The typical presentation is high fever for several days, followed by rapid defervescence and the development of an erythematous maculopapular rash. Several studies have documented HHV-6 in the livers of infants and adults with acute liver failure.148,149 However, whether HHV-6 is the actual etiologic agent causing the liver disease is controversial, since almost all children are seropositive for the virus by 2 years of age, and the virus may persist and reactivate.150,151 This claim was further substantiated by reports
1443
Section XII. Inherited and Pediatric Diseases of the Liver
that HHV-6 was detected in serum or bone marrow by PCR in hospitalized children with other chronic medical problems152 and has been detected via PCR in the livers of 36 of 48 children (75%) with various liver diseases.153 Chronic hepatitis associated with HHV-6 has been reported in 1 infant.154
CONCLUSION Although adults and children may become infected with the same viruses causing hepatitis, the epidemiology, clinical findings, and natural history of these infections may be quite different. Identifying and recognizing these differences are important in terms of understanding pathogenesis, prevention, and treatment. Children may have substantial morbidity later in life from chronic hepatitis, such as cirrhosis or HCC. Determination of optimal treatment regimens for chronic infections in children requires further research through large trials with new medications or combinations of medications. Prevention is still the most important mechanism for avoiding these viral sequelae. Broadening HAV and HBV vaccine usage, and development of other vaccines, as well as prevention of perinatal transmission of hepatitis C, will further decrease the numbers of children with acute and chronic viral hepatitis in the future.
REFERENCES 1. Bell BP, Shapiro CN, Alter MJ, et al. The diverse patterns of hepatitis A epidemiology in the United States – implications for vaccination strategies. J Infect Dis 1998; 178:1579–1584. 2. CDC. Hepatitis surveillance report 58. Atlanta, GA: US Department of Health and Human Services, Public Health Service, CDC, 2003. 3. CDC. Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR. Morbid Mortal Week Rep 1999; 48. 4. CDC. Hepatitis surveillance report no. 56. Atlanta, GA: US Department of Health and Human Services, Public Health Service, CDC, 1996. 5. Siegl G, Weitz M. Pathogenesis of hepatitis A: persistent viral infection as basis of an acute disease? Microb Pathogen 1993; 14:1–8. 6. Gabriel P, Vallbracht A, Flehmig B. Lack of complementdependent cytolytic antibodies in hepatitis A virus infection. J Med Virol 1986; 20:23–31. 7. Vallbracht A, Maier K, Stierhof YD, et al. Liver-derived cytotoxic T cells in hepatitis A virus infection. J Infect Dis 1989; 160:209–217. 8. Ertekin V, Selimoglu M, Orbak Z. An unusual combination of relapsing and cholestatic hepatitis A in childhood. Yonsei Med J 2003; 44:939–942. 9. Arslan S CH, Oner AF, Odabas D, et al. Relapsing hepatitis A in children: report of two cases. Acta Paediatr Taiwan 2002; 43:358–360. 10. Press J, Maslovitz S, Avinoach I. Cutaneous necrotizing vasculitis associated with hepatitis A virus infection. J Rheumatol 1997; 24:965–967. 11. Inman RD, Hodge M, Johnston MEA, et al. Arthritis, vasculitis, and cryoglobulinemia associated with relapsing hepatitis A infection. Ann Intern Med 1986; 105:700–703.
1444
12. Agarwal K, Puliyel J, Mathew A, et al. Acute pancreatitis with cholestatic hepatitis: an unusual manifestation of hepatitis A. Ann Trop Paediatr 1999; 19:391–394. 13. Amarapurkar DN, Begani MM, Mirchandani K. Acute pancreatitis in hepatitis A infection. Trop Gastroenterol 1996; 17:30–31. 14. Debray D, Cullufi P, Devictor D, et al. Liver failure in children with hepatitis A. Hepatology 1997; 26:1018–1022. 15. Vento S, Garofano T, Renzini C, et al. Fulminant hepatitis associated with hepatitis A virus superinfection in patients with chronic hepatitis C. N Engl J Med 1998; 338:286–290. 16. Anonymous. Studies of pediatric liver transplantation 2003 annual report, 2003. 17. Anonymous. Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). Morbid Mortal Week Rep 1999; 48(RR12):1–37. 18. Das A. An economic analysis of different strategies of immunization against hepatitis A virus in developed countries. Hepatology 1999; 29:548–52. 19. Rosenthal P. Cost-effectiveness of hepatitis A vaccination in children, adolescents, and adults. Hepatology 2003; 37:44–51. 20. American Academy of Pediatrics. Hepatitis A. In: Pickering L, ed. 2000 Red book: report of the Committee on Infectious Diseases. Elk Grove Village, IL: American Academy of Pediatrics; 2000:280–289. 21. Ferreira C, da Silveira T, Vieira S, et al. Immunogenicity and safety of hepatitis A vaccine in children with chronic liver disease. J Pediatr Gastroenterol Nutr 2003; 37:258–261. 22. Majda-Stanislawska EBM, Kuydowicz J. Immunogenicity of inactivated hepatitis A vaccine in children with chronic liver disease. Pediatr Infect Dis J 2004; 23:571–574. 23. Hsu HY, Chang MH, Liaw SH, et al. Changes of hepatitis B surface antigen variants in carrier children before and after universal vaccination in Taiwan. Hepatology 1999; 30:1312–1317. 24. Stevens C, Beasley R, Tsui J, et al. Vertical transmission of hepatitis B antigen in Taiwan. N Engl J Med 1975; 292:771–774. 25. McMahon BJ, Alward WLM, Hall DB, et al. Acute hepatitis B virus infection: relation of age to the clinical expression of disease and subsequent development of the carrier state. J Infect Dis 1985; 151:599–603. 26. Centers for Disease Control. Hepatitis surveillance report no. 54. Atlanta, GA: Centers for Disease Control, 1992:1. 27. Armstrong GL, Mast EE, Wojczynski M, et al. Childhood hepatitis B virus infections in the United States before hepatitis B immunization. Pediatrics 2001; 108:1123–1128. 28. Anonymous. Achievements in public health: hepatitis B vaccination – United States, 1982–2002. Morbid Mortal Week Rep 2002; 51:549–52, 563. 29. Centers for Disease Control and Prevention. Incidence of acute hepatitis B – United States, 1990–2002. MMWR. Morbid Mortal Week Rep 2004; 52:1252–1254. 30. Hsu H, Chang M, Hsieh K, et al. Cellular immune response to HBcAg in mother-to-infant transmission of hepatitis B virus. Hepatology 1992; 15:770–776. 31. Sinatra F, Shah P, Weissman J, et al. Perinatal transmitted acute icteric hepatitis B in infants born to hepatitis B surface antigenpositive and antihepatitis B e-positive carrier mothers. Pediatrics 1982; 70:557–559. 32. Delaplane D, Yogev R, Crussi F, et al. Fatal hepatitis B in early infancy: the importance of identifying HBsAg-positive pregnant women and providing immunoprophylaxis to their newborns. Pedatrics 1983; 72:176–180. 33. Chen H-L, Chang C-J, Kong M-S, et al. Pediatric fulminant hepatic failure in endemic areas of hepatitis B infection: 15 years after universal hepatitis B vaccination. Hepatology 2004; 39:58–63.
Chapter 73 PEDIATRIC VIRAL HEPATITIS
34. Chang M, Lee C, Chen D, et al. Fulminant hepatitis in children in Taiwan: the important role of hepatitis B virus. J Pediatr 1987; 111:34–39. 35. Gianotti F. Papular acrodermatitis of childhood: an Australian antigen disease. Arch Dis Child 1973; 48:794–799. 36. Kleinknecht C, Levy M, Peix A, et al. Membranous glomerulonephritis and hepatitis B surface antigen in children. J Pediatr 1979; 95:946–952. 37. Southwest Pediatric Nephrology Study Group. Hepatitis B surface antigenemia in North American children with membranous glomerulonephropathy. J Pediatr 1985; 106:571–578. 38. Ozdamar S, Gucer S, Tinaztepe K. Hepatitis-B virus associated nephropathies: a clinicopathological study in 14 children. Pediatr Nephrol 2003; 18:23–28. 39. Filler G, Feber J, Weiler G, et al. Another case of HBV associated membranous glomerulonephritis resolving on lamivudine. Arch Dis Child 2003; 88:460. 40. Connor F, Rosenberg A, Kennedy S, et al. HBV associated nephrotic syndrome: resolution with oral lamivudine. Arch Dis Child 2003; 88:446–449. 41. Benador N, Mannhardt W, Schranz D, et al. Three cases of neonatal herpes simplex virus infection presenting as fulminant hepatitis. Eur J Pediatr 1990; 149:555–559. 42. Modlin JF. Perinatal echovirus infection: insights from a literature review of 61 cases of serious infection and 16 outbreaks in nurseries. Rev Infect Dis 1986; 8:918–926. 43. Abzug MJ, Levin MJ. Neonatal adenovirus infection: four patients and review of the literature. Pediatrics 1991; 87:890–893. 44. Yoto Y, Kudoh T, Haseyama K, et al. Human parvovirus B19 infection associated with acute hepatitis. Lancet 1996; 347:868–869. 45. Maggiore G, Hadchouel M, Sessa F, et al. A retrospective study of the role of delta agent infection in children with HBsAgpositive chronic hepatitis. Hepatology 1985; 5:7–9. 46. Kao J, Hsu H, Shau W, et al. Universal hepatitis B vaccination and the decreased mortality from fulminant hepatitis in infants in Taiwan. J Pediatr 2001; 139:349–352. 47. American Academy of Pediatrics. Hepatitis B. In: Pickering L, ed. 2000 Red book: report of the Committee on Infectious Diseases. Elk Grove Village, IL: American Academy of Pediatrics; 2000:289–301. 48. Lok ASF, Lai CL. A longitudinal follow-up of asymptomatic hepatitis B surface antigen-positive Chinese children. Hepatology 1988; 8:1130–1133. 49. Hsu H, Chang M, Chen D, et al. Baseline seroepidemiology of hepatitis B virus infection in children in Taipei, 1984: a study just before mass hepatitis B vaccination program in Taiwan. J Med Virol 1986; 18:301–307. 50. Bortolotti F, Cadrobbi P, Crivellaro C, et al. Long-term outcome of chronic type B hepatitis in patients who acquire hepatitis B virus infection in childhood. Gastroenterology 1990; 99:805–810. 51. Bortolotti F, Jara P, Crivellaro C, et al. Outcome of chronic hepatitis B in Caucasian children during a 20-year observation period. J Hepatol 1998; 29:184–190. 52. Chang M, Hwang L, Hsu H, et al. Prospective study of asymptomatic HBsAg carrier children infected in the perinatal period: clinical and liver histologic studies. Hepatology 1988; 8:374–377. 53. Czauderna P, Mackinlay G, Perilongo G, et al. Hepatocellular carcinoma in children: results of the first prospective study of the International Society of Pediatric Oncology group. J Clin Oncol 2002; 20:2798–2804. 54. Pontisso P, Basso G, Perilongo G, et al. Does hepatitis B virus play a role in primary liver cancer in children of western countries? Cancer Detect Prevent 1991; 15:363–368. 55. Giacchino R, Navone C, Facco F, et al. HBV-DNA-related hepatocellular carcinoma occurring in childhood. Report of three cases. Dig Dis Sci 1991; 36:1143–1146.
56. Chang MH, Chen CJ, Lai MS, et al. Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. Taiwan Childhood Hepatoma Study Group. N Engl J Med 1997; 336:1906–1907. 57. Young JJ, Ries L, Silverberg E, et al. Cancer incidence, survival, and mortality for children younger than age 15 years. Cancer 1986; 58 (Suppl):598–602. 58. Wen W, Chang M, Hsu H, et al. The development of hepatocellular carcinoma among prospectively followed children with chronic hepatitis B virus infection. J Pediatr 2004; 144:397–399. 59. Hsu H, Wu M, Chang M, et al. Childhood hepatocellular carcinoma develops exclusively in hepatitis B surface antigen carriers in three decades in Taiwan. Report of 51 cases strongly associated with rapid development of liver cirrhosis. J Hepatol 1987; 5:260–267. 60. Lok AS. Alpha-interferon therapy for chronic hepatitis B virus infection in children and Oriental patients. J Gastroenterol Hepatol 1991 (Suppl. 1):15–17. 61. Lai CL, Lin HJ, Lau JN, et al. Effect of recombinant alpha2 interferon with or without prednisone in Chinese HBsAg carrier children. Q J Med 1991; 78:155–163. 62. Lai CL, Lok AS, Lin HJ, et al. Placebo-controlled trial of recombinant alpha2-interferon in Chinese HBsAg-carrier children. Lancet 1987; 2:877–880. 63. Sokal EM, Conjeevaram HS, Roberts EA, et al. Interferon alfa therapy for chronic hepatitis B in children: a multinational randomized controlled trial. Gastroenterology 1998; 114:988–995. 64. Iorio R, Pensati P, Botta S, et al. Side effects of alpha-interferon therapy and impact on health-related quality of life in children with chronic viral hepatitis. Pediatr Infect Dis J 1997; 16:984–990. 65. Comanor L, Minor J, Conjeevaram HS, et al. Impact of chronic hepatitis B and interferon-alfa therapy on growth of children. J Viral Hepatitis 2001; 8:139–147. 66. Gottrand F, Michaud L, Guimber D, et al. Influence of recombinant interferon alpha on nutritional status and growth pattern in children with chronic viral hepatitis. Eur J Pediatr 1996; 155:1031–1034. 67. Sokal EM, Roberts EA, Mieli-Vergani G, et al. A dose ranging study of the pharmacokinetics, safety, and preliminary efficacy of lamivudine in children and adolescents with chronic hepatitis B. Antimicrob Agents Chemother 2000; 44:590–597. 68. Jonas MM, Kelly DA, Mizerski J, et al. Clinical trial of lamivudine in children with chronic hepatitis B. N Engl J Med 2002; 346:1706–1713. 69. Hom X, Little N, Gardner S, Jonas M. Predictors of virologic response to lamivudine treatment in children with chronic hepatitis B infection. Pediatr Infect Dis J 2004; 23:441–445. 70. Sokal E, Mizerski J, Woynarowski M, et al. An extended, stratified, follow-on study in children with chronic hepatitis B (CHB) who have completed previous lamivudine studies (abstract). Hepatology 2003; 38 (suppl. 1):710A. 71. Anonymous. Update: recommendations to prevent hepatitis B virus transmission – United States. MMWR Morbid Mortal Week Rep 1999; 48:33–34. 72. Luman E, Fiore A, Strine T, et al. Impact of thimerosal-related changes in hepatitis B vaccine birth – dose recommendations on childhood vaccination coverage. JAMA 2004; 291:2351–2358. 73. Clark S, Cabana M, Malik T, et al. Hepatitis B vaccination practices in hospital newborn nurseries before and after changes in vaccination recommendations. Arch Pediatr Adolesc Med 2001; 155:915–920. 74. Harpaz R, McMahon B, Margolis H, et al. Elimination of new chronic hepatitis B virus infections: results of the Alaska immunization program. J Infect Dis 2000; 181:413–418.
1445
Section XII. Inherited and Pediatric Diseases of the Liver
75. Yuen MF, Lim WL, Cheng CC, et al. Twelve-year follow-up of a prospective randomized trial of hepatitis B recombinant DNA yeast vaccine versus plasma-derived vaccine without booster doses in children. Hepatology 1999; 29:924–927. 76. Committee on Infectious Diseases, American Academy of Pediatrics. Hepatitis C virus infection. Pediatrics 1998; 101:481–485. 77. Jonas MM, Robertson LM, Middleman AB. Low prevalence of antibody to hepatitis C virus in an urban adolescent population. J Pediatr 1997; 131:314–316. 78. El-Kamary SS, Serwint JR, Joffe A, et al. Prevalence of hepatitis C virus infection in urban children. J Pediatr 2003; 143:54–59. 79. Tanaka E, Kiyosawa K, Soeyama T, et al. Prevalence of antibody to hepatitis C virus in Japanese schoolchildren: comparison with adult blood donors. Am J Trop Med Hyg 1992; 46:460–464. 80. Lee S-D, Chan C-Y, Wnag Y-J, et al. Seroepidemiology of hepatitis C virus infection in Taiwan. Hepatology 1991; 13:830–833. 81. Khalifa AS, Mitchell BS, Watts DM, et al. Prevalence of hepatitis C viral antibody in transfused and nontransfused Egyptian children. Am J Trop Med Hyg 1993; 49:316–321. 82. Gessoni G, Manoni F. Prevalence of anti-hepatitis C virus antibodies among teenagers in the Venetian area: a seroepidemiological study. Eur J Med 1993; 2:79–82. 83. Romano L, Azara A, Chiaramonte M, et al. Low prevalence of anti-HCV antibody among Italian children. Infection 1994; 22:350–352. 84. al-Faleh FZ, Ayoola EA, al-Jeffry M, et al. Prevalence of antibody to hepatitis C virus among Saudi Arabian children: a community-based study. Hepatology 1991; 14:215–218. 85. Ngatchu T, Stroffolini T, Rapicetta M, et al. Seroprevalence of anti-HCV in an urban child population: a pilot survey in a developing area, Cameroon. J Trop Med Hyg 1992; 95:57–61. 86. Bortolotti F, Resti M, Giacchino R, et al. Changing epidemiologic pattern of chronic hepatitis C virus infection in Italian children. J Pediatr 1998; 133:378–381. 87. Ohto H, Terazawa S, Sasaki N, et al. Transmission of hepatitis C virus from mothers to infants. N Engl J Med 1994; 330:744–750. 88. Wejstal R, Widell A, Mansson A-S, et al. Mother to infant transmission of hepatitis C virus. Ann Intern Med 1992; 117:887–890. 89. Lam JPH, McOmish F, Burns SM, et al. Infrequent vertical transmission of hepatitis C virus. J Infect Dis 1993; 167:572–576. 90. Lin HH, Kao JH, Hsu HY, et al. Possible role of high-titer maternal verimia in perinatal transmission of hepatitis C virus. J Infect Dis 1994; 169:638–641. 91. Resti M, Azzari C, Mannelli F, et al. Mother to child transmission of hepatitis C virus: prospective study of risk factors and timing of infection in children born to women seronegative for HIV-1. Br Med J 1998; 317:437–441. 92. Zanetti AR, Tanzi E, Paccagnini S, et al. Mother-to-infant transmission of hepatitis C virus. Lombardy Study Group on vertical HCV transmission. Lancet 1995; 345:289–291. 93. Manzini P, Saracco G, Cerchier A, et al. Human immunodeficiency virus infection as a risk factor for motherto-child hepatitis C virus transmission; persistence of antihepatitis C virus in children is associated with the mother’s anti-hepatitis C virus immunoblotting pattern. Hepatology 1995; 21:328–332. 94. Zuccotti GV, Ribero ML, Giovannini M, et al. Effect of hepatitis C genotype on mother-to-infant transmission of virus. J Pediatr 1995; 127:278–280. 95. Paccagnini S, Principi N, Massironi E, et al. Perinatal transmission and manifestation of hepatitis C virus infection in a high risk population. Pediatr Infect Dis J 1995; 14:195–199.
1446
96. Ferrero S, Lungaro P, Bruzzone B, et al. Prospective study of mother-to-infant transmission of hepatitis C virus: a 10-year survey (1990–2000). Acta Obstet Gynecol Scand 2003; 82:229–234. 97. Pramoolsinsap C, Kurathong S, Lerdverasirikul P. Prevalence of anti-HCV antibody in family members of anti-HCV-positive patients with acute and chronic liver disease. Southeast Asian J Trop Med Public Health 1992; 23:12–16. 98. Chang TT, Liou TC, Young KC, et al. Intrafamilial transmission of hepatitis C virus: the important role of inapparent transmission. J Med Virol 1994; 42:91–96. 99. American Academy of Pediatrics. Hepatitis C. In: Peter G, ed. Red book: report of the Committee on Infectious Diseases, vol. 99. Elk Grove Village, IL: American Academy of Pediatrics, 1997:260–264. 100. Vogt M, Lang T, Frosner G, et al. Prevalence and clinical outcome of hepatitis C infection in children who underwent cardiac surgery before the implementation of blood-donor screening. N Engl J Med 1999; 341:866–870. 101. Bortolotti F, Resti M, Giacchino R, et al. Hepatitis C virus infection and related liver disease in children of mothers with antibodies to the virus. J Pediatr 1997; 130:990–993. 102. Rehermann B. Immunopathogenesis of hepatitis C. In: Liang T, Hoofnagle J, eds. Hepatitis C. San Diego, California: Academic Press; 2000:147–166. 103. Locasciulli A, Gornati G, Tagger A, et al. Hepatitis C virus infection and chronic liver disease in children with leukemia in long-term remission. Blood 1991; 78:1619–1622. 104. Ni Y-H, Chang M-H, Lue H-C, et al. Posttransfusion hepatitis C virus infection in children. J Pediatr 1994; 124:709–713. 105. Bortolotti F, Resti M, Giacchino R, et al. Hepatitis C virus infection and related liver disease in children of mothers with antibodies to the virus. J Pediatr 1997; 130:990–993. 106. Bortolotti F, Jara P, Diaz C, et al. Posttransfusion and community-acquired hepatitis C in childhood. J Pediatr Gastroenterol Nutr 1994; 18:279–283. 107. Jonas MM, Baron MJ, Bresee JS, et al. Clinical and virologic features of hepatitis C virus infection associated with intravenous immunoglobulin. Pediatrics 1996; 98:211–215. 108. Conte D, Fraquelli M, Prati D, et al. Prevalence and clinical course of chronic hepatitis C virus (HCV) infection and rate of HCV vertical transmission in a cohort of 15 250 pregnant women. Hepatology 2000; 31:751–755. 109. Agnello V, Chung RT, Kaplan LM. A role for hepatitis C virus infection in type II cryoglobulinemia. N Engl J Med 1992; 327:1490–1495. 110. Marcellin P, Descamps V, Martinot-Peignoux M, et al. Cryoglobulinemia with vasculitis associated with hepatitis C virus infection. Gastroenterology 1993; 104:272–277. 111. Johnson RJ, Gretch DR, Yamabe H, et al. Membranoproliferative glomerulonephritis associated with hepatitis C virus infection. N Engl J Med 1993; 328:465–470. 112. Bortolotti F, Vajro P, Balli F, et al. Non-organ specific autoantibodies in children with chronic hepatitis C. J Hepatol 1996; 25:614–620. 113. Gregorio G, Pensati P, Iorio R, et al. Autoantibody prevalence in children with liver disease due to chronic hepatitis C virus (HCV) infection. Clin Exp Immunol 1998; 112:471–476. 114. Bortolotti F, Muratori L, Jara P, et al. Hepatitis C virus infection associated with liver–kidney microsomal antibody type 1 (LKM1) autoantibodies in children. J Pediatr 2003; 142:185–190. 115. Castellino S, Lensing S, Riely C, et al. The epidemiology of chronic hepatitis C infection in survivors of childhood cancer: an update of the St Jude Children’s Research Hospital hepatitis C seropositive cohort. Blood 2004; 103:2460–2466.
Chapter 73 PEDIATRIC VIRAL HEPATITIS
116. Strickland DK, Riely CA, Patrick CC, et al. Hepatitis C infection among survivors of childhood cancer. Blood 2000; 95:3065–3070. 117. Resti M, Jara P, Hierro L, et al. Clinical features and progression of perinatally acquired hepatitis C virus infection. J Med Virol 2003; 70:373–377. 118. Jara P, Resti M, Hierro L, et al. Chronic hepatitis C virus infection in childhood: clinical patterns and evolution in 224 white children. Clin Infect Dis 2003; 36:275–280. 119. Badizadegan K, Jonas MM, Ott MJ, et al. Histopathology of the liver in children with chronic hepatitis C viral infection. Hepatology 1998; 28:1416–1423. 120. Kage M, Fujisawa T, Shiraki K, et al. Pathology of chronic hepatitis C in children. Hepatology 1997; 26:771–775. 121. Guido M, Rugge M, Jara P, et al. Chronic hepatitis C in children: the pathological and clinical spectrum. Gastroenterology 1998; 115:1525–1529. 122. Schaffner F, Thung SN. Clinicopathology conferences. End-stage liver disease in a young woman. Hepatology 1994; 19:534–537. 123. Strickland D, Jenkins J, Hudson M. Hepatitis C infection and hepatocellular carcinoma after treatment of childhood cancer. J Pediatr Hematol Oncol 2001; 23:527–529. 124. Gonzalez-Peralta R, Langham M, Mohan P, et al. Hepatocellular carcinoma in two adolescents with cirrhosis secondary to hepatitis C virus infection. J Pediatr Gastroenterol Nutr 2003; 37:380. 125. Jacobson KR, Murray K, Zellos A, et al. An analysis of published trials of interferon monotherapy in children with chronic hepatitis C. J Pediatr Gastroenterol Nutr 2002; 34:52–58. 126. Bunn S, Kelly D, Murray KF, et al. Safety, efficacy and pharmacokinetics of interferon-alfa-2b and ribavirin in children with chronic hepatitis C (abstract). Hepatology 2000; 32:350A. 127. Gonzalez-Peralta R, Haber BA, Jonas MM, et al. Interferonalfa2b in combination with ribavirin for the treatment of chronic hepatitis C in children (abstract). Hepatology 2002; 36 (part 2):311A. 128. Schwarz KB, Mohan P, Narkewicz MR, et al. The safety, efficacy and pharmacokinetics of peginterferon alfa-2a (40kD) in children with chronic hepatitis C (abstract). Gastroenterology 2003; 124 (suppl 1):A-700. 129. Farci P, Barbera C, Navone C, et al. Infection with the delta agent in children. Gut 1985; 26:4–7. 130. Di Marco V, Giacchino R, Timitilli A, et al. Long-term interferon-alpha treatment of children with chronic hepatitis delta: a multicentre study. J Viral Hepat 1996; 3:123–128. 131. Dalekos G, Galanakis E, Zervou E, et al. Interferon-alpha treatment of children with chronic hepatitis D virus infection: the Greek experience. Hepatogastroenterology 2000; 47:1072–1076. 132. Hyams KC, Purdy MA, Kaur M, et al. Acute sporadic hepatitis E in Sudanese children: analysis based on a new Western blot assay. J Infect Dis 1992; 165:1001–1005. 133. Hyams K, McCarthy M, Kaur M, et al. Acute sporadic hepatitis E in children living in Cairo, Egypt. J Med Virol 1992; 37:274–277. 134. Arora NK, Nanda SK, Gulati S, et al. Acute viral hepatitis types E, A, and B singly and in combination in acute liver failure in children in north India. J Med Virol 1996; 48:215–221. 135. Aradottir EAE, Shulman ST. Severe neonatal enteroviral hepatitis treated with pleconaril. Pediatr Infect Dis J 2001; 20:457–459. 136. Abzug M, Loeffelholz M, Rotbart H. Diagnosis of neonatal enterovirus infection by polymerase chain reaction. J Pediatr 1995; 126:447–450. 137. Egawa H, Inomata Y, Nakayama S, et al. Fulminant hepatic failure secondary to herpes simplex virus infection in a neonate: a case report of successful treatment with liver transplantation and perioperative acyclovir. Liver Transpl Surg 1998; 4:513–515.
138. Johnson JR, Egaas S, Gleaves CA, et al. Hepatitis due to herpes simplex virus in marrow-transplant recipients. Clin Infect Disease 1992; 14:38–45. 139. Michaels MG, Green M, Wald ER, et al. Adenovirus infection in pediatric liver transplant recipients. J Infect Dis 1992; 165:170–173. 140. Berenberg W, Nankervis G. Long-term follow up of cytomegalic inclusion disease of infancy. Pedatrics 1970; 46:403–410. 141. Cheung T, Teich S. Cytomegalovirus infection in patients with HIV infection. Mt Sinai J Med 1999; 66:113–124. 142. Kanj S, Shahara A, Clavien P, et al. Cytomegalovirus infection following liver transplantation: review of the literature. Clin Infect Dis 1996; 22:537–549. 143. Sokal EM, Melchior M, Cornu C, et al. Acute parvovirus B19 infection associated with fulminant hepatitis of favourable prognosis in young children. Lancet 1998; 352:1739–1741. 144. Pardi DS, Romero Y, Mertz LE, et al. Hepatitis-associated aplastic anemia and acute parvovirus B19 infection: a report of two cases and a review of the literature. Am J Gastroenterol 1998; 93:468–470. 145. Lloyd-Still JD, Scott JP, Crussi F. The spectrum of Epstein–Barr virus hepatitis in children. Pediatr Pathol 1986; 5:337–351. 146. Schooley R. Epstein–Barr virus (infectious mononucleosis). In: Mandell GLBJ, Dolin R, eds. Principles and practice of infectious diseases. Philadelphia: Churchill Livingstone; 2000:1599–1613. 147. Sullivan JL, Woda BA, Herrod HG, et al. Epstein–Barr virusassociated hemophagocytic syndrome: virologic and immunopathological studies. Blood 1985; 65:1097–1104. 148. Asano YYT, Suga S, Yazaki T, et al. Fatal fulminant hepatitis in an infant with human herpesvirus-6 infection. Lancet 1990; 335:862–863. 149. Harma M, Hockerstedt K, Lautenschlager I. Human herpesvirus-6 and acute liver failure. Transplantation 2003; 76:536–539. 150. American Academy of Pediatrics. Human herpesvirus 6 (including roseola) and 7. In: Pickering L, ed. 2000 Red book: report of the committee on infectious diseases. Elk Grove Village, IL: American Academy of Pediatrics, 2000:322–324. 151. Saxinger C, Polesky H, Eby N, et al. Antibody reactivity with HBLV (HHV-6) in US populations. J Virol Methods 1988; 21:199–208. 152. Li D-Y, Boitnott J, Schwarz K. Human herpes virus 6 (HHV-6); a cause for fulminant hepatic failure or an innocent bystander? J Pediatr Gastroenterol Nutr 2003; 37:95. 153. Ozaki Y, Tajiri H, Tanaka-Taya K, et al. Frequent detection of the human herpesvirus 6-specific genomes in the livers of children with various liver diseases. J Clin Microbiol 2001; 39:2173–2177. 154. Tajiri H, Tanaka-Taya K, Ozaki Y, et al. Chronic hepatitis in an infant, in association with human herpesvirus-6 infection. J Pediatr 1997; 131:473–475. 155. Broderick AL, Jonas MM. Management of hepatitis B in children. In: Tran TT, Martin P, eds. Clinical liver disease, vol. 8. Philadelphia, Pennsylvania: WB Saunders, 2004:387–401. 156. Bruguera M, Amat L, Garcia O, et al. Treatment of chronic hepatitis B in children with recombinant alfa interferon. Different response according to age at infection. J Clin Gastroenterol 1993; 17:296–299. 157. Torre D, Tambini R. Interferon-alpha therapy for chronic hepatitis B in children: a meta-analysis. Clin Infect Dis 1996; 23:131–137. 158. Brook MG, Karayiannis P, Thomas HC. Which patients with chronic hepatitis B virus infection will respond to alpha interferon therapy? A statistical analysis of predictive factors. Hepatology 1989; 10:761–763.
1447
Section XII. Inherited and Pediatric Diseases of the Liver
159. Ruíz-Moreno M, Fernández P, Leal A, et al. Pilot interferon-b trial in children with chronic hepatitis B who had previously not responded to interferon-alfa therapy. Pediatrics 1997; 99:222–225. 160. Comanor L, Minor J, Conjeevaram H, et al. Statistical models for predicting response to interferon-alfa and spontaneous
1448
seroconversion in children with chronic hepatitis B. J Viral Hepatitis 2000; 7:144–152. 161. Jonas MM. Treatment of chronic hepatitis C in pediatric patients. In: Keeffe EB, ed. Treatment of chronic hepatitis C, vol. 3. Philadelphia: WB Saunders; 1999:855–867.
Section XII: Inherited and Pediatric Diseases of the Liver
BILIRUBIN METABOLISM AND ITS DISORDERS
74
Jayanta Roy-Chowdhury, Namita Roy-Chowdhury, and Peter L.M. Jansen Abbreviations ABC ATP-binding cassette ATF-2 activating transcription factor-2 BSEP bile-salt export pump BSP bromosulfophthalein CAR the constitutive androstane receptor CO carbon monoxide DBSP dibromosulfophthalein DDT dichlorodiphenyltrichloroethane ELB “early-labeled peak” of bilirubin ER endoplasmic reticulum GSTs glutathione-S-transferases
HO HPLC ICG IL mrp2 NADPH NHANES III ntcp
heme oxygenase high-pressure liquid chromatography indocyanine green interleukin multidrug resistance-associated protein-2 nicotinamide-adenine dinucleotide phosphate third national health and nutrition examination survey Na+/taurocholate co-transporting polypeptide
INTRODUCTION Bilirubin is the degradation product of heme, the bulk of which is derived from hemoglobin of senescent erythrocytes and hepatic hemoproteins. Bilirubin is potentially toxic, but is normally rendered harmless by binding to plasma albumin, and efficient hepatic clearance. In some disease states, severe unconjugated hyperbilirubinemia can result in encephalopathy (kernicterus). Perhaps because of its distinctive color, bilirubin has attracted the attention of physicians since antiquity. Hippocrates considered it one of the four important humors of the body: blood, phlegm, black bile and yellow bile.1 The ancient Indian book of medicine, Ayurveda, also included it among the three principal factors – gases, bile and phlegm – the proper balance of which was considered critical for health. During the last two centuries, the chemistry, metabolism, and disposal of bilirubin have been investigated meticulously by generations of chemists, biologists, and clinical investigators. Excretion of bilirubin by the liver has also been studied as a model for hepatic disposal of other biologically important organic anions of limited aqueous solubility. Several inherited disorders of bilirubin metabolism and excretion have been described in humans and animals. Investigation of these inborn errors has provided important information regarding its metabolic pathways. Definitive treatment of some of these disorders continues to be a therapeutic challenge and an impetus for further research. Although bilirubin has interested physiologists mainly as a toxic metabolic product, as an antioxidant, it may serve as a defense mechanism against oxidative damage.
SHP gene short heterodimer partner TCDD tetrachlorodibenzo-p-dioxin UGT uridinediphosphoglucuronate glucuronosyltransferase UGT1A1 uridinediphosphoglucuronate glucuronosyltransferase UGT1A1 bilirubin uridine diphosphoglucuronate glucuronosyltransferase.
JAUNDICE AS AN INDICATOR OF HEPATIC DYSFUNCTION Jaundice is a sensitive indicator of liver dysfunction. As a sign and symptom, jaundice and hyperbilirubinemia are among the frequently used liver function tests. In acute hepatitis, jaundice is common and usually transient. In contrast, in other hepatocellular diseases such as alcoholic hepatitis and alcoholic or non-alcoholic liver cirrhosis and drug-induced hepatitis, jaundice has a dismal prognosis. In the intensive care unit, in septic or multitrauma patients, jaundice is associated with a high mortality rate. In primary biliary cirrhosis, jaundice is a major indicator of poor prognosis and serial serum bilirubin measurement is one of the tests used for determining the appropriate timing of liver transplantation. Impairment of bile flow caused by obstruction of the intrahepatic or extrahepatic biliary tract leads to jaundice. As this is a post-conjugation event, conjugated bilirubin accumulates in the blood. After relief of obstruction of the bile duct, jaundice usually resolves within a week, although elevated plasma bilirubin levels may linger because of the covalent binding of conjugated bilirubin to albumin. Acquired causes of hyperbilirubinemia, which include hemolysis, liver disease, or biliary obstruction, need to be differentiated from inborn errors of bilirubin metabolism. Although jaundice is a common symptom, its clinical significance varies according to the underlying disease. In some cases, a simple bilirubin determination has more clinical predictive power than a
1449
Section XII. Inherited and Pediatric Diseases of the Liver
erythropoietic porphyria.4 This phase is also increased during accelerated erythropoiesis, probably due to intramedullary breakdown of normoblasts, destruction of reticulocytes in the peripheral circulation,7 and trimming of reticulocytes during maturation.8 The “late-labeled peak,” normally comprising 80% of the radiolabel, is derived from the hemoglobin of senescent erythrocytes and is associated with the lifespan of erythrocytes (approximately 50 days in rats and 110 days in humans). When the erythrocyte lifespan is reduced, as in hemolytic syndromes, or intravascular or extravascular hemolysis, the “late-labeled peak” appears earlier.
battery of expensive diagnostic tests, including invasive techniques. However, a sound knowledge of the pathophysiology of bilirubin metabolism is required for interpretation of this simple and valuable liver function test.
FORMATION OF BILIRUBIN Breakdown of heme results in the daily production of 250–400 mg of bilirubin in humans. Normally, about 80% of bilirubin originates from the hemoglobin of senescent erythrocytes,2 and the remainder is derived from heme-containing enzymes, such as tissue cytochromes, catalase, peroxidase, and tryptophan pyrrolase, and from myoglobin. A fraction of bilirubin is also derived from free heme. After injection of radiolabeled heme precursors, glycine or daminolevulinic acid in humans or rat, radioactivity is incorporated into bile pigments in two phases.2–4 The “early-labeled peak” of bilirubin (ELB) contains 20% of the radiolabel and is excreted in bile within 3 days. The initial “fast” component of ELB comprises twothirds of the peak in humans and is largely derived from hepatic hemoproteins such as cytochromes, catalase, peroxidase, and tryptophan pyrrolase,4 and a rapidly turning-over pool of free heme in the cytosol of hepatocytes, a fraction of which may be degraded without incorporation into heme proteins.5 Induction of hepatic cytochrome P450 increases the ELB.6 As d-aminolevulinic acid is preferentially incorporated into hepatic hemoproteins, when labeled d-aminolevulinic acid is used as a precursor, only the initial component of the ELB incorporates radioactivity. The relatively “slower” phase of the ELB, which normally comprises one-third of the peak, is derived from both erythroid and non-erythroid sources. This slower phase is enhanced in conditions associated with “ineffective erythropoiesis,” such as congenital dyserythropoietic anemias, megaloblastic anemias, iron-deficiency anemia and lead poisoning, and
Iron protoporphyrin IX (HEME) a
V
OPENING OF THE HEME RING BY HEME OXYGENASE Heme (ferroprotoporphyrin IX) is a ring of four tetrapyrroles connected by methene bridges (Figure 74-1). The ring is opened by cleavage of the a-methene bridge, catalyzed by microsomal heme oxygenase (HO). Initially, an electrophilic attack at Fe (II) by a reducing agent, such as nicotinamide-adenine dinucleotide phosphate (NADPH), and oxygen, results in the formation of aoxyheme (Figure 74-1).9 Subsequently, the a-methene bridge carbon is eliminated as carbon monoxide (CO) and the porphyrin ring carbons that flank the a-methene bridge are oxidized, utilizing two additional oxygen molecules, resulting in the two lactam oxygens of biliverdin and bilirubin.10 Iron is released from the open tetrapyrrole after addition of electrons, suggesting that conversion of ferric to ferrous iron is required.11 Only a minute fraction of heme is opened at the b, g, or d bridges, resulting in the excretion of traces of bilirubin IXb, IXg, and IXd in bile. HO catalyzes physiological heme degradation. It consists of three structurally related isozymes, HO-1, HO-2, and HO-3. HO-1 is the inducible form, HO-2 is a constitutive isoform, and HO-3 is a minor isoform present in spleen, liver, thymus, heart, kidney, brain, and Figure 74-1. Mechanism of heme ring opening and subsequent reduction of biliverdin to bilirubin. The a-carbon bridge, marked with an interrupted arc, is the site of cleavage catalyzed by heme oxygenase.
Bilirubin IXa
M
V
M OO
M
A
D N
d
V
C g
M V
Heme Oxygenase (NADPH + O2) M = CH3 V = CH = CH2 P = CH2 CH2 COOH
M
M
A
D NH
HN
NH
HN H C H g
C
M
P
V HC b
NH
N
B P
V
HN
d CH M
B P
OO CO
H
1450
M
P Fe
D NH
HC b
N
B P
A
d CH
b
Fe N
M
M
N
C C H g
P
Biliverdin IXa
M
Biliverdin reductase (NADH or NADPH)
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
testis.12 HO-2 is the isoform in hepatocytes and spleen responsible for biliverdin and CO production under normal physiological conditions. High levels of HO-2 activity are present in cells involved in the breakdown of hemoproteins, such as the spleen, where senescent erythrocytes are sequestered. In the liver, both hepatocytes and Kupffer cells have HO activity; the activity in the Kupffer cells is as high as in the spleen.13 Apart from breakdown of heme from circulating hemoglobin, constitutive HO-2 is important for cellular hemoprotein homeostasis. HO-2 also appears to function as an oxygen sensor in the lungs. HO-2-deficient mice are severely hypoxic and data suggest that HO-2 is responsible for matching the ventilation to perfusion. HO-1 is a 32-kDa protein that is induced by lipopolysaccharide, cytokines, heavy metals, reactive oxygen species, protoheme IX, oxidized low-density lipoprotein, hypoxia, and probably also by shear stress in endothelial cells in the cirrhotic liver.14,15 NF-kappaB and p38 MAPK signaling pathways mediate the lipopolysaccharide-dependent induction of HO-1 gene expression via DNA sequences in the proximal promoter region. HO-1 acts as a stress-response protein and, by converting pro-oxidant heme to antioxidant biliverdin and bilirubin, it plays a role in the cellular defense against oxidative injury. A prerequisite for this antioxidant action is that toxic ferrous iron that is released upon cleavage of the porphyrin ring is efficiently scavenged by ferritin. HO-1 deficiency in humans is associated with growth retardation, hyperlipidemia, endothelial cell damage with consumption coagulopathy, and microangiopathic hemolytic anemia. Inducible HO-1 is particularly important for cytoprotection in vascular endothelium and renal tubular epithelium. HO-1-deficient mice show early atherosclerosis in particular when they are also hypercholesterolemic. In addition to these cellular effects, it should be realized that the products of the HO reaction, CO and bilirubin, may have more distant effects. CO is a signaling molecule with vasodilatory effects16 and effects on intestinal motility and sphincters.17 Moreover, a possible negative correlation between circulating bilirubin levels and coronary heart disease has been reported. Whether this protective effect of bilirubin also holds for patients with Gilbert’s syndrome is unclear. Binding of heme to HO requires the propionic acid substituents in the C-6 and C-7 positions and a metal, such as iron, tin, or zinc. Oxygen binds to ferrous heme and undergoes reductive activation. Non-iron metalloprotoporphyrins, such as tin- and zinc-protoporphyrin, bind HO with even greater affinity, but do not activate oxygen and are therefore not degraded by HO. These metalloporphyrins are dead-end inhibitors of heme degradation.18 Tin- and zinc-protoporphyrins also disrupt the integrity of HO-2, but not of HO-1. The loss of integrity of HO-2, the more abundant form of HO, may partly account for the suppression of bilirubin formation by tin-protoporphyrin.
CONVERSION OF BILIVERDIN TO BILIRUBIN The immediate product of HO-mediated ring opening is the green pigment biliverdin, which is the major bile pigment in many amphibian, avian, and fish species. In most mammals, biliverdin is converted to the orange pigment, bilirubin. Being less polar, bilirubin crosses placental membranes more readily than does biliverdin,19 although some placentate animals, such as nutria and rabbits, excrete biliverdin as the main bile pigment.20 Conversely, bilirubin forma-
tion has been found in early vertebrates, such as teleost and elasmobranch fish,21,22 that precede the evolution of the placenta. Reduction of biliverdin to bilirubin is catalyzed by biliverdin reductase (Figure 74-1), a family of cytosolic enzymes that utilize NADH at pH 6.7 and NADPH at pH 8.5 as cofactors. Guinea pig liver biliverdin reductase is a 70-kDa protein.23 Several interconverting forms of biliverdin reductase found in rat liver and spleen are produced by tissue-specific post-translational modification of a single gene product.24 Recently biliverdin reductase was shown to induce activating transcription factor-2 (ATF-2). ATF-2 controls the transcription of the HO-1 gene. This thus provides for a regulatory loop in which HO-1 and biliverdin reductase expression are interdependent.
Potential Beneficial Effects of Products of Heme Breakdown A number of natural products that were previously considered to be mere waste products now appear to double as agents with potent biological activity. Both biliverdin and bilirubin have strong antioxidant activities, which may be particularly important in the newborn period, when the level of other natural antioxidants is low. Biliverdin appears to attenuate graft rejection in both cardiac and smallintestine transplant models.25,26 Bilirubin is also a strong antioxidant and has been reported to have cytoprotective activity, although at higher concentrations it is neurotoxic. The evolutionary development and conservation of the energetically expensive mechanisms of bilirubin production and elimination suggest a physiological benefit for bilirubin. An inverse relationship between serum bilirubin levels and risk of coronary artery disease supports this hypothesis,27 although whether such protective effect extends to subjects with Gilbert syndrome is questionable.28 Analysis of over 176 million individuals in the Third National Health and Nutrition Examination Survey (NHANES III) in the USA showed a marked inverse relationship between serum bilirubin concentrations and history of colorectal cancer.29 The odds ratios for colorectal cancer per 1 mg/dl increase in serum bilirubin levels for men and women were 0.295 (confidence interval 0.291–0.299) and 0.186 (confidence interval 0.183–0.189). This observation is consistent with the analysis of a previous large study showing an inverse relationship between serum bilirubin levels and cancer mortality in a Belgian population.30 However, despite the impressive difference in odds ratios, such database analyses do not establish conclusively a causeand-effect relationship, because of the possible existence of known and unknown confounding variables.
MEASUREMENT OF BILIRUBIN PRODUCTION At steady state, bilirubin production equals the synthesis and breakdown of biologically important hemoproteins. Because bilirubin is almost quantitatively excreted in bile, bilirubin production can be measured by determination of its biliary excretion in bile ductcanulated experimental animals. A portion of the unconjugated bilirubin excreted in bile may undergo enterohepatic cycling. This may become important in patients with terminal ileum dysfunction, such as in Crohn’s disease, wherein bile acids spill over into the cecum where they solubilize deconjugated bilirubin, which may be reabsorbed.31,32 In humans, and in animals with an intact entero-
1451
Section XII. Inherited and Pediatric Diseases of the Liver
hepatic circulation, bilirubin production is estimated by quantification of fecal and urinary urobilinogen and stercobilinogen, the bacterial degradation products of bilirubin.33 A more convenient method utilizes the turnover of radioisotopically labeled bilirubin. Following intravenous injection of radiolabeled albumin-bound bilirubin, blood samples are collected at frequent intervals, and plasma bilirubin concentration and radioactivity are measured.34 Plasma bilirubin clearance (the fraction of plasma from which bilirubin is irreversibly extracted) is calculated from the area under the radiobilirubin disappearance curve.35 Bilirubin removal is measured as the product of plasma bilirubin concentration and clearance. At steady-state levels of plasma bilirubin, the removal rate of bilirubin equals the rate of its synthesis. This method does not take into account a small portion of bilirubin that is produced in the liver and excreted directly into bile without appearing in the circulation and, therefore, slightly underestimates bilirubin production. Because heme ring opening is the main source of endogenous CO production, bilirubin formation can also be quantified by measuring CO production. This is done by placing a subject in a closed rebreathing system to prevent CO excretion. In the absence of anoxia, the body CO stores rapidly equilibrate with the CO in the rebreathed air. Therefore, CO production can be calculated from the CO concentration in the breathing chamber or from an increment in blood carboxyhemoglobin saturation.36 A fraction of CO in expired air may be derived from non-heme sources, e.g., halogenated methane and polyphenolic compounds, including catecholamines.37 In addition, intestinal bacteria contribute a small fraction of the CO.38 Therefore, CO production exceeds plasma bilirubin turnover by 12–18%.
CHEMISTRY OF BILIRUBIN The systemic name of bilirubin IXa is 1¢8¢-dioxo-1,3,6,7-tetramethyl-2,8-divinylbiladiene-a,c-dipropionic acid. 39 The planar chemical structure of bilirubin, as determined by Fischer and Plieninger,40 was confirmed by X-ray diffraction analysis.41 The carbon bridges between pyrrolenone rings A and B (C-4), and C and D (C-15) are in trans- or Z configuration. The oxygen attached to the outer pyrrolenone ring is in a lactam rather than lactim configuration. Titration of bilirubin in aqueous solutions suggests a pK value of 7–8.42 Since bilirubin tends to form insoluble aggregates below pH 8.0, determination of pK by titration of aqueous solutions of bilirubin can be misleading.37 The pK values for bilirubin have been determined by 13C nuclear magnetic resonance spectra, and potentiometric and spectrophotometric titrations in aqueous solutions. These studies indicate that the pK value of the two carboxyl groups is 4.4 and that of the two lactam groups is 13.0.43
1452
bin IXa-Z, with two protonated carboxyl groups, is virtually insoluble in water. At acidic, neutral, or mildly alkaline pH, bilirubin partitions from aqueous solutions to water-immiscible solvents such as chloroform, ethyl acetate, or methylethyl ketone. Unconjugated bilirubin appears to stain adipose tissues. Although these observations suggest that bilirubin is lipophilic, the pigment is readily soluble in polar solvents, provided the intramolecular hydrogen bonds can be disrupted.45 The apparent yellow discoloration of adipose tissues or brain during hyperbilirubinemia results from the aggregation of bilirubin at the surface of phospholipid membranes, rather than its presence in the fat. Bilirubin and polar ligands, such as sulfonamides, share a binding site in the polar region of albumin43 with other polar substances, such as sulfonamides.46 Therefore, despite its insolubility in water at physiological pH, bilirubin should be considered a relatively polar substance, the mechanism of toxicity of which may differ from that of truly lipophilic toxins, such as dichlorodiphenyltrichloroethane (DDT).46 Despite the presence of two propionic acid side chains, four amino groups, and two lactam oxygens, bilirubin IX is insoluble in water at physiologic pH. A possible explanation for this paradox was suggested by Fog and Jellum47 and Kuenzle et al.,48,49 who proposed that bilirubin IXa may be internally stabilized by hydrogen bonding between the carboxyl and the two external pyrrolenone rings (Figure 74-2). X-ray diffraction studies of crystalline bilirubin confirm hydrogen bonding between each propionic acid side chain and the pyrrolic and lactam sites in the opposite half of the molecule.41 These hydrogen bonds constrain the molecule into a “ridge-tile” conformation, in which the two dipyrrolic halves of the molecule lie in two different planes with an interplanar angle of 98–100o (Figure 74-2). The integrity of the hydrogen-bonded structure requires the interpyrrolic bridges at the 4 and 15 position of bilirubin to be in trans- or Z configuration.50 Because both carboxylic groups, all NH groups, and the two lactam oxygens are engaged by hydrogen bonding, the molecule is insoluble in water. Addition of methanol, ethanol, or 6 mol/l urea interferes with the hydrogen-bonded structure and makes bilirubin more labile,49 water-soluble, and rapidly reactive with diazo reagents. As bilirubin IXb, IXg, and IXd lack the internally hydrogen-bonded structure, these are more polar than is bilirubin IXa and are readily excreted in bile in the unconjugated form following intravenous injection in rats.51 In contrast, bilirubin IXa requires conversion to a polar molecule before excretion. Conjugation of the propionic acid carboxyls with sugar moieties disrupts the hydrogen bonds, resulting in the formation of water-soluble conjugates that are readily excreted in bile. Resonance Raman spectroscopic studies of bilirubin–sphyngomyelin complexes suggest that the intramolecular hydrogen bonds are disrupted in such complexes, and the propionic acid carboxyls form ion pairs with the quaternary ammonium ion of the choline moiety of sphyngomyelin.52
PHYSICAL CONFORMATION AND SOLUBILITY OF BILIRUBIN IXa
ABSORPTION SPECTRA AND CIRCULAR DICHROISM
Precise determination of the solubility of bilirubin IXa is difficult because the pigment is unstable in aqueous solutions and tends to form dimers, colloids, or surface films.37 Lee and Gartner propose that bilirubin toxicity may result from interaction of the colloidal solution of bilirubin with the surface of cells.44 Crystallized biliru-
Bilirubin IXa has a main absorption band at 450–474 nm in most organic solvents with an extinction coefficient of 48.0–63.4/mmol/l at its absorption maximum at 1 cm path length. In alkaline aqueous solutions, there is a 10–30-nm shift of the absorption band toward shorter wavelengths (hypsochromic shift) and a weaker absorption
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
Me O
98-100º
N H
C O CH2
0
CH2
Me
A NH Vn O
H N C
B N H
Me
Vn
D
H O
Me
C H2 CH2 O HO
C
CH2
Figure 74-2. X-ray crystallographic structure of bilirubin. The linear structure shown in Figure 74-1 is contorted into a ridge-tile-like configuration caused by internal hydrogen bonding (interrupted line) of the propionic acid carboxyls to the amino groups and the lactam oxygen of the pyrrolenone rings of the opposite half of the molecule. The carbon bridges connecting pyrrolenone rings A and B (C-4) and C and D (C-15) are in the Z (trans-) configuration. Because the polar groups, the propionic acid carboxyls, and the amino and lactam groups are all engaged by hydrogen bonding, bilirubin is very sparsely soluble in water. The hydrogen bonding bends the molecule along the central carbon bridge (C-10) and buries the central bridge deep within the molecule, thereby restricting the access of diazo reagents to the central bridge. Therefore, bilirubin reacts very slowly with diazo reagents, unless the hydrogen bonds are opened by adding “accelerator” reagents (“indirect diazo reaction,” discussed later in the text).
band at 280–300 nm. The main absorption band of bilirubin derivatives in which hydrogen bonds are disrupted (e.g., bilirubin IXa dimethylester) is much more sensitive to change in solvents.53 Circular dichroism spectroscopy shows that biliverdin preferentially adopts a minus-helicity conformation when bound to human serum albumin, whereas bilirubin-IXa prefers a plus-helicity. Reduction of human serum albumin-bound biliverdin to bilirubin results in a conformational inversion from minus to plus helicity.54 Such sign inversion also occurs on addition of halothane, chloroform, or other volatile anesthetics to the albumin–bilirubin complex, suggesting that the volatile anesthetics alter the internal topography of receptor sites, influencing the stereoselectivity of ligand binding.55
PHOTOCHEMISTRY OF BILIRUBIN Conformational Isomerization and Cyclization (Figure 74-3) Exposure of circulating bilirubin to light changes the configuration of one or both of the interpyrrolic bridges at positions 5 and 15 from Z (trans-) to E (cis-) configuration. The resulting 4Z-15E or 4E-15Z
isomers lack hydrogen bonds in one half of the molecule, while bilirubin IXa-EE lacks hydrogen bonds on both halves of the molecule. Of these conformations, the 4Z-15E isomer is more abundant.56 Blue light is more efficient in mediating the conformational changes. The conversion of the E forms back to the Z forms may be reduced in serum albumin-bound bilirubin by competitive internal conversion processes.57 Following absorption of two photons,56 the vinyl substituent at position C3 of bilirubin IXa-4E-15Z is cyclized with the methyl substituent on the internal pyrrole ring forming the structural isomer, E-cyclobilirubin, or lumirubin.58,59 Although cyclization of bilirubin occurs at a slower rate than formation of configurational isomers, because of the relative stability of cyclobilirubin, this form may be quantitatively more important in phototherapy of neonatal jaundice.58 The conformational isomers are more polar than is bilirubin IXa-ZZ and can be excreted in bile without conjugation.60
Fluorescence Although pure bilirubin does not fluoresce, when dissolved in detergents, albumin, or alkaline methanol it exhibits intense fluorescence at 510–530 nm,61 which has been used for quantification of blood bilirubin concentrations and the unsaturated bilirubin-binding capacity of albumin.
Photo-Oxidation and Degradation In the presence of light and oxygen, bilirubin undergoes a selfsensitized reaction involving singlet oxygen, resulting in the formation of colorless fragments, chiefly maleimides and propentdyopent adducts.62 A small amount of biliverdin is also formed.
Dismutation When bilirubin IXa is irradiated in deoxygenated aqueous solution, free radical disproportionation results in formation of the symmetrical molecules, bilirubin IIIa and bilirubin XIIIa, which are nonphysiologic isomers of bilirubin.63 The reaction is enhanced in the presence of acid and oxygen and is inhibited by ascorbate.
BILIRUBIN TOXICITY Cerebral toxicity of bilirubin in neonatal jaundice has been known for at least five centuries. Degeneration of brain tissues associated with yellow pigmentation was reported in 1949.64 The association of kernicterus, or bilirubin-induced encephalopathy with severe unconjugated hyperbilirubinemia, was subsequently established.65 The study of mutant rats (Gunn strain) that lack hepatic bilirubin glucuronidating activity has contributed greatly to the current knowledge of bilirubin toxicity. While the cerebral toxicity of bilirubin has been studied in detail, other organ systems are also affected.
BILIRUBIN-ENCEPHALOPATHY IN GUNN RATS Bilirubin deposition at specific areas of the brain accompanied by structural damage is termed kernicterus. The Gunn rat is the only animal model in which spontaneous bilirubin-induced brain damage has been observed. Normally, albumin binding inhibits bilirubin deposition in the brain. Displacement of bilirubin from albumin-binding sites by drugs such as salicylates or sulfonamides increases bilirubin
1453
Section XII. Inherited and Pediatric Diseases of the Liver
Bilirubin IXa-4Z,15Z HOOC
COOH
4
HOOC
10
N H
O
Bilirubin IXa-4Z,15E
15
N H
4
N H
N H
O
10
N H
O
COOH
N H
N H
N H
15
O
H H O
O C
O H
4
O
N
N
15
10
H
10 N
O C
O
O
HOOC
H O C
H
O
Lumirubin
COOH
HOOC
10 N H
15
N H
4
N H
O C
COOH
10 N H
O
N H
O
15
N
4
N H
H O
O C
O H N H
N H
O H N H
15
N
H
H
O H O
10
4 O C
15 N
10
4 N
O
H O
N
O
15
H
Bilirubin IXa-4E,15Z
O
H
N
H O
N
H
4
N
H
O
H
N N
O
C
N H
N
O H O
C
Figure 74-3. Photoisomerization of bilirubin IXa. The linear (upper row) and hydrogen-bonded structures of bilirubin IXa and its photoisomers are shown. Bilirubin IXa-4Z,15Z is the preferred configuration of unconjugated bilirubin, in which carbon bridges at both 4 and 15 position are in Z or trans-configuration, allowing hydrogen bonding on both halves of the molecule. Upon exposure to light, configurational isomers, 4Z,15E and 4E,15Z, are formed. As shown in the figure, there is disruption of hydrogen bonds in the dipyrrolic half of the molecule that bears an E configuration. The 4E,15Z isomer becomes cyclized to produce the stable structural isomer, lumirubin, which, because of its stability, is quantitatively the most important photoisomer of bilirubin formed during phototherapy.
1454
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
accumulation in the brain and may precipitate kernicterus.66 Gunn rats that are rendered genetically analbuminemic by cross-breeding with Nagase analbuminemic rats have serum bilirubin levels that are only 25% that of other Gunn rats, whereas their cerebral bilirubin content is 1.2–2.7-fold higher.67 Such hybrid rats die within 3 weeks of birth. Therefore, for clinical purposes, it is important to calculate the molar ratio between plasma albumin and bilirubin. However, the plasma free bilirubin level does not correlate well with brain bilirubin concentration68 and it is not certain whether unbound bilirubin is the only toxic species of the pigment. Abnormalities of the cochlear nuclei, associated with various degrees of hearing deficiency, commonly occur as a complication of neonatal hyperbilirubinemia. Brainstem auditory evoked potential studies in Gunn rats indicate functional abnormalities of the central auditory pathways at and rostral to the cochlear nuclei69 beginning at 17 days of age. Similar changes are found in human neonates with severe hyperbilirubinemia. Sulfonamides displace bilirubin from albumin binding, thereby promoting the net transfer of bilirubin into neural tissues. Administration of sulfonamides results in reversible brainstem auditory evoked potential abnormalities in Gunn rats.70 Under these conditions, focal bilirubin staining occurs in Purkinje cells of the cerebellum, hippocampus, and basal ganglia. Similar changes occur in human infants with fully developed kernicterus. A large number of Purkinje cells are affected in the cerebellum of Gunn rats at the age of 7 days; most of these cells degenerate and disappear between days 12 and 30, resulting in cerebellar hypoplasia.71 The remaining Purkinje cells recover and persist into adult life. However, synapse formation among these Purkinje cells or with other neural cells may be abnormal. Cerebellar mitochondria are enlarged and distorted in Gunn rats.72 Increased activities of the lysosomal enzymes, arylsulfatase and cathepsin, occur in the cerebellum of Gunn rats by the 8th day of life.73 Cerebellar cyclic guanosine monophosphate concentrations decrease progressively from day 15 to day 30, but cyclic adenosine monophosphate (cAMP) levels remain normal.74
CLINICAL FEATURES OF BILIRUBIN ENCEPHALOPATHY Except in patients with inherited deficiency of bilirubin glucuronidation (see below), the occurrence of kernicterus is usually limited to the neonatal period and the first few months of life. Bilirubin encephalopathy may present with a broad spectrum of neurological features. In the most severe cases, overt kernicterus presents between the third and sixth day of life. The normal Moro reflex is lost, the muscles become hypotonic, the cry is high-pitched, athetoid movements appear, and reflex opisthotonus occurs in response to a startling stimulus. This may progress to lethargy, atonia, and death. Occasionally, in some children with Crigler–Najjar syndrome type 1, bilirubin encephalopathy may present late with cerebellar symptoms as the presenting feature.75 Those who survive acute kernicterus may develop chronic hearing abnormalities, athetoid movements, paralysis of upward gaze, and mental retardation, in various combinations. The cochlear nucleus is commonly affected by hyperbilirubinemia. Cells of the auditory system that receive synaptic input from end-bulbs or calyces appear to be early targets.76 In Gunn rat pups, these morphological changes correlate
with abnormalities of brainstem auditory evoked potentials.77 Sensitivity of auditory evoked potential testing can be increased by recording binaural difference waves obtained by subtracting the sum of two monaural brainstem auditory evoked potentials from a binaural brainstem auditory evoked potential.78 Bilirubin staining of the hippocampus, basal ganglia, and nuclei of the cerebellum and brainstem occurs in infants dying in the acute phase of bilirubin encephalopathy;79 however, such staining is not found in children dying in the chronic stage of the disorder. Clinical manifestations precede histological evidence of brain damage by approximately 72 hours.80 Focal necrosis of neurons and glial cells occurs later. Gliosis of the affected areas is seen in chronic cases.80 Because these histological lesions are not present from the onset of clinical kernicterus, these may not be the initiating pathophysiological events in bilirubin-induced brain damage. Non-specific signs of encephalopathy in the neonate may result from other causes, such as cerebral hemorrhage,81 and therefore kernicterus cannot always be diagnosed without pathological documentation. Conversely, focal bilirubin staining of the brain may occur in other forms of brain injury. Thus, in the absence of neuronal degeneration, bilirubin staining alone does not establish the diagnosis of classic kernicterus.82 Peak serum bilirubin levels of up to 10–12 mg/dl are usually considered safe. The prognostic significance of a moderate degree of hyperbilirubinemia is not entirely clear. Serum bilirubin levels that are not high enough to cause kernicterus have been reported to result in an increased incidence of neurologic abnormalities or decreased intellectual performance later in life.83
THE BLOOD–BRAIN BARRIER AND CEREBRAL BILIRUBIN CLEARANCE The equilibration of hydrophilic water-soluble substances and proteins between the blood and the brain is restricted by a functional blood–brain barrier.84 Tight junctions between capillary endothelial cells and foot processes of astroglial cells represent the structural component of this barrier.84 Specific transport mechanisms that are involved in the translocation of ions, water, and nutrients from plasma to brain may provide the functional counterpart of the blood–brain barrier. Conventionally, immaturity of the blood–brain barrier in neonates has been implicated in the high incidence of kernicterus in neonates. However, it has been difficult to confirm a more rapid passage of labeled markers85 or lipophilic substances86 into the immature brain. Therefore, there is no firm evidence to support the concept of an immature blood–brain barrier in the neonate. The efficiency of cerebral bilirubin clearance may be inversely related to the cerebral toxicity of bilirubin. Experimentally, the blood–brain barrier can be unilaterally and reversibly opened without causing brain damage by infusion of hypertonic urea87 or arabinose.84 The hyperosmolarity-associated shrinkage of the capillary endothelial cells results in temporary opening of the tight junctions. When the blood–brain barrier is opened in newborn rats by this technique, intravenously administered albumin-bound bilirubin rapidly enters the brain. Following the reversal of the blood–brain barrier, bilirubin is rapidly cleared from the brain. The clearance of bilirubin from brain parallels its clearance from serum, suggesting that bilirubin is cleared by diffusion or actively transported back into
1455
Section XII. Inherited and Pediatric Diseases of the Liver
the general circulation.88 Damaged brain, which may be edematous, may bind bilirubin89 to such an extent that it is unable to clear bilirubin rapidly. The damaged brain, therefore, may be more vulnerable to bilirubin toxicity. Endothelial cells of cerebral microvessels and the choroid plexus together form the blood–brain barrier and cerebrospinal fluid–blood barrier. Recent studies show that these endothelial cells are richly supplied with adenosine triphosphate (ATP)-dependent transporters belonging to the ATP-binding cassette (ABC) superfamily of transporting proteins. Notably, MDR1 (multidrug-resistant 1, a Pglycoprotein) and the multidrug resistance-associated proteins MRP1, MRP4, MRP5 and, to a lesser extent, MRP6 are expressed in these tissues.90–93 The MRPs located in the microvascular endothelial cells and the basolateral membranes of the choroid plexus epithelial cells act as efflux pumps stimulating the efflux of drugs from the brain and the cerebrospinal fluid to the blood.91–93 Mrp1 is preferentially expressed in the astroglial component of the blood–brain barrier. Recent evidence indicates that bilirubin is a substrate for Mrp1 (ABCC1). According to these concepts the blood–brain barrier is not merely a passive anatomical structure, but is an active tissue that can pump bilirubin and other metabolites and drugs out of the brain, thus reducing their intracellular concentrations. Substrate competition at the level of these pumps may be another way of looking at the importance of drug exposure and bilirubin brain toxicity.
BIOCHEMICAL BASIS OF BILIRUBIN TOXICITY Bilirubin has a broad range of toxicity on organs, cells, subcellular organelles, and cellular functions. It is not clear which of these toxic effects are relevant in bilirubin encephalopathy. Bilirubin may also uncouple oxidative phosphorlyation and inhibit ATPase activity of brain mitochondria.94 Bilirubin reduces local cerebral metabolic rates for glucose in immature rats.95 All areas of the brain are affected, but decreased glucose utilization is particularly pronounced in auditory, visual, hypothalamic, and thalamic regions. The finding of large glycogen vacuoles in neonatal Gunn rat brains,69 reduced brain pyruvate concentration, and decreased activity of glycolytic enzymes96 suggest an inhibition of glycogenolysis by bilirubin. Decreased choline acetyltransferase activity has been observed in substantia nigra and amygdala of Gunn rat brain, whereas the enzyme activity was increased in the olfactory bulb.97 Bilirubin has been shown to inhibit cAMP-dependent protein kinase activity in vitro98 and noncAMP-dependent protein kinase activity in vivo.99 In a cell-free system, bilirubin irreversibly inhibits Ca+-activated, phospholipiddependent protein kinase C activity and cAMP-dependent protein kinase activity.100 Since protein phosphorylation is the final common pathway of neuronal signal transmission, inhibition of protein kinase C by bilirubin may play a role in bilirubin encephalopathy in the newborn.
BILIRUBIN NEPHROTOXICITY Renal medullary deposition of unconjugated bilirubin results in medullary necrosis and formation of bilirubin crystals on the renal papillae101 in Gunn rats and in hyperbilirubinemic infants.102 In adult Gunn rats, abnormality of the ascending loop of Henle101 leads to an impairment of urinary concentration, which is ameliorated by reduction of serum bilirubin levels.103 The urinary concentration
1456
defect has not been found in mature neonates with hyperbilirubinemia104 or in patients with Crigler–Najjar syndrome type 1 who survive to adult age.105
DISPOSITION OF BILIRUBIN Bilirubin is carried in circulation bound to plasma proteins, predominantly to albumin. In the hepatic sinusoids, the albumin–bilirubin complex dissociates and bilirubin is internalized. Bilirubin is stored in the hepatocytes bound to cytosolic proteins. Bilirubinuridinediphosphoglucuronate glucuronosyltransferase (UGT1A1) located in the endoplasmic reticulum (ER) mediates the conjugation of bilirubin with glucuronic acid. Conjugated bilirubin first needs to cross the endoplamic reticulum membrane to be transported across the bile canalicular membrane into the bile. This multistep process is summarized in Figure 74-4. Many of these steps are shared by other organic anions and cholephilic metabolites. These processes are briefly discussed below.
THE ROLE OF ALBUMIN Bilirubin is carried in the circulation bound to plasma albumin. As discussed above, unconjugated bilirubin is only sparingly soluble in water at physiological pH. Binding to albumin prevents its precipitation and deposition in tissues, thereby facilitating the transport of bilirubin from its site of production to the organ of elimination, the liver. Albumin-binding prevents bilirubin from entering the brain. The reserve binding capacity of albumin acts as a buffer by accommodating for sudden rises in serum bilirubin concentrations, e.g., during acute hemolysis. The endothelial lining of the hepatic sinusoids is fenestrated and the albumin-bilirubin complex entering the liver through the portal circulation passes through the fenestrae to reach the space of Disse, where bilirubin comes in direct contact with the sinusoidal and basolateral plasma membrane domains of the hepatocyte (Figure 74-4). Albumin enables bilirubin to get across the so-called “unstirred” water layer, a thin layer of water that surrounds the hepatocytes. By gradually releasing ligands to hepatocytes along the acinus, albumin distributes the ligands to all zones of the acinus, thereby establishing a considerable metabolic reserve. Bilirubin, but not albumin, passes into the hepatocyte, indicating that bilirubin dissociates from albumin close to or at the hepatocyte surface. It is unclear whether or not an albumin receptor at the hepatocyte surface facilitates this dissociation.
Bilirubin Binding Sites and Competition by Other Ligands Binding Sites. There is a primary and a secondary binding site on albumin for bilirubin.106 Affinity labeling studies indicate that the primary binding site of bilirubin is located at residues 124–297 and, to a lesser extent, at residues 446–547.107 The lysine 240 in human albumin and lysine 238 in bovine serum albumin appear to be the sites of bilirubin binding.108 Binding Capacity and Effect of Competitive Substrates. Albumin inhibits the neurotoxic effects of unconjugated bilirubin following intravenous injection in puppies.109 The role of albumin in preventing bilirubin toxicity is clearly shown in analbuminemic–Gunn
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
Liver Sinusoid
Hepatic sinusoidal endothelial cell
Bilirubin + Albumin
Bilirubin – Albumin
GSTs Storage
SLC21A6 (?) Uptake
Contiguous surface
UDPGA
Space of Disse
Bilirubin Tight junction
Canalicular (apical) surface
UGT1A1 Conjugation
Sinusoidal (basolateral) surface
UDP
ABC-C2 (MRP2) Excretion
Bilirubin glucuronides
MRP3 Transport into plasma during cholestasis
Figure 74-4. A schematic diagram summarizing bilirubin metabolism by hepatocytes. During its carriage in plasma, bilirubin is strongly, but reversibly, bound to albumin. In hepatic sinusoids, this complex comes in direct contact with the basolateral domain of the hepatocyte plasma membrane through fenestrae of the specialized hepatic endothelial cells. At the hepatocyte surface, dissociation of the albumin–bilirubin complex occurs, and bilirubin enters hepatocytes by a specific uptake mechanism(s). A fraction of the bilirubin is also derived from catabolism of hepatocellular heme proteins. Storage within the hepatocyte is accomplished by binding of bilirubin to a group of cytosolic proteins, glutathione-S-transferases (GSTs) (also termed ligandin or Y-protein). Binding to these proteins keeps bilirubin in solution and inhibits its efflux from the cell, thereby increasing the net uptake. Conjugation of bilirubin in the endoplasmic reticulum is catalyzed by bilirubin uridinediphosphoglucuronate glucuronosyltransferase (UGT1A1), forming bilirubin monoglucuronide and diglucuronide. Both conjugates may bind to GSTs in the cytosol. Conjugation is obligatory for efficient transport across the bile canaliculus. Bilirubin glucuronides are secreted across the bile canaliculus by an energy-consuming process mediated by MRP2 (previously termed cMOAT). This process is normally rate-limiting in bilirubin throughput, and is shared by other organic anions, but not bile salts.
hybrid rats, which die of bilirubin encephalopathy shortly after birth. Binding of other ligands to albumin affects its bilirubin binding capacity. The other ligand may bind at the same site as does bilirubin, resulting in competitive displacement, or bind at a different site, causing non-competitive displacement of bilirubin. Non-competitive binding may not affect bilirubin binding or may produce conformational changes which enhance (cooperative binding) or decrease (anti-cooperative) bilirubin binding. Sulfonamides, antiinflammatory drugs, and contrast media used for cholangiography displace bilirubin competitively from albumin.110 Prophylactic use of sulfonamides in newborn babies enhances bilirubin encephalopathy,111 probably by enhancing the dissociation of bilirubin from albumin, thereby increasing the net uptake of bilirubin into neural tissues.112 “Reserve” Binding Capacity. Binding of bilirubin to albumin is normally reversible. Because of the influence of many metabolites and drugs on albumin binding of bilirubin and its transfer from plasma to the central nervous system, measurement of plasma bilirubin concentration does not accurately estimate the risk of brain damage from unconjugated bilirubin. Unbound bilirubin in serum has been quantified by gel chromatography,113 peroxidase treatment,114 electrophoresis on cellulose acetate,115 and fluorimetry of serum with or without detergent treatment.116 Free bilirubin concentration may be roughly estimated as the product of serum bilirubin concentration, concentration of reserve bilirubin binding sites on albumin, and the association constant for bilirubin.117 Competitive binding by a 14C-labeled ligand, monoacetyl-4,4¢-diaminodiphenyl
sulfone, or a spin-labeled ligand 1-N-(2,2,6,6-tetramethyl-1-oxyl-4piperidinyl)5-N-(1-aspartate)-2,4-dinitrobenzene118 has been used to determine reserve binding capacity. Irreversible Binding of Bilirubin to Albumin. During prolonged conjugated hyperbilirubinemia, bilirubin becomes covalently bound to albumin.119 This fraction of bile pigments is not cleared by the liver or kidney, and persists for a long time in circulation, reflecting the long half-life of serum albumin.
Bilirubin Metabolism in Genetic Analbuminemia In view of this discussion it is interesting to note that analbuminemia is compatible with life and that elimination of amphipathic compounds such as bilirubin and bromosulfophthalein (BSP) is disturbed to a relatively minor degree in analbuminemic rats.120,121 For example, elimination of a tracer dose of bilirubin was entirely normal, although the biliary recovery after a loading dose was decreased.121 Other plasma proteins, such as high-density lipoproteins, can take over some of the assigned roles of albumin.
HEPATIC BILIRUBIN UPTAKE To be taken up in the liver efficiently, bilirubin needs to be delivered to hepatocytes via the sinusoidal blood. In the presence of portosystemic collaterals that develop in patients with portal hypertension, or surgically produced portosystemic shunts, bilirubin generated in the spleen is diverted to the systemic circulation. In these circumstances, first-pass clearance of bilirubin by the liver does not occur, resulting in mild hyperbilirubinemia. Similarly, an open
1457
Section XII. Inherited and Pediatric Diseases of the Liver
Table 74-1. Organic Anion Uptake Transporters in the Basolateral Membrane of Human Hepatocytes
Bilirubin Bromosulfophthalein Taurocholate Estrone-3-sulfate Estradiol17b-glucuronide Dehydroepiandrosterone sulfate Ouabain Digoxin Pravastatin N-methyl quinine Leukotriene C4 Prostaglandin E2 Tissue distribution References
NTCP SLC10A1
OATP-A SLC21A3 OATP1 OATP
OATP-B SLC21A9
OATP-C SLC21A6 LST-1 OATP2
OATP8 SLC21A8
++ -
+ + + ++ + -
+ + -
+/+/+ + + +
-
+ + + + + + + + +
-
+++ + +
H 122,125
B 125
H,B 125
H 124,136
H 124
+ +
PGT SLC21A2
++
W 124,125
H, hepatocytes; B, brain; W, wide tissue distribution; He, heart; M, muscle.
ductus venosus in the newborn may intensify and prolong the “physiological” jaundice in premature infants. Hepatic uptake of amphiphilic organic anions is carrier-mediated. Recent studies have elucidated various aspects of this transport process. A brief discussion of these mechanisms follows.
Energy Requirements The intrahepatic concentration of a ligand can exceed the plasma concentration by about 10-fold. However, because ligands are firmly protein-bound in both plasma and hepatocytes, whether a concentration gradient exists with respect to a free (unbound) ligand depends not only on respective concentrations of the ligand on both sides of the hepatocyte sinusoidal plasma membrane, but also on the affinities of the ligand for albumin and intrahepatic binding proteins. These considerations are relevant for a discussion on the driving forces for this potentially uphill transport. It is clear that there are multiple uptake systems that differ in energy requirement.122 In any uptake system, organic anions must be translocated against an inside negative plasma membrane potential of about –35 mV that is generated by Na+/K+-ATPase. Electrical neutrality must be maintained by intracellular association with counterions, by cotransport with cations such as H+, or by countertransport with anions such as OH– or Cl–.
Transporters The basolateral hepatocyte membrane contains various transporter proteins that function as carriers for the uptake of a multitude of endogenous and exogenous substances. The uptake carriers in human hepatocytes are listed in Table 74-1. NTCP (SLC10A1) in humans (ntcp, Slc10a1 in rodents) has a narrow range of substrate specificity and is a specialized carrier for the Na+-dependent hepatic uptake of bile salts.123 BSP and its glutathione conjugate are widely used model substrates in transport studies and indeed are substrates for a number of uptake transporters. Also the thyroid hormones, thyroxine and triiodothryonine, are transported by a number of uptake carriers. Organic cations are transported by members of the
1458
OCT/oct family. These are particularly active in the kidney. Hepatic uptake of unconjugated bilirubin has been investigated in numerous classic studies using membrane fractions of various degrees of purity, isolated hepatocytes, isolated perfused rat livers, or whole animals. Concentrative bilirubin uptake in the liver can partly be explained by sinusoidal dissociation from albumin and binding to intrahepatocellular proteins. Before bilirubin binds to intracellular proteins, it must cross the sinusoidal domain of the hepatocyte plasma membrane. Bilirubin is of low enough polarity that it can cross membranes without the aid of transporter proteins. Zucker and Goessling have proposed a non-carrier-mediated transmembrane diffusion of unconjugated bilirubin, in which dissociation from albumin is ratelimiting.124 However, since concentrative bilirubin uptake is a liverspecific function, a hepatocyte-specific carrier protein is thought to be involved. OATP2 (also termed OATP-C or SLC21A6) in human liver and Oatp4 (Slc21a10) in rat liver are high-affinity transporters of organic anions, such as BSP, taurocholate, estradiol-17-bglucuronide, LTC4, estrone-3- and 1-sulfate, dehydroepiandrosterone sulfate, triiodothyronine and thyroxine.125,126 OATP-C has also been reported to transport unconjugated bilirubin,125 but other studies did not confirm this conclusion.127 Thus, the transport mechanism of bilirubin across the sinusoidal surface membrane of the hepatocytes remains to be identified conclusively.
Acquired and Genetic Abnormalities of Hepatic Uptake The expression of hepatic uptake carrier proteins is regulated at both transcriptional and post-transcriptional levels. For instance, cell swelling in the isolated perfused rat liver causes a cAMP-mediated translocation of ntcp from intracellular storage sites to the basolateral plasma membrane.128 Delivery and insertion of ntcp into the plasma membrane require microtubules and microfilaments.129 Prolactin induces ntcp by Stat-5-regulated enhancement of transcription.130 The transcriptional and post-transcriptional regulation of uptake transporters has been studied. Endotoxin, tumor necrosis factor-a, interleukin (IL)-1b and IL6 reduce bile salt and organic
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
anion uptake. These cytokines down-regulate expression of Ntcp and Oatp4.131,132 They mediate their effect most probably via the nuclear hormone receptors RXR and RAR.133 They bind to the RXRRAR heterodimer response elements within the promoter region of the ntcp and oatp4 genes. High intracellular bile salt concentrations decrease ntcp expression. The mechanism of this down-regulation has been elucidated. Bile salts bind to the farnesoid X-receptor (FXR), a cytosolic nuclear hormone receptor. Upon binding bile salts, this receptor associates with RXR and the heterodimer then translocates to the nucleus. Here it binds to an FXR:RXR response element that is present in a number of genes, including the SHP gene (short heterodimer partner). Thus, high bile salt concentration activates SHP expression and the up-regulated SHP protein interferes with RAR:RXR binding to the ntcp gene and this inhibits transcription. Also OATP-C expression in humans and Oatp4 expression in rats is down-regulated in cholestasis. OATP-C is under transcriptional control of hepatocyte nuclear factor 1 (HNF1a) and HNF1a expression is controlled by HNF4a. SHP inhibits HNF4a-mediated transactivation of the HNF1a promoter in cotransfection assays. In addition, the human HNF4a alpha gene promoter is repressed by CDCA through an SHP-independent mechanism. This explains why, in addition to NTCP, OATP-C is also down-regulated in cholestatic liver disease. Since in most liver diseases conjugated bilirubin rather than unconjugated bilirubin is the dominant bile pigment in serum, it is likely that down-regulation of bilirubin transport is a postconjugation event affecting canalicular transport. As discussed later, an elegant explanation for the accumulation of conjugated bilirubin in plasma in cholestatic conditions is now available. MRP2, the biliary export pump for conjugated bilirubin, is down-regulated during cholestasis,134–136 while MRP-3, a transporter with affinity for conjugated bilirubin, is up-regulated in the basolateral membrane. Via this transporter, conjugated bilirubin is pumped from the hepatocyte into blood.137,138 Subsequent clearance of bilirubin conjugates by the kidneys in cholestasis is helped by the up-regulation of renal mrp-2 expression. From this discussion it is clear that bile acid and bilirubin are taken up by the hepatocyte partly via separate mechanisms. This explains the discrepancy between the extent of elevation of plasma bile acid and bilirubin levels in some clinical situations. For example, early in the course of primary biliary cirrhosis, bile acid levels are eleated, but bilirubin levels may still be normal.139,140 Reduced hepatic bilirubin uptake has been observed in some cases of Gilbert syndrome (discussed later).
proteins, identical to the alpha class of glutathione-S-transferases (GSTs).142 Studies of bilirubin transport in isolated perfused rat liver showed that hepatic ligandin concentration did not affect the influx rate of bilirubin, but increased the net uptake by reducing the efflux rate.143
CONJUGATION OF BILIRUBIN Biliary excretion of bilirubin requires its conversion to polar derivatives by enzyme-catalyzed conjugation of the propionic acid carboxyls with sugar moieties, which disrupts the internal hydrogen bonds (Figure 74-5). Glucuronic acid is the predominant conjugating sugar in vertebrate bile; glucuronidation of one or both propionic acid carboxyls results in the formation of bilirubin monoglucuronide or diglucuronide, respectively, both of which are efficiently excreted in bile.144 In normal bile from humans and most other mammals, bilirubin diglucuronide is the predominant conjugate.145,146 In addition, small amounts of glucosyl and xylosyl conjugates have been described in mammalian bile.147
Enzyme-Catalyzed Glucuronidation of Bilirubin Glucuronidation of bilirubin is catalyzed by a specific isoform of a family of enzymes termed uridinediphosphoglucuronate glucuronosyltransferase (UDP-glucuronosyltransferase, EC 2.4.1.17, recommended abbreviation: UGT). UGTs are concentrated in the ER and nuclear envelope of a variety of cells.148 These enzymes catalyze the
O H 4 O
N 10 H N H
N H
N H
O
15
O H O Bilirubin IXa-4Z,15Z
O COOH OH O
O
HO
A
B
N 10 H
N H
N H
C
COOH O HO HO
O
OH OH
O
4
STORAGE OF BILIRUBIN WITHIN THE HEPATOCYTE Within the hepatocyte bilirubin is kept in solution predominantly by binding to cytosolic proteins. Gel permeation chromatography of hepatic cytosol containing 3H-bilirubin revealed that radioactivity is associated with two protein fractions, which were originally termed Y and Z.141 Tracer quantities of anions bind almost exclusively to the Y fraction, while in the presence of higher concentrations, binding to the Z fraction becomes apparent, suggesting that components of the Y fraction are the predominant proteins to which organic anions bind within hepatocyes.141 The Y protein(s) bind many compounds, including drugs, hormones, and organic anions. This led to their being termed “ligandin.” Later, ligandin was found to be a family of
O
N H D
O
15
O Bilirubin IXa-diglucuronide
Figure 74-5. Effect of sugar conjugation on the structure of bilirubin: the fully hydrogen-bonded structure of bilirubin IXa “-Z,Z is shown in the top panel. In bilirubin glucuronides, glucuronidation of the propionic acid carboxyls disrupts the hydrogen bonds. This makes the molecule water-soluble and rapidly excretable in the bile. Disruption of the hydrogen bonds exposes the central carbon bridge (C-10) to diazo reagents, resulting in “direct diazo reaction.”
1459
Section XII. Inherited and Pediatric Diseases of the Liver
transfer of the glucuronosyl moiety of UDP-glucuronate to a wide spectrum of aglycones, forming ether, ester, thiol, and Nglucuronides.149 Substrates of UGT include hormones (e.g., steroid hormones, thyroid hormones, and catecholamines), endogenous metabolites (e.g., bile salts and bilirubin), numerous drugs and their intermediate metabolites, toxins (e.g., carcinogens), and laboratory xenobiotics.149 Glucuronidation renders the aglycone substrates more polar and, in general, less biologically active. Thus, UGTs constitute one of the most important detoxification mechanisms of the body. Latency of UGTs. UGTs are integral membrane proteins that require specific membrane lipids for function.150 UGT activity in microsomal vesicles is partially latent. Membrane-perturbation by physical,151,152 enzymatic,153 or detergent treatment152 removes the latency, making the enzyme fully active. Two mechanisms have been proposed to explain the latency and activation of UGT in microsomal membranes. In the compartmental model, the catalytic site of the enzymes is envisioned to be located inside the lumen of the ER. Analysis of their amino acid sequence reveals a putative transmembrane domain near the carboxy-terminal end, leaving only a small segment of the proteins on the cytosolic side of the ER. This cytosolic tail is believed to contain the ER-localizing signal.154 In view of the hydrophilic nature of UDP-glucuronic acid, the presence of a transporter protein to mediate the movement of UDP-glucuronic acid from the cytosol to the ER lumen is likely. Mechanisms of transport of UDP-glucuronic acid into the ER cisternae are being investigated, but the putative transporters have not been fully characterized or cloned.155–157 UDP-N-acetylglucosamine activates hepatic microsomal UGT activity at low concentrations and has been postulated to be a physiological activator of the transferase. Two alternatives have been proposed to explain this. UDP-Nacetylglucosamine could increase the access of UDP-glucuronate either by activating a putative transporter or by physically increasing the permeability of the lipid membranes to UDP-glucuronic acid.158 An alternative hypothesis proposes that the UGT activity is “constrained” by the membrane. Physical or chemical activators, such as UDP-N-acetylglucosamine, may act by releasing the enzyme from such constraint.159 In this model, UDP-N-acetylglucosamine is envisioned as an allosteric activator of UGTs. The two models may not be mutually exclusive and neither has been reliably excluded by firm experimental evidence. Multiple Forms of UGT. The UGT system consists of a family of structurally related enzymes that are functionally heterogeneous. The isoforms differ from each other in ontogenic development160 and the effect of enzyme-inducing agents.161,162 Several laboratories have isolated multiple UGT isoforms from solubilized liver microsomes.163 Cloning of cDNAs and genomic DNA has provided a large amount of information on the structure and evolutionary divergence of UGTs. This topic has been reviewed.164 Based on the degree of structural homology among the various UGT cDNAs, UGTs may be separated into two major families.164 One family (UGT1) contains the bilirubin conjugating form in human and rat liver, and at least two phenol conjugating UGTs. The second family (UGT2) includes a number of UGT isoforms that mediate the conjugation of steroids and other endogenous and exogenous substrates. One of these isoforms is inducible by phenobarbital. The UGT2 subfamily also
1460
includes two UGT isoforms located in the olfactory tissues that are thought to be responsible for the turn-off of chemical olfactory signals.165 Organization of the UGT1A Gene Locus. Members of the UGT1A family, including the form that accepts bilirubin as a substrate (UGT1A1) and two phenol-UGT isoforms (UGT1A6 and UGT1A7), are expressed from a locus, termed UGT1A.166 This locus, shown schematically in Figure 74-6, is located on human chromosome 2 at region 2q37.167 The 3¢ domain of this gene contains four consecutive exons (exons 2 through 5) that are shared by all UGT isoforms expressed from the UGT1A locus and encode their identical carboxy-terminal domains. This “common region” is highly conserved in evolution and is responsible for UDPglucuronate binding.168 Upstream to these exons are a series of at least 12 exons (exon 1A1 through 1A12), each encoding the variable NH2-terminal region of a different UGT1A isoform. Each variable region exon is preceded by a separate promoter sequence. Depending on promoter selection during transcription, transcripts of various lengths are produced. During processing of the transcript to mRNA, the exon which is located at the 5¢ end of the transcript is always spliced to exon 2 (the first of the four common region exons), the whole intervening sequence being spliced out. The amino-terminal domain of a given UGT1A isoform encoded by a unique exon confers the aglycone substrate specificity to that isoform.169 The genes for the various UGT1A isoforms are named according to the unique exon used in that mRNA. For example, exon 1A1 encodes the amino-terminal domain of UGT1A1 (bilirubinUGT) and this gene is termed UGT1A1. UGT1A1 is the only isoform that significantly contributes to the conjugation of bilirubin.170 Two other genes, UGT1A6 and UGT1A7, encode isoforms that conjugate phenols.164 The presence of a separate regulatory cis-element upstream to each unique exon permits UGT1A isoforms to be independently regulated, explaining their different organ distribution and expression during ontogeny, enzyme induction, or carcinogenesis. Enzyme activity toward 4-nitrophenol and other simple phenolic substrates develops in late fetal life in rats, whereas activity toward bilirubin develops after birth.160 UGT1A6, a 3-methyl cholanthrene-inducible isoform, is permanently overexpressed in carcinogen-induced preneoplastic nodules in rat liver.171 Triiodothyronine treatment results in a threefold increase in rat liver phenol-UGT activity, whereas bilirubin glucuronidation is reduced by 80%.162
BILE CANALICULAR SECRETION OF BILIRUBIN CONJUGATES AND RELATED ORGANIC ANIONS Excretion across the canalicular membrane is rate-limiting in overall disposition of many cholephilic compounds and represents the most important concentrative step. Indeed, for all classes of organic compounds, including cations, anions, and bile acids, this step creates a larger concentration gradient than that observed across the sinusoidal/basolateral membrane. For example, for organic anions like dibromosulfophthalein (DBSP), a liver-to-bile concentration ratio of 1 : 1000 can be reached.172 These concentration gradients are too large to be accounted for by the membrane potential difference over the canalicular membrane. For most classes of compounds, active, energy-consuming transport systems have been demonstrated.
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
1A12
1A7
1A6
1A5
1A4
1A3
1A2
1A1
2 3 4
5
Stop CN-I Stop codon or frame-shift
Signal peptide Stop
Substitution CN-II Splice-site mutation
Stop Gilbert Normal TATAA box: A(TA)6 TAA Gilbert type TATAA box: A(TA)7 TAA Figure 74-6. A schematic representation of the human UGT1A locus at human chromosome 2q37, that comprises several genes encoding the UGT1A family of isoforms. At the 3’ end of the locus there are 4 exons (exons 2, 3, 4, and 5, shown as solid bars), which are used in all UGT isoforms expressed from this locus. Upstream to these common region exons there is a series of variable region exons, designated 1A1 through 1A12, only one of which is used in a given UGT isoform. Each of these exons encodes the variable NH2-terminal region of one UGT isoform. Each variable region exon has an upstream promoter element, and is differentially regulated. Depending on which promoter is used, transcripts of various lengths are produced. In each transcript, the exon located at the 5’ end of the transcript is spliced to exon 2; the intervening stretch of RNA is spliced out. The genes are named according to their unique region exons. For example, when the transcription starts at exon 1A1, the processed mRNA will consist of exon 1A1, 2, 3, 4, and 5. This gene is termed UGT1A1. If the transcription starts at exon 1A6, the mRNA will consist of exons 1A6, 2, 3, 4, and 5 and the gene is termed UGT1A6. UGT1A1 expresses the only UGT isoform that significantly contributes to bilirubin glucuronidation. Genetic nonsense (q) or missense (s) mutations of any of the five exons of UGT1A1 can abolish or reduce UGT1A1 activity, causing CN-1. In the case of CN-2 all mutations are of the missense type (s). Genetic lesions of exon 1A1 only affect bilirubin glucuronidation, whereas those of exons 2, 3, 4, or 5 affect all isoforms of the UGT1A subfamily. Gilbert syndrome is associated with the insertion of a TA dinucleotide within the TATAA element upstream to exon 1A1.
ATP-Dependent Organic Anion Transport Proteins of the MRP family are important in the canalicular excretion of bilirubin glucuronides. MRP2 (mrp2 in rodents) is a canalicular efflux pump for organic anions. This protein is also expressed in intestine and kidneys.173–175 In the liver, MRP2 functions as the efflux pump for many organic anions, most of which are products of phase II drug metabolism. Bilirubin mono- and diglucuronide are among its substrates.176 Genetic deficiency of MRP2 in humans leads to the Dubin–Johnson syndrome, characterized by hyperbilirubinemia.177,178 The TR– rat is an animal model with conjugated bilirubinemia caused by a single nucleotide deletion in the mrp2 gene.179 MRP1 (mrp1 in rodents) is expressed during liver regeneration and endotoxin-mediated cholestasis.135,180 MRP3 (mrp3 in rodents) is expressed at very low levels in normal liver, but is greatly induced during cholestasis and hyperbilirubinemia.181–183 Both mrp1 and mrp3 are located in the basolateral membrane of the hepatocytes. Mrp1 and mrp3 are mainly pumps for glucuronides, including bilirubin glucuronides,184,185 whereas mrp1 pumps glutathione-S-conjugates138 and, as has recently been reported, unconjugated bilirubin. Mrp3 expression is strongly enhanced during cholestasis, in the liver of mrp2-deficient EHBR rats and in UGT1A1-deficient Gunn rats.185,186 Thus, MRP3/mrp3 functions as a reverse transporter, which pumps substrates back to the blood when biliary excretion of glucuronides is reduced because of an
inherited deficiency of MRP2/mrp2 or its down-regulation in cholestatic conditions. CAR, the constitutive androstane receptor, is the dominant transcriptional regulator of MRP3. Phenobarbital is the prototypical substrate of CAR. MRP3 expression is enhanced in patients with Dubin–Johnson syndrome and Eisai hyperbilirubinemic rats, suggesting that conjugated bilirubin may also be a CAR substrate. The bilirubin conjugate transporter MRP2 (ABCC2) is a multispecific pump that contributes to the bile formation by transporting GSH, a major driving force for bile salt-independent bile flow. In addition, MRP2 also has a role in canalicular anionic phase II conjugate transport. Mrp2 expression in rat liver responds to inducing agents and is down-regulated by endotoxin, cytokines, and bile duct ligation. A dose- and time-dependent induction of mrp2 expression was observed in isolated rat hepatocytes cultured in the presence of xenobiotics, including vincristine, tamoxifen, or the PXR-ligand rifampicin,187 indicating that mrp2-gene transcription may respond to substrates of MRP2 itself. The promoter regions of the human MRP2- and the rat mrp2-genes have been isolated.188,189 Sequence analysis of the human MRP2 promoter showed a number of putative consensus binding sites for both ubiquitous and liver-enriched transcription factors, including activating protein AP1, SP1, HNF1, and HNF3b188,190 as well as CAR, PXR, and FXR. An unusual sequence was identified 440 bp upstream of the MRP2 transcrip-
1461
Section XII. Inherited and Pediatric Diseases of the Liver
Table 74-2. Organic Anion Secretion Transporters in Human Liver
Glutathione (GSH) Glutathione disulfide (GSSG) Leukotriene C4 Gluthatione-S-conjugates Bilirubin monoglucuronide Bilirubin diglucuronide Estradiol 17-b-D-glucuronide Taurocholate Glycocholate 3a-Sulfotaurochenodeoxycholate 6a-Glucuronosyl hyodeoxycholate 3a-Sulfotaurolithocholate Ochratoxin A Methotrexate BQ-123 Regulation
Polarity Tissue distribution References
MRP1 (ABCC1)
MRP2 (ABCC2 or cMOAT)
+ + + + + + +
+ + + + + + +
+ +
+ +
+
MRP3 (ABCC3)
MRP6 (ABCC6)
* * + + + + + +
LPS≠29 PH≠45
LPSØ30≠29 BDLØdexamethasone ≠122,123
Basolateral
Canalicular
BDL124,125 EHBR rats≠126 Gunn rats≠127 phenobarbital≠50 Basolateral
H, E, B 137,190
H, I, K 181,182,185–187
H, C 141,188–192
+
Basolateral, Canalicular H 190
H, hepatocytes; C, cholangiocytes; E, erythrocytes; K, kidney; B, brain, PH, partial hepatectomy; BDL, bile duct ligation; *, probable. Alternative names are given in parentheses.
tion initiation site that binds with high-affinity PXR, CAR, and FXR as heterodimers with RXR. Human and rodent hepatocytes reacted with a robust induction of MRP2 mRNA levels upon exposure to the PXR, CAR, and FXR agonists, rifampicin, dexamethasone, pregnenolone 16a-carbonitrile, spironolactone (PXR), phenobarbital (CAR), and to chenodeoxycholic acid (FXR). Ugt1A1 in rodents has also been reported to be a PXR target gene. Like MRP2, UGT1A1 contains a multicomponent enhancer element in its promotor region with CAR, PXR, and AhR motifs. In addition, both glucuronidation and secretion are induced by PXR and CAR agonists. Thus PXR and CAR are major regulators of bilirubin uptake, glucuronidation, and secretion. The various secretion transporters and the substances that they handle are listed in Table 74-2.
Transport of Bile Acids and Non-Bile Acid Organic Anions Human MDR1 (ABCB1), MDR3 (ABCB4), and BSEP (ABCB11), and their rat orthologues, Mdr1a and 1b, Mdr2 and Bsep, are Pglycoproteins that are constitutively expressed in the canalicular membrane of the hepatocyte. MDR3/Mdr2 and BSEP/Bsep are expressed in the liver only, whereas MDR1 is also expressed in various non-hepatic secretory epithelia. The canalicular bile-salt export pump (BSEP) is of paramount importance for bile formation and liver function. BSEP appears to be the principal driving force in the enterohepatic circulation of bile salts. Also the bilesalt-dependent fraction of bile flow depends on BSEP. Bile salts are the major, if not the only, substrates of BSEP. Rat, mouse, and human BSEP-genes have been cloned recently.190–194 MDR3 in humans (mdr2 in mice) is involved in the biliary secretion of phosphatidylcholine, the only phospholipid in bile. Mice with a knock-
1462
out mutation of the mdr2-gene produce bile in which phospholipids are absent.195 These mice and humans with a similar defect develop severe liver disease characterized by bile duct proliferation, portal fibrosis, and eventually cirrhosis.196,197 Heterozygosity for MDR3 deficiency is also in part responsible for intrahepatic cholestasis of pregnancy as well as for intrahepatic and gallbladder cholesterol stone formation. Role of Nuclear Receptors in the Orchestration of Various Steps Involved in Bilirubin Throughput. Capacity of the various steps involved in bilirubin throughput appears to be matched in vivo. Thus, reduction of transport, storage, conjugation, or excretion can all lead to hyperbilirubinemia. On the other hand, enhanced excretion of bilirubin, for example, in response to an increased bilirubin load, would require coordinated increase in the capacity of each step involved in the uptake and elimination process. It has been proposed that the nuclear receptor, CAR, serves as the coordinating mechanism for physiological modulation of each of these steps.198,199
FATE OF BILIRUBIN IN THE GASTROINTESTINAL TRACT Conjugated bilirubin reaching the intestinal tract is not substantially absorbed.200 Absorption of unconjugated bilirubin from the intestine may be enhanced in the presence of maternal milk and may contribute to neonatal hyperbilirubinemia.201 A small amount of bilirubin may be reabsorbed from the gallbladder in animals.202 Bilirubin glucuronides are deconjugated by intestinal bacteria203 and degraded to a series of urobilinogens and related products.204 Urobilinogens are not glucuronidated; it is not known whether deconjugation precedes or follows bilirubin degradation. After absorption from the
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
intestine, urobilinogen is excreted in the bile and, to a lesser extent, in the urine. Variability of reabsorption of urobilinogen by renal tubules and instability of the pigment in acid urine make the measurement of urinary urobilinogen concentration an unreliable indicator of metabolism. However, the absence of urobilinogen in stool and urine indicates complete obstruction of the bile duct or severe cholestasis, as seen in the early days of acute hepatitis. In most liver diseases and in states of increased bilirubin production, urinary urobilinogen excretion is increased. Urobilinogen is colorless. Its oxidation product, urobilin, is yellow and contributes to the color of normal urine and stool.
EXTRAHEPATIC HANDLING OF BILIRUBIN Renal Handling of Bilirubin Only 3% of labeled unconjugated bilirubin is excreted by the kidney after intravenous administration. Even in the presence of high bilirubin levels, bile remains the main route of bilirubin excretion. Renal excretion of conjugated bilirubin depends on the glomerular filtration of a small, non-protein-bound fraction of conjugated bilirubin.205 Unconjugated bilirubin, which is tightly bound to albumin, is not filtered to a significant extent by normal renal glomeruli and does not appear in the urine. There is evidence for tubular reabsorption but not tubular secretion of bilirubin.206 Because, normally, less than 5% of total bilirubin in plasma is conjugated, bilirubin is not present in urine in the absence of albuminuria. During cholestasis, the kidney assumes a major role in bilirubin disposal. Upon infusion of radiolabeled bilirubin into plasma of animals with experimental bile duct ligation207,208 and in children with biliary atresia,209 50–90% of injected radioactivity is excreted in urine. Urinary excretion becomes the major excretory pathway for bilirubin in the presence of complete biliary obstruction.205 Intestinal Bilirubin Metabolism. Intestinal mucosal epithelia, particularly cells of the proximal small intestinal villi, have bilirubinUGT activity.148 The relative contribution of small intestinal glucuronidation in the overall disposition of bilirubin is not clearly known.
ALTERNATIVE PATHWAYS OF BILIRUBIN ELIMINATION Photoisomerization Configurational and cyclic isomers of bilirubin formed in the presence of ambient light or during phototherapy (discussed above in this chapter), are excreted in bile in unconjugated form.210,211 A significant amount of bilirubin is degraded to polar diazo-negative compounds that are excreted in bile and urine.212
Enzyme-Catalyzed Oxidation Oxidation of bilirubin may be mediated by mixed-function oxidases in liver and other organs. Inducers of the mixed-function oxidase, cytochrome P450c, such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) or 3-methylcholanthrene reduce plasma bilirubin levels in bilirubin-UGT-deficient Gunn rats.213 Induction of cytochrome P450c by chlorpromazine resulted in the reduction of serum bilirubin levels in 1 patient with Crigler–Najjar syndrome type 1.214 Degradation of bilirubin in vitro by cytochrome P450-1A1 supports
the concept that bilirubin oxidation in the liver may be a functional alternative pathway of bilirubin disposal. Induction of this enzyme system by substances such as indole-3-carbinol may convert bilirubin to colorless metabolites.215 The relative contribution of microsomal oxidation to the overall turnover of bilirubin in vivo is not clear, but this pathway may play a significant role in bilirubin disposition in cases where the normal pathways are deficient, for example, in patients with Crigler–Najjar syndrome type 1 or Gunn rats.216
QUANTIFICATION OF BILIRUBIN Serum bilirubin is an important marker of liver function. In the newborn period, the total serum bilirubin concentration and the fractional concentration of free (non-protein-bound) bilirubin need to be monitored to determine the need to institute therapy to reduce serum bilirubin level. Clinically, serum bilirubin is usually measured after conversion to stable azoderivatives. Quantification of the various species of bilirubin as intact tetrapyrroles is more accurate and of value mainly for research purposes. Currently available methods of bilirubin analysis have been reviewed.217
CONVERSION OF BILIRUBIN TO AZODERIVATIVES Quantification of bilirubin is facilitated by derivatization with diazo reagents that transform it to stable dipyrrolic derivatives. Electrophilic attack by a diazonium ion at the carbons flanking the central carbon bridge (C9 and C11 positions of bilirubin)218 converts the tetrapyrrole to two diazotized azodipyrroles, the central bridge carbon being eliminated as formaldehyde. Unconjugated bilirubin yields two unconjugated dipyrroles, bilirubin diconjugates form two conjugated azodipyrroles, and bilirubin monoconjugates form one conjugated and one unconjugated azodipyrrole. In 1916, van den Bergh and Muller described that one species of serum bilirubin reacts with the sulfanilic acid diazo reagent within minutes (“direct” reacting fraction), whereas another species only reacts rapidly when accelerator substances, such as methanol or caffeine, are present (“indirect” reacting fraction).219 Later, it was understood that indirect-reacting bilirubin represents unconjugated bilirubin, while direct-reacting fraction roughly corresponds to conjugated bilirubin.220 Variations of the van den Bergh reaction are currently used for the clinical determination of bilirubin conjugates. The direct diazo reacting fraction of bilirubin overestimates the levels of conjugated bilirubin, as solutions of unconjugated bilirubin may show as much as 10–15% of the total pigment as direct-reacting. Usually, a direct-reacting bilirubin concentration of less than 15% of total is considered normal. The diazo reaction does not differentiate between non-covalently albumin-bound conjugated bilirubin and the fraction of bilirubin that becomes irreversibly bound to serum proteins, particularly albumin, during conjugated hyperbilirubinemia,221 because both fractions give direct diazo reaction. The irreversibly protein-bound bilirubin is slowly cleared from serum after reversal of biliary obstruction. Therefore, finding direct-reacting bilirubin during this period may give a false impression of continued biliary obstruction. Certain metabolites, such as indican, which accumulate in serum during renal failure, may
1463
Section XII. Inherited and Pediatric Diseases of the Liver
interfere with the diazo reaction.222 In circumstances in which it is critical to know whether conjugated bilirubin is present in the plasma, it is necessary to perform chromatographic analysis of bile pigments.
CHROMATOGRAPHIC ANALYSIS OF BILIRUBIN SPECIES AS INTACT TETRAPYRROLES Thin-Layer and High-Pressure Liquid Chromatography Bilirubin and its conjugates can be separated by thin-layer chromatography.223 High-resolution analysis of bilirubin and its conjugates is possible using high-pressure liquid chromatography (HPLC). Methyl esters formed by alkaline methanolysis of bilirubin monoand diconjugates are easy to extract from serum and analyze by HPLC.224 However, as the conjugating sugars are replaced by methyl groups, this method does not identify the type of sugar conjugate. Methods for separation and quantification of intact bilirubin tetrapyrrole conjugates by HPLC have been developed225,226 and offer accurate and sensitive means to identify and quantify bilirubin conjugates in body fluids and in vitro.
Measurement of the Irreversibly Protein-Bound Fraction Most chromatographic methods require extraction of serum bile pigments into organic solvents, whereby the irreversibly protein-bound fraction of bilirubin (also known as delta-bilirubin) is lost. To measure this fraction, along with other fractions of unconjugated and conjugated bilirubin, reverse-phase HPLC is performed on incompletely deproteinated serum.221 Investigations utilizing this method indicate that the irreversibly protein-bound serum bilirubin fraction is formed in conditions associated with conjugated hyperbilirubinemia.
SLIDE TESTS A convenient slide test measures conjugated, unconjugated, and irreversibly protein-bound bilirubin. The Ektachem TIBL slide is used to quantify total bilirubin by a diazo technique.227 Another slide is specially coated to allow only the free and reversibly protein-bound forms of bilirubins to react with the diazo reagent.228 Irreversibly protein-bound bilirubin can be estimated from the difference between total bilirubin and the sum of conjugated and unconjugated bilirubin, as estimated by the second slide. Corroboration with chromatographic methods indicates that results obtained by the Ektachem slide tests are consistent and reliable.227,228
centrations that correlated well with values obtained by a standard diazo method using serum samples.230
THE NATURE AND SIGNIFICANCE OF BILIRUBIN IN BODY FLUIDS BILIRUBIN IN PLASMA When measured by accurate chromatographic techniques, unconjugated bilirubin is normally the predominant plasma bile pigment: only up to 4% of the pigments are conjugated. When there is an overproduction of bilirubin, the proportion of the unconjugated and conjugated fractions remains unchanged despite the increase of total bilirubin levels. In contrast, when the serum level increases because of a deficiency of bilirubin glucuronidation, the absolute concentration of conjugated bilirubin may remain normal or may be reduced, resulting in a reduced proportion of conjugated bilirubin. During biliary obstruction, intrahepatic cholestasis, or hepatocellular diseases, both conjugated and unconjugated bilirubin accumulate in plasma, and the proportion of conjugated bilirubin increases. As discussed above, during cholestasis MRP2 is down-regulated, reducing biliary excretion of conjugated bilirubin. Bile pigments accumulating in hepatocytes may be pumped out into plasma by MRP3, which is up-regulated in basolateral membranes during cholestasis. When there is a prolonged accumulation of conjugated bilirubin, a fraction of the pigment becomes irreversibly bound to albumin. This fraction, termed delta-bilirubin, gives a direct diazo reaction and can be identified by chromatographic analysis. After successful surgical correction of biliary obstruction, the reversibly protein-bound fraction of serum bilirubin is rapidly excreted in bile, resulting in an increase in the proportion of the irreversibly proteinbound fraction of serum bilirubin. If biliary obstruction persists, both reversibly protein-bound and irreversibly protein-bound fractions are retained; therefore the proportion of the irreversibly bound fraction does not show a marked increase.119
BILIRUBIN IN URINE Because unconjugated bilirubin is tightly bound to albumin, it is not normally filtered in renal glomeruli. Conjugated bilirubin is less strongly bound to albumin and, therefore, appears in urine. In the absence of proteinuria, excretion of bilirubin in the urine indicates the presence of conjugated bilirubin in the plasma. Covalently protein-bound bile pigments (delta-bilirubin) is not excreted in urine.
BILIRUBIN IN BILE TRANSCUTANEOUS BILIRUBINOMETRY Determination of the rate of increase of serum bilirubin levels during the first 24–48 hours of life by repeated measurement of serum bilirubin levels can be helpful in predicting peak serum bilirubin level during the neonatal period. Such measurements can be performed painlessly and at low expense by the measurement of the yellow color of the skin by analysis of reflected light.229 On-board computers in these analyzers are programmed to measure the yellow color without interference by underlying skin pigmentation or erythema. Transcutaneous bilirubinometry in 900 term and premature infants of different races provided estimated serum bilirubin con-
1464
Over 80% of bilirubin in normal hepatic bile is the diglucuronide. Unconjugated bilirubin accounts for only 4% of biliary pigments. In the presence of complete deficiency of bilirubin-UGT activity, as in the case of Crigler–Najjar syndrome type 1, little or no bilirubin glucuronides are excreted in bile. When bilirubin glucuronidation is partially deficient, as in Crigler–Najjar syndrome type 2 or Gilbert syndrome, the proportion of bilirubin monoglucuronide and unconjugated bilirubin increases in bile. The presence of a significant amount of conjugated bilirubin in bile reliably differentiates Crigler–Najjar syndrome type 1 from Crigler–Najjar syndrome type 2 (see below).
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
Tissue fluids with high protein content contain albumin-bound bilirubin. Therefore, a comparison of the serum bilirubin concentration with the concentration of bilirubin in body fluids can help to differentiate between exudates and transudates. A pleural fluid-toserum bilirubin ratio of 0.6 or higher is strongly indicative of an exudate.231 Bilirubin is excreted in sweat, semen, and milk in hyperbilirubinemic patients. Bilirubin is present in synovial and ocular fluids. However, xanthopsia is extremely rare in jaundice. Interestingly, paralyzed limbs are less deeply jaundiced than other parts of the body.
predominantly unconjugated. Exaggeration of physiologic jaundice may expose infants to the risk of kernicterus. In approximately 16% of newborn babies, maximal serum bilirubin concentrations reach or exceed 10 mg/dl; in 5% the level exceeds 15 mg/dl.233 Physiologic jaundice of the newborn may result from a combination of increased bilirubin production and a lower than adult level of the capability of the liver to dispose of bilirubin. Exaggeration of one or more of the normal restrictions characteristic of the newborn period and/or superimposition of additional mechanisms may result in a pathological level of neonatal hyperbilirubinemia. A brief discussion of these mechanisms follows.
BILIRUBIN IN CEREBROSPINAL FLUID
Increased Bilirubin Load
BILIRUBIN IN TISSUE FLUIDS
Because of the low protein concentration of the cerebrospinal fluid, bilirubin concentration in cerebrospinal fluid is much lower than that in the serum. When an increase in the cerebrospinal fluid protein concentration coexists with jaundice, e.g., in leptospirosis, the cerebrospinal fluid may contain bilirubin.
BILIRUBIN IN SKIN AND SCLERA Bilirubin binds avidly to the elastic tissue of skin and sclera. Therefore, scleral icterus is a sensitive clinical feature of hyperbilirubinemia. Scleral icterus often outlasts hyperbilirubinemia. Yellow discoloration of skin and mucous membranes is more intense in conjugated hyperbilirubinemia, probably because of better penetration of the water-soluble conjugates into body fluids. During prolonged conjugated hyperbilirubinemia, oxidation of bilirubin to biliverdin may produce a greenish pigmentation of the skin.
DISORDERS OF BILIRUBIN METABOLISM Increased bilirubin production and abnormalities of any of the four distinct but interactive steps of hepatic bilirubin throughput, namely, uptake from the circulation, intracellular binding or storage, conjugation, and biliary excretion, may result in hyperbilirubinemia. Many clinical disorders, such as hepatitis or cirrhosis, affect multiple steps of this process. In contrast, in several inherited disorders, a specific step of bilirubin throughput may be involved. From the viewpoint of bilirubin metabolism, these disorders may be classified into those that cause predominantly unconjugated hyperbilirubinemia, and those that are characterized by elevation of both conjugated and unconjugated bilirubin in plasma.
DISORDERS ASSOCIATED WITH UNCONJUGATED HYPERBILIRUBINEMIA NEONATAL HYPERBILIRUBINEMIA All newborn babies have higher serum bilirubin concentrations than do normal adults. Clinically obvious jaundice occurs in about 50% of neonates during the first 5 days of life. In normal, full-term babies, serum bilirubin levels increase from 1–2 to 5–6 mg/dl in approximately 72 hours and subsequently decrease to below 1 mg/dl in 7–10 days.232 In this “physiological jaundice,” serum bilirubin is
Premature breakdown of erythrocytes or ineffective erythropoiesis cause hyperbilirubinemia despite normal liver function. Increased bilirubin production in the newborn period is evidenced by increased endogenous carbon monoxide production,234 early-labeled peak from erythroid and non-erythroid sources and decreased erythrocyte halflife.235 Hemolytic diseases of the fetus, such as rhesus-incompatibility between mother and fetus, used to be a common cause of severe neonatal jaundice prior to the treatment of the mother with anti-rhesus immunoglobulins.236 However, ABO blood group incompatibility remains a common cause of exaggerated neonatal hyperbilirubinemia.237 Inherited disorders, such as sickle-cell disease and hereditary spherocytosis, and toxic or idiosyncratic drug reactions are common causes of hemolytic jaundice in the newborn period. Ineffective erythropoiesis occurs in thalassemia, vitamin B12 deficiency, and congenital dyserythropoietic anemias. In the presence of normal liver function, serum bilirubin levels rarely exceed 4 mg/dl. In some cases, the high bilirubin throughput may result in the accumulation of some conjugated bilirubin in serum in addition to the predominantly unconjugated fraction.
Immaturity of Hepatic Bilirubin Uptake Hepatic bilirubin uptake capacity is low at birth compared to adult levels, and may remain so for the first few days of life. Maturation of the net hepatic bilirubin uptake may correlate with the expression of hepatic GSTs238 that bind bilirubin, thereby reducing its efflux from the liver. When the closure of the ductus venosus is delayed, portal blood bypasses the liver, thereby reducing hepatic uptake.239 Reduced caloric intake may contribute to hyperbilirubinemia by reducing hepatic bilirubin clearance.
Bilirubin Conjugation Hepatic UGT1A1 activity is very low in mammalian fetuses and is only 1% of normal adult levels at birth in humans.240 Regardless of gestational age at birth, bilirubin glucuronidating activity rapidly increases to adult levels by the 14th week of life.241 Reduced hepatic levels of UGT1A1 may linger in some cases because of inherited inhibitory factor(s) in maternal milk or serum as described below. Maternal Milk Jaundice. Breast-fed infants have higher serum bilirubin levels than do formula-fed babies.242 Occasionally, serum bilirubin levels may increase to 15–24 mg/dl within 10–19 days of life. This transient non-hemolytic unconjugated hyperbilirubinemia may take up to 4 weeks to disappear, but promptly resolves upon discontinuation of breast-feeding. Kernicterus is rare, but has been reported.243 The presence of an inhibitor of bilirubin glucuronida-
1465
Section XII. Inherited and Pediatric Diseases of the Liver
tion in maternal milk244,245 is responsible for this syndrome. A progestational steroid, 3¢-20b-pregnanediol, isolated from the milk of mothers of infants with this syndrome, was reported to inhibit o-aminophenol glucuronidation by guinea pig, rat, and rabbit liver,244,245 but not by human liver. The free fatty acid concentration in maternal milk correlates positively with its inhibitory effect on human UGT1A1 activity. The inhibition is more pronounced with polyunsaturated free fatty acids.246 The presence of a lipolytic enzyme in specific maternal milk samples has been reported by some investigators, who have suggested that lipolysis may be responsible for the increased free fatty acid concentrations in the milk. The inhibitory effect of maternal milk on bilirubin glucuronidation increases on storage and is destroyed by heating at 56oC.246 Maternal Serum Jaundice. In this condition, termed Lucey– Driscoll syndrome,247,248 jaundice occurs within the first 4 days of life. Peak serum bilirubin concentrations of 8.9–65 mg/dl are reached within 7 days. An unidentified inhibitor of UGT1A1 is present in the serum of mothers of these infants. Jaundice begins earlier than in maternal milk jaundice, is more severe, and can be associated with kernicterus.
Canalicular Bilirubin Excretion During the late newborn period, uptake and conjugation of bilirubin mature to adult levels, but the bilirubin load continues to be high. During this period of life, canalicular excretion becomes ratelimiting in hepatic disposition of bilirubin. In cases where the bilirubin load is further increased, conjugated bilirubin accumulates in serum.249 However, as discussed above, a small amount of conjugated bilirubin may accumulate in plasma even when the bilirubin load does not exceed the maximal canalicular excretory capacity.
tinal flora is not yet established in the newborn, there is a reduced level of bacterial degradation of bilirubin, resulting in an increased absorption of unconjugated bilirubin.250 Maternal milk may also increase intestinal absorption of unconjugated bilirubin.
HYPERBILIRUBINEMIA DUE TO BILIRUBIN OVERPRODUCTION Except in the presence of abnormal liver function, overproduction of bilirubin rarely causes serum bilirubin to reach levels greater than 3–4 mg/dl. Common causes of hemolytic jaundice include sicklecell anemia, hereditary spherocytosis, and toxic or idiosyncratic drug reactions in susceptible individuals. In hemolytic jaundice, a small amount of conjugated bilirubin produced in the liver may appear in the circulation;251 however, the unconjugated-to-conjugated bilirubin ratio remains normal. Ineffective erythropoiesis that occurs in thalassemia and other hematologic disorders is often associated with hyperbilirubinemia.252 Hereditary dyserythropoietic anemias are a group of rare disorders characterized by ineffective erythropoiesis, intramedullary normoblastic hyperplasia, secondary hemochromatosis, and unconjugated hyperbilirubinemia.253
INHERITED DISORDERS OF BILIRUBIN GLUCURONIDATION Three grades of inherited deficiency of UGT1A1 activity have been described in humans. A virtual absence of UGT1A1 activity results in the most severe of these disorders, Crigler–Najjar syndrome type 1. Severe but incomplete deficiencies of the transferase activity lead to Crigler–Najjar syndrome type 2, also known as Arias syndrome. A mild reduction of UGT1A1 activity is found in the common, mostly innocuous disorder, termed Gilbert syndrome. Table 74-3 summarizes the clinical and diagnostic features of these disorders.
Increased Intestinal Reabsorption
Crigler–Najjar Syndrome Type 1
Hydrolysis of conjugated bilirubin by intestinal b-glucuronidase releases unconjugated bilirubin in the intestine.250 Because the intes-
This rare disorder was described by Crigler and Najjar in 1952 in six infants from three unrelated families.254 Subsequently, the disease
Table 74-3. Features of Inherited Disorders Resulting in Unconjugated Hyperbilirubinemia Crigler–Najjar syndrome type 1
Crigler–Najjar syndrome type 2
Gilbert syndrome
<340 mmol/l Normal Normal Normal Normal Increased proportion of bilirubin monoglucuronide
Usually <50 mmol/l Normal Normal Normal Normal Increased proportion of bilirubin monoglucuronide
Hepatic UGT1A1 activity Effect of phenobarbital on serum bilirubin Mode of inheritance Prevalence
340–850 mmol/l Normal Normal Normal Normal Usually pale: contains small amounts of unconjugated bilirubin None None
10% of normal or less Reduction by 25% or more
25–40% of normal Reduction
Autosomal recessive Rare
Autosomal recessive Uncommon
Prognosis Animal model
Kernicterus is the rule Gunn rat
Usually benign; kernicterus is rare —
Autosomal recessive Common, ~5% in general population Benign Bolivian squirrel monkey ? Mutant Southdown sheep
Serum bilirubin levels Routine liver function tests Serum bile acid levels Oral cholecystography Liver histology Bile conjugates
UGT1A1, bilirubin uridine diphosphoglucuronate glucuronosyltransferase.
1466
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
was found to result from an absence of bilirubin-UGT activity.255 Severe indirect hyperbilirubinemia in the absence of hemolysis was observed in all cases within the first few days of life, and persisted lifelong. Five of the six infants in the initial series died of kernicterus by the age of 15 months. The single survivor was free of neurologic disease until 15 years of age, but then developed kernicterus and died 6 months later. In a related patient neurologic symptoms developed at 18 years of age; she died at the age of 24.256 These early cases suggested the heterogeneity of Crigler–Najjar syndrome, which became clearly established when Arias and associates subsequently described a milder variant of the condition, termed Crigler–Najjar syndrome type 2257 (see below). Several other recessively inherited traits, such as Morquio’s syndrome, homocystinuria, metachromatic leukodystrophy, and bird-headed dwarfism were found in some families, but were subsequently not found to be associated with Crigler–Najjar syndrome. Since the initial report, several hundred patients with Crigler–Najjar syndrome type 1 have been described.258 The syndrome occurs in all races and is transmitted as an autosomal recessive trait.254,255 The incidence appears to be approximately one in a million live births, although the precise frequency of this disease in various populations is not known. As in all rare autosomal recessive disorders, a high incidence of consanguinity among the parents has been observed. Physical examination is normal except for jaundice, and sometimes, evidence of neurological damage. The institution of routine use of phototherapy and intermittent application of exchange transfusion for clinical emergencies has extended the life expectancy beyond adolescence. However, around puberty phototherapy becomes less effective and the bilirubin load increases. Therefore, patients always remain at the risk of kernicterus258 and liver transplantation remains the only definitive treatment. Laboratory Tests. Biochemical tests of the serum are normal in Crigler–Najjar syndrome type 1, except for a high serum bilirubin level, which usually ranges from 18 to 30 mg/dl, but may be as high as 50 mg/dl.258 As serum bilirubin is all unconjugated, bilirubin is not present in the urine. Plasma bilirubin concentrations fluctuate according to the level of exposure to the sun or other light, and increases during intercurrent illness.255,258 The bile lacks bilirubin glucuronides and may contain only small amounts of unconjugated bilirubin. Although fecal urobilinogen excretion is reduced, stool color remains normal.254 Bilirubin production rate is normal254,259 and there is no evidence of hemolysis.254,257 Intact bile canalicular excretion mechanisms are evidenced by normal plasma disappearance of BSP256 and indocyanine green,254 and radiologic visualization of the biliary tree by cholecystographic agents. Liver biopsy reveals normal histology, except for bilirubin plugs in bile canaliculi and bile ducts,254,258 probably resulting from biliary excretion of unconjugated bilirubin or its photoisomers. A high incidence of pigment gallstones, often requiring cholecystectomy, is seen in patients with Crigler–Najjar syndrome. Electron microscopy of the liver reveals no specific pathologic change.260 Abnormalities of Hepatic UGTs and their Molecular Mechanisms. All Crigler–Najjar syndrome type 1 patients lack hepatic UGT activity toward bilirubin, but some have additional abnormalities of phenol glucuronidation.261 The mechanism of the abnormality of single or multiple isoforms of hepatic UGTs in Crigler–Najjar syndrome type 1 was clarified when the molecular basis of this disor-
der became known in 1992.262–264 The structure of the UGT1A locus dictates that genetic lesions located in any of the four common region exons (exon 2-5) should cause a defect of all isoforms expressed from the UGT1A locus (Figure 74-6), whereas sequence abnormalities within the unique exon 1A1 should affect only bilirubin glucuronidation, which is mediated by UGT1A1. Since these initial reports, over 50 genetic lesions have been identified in more than 100 Crigler–Najjar syndrome patients and in many of their immediate relatives. These studies have been reviewed and tabulated in reference 265. Analysis of this cumulative experience shows that genetic lesons, such as point mutations, deletions, or insertions, within any of the five exons constituting UGT1A1 mRNA, can inactivate the enzyme. Effects of singleamino-acid substitutions are being studied by site-directed mutagenesis of expression vectors, followed by transfection into COS cells. Furthermore, in three cases, intronic mutations at the splice donor or splice acceptor sites have been found to cause abnormal splicing at cryptic splice sites within exons, leading to frameshift and truncation of the enzyme.265,266 The molecular genetic studies confirm the autosomal recessive inheritance pattern of Crigler–Najjar syndrome type 1. A high incidence of consanguinity in the families is reflected by the observation that, in the majority of patients, both alleles contain the same genetic lesion. However, in some cases no history of consanguinity exists and the patients are compound heterozygotes, that is, a different genetic lesion is present in each allele. Although in a great majority of cases, the genetic lesion is inherited from both parents, recently an instance of uniparental isodisomy has been reported, in which both mutant alleles were inherited from the father.267 The mother’s UGT1A1 genotype was normal. This case underscores the need for analyzing the genotype of both parents in order to ascertain the mode of inheritance of inherited jaundice. Availability of modern molecular biological techniques has made it feasible to identify sequence abnormalities of the UGT1A1 gene from genomic DNA extracted from any nucleated cell (e.g., blood leukocytes or buccal smear). This should assist genetic counseling, by the identification of heterozygous carriers and prenatal diagnosis based on genomic DNA analysis of cultured chorionic villous cells. The Gunn Rat, an Animal Model of Crigler–Najjar Syndrome Type 1. In 1938, Gunn described a mutant strain of Wistar rats that exhibited lifelong hyperbilirubinemia, inherited as an autosomal recessive characteristic.268 Although the mechanism of the nonhemolytic unconjugated hyperbilirubinemia was not known at that time, Professor William E. Castle maintained the mutants for over 15 years, until the deficiency of bilirubin glucuronidation was shown to be the mechanism of jaundice in this strain.269 The Gunn rat is both a metabolic and molecular model of Crigler–Najjar syndrome type 1.270 Studies using this animal model have led to major advances in understanding bilirubin throughput and toxicity. As in Crigler–Najjar syndrome type 1, Gunn rats have high concentrations of serum bilirubin, which is all unconjugated. The bile contains no conjugated bilirubin and liver histology is normal.270 The Gunn rat is the only animal model that develops bilirubin encephalopathy spontaneously.271 Much of the present information on bilirubin encephalopathy was derived from studies in the Gunn rat (see the section on bilirubin toxicity, above). Studies performed
1467
Section XII. Inherited and Pediatric Diseases of the Liver
in the Gunn rat also helped in developing new treatment modalities for hyperbilirubinemia, including cell transplantation and gene therapies. Enzyme and molecular abnormalities in Gunn rats include that Gunn rat livers have no UGT activity toward bilirubin.269 Canalicular transport of exogenously administered conjugated bilirubin is normal.272 In addition to lacking hepatic bilirubin-UGT activity, Gunn rat livers lack UGT activity toward digitoxogenin monodigitoxoside273 and the methylcholanthrene-inducible phenol-UGT activity.274,275 Abnormality of multiple isoforms of the UGT1A family is explained by the genetic lesion in Gunn rats, which consists of the deletion of a single guanosine residue from exon 4 of the ugt1a1 gene. This genetic lesion results in a premature stop codon, leading to the truncation of carboxy-terminal 150 amino acids at the carboxy-terminal end of the UGTs and the inactivation of their catalytic function.276 Because this deletion is in one of the common region exons, all isoforms of the UGT1A subfamily are affected. UGT isoforms expressed from other genes are normal. Several UGT isoforms with normal catalytic activity have been isolated from the liver of Gunn rats.274 Treatment of Crigler–Najjar Syndrome Type 1. Conventional treatment aims at reducing serum bilirubin levels. Unlike the results in patients with Crigler–Najjar syndrome type 2, and Gilbert syndrome, serum bilirubin levels in Crigler–Najjar syndrome type 1 are not significantly reduced by phenobarbital administration.257 Phototherapy. Phototherapy is the mainstay of medical therapy for severe unconjugated hyperbilirubinemia.277 Banks of fluorescent lamps with devices for shielding the eyes or “light blankets” effectively lower serum bilirubin levels by mechanisms that have been discussed in the section on bilirubin chemistry. About the time of puberty, phototherapy becomes progressively less effective due to thickening of the skin, increased skin pigmentation, and decreased surface area in relation to body mass.258 Plasmapheresis. In neurological emergencies, plasmapheresis is an efficient method for reducing serum bilirubin concentration acutely.256,258 Because bilirubin is tightly bound to plasma albumin, removal of albumin results in the withdrawal of equimolar amounts of bilirubin. This is followed by mobilization of tissue bilirubin stores to the plasma. Attempts to remove plasma bilirubin by affinity chromatography on albumin-conjugated agarose gel columns were successful in Gunn rats,278 but were limited by the additional removal of formed elements of simian or human blood.279 Orthotopic Liver Transplantation. Because, at present, there is no other definitive long-term treatment for patients with this condition, orthotopic liver transplantation has become a standard treatment of Crigler–Najjar syndrome type 1. Although this procedure is not without risk, it has been curative in several patients and has dramatically changed the prognosis of Crigler–Najjar syndrome type 1 patients.280
Experimental Methods Aimed at Reduction of Serum Bilirubin Levels Inhibition of Heme Oxygenase. Administration of non-iron metalloporphyrins for inhibition of microsomal HO activity19 has resulted in the suppression of neonatal hyperbilirubinemia in rats281 and rhesus monkey.282 Administration of tin-mesoporphyrin,
1468
0.5 mmol/kg three times a week for 13–23 weeks in two 17-yearold boys with Crigler–Najjar syndrome type 1 resulted in only a modest and variable degree of reduction in serum bilirubin levels.283 Although the place of this agent in the treatment of Crigler–Najjar syndrome type 1 is not fully clarified, these agents may have a role in reducing serum bilirubin levels during acute emergencies. Bilirubin Degradation by Bilirubin Oxidase. Bilirubin oxidase from Myrothecium verrucaria284 catalyzes the oxidation of bilirubin to a colorless product. Bilirubin oxidase, linked to polyethyleneglycol for increasing its half-life in circulation, has been administered intravenously,285 and resulted in substantial reduction of serum bilirubin levels in Gunn rats for 3 hours. Induction of P450c. As discussed earlier, the induction of P450c results in increased oxidative degradation of serum bilirubin in Gunn rat liver, resulting in the reduction of serum bilirubin levels. Several naturally occurring indoles extracted from cruciferous vegetables, such as cabbage, cauliflower, and Brussels sprouts, induce the P450 isoforms, CYP1A1 and CYP1A2, in rat liver and intestine.286 Indole3-carbinol, an inducer of CYP1A2, has been reported to reduce serum bilirubin levels temporarily in patients with Crigler–Najjar syndrome type 1.287 Methods Aimed at Replacing Hepatic UGT1A1 Activity. Because UGT1A1 activity is present in excess in normal liver, partial replacement of the enzyme should ameliorate hyperbilirubinemia in Crigler–Najjar syndrome type 1. As rat kidney contains a low but significant level of UGT1A1 activity, transplantation of a normal Wistar rat kidney into homozygous Gunn rat results in the excretion of bilirubin glucuronides in bile and reduction of serum bilirubin concentrations.288 However, human kidney lacks bilirubin-UGT activity and, therefore, renal transplantation cannot be used for the treatment of Crigler–Najjar syndrome type 1. Liver Cell Transplantation. Transplantation of hepatocytes is technically easier than liver transplantation and, because the host liver is retained, the consequence of graft loss is minimized. For these reasons, liver cell transplantation is being evaluated as a potential treatment for Crigler–Najjar syndrome type 1. Based on experience in rodent and murine models,289,290 hepatocytes were transplanted into a 10-year-old girl with Crigler–Najjar syndrome type 1.291 Transplantation of 7.5 billion hepatocytes resulted in partial amelioration of jaundice, and permitted reduction of the daily duration of phototherapy. Two and a half years later, bilirubin glucuronide excretion in bile continued, but the serum bilirubin level gradually increased, probably due to increased bilirubin production or reduced effectiveness of phototherapy. The patient received an auxiliary liver transplantation, which has kept her serum bilirubin within nomal limits (J. Roy Chowdhury, personal communication). Experience in this case, as well as worldwide experience with hepatocyte transplantation, indicates that the number of adult hepatocytes that can be transplanted at a single procedure is not likely to be sufficient for a complete cure of inherited metabolic diseases of the liver.292 Moreover, the shortage of good-quality donor livers for hepatocyte isolation has limited more general application of hepatocyte transplantation.292,293 Preparative irradiation and other manipulations of the host liver are being explored for inducing preferential proliferation of the engrafted hepatocytes.294,295
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
Gene Therapy. Because the metabolic defect in Gunn rats and in Crigler–Najjar syndrome type 1 results from lesions of a single gene, supplementation with a normal bilirubin-UGT gene would be an attractive potential therapeutic modality. Methods for gene introduction into the liver using recombinant viruses or ligands that mediate receptor-directed endocytosis are being evaluated for this purpose. In the ex-vivo approach, liver cells harvested from a mutant subject or animal by partial hepatectomy are established in primary culture and transduced with normal (therapeutic) genes and transplanted back into the donor.296 In the in-vivo approach, genes are introduced into the organ of live organisms using recombinant viruses or non-viral vectors.297 Recombinant adenoviral vectors are highly efficient in transferring therapeutic genes into the liver. However, as adenoviruses are episomal, the effect is not permanent. Moreover, the host immune response toward adenoviral proteins can cause toxicity and precludes repeated injection of the vector. Adenoviral vectors have been developed that do not evoke an immune response, because of the co-expression of an immunosuppressive gene product.298 However, the safety of abrogating host immunity toward adenoviruses, which are potential human pathogens, remains doubtful. A T-antigen-deleted recombinant simian virus 40 has been used successfully in gene therapy in Gunn rats.299 These vectors integrate into the host genome and are not immunogenic.300 Site-directed gene repair by triggering the cells’ gene repair enzymes using RNA-DNA hybrid oligonucleotides has been used to replace the deleted G residue in Gunn rats, resulting in expression of normal UGT1A1 and partial amelioration of jaundice.301 This last approach, although highly interesting, has not yet provided enough efficiency for application in clinical gene therapy. These, and other approaches that are currently being investigated, raise the hope that hepatocyte transplantation, gene therapy, or the combination of the two should eventually result in curative medical therapy for Crigler–Najjar syndrome and other inherited metabolic liver diseases.302
Crigler–Najjar Syndrome Type 2 (Arias’ Syndrome) Clinical Features. In 1962, Arias showed that some patients with chronic unconjugated hyperbilirubinemia differ from patients with classic Crigler–Najjar syndrome type 1 in that they have somewhat lower serum bilirubin concentrations, their serum bilirubin levels are reduced by at least 25% following phenobarbital treatment and, in general, they have a better prognosis.257,303 In most patients jaundice is noted before the age of 1 year, but occasionally, it may only come to the attention of the physician in adult life. Serum bilirubin concentrations usually range from 8 to 18 mg/dl, and are mostly indirect-reacting. Hepatic bilirubin-UGT activities are markedly reduced. As in Crigler–Najjar syndrome type 1, red-cell survival is normal and there is no other clinical abnormality. Kernicterus is uncommon in Crigler–Najjar syndrome type 2, but can occur under stressful situations. Serum bilirubin may rise to very high levels following general anesthesia and surgery, with precipitation of encephalopathy. In a female patient who developed signs of bilirubin encephalopathy at the age of 43 years and died at the age of 44, autopsy showed that the brain was small.257 There was no bilirubin staining, but histological features typical of kernicterus
were seen. Several additional cases with Crigler–Najjar syndrome type 2, who developed neurological lesions were reported.304,305 Laboratory Tests. As in Crigler–Najjar syndrome type 1, laboratory examination is normal except for serum bilirubin concentrations of 8–18 mg/ml. Serum bilirubin levels may increase to as high as 40 mg/dl during fasting303 or intercurrent illness.304 By diazo analysis, serum bilirubin is mostly indirect reacting. In contrast to Crigler–Najjar syndrome type 1, the bile contains significant amounts of bilirubin glucuronides, although less than 50% of estimated daily bilirubin production is excreted into bile.257,305 The proportion of bilirubin monoglucuronide in bile exceeds 30% of total conjugated bilirubin,306,307 reflecting the reduced UGT1A1 activity in the liver. The liver histology is normal. Hepatic UGT1A1 activity is markedly reduced, but is detectable using sensitive techniques.305,306 Molecular Mechanism and Inheritance. As in Crigler–Najjar syndrome type 1, this syndrome is caused by genetic lesions within the various exons that constitute the coding region of UGT1A1.307 However, in Crigler–Najjar syndrome type 2, the genetic lesions always consist of single amino acid substitutions that reduce, but do not completely abolish, UGT1A1 activity. In some cases, the residual enzyme activity is enough to result in only a minor elevation of serum bilirubin levels that are compatible with the diagnosis of Gilbert syndrome. Mutations of the coding region of UGT1A1 that are known to partially reduce the enzyme activity have been reviewed and tabulated in reference 265. Both autosomal dominant transmission with incomplete penetrance257 and autosomal recessive transmission308 had been suggested for Crigler–Najjar syndrome type 2. Molecular genetic studies have confirmed autosomal recessive inheritance. Part of the confusion about the mode of inheritance arose from the observation that some family members of patients with Crigler–Najjar syndrome types 1 and 2 have intermediate levels of hyperbilirubinemia. A molecular explanation for this phenomenon is now available. In heterozygous carriers of Crigler–Najjar syndrome, if the allele with normal coding region carries a variant promoter (“Gilbert-type” TATAA element; see below), expression of the only normal allele is reduced. As Gilbert syndrome is very common in the general population, this type of compound heterozygosity is a more common cause of intermediate levels of hyperbilirubinemia than is homozygosity for a coding region mutation.309
GILBERT SYNDROME A common disorder, characterized by mild, chronic, fluctuating increase of serum unconjugated bilirubin levels, was described by Gilbert and Lereboullet in 1901.310 Although various investigators have used other names for this disorder, such as constitutional hepatic dysfunction, hereditary hemolytic bilirubinemia, and familial non-hemolytic jaundice, Gilbert syndrome is the most commonly used name for this condition.
Clinical Features Gilbert syndrome is usually diagnosed in young adults who are found to have mild, predominantly unconjugated hyperbilirubinemia, often during blood tests performed for other reasons, such as preemployment or pre-insurance screening, or intercurrent illness. In
1469
Section XII. Inherited and Pediatric Diseases of the Liver
most cases, bilirubin levels are less than 3 mg/dl, but fluctuate with time, and increase during intercurrent illness, stress, or menstrual periods.311 Jaundice is the only positive finding on physical examination. The vague constitutional symptoms, such as fatigue and abdominal discomfort, experienced by some patients311 may be manifestations of anxiety. Routine laboratory tests are normal, except the predominantly unconjugated hyperbilirubinemia. Oral cholecystography allows visualization of the gallbladder. Liver biopsy is not routinely indicated, but when performed shows normal histology. Hepatic UGT1A1 activity is reduced to approximately 30% of normal.312 Gilbert syndrome is diagnosed with much more frequency in males than in females,313 probably reflecting higher rates of bilirubin production in males.29 Gilbert syndrome is often diagnosed around puberty, which may be related to increased hemoglobin turnover, and possibly the inhibition of bilirubin glucuronidation by endogenous steroid hormones.314
The Genetic Basis of Gilbert Syndrome The molecular background of reduced UGT activity in patients with Gilbert syndrome has been unraveled. The promoter region of the UGT1A1 gene contains a TATAA element with a common A(TA)6TAA motif. Caucasian patients with Gilbert syndrome are homozygous for a longer TATAA element, A(TA)7TAA.315 The UGT1A1 gene with this variant TATAA element has been termed UGT1A1*28. The frequency of the variant promoter is approximately 30% among Caucasian and black populations,315,316 resulting in homozygosity in about 9%. Although homozygosity for the Gilbert-type promoter is required for Gilbert syndrome in these populations, all subjects who are homozygous for this polymorphism do not manifest the full clinical phenotype of Gilbert syndrome. For example, Gilbert syndrome is diagnosed infrequently in women. Additional factors, such as increased bilirubin production, may be required for manifestation of the Gilbert phenotype. In some, but not all, subjects with Gilbert syndrome, a defect of hepatic bilirubin uptake has been observed. However, the relationship between the uptake defect and reduced bilirubin glucuronidation is unclear. Promoter-reporter studies show that an increased TATAA box length reduces UGT1A1 expression.315 Patients with the Gilbert genotype have been shown to have lower hepatic microsomal UGT1A1 activity.317 In peoples of African origin, a small percentage of the population have an even longer TATAA element, A(TA)8TAA. Others have a shorter than usual TATAA element, A(TA)5TAA. There appears to be an inverse relation between the TATA box length and gene expression: TATA boxes with seven and eight repeats are associated with a decreased expression and five TA repeats with an increased expression.318 The A(TA)7TAA variant is less frequent among Japanese populations. In Japanese, Korean, and Chinese people heterozygosity for missense mutations in the UGT1A1 coding region has been reported to be a more common cause of Gilbert syndrome.319–321 Compound heterozygotes for the common structural G71R mutation and the variant TATAA element have also been described. So far, these mutations have only been found in races originating in the Far East. Gilbert-syndrome genotypes have been reported to be associated with accelerated or prolonged neonatal jaundice.322–324 In children with a combination of glucose-6-phosphate dehydrogenase defi-
1470
ciency and the Gilbert-type TATAA element, neonatal serum bilirubin concentrations can rise to dangerously high levels.325 In patients with splenomegaly (P.L.M. Jansen, unpublished observation) and in liver transplant recipients, the Gilbert-type promoter can cause prolonged spontaneous unconjugated hyperbilirubinemia.326,327 Gilbert syndrome is also correlated with early development of gallstones in patients with hereditary spherocytosis.328 Oxidative acetaminophen metabolism is associated with drug toxicity. Conflicting evidence has been obtained for a possibly increased oxidative metabolism and a decreased acetaminophen glucuronidation in patients with Gilbert syndrome.329–331 Gilbert syndrome is associated with a high incidence of diarrhea in patients treated with the anticancer drug, irinotecan.332
Organic Anion Transport While all subjects with Gilbert syndrome have reduced bilirubinUGT activity, some patients exhibit additional abnormalities of organic anion transport. Clearance of intravenously administered bilirubin is reduced in Gilbert syndrome.333 Goresky and associates found normal initial uptake in Gilbert syndrome, suggesting that the reduced clearance is due to decreased glucuronidation.334 However, multicompartmental analysis suggests that the reduction of plasma clearance may also be related to decreased hepatic bilirubin uptake.335 Subsequent studies have shown that subjects with Gilbert syndrome may be heterogeneous in terms of organic anion transport. In two subsets, BSP336 and indocyanine green (ICG)337 plasma disappearance is abnormal. In one group, reduced plasma clearance of BSP and ICG appears to be due to decreased hepatic uptake. In the other group compartmental analysis showed a defect in BSP transport at a later stage in the transport process. Since the excretion of BSP and ICG into bile is normal in Crigler–Najjar syndrome type 1 and Gunn rats, decreased clearance of these organic anions in Gilbert syndrome cannot be attributed to reduced UGT1A1 activity. At this time, no mechanism for the association of decreased UGT1A1 activity and reduced organic anion uptake is known, and the presence of the two apparently unrelated abnormalities in subsets of subjects with Gilbert syndrome appears be coincidental, perhaps owing to the high prevalence of Gilbert syndrome in the general population.
Effect of Fasting A two- to threefold increase in serum bilirubin levels occur in Gilbert syndrome upon reduction of daily caloric intake to 400 kcal for 48 hours.338 Because fasting also increases serum bilirubin levels in normal individuals339,340 and in individuals with other hepatobiliary disorders,340 the fasting test is of limited use in the differential diagnosis of Gilbert syndrome. Fasting-induced hyperbilirubinemia in normal individuals probably results from multiple physiological factors. Clearly, reduction of bilirubin glucuronidation cannot be the sole reason, since fasting also exacerbates hyperbilirubinemia in homozygous Gunn rats.341 Increased heme catabolism during fasting has been reported.329,331,340,342 Decreased intestinal motility during fasting may contribute to the bilirubin load by increasing bilirubin reabsorption from the gut.343 However, earlier kinetic studies had indicated that reduced hepatic clearance of bilirubin from plasma was more impor-
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
tant in fasting hyperbilirubinemia than increased bilirubin load.344 Reduced hepatic uptake of bilirubin due to down-regulation of organic anion transport345 or competition with serum free fatty acids may explain these observations.346,347 While increased bilirubin load and reduced hepatic uptake appear to contribute to fasting hyperbilirubinemia, an inverse relationship between hepatic UGT1A1 activity and the increase of serum bilirubin concentration during caloric restriction has been demonstrated.348 In normal subjects, homozygous for a normal UGT1A1 TATA element sequence [A(TA)7TAA], the mean increase in serum bilirubin levels after caloric restriction was 9.6 and 4.1 mmol/l, in males and females respectively. In subjects who were homozygous for the Gilbert-type promoter [A(TA)8TAA], the increment was significantly greater (20.5 ± 7.2 mmol/l), but in these subjects there was no significant gender difference. Interestingly, the mean increment in serum bilirubin levels was also slightly, but significantly increased in heterozytes for the promoter (A(TA)7TAA/A(TA)8TAA), suggesting a rateliming role of UGT1A1 in serum bilirubin clearance, particularly during fasting. It has suggested that a reduction of hepatic UDPglucuronic acid content during fasting may result in a reduction of UGT1A1 activity in vivo.349
Effect of Nicotinic Acid Administration Intravenous nicotinic acid administration increases unconjugated hyperbilirubinemia, probably by increasing erythrocyte fragility, splenic HO activity, and splenic bilirubin formation.350 Consistent with this, splenectomy prevents nicotinic acid-induced unconjugated hyperbilirubinemia.351 Although nicotinic acid administration has also been proposed as a provocative test for the diagnosis of Gilbert syndrome,351,352 like fasting, it does not clearly separate patients with Gilbert syndrome from normal subjects or those with hepatobiliary disease.352
Bilirubin Conjugates in Bile As in Crigler–Najjar syndrome type 2, an increased proportion of bilirubin glucuronides excreted in bile in Gilbert syndrome is bilirubin monoglucuronide.306 The alteration of bilirubin diglucuronideto-monoglucuronide ratio may reflect reduced hepatic bilirubin-UGT activity in these syndromes.
Diagnosis Gilbert syndrome is conventionally diagnosed in individuals with mild unconjugated hyperbilirubinemia without evidence of hemolysis or structural liver disease. Hemolysis is not a feature of Gilbert syndrome; however, coexistent compensated hemolysis is found in many patients with Gilbert syndrome, because increased bilirubin production makes the hyperbilirubinemia more clinically apparent.352 A presumptive diagnosis of Gilbert syndrome is made when mild unconjugated hyperbilirubinemia is noted on several occasions, and serum levels of glutamic oxaloacetic acid pyruvic transaminase, alkaline phosphatase, g-glutamyl transpeptidase and fasting and postcibal bile acids are normal. If confirmation of diagnosis is essential, chromatographic determination of the relative content of bilirubin monoglucuronide and diglucuronide in bile is of potential use in the diagnosis of Gilbert syndrome. Genetic analysis has greatly facilitated the diagnosis of Gilbert syndrome.
Animal Model Bolivian squirrel monkeys have higher serum unconjugated bilirubin concentration than do the closely related Brazilian squirrel monkeys.353 The difference in bilirubin levels becomes exaggerated in the fasting state. Compared to the Brazilian population, Bolivian monkeys have slower plasma clearance of intravenously administered bilirubin, a lower level of hepatic bilirubin-UGT activity, and a higher bilirubin monoglucuronide-to-diglucuronide ratio in bile. The two populations of squirrel monkeys have comparable erythrocyte lifespan and hepatic GST activity. In these respects, the Bolivian squirrel monkeys are a model of human Gilbert syndrome. Fasting hyperbilirubinemia is rapidly reversed by oral or intravenous administration of carbohydrates, but not by lipid administration.354
DISORDERS RESULTING IN PREDOMINANTLY CONJUGATED HYPERBILIRUBINEMIA Conjugated bilirubin may accumulate in the serum because of leakage of bilirubin glucuronides from hepatocytes, reverse transport from hepatocyte to the plasma, or regurgitation from bile. When the accumulation of conjugated bilirubin in plasma is due to inflammatory diseases of the hepatocyte, intrahepatic cholestasis or biliary obstruction, plasma concentrations of bile salts and various hepatocellular proteins are also expected to increase. However, in specific disorders of organic anion storage or transport, e.g., Rotor or Dubin–Johnson syndrome, respectively, plasma bile acid concentrations and liver enzyme levels remain normal, and hyperbilirubinemia is the predominant biochemical abnormality. Because of partial hydrolysis by hepatic b-glucuronidase and reversibility of the UGT1A1-catalyzed reaction, conjugated hyperbilirubinemia is always associated with various degrees of unconjugated hyperbilirubinemia.
ACQUIRED DEFECTS OF HEPATOBILIARY TRANSPORT Cholestasis is the result of impaired hepatobiliary transport. Hepatobiliary transport starts at the level of the basolateral membrane of the hepatocytes and includes transcellular and canalicular transport; cholestasis may result from disturbances of either of these. In addition, bile flow in the intrahepatic or extrahepatic part of the biliary tree may be impaired. Clinical cholestasis mainly results from disturbances of bile flow through inflammation or obstruction of intrahepatic or extrahepatic biliary tree. Common cholestatic diseases include primary biliary cirrhosis, primary sclerosing cholangitis, cholangiocarcinoma, gallstone disease, and papillary or pancreatic tumors. Also inflammation of the liver parenchyma, as occurs in hepatitis, usually causes some degree of cholestasis. Some drugs are well-known for cholestatic reactions.355 Experimental cholestasis can result from impaired hepatic uptake, reduced Na+/K+-ATPase activity, increased permeability of tight junctions, disturbed function of microtubules or microfilaments, reduced ATP generation, impaired canalicular transport, or the formation of precipitates within the bile canaliculi. Table 74-2 reviews
1471
Section XII. Inherited and Pediatric Diseases of the Liver
the canalicular transporters and their substrates. Ethinylestradiol is one of the best-studied cholestatic agents, and results obtained from using this drug have led to major insights into the pathophysiology of bile secretion. Ethinylestradiol alters the membrane fluidity and decreases Na+/K+-ATPase activity in the basolateral membrane. It inhibits the hepatic uptake and the canalicular secretion of bile acids and, to a lesser extent, that of conjugated bilirubin.356,357 In rodents, estrogen administration reduces the expression and/or activity of several transporters, including ABC, bsep,358 mdr1a/1b, two MDR1 isoforms,359 and the multidrug resistance-associated protein-2 (mrp2).360 In addition, estrogen-induced cholestasis is associated with a reduced activity of the Na+/taurocholate co-transporting polypeptide (ntcp)359,361 and a reduction of endogenous bile acid synthesis.361 Most known physiological actions of estrogens are mediated by binding of the estrogen receptor to an estrogen response element in the cis-regulatory elements in genes. In addition, estrogens increase the hepatic expression of the short heterodimer partner, an atypical nuclear receptor that does not have a DNA-binding domain, represses the activity of several nuclear hormone receptors in vitro and is a target for FXR ligands.362 6-Ethyl chenodeoxycholic acid, a FXR ligand, protects against estrogeninduced cholestasis.363 Cholestasis is characterized by the accumulation of both bilirubin and bile acid conjugates in the blood. Under normal physiological conditions the basolateral membrane of the hepatocyte contains efflux carriers that mediate the secretion of metabolites from liver to blood. Physiologically these transporters may transport compounds that are metabolized in the liver to the blood for subsequent excretion by the kidney. In experimental cholestasis, as occurs during ethinylestradiol administration, bile duct ligation or endotoxin administration, a reorientation of ABC transporters takes place, resulting in a loss of polarity of hepatocytes. Under these conditions, ATP-dependent transporters such as mrp3 and mrp1, which are expressed at very low levels in normal liver, are induced and localize to the basolateral membrane of hepatocytes.185 Canalicular mrp2, and, to a lesser extent, bsep, are down-regulated during cholestasis.135,364 Mrp3 and mrp1 mediate the active transport of glucuronides, glutathione-conjugates, and bile salts.138,365 The basolateral expression of these ABC transporters may help to remove metabolites that could otherwise inhibit hepatic metabolism. Downregulation of the Na+-dependent taurocholate co-transporter may also serve to protect the hepatocyte in cholestatic conditions.366 Common cholestatic agents, such as chlorpromazine and ciclosporin, inhibit hepatic transport at multiple levels. Ciclosporin therapy is frequently associated with jaundice. Ciclosporin inhibits the hepatic uptake of both bile acids and nonbile-acid organic anions.367 It is a competitive inhibitor of taurocholate uptake in both basolateral and canalicular membrane vesicles but not of Na+-dependent alanine uptake.368 In addition to canalicular ATP-dependent transport of taurocholate, the drug also strongly inhibits the ATP-dependent transport of leukotriene C4 (a mrp2 substrate) and of daunorubicin (a P-glycoprotein substrate).368 In the rat, ciclosporin impairs both the bile acid-dependent and the bile acid-independent components of bile flow. As with ethinylestradiol, administration of ciclosporin to rats increases serum bile acid levels.368 With these drugs, serum bile acid levels are more sensitive indicators of cholestasis than are serum bilirubin levels.
1472
INHERITED DISORDERS OF EXCRETION OF CONJUGATED BILIRUBIN Dubin–Johnson Syndrome In 1954, Dubin and Johnson369 and Sprinz and Nelson370 described a syndrome with chronic non-hemolytic jaundice characterized by accumulation of conjugated bilirubin in serum and grossly pigmented, but otherwise histologically normal, livers. Clinical Findings. Except for mild icterus, physical examination is within normal limits. Patients are usually asymptomatic, although an occasional patient complains of weakness and vague abdominal pain, and, rarely, hepatosplenomegaly is observed.371,372 As serum total bile acid levels are normal,373 pruritus is absent. Hyperbilirubinemia is increased by intercurrent illness, oral contraceptives, and pregnancy.373 The diagnosis is often made after puberty, although some patients have been diagnosed during the neonatal period.372,374 Sometimes the disorder is noted for the first time when a woman becomes pregnant or receives oral contraceptives, which increase the hyperbilirubinemia to a clinically detectable level.373 Laboratory Tests. Complete blood count, prothrombin time, and serum levels of bile acids, transaminases, alkaline phosphatase, and albumin are normal.372,373 Serum bilirubin concentration is usually between 2 and 5 mg/dl but can be as high as 20–25 mg/dl. Over 50% of total serum bilirubin is direct-reacting. Serum bilirubin levels fluctuate and individual measurements may yield normal results. Because of the general abnormality of canalicular transport of nonbile-acid organic anions, oral cholecystography, even using a “double dose” of contrast material, usually does not visualize the gallbladder. However, visualization may occur 4 hours after administration of intravenous contrast medium.375 The liver is grossly black and light microscopy reveals a dense pigment (Figure 74-7), which on electron microscopy appears to be contained within lysosomes.376 Histochemical staining and physicochemical properties of the pigment suggest that the pigment is related to melanin.377 Following infusion of [3H] epinephrine into mutant Corriedale sheep (an animal model for Dubin–Johnson syndrome), the isotope is incorporated into the hepatic pigment,378 supporting the concept that the pigment is a melanin-like derivative. However, electron spin resonance spectroscopy shows that the pigment differs from authentic melanin.379 It may be composed of polymers of epinephrine metabolites.380,381 Interestingly, following liver disease, such as acute viral hepatitis, the pigment is cleared from the liver.382 Following recovery, the pigment reaccumulates slowly, starting from the centrizonal region. Organic Anion Transport. The hepatic secretion of bilirubinglucuronides and the glutathione conjugate of BSP is disturbed in these patients.272,383 The hepatic secretion of negatively charged iopanoic acid is disturbed whereas the secretion of neutral iopamide is normal. The bile acid secretion is unaffected.383 Pharmacokinetic analysis of plasma disappearance of bilirubin, BSP and ICG revealed normal hepatic storage but impaired secretion.272,383,384 Impaired canalicular secretion of non-bile-acid organic anions represents the basic defect of this syndrome. After intravenous injection of BSP, plasma BSP concentration decreases at near-normal rate for 45 minutes. However, in 90% of patients, plasma BSP concentration increases after this time, so that the concentration at 90 minutes is
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS Figure 74-7. Pigmentation of the liver in DubinJohnson syndrome: hematoxylineosin stained section of the liver shows deep brown pigments in hepatocytes, particularly in the perivenous zone (Zone 3).
greater than that at 45 minutes.360 This secondary rise results from reflux of glutathione-conjugated BSP from hepatocytes into the circulation. A similar secondary rise occurs after intravenous administration of bilirubin.383 The secondary rise of plasma BSP has been observed in other hepatobiliary disorders,384 and is, therefore, not diagnostic of Dubin–Johnson syndrome. The Genetic Background of Dubin–Johnson Syndrome. Dubin–Johnson syndrome is caused by mutations of the MRP2 gene causing a deficiency of canalicular MRP2 expression.177–179 More than a dozen genetic lesions, including nucleotide transition of a single nucleotide deletion, resulting in amino acid substitutions, premature truncation, or exon skipping, have been reported to cause Dubin–Johnson syndrome.385 Some of the reported mutations may lead to impaired glycosylation of the MRP2 protein, impaired sorting to the canalicular membrane, and premature proteasomedependent degradation.386 As in TR– rats and EHBR rats, the absence of MRP2 (mrp2 in rats) in the canalicular membrane causes severe impairment of canalicular secretion of bilirubin conjugates, the leukotriene LTC4, reduced and oxidized glutathione, and numerous glucuronide and glutathione-conjugates.387 As a consequence, these patients and animals have a mild conjugated hyperbilirubinemia. Experiments with the animal models provided firm evidence for the existence of different pathways for the canalicular secretion of bilirubin conjugates and bile acids.388 In contrast to bile acids with a free 3-hydroxyl, 3-OH-conjugated bile acids are transported by
mrp2, rather than bsep, as evidenced by impaired secretion of bile acid conjugates in TR– and EHBR rats.389 Upon feeding TR– rats a diet enriched with tryptophan, tyrosine, and phenylalanine, intravenous injection of metanephrine results in the accumulation of a black lysosomal pigment in hepatocytes, identical to that seen in patients with Dubin–Johnson syndrome.390 Despite the absence of MRP2, the serum bilirubin levels are only mildly elevated in Dubin–Johnson syndrome and in the two rat models of the disease, suggesting alternative canalicular secretion pathways for bilirubin conjugates.391 Members of the mrp family, other than Mrp2, have been described in the rat bile canaliculi.392,393 Whether these mrps accept bilirubin glucuronides as substrates needs to be studied. MRP3 is expressed in the basolateral domain of the hepatocyte plasma membrane in patients with Dubin–Johnson syndrome. Activity of this transporter contributes to the conjugated hyperbilirubinemia by actively pumping bilirubin conjugates from liver to blood in these patients.137 Thus, the accumulation of conjugated bilirubin in the plasma of Dubin–Johnson syndrome patients may not be caused by passive “leakage” only, but also by active transport of bilirubin conjugates out of the hepatocytes. Inheritance. Dubin–Johnson syndrome is rare, but occurs in both sexes and in virtually all races. The syndrome occurs frequently (1 : 1300) in Persian Jews,372 in whom it is associated with clotting factor VII deficiency.394,395 It had been difficult to ascertain the inheritance pattern from clinical analysis,373 but with respect to
1473
Section XII. Inherited and Pediatric Diseases of the Liver
urinary coproporphyrin excretion (see below), Dubin–Johnson syndrome is inherited as an autosomal recessive characteristic.396,397 Urinary Coproporphyrin Excretion. Urinary coproporphyrin I excretion is increased in patients with Dubin–Johnson syndrome to a greater degree than in patients with other hepatobiliary disorders.397 Of the two isomers of coproporphyrin, isomer I and isomer III, coproporphyrin III is a precursor of heme, whereas other porphyrin isomers are metabolic byproducts of unknown significance and are excreted in urine and bile.396 Normally, approximately 75% of total urinary coproporphyrin is coproporphyrin isomer III. In Dubin–Johnson syndrome, total urinary coproporphyrin excretion is normal, but over 80% is coproporphyrin I (Figure 74-8).397,398 Although neonates normally have elevated urinary content of coproporphyrin I as compared to adults, levels are not as high as seen in Dubin–Johnson syndrome.399 In obligate heterozygotes (e.g., unaffected parents or children of subjects with Dubin–Johnson syndrome), total urinary coproporphyrin excretion was reduced by 40% of normal.397,399 The mechanism of the abnormal urinary porphyrin excretion and its relationship to the organic anion transport defect are not known. When the history and physical examination are consistent, the urinary coproporphyrin excretion pattern is diagnostic of Dubin–Johnson syndrome. However, the overlap of results in carriers with those in controls397 makes identification of heterozygotes difficult.
Animal Models
Mutant Corriedale Sheep. This mutant strain has an inherited defect that closely resembles the Dubin–Johnson syndrome. Biliary excretion of conjugated bilirubin, glutathione-conjugated BSP, iopanoic acid, and ICG is decreased, whereas taurocholate transport is normal.399,400 The secretion of the organic cation procaine amide ethobromide is unaffected400 and, interestingly, the secretion of unconjugated BSP is unimpaired.401 These sheep have a mild hyperbilirubinemia, with 60% of the bilirubin being conjugated. The liver is pigmented as in Dubin–Johnson syndrome,402 but the histology is otherwise normal. Total urinary coproporhyrin excretion is normal with increased excretion of coproporphyrin isomer I and decreased isomer III excretion. TR- Rats and EHBR Rats. These rats have a hepatic excretory abnormality that strongly resembles that of the mutant Corriedale sheep and patients with Dubin–Johnson syndrome. The biliary excretion of conjugated bilirubin and many other organic anions is impaired.403–405 As in Dubin–Johnson syndrome, coproporphyrin I constitutes a major fraction of the total urinary coproporphyrins.403 As explained above, the Mrp2 gene defect of TR– and EHBR rats has been characterized. Different single nucleotide deletions cause an absence of canalicular Mrp2 expression in these two animal models.179,406
Rotor Syndrome Clinical Findings. In 1948, Rotor et al. described several individuals from two families who had chronic predominantly conjugated hyperbilirubinemia.407 As in Dubin–Johnson syndrome, routine hematological tests and routine blood biochemistries are normal, except for conjugated hyperbilirubinemia.407,408 There is no evidence of hemolysis. Histomorphology of the liver is normal and, in contrast to the findings in Dubin–Johnson syndrome, the liver is not
1474
pigmented.409 Rotor syndrome is rare, but has been described in several races. Although Rotor syndrome shares many clinical features with Dubin–Johnson syndrome, the basic abnormality of these disorders are different.408 Organic Anion Excretion. In contrast to the findings in Dubin–Johnson syndrome, over 25% of injected BSP is retained in serum at 45 minutes after intravenous administration of 5 mg/kg BSP.410 There is no secondary rise of plasma BSP level in Rotor syndrome and conjugated BSP is not found in plasma.409 Following intravenous administration, there is also marked plasma retention of unconjugated bilirubin410 and ICG.411 Phenotypically normal obligate heterozygotes for Rotor syndrome have mildly abnormal BSP retention at 45 minutes, which is intermediate between results in affected patients and normal controls.410 Oral cholecystographic agents usually do not visualize the gallbladder in the Dubin–Johnson syndrome, whereas roentgenologic visualization is usually possible in Rotor’s syndrome.409 Hepatic storage and transport maximum of BSP have been determined by a constant infusion technique.410,411 In Dubin–Johnson syndrome the transport is severely abnormal, whereas the hepatic storage capacity is normal. In contrast, in Rotor’s syndrome, the storage capacity was reduced by 75–90%, whereas the transport maximum was reduced by only 50%.410,412 The findings in Rotor syndrome are similar to those in hepatic storage disease, a familial disorder manifested by predominantly conjugated hyperbilirubinemia and normal liver histology.413,414 These two disorders may represent a single pathophysiologic entity. Urinary Coproporphyrin Excretion. In contrast to the findings in Dubin–Johnson syndrome, total urinary coproporphyrin is increased by 250–500% over normal in Rotor syndrome, and the proportion of coproporphyrin I in urine is approximately 65% of total (Figure 74-8).408 In one report, however, two brothers with clinical Rotor’s syndrome had over 80% of urinary coproporphyrins as isomer I.415 These results are similar to those seen in many other hepatobiliary disorders.416 Obligate heterozygotes have a coproporphyrin excretory pattern that is intermediate between that of control subjects and patients with Rotor’s syndrome. With respect to urinary coproporphyrin excretion, Rotor’s syndrome is inherited as an autosomal recessive characteristic.408 The urinary coproporphyrin abnormality in Rotor syndrome is most likely caused by reduced biliary excretion of coproporphyrins, with a concomitant increase in renal excretion. The nature of the organic anion transport defect in Rotor’s syndrome is unknown. Similar and differential features of Dubin–Johnson syndrome and Rotor syndrome are listed in Table 74-4.
ACKNOWLEDGMENT This work was partly supported by the following National Institutes of Health grants: RO1-DK 39137, RO1-DK 46057, R01 A1 42295, P30-DK 41296, and 1MO1 RR 12248.
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
Table 74-4. Inherited Disorders Associated with Elevation of Conjugated Bilirubin
Serum bilirubin pattern and concentrations Routine liver function tests Serum bile salt levels Plasma BSP retention BSP infusion studies Oral cholecystography Urinary coproporphyrin excretion pattern Appearance of liver Histology of liver Mode of inheritance Prevalence Prognosis Animal model
Dubin–Johnson syndrome
Rotor syndrome
Predominantly conjugated Usually 50–85 mmol/l, can be as high as 340 mmol/l Normal except for hyperbilirubinemia Normal Normal at 45 minutes; secondary rise at 90 minutes Tm very low; storage normal Usually does not visualize the gallbladder Total: normal; >80% as coproporphyrin I
Predominantly conjugated Usually 50–100 mmol/l, occasionally as high as 340 mmol/l Normal except for hyperbilirubinemia Normal Elevated; but no secondary rise at 90 minutes
Grossly black Dark pigments, predominantly in centrilobular areas; otherwise normal Autosomal recessive Rare (1 : 1300 in Middle Eastern Jews) Benign Mutant TR- rats/Mutant Corriedale sheep/ Golden lion tamarind monkey
Normal Normal, no increase in pigmentation Autosomal recessive Rare Benign None
Tm and storage both reduced Usually visualizes the gallbladder Total: elevated; ~50–75% coproporphyrin I
BSP, Bromosulfophthalein.
REFERENCES
mg of coproporphyrin per g creatinine
500
400
300
200
100
0 Normal
D-J
D-J hetero
Rotor Rotor hetero
Figure 74-8. Urinary coproporphyrin excretion in normal subjects, and in Dubin–Johnson (D-J) syndrome and Rotor syndrome. The hatched bars represent total urinary coproporphyrin excretion, normalized to creatinine excretion. Coproporphyrin I (solid bar) and coproporphyrin III (open bar) excretion are also shown. Vertical bars represent 1 SEM. In Dubin–Johnson syndrome (D-J syndrome), the total urinary coproporphyrin excretion is normal, but the proportion of coproporphyrin I is markedly elevated (over 80%). In contrast, both total urinary coproporphyrin and the proportion of coproporphyrin I are increased in Rotor syndrome. (Data from Wolkoff AW, Wolpert E, Pascasio FN, Arias IM. Rotor’s syndrome: a distinct inheritable pathophysiologic entity. Am J Med 1976; 60:173.)
1. Chen TS, Chen PS. Understanding the liver. A history. Greenwood Press, CT: 1984:99. 2. Berk PD, Howe RB, Bloomer JR, et al. Studies on bilirubin kinetics in normal adults. J Clin Invest 1979; 48:2176. 3. London IM, West R, Shemin D, et al. On the origin of bile pigment in normal man. J Biol Chem 1950; 184:351. 4. Schwartz S, Johnson JA, Stephenson BD, et al. Erythropoietic defects in protoporphyria: a study of factors involved in labeling of porphyrins and bile pigments from ALA-3H and glycine-14C. J Lab Clin Med 1971; 78:411. 5. Grandchamp B, Bissel DM, Licko V, et al. Formation and disposition of newly synthesized heme in adult rat hepatocytes in primary cultures. J Biol Chem 1981; 256:11677. 6. Levitt M, Schacter BA, Zipursky A, et al. The nonerythropoietic component of early bilirubin. J Clin Invest 1968; 47:1281. 7. Robinson SH, Tsong M. Hemolysis of “stress” reticulocytes: a source of erythropoietic bilirubin formation. J Clin Invest 1970; 49:1025. 8. Come SE, Shohet SB, Robinson SH. Surface remodeling vs. whole-cell hemolysis of reticulocytes produced with erythroid stimulation or iron deficiency anemia. Blood 1974; 44:817. 9. Tenhunen R, Marver HS, Schmidt R. Microsomal heme oxygenase: characterization of the enzyme. J Biol Chem 1969; 244:6388. 10. Brown SB, King RFGJ. The mechanism of heme catabolism: bilirubin formation in living rats by [18O]oxygen labeling. Biochem J 1978; 170:297. 11. Yoshida T, Kikuchi G. Features of the reaction of heme degradation catalyzed by the reconstituted microsomal hemeoxygenase system. J Biol Chem 1978; 53:4230. 12. Maines MD, Trakshel GM, Kutty RK. Characterization of two constitutive forms of rat liver microsomal heme oxygenase. Only one molecular species of the enzyme is inducible. J Biol Chem 1986; 261:411.
1475
Section XII. Inherited and Pediatric Diseases of the Liver
13. Bissel DM, Hammaker L, Schmid R. Liver sinusoidal cells. Identification of a subpopulation for erythrocyte catabolism. J Cell Biol 1972; 54:107. 14. Makino N, Suematsu M, Sugiura Y, et al. Altered expression of heme oxygenase-1 in the livers of patients with portal hypertensive diseases. Hepatology 2001; 33:32. 15. Fernandez M, Bonkovsky HL. Increased heme oxygenase-1 gene expression in liver cells and splanchnic organs from portal hypertensive rats. Hepatology 1999; 29:1672. 16. Nishikawa Y, Stepp DW, Merkus D, et al. In vivo role of heme oxygenase in ischemic coronary vasodilation. Am J Physiol Heart Circ Physiol 2004; 286:H2296–H2304. 17. Rattan S, Al Haj R, De Godoy MA. Mechanism of internal anal sphincter relaxation by CORM-1, authentic CO, and NANC nerve stimulation. Am J Physiol Gastrointest Liver Physiol 2004; 287:G605–G611. 18. Yoshinaga T, Sassa S, Kappas A. The occurrence of molecular interactions among NADPH-cytochrome C reductase, heme oxygenase and biliverdin reductase in heme degradation. J Biol Chem 1982; 257:7778. 19. Cornelius CE. Comparative bile pigment metabolism in vertebrates. In: Ostrow JD, ed. Bile pigments and jaundice. New York: Marcel Dekker; 1986:601. 20. McDonagh AF, Palma LA, Schmid R. Reduction of biliverdin and placental transfer of bilirubin and biliverdin in the pregnant guinea pig. Biochem J 1981; 194:273. 21. Roy Chowdhury J, Roy Chowdhury N, Arias IM. Bilirubin conjugates in the spiny dogfish, Squalus acanthias, the small skate, Raja erinacea and the winter flounder Pseudopleuronectes americanas. Comp Biochem Physiol 1980; 66b:523. 22. Roy Chowdhury N, Roy Chowdhury J, Arias IM. Bile pigment composition and hepatic UDP-glucuronyl transferase activity in adult and fetal dogfish shark Squalus acanthias. Comp Biochem Physiol 1982; 73B:651. 23. Huang TJ, Trakshel GM, Maines MD. Detection of ten variants of biliverdin reductase in rat liver by two-dimensional gel electrophoresis. J Biol Chem 1989; 264:7844. 24. Fakhrai H, Maines MD. Expression and characterization of a rat kidney biliverdin reductase: evidence suggesting that the liver and kidney enzymes are the same gene transcript product. J Biol Chem 1992; 267:4023. 25. Nakao A, Otterbein LE, Overhaus M, et al. Biliverdin protects the functional integrity of a transplanted syngeneic small bowel. Gastroenterology 2004; 127:595–606. 26. Yamashita K, McDaid J, Ollinger R, et al. Biliverdin, a natural product of heme catabolism, induces tolerance to cardiac allografts. FASEB J 2004; 18:765–767. 27. Breimer LH, Wannamethee G, Ebrahim S, et al. Serum bilirubin and risk of ischemic heart disease in middle-aged British men. Clin Chem 1995; 41:1504–1508. 28. Bosma PJ, van dM I, Bakker CT, et al. UGT1A1*28 allele and coronary heart disease: the Rotterdam study. Clin Chem 2003; 49:1180–1181. 29. Zucker SD, Horn PS, Serman KE. Serum bilirubin levels in the US population: gender effect and inverse correlation with colorectal cancer. Hepatology 2004; 40:827–835. 30. Temme EHM, Zhang J, Schouten EG, et al. Serum bilirubin and 10-year mortality risk in a Belgian population. Cancer Causes control 2001; 12:887–894. 31. Brink MA, Slors JF, Keulemans YC, et al. Enterohepatic cycling of bilirubin: a putative mechanism for pigment gallstone formation in ileal Crohn’s disease [see comments]. Gastroenterology 1999; 116:1420. 32. Brink MA, Mendezsanchez N, Carey MC. Bilirubin cycles enterohepatically after ileal resection in the rat. Gastroenterology 1996; 110:1945. 33. Dhar GJ. Enterohepatic circulation and plasma transport of urobilinogen. In: Berk PD, Berlin NI, eds. Bile pigments:
1476
34.
35.
36.
37.
38. 39. 40.
41. 42.
43. 44. 45.
46.
47. 48. 49. 50.
51.
52.
53.
54.
55.
56.
chemistry and physiology. Washington: DHEW, NIH; 1977:526. Jones EA, Shrager R, Bloomer JR, et al. Quantitative studies of the delivery of hepatic synthesized bilirubin to plasma utilizing d-aminolevulinic acid-4-14C and bilirubin-3H in man. J Clin Invest 1972; 51:2450. Jones EA, Bloomer JR, Berk PD, et al. Quantitation of hepatic bilirubin synthesis in man. In: Berk PD, Berlin NI, eds. The chemistry and physiology of bile pigments. Washington: USDEW, NIH; 1977:189. Berk PD, Rodkey FL, Blaschke Collison HA, et al. Comparison of plasma bilirubin turnover and carbon monoxide production in man. J Lab Clin Med 1974; 83:29. Engel RR. Alternative sources of carbon monoxide. In: Berk PD, Berlin NI, eds. Bile pigments: chemistry and physiology. Washington: DHEW, NIH; 1977:148. Westlake DWS, Roxburgh JM, Talbot G. Microbial production of carbon monoxide from flavinoids. Nature 1961; 189:510. Merritt JE, Loening KL. Nomenclature of tetrapyrroles. Pure Appl Chem 1979; 51:2251. Fischer H, Plieninger H. Synthese des biliverdins (uteroverdins) und bilirubins der biliverdine XIII, und III, sowie der Vinulneoxanthosaure. Hoppe–Seylers A Physiol Chem 1942; 274:231. Bonnet RJ, Davis E, Hursthouse MB. Structure of bilirubin. Nature 1976; 262:326. Gray CH, Kulszycka A, Nicholson DC. The chemistry of bile pigments, IV. Spectrophotometric titration of the bile pigments. J Chem Soc 1961; 442:2276. Krasner J, Yaffe SJ. The automatic titration of bilirubin. Biochem Med 1973; 7:128. Lee KS, Gartner LM. Spectrophotometric characteristics of bilirubin. Pediatr Res 1976; 10:782. Brodersen R. Aqueous solubility, albumin binding and tissue distribution of bilirubin. In: Ostrow JD, ed. Bile pigments and jaundice. New York: Marcel Dekker; 1986:157. Brodersen R. Free bilirubin in the blood plasma of the newborn. Effects of albumin, fatty acids, pH, displacing drugs and phototherapy. Appendix A: Provisional survey of the bilirubin displacing effect of 150 drugs. In: Stern L, ed. Intensive care of the newborn, vol. 2. New York: Masson; 1978:331. Fog J, Jellum E. Structure of bilirubin. Nature 1963; 198:88. Kuenzle CC, Weibel MH, Pelloni RR. A proposed novel structure for the metal chelates of bilirubin. Biochem J 1973; 130:1147. Kuenzle CC, Weibel MH, Pelloni RR. The reaction of bilirubin with diazomethane. Biochem J 1973; 133:357. Lightner DA, Wooldridge TA, McDonagh AF. Configurational isomerization of bilirubin and the mechanism of phototherapy. Biochem Biophys Res Commun 1979; 86:235–243. Blanckaert N, Heirwegh KPM, Zaman Z. Comparison of the biliary excretion of the four isomers of bilirubin-IX in Wistar and homozygous Gunn rats. Biochem J 1977; 164:229. Yang BJ, Morris MD, Xie MQ, et al. Resonance Raman spectrometry of bilirubins: band assignments and application to bilirubin/lipid complexation. Biochemistry 1991; 30:688–94. McDonagh AF. Bilitrienes and 5,15 biladines. In: Dolphin D, ed. The porphyrins, vol VI, bile pigments. New York: Academic Press; 1979:293. Trull FR, Ibars O, Lightner DA. Conformational inversion of bilirubin formed by reduction of the biliverdin–human serum albumin complex: evidence from circular dichroism. Arch Biochem Biophys 1992; 298:710. McDonagh AF, Pu YM, Lightner DA. Effect of volatile anesthetics on the circular dichroism of bilirubin bound to human serum albumin. Experientia 1992; 48:246. Moan J. Light therapy of newborns with hyperbilirubinemia (translated from Norwegian). Tidsskr Norske Laegefor 1992; 112:1459.
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
57. Agati G, Fusi F, Pratesi R. Evaluation of the quantum yield for E to Z isomerization of bilirubin bound to human serum albumin. Evidence of internal conversion processes competing with configurational photoisomerization. J Photochem Photobiol 1993; 17:173. 58. Itoh S, Onishi S. Kinetic study of the photochemical changes of (ZZ)-bilirubin IX¢ bound to human serum albumin. Demonstration of (EZ)-bilirubin IX¢ as an intermediate in photochemical changes from (ZZ)-bilirubin IX≤ to (EZ)cyclobilirubin IX.≤ Biochem J 1985; 226:251. 59. Ennever JF, Dreising TJ. Quantum yields for the cyclization and configurational isomerization of 4E, 15Z bilirubin. Photochem Photobiol 1991; 53:25. 60. McDonagh AF, Palma LA, Lightner DA. Phototherapy for neonatal jaundice. Stereospecific and regiospecific photoisomerization of bilirubin bound to human serum albumin and NMR characterization of intramolecularly cyclized photoproducts. J Am Chem Soc 1982; 104:6867. 61. Cu A, Bellah GG, Lightner DA. On the fluorescence of bilirubin. J Am Chem Soc 1975; 97:2579. 62. Lightner DA. Products of bilirubin photodegradation. In: Berlin NI, Berk PD, Watson CJ, eds. Chemistry and physiology of bile pigments. Washington: DHEW, NIH; 1977:93. 63. McDonagh AF. Thermal and photochemical reactions of bilirubin IX. Ann NY Acad Sci 1975; 244:553. 64. Mollison PL, Cutbush M. Hemolytic disease of the newborn: criteria of severity. Br Med J 1949; 1:123. 65. Hsia DYY, Allen FH, Gellis SS, et al. Erythroblastosis fetalis, VIII. Studies of serum bilirubin in relation to kernicterus. N Eng J Med 1952; 247:668. 66. Diamond I, Schmidt R. Experimental bilirubin encephalopathy. The mode of entry of bilirubin 14C into the central nervous system. J Clin Invest 1966; 45:678. 67. Takahashi M, Sugiyama K, Shumiya S, et al. Penetration of bilirubin into the brain of albumin-deficient and jaundiced rats (AJR) and Nagase analbuminemic rats (NAR). J Biochem 1984; 96:1705. 68. Sata H, Temba R. Relationship between plasma unbound bilirubin concentration and cerebellar bilirubin content in homozygous Gunn rat sucklings. Experientia 1978; 34:221. 69. Jew JY, Williams TH. Ultrastructural aspects of bilirubin encephalopathy in cochelar nuclei of the Gunn rat. J Anat 1977; 124:599. 70. Sapiro SM. Reversible brainstem auditory evoked potential abnormalities in jaundiced Gunn rats given sulfonamide. Pediatr Res 1993; 34:629. 71. Takagishi Y, Yamamura H. Purkinje cell abnormalities and synaptogenesis in genetically jaundiced rats (Gunn rats). Brain Res 1989; 492:116. 72. Schutta HS, Johnson L, Neville HS. Mitochondrial abnormalities in bilirubin encephalopathy. J Neuropathol Exp Neurol 1970; 29:296. 73. Sato H, Keino H, Aono S, et al. Different behaviours among lysosomal enzymes in the cerebellum of jaundiced Gunn rats with cerebellar hypoplasia. J Neurochem 1987; 48:1823. 74. Aono S, Kato T, Tanaka R, et al. Cyclic nucleotides and the activity of glia maturation factor in the hypoplastic cerebellum of developing jaundiced Gunn rats. J Neurochem 1988; 50:700. 75. Labrune PH, Myara A, Francoual J, et al. Cerebellar symptoms as the presenting manifestations of bilirubin encephalopathy in children with Crigler–Najjar type I disease. Pediatrics 1992; 89:768. 76. Conlee JW, Shapiro SM. Morphological changes in the cochlear nucleus and nucleus of the trapezoid body in Gunn rat pups. Hearing Research 1991; 57:23. 77. Shapiro SW, Conlee JW. Brainstem auditory evoked potentials correlate with morphological changes in Gunn rat pups. Hearing Research 1991; 57:16.
78. Shapiro SM. Binaural effects in brainstem auditory evoked potentials of jaundiced Gunn rats. Hearing Res 1991; 53:41. 79. Zuelzer WW, Mudgett RT. Kernicterus. Etiologic study based on an analysis of 55 cases. Pediatrics 1950; 6:452. 80. Vaughan VC, Allen FC, Diamond LK. Erythroblastosis fetalis. IV. Further observations on kernicterus. Pediatrics 1950; 6:706. 81. Lucey JF. Bilirubin and brain damage – a real mess. Pediatrics 1982; 69:381. 82. Turkel SB, Miller CA, Guttenberg ME, et al. A clinical pathologic reappraisal of kernicterus. Pediatrics 1982; 69:267. 83. Van de Bor M, Ens-dokkum M, Schreuder AM, et al. Hyperbilirubinemia in low birth weight infants and outcome at 5 years of age. Pediatrics 1992; 89:359. 84. Rappaport SI. Sites and functions of blood–brain barrier. In: Blood–brain barrier in physiology and medicine. New York: Raven Press; 1976:43. 85. Purpura DP, Carmichael MW. Characteristics of blood–brain barrier to gamma-aminobutyric acid in neonatal cat. Science 1960; 131:410. 86. Cornford EM, Braun LD, Oldendorp WH, et al. Comparison of lipid-mediated blood–brain barrier penetrability in neonates and adults. Am J Physiol 1982; 243:C161. 87. Laas R, Helmke K. Regional cerebral blood flow following unilateral blood–brain barrier alteration induced by hyperosmolar perfusion in the albino rat. In: Cervos-Navarro J, Fritschka E, eds. Cerebral circulation and metabolism. New York: Raven Press; 1981:317. 88. Levine RL, Fredericks WR, Rappaport SI. Clearance of bilirubin from rat brain after reversible osmotic opening of the blood–brain barrier. Pediatr Res 1985; 19:1040. 89. Lee K-S, Gartner LM. Management of unconjugated hyperbilirubinemia in the newborn. Semin Liver Dis 1983; 3:52. 90. Cordon-Cardo C, O’Brien JP, Casals D, et al. Multidrugresistance gene (P-glycoprotein) is expressed by endothelial cells at blood–brain barrier sites. Proc Natl Acad Sci USA 1989; 86:695. 91. Rao VV, Dahlheimer JL, Bardgett ME, et al. Choroid plexus epithelial expression of MDR1 P glycoprotein and multidrug resistance-associated protein contribute to the blood–cerebrospinal–fluid drug-permeability barrier. Proc Natl Acad Sci USA 1999; 96:3900. 92. Wijnholds J, deLange EC, Scheffer GL, et al. Multidrug resistance protein 1 protects the choroid plexus epithelium and contributes to the blood–cerebrospinal fluid barrier. J Clin Invest 2000; 105:279. 93. Zhang Y, Han H, Elmquist WF, et al. Expression of various multidrug resistance-associated protein (MRP) homologues in brain microvessel endothelial cells. Brain Res 2000; 876:148. 94. Mustafa MG, Cowger ML, Kind TE. Effects of bilirubin on mitochondrial reactions. J Biol Chem 1969; 244:6403. 95. Roger C, Koziel V, Vert P, et al. Effects of bilirubin infusion on cerebral glucose utilization in the immature rat. Dev Brain Res 1993; 76:115. 96. Rudman D, Bixler TJ, Del Rio AE. Effect of free fatty acid on binding of drugs by bovine serum albumin, by human serum albumin and by rabbit serum. J Pharmacol Exp Ther 1971; 176:261. 97. Brodersen R, Cashore W, Wennberg RP, et al. Kinetics of bilirubin oxidation with peroxidase, as applied to studies of bilirubin-albumin binding. Scand J Clin Lab Invest 1979; 39:143. 98. Constantopoulos A, Matsaniotis N. Bilirubin inhibition of protein kinase: its prevention by cyclic AMP. Cytobios 1976; 17:17. 99. Morphis I, Constantopoulos A, Matsaniotis N, et al. Bilirubininduced modulation of cerebral protein phosphorylation in neonate rabbits in vivo. Science 1982; 218:156. 100. Sano K, Nakamura H, Tamotsu M. Mode of inhibitory action of bilirubin on protein kinase C. Pediatr Res 1985; 19:587.
1477
Section XII. Inherited and Pediatric Diseases of the Liver
101. Martinez-Maldonado M, Suki WN, Schenker S. Nature of the urinary concentrating defect in the Gunn strain of rat. Am J Physiol 1969; 216:1386. 102. Zuelzer WW, Mudgett RT. Kernicterus. Etiologic study based on an analysis of 55 cases. Pediatrics 1950; 6:452. 103. Odell GB, Bolen JL, Poland RL, et al. Protection from bilirubin nephropathy in jaundiced Gunn rats. Gastroenterology 1974; 66:1218. 104. Engle WD, Avant BS Jr. Neonatal hyperbilirubinemia and renal function. J Pediatr 1982; 100:113. 105. Axelson RA. Spontaneous renal papillary necrosis in the Gunn rat. Pathology 1973; 5:43. 106. Brodersen R. Bilirubin solubility and interaction with albumin and phospholipid. J Biol Chem 1979; 254:2364. 107. Gitzelmann-Cumarasamy N, Kuenzle CC, Wilson KJ. Mapping of the primary bilirubin binding site of human serum albumin. Experientia 1976; 32:768. 108. Jacobsen C. Lysine residue 240 of human serum albumin is involved in high-affinity binding of bilirubin. Biochem J 1978; 171:453. 109. Odell GB, Cohen SN, Gordes EH. Administration of albumin in the management of hyperbilirubinemia by exchange transfusion. Pediatrics 1962; 30:613. 110. Brodersen R. Bilirubin solubility and interaction with albumin and phospholipid. J Biol Chem 1979; 254:2364. 111. Odell GB. The dissociation of bilirubin from albumin and its clinical implications. J Pediatr 1959; 55:268. 112. Schmid R, Diamond I, Hammaker I, et al. Interaction of bilirubin with albumin. Nature 1965; 204:1041. 113. Kapitulnik J, Valaes T, Kaufmann NA, et al. Clinical evaluation of Sephadex gel filtration in estimation of bilirubin binding in serum in neonatal jaundice. Arch Dis Child 1974; 49:886. 114. Brodersen R, Cashore W, Wennberg RP, et al. Kinetics of bilirubin oxidation with peroxidase, as applied to studies of bilirubin-albumin binding. Scand J Clin Lab Invest 1979; 39:143. 115. Athanassiadis S, Chopra DR, Fisher M, et al. An electrophoretic method for detection of unbound bilirubin and reserve bilirubin binding capacity in serum of newborns. J Lab Clin Med 1974; 83:968. 116. Lamolla AA, Eisinger J, Blumberg WE, et al. Fluorometric study of the partition of bilirubin among blood components: basis for rapid microassays of bilirubin and bilirubin binding capacity in whole blood. Anal Biochem 1979; 15:25. 117. Brodersen R. Determination of the vacant amount of high affinity bilirubin binding site on serum albumin. Acta Pharmacol Toxicol 1978; 42:153. 118. Hsia JC, Kwan NH, Er SS, et al. Development of a spin assay for reserve bilirubin loading capacity of human serum. Proc Natl Acad Sci USA 1978; 75:1542. 119. Weiss JS, Gautam JJ, Lauff MW, et al. The clinical importance of a protein-bound fraction of serum bilirubin in patients with hyperbilirubinemia. N Engl J Med 1983; 309:147. 120. Inoue M, Okajima K, Nagase S, et al. Plasma clearance of sulfobromophthalein and its interaction with hepatic binding proteins in normal and analbuminemic rats: is plasma albumin essential for vectorial transport of organic anions in the liver? Proc Natl Acad Sci USA 1983; 80:7654. 121. Inoue M, Hirata E, Morino Y, et al. The role of albumin in the hepatic transport of bilirubin: studies in mutant analbuminemic rats. J Biochem 1985; 97:737. 122. Yamazaki M, Suzuki H, Sugiyama Y, et al. Uptake of organic anions by isolated rat hepatocytes: a classification in terms of ATP-dependency. J Hepatol 1992; 14:41. 123. Kouzuki H, Suzuki H, Ito K, et al. Contribution of sodium taurocholate co-transporting polypeptide to the uptake of its possible substrates into rat hepatocytes. J Pharmacol Exp Ther 1998; 286:1043.
1478
124. Zucker SD, Goessling W. Mechanism of hepatocellular uptake of albumin-bound bilirubin. Biochim Biophys Acta 2000; 1463:197. 125. Cui Y, Konig J, Leier I, et al. Hepatic uptake of bilirubin and its conjugates by the human organic anion-transporting polypeptide 2 (symbol SLC21A6). J Biol Chem 2000; 276: 9626. 126. Suzuki H, Sugiyama Y. Transport of drugs across the hepatic sinusoidal membrane: sinusoidal drug influx and efflux in the liver. Semin Liver Dis 2000; 20:251. 127. Wang P, Kim RB, Roy-Chowdhury J, et al. The human organic anion transport protein SLC21A6 (OATP2) is not sufficient for bilirubin transport. J Biol Chem 2003; 278:20695–20699. 128. Webster CR, Blanch CJ, Phillips J, et al. Cell swelling-induced translocation of rat liver Na+/taurocholate cotransport polypeptide is mediated via the phosphoinositide 3-kinase signaling pathway. J Biol Chem 2000; 275:29754. 129. Dranoff JA, McClure M, Burgstahler AD, et al. Short-term regulation of bile acid uptake by microfilament-dependent translocation of rat ntcp to the plasma membrane. Hepatology 1999; 30:223. 130. Ganguly TC, O’Brien ML, Karpen SJ, et al. Regulation of the rat liver sodium-dependent bile acid cotransporter gene by prolactin. Mediation of transcriptional activation by Stat5. J Clin Invest 1997; 99:2906. 131. Denson LA, Auld KL, Schiek DS, et al. Interleukin-1a suppresses retinoid transactivation of two hepatic transporter genes involved in bile formation. J Biol Chem 2000; 275:8835. 132. Trauner M, Arrese M, Lee H, et al. Endotoxin downregulates rat hepatic ntcp gene expression via decreased activity of critical transcription factors. J Clin Invest 1998; 101:2092. 133. Beigneux AP, Moser AH, Shigenaga JK, et al. The acute phase response is associated with retinoid X receptor repression in rodent liver. J Biol Chem 2000; 275:16390. 134. Jedlitschky G, Leier I, Buchholz U, et al. ATP-dependent transport of bilirubin glucuronides by the multidrug resistance protein MRP1 and its hepatocyte canalicular isoform MRP2. Biochem J 1997; 327:305. 135. Vos TA, Hooiveld GJ, Koning H, et al. Up-regulation of the multidrug resistance genes, mrp1 and mdr1b, and downregulation of the organic anion transporter, mrp2, and the bile salt transporter, spgp, in endotoxemic rat liver. Hepatology 1998; 28:1637. 136. Trauner M, Arrese M, Soroka CJ, et al. The rat canalicular conjugate export pump (mrp2) is down-regulated in intrahepatic and obstructive cholestasis. Gastroenterology 1997; 113:255. 137. Konig J, Rost D, Cui Y, et al. Characterization of the human multidrug resistance protein isoform MRP3 localized to the basolateral hepatocyte membrane. Hepatology 1999; 29:1156. 138. Hirohashi T, Suzuki H, Sugiyama Y. Characterization of the transport properties of cloned rat multidrug resistance-associated protein 3 (MRP3). J Biol Chem 1999; 274:15181. 139. Jansen PLM, Cuypers HT, Peters WHM. Quantitation of bilirubin conjugates with high-performance liquid chromatography in patients with low total serum bilirubin levels. Eur J Clin Invest 1984; 14:295. 140. Jansen PLM, Peters WHM, Jansssens AR. Clinical value of serum bilirubin subfractionation by high-performance liquid chromatography and conventional methods in patients with primary biliary cirrhosis. J Hepatol 1986; 2:485. 141. Levi AJ, Gatmaitan Z, Arias IM. Two hepatic cytoplasmic protein fractions, Y and Z, and their possible role in the hepatic uptake of bilirubin, sulfobromophthalein, and other anions. J Clin Invest 1969; 48:2156. 142. Rowe JD, Nieves E, Listowsky I. Subunit diversity and tissue distribution of human glutathione S-transferases: interpretations based on electrospray ionization-MS and peptide sequencespecific antisera. Biochem J 1997; 325:481.
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
143. Wolkoff AW, Goresky CA, Sellin J, et al. Role of ligandin in transfer of bilirubin from plasma into liver. Am J Physiol 1979; 236:E638. 144. Compernolle F, Van Hees GP, Fevery JJ, et al. Massspectrometric structure elucidation of a dog bile axopigments as the acyl glycosides of glucopyranose and xylopyranose. Biochem J 1971; 125:811. 145. Fevery JB, Van Damme R, Michiels R, et al. Bilirubin conjugates in bile of man and rat in the normal state and in liver disease. J Clin Invest 1972; 51:2482. 146. Spivak W, Carey MC. Reverse-phase HPLC separation, quantification and preparation of bilirubin and its conjugates from native bile. Biochem J 1985; 225:787. 147. Fevery J, Van Hees GP, Leroy P, et al. Excretion in dog bile of glucose and xylose conjugates of bilirubin. Biochem J 1971; 125:803. 148. Roy Chowdhury J, Novikoff PM, Roy Chowdhury N, et al. Distribution of UDP-glucuronosyltransferase in rat tissue. Proc Natl Acad Sci USA 1985; 82:2990. 149. Dutton GJ, Burchell B. Newer aspects of glucuronidation. Prog Drug Metab 1977; 2:1. 150. Zakim D, Eibl H. The influence of charge and the distribution of charge in the polar region of phospholipids on the activity of UDP-glucuronosyltransferase. J Biol Chem 1992; 267:13166. 151. Henderson P. Activation in vitro of rat hepatic UDP glucuronyltransferase by ultrasound. Life Sci 1970; 9:511. 152. Graham AD, Wood GC. Factors affecting the response of microsomal UDP-glucuronyltransferase to membrane perturbants. Biochim Biophys Acta 1973; 311:45. 153. Vessey DA, Zakim D. Regulations of microsomal enzymes by phospholipids. J Biol Chem 1971; 246:4649. 154. Meech R, Mackenzie PI. Structure and function of uridine diphosphate glucuronosyltransferases. Clin Exp Pharmacol Physiol 1997; 24:907–915. 155. Hauser SC, Ziurys JC, Gollan JL. A membrane transporter mediates access of uridine-5¢-diphosphoglucuronic acid from the cytosol into the endoplasmic reticulum of rat hepatocytes. Implications for glucuronidation reactions. Biochim Biophys Acta 1988; 967:149. 156. Battaglia E, Radominska-Pandya A. A functional role for histidyl residues of the UDP-glucuronic acid carrier in rat liver endoplasmic reticulum membranes. Biochemistry 1998; 37:258. 157. Bossuyt X, Blanckaert N. Carrier-mediated transport of uridine diphosphoglucuronic acid across the endoplasmic reticulum membrane is a prerequisite for UDP-glucuronosyltransferase activity in rat liver. Biochem J 1997; 323:645. 158. Heirwegh KPM, Campbell M, Meuwissen JATP. Compartmentation of membrane bound enzymes. Some basic concepts and consequences for kinetic studies. In: Aitio A, ed. Conjugation reactions in drug biotransformation. Amsterdam: Elsevier; 1978:191. 159. Hallinan T. Comparison of compartmented and of conformational phospholipid-constraint models for the intramembranous arrangement of UDP-glucuronyltransferase. In: Aitio A, ed. Conjugation reactions in drug biotransformation. Amsterdam: Elsevier; 1978:257. 160. Wishart GF. Functional heterogeneity of UDP glucuronosyl transferase as indicated by its differential development and inducibility by glucocorticoids. Biochem J 1978; 174:485. 161. Lillienblum W, Walli AK, Bock KW. Differential induction of rat liver microsomal UDP-glucuronosyltransferase activities by various inducing agents. Biochem Pharmacol 1982; 31:907. 162. Roy Chowdhury J, Roy Chowdhury N, Moscioni AD, et al. Differential regulation by triiodothyronine of substrate-specific uridinediphosphoglucuronate glucuronosyl transferases in rat liver. Biochim Biophys Acta 1983; 761:58. 163. Roy Chowdhury J, Roy Chowdhury N, Falany CN, et al. Isolation and characterization of multiple forms of rat liver
164.
165. 166.
167.
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
UDP-glucuronoate glucuronosyltransferase. Biochem J 1986; 233:827. Mackenzie PI, Owens IS, Burchell B, et al. The UDP glucuronosyltransferase gene superfamily: recommended nomenclature update based on evolutionary divergence. Pharmacogenetics 1997; 7:255. Lazard D, Zupko K, Poria Y, et al. Odorant signal termination by olfactory UDP glucuronosyl transferase. Nature 1991; 349:790. Ritter JK, Chen F, Sheen YY, et al. A novel complex locus UGT1 encodes human bilirubin, phenol and other UDPglucuronosyltransferase isozymes with identical carboxy termini. J Biol Chem 1992; 267:3257. Van Es HHG, Bout A, Liu J, et al. Assignment of the human UDP-glucuronosyltransferase gene UGT1A1 to chromosome 2, region 2q37. Cytogenet Cell Genet 1993; 63:114. Drake RR, Igari Y, Lester R, et al. Application of 5-azido-UDPglucose and 5-azido-UDP-glucuronic acid photoaffinity probes for the determination of the active site orientation of microsomal UDP-glucosyltransferases and UDPglucuronosyltransferases. J Biol Chem 1992; 267:11360. Mackenzie PI. Expression of chimeric cDNAs in cell culture defines a region of UDP-glucuronosyltransferase involved in substrate selection. J Biol Chem 1990; 265:3432. Bosma PJ, Seppen J, Goldhoorn B, et al. Bilirubin UDPglucuronosyltransferase 1 is the only relevant bilirubin glucuronidating isoform in man. J Biol Chem 1994; 269:17960. Roy Chowdhury N, Saber MA, Lahiri P, et al. Expression of specific UDP-glucuronosyltransferase isoforms in carcinogeninduced preneoplastic rat liver nodules. Hepatology 1991; 13:31. Meijer DKF. Transport and metabolism in the hepato-biliary system. In: Schulz SG, Forte JG, Rauner BB, eds. Handbook of physiology. The gastrointestinal system, vol. III. New York: Oxford University Press; 1989:717. Keppler D, Konig J. Hepatic canalicular membrane 5: expression and localization of the conjugate export pump encoded by the MRP2 (cMRP/cMOAT) gene in liver. FASEB J 1997; 11:509. Van Aubel RA, Peters JG, Masereeuw R, et al. Multidrug resistance protein Mrp2 mediates ATP-dependent transport of classic renal organic anion p-aminohippurate. Am J Physiol Renal Physiol 2000; 279:F713. Van Aubel RA, Hartog A, Bindels RJ, et al. Expression and immunolocalization of multidrug resistance protein 2 in rabbit small intestine. Eur J Pharmacol 2000; 400:195. Kamisako T, Leier I, Cui Y, et al. Transport of monoglucuronosyl and bisglucuronosyl bilirubin by recombinant human and rat multidrug resistance protein 2. Hepatology 1999; 30:485. Paulusma CC, Kool M, Bosma PJ, et al. A mutation in the human canalicular multispecific organic anion transporter gene causes the Dubin–Johnson syndrome. Hepatology 1997; 25:1539. Kartenbeck J, Leuschner U, Mayer R, et al. Absence of the canalicular isoform of the MRP gene-encoded conjugate export pump from the hepatocytes in Dubin–Johnson syndrome. Hepatology 1996; 23:1061. Paulusma CC, Bosma PJ, Zaman GJ, et al. Congenital jaundice in rats with a mutation in a multidrug resistance-associated protein gene. Science 1996; 271:1126. Vos TA, Ros JE, Havinga R, et al. Regulation of hepatic transport systems involved in bile secretion during liver regeneration in rats. Hepatology 1999; 29:1833. Hirohashi T, Suzuki H, Ito K, et al. Hepatic expression of multidrug resistance-associated protein-like proteins maintained in Eisai hyperbilirubinemic rats. Mol Pharmacol 1998; 53:1068. Ogawa K, Suzuki H, Hirohashi T, et al. Characterization of inducible nature of MRP3 in rat liver. Am J Physiol Gastrointest Liver Physiol 2000; 278:G438. Konig J, Rost D, Cui Y, et al. Characterization of the human multidrug resistance protein isoform MRP3 localized to the basolateral hepatocyte membrane. Hepatology 1999; 29:1156.
1479
Section XII. Inherited and Pediatric Diseases of the Liver
184. Keppler D, Konig J. Hepatic secretion of conjugated drugs and endogenous substances. Semin Liver Dis 2000; 20:265. 185. Stockel B, Konig J, Nies AT, et al. Characterization of the 5’flanking region of the human multidrug resistance protein 2 (MRP2) gene and its regulation in comparison with the multidrug resistance protein 3 (MRP3) gene. Eur J Biochem 2000; 267:1347. 186. Kauffmann HM, Keppler D, Kartenbeck J, et al. Induction of cMrp/cMoat gene expression by cisplatin, 2acetylaminofluorene, or cycloheximide in rat hepatocytes. Hepatology 1997; 26:980. 187. Kauffman HM, Pfannschmidts S, Zoller H, et al. Influence of redox-active compounds and PXR-activators on human MRP1 and MRP2 gene expression. Toxicology 2002; 171:137. 188. Tanaka T, Uchiumi T, Hinoshita E, et al. The human multidrug resistance protein 2 gene: functional characterization of the 5¢flanking region and expression in hepatic cells. Hepatology 1999; 30:1507. 189. Kauffmann HM, Schrenk D. Sequence analysis and functional characterization of the 5¢-flanking region of the rat multidrug resistance protein 2 (mrp2) gene. Biochem Biophys Res Commun 1998; 245:325. 190. Stockel B, Konig J, Nies AT, et al. Characterization of the 5¢flanking region of the human multidrug resistance protein 2 (MRP2) gene and its regulation in comparison with the multi drug resistance protein 3 (MRP3) gene. Eur J Biochem 2000; 267:1347. 191. Gerloff T, Stieger B, Hagenbuch B, et al. The sister of Pglycoprotein represents the canalicular bile salt export pump of mammalian liver. J Biol Chem 1998; 273:10046. 192. Strautnieks SS, Bull LN, Knisely AS, et al. A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis. Nat Genet 1998; 20:233. 193. Lecureur V, Sun D, Hargrove P, et al. Cloning and expression of murine sister of P-glycoprotein reveals a more discriminating transporter than MDR1/P-glycoprotein. Mol Pharmacol 2000; 57:24. 194. Green RM, Hoda F, Ward KL. Molecular cloning and characterization of the murine bile salt export pump. Gene 2000; 241:117. 195. Smit JJM, Schinkel AH, Elferink RPJO, et al. Homozygous disruption of the murine mdr2 P-glycoprotein gene leads to a complete absence of phospholipid from bile and to liver disease. Cell 1993; 75:451. 196. de Vree JM, Jacquemin E, Sturm E, et al. Mutations in the MDR3 gene cause progressive familial intrahepatic cholestasis. Proc Natl Acad Sci USA 1998; 95:282. 197. Crawford AR, Smith AJ, Hatch VC, et al. Hepatic secretion of phospholipid vesicles in the mouse critically depends on mdr2 or MDR3 P-glycoprotein expression. Visualization by electron microscopy. J Clin Invest 1997; 100:2562. 198. Huang W, Shang J, Chua SS, et al. Proc Natl Acad Sci USA 2003; 100:4156–4161. 199. Roy-Chowdhury J, Locker J, Roy-Chowdhury N. Nuclear receptors orchestrate detoxification pathways. Dev Cell 2003; 4:607–608. 200. Lester R, Schmid R. Intestinal absorption of bile pigments. II. Bilirubin absorption in man. N Engl J Med 1963; 269:178. 201. Alonso EM, Whitington PF, Whitington SH, et al. Enterohepatic circulation of nonconjugated bilirubin in rats fed with human milk. J Pediatr 1991; 118:425. 202. Ostrow JD. Absorption of bile pigments by the gall bladder. J Clin Invest 1967; 46:2035. 203. Brodersen R, Herman LS. Intestinal reabsorption of unconjugated bilirubin: a possible contributing factor in neonatal jaundice. Lancet 1963; 1:1242. 204. Stoll MS, Lim CD, Gray CH. Chemical variants of the uroblins. In: Berk PD, Berlin NI, eds. Bile pigments, chemistry and
1480
205.
206.
207.
208. 209.
210.
211.
212.
213.
214.
215.
216.
217. 218.
219.
220. 221.
222.
223.
224.
physiology. Washington, DC: US Government Printing Office; 1977:483. Fulop M, Sanson J, Brazeau P. Dialyzability, protein binding, and renal excretion of plasma conjugated bilirubin. J Clin Invest 1965; 44:666. Gollan JL, Dallinger KJC, Billing BH. Excretion of conjugated bilirubin in the isolated perfused rat kidney. Clin Sci Mol Med 1978; 54:381. Tanaka Y, Kobayashi Y, Gabazza EC, et al. Increased renal expression of bilivubin glucronide transporters in a rat model of obstructive jaundice. Am J Physiol Gastrointest Liver Physiol 2002; 282:G656. Cameron JL, Pulaski EJ, Abel T, et al. Metabolism and disposition of C14-bilirubin in experimental obstructive jaundice. Ann Surg 1966; 163:330. Cameron JL, Filler RM, Iber EL, et al. Metabolism and excretion of C14 labeled bilirubin in children with biliary atresia. N Engl J Med 1966; 274:231. McDonagh F, Palma LA, Lighner DA. Phototherapy for neonatal jaundice. Stereospecific and regiospecific photoisomerization of bilirubin bound to human serum albumin and NMR characterization of intramolecularly cyclized photoproducts. J Am Chem Soc 1982; 104:6867. Itoh S, Onishi S. Kinetic study of the photochemical changes of (ZZ)-bilirubin IX bound to human serum albumin. Demonstration of (EZ)-bilirubin IX as an intermediate in photochemical changes from (ZZ)-bilirubin IX to (EZ)cyclobilirubin IX. Biochem J 1985; 226:251. Berry CS, Zarembo JE, Ostrow JD. Evidence for conversion of bilirubin to dihydroxyl derivates in the Gunn rat. Biochem Biophys Res Commun 1972; 49:1366. Kapitulnik J, Ostrow JD. Stimulation of bilirubin catabolism in jaundiced Gunn rats by an inducer of microsomal mixed function mono-oxygenases. Proc Natl Acad Sci USA 1978; 75:682. Kapitulnik J, Bircher J, Hedorn HB. Chlorpromazine reduces plasma bilirubin levels in Crigler–Najjar syndrome type I (CNSI). Hepatology 1989; 10:A708. Cohen AN, Kapitulnik J, Ostrow JD, et al. Effects of phenobarbital on bilirubin metabolism and its response to phototherapy in the jaundiced Gunn rat. Hepatology 1985; 5:310. Tiribelli C, Ostrow JD. New concepts in bilirubin chemistry, transport and metabolism. Report of the International Bilirubin Workshop, Trieste, Italy. Hepatology 1990; 11:303. Doumas BT, Wu TW. The measurement of bilirubin fractions in serum. Crit Rev Clin Lab Sci 1991; 28:415. Hutchinson DW, Johnson B, Knell AJ. The reaction between bilirubin and aromatic diazo compounds. Biochem J 1972; 127:907. Van den Bergh AAH, Muller P. Ueber eine direkte und eine indirekte Diazoreaktion auf Bilirubin. Biochem Z 1916; 77:90. Talafant E. Properties and composition of bile pigment giving direct diazo reaction. Nature (Lond) 1956; 178:312. Lauff JJ, Kasper ME, Ambros RT. Quantitative liquid chromatographic estimation of bilirubin species in pathological serum. Clin Chem 1983; 29:800. Poon R, Hinberg IH. Indican interference with six commercial procedures for measuring total bilirubin. Clin Chem 1985; 31:92. Heirwegh KPM, Fevery JB, Meuwissen JATP, et al. Recent advances in the separation and analysis of diazo-positive bile pigments. Methods Biochem Anal 1974; 22:205. Blanckaert N, Kabra PM, Farina FA, et al. Measurement of bilirubin and its mono- and diconjugates in human serum by alkaline methanolysis and high performance liquid chromatography. J Lab Clin Med 1980; 96:198.
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
225. Onishi S, Itho S, Kawade N, et al. An accurate and sensitive analysis by high pressure liquid chromatography of conjugated and unconjugated bilirubin IX and in various biological fluids. Biochem J 1980; 185:281. 226. Roy Chowdhury J, Roy Chowdhury N, Wu G, et al. Bilirubin monoglucuronide and diglucuronide formation by human liver in vitro: assay by high pressure liquid chromatography. Hepatology 1981; 1:622. 227. Dappen GM, Sundberg MW, Wu TW, et al. A diazo-based dry film for determination of total bilirubin in serum. Clin Chem 1983; 29:37. 228. Kubasik NP, Mayer TK, Baskar AG, et al. The measurement of fractionated bilirubin by Ektachem film slides. Method validation and comparison of conjugated bilirubin measurements with direct bilirubin in obstructive and hepatocellular jaundice. Am J Clin Pathol 1985; 84:518. 229. Maisels MJ, Kring E. Transcutaneous bilirubinometry decreases the need for serum billirubin measurement and saves money. Pediatrics 1997; 99:599. 230. Tayba R, Gribetz D, Gribetz I, et al. Non-invasive estimation of serum bilirubin. Pediatrics 1998; 102:28. 231. Meisel S, Shamiss A, Thaler M, et al. Pleural fluid to serum bilirubin concentration ratio for the separation of transudates from exudates. Chest 1990; 98:141. 232. Gartner LM, Lee K, Vaisman S, et al. Development of bilirubin transport and metabolism in the newborn Rhesus monkey. J Pediatr 1977; 90:513. 233. Hardy JB, Peeples MO. Serum bilirubin levels in newborn infants. Distributions and associations with neurological abnormalities during the first year of life. Johns Hopkins Med J 1971; 128:265. 234. Maisels MJ, Pathak A, Nelson NM, et al. Endogenous production of carbon monoxide in normal and erythroblastic newborn infants. J Clin Invest 1971; 50:1. 235. Vest M, Strebel L, Hauensiein D. The extent of “shunt” bilirubin and erythrocyte survival in the newborn infant measured by the administration of (15N) glycine. Biochem J 1965; 95:11c. 236. Clarke CA, Donohoe WTA, Finn R, et al. Combined study: prevention of Rh hemolytic disease: final results of the “high risk” clinical trial. A combined study from centers in England and Baltimore. Brit Med J 1971; 2:607. 237. Haberman S, Kraft EJ, Leucke PE, et al. ABO isoimmunization: the use of the specific Coombs and best elution tests in the detection of hemolytic disease. J Pediatr 1960; 56:471. 238. Levi AJ, Gatmaitan Z, Arias IM. Deficiency of hepatic organic anion-binding protein, impaired organic anion uptake by liver and “physiologic” jaundice in newborn monkeys. N Engl J Med 1970; 283:1136. 239. Odell GB. “Physiologic” hyperbilirubinemia in the neonatal period. N Engl J Med 1967; 277:193. 240. Kawade N, Onishi S. The prenatal and postnatal development of UDP-glucuronosyltransferase activity towards bilirubin and the effect of premature birth on this activity in human liver. Biochem J 1981; 196:257. 241. Gollan JL, Knapp AB. Bilirubin metabolism and congenital jaundice. Hosp Pract 1985; 15:83. 242. Arthur LJH, Bevan BR, Holton JB. Neonatal hyperbilirubinemia and breast feeding. Dev Med Child Neurol 1966; 8:279. 243. Maisels MJ, Newman TB. Kernicterus in otherwise healthy breast-fed term newborns. Pediatrics 1995; 96:730. 244. Arias IM, Gartner LM, Seifter S, et al. Prolonged neonatal unconjugated hyperbilirubinemia associated with breast feeding and a steroid, pregnana-3(alpha), 20(beta)-diol, in maternal milk that inhibits glucuronide formation in vitro. J Clin Invest 1964; 43:2037. 245. Hargreaves T, Piper RF. Breast milk jaundice: Effect of inhibitory breast milk and 3(alpha), 20(beta)-diol, in maternal
246.
247.
248. 249.
250. 251. 252.
253. 254. 255.
256.
257.
258.
259. 260.
261.
262.
263.
264.
265.
milk that inhibits glucuronide formation in vitro. J Clin Invest 1964; 43:2037. Foliot A, Ploussard JP, Housett E, et al. Breast milk jaundice: in vitro inhibition of rat liver bilirubin-uridine diphosphate glucuronosyl transferase activity and Z proteinbromosulfophthalein binding by human breast milk. Pediatr Res 1976; 10:594. Lucey JF, Driscol JJ. Physiological jaundice re-examined. In: Sass-Kortsak A, ed. Kernicterus. Toronto: University of Toronto Press; 1961:29. Arias IM, Wolfson S, Lucey JF, et al. Transient familial neonatal hyperbilirubinemia. J Clin Invest 1965; 44:1442. Hsia DY-T, Patterson P, Allen FH, et al. Prolonged obstructive jaundice in infancy: general survey of 156 cases. Pediatrics 1952; 10:243. Poland RL, Odell GB. Physiologic jaundice: the enterohepatic circulation of bilirubin. N Engl J Med 1971; 284:1. Schalm L, Weber AP. Jaundice with conjugated bilirubin in hyperhaemolysis. Acta Med Scand 1964; 176:549. Robinson S, Vanier T, Desforges JF, et al. Jaundice in thalassemia minor: a consequence of “ineffective erythropoiesis.” N Engl J Med 1962; 267:512. Verwilghen R, Lewis S, Dacie J, et al. Hempas: congenital dyserythropoietic anaemia (type II). Q J Med 1973; 42:257. Crigler JF, Najjar VA. Congenital familial non-hemolytic jaundice with kernicterus. Pediatrics 1952; 10:169. Childs B, Sidbury JB, Migeon CJ. Glucuronic acid conjugation by patients with familial non-hemolytic jaundice and their relatives. Pediatrics 1959; 23:903. Blaschke TF, Berk PD, Scharschmidt BF, et al. Crigler–Najjar syndrome: an unusual course with development of neurologic damage at age eighteen. Pediatr Res 1974; 8:573. Arias IM, Gartner LM, Cohen M, et al. Chronic nonhemolytic unconjugated hyperbilirubinemia with glucuronosyl transferase deficiency: clinical, biochemical, pharmacologic, and genetic evidence for heterogeneity. Am J Med 1969; 47:395. Wolkoff AW, Roy Chowdhury J, Gartner LA, et al. Crigler–Najjar syndrome (type I) in an adult male. Gastroenterology 1979; 76:3380. Schmid R, Hammaker L. Metabolism and disposition of C14bilirubin in congenital nonhemolytic jaundice. J Clin Invest 1963; 42:1720. Huang PWH, Rozsilsky B, Gerrard JW, et al. Crigler–Najjar syndrome in four of five siblings with post-mortem findings in one. Arch Pathol 1970; 90:536. Van Es HHG, Goldhoorn B, Paul-Abrahamse M, et al. Immunochemical characterization of UDPglucuronosyltransferase in four patients with the Crigler–Najjar type I syndrome. J Clin Invest 1990; 85:1199. Bosma PJ, Roy Chowdhury N, Goldhoorn BG, et al. Sequence of exons and the flanking regions of human bilirubin-UDPglucuronosyltransferase gene complex and identification of a genetic mutation in a patient with Crigler–Najjar syndrome, type I. Hepatology 1992; 15:941. Bosma PJ, Roy Chowdhury J, Huang TJ, et al. Mechanism of inherited deficiencies of multiple UDP-glucuronosyltransferase isoforms in two patients with Crigler–Najjar syndrome, type I. FASEB J 1992; 6:2859. Ritter JK, Yeatman MT, Ferriera P, et al. Identification of a genetic alteration in the code for bilirubin UDPglucuronosyltransferase in the UGT1 gene complex of a Crigler–Najjar syndrome, type I. J Clin Invest 1992; 90:150. Kadakol A, Ghosh SS, Sappal BS, et al. Genetic lesions of bilirubin uridinediphosphoglucuronate glucuronosyltransferase causing Crigler–Najjar and Gilbert’s syndromes: correlation of genotype to phenotype. Human Mutations 2000; 16:297.
1481
Section XII. Inherited and Pediatric Diseases of the Liver
266. Gantla S, Bakker CTM, Deocharan B, et al. Splice site mutations: a novel genetic mechanism of Crigler–Najjar syndrome type 1. Am J Hum Gen 1998; 62:585. 267. Petit FM, Gajdos V, Parisot F, et al. Parental isodisomy for chromosome 2 as the cause of Crigler–Najjar syndrome type I syndrome. Eur J Hum Gen (advance online publication) December 2004. 268. Gunn CH. Hereditary acholuric jaundice in a new mutant strain of rats. J Hered 1938; 29:137. 269. Carbone JV, Grodsky GM. Constitutional nonhemolytic hyperbilirubinemia in the rat: defect of bilirubin conjugation. Proc Soc Exp Biol Med 1957; 94:461. 270. Roy Chowdhury J, Van Es HHS, Roy Chowdhury N. Gunn rat: an animal model for deficiency of bilirubin conjugation. In: Berk PD, Tavoloni N, eds. Pathophysiology of bile secretion. New York: Raven Press; 1992:713. 271. Blanc WA, Johnson L. Studies on kernicterus. J Neuropathol Exp Neurol 1959; 18:165. 272. Shupeck M, Wolkoff AW, Scharschmidt BF, et al. Studies of the kinetics of purified conjugated bilirubin-3H in the rat. Am J Gastroenterol 1978; 70:259. 273. Flock EV, Bollman JL, Owen CA, et al. Conjugation of thyroid hormones and analogues by the Gunn rat. Endocrinology 1965; 77:303. 274. Roy Chowdhury N, Gross F, Moscioni AD, et al. Isolation and purification of multiple normal and functionally defective forms of UDP-glucuronosyltransferase from livers of inbred Gunn rats. J Clin Invest 1987; 79:327. 275. ElAwady M, Roy Chowdhury J, Kesari K, et al. Mechanism of the lack of induction of UDP-glucuronosyltransferase activity by 3-methylcholanthrene in Gunn rats. J Biol Chem 1990; 265:10752. 276. Roy Chowdhury J, Huang TJ, Kesari K, et al. Molecular basis for the lack of bilirubin-specific and 3-methylcholanthrene-inducible UDP-glucuronosyltransferase activities in Gunn rats: the two isoforms are encoded by distinct mRNA species that share an identical single base deletion. J Biol Chem 1991; 266:18294. 277. Gorodischer R, Levy G, Krasner, et al. Congenital nonobstructive, nonhemolytic jaundice: effect of phototherapy. N Engl J Med 1970; 282:375. 278. Plotz PH, Berk PD, Sscharschmidt BF, et al. Removing substances from blood by affinity chromatography. I. Removing bilirubin and other albumin-bound substances from plasma and blood with albumin-conjugated agarose beads. J Clin Invest 1974; 53:778. 279. Scharschmidt BF, Martin JF, Shapiro LJ, et al. Hemoperfusion through albumin-conjugated agarose gel for the treatment of neonatal jaundice in Rhesus monkeys. J Lab Clin Med 1977; 89:101. 280. Schauer R, Stangl M, Lang T, et al. Treatment of Crigler–Najjar type 1 disease: relevance of early liver transplantation. J Pediatr Surg 2003; 38:1227–1231. 281. Drummond GS, Kappas A. Chemoprevention of neonatal jaundice, potency of tin protoporphyrin in an animal model. Science 1982; 217:6466. 282. Cornelius CS, Rodgers PA. Prevention of neonatal hyperbilirubinemia in rhesus monkey by tinprotoporphyrin. Pediatr Res 1984; 18:728. 283. Galbraith RA, Drummond GS, Kappas A. Suppression of bilirubin production in the Crigler–Najjar, type I syndrome: studies with heme oxygenase inhibitor tinmesoporphyrin. Pediatrics 1992; 89:175. 284. Lavin A, Sung C, Klibanov AM, et al. Enzymatic removal of bilirubin from blood: a potential treatment for neonatal jaundice. Science 1985; 230:543. 285. Sugi K, Inoue M, Morino Y. Degradation of plasma bilirubin by a bilirubin oxidase derivative which has a relatively long half-life in the circulation. Biochim Biophys Acta 1989; 991:405.
1482
286. Wattenberg LW, Loub WD. Inhibition of polycyclic aromatic hydrocarbon-induced neoplasia by naturally occurring indoles. Cancer Res 1978; 38:1410. 287. Kapitulnik J. The role of cytochrome P-450 in the alternate pathways of bilirubin metabolism in congenitally jaundiced Gunn rats and infants with the Crigler–Najjar syndrome, type I. Trieste: International Bilirubin Workshop; 1992:53. 288. Roy Chowdhury J, Fischberg EB, Daniller A, et al. Hepatic conversion of bilirubin monoglucuronide to bilirubin diglucuronide in uridinediphosphate glucuronosyltransferase deficient man and rat by bilirubin glucuronoside glucuronosyltransferase. J Clin Invest 1978; 21:191. 289. Demetriou AA, Levenson SM, Whiting J, et al. Replacement of hepatic functions in rats by transplantation of microcarrierattached hepatocytes. Science 1986; 233:1190. 290. Gupta S, Aragona E, Vemuru RP, et al. Permanent engraftment and function of hepatocytes delivered to the liver: implications for gene therapy and liver repopulation. Hepatology 1991; 14:144. 291. Fox IJ, Roy Chowdhury J, Kaufman SS, et al. Treatment of Crigler–Najjar syndrome type I with hepatocyte transplantation. N Engl J Med 1998; 333:1422. 292. Fox IJ, Roy-Chowdhury J. Hepatocyte transplantation. J Hepatol 2004; 40: 878–886. 293. Roy Chowdhury J, Strom SC, Roy Chowdhury N, et al. Hepatocyte transplantation in humans: gene therapy and more. Pediatrics 1998; 102:647. 294. Guha C, Parashar B, Deb NJ, et al. Long-term normalization of serum bilirubin levels by massive repopulation of Gunn rat liver by normal hepatocytes, transplanted after preparative hepatic irradiation and partial hepatectomy. Hepatology 2002; 36: 354–362. 295. Takahashi M, Deb NJ, Kawashita Y, et al. A novel strategy for in vivo expansion of transplanted hepatocytes using preparative hepatic irradiation and FasL-induced hepatocellular apoptosis. Gene Ther 2003; 10:304–313. 296. Roy Chowdhury J, Grossman M, Gupta S, et al. Long-term improvement of hypercholesterolemia after ex vivo gene therapy in LDL-receptor deficient rabbits. Science 1991; 54:1802. 297. Kren B, Roy Chowdhury N, Roy Chowdhury J, et al. Gene therapy as an alternative for liver transplantation. Liver Transplant 2002; 8:1089. 298. Thummala NR, Ghosh SS, Lee SW, et al. A non-immunogenic adenoviral vector, coexpressing CTLA4Ig and bilirubinuridinediphosphoglucuronateglucuronosyltransferase permits long-term, repeatable transgene expression in the Gunn rat model of Crigler–Najjar syndrome. Gene Ther 2002; 9:981–990. 299. Sauter BV, Parashar B, Roy Chowdhury N, et al. Gene transfer to the liver using a replication-deficient recombinant SV40 vector results in long-term amelioration of jaundice in Gunn rats. Gastroenterology 2000; 119:1348. 300. Strayer DS, Branco F, Zern MA, et al. Durability of transgene expression and vector integration: recombinant SV40-derived gene therapy vectors. Mol Ther 2002; 6:227–237. 301. Kren B, Parashar B, Bandopadhyay P, et al. Correction of the UDP-glucuronosyltransferase gene defect in the Gunn rat model of Crigler–Najjar syndrome type I. Proc Natl Acad Sci USA 1999; 96:10349–10353. 302. Ghosh SS, Takahashi M, Thummala NR, et al. Liver-directed gene therapy: promises, problems and prospects at the turn of the century. J. Hepatol 2000; 32 (suppl 1):238–252. 303. Arias IM. Chronic unconjugated hyperbilirubinemia without overt signs of hemolysis in adolescents and adults. J Clin Invest 1962; 41:2233. 304. Gollan JL, Huang SM, Billing B, et al. Prolonged survival in three brothers with severe type II Crigler–Najjar syndrome. Ultrastructural and metabolic studies. Gastroenterology 1975; 68:1543.
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
305. Gordon ER, Shaffer EA, Sass-Kortsak A. Bilirubin secretion and conjugation in the Crigler–Najjar syndrome type II. Gastroenterology 1976; 70:761. 306. Fevery J, Blanckaert N, Heirwegh KPM, et al. Unconjugated bilirubin and an increased proportion of bilirubin monoconjugates in the bile of patients with Gilbert’s syndrome and Crigler–Najjar syndrome. J Clin Invest 1977; 60:970. 307. Bosma PJ, Golhoorn B, Oude Elferink RP, et al. A mutation in bilirubin uridine 5¢-diphosphate glucuronosyltransferase isoforms 1 causing Crigler–Najjar syndrome type II. Gastroenterology 1993; 105:216. 308. Hunter JO, Thompson RPH, Dunn PM, et al. Inheritance of type II Crigler–Najjar hyperbilirubinemia. Gut 1973; 14:46. 309. Kadakol A, Sappal BS, Ghosh SS, et al. Interaction of coding region mutations and the gilbert-type promoter abnormality of the UGT1A1 gene causes moderate degrees of unconjugated hyperbilirubinemia and may lead to neonatal kernicterus. J Med Genet 2001; 38:244. 310. Gilbert A, Lereboullet P. La cholamae simple familiale. Semin Med 1901; 21:241. 311. Foulk WT, Butt HR, Owen CA, et al. Constitutional hepatic dysfunction (Gilbert’s disease): its natural history and related syndrome. Medicine (Baltimore) 1959; 38:25. 312. Arias IM, London IM. Bilirubin glucuronide formation in vitro: demonstration of a defect in Gilbert’s disease. Science 1957; 126:563. 313. Powell LW, Hemingway E, Billing BH, et al. Idiopathic unconjugated hyperbilirubinemia (Gilbert’s syndrome): a study of 42 families. N Engl J Med 1967; 277:1108. 314. Muraca M, Fevery J. Influence of sex and sex steroids on bilirubin-uridinediphosphate glucuronosyltransferase activity of rat liver. Gastroenterology 1984; 87:308. 315. Bosma PJ, Roy Chowdhury J, Bakker C, et al. The genetic basis of the reduced expression of bilirubin UDPglucuronosyltransferase 1 in Gilbert’s syndrome. N Engl J Med 1995; 333:1171. 316. Monaghan G, Ryan M, Seddon R, et al. Genetic variation in bilirubin UDP-glucuronosyltransferase gene promoter and Gilbert’s syndrome. Lancet 1996; 347:578. 317. Raijmakers MT, Jansen PL, Steegers EA, et al. Association of human liver bilirubin UDP-glucuronyltransferase activity with a polymorphism in the promoter region of the UGT1A1. J Hepatol 2000; 33:348. 318. Beutler E, Gelbart T, Demina A. Racial variability in the UDPglucuronosyltransferase 1 (UGT1A1) promoter: a balanced polymorphism for regulation of bilirubin metabolism? Proc Natl Acad Sci USA 1998; 95:8170. 319. Maruo Y, Sato H, Yamano T, et al. Gilbert syndrome caused by a homozygous missense mutation (Tyr486Asp) of bilirubin UDPglucuronosyltransferase gene. J Pediatr 1998; 132:1045. 320. Koiwai O, Nishizawa M, Hasada K, et al. Gilbert’s syndrome is caused by a heterozygous missense mutation in the gene for bilirubin UDP-glucuronosyltransferase. Hum Mol Genet 1995; 4:1183. 321. Aono S, Adachi Y, Uyama E, et al. Analysis of genes for bilirubin UDP-glucuronosyltransferase in Gilbert’s syndrome. Lancet 1995; 345:958. 322. Akaba K, Kimura T, Sasaki A, et al. Neonatal hyperbilirubinemia and a common mutation of the bilirubin uridine diphosphateglucuronosyltransferase gene in Japanese. J Hum Genet 1999; 44:22. 323. Bancroft JD, Kreamer B, Gourley GR. Gilbert syndrome accelerates development of neonatal jaundice. J Pediatr 1998; 132:656. 324. Roy-Chowdhury N, Deocharan B, Bejjanki HR, et al. Presence of the genetic marker for Gilbert syndrome is associated with increased level and duration of neonatal jaundice. Acta Paediatr 2002; 91:100–102.
325. Kaplan M, Renbaum P, Levy-Lahad E, et al. Gilbert syndrome and glucose-6-phosphate dehydrogenase deficiency: a dosedependent genetic interaction crucial to neonatal hyperbilirubinemia. Proc Natl Acad Sci USA 1997; 94:12128. 326. Jansen PL, Bosma PJ, Bakker C, et al. Persistent unconjugated hyperbilirubinemia after liver transplantation due to an abnormal bilirubin UDP-glucuronosyltransferase gene promoter sequence in the donor. J Hepatol 1997; 27:1. 327. Vajro P, DeVincenzo A, Lucariello S, et al. Unusual early presentation of Gilbert syndrome in pediatric recipients of liver. J Pediatr Gastroenterol Nutr 2000; 31:238. 328. del Giudice EM, Perrotta S, Nobili B, et al. Coinheritance of Gilbert syndrome increases the risk for developing gallstones in patients with hereditary spherocytosis. Blood 1999; 94:2259. 329. de Morais SM, Uetrecht JP, Wells PG. Decreased glucuronidation and increased bioactivation of acetaminophen in Gilbert’s syndrome. Gastroenterology 1992; 102:577. 330. Ullrich D, Sieg A, Blume R, et al. Normal pathways for glucuronidation, sulphation and oxidation of paracetamol in Gilbert’s syndrome. Eur J Clin Invest 1987; 17:237. 331. Wells PG, Mackenzie PI, Chowdhury JR, et al. Glucuronidation and the UDP-glucuronosyltransferases in health and disease. Drug Metab Dispos 2004; 32:281–290. 332. Iyer L, King CD, Whitington PF, et al. Genetic predisposition to the metabolism of irinotecan (CPT-11). Role of uridine diphosphate glucuronosyltransferase isoform 1A1 in the glucuronidation of its active metabolite (SN-38) in human liver microsomes. J Clin Invest 1998; 101:847. 333. Billing BH, Williams R, Richards TG. Defects in hepatic transport of bilirubin in congenital hyperbilirubinaemia. An analysis of plasma bilirubin disappearance curves. Clin Sci 1964; 27:245. 334. Goresky CA, Gordon ER, Shaffer EA, et al. Definition of a conjugation dysfunction in Gilbert’s syndrome: studies of the handling of bilirubin loads and of the pattern of bilirubin conjugates secreted in bile. Clin Sci Mol Med 1978; 1:63. 335. Cobelli C. Modeling, identification and parameter estimation of bilirubin kinetics in normal, hemolytic and Gilbert’s states. Comput Biomed Res 1975; 8:522. 336. Berk PD, Blaschke TF, Waggoner JG. Defective BSP clearance in patients with constitutional hepatic dysfunction (Gilbert’s syndrome). Gastroenterology 1972; 63:472. 337. Cobelli C, Ruggeri A, Toffolo G, et al. BSP vs bilirubin kinetics in Gilbert’s syndrome. In: Okolicsanyl L, ed. Familial hyperbilirubinemia. New York: Wiley; 1981:121. 338. Gollan J, Bateman C, Billing BH. Effect of dietary composition on the unconjugated hyperbilirubinemia of Gilbert’s syndrome. Gut 1976; 5:335. 339. Felsher BR, Carpio NM. Caloric intake and unconjugated hyperbilirubinemia. Gastroenterology 1975; 69:42. 340. Bensinger TA, Maisels MJ, Marlson DE, et al. Effect of low caloric diet on endogenous carbon monoxide production: normal adults and Gilbert’s syndrome. Proc Soc Exp Biol Med 1973; 144:417. 341. Gollan JL, Hatt KJ, Billing BH. The influence of diet on unconjugated hyperbilirubinemia in the Gunn rat. Clin Sci Mol Med 1975; 49:229. 342. Lundh B, Johansson M, Mercke C, Cavallin-Stahl E. Enhancement of heme catabolism by caloric restriction in man. Scand J Clin Lab Invest 1972; 30:421. 343. Kotal P, Vitek L, Fevery J. Fasting-related hyperbilirubinemia in rats: the effect of decreased intestinal motility. Gastroenterology 1996; 111:217–223. 344. Kirschebaum G, Shames DM, Schmid R. An expanded model of bilirubin kinetics: effect of feeding, fasting and phenobarbital in Gilbert’s syndrome. J Pharmacokinet Biopharm 1976; 2:115.
1483
Section XII. Inherited and Pediatric Diseases of the Liver
345. Laperche Y, Praux AM, Feldman G, et al. Effect of fasting on organic anion uptake by isolated rat liver cells. Hepatology 1981; 1:617–621. 346. Cowan RE, Thompson RPH, Kaye JP, et al. The association between fasting hyperbilirubinaemia and serum non-esterified fatty acids in man. Clin Sci Mol Med 1977; 53:155. 347. Sorrenito D, Stump DD, Zhou S, et al. The hepatocellular uptake of free fatty acid is selectively preserved during starvation. Gastroenterology 1994; 107:1415–1424. 348. Ishihara T, Kaito M, Takeuchi K, et al. Role of UGT1A1 mutation in fasting hyperbilirubinemia. J Gastroenterol Hepatol 2001; 16:678–682. 349. Felscher BF, Carpio NM, Van Couvering K. Effect of fasting and phenobarbital on hepatic UDP-glucuronic acid formation in the rat. J Lab Clin Med 1979; 93:414. 350. Ohkubo H, Musha H, Okuda K. Studies on nicotinic acid interaction with bilirubin metabolism. Dig Dis Sci 1979; 24:700. 351. Fromke VL, Miller D. Constitutional hepatic dysfunction (CHD: Gilbert’s disease): a review with special reference to a characteristic increase and prolongation of the hyperbilirubinemic response to nicotinic acid. Medicine (Baltimore) 1972; 51:451. 352. Davidson AR, Rojas-Beuno A, Thompson RPH, Williams R. Reduced caloric intake and nicotinic acid provocation tests in diagnosis of Gilbert’s syndrome. Br Med J 1975; 2:480. 353. Portman OW, Roy Chowdhury J, Roy Chowdhury N, et al. A non-human primate model for Gilbert’s syndrome. Hepatology 1984; 4:175. 354. Portman OW, Alexander M, Roy Chowdhury J, et al. Effects of nutrition on hyperbilirubinemia in Bolivian squirrel monkeys. Hepatology 1984; 4:454. 355. Zimmerman HJ. Drug-induced liver disease. Clin Liver Dis 2000; 4:73–96. 356. Berr F, Simon FR, Reichen J. Ethynylestradiol impairs bile salt uptake and Na-K pump function of rat hepatocytes. Am J Physiol 1984; 247:G437. 357. Rosario J, Sutherland E, Zaccaro L, et al. Ethinylestradiol administration selectively alters liver sinusoidal membrane lipid fluidity and protein composition. Biochem 1988; 27:3939. 358. Stieger B, Fattinger K, Madon J, et al. Drug- and estrogeninduced cholestasis through inhibition of the hepatocellular bile salt export pump (bsep) of rat liver. Gastroenterology 2000; 118:422. 359. Bossard R, Stieger B, O’Neill B, et al. Ethinylestradiol treatment induces multiple canalicular membrane transport alterations in rat liver. J Clin Invest 1993; 91:2714–2720. 360. Huang L, Smit JW, Meijer DK, et al. MRP2 is essential for estradiol-17 beta (beta-D-glucuronide)-induced cholestasis in rats. Hepatology 2000; 32:66–72. 361. Simon FR, Fortune J, Iwahashi M, et al. Ethinylestradiol cholestasis involves alterations in expression of liver sinusoidal transporters. Am J Physiol 1996; 271:G1043–G1052. 362. Lai K, Harnish DC, Evans MJ. Estrogen receptor a regulates expression of the orphan receptor small heterodimer partner. J Biol Chem 2003; 278:36418–36429. 363. Fiorucci S, Clerici C, Antonelli E, et al. Protective effects of 6ethyl chenodeoxycholic acid, a farnesoid X receptor (FXR) ligand, in estrogen induced cholestasis. J Pharmacol Exp Ther 2005 (Epub ahead of print). 364. Lee JM, Trauner M, Soroka CJ, et al. Expression of the bile salt export pump is maintained after chronic cholestasis in the rat. Gastroenterology 2000; 118:163. 365. Hirohashi T, Suzuki H, Takikawa H, et al. ATP-dependent transport of bile salts by rat multidrug resistance-associated protein 3 (mrp3). J Biol Chem 2000; 275:2905. 366. Liang D, Hagenbuch B, Stieger B, et al. Parallel decrease of Na+-taurocholate cotransport and its encoding messenger
1484
367.
368.
369.
370.
371. 372.
373.
374. 375.
376.
377.
378.
379.
380. 381. 382. 383.
384.
385.
386.
RNA in primary cultures of rat hepatocytes. Hepatology 1993; 18:1162. Bohme M, Buchler M, Muller M, et al. Differential inhibition by cyclosporins of primary-active ATP-dependent transporters in the hepatocyte canalicular membrane. FEBS Lett 1993; 333:193. Stone BG, Udani M, Sanghvi A, et al. Cyclosporin A-induced cholestasis. The mechanism in a rat model. Gastroenterology 1987; 93:344. Dubin IN, Johnson FB. Chronic idiopathic jaundice with unidentified pigment in liver cells: a new clinocopathologic entity with a report of 12 cases. Medicine (Baltimore) 1954; 33:155. Sprinz H, Nelson RS. Persistent nonhemolytic hyperbilirubinemia associated with lipochrome-like pigment in liver cells: report of four cases. Ann Intern Med 1954; 41:952. Dubin IN. Chronic idiopathic jaundice: a review of fifty cases. Am J Med 1958; 23:268. Shani M, Seligshoh U, Gilon E, et al. Dubin–Johnson syndrome in Israel. I. Clinical laboratory, and genetic aspects of 101 cases. West J Med 1970; 39:549. Cohen L, Lewis C, Arias IM. Pregnancy, oral contraceptives, and chronic familial jaundice with predominantly conjugated hyperbilirubinemia (Dubin–Johnson syndrome). Gastroenterology 1972; 62:1182. Nakata F, Oyanagi K, Fujiwara M, et al. Dubin–Johnson syndrome in a neonate. Eur J Pediatr 1979; 132:299. Morita M, Kihava T. Intravenous cholecystography and metabolism of meglumine iodipamide (biligrafin) in Dubin–Johnson syndrome. Radiology 1971; 99:57. Muscatello U, Mussini L, Agnolucci MT. The Dubin–Johnson syndrome: an electronmicroscopic study of the liver cell. Acta Hepatosplen 1967; 14:162. Wegmann R, Rangier M, Eteve J, et al. Melanose hepatosplenique avec ictere chronique a bilirubine directe: maladie de Dubin–Johnson? Etude clinique et biologique de la maladie. Etudie histochimique, chimique et spectographique du pigment anormal. Semin Hop Paris 1960; 26:1761. Arias IM, Bernstein L, Roffler R, et al. Black liver diseases in Corriedale sheep: metabolism of tritiated epinephrine and incorporation of isotope into the hepatic pigment in vivo. J Clin Invest 1965; 44:1026. Swartz HM, Sarna T, Varma RR. On the nature and excretion of the hepatic pigment in the Dubin–Johnson syndrome. Gastroenterology 1979; 76:958. Arias IM, Blumberg W. The pigment in Dubin–Johnson syndrome. Gastroenterology 1979; 77:820. Swartz HM, Sarnat T, Varma RR. The pigment in Dubin–Johnson syndrome. Gastroenterology 1979; 77:821. Ware A, Eigenbrodt E, Naftalis J, et al. Dubin–Johnson syndrome and viral hepatitis. Gastroenterology 1974; 67:560. Schoenfield LJ, McGill DB, Hunton DB, et al. Studies of chronic idiopathic jaundice (Dubin–Johnson syndrome). I. Demonstration of hepatic excretory defect. Gastroenterology 1963; 44:101. Rodes J, Zubizarreta A, Bruguera M. Metabolism of the bromsulphalein in Dubin–Johnson syndrome. Diagnostic value of the paradoxical in plasma levels of BSP. Dig Dis 1972; 17:545. Machida I, Inagaki Y, Suzuki S, et al. Mutation analysis of the multidrug resistance protein 2 (MRP2) gene in a Japanese patient with Dubin–Johnson syndrome. Hepatol Res 2004; 30:86–90. Keitel V, Kartenbeck J, Nies AT, et al. Impaired protein maturation of the conjugate export pump multidrug resistance protein 2 as a consequence of a deletion mutation in Dubin–Johnson syndrome. Hepatology 2000; 32:1317.
Chapter 74 BILIRUBIN METABOLISM AND ITS DISORDERS
387. Bohan A, Boyer JL. Mechanisms of hepatic transport of drugs: implications for cholestatic drug reactions. Semin Liver Dis 2002; 22:123–136. 388. Kitamura T, Jansen P, Hardenbrook C, et al. Defective ATPdependent bile canalicular transport of organic anions in mutant (TR-) rats with conjugated hyperbilirubinemia. Proc Natl Acad Sci USA 1990; 87:3557. 389. Mills CO, Milkiewicz P, Müller M, et al. Different pathways of canalicular secretion of sulfated and non-sulfated fluorescent bile acids: a study in isolated hepatocyte couplets and TR– rats. J Hepatol 1999; 31:678. 390. Kitamura T, Alroy J, Gatmaitan Z, et al. Defective biliary excretion of epinephrine metabolites in mutant (TR-) rats: relation to the pathogenesis of black liver in the Dubin–Johnson syndrome and Corriedale sheep with an analogous excretory defect. Hepatology 1992; 15:1154. 391. Jansen PL, Van Klinken JW, van Gelder M, et al. Preserved organic anion transport in mutant TR- rats with a hepatobiliary secretion defect. Am J Physiol 1993; 265:G445. 392. Borst P, Evers R, Kool M, et al. The multidrug resistance protein family. Biochim Biophys Acta 1999; 1461:347. 393. Kool M, van der Linden M, de Haas M, et al. Expression of human MRP6, a homologue of the multidrug resistance protein gene MRP1, in tissues and cancer cells. Cancer Res 1999; 59:175. 394. Seligsohn U, Shani M, Ramot B, et al. Dubin–Johnson syndrome in Israel. II. Association with factor-VII deficiency. Q J Med 1970; 39:569. 395. Levanon M, Rimon S, Shani M, et al. Active and inactive factorVII in Dubin–Johnson syndrome with factor-VII deficiency, hereditary factor-VII deficiency and on coumadin administration. Br J Haematol 1972; 23:669. 396. Kaplowitz N, Javitt N, Kappas A. Coproporphyrin I and III excretion in bile and urine. J Clin Invest 1972; 51:2895. 397. Wolkoff AW, Cohen LE, Arias IM. Inheritance of the Dubin–Johnson syndrome. N Engl J Med 1973; 288:113. 398. Kondo T, Kuchiba K, Shimizu Y. Coproporphyrin isomers in Dubin–Johnson syndrome. Gastroenterology 1976; 70:1117. 399. Wolkoff AW, Arias IM. Coproporphyrin excretion in amniotic fluid and urine from premature infants: a possible maturation defect. Pediatr Res 1974; 8:591. 400. Cornelius CE, Arias IM, Osburn BI. Hepatic pigmentation with photosensitivity: a syndrome in Corriedale sheep resembling Dubin–Johnson syndrome in man. J Am Vet Med Ass 1965; 146:709. 401. Alpert S, Mosher M, Shanske A, et al. Multiplicity of hepatic excretory mechanisms for organic anions. J Gen Physiol 1969; 53:238.
402. Barnhart JL, Gronwall RR, Combes B. Biliary excretion of sulfobromophthalein compounds in normal and mutant Corriedale sheep. Evidence for a disproportionate transport defect for conjugated sulfobromophthalein. Hepatology 1981; 1:441. 403. Arias IM, Bernstein L, Toffler R, et al. Black liver disease in Corriedale sheep: metabolism of tritiated epinephrine and incorporation of isotope into the hepatic pigment in vivo. J Clin Invest 1965; 44:1026. 404. Jansen PLM, Peters WHM, Lamers WH. Hereditary chronic conjugated hyperbilirubinemia in mutant rats caused by defective hepatic anion transport. Hepatology 1985; 5:573. 405. Jansen PLM, Groothuis GMM, Peters WHM, et al. Selective hepatobiliary transport defect for organic anions and neutral steroids in mutant rats with hereditary conjugated hyperbilirubinemia. Hepatology 1987; 7:71. 406. Jansen, PLM, van Klinken JW, van Gelder M, et al. Preserved organic anion transport in mutant TR-rats with a hepatobiliary secretion defect. Am J Physiol 1993; 265:G445. 407. Ito K, Suzuki H, Hirohashi T, et al. Molecular cloning of canalicular multispecific organic anion transporter defective in EHBR. Am J Physiol 1997; 272:G16. 408. Rotor AB, Manahan L, Florentin A. Familial nonhemolytic jaundice with direct van den Bergh reaction. Acta Med Phil 1948; 5:37. 409. Wolkoff AW, Wolpert E, Pascasio FN, et al. Rotor’s syndrome: a distinct inheritable pathophysiologic entity. Am J Med 1976; 60:173. 410. Pereira-Lima JE, Utz E, Rosenberg I. Hereditary nonhemolytic conjugated hyperbilirubinemia without abnormal liver cell pigmentation. A family study. Am J Med 1966; 40:628. 411. Wolpert E, Pascasio FM, Wolkoff AW, et al. Abnormal sulfobromophthalein metabolism in Rotor’s syndrome and obligate heterozygotes. N Engl J Med 1977; 296:1099. 412. Schiff L, Billing BH, Oikawa Y. Familial nonhemolytic jaundice with conjugated bilirubin in the serum. N Engl J Med 1959; 260:1315. 413. Kawasaki H, Kinwa N, Irisa T, et al. Dye clearance studies in Rotor’s syndrome. Am J Gastroenterol 1979; 71:380. 414. Hadchouel P, Charbonnier A, Lageron A, et al. A propos d’une nouvelle forme d’ictere chronique idiopathique. Hypothese physiopathologique. Rev Med Chir Mal Foie 1971; 46:61. 415. Dhumeaux D, Berthelot P. Chronic hyperbilirubinemia associated with hepatic uptake and storage impairment: a new syndrome resembling that of the mutant Southdown sheep. Gastroenterology 1975; 69:988. 416. Rapacini GL, Topi GC, Anti M, et al. Porphyrins in Rotor syndrome: a study on an Italian family. Hepato-Gastroenterol 1986; 33:11.
1485
Index Aagenaes’s syndrome 80, 1270, 1306, 1374–5 Abacavir 525 Abciximab 529 ABC-transporter proteins 73–5 Abscess amebic 283 fungal 283–5 pyogenic 282–3 Acalculous gallbladder disease 1182 ACE inhibitors 528–9 acute intermittent porphyria 1400 Aceruloplasminemia 1248 Acetaldehyde 587–8 oxidation to acetate 582–3 oxidation of alcohol to 581–2 Acetaldehyde adducts in alcoholic liver diseases 590 antibodies to 589–90 Acetaminophen acute intermittent porphyria 1400 cell death intracellular proteases 46–7 mitochondrial dysfunction 45–6 nuclear DNA damage 46 and granuloma 774 hepatotoxicity 387–9, 394 necrosis 207 Acetazolamide, acute intermittent porphyria 1400 N-Acetyltransferase 2 60–1 Aciclovir cytomegalovirus 726 human herpes virus-6 728 Acid-base management, acute liver failure 396–7 Actinomycin D, and sinusoidal obstruction syndrome 897 Acute cellular rejection 967 Acute fatty liver of pregnancy 1014–17 clinical features 1015–16 etiology 1014–15 incidence 1014 natural history and prognosis 1016–17 pathology 1016 therapy 1017 Acute hepatitis, histopathology 208–9 Acute icteric hepatitis 706–7 Acute intermittent porphyria 1397–404 acute attacks prevention 1403–4 treatment 1401–3 clinical manifestations 1397–8 drugs 1399–400 endocrine factors 1399 etiology and pathogenesis 1398–9 laboratory evaluation and diagnosis 1401 neurological mechanisms 1401 nutritional factors 1400 pathogenesis of acute attacks 1399
Acute liver failure 383–415 complications bleeding 397–8 cardiovascular 398 cerebral edema and intracranial hypertension 401–2 hepatic encephalopathy 401 infection 398–9 multiorgan failure 401 psychomotor agitation 402 pulmonary 398 renal insufficiency 399–400 seizures 402 definition 383 etiology and epidemiology 383–9 biological toxins 389 diffuse malignant infiltration 389 drug-induced 386–9 hepatotrophic viruses 385–6 ischemic causes 389 metabolic causes 389 rare causes 389 management 394–403 fluids, electrolytes and acid-base abnormalities 396–7 initial evaluation and triage 394–5 nutrition 397 orthotopic liver transplantation 403 protection/regeneration of liver 402–3 specific therapies 395–6 pathogenesis and clinical features cardiovascular consequences 391 early presentation 390 failure of biosynthesis 390–1 failure of hepatobiliary excretion 390 failure of intermediary metabolism 390 failure to metabolize toxic substances 390 gastrointestinal consequences 393 hematologic disturbances 392 hepatic encephalopathy 393–4 immune system breakdown 392–3 intracranial hypertension and cerebral edema 394 microcirculatory dysfunction 391 neurologic consequences 393 pulmonary consequences 391–2 renal and electrolyte disturbances 392 prognosis and natural history 403–5 Acute renal failure 392 management 399–401 Adaptive immune response 117–18, 149 antigen-presenting cells 117 B cells 118 CD4+ cells 117–18 Adaptive immune system alcoholic liver disease 589–92 CD8+ T cells 118
Addison’s disease 236 Adefovir dipivoxil 525 hepatitis B 653 Adenomatous polyposis 26 Adenosine 433, 529 S-Adenosyl-methionine 101 Adenovirus 728 in children 1443 Adriamycin 536 Aflatoxin 169–70, 574–5 and hepatocellular carcinoma 1111 African green-monkey kidney cell model 627, 629 Age, as contraindication to liver transplantation 936 Alagille’s syndrome 207, 217, 225, 226, 947, 1270, 1306–8, 1371–4 clinical features 1372–3 diagnosis 1373–4 etiology and pathogenesis 1371–2 management 1374 prognosis 1374 Alanine aminotransferase 235, 236–7 Albendazole 523, 524, 742 Albumin 239 bilirubin binding by 1456–7 Alcohol absorption, distribution and excretion 580–1 abstinence from 607 acute intermittent porphyria 1400 and anemia 493 beverage content 1043 and cancer 605 dose 594–5 and hepatocellular carcinoma 1110–11 metabolism 581–8 and alcoholic liver disease 583–8 alterations following chronic consumption 583 antibodies to metabolic products 590–1 oxidation to acetaldehyde 581–2 oxidation of acetaldehyde to acetate 582–3 site of alcohol oxidation 581 and porphyria cutanea tarda 1408 and pregnancy 1006 Alcohol abuse, as contraindication to liver transplantation 935–6 Alcohol dehydrogenase 581–2 Alcoholic cirrhosis 610–11 liver transplantation 611–12 treatment antioxidants 610 colchicine 610 phosphatidylcholine 610–11 propylthiouracil 610 Alcoholic dementia 604
1487
Index
Alcoholic fatty liver disease 217, 597–8 clinical features 598 pathogenesis 583–5 altered redox state 583 altered triglyceride export 584 dietary fat 583 inhibition of methionine cycle 584–5 PPAR-a inhibition 583–4 steatosis 585 tumor necrosis factor-a 584 pathology 597–8 prognosis 598 see also Alcoholic liver disease Alcoholic hepatitis 593, 598–9 clinical features 598–9 liver transplantation 612 pathology 598 prognosis 599–600 treatment 607–10 antioxidants 609 corticosteroids 608 experimental therapies 609–10 hepatic mitogens 609 hepatorenal system 610 nutritional supplementation 608–9 pentoxifylline 608 propylthiouracil 609 Alcoholic liver disease 89, 360, 579–623 cardiovascular complications 602–3 arrhythmias and sudden cardiac death 603 cardiomyopathy 603 cerebrovascular disease 603 coronary artery disease 603 hypertension 602–3 clinical examination 597 diagnosis 596–7 endocrine complications 606 epidemiology 579–80 gastrointestinal complications 600–2 colon 602 esophagus 601 pancreas 602 salivary glands and oropharynx 600–1 small intestine 601–2 stomach 601 hematological complications 605–6 coagulation 606 erythrocytes 605–6 leukocytes 606 platelets 606 history 597 laboratory investigations 597 liver biopsy 597 nervous system 603–4 alcoholic dementia 604 brainstem disease 604 cerebellar disease 604 neuropathies 604 Wernicke-Korsakoff syndrome 603–4 pathogenesis 580–94 absorption, distribution and excretion 580–1 adaptive immune system 589–92 alcohol-induced fibrosis 592–3 alcohol metabolism 581–8 cellular immune response 592 hepatocellular carcinoma 593–4 innate immune system 588–9
1488
Alcoholic liver disease—cont’d post-transplant 998–9 susceptibility to 594–6 diet 595 dose of ethanol 594–5 gender 595 genetic factors 595–6 treatment 607–12 abstinence 607 see also Alcoholic fatty liver disease; Alcoholic hepatitis; Cirrhosis Alcoholic myopathy 604–5 Alcoholic steatohepatitis 1049–50 Alcohol-induced fibrosis 592–3 Alkaline phosphatase 235, 237–8 in pregnancy 237 Alkylating agents 536–7 busulfan 536 chlorambucil 536–7 cyclophosphamide 536 dacarbazine 537 ifosfamide 536 nitrosoureas 537 Allopurinol acute intermittent porphyria 1400 and granuloma 774 Alopecia 797 Alosetron 505 a-1-antitrypsin deficiency 168, 1257–67, 1270 clinical features 1261–2 diagnosis 1262–3 epidemiology 1257 and hepatocellular carcinoma 1111 histopathology 210, 212, 221 liver cell therapy 183 pathogenesis 1257–60 pediatric 1379–80 investigation 1379–80 management 1380 presentation 1379 treatment 1263–5 ALT : AST ratio 1044 Alternative reading frame protein 128 Alveolar echinococcosis, post-transplant 998 Amanita phalloides 389, 571–2 Amanita verna 389, 571–2 Amanita virosa 389, 571–2 Amantadine 517 Amebic abscess 283 Amfenac 532 Amikacin 517 Amiloride, acute intermittent porphyria 1400 g-Aminolevulinic acid dehydratase porphyria 1395–7 clinical manifestations 1396 etiology and pathogenesis 1396 history, definition and prevalence 1395–6 laboratory evaluation and diagnosis 1396–7 treatment 1397 Aminopyrine, acute intermittent porphyria 1400 14 C-Aminopyrine breath test 246 Aminotransferases 236–7 non-alcoholic steatohepatitis 1044 normal reference ranges 1044 Amiodarone 517, 529, 530 and non-alcoholic steatohepatitis 1042
Amitryptiline8 517 Ammonia blood testing 318 reduction of production/absorption 320–1 Ammonia hypothesis of hepatic encephalopathy original 312–13 unifying 313 Ammonium tetrathiomolybdate 1231 Amniotic fluid 180 Amodiaquine 523 Amoxicillin 527 Amoxycillin-clavulinic acid, and granuloma 774 Amphiregulin 27 Amphotericin B 522 Ampicillin 527 Amprenavir 525 Amsinckia intermedia 573 Amylo-1,6-glucosidase deficiency 1289–91 biochemical characteristics and laboratory findings 1290 clinical characteristics 1290 molecular basis 1289–90 pathology 1290 treatment 1290–1 Amyloidosis 280–2, 1077–9 clinical features 1078–9 diagnosis 1078–9 post-transplant 998 primary and secondary 1077–8 treatment 1078–9 Anabolic steroids 517 Anagrelide 529 Analbuminemia 1457 Anaplasma phagocytophilum 759 Anemia 491–3 and alcohol consumption 493 hemolytic 492 hepatitis-associated 492 macrocytic 491–2 microcytic 491 Anesthetic-induced liver injury 1101–2 Anesthetics 517–19 Angiodysplasia 918 Angiogenesis 170, 1136 Angiomyolipoma 262–4 Angiotensin-converting enzyme inhibitors see ACE inhibitors Angiotensin II receptor blockers 529 Anicteric hepatitis 707 Anistreplase 529 Anorexia nervosa 236 Antiacetaldehyde antibodies 590 Anti-Alzheimer’s drugs 521 Antiandrogens, hepatocellular carcinoma 1122 Antiarrhythmic drugs 529–30 Antibacterials 526–8 macrolides 527 penicillins and cephalosporins 526–7 quinolones 527 sulfonamides 527–8 tetracyclines 528 Antibodies antiacetaldehyde 590 products of ethanol metabolism 590–1
Index
Anticoagulants 528 Anticonvulsants 519–21 See also individual drugs Antidiabetic agents 522 Antidiuretic hormone 430–1 Antifibrotic agents hepatic fibrosis 102–3 primary biliary cirrhosis 812–13 Antifungal drugs 522–5 Antigen-presenting cells 117 induction of T-cell tolerance 155 organ-resident 157 Anti-HIV agents 525 Antihypertensive drugs 531 Anti-inflammatory agents, parenteral nutrition-induced liver disease 1094 Antilymphocyte globulin, liver transplantation 964 Antimalarial drugs 522–5 Antimetabolites 534–6 adriamycin 536 L-asparaginase 535–6 ciclosporin A 536 cytosine arabinoside 535 dactinomycin 536 etoposide 536 5-fluorouracil and floxuridine 535 6-mercaptopurine and azathioprine 534–5 methotrexate 534 mithramycin 536 6-thioguanine 535 vinca alkaloids 536 Antimicrobial therapy, parenteral nutritioninduced liver disease 1094 Antimigraine drugs 521 Antimycobacterial therapy, drug-induced liver injury 783 Antineoplastic and immunosuppressive agents 533–7 alkylating agents 536–7 busulfan 536 chlorambucil 536–7 cyclophosphamide 536 dacarbazine 537 ifosfamide 536 nitrosoureas 537 antimetabolites 534–6 adriamycin 536 L-asparaginase 535–6 ciclosporin A 536 cytosine arabinoside 535 dactinomycin 536 etoposide 536 5-fluorouracil and floxuridine 535 6-mercaptopurine and azathioprine 534–5 methotrexate 534 mithramycin 536 6-thioguanine 535 vinca alkaloids 536 biologic response modulators 537 Antinuclear antibody 807 Antioxidant agents 101–2 polyenylphosphatidylcholine 101 silymarin 101 Antioxidant defences, depletion in alcoholic liver disease 586–7
Antioxidants alcoholic cirrhosis 610 alcoholic hepatitis 609 parenteral nutrition-induced liver disease 1094 porphyria cutanea tarda 1408 Wilson’s disease 1232 Antiparasitic drugs 522–5 Anti-Parkinson’s drugs 521 Antiphospholipid autoantibodies 591 Antiphospholipid syndrome 879, 1073 Antiplatelet agents 528 Antiproliferative agents, liver transplantation 965 Antipyrine, acute intermittent porphyria 1400 Antiretroviral therapy drug-induced liver injury 783 and hepatitis B flares 646 Antithrombin III 490, 529 Antithymocyte globulin graft-versus-host disease 869 liver transplantation 964 Antitubercular agents 522–5 Antiviral therapy 525–6 hepatitis B 650–6 agents under development 653–4 choice of agents 650 entecavir 653 guidelines 650–1 HIV/HBV coinfected patients 655 innovative approaches 655–6 interferon-a 651–2 novel therapies 654–5 nucleoside analogs 652–3 virologic endpoints 650 and hepatitis B flares 645–6 post-transplantation 978–9, 988 pretransplantation 977–8, 986–8 See also individual drugs APACHE-II score 299 Apoferritin 13 Apolipoprotein E deficiency, liver cell therapy 183 Apoptosis 37–40, 506–7, 1136 biochemical markers 40 Fas receptor-mediated 40–2 and inflammation 39–40 intrinsic (mitochondrial) pathway 44 and secondary necrosis 39 TNF receptor mediated NF-kB activation 42–3 Apoptosis protease-activating factor-1 41 APRI index 241 Argatroban 529 Arginine vasopressin 430–1 Arias’ syndrome 1469 Aromatic hydrocarbons 566 Arsenic 517, 562, 568 Arsenic, causing portal hypertension 58 Arterial embolization 1122–5 Arterial vasodilatation theory 438–40 Arteriohepatic dysplasia see Alagille’s syndrome Arteriovenous fistula 358–9 Arthralgias 671 Arthropathy 643 of hemochromatosis 1242
Ascaris lumbricoides 737–40 Ascites 333–46 associated conditions 336 cirrhotic 302–3 clinical features 334 as contraindication to liver biopsy 201 diagnosis 334–6 differential diagnosis 334–6 epidemiology 333 in HIV patients 787–8 pathogenesis 333–4 sinusoidal hypertension 333 sodium retention 333–4 prognosis and natural history 343 renal dysfunction 418–21 and spontaneous bacterial peritonitis 336–9 treatment 339–42 diuretics 340 large-volume paracentesis 340–1 peritoneovenous shunt 342 sodium restriction 340 TIPS 341–2 treatment of associated conditions hepatic hydrothorax 342–3 hyponatraemia 342 see also Sodium retention Ascites total protein 335 L-Asparaginase 535–6 Aspartate aminotransferase 235, 236–7 Aspergillus spp. 761 Aspergillus flavus 169 Aspergillus parasiticus 169 Aspirin 531–2 acute intermittent porphyria 1400 Asterixis 317 Astrocyte swelling 312 Atazanavir 525 ATP-binding cassette transporter proteins 72 Atractylis gummifera 552–3 Atropine, acute intermittent porphyria 1400 Atteplase 529 Autoantibodies 591, 855 alcoholic liver disease 591–2 antiphospholipid 591 non-specific 591 primary sclerosing cholangitis 824 Autoimmune hemolytic anemia 797 Autoimmune hepatitis 210, 211, 795–801 alternative treatments 799 causes, risk factors and disease associations 795 clinical presentation 798 diagnosis 798–9 differential diagnosis 798 epidemiology 795 extrahepatic disease associations 797 with features of primary biliary cirrhosis 858, 860 with features of primary sclerosing cholangitis 859, 860 and hepatitis C 672 and hepatocellular carcinoma 1111 laboratory testing 244, 245 pathogenesis 795–6 pathology 797–8 post-transplant 996 in pregnancy 1006
1489
Index
Autoimmune hepatitis—cont’d primary biliary cirrhosis 808 and primary sclerosing cholangitis in adults 856–7 in children 855–6 prevention 858 specific treatment 857–8 treatment and prevention 799 see also Overlap syndromes Autoimmune liver disease 701 dynamic nature of 855 see also Autoimmune hepatitis Autoimmune thyroid disease 797 Autonomic dysfunction 460–1 Autophagosomes 1259 Autosomal dominant polycystic kidney disease 1332–3 events contributing to 1334–5 formation of mechanosensory complex 1334 histopathology 1336–7, 1338 moderation of transcription factor activities 1334 PKD1 and polycystin-1 1332–3 PKD2 and polycystin-2 1333 Autosomal recessive polycystic kidney disease 1336 histopathology 1337 Axin 26 Azathioprine 517, 534–5 autoimmune hepatitis 799 graft-versus-host disease 869 and hepatitis C recurrence 984 liver transplantation 964 primary biliary cirrhosis 813 and sinusoidal obstruction syndrome 897 Azithromycin 527
Bacillary peliosis 286 Bacille Calmette-Guérin, granuloma formation 772 Bacteremia 198 Bacterial and fungal hepatitis 755–9 Gram-negative bacilli 756–7 bartonellosis and peliosis hepatitis 757 brucellosis 756 Burkhoderia pseudomallei 757 legionellosis 756 salmonella 756 Gram-negative cocci 757–9 Borreliosis 758 Leptospirosis 758–9 Mycobacterium avium-iontracellulare 758 Mycobacterium tuberculosis 757–8 Neisseria spp. 757 Treponema pallidum 758 Gram-positive bacilli 756 Clostridium 756 Listeria 756 Gram-positive cocci 756 Bacterial infections 747–65 bacterial and fungal hepatitis 755–9 endotoxin and inflammatory-related hepatitis 755 graft-versus-host disease 868 neonatal hepatitis 1369 post-transplantation 967
1490
Bacterial infections—cont’d pyelophlebitis 755 pyogenic liver abscess 282–3, 747–54 see also individual bacterial infections Bacterial overgrowth 1091 Balloon tamponade of varices 368 Band ligation of varices 371–2 Banti, Guido 359 Barbiturates, acute intermittent porphyria 1400 Bariatric surgery 1052 Barium esophagram 349 Bartonella henselae 787 Bartonellosis 757 Basiliximab daclizumab, liver transplantation 964 Basolateral anionic conjugate transporters 73 Basolateral transport proteins 71 B cells 118 Bedside diagnostic tools 96 Behçet’s disease 879 Benign liver tumors 1147–67 focal nodular hyperplasia 227, 252–4, 362, 1152–9 associated conditions 1158 clinical features 1152 diagnosis 1152–3 diagnostic workup 1153–8 epidemiology 1152 imaging 1153 pathogenesis 1152 pathology 1152 in pregnancy 1010 prognosis and natural history 1158–9 treatment 1158 hemangioma 257–61, 1147–52 clinical features 1147 diagnosis 1147–8 diagnostic workup 1150–1 epidemiology 1147 imaging 1148–50 pathogenesis 1147 in pregnancy 1010 prognosis and natural history 1151–2 treatment 1151 hepatocellular adenoma 251–2, 1159–62 associated conditions 1161 clinical features 1159 diagnosis 1159 diagnostic workup 1161 epidemiology 1159 imaging 1159–61 pathogenesis 1159 pathology 1159 in pregnancy 1010 prognosis and natural history 1162 treatment 1161–2 nodular regenerative hyperplasia 228, 255, 362, 1162–3 associated conditions 1163 clinical features 1162 diagnosis 1162 disease complications 1163 epidemiology 1162 imaging 1163 pathogenesis 1162 prognosis and natural history 1163 treatment 1163
Benign recurrent intrahepatic cholestasis type 1 77–8 type 2 79 Benorilate 532 Benoxaprofen 505, 532 Beryllium 562, 569 b-blockers hepatotoxicity 529, 530 prevention of variceal bleeding 369–70 Betaine, non-alcoholic steatohepatitis 1053 Bethanidine, acute intermittent porphyria 1400 Bezafibrate, primary biliary cirrhosis 814 Bile acid dissolution therapy 1183 Bile acid metabolism disorders 1304–8 Alagille’s syndrome 207, 217, 225, 226, 947, 1270, 1306–8 hereditary lymphedema with recurrent cholestasis 80, 1270, 1306 progressive familial intrahepatic cholestasis 1304–5 progressive intrahepatic cholestasis type I (Byler’s disease) 1304–5 progressive intrahepatic cholestasis type II 1305 progressive intrahepatic cholestasis type III 1305 progressive intrahepatic cholestasis type I (Byler’s disease) 76–7, 207–8, 1270, 1304–5 Bile acid synthesis defects 79, 1379 Bile canaliculi 8 Bile duct adenoma 227, 257 agenesis 1367 extramural compression 1367 hamartoma 257 histology 207–8 imaging 292–3 injuries 1194–6, 1209–10 intrahepatic diseases, histopathology 216–17 necrosis 219 obstruction, postoperative 1103 paucity 1371–5 Aagenaes’s syndrome see Aagenaes’s syndrome Alagille’s syndrome see Alagille’s syndrome non-syndromic 1374 progressive familial intrahepatic cholestasis 1375 proliferation 214 spontaneous perforation 1366 stenosis 1366–7 strictures 1211 tumors 1367 vanishing bile duct syndrome 229 Bile formation, in pregnancy 1004 Bile infarct 215 Bile lakes 215 Bile pigments 208 Bile plug syndrome 1366 Bile salts 67 excretory pump 73 transporters 73
Index
Bile secretion 3, 67–85 bile secretion unit 68–70 genetic defects 76–80 benign recurrent intrahepatic cholestasis type 1 77–8 type 2 79 bile acid synthesis defects 79 Dubin-Johnson syndrome see DubinJohnson syndrome familial hypercholanemia 79 intrahepatic cholestasis of pregnancy 79–80 progressive familial intrahepatic cholestasis type 1 76–7 progressive familial intrahepatic cholestasis type 2 78–9 progressive familial intrahepatic cholestasis type 3 79 hepatic transport proteins 70–5 basolateral transport proteins 71 canalicular transport proteins 72–5 regulation of 75–6 Bile synthesis 3 Biliary atresia, extrahepatic 225, 1356–65 clinical features 1359–60 complications 1364–5 etiology 1356–7 incidence 1356 liver transplantation 1365 outcome 1365 pathogenesis 1358–9 pathology 1359 postoperative care 1364 specific investigations 1360–2 surgical management 1363–4 surgical prognostic factors 1364 treatment and prognosis 1362–3 Biliary atresia splenic malformation syndrome 1357 Biliary cirrhosis 216 Biliary cystadenocarcinoma 271–2 Biliary cystadenoma 256–7 Biliary disease and graft-versus-host disease 867 histology 212–13 Biliary diversion 1386 Biliary hamartoma 258 Biliary imaging 291–6 biliary strictures 294 common bile duct and gallbladder stones 292–3 gallstones and cholecystitis 291–2 intrahepatic stones 293–4 right upper quadrant pain 291 Biliary leaks 1194 Biliary lesions bile duct adenoma 257 biliary cystadenoma 256–7 hepatic (biliary) cysts 255–6 Biliary obstruction, as contraindication to liver biopsy 201 Biliary stents 1142 Biliary strictures 294 Biliary tract 8 abnormalities in HIV patients 788–9 cholangiographic patterns 788 clinical presentations 788 diagnosis 788
Biliary tract—cont’d abnormalities in HIV patients—cont’d etiology 788 treatment 789 complications of liver transplantation 957, 967 disease in pregnancy 1009 sickle-cell anemia 1079–80 see also Gallstones reconstruction 953 surgery 1201–17 see also Gallbladder Biliary tree, duplication of 1367 Bilirubin 3, 235, 238–9 in body fluids 1464–5 bile 1464 cerebrospinal fluid 1465 plasma 1464 skin and sclera 1465 tissue fluids 1465 urine 1464 chemistry 1452–3 absorption spectra and circular dichroism 1452–3, 1454 photochemistry 1453 physical conformation and solubility of bilirubin IXa 1452 disposition of 1456–63 albumin 1456–7 alternative pathways of elimination 1463 bile canalicular secretion of bilirubin conjugates 1460–2 conjugation 1459–60 extrahepatic handling 1463 fate in gastrointestinal tract 1462–3 hepatic uptake 1457–9 acquired and genetic abnormalities of 1458–9 energy requirements 1458 transporters 1458 storage in hepatocytes 1459 formation of 1450–2 conversion of biliverdin to bilirubin 1451 opening of heme ring 1450–1 production 1451–2 quantification of 1463–4 chromatographic analysis 1464 conversion to azoderivatives 1463–4 slide tests 1464 transcutaneous bilirubinometry 1464 toxicity 1453–6 bilirubin encephalopathy clinical features 1455 in Gunn rat 1453, 1455 biochemical basis of 1456 blood-brain barrier and cerebral bilirubin clearance 1455–6 nephrotoxicity 1456 Bilirubin encephalopathy clinical features 1455 in Gunn rat 1453, 1455 Bilirubin glucuronidation disorders 1466–9 Bilirubin metabolism disorders 1465–75 acquired defects of hepatobiliary transport 1471–2 bilirubin overproduction 1466
Bilirubin metabolism disorders—cont’d Dubin-Johnson syndrome see DubinJohnson syndrome Gilbert’s syndrome 1469–71 inherited disorders of bilirubin glucuronidation 1466–9 jaundice 1449–50 neonatal hyperbilirubinemia 1465–6 rotor syndrome 1474 Bilirubin oxidase 1468 Biloma 1210–11 Biochemical markers 40 Biological toxins 389 Biologic DT system 324 Biologic response modulators 537 Biopsy gun 196 Biosynthesis, failure of 390–1 Bipotency 177 Bivalirudin 529 Black cohosh 553 Bleeding, acute liver failure 397–8 Blood-based markers 96–8 Blood products, transfusion of 494–5 Blood volume 459–60 Borrelia burgdorferi 759 Borreliosis 758 Bosentan 529, 531 Bovine viral diarrhea virus 129 Brachytherapy, intraluminal 1143 Brain imaging, hepatic encephalopathy 318 Brainstem disease 604 Bretylium 529 BRIC 1376–7 genetics 1376–7 investigation 1377 liver biopsy 1377 management 1377 presenting features 1377 Bromfenac 505 Bromides, acute intermittent porphyria 1400 Bronze diabetes see Hemochromatosis Brucellosis 756 granuloma formation 773–4 Budd-Chiari syndrome 219, 220, 877–96 acute 887 acute liver failure 389 causes 877–80 chronic 887 diagnosis 887–9 diagnostic workup 889 direct evidence 887–8 indirect evidence 888–9 epidemiology 877 fulminant 887 imaging 278–9 liver damage 881–4 manifestations and course 884–7 pathology and pathogenesis 880–1 local factors 881 thrombosis 880–1 post-transplant 998 in pregnancy 1010–11 primary 877–9 secondary 879–80 subacute 887 survival and prognosis 891–3
1491
Index
Budd-Chiari syndrome—cont’d therapy 889–91 complications 889 liver decompression 889–90 liver transplantation 890–1 management 891 underlying condition 889 TIPS 303 Budesonide, primary biliary cirrhosis 813 Bumetanide, acute intermittent porphyria 1400 Burkhoderia pseudomallei 757 Busulfan 536 Byler’s disease 76–7, 207–8, 1270, 1304–5
Cadmium 570 Calcineurin inhibitors, liver transplantation 964–5 Calcitonin gene-related peptide 8 Calcium channel blockers 530 acute intermittent porphyria 1400 Caliciviruses 693 Callilepsis laureola 552–3 Calot’s triangle 1210 Calpains 47 Camphor oil 553–4 Canalicular membrane 68 Canalicular transport proteins 72–5 ABC-transporter proteins 73–5 basolateral anionic conjugate and bile salt transporters 73 bile salt excretory pump 73 MDR1 P-glycoprotein 72 multidrug resistance protein MRP2 73 phospholipid transporter 72 Candida spp. 761 Candida albicans 283 Caput medusae 275 Carbamazepine 519–20 acute intermittent porphyria 1400 and granuloma 774 Carbohydrate loading 1402 Carbohydrate metabolism, errors of 1269–70 Carbon monoxide 457 Carbon tetrachloride 510, 562, 566–7 Cardiac arrhythmias 603 Cardiac complications 461–72 Alagille’s syndrome 1372 cardiovascular function and liver transplantation 470–1 cirrhotic cardiomyopathy 462–70, 471 coronary artery disease 470 endocarditis and pericarditis 470 hemochromatosis 1242 natural history 471–2 Cardiac disease, post-transplantation 969 Cardiocirculatory theory 440–1 Cardiomyopathy 603 Cardiovascular agents 528–31 angiotensin-converting enzyme inhibitors 528–9 angiotensin II receptor blockers 529 antiarrhythmic drugs 529–30 antihypertensive drugs 531 antiplatelet/anticoagulant/thrombolytic agents 528 b-blockers 530
1492
Cardiovascular agents—cont’d cholesterol-lowering agents 530–1 diuretics 531 Cardiovascular complications 453–75 acute liver failure 391, 398 alcoholic liver disease 602–3 arrhythmias and sudden cardiac death 603 cardiomyopathy 603 cerebrovascular disease 603 coronary artery disease 603 hypertension 602–3 heart see Cardiac complications hyperdynamic circulation 453–61 clinical features 453–4 complications 460–1 management 461 pathogenesis 454–60 Cardiovascular disease as contraindication to liver transplantation 936 post-transplantation 968 Carisoprodol, acute intermittent porphyria 1400 Carmustine 537 Caroli’s disease 1347 characteristics 1347 clinical features 1347 diagnosis 1347 histopathology 1337, 1341 pediatric 1365–6 treatment 1347 Carpofen 532 Cascara 553, 554 Caspase-dependent DNase 41 Caspofungin 523 Catalase 582 b-Catenin 26, 169, 170–3 Cat-scratch disease 774 CD4+ cells 117–18 CD8+ T cells 118 CD81 135 Cefaclor 527 Cefdinir 527 Ceftriaxone 527 Celiac disease 236, 797 Cell cycle regulation 24–5 Cellular immune response, in alcoholic liver disease 592 b-Cellulin 27 Central nervous system complications, Osler-Weber-Rendu syndrome 918 Central veins 220 Cephalosporins 526–7 Cerebellar disease 604 Cerebral circulation 461 Cerebral edema 394 management 401–2 Cerebrohepatorenal syndrome see Zellweger’s syndrome Cerebrovascular disease 603 Cerivistatin 505 Ceroid 208 Ceruloplasmin 1227 C-FLIP 42 Chaparral 553, 554 Chaso 553, 554–5
Checkpoints 24 Chelation therapy 1230–1 Chemokines 31 Chemotherapy neoadjuvant 1174 and sinusoidal obstruction syndrome 897 Cherry-red spots 365 Chilaiditi syndrome 195 Child-Pugh score 299, 343 Children endoscopic retrograde cholangiopancreatography 1193 fatty liver disease 1034 see also Pediatric Chlorambucil 536–7 primary biliary cirrhosis 813, 814 Chloramphenicol 517 Chlorcyclizine 517 Chlordecone 568 Chloride, causing portal hypertension 58 Chlorinated ethylenes 565–6 Chloripramine 517 Chloroquine 517 Chlorpheniramine 517 Chlorpromazine 517 Chlorpropamide, and granuloma 774 Cholangiocarcinoma 271, 836–41, 1133–46 as contraindication to liver transplantation 938 diagnosis 836–40 biliary tumor markers 839–40 clinical symptoms/signs 836–7 histology and brush cytology 837–8 serum tumor markers 838–9 epidemiology 1133–4 extrahepatic 1137–8 intrahepatic 1138–40 molecular pathogenesis 1134–6 pathogenesis 836 pathology and classification 1136–7 risk factors 840–1, 1134 staging 1140 therapy 1140–4 biliary stents 1142 high-intensity intraductal ultrasound 1143 intraluminal brachytherapy 1143 liver transplantation 1143–4 palliative 1142 photodynamic 1142–3 surgical 1141–2 treatment 841 Cholangiocytes 824–5 Cholangiography 1362 Cholangitis 1386 acute 215 ascending 1364 as contraindication to liver biopsy 201 fibro-obliterative 215 as indication for cholecystectomy 1202–3 neonatal sclerosing 1366 Cholangitis lenta 214 Cholate stasis 216 Cholecystectomy 1183 after drainage 1204 bile duct injury 1209–10 in cirrhosis 1211–12 complications 1193–4
Index
Cholecystectomy—cont’d indications 1201–3 acute cholecystitis 1202 cholangitis 1202–3 laparoscopic 1207–9 in pregnancy 1212–13 Cholecystitis acute 1202 in pregnancy 1009 Cholecystostomy, percutaneous 1187–8 Choledochal cysts 1347–8, 1349 characteristics 1347 clinical features 1347 diagnosis 1347–8 histopathology 1337, 1341 pediatric 1365 treatment 1348 Choledocholithiasis 292, 1188–94 complex stone extraction 1191–2 diagnosis 1188–9 endoscopic balloon dilation 1190–1 endoscopic retrograde cholangiopancreatography 1189–90 in pediatrics 1193 in pregnancy 1193 endoscopy in cholecystectomy complications 1193–4 Cholelithiasis 1187, 1201 complications 1202 surgical decompression 1203–4 symptoms 1202 Wilson’s disease 1225 Choleretics 1093–4 Cholescintigraphy 292 Cholescystitis, imaging 291–2 Cholestasis 214 chronic 834 conditions associated with 213 histopathology 212–13 pediatric etiology 1380–2 management 1382–3 postoperative 1102–3 see also individual disorders Cholestatic enzymes 1044 Cholestatic hepatitis, prolonged 707 Cholestatic parenchymal diseases, histopathology 217 Cholestatic pattern injury 512–13, 514–15 Cholesterol gallstones 1181 intracellular metabolism 1298–9 Cholesterol ester storage disease 1270, 1301–4 biochemical characteristics 1302 clinical picture 1301 genetics 1302 laboratory findings 1301–2 pathology 1302 treatment and diagnosis 1302–4 Cholesterol-lowering agents 530–1 Chromosomal disorders 1369 Chronic active hepatitis 1032 Chronic hepatitis histology 209–12 scoring 210, 212 Chronic rejection 229, 970
Ciclosporin 536 graft-versus-host disease 869 and hepatitis C recurrence 984 liver transplantation 964 Cidofovir 525, 526 cytomegalovirus 726 Cilostazol 529 Cimetidine, acute intermittent porphyria 1400 Cinchophen 532 Cirrhosis 87–109, 120, 360, 600 age-adjusted death rates 580 alcoholic see Alcoholic cirrhosis; Alcoholic liver disease biliary 216 cholecystectomy in 1211–12 clinical features 600 compensated 987 coronary artery disease 470 cryptogenic and non-alcoholic steatohepatitis 1055 post-transplant 999 decompensated 987 extrahepatic biliary atresia 1364–5 hepatitis C superinfection 678–9 hepatitis E superinfection 7–9 and hepatocellular carcinoma 1111 histopathology 222–4 imaging 273–5 iron overload 1253 and non-alcoholic steatohepatitis 1055 pathology 600 pathophysiology 90–5 disease-specific mechanisms 95 extracellular matrix 90–1 extracellular matrix-cell interactions 91–2 hepatic stellate cell activation 92–5 resolution 95 and portal vein thrombosis 906 preascitic 420–1 in pregnancy 1006–7 prognosis 600 renal dysfunction 425–45 circulatory abnormalities 425–6 management 441–5 neurohumoral systems 426–35 pathophysiology 438–41 time-course of renal function abnormalities 435–8 reversibility 90 Cirrhotic ascites 302–3 Cirrhotic cardiomyopathy 462–70, 471 cardiac chamber dimensions/histology 464–5 complications 466 electrophysiological abnormalities 465 management 471 pathogenesis 466–70 cardiomyocyte membrane mechanisms 466–7 cellular calcium kinetics 467–8 cGMP-mediated mechanisms 468–70 delayed repolarization 470 serum markers 465–6 systolic and diastolic function 462–4 Cis-acting replication elements 127
Cisapride 505 Cisplatin, hepatocellular carcinoma 1122 Clades 125 Clarithromycin 527 Clevudine 654 Clinical Diagnostic Scale 506 Clofazimine, graft-versus-host disease 870 Clometacin 532 Clonazepam 520 acute intermittent porphyria 1400 Clonidine 529 Clonorchis sinensis 735–7, 836, 1134 clinical presentation 735 diagnosis 735–6 treatment 736–7 Clopidogrel 529 Clostridium 756 C-met 27 Coagulation, in alcoholic liver disease 606 Coagulation cascade 490–1 Coagulation factors 239–40 Coagulopathy 493–4 Cobalamin 491 Cocaine 574 Coccidiodes immitis 760 Coccidioidomycoses 760 Coding region variants 54 Colchicine alcoholic cirrhosis 610 hepatic fibrosis 102–3 primary biliary cirrhosis 812–13 Colon, alcoholic liver disease 602 Colony-stimulating factor 489 Combivir, primary biliary cirrhosis 814 Comfrey 553, 555 Computed tomography fatty liver disease 1045 focal nodular hyperplasia 1153 hemangioma 1148–9 hepatocellular adenoma 1160 Congenital erythropoietic porphyria 1404–5 clinical manifestations 1404 etiology and pathogenesis 1404–5 history, definition and prevalence 1404 laboratory evaluation and diagnosis 1405 treatment 1405 Congenital hepatic fibrosis 361, 1346–7 characteristics 1346 clinical features 1346 histopathology 1337, 1339, 1340 treatment 1346–7 Congestive hepatopathy 1065–7 Connective tissue diseases 1071–4 antiphospholipid syndrome 879, 1073 rheumatoid arthritis 797, 1072–3 scleroderma 1074 Sjögren’s syndrome 797, 1073–4 systemic lupus erythematosus 1071–2 Constrictive pericarditis 208–9, 1069 Contraceptive steroids 517 Copper 569 causing portal hypertension 58 deficiency 1233 disorders of 1221 storage 1224 transport 1223–4 see also Wilson’s disease Copper-binding protein 212
1493
Index
Copper pathway 1221–2 Corkscrew vessels 274 Coronary artery disease 470, 603 Cor pulmonale 219–20 Corticosteroids alcoholic hepatitis 608 graft-versus-host disease 868–9 hepatic fibrosis 100 and hepatitis C recurrence 984 liver transplantation 963 primary biliary cirrhosis 812 withdrawal, and hepatitis B flares 646 Coumarins, acute intermittent porphyria 1400 Coxiella burnetii 389, 759–60 granuloma formation 772–3 CREST syndrome 797 Crigler-Najjar syndrome liver cell therapy 183 type 1 1466–9 abnormalities of hepatic UGTs 1467 Gunn rat as animal model 1467–8 laboratory tests 1467 reduction of serum bilirubin levels 1468–9 treatment 1468 type 2 1469 Crotolaria spp. 573 Cruveilhier-Baumgarten murmur 349 Cryptococcosis 760 Cryptococcus infection, post-transplantation 970 Cryptococcus neoformans 760 Cryptogenic cirrhosis and non-alcoholic steatohepatitis 1055 post-transplant 999 CTLA4lg, graft-versus-host disease 870 CTP scoring system 939, 941–2 Cyanobacterial toxins 575 Cyclin-dependent kinase 24 Cyclooxygenase-2 1135 Cyclophosphamide 536 Cyclosporine, primary biliary cirrhosis 812 CYP2E1 586, 591 CYP2E2 591 Cystic fibrosis 362, 1270, 1311–14 biochemical characteristics 1312 clinical features 1311 genetics 1314 laboratory findings 1313 liver disease in childhood and adolescence 1311–12 molecular basis 1312–13 pathology 1313–14 therapy 1314 Cystogenesis 1331–2 Cysts choledochal 1347–8, 1349 pediatric 1365 hepatic (biliary) 255–6, 1364 hydatid 285 solitary hepatic 1348, 1350 see also Fibrocystic disease Cytochrome P4502C9 57–8 Cytochrome P4502C19 58 Cytochrome P4502D6 55–6 Cytokine genes 596
1494
Cytokines 28–31 hepatitis B treatment 654 hepatitis C treatment 681 insulin resistance 1041 interferon-g 31 interleukin-6 30–1 release by stellate cells 94–5 tumor necrosis factor-a 28–30 Cytomatrix 13 Cytomegalovirus 725–6 in children 1443 graft-versus-host disease 867 histopathology 231 intrauterine infection 1368 post-transplantation 967–8, 970–1 Cytoplasmic inclusions 13 Cytoprotective agents 102 Cytosine arabinoside 535 Cytoskeleton 13 Cytotoxic T-cells 156
Dacarbazine 537 Daclizumab, graft-versus-host disease 870 Dactinomycin 536 Dalteparin 529 Danazol 517 acute intermittent porphyria 1400 Danshen 552 Dapsone 523, 524–5 DC-SIGN 135, 153 Defibrotide 901 Delavirdine 525 Demeclocyline 527 Dendritic cells 153–4 immune functions 155–6 Desipramine 517 Devil’s claw 552 Diabetes mellitus 671, 797 and hemochromatosis 1242 and hepatocellular carcinoma 1111 post-transplantation 969 Dichlorodiphenyltrichloroethane 567–8 Diclofenac 532, 533 acute intermittent porphyria 1400 Didanosine 525 Diet alcoholic hepatitis 608–9 and alcoholic liver disease 595 Wilson’s disease 1232 See also Nutrition Dietary protein restriction 321–2 Diethylstilbestrol 517 Diffuse liver disease 273–82 amyloidosis 280–2 Budd-Chiari syndrome see Budd-Chiari syndrome cirrhosis see cirrhosis iron storage disorders 275–8 sarcoidosis 225, 279–80 steatosis 278 Wilson’s disease see Wilson’s disease Diffuse malignant infiltration 389 Diflunisal 532 Dihydropyrimidine dehydrogenase 58–9 Dilevalol 505 Diltiazem 529 and granuloma 774
Dilutional hyponatremia 421–2 management 443 Dimethyl acetamide 567 Dimethylformamide 567 Dinitrophenol 564–5 Dioxane 510, 562 Dipyridamole 529 Direct transhepatic portal venous pressure 353 Disaccharidase inhibitors 322 Disopyramide 529 Disseminated intravascular coagulation 494 Distal splenorenal shunt 301 Diuretics ascites 340 hepatotoxicity 531 DNA integration 166–7 mismatch repair 170 DNA methylation, reduced 594 DNA mutagenesis 594 Dofetilide 529 Domperidone, prevention of variceal bleeding 370 Dong quai 552 Dopaminergic agents 323 Doppler ultrasound 354 Doxazosin 529 Doxorubicin, hepatocellular carcinoma 1122 Doxycycline 527 Droxican 532 Drug-induced hepatitis 774–5, 1102 Drug-induced liver injury 386–9, 503–50, 782 antimicrobial agents 522–33 antibacterials 526–8 antifungal/antiparasitic/antimalarial/antit ubercular agents 522–5 antivirals 525–6 antimycobacterial therapy 783 antineoplastic and immunosuppressive agents 533–7 alkylating agents 536–7 antimetabolites 534–6 biologic response modulators 537 antiretroviral agents 783 bioactivation of xenobiotic agents 508–11 glutathione 508–9 oxidative stress and free radical reactions 510–11 cardiovascular agents 528–31 ACE inhibitors 528–9 angiotensin II receptor blockers 529 antiarrhythmic drugs 529–30 antihypertensive drugs 531 antiplatelet/anticoagulant/thrombolytic agents 528 b-blockers 530 cholesterol-lowering agents 530–1 diuretics 531 causality assessment 505–6 clinicopathologic patterns 511–17 cholestatic pattern 514–15 hepatocellular 511–14 macrovesicular/mixed micro- and macrovesicular steatosis 516 mixed pattern 515 predictable vs unpredictable 516–17 steatosis 515–16
Index
Drug-induced liver injury—cont’d epidemiology 504–5 idiosyncratic 518 intrinsic 518 mechanisms 506–7 non-steroidal anti-inflammatory drugs 531–3 aspirin 531–2 diclofenac 533 diflunisal 532 indomethacin 532 sulindac 532–3 specific agents anesthetics 517–19 anticonvulsants 519–21 antidiabetic agents 522 anti-Parkinson’s/antimigraine/antiAlzheimer’s agents 521 trimethoprim-sulfamethoxazole 783 Drug metabolism 3 Dual porphyria 1417 Dubin-Johnson pigment 208 Dubin-Johnson syndrome 73, 80, 1472–4 animal models 1474 clinical findings 1472 genetics 1473 inheritance 1473–4 laboratory tests 1472 organic ion transport 1472–3 in pregnancy 1008 urinary coproporphyrin excretion 1474, 1475 Ductal cells 178 Ductal plate malformation hypothesis 1329–30 Ductopenia 216–17 Ductular reaction 213 Dyslipidemia 810 Dysplastic nodules 268–9
E-cadherin 26 Echinococcal disease 285–6 Echinococcus spp. 740–2 clinical presentation 741 diagnosis 741–2 prognosis 742 treatment 742 Echinococcus granulosum 285 Echinococcus multilocularis 285 Echium lycopsis 573 Ectopic varices 365 Efavirenz 525 EHBR rat 1474 Ehrlichia chaffeensis 759 Eisocanoids, vasoconstrictor 433 Electrolyte management, acute liver failure 396–7 Electrolytes, acute liver failure 392 Emtricitabine 525 hepatitis B 653 Encainide 505 Endocannabinoids 456–7 Endocarditis 470 Endocrine complications alcoholic liver disease 606 hemochromatosis 1242 Endocytosis, receptor-mediated 152 Endoplasmic reticular stress 584–5
Endoplasmic reticulum, signal peptidase 127 Endoscopic balloon dilation 1190–1 Endoscopic retrograde cholangiopancreatography 1189–90 in pediatrics 1193 in pregnancy 1193 Endothelins 357, 431–3 Endotoxin 588 and insulin resistance 1042 Endotoxin receptor 596 Endotoxin-related hepatitis 755 Energy metabolism 31–2 Enoxaparin 529 Entamoeba histolytica 283, 742–3 clinical presentation 742 diagnosis 742–3 importance and prevalence 742 prognosis 743 treatment 743 Entecavir, hepatitis B 653 Enteral stimulation 1094 Enteroviruses 1443 Envelope glycoproteins 128–9 Envelope proteins 115 Enzyme immunoassays 243, 708 Eosinophilia 797 Epidermal growth factor 27–8, 92, 1135 Epiregulin 27 Epirubicin, hepatocellular carcinoma 1122 Epistaxis 917 Epithelioid hemangioendothelioma 272–3 Eplerenone 529 Epoprostenol 529 Epstein-Barr virus 726–7 in children 1443 post-transplantation 971 Eptifibatide 529 Ergot alkaloids, acute intermittent porphyria 1400 Erythema nodosum 797 Erythrocytes, in alcoholic liver disease 605–6 Erythromycin 527 Erythropoietic protoporphyria 1414–17 clinical manifestations 1414–15 etiology and pathogenesis 1415–16 history, definition and prevalence 1414 laboratory evaluation and diagnosis 1416–17 post-transplant 998 treatment 1417 Erythropoietin 489 Escherichia coli 805 Esophageal cancer 601 Esophageal/gastric varices 349, 362–73 balloon tamponade 368 band ligation 371–2 bleeding 347–82 cause of 362–4 drug treatment 367–88 endoscopic therapy 368 non-endoscopic/non-surgical management 366–7 portal hypertensive gastropathy 364–5 prevention 369–73 pre-primary 372 primary 300–1 secondary 301–2, 372–3 risk factors 364
Esophageal/gastric varices—cont’d diagnosis 365 identification of 350 injection sclerotherapy 371 natural history and screening 365–6 predicted bleeding from other sites 365 pressure and flow measurement 353–4 Esophagus, alcoholic liver disease 601 Estrogens, and porphyria cutanea tarda 1408–9 Estrone 517 Ethambutol 523, 524 Ethchlorvynol, acute intermittent porphyria 1400 Ethionamide 523, 525 Etodolac 532 Etoposide 536 hepatocellular carcinoma 1122 Euglobulin clot lysis 494 Exercise 1051 Extracellular matrix 25–6, 90–1 cellular interactions 91–2 Extracorporeal shock-wave lithotripsy 1183–4 Extrahepatic biliary atresia 225, 1356–65 Ezetimibe 529
Factor V, acetaminophen-induced acute liver failure 405 Factor VII deficiency, liver cell therapy 183 Falciparum malaria 389 False neurotransmitter hypothesis of hepatic encephalopathy 313–14 Familial hypercholanemia 79 Familial hypercholesterolemia, liver cell therapy 183 Fas-associated death domain 28 Fasciola hepaticum 737 Fas receptor-mediated apoptosis 40–2 internal regulators 42 intrinsic (mitochondrial) pathway 44 Fatigue 809–10 hemochromatosis 1243 Fatigue syndrome 672 Fat-soluble vitamin deficiency 811 Fatty liver disease 217 acute fatty liver of pregnancy 1014–17 clinical features 1015–16 etiology 1014–15 incidence 1014 natural history and prognosis 1016–17 pathology 1016 therapy 1017 alcohol history 1043 clinical estimation of insulin resistance 1046–8 clinical features 1043 diagnosis 1042–3 drugs causing amiodarone 1042 tamoxifen 1042 examination findings 1044 histology 217–18 imaging studies 1045–6 laboratory tests 1044 liver biopsy 1044–5
1495
Index
Fatty liver disease—cont’d pathogenesis 1035–42 accumulation of triglyceride in hepatocytes 1035–8 insulin resistance 1038–42 post-transplant 999 prevalence 1033–4 in children 1034 progression 1034–5 risk factors 1033 symptoms 1043 see also Non-alcoholic steatohepatitis Felbamate 520 Fenbufen 532 Fenclofenac 532 Fenclofenamic acid 532 Fenclozic acid 532 Fenestrae 153 Fenofibrate 529 Fenoldapam 529 Fenoprofen 532 Fentiazac 532 Ferritin, serum levels 1244–5 Ferroportin disease 1248 Fetal alcohol syndrome 605 Fetal stem cells 179 Feverfew 552 Fever of unknown origin 770 Fibrin 490 Fibrinolysis 494 Fibroblast growth factor 92 Fibrocystic disease 1329–53 autosomal dominant polycystic kidney disease 1332–3 events contributing to 1334–5 formation of mechanosensory complex 1334 histopathology 1336–7, 1338 moderation of transcription factor activities 1334 PKD1 and polycystin-1 1332–3 PKD2 and polycystin-2 1333 autosomal recessive polycystic kidney disease 1336 histopathology 1337 biology and pathobiology 1329–36 Caroli’s disease 1348 characteristics 1347 clinical features 1347 diagnosis 1347 histopathology 1337, 1341 treatment 1347 choledochal cysts 1347–8, 1349 characteristics 1347 clinical features 1347 diagnosis 1347–8 histopathology 1337, 1341 treatment 1348 congenital hepatic fibrosis 1346–7 characteristics 1346 clinical features 1346 histopathology 1337, 1339, 1340 treatment 1346–7 cystogenesis and primary cilia 1331–2 ductal plate malformation hypothesis 1329–30 genes and proteins 1332 genetics 1330–1
1496
Fibrocystic disease—cont’d histopathology 1336–42 polycystic liver disease 1335–6, 1342–6 clinical features 1343 complications 1345 gastrointestinal symptoms 1344 mechanism of 1336 molecular diagnostics 1342 natural history 1342 PRKCSH and hepatocystin 1335–6 Sec63 and Sec63p 1336 treatment cyst fenestration 1344–5 liver resection 1345–6 liver transplantation 1346 medical 1343–4 radiological cyst aspiration and sclerosis 1344 solitary hepatic cysts 1348, 1350 characteristics 1348, 1350 histopathology 1341–2 treatment 1350 Fibrocystin 1336 Fibrolamellar carcinoma 229, 269–71, 1114 see also Hepatocellular carcinoma Fibro-obliterative cholangitis 215 Fibrosing cholestatic hepatitis 231 Fibrosis 120 FibroTest 241, 242 Fine-needle aspiration, ultrasound-guided 199, 201 Flaviviruses 125, 126 Flecanide 529 FLICE-inhibitory protein 42 Flosequinan 505 Floxuridine 535 Fluconazole 523 Flucytosine 523 Fluid management, acute liver failure 396–7 Flumazenil 322–3 5-Fluorouracil 535 hepatocellular carcinoma 1122 Fluoxetine, acute intermittent porphyria 1400 Flurbiprofen 532 Focal fat 1046 Focal malignant lesions biliary cystadenocarcinoma 271–2 cholangiocarcinoma 271 epithelioid hemangioendothelioma 272–3 fibrolamellar carcinoma 269–71 hepatic angiosarcoma 272 hepatocellular carcinoma 266–9 liver metastases 264–6 lymphoma 272 Focal nodular hyperplasia 227, 252–4, 362, 1152–9 associated conditions 1158 clinical features 1152 diagnosis 1152–3 diagnostic workup 1153–8 epidemiology 1152 imaging 1153 pathogenesis 1152 pathology 1152 in pregnancy 1010 prognosis and natural history 1158–9 treatment 1158
Focal sparing 1046 Folate 491 Fondaparinux 529 Forns index 241 Fosamprenavir 525 Foscarnet cytomegalovirus 726 human herpes virus-6 728 Fosfomycin 527, 528 Foxa transcription factor 178 Free fatty acid toxicity 1041 Free hepatic vein pressure 298 Free radicals 510–11 Frizzled-7 26, 170–3 Fructose metabolism disorders 1270–5 fructose-1,6-diphosphatase deficiency 1269–70 fructose diphosphatase deficiency 1275 fructose phosphate aldolase deficiency 1269–70 hepatic enzyme elevation 1273 hereditary fructose intolerance 1270–3 hyperuricemia and increased urate excretion 1274–5 hypoglycemia 1273 hypophosphatemia 1273 Fulminant hepatic failure 629–30, 709 Wilson’s disease 1233 Fungal hepatitis 760–1 coccidioido- and paracoccidioidomycoses 760 cryptococcosis 760 histoplasmosis 760 invasive fungal infections 761 Pneumocystis spp. 760 Fungal infections abscess 283–5 graft-versus-host disease 868 in HIV patients 786–7 post-transplantation 967, 970 see also Fungal hepatitis Fusarium 761
GABA/benzodiazepine hypothesis of hepatic encephalopathy 314 Gabapentin, acute intermittent porphyria 1400 Galactose metabolism disorders 1275–81 biochemical characteristics and pathogenesis brain 1278 gonads 1278 hypoglycemia 1278 kidneys 1278 lenticular changes 1278 liver 1278 diagnosis 1279–80 diagnostic screening 1279 galactokinase-deficiency galactosemia 1281 genetics 1279 pathology 1278–9 prognosis 1280–1 transferase-deficiency galactosemia 1275–7 treatment 1280 Galactosemia 1269–70 Duarte variant 1279–80 galactokinase-deficiency 1281
Index
Galactosemia—cont’d Indiana variant 1280 Los Angeles variant 1280 Negro variant 1280 Rennes variant 1280 transferase-deficiency 1275–7 uridine disphosphate galactose-4epimerase-deficiency 1281 Gallbladder 69, 1101–11 acalculous disease 1182 bile duct strictures 1211 biloma 1210–11 carcinoma 841–2 cholecystectomy 1183 after drainage 1204 bile duct injury 1209–10 in cirrhosis 1211–12 complications 1193–4 indications 1201–3 acute cholecystitis 1202 cholangitis 1202–3 laparoscopic 1207–9 in pregnancy 1212–13 cholelithiasis 1187, 1201 complications 1202 surgical decompression 1203–4 symptoms 1202 gallstone pancreatitis 1204–7 gallstone ileus 1206–7 mild/moderate 1204–5 Mirizzi syndrome 1205–6 severe 1205 in HIV patients 789 percutaneous transhepatic biliary decompression 1203 polyps 1184 see also Gallstones Gallstone ileus 1206–7 Gallstone pancreatitis 1204–7 gallstone ileus 1206–7 mild/moderate 1204–5 Mirizzi syndrome 1205–6 severe 1205 Gallstones 1181–6 clinical presentation 1182 diagnosis 1182 imaging 291–2 management cholecystectomy 1183 expectant management and preventive measures 1183 extracorporeal shock-wave lithotripsy 1183–4 oral bile acid dissolution therapy 1183 topical dissolution therapy 1184 microlithiasis 1184 natural history 1202 non-surgical management 1187–99 biliary leaks 1194 choledocholithiasis 1188–94 cholelithiasis 1187 duct injuries 1194–6 percutaneous cholecystostomy 1187–8 pathogenesis and risk factors 1181–2 acalculous gallbladder disease 1182 cholesterol gallstones 1181 pigment stones 1181 pediatric 1367
Gallstones—cont’d in pregnancy 1009 recurrence 1184 see also Gallbladder g-Globulins 245 g-Glutamyl transferase 235, 238 Ganciclovir cytomegalovirus 726 human herpes virus-6 728 Gangliosidosis 1270 Garlic 552 Gastric antral vascular ectasia 365 Gastric varices see Esophageal/gastric varices Gastrointestinal bleeding 494 Gastrointestinal complications acute liver failure 393 alcoholic liver disease 600–2 colon 602 esophagus 601 pancreas 602 salivary glands and oropharynx 600–1 small intestine 601–2 stomach 601 graft-versus-host disease 86 Osler-Weber-Rendu syndrome 918–19 parenteral nutrition-induced liver disease 1091 Gastrointestinal hormones 1094 Gaucher’s disease 1270 Gemcitabine, hepatocellular carcinoma 1122 Gemfibrazole 529 Gemtuzumab ozogamicin, and sinusoidal obstruction syndrome 897 Gender, and risk of alcoholic liver disease 595 Gene therapy Crigler-Najjar syndrome 1469 Wilson’s disease 1232 Genetics Alagille’s syndrome 1373–4 alcoholic liver disease 595–6 BRIC 1376–7 cholesterol ester storage disease 1302 cystic fibrosis 1314 Dubin-Johnson syndrome 1473 fibrocystic disease 1330–1 galactose metabolism disorders 1279 Gilbert’s syndrome 1470 hemochromatosis 1246–8 hyperuricemia 1274 PFIC1 1375 PFIC2 1377 PFIC3 1378 primary biliary cirrhosis 804 primary sclerosing cholangitis 826–7 THCA syndrome 1311 Wolman’s disease 1301 Zellweger’s syndrome 1310 see also Pharmacogenetics Genotype 56, 243 Gentamicin 517 acute intermittent porphyria 1400 Germander 552, 553, 555 Giant cell hepatitis 225 post-transplant 998 Gilbert’s syndrome 238, 1100, 1469–71 animal model 1471 bilirubin conjugates in bile 1471
Gilbert’s syndrome—cont’d clinical features 1469–70 diagnosis 1471 effect of fasting 1470–1 effect of nicotinic acid administration 1471 genetics 1470 organic anion transport 1470 Gingko 552 Ginseng 552 Glafenine 532 Glasgow Alcoholic Hepatitis Score 600 Glasgow Coma Scale 317 Glatiramer 521 Gliburide, and granuloma 774 Glisson’s capsule 4 Glomerulonephritis 643–4 Glossitis 600–1 a-1,4 Glucan-6-glycosyl transferase deficiency 1291–3 biochemical characteristics 1292 clinical characteristics 1291–2 laboratory findings 1292 molecular basis 1293 pathology 1292–3 treatment 1293 Glucocorticoids 517 acute intermittent porphyria 1400 Glucose-6-phosphatase deficiency 1282–9 biochemical characteristics 1283–7 blood glucose changes 1283 glucose production 1286–7 hepatic adenomas and carcinomas 1285–6 hyperlipidemia 1283–4 hyperuricemia 1284–5 hypophosphatemia 1285 lactic acid changes 1283 platelet dysfunction 1285 severity of illness 1286 classification 1282 clinical characteristics 1283 diagnosis 1287–8 molecular basis 1282–3 pathology 1287 prognosis 1288–9 Glucose homeostasis 3 GLUS syndrome 775 Glutamate dehydrogenase 240 Glutathione 508–10 Glutathione S-transferases 61 Glutethimide, acute intermittent porphyria 1400 Glyceraldehyde-3-phosphate dehydrogenase 113 GlycoCirrhoTest 241 Glycogen granules 13 Glycogen nuclei 221 Glycogen storage disease 221, 1269–70 liver cell therapy 183 type IB 1289 type I (glucose-6-phosphatase deficiency) 1282–9 biochemical characteristics 1283–7 classification 1282 clinical characteristics 1283 diagnosis 1287–8 molecular basis 1282–3 pathology 1287 prognosis 1288–9
1497
Index
Glycogen storage disease—cont’d type III (amylo-1,6-glucosidase deficiency) 1289–91 biochemical characteristics and laboratory findings 1290 clinical characteristics 1290 molecular basis 1289–90 pathology 1290 treatment 1290–1 type IV (a-1,4 glucan-6-glycosyl transferase deficiency) 1291–3 biochemical characteristics 1292 clinical characteristics 1291–2 laboratory findings 1292 molecular basis 1293 pathology 1292–3 treatment 1293 Graft-versus-host disease 863–74, 868–70 acute 865, 868–70 chronic 865, 870 clinical manifestations 864–6 gastrointestinal tract 866 hematopoietic system 866 liver involvement 865–6 skin 866 differential diagnosis 867 early hepatic complications 867–8 bacterial infection 868 biliary disease 867 cytomegalovirus infection 867 fungal infections 868 hepatitis B infection 867 hepatitis C infection 867 hepatotoxic agents 867 mycobacterial infection 868 veno-occlusive disease 867 epidemiology 863 histopathology 865–6 immunopathogenesis 863–4 late post-transplantation liver complications 868 liver transplantation 870 prognosis 870 staging 866 see also Liver transplantation Granuloma 767–80 causes 770–5 atypical mycobacteria 772 bacille Calmette-Guérin 772 brucellosis 773–4 cat-scratch disease 774 drug-induced hepatitis 774–5 granulomatous hepatitis 775 Q fever 772–3 sarcoidosis 770–1 schistosomiasis 771–2 tuberculosis 772 definition 767 diseases associated with 769–70 fever of unknown origin 770 hepatitis C 769–70 human immunodeficiency virus 770, 782 primary biliary cirrhosis 770 histopathology 768 immunology 767–8
1498
Granuloma—cont’d incidental 775–6 leprosy 775–6 lymphoma 775 lipogranulomas 768–9 Granulomatous hepatitis 775 Greater celandine 553, 555–6 Griseofulvin 523 acute intermittent porphyria 1400 Ground-glass hepatocytes 219, 644 Growth hormone therapy 1385 Guanabenz 529 Guanethidine, acute intermittent porphyria 1400 Guillain-Barré syndrome 710 Gunn rat 1453, 1455, 1467–8
HAART therapy 655 Halogenated aliphatic hydrocarbons 566–7 Halogenated aromatic hydrocarbons 563–4 Halothane, and granuloma 774 Hamartoma biliary 258 mesenchymal 226, 261 Haplotype 54 HBV see Hepatitis B Heart disease 1065–70 constrictive pericarditis 1069 left heart failure and ischemic hepatitis 1067–9 portal hypertension 362 right heart failure (congestive hepatopathy) 1065–7 Heat-shock proteins 116 Helicase inhibitors 681–3 Helicases 130 Helicobacter pylori 601, 805 Heliotropium spp. 573 HELLP syndrome 1018–19 Hemangioma 257–61, 1147–52 clinical features 1147 diagnosis 1147–8 diagnostic workup 1150–1 epidemiology 1147 imaging 1148–50 pathogenesis 1147 in pregnancy 1010 prognosis and natural history 1151–2 treatment 1151 Hematological complications acute liver failure 392 alcoholic liver disease 605–6 coagulation 606 erythrocytes 605–6 leukocytes 606 platelets 606 Hematologic disease 1079–83 Hodgkin’s lymphoma 1080–1 mastocytosis 1082–3 multiple myeloma 1081–2 non-Hodgkin’s lymphoma 1081 sickle-cell anemia 1079 biliary tract disease 1079–80 hepatic crisis 1079 hepatic histology 1080 viral hepatitis 1080
Hematopoiesis abnormalities 489–99 clinical features and complications 491–4 anemia 491–3 coagulopathy 493–4 disseminated intravascular coagulation 494 fibrinolysis 494 gastrointestinal bleeding 494 thrombocytopenia and hypersplenism 493 thrombotic disorders 493 epidemiology and pathogenesis 489–91 coagulation cascade 489–91 hematopoiesis 489 platelet plug formation 489–91 graft-versus-host disease 86 treatment 494–5 transfusion of blood products 494–5 Hematoxylin and eosin 205, 206, 207 Heme oxygenase 457 Heme pathway intermediates 1394 Heme synthesis 1391–3 Hemihepatic vascular clamping 1169 Hemin 1402 Hemochromatosis 168, 221, 222, 1239–56 aceruloplasminemia 1248 clinical features 1240–3 arthropathy 1242 cardiac disease 1242 diabetes mellitus 1242 endocrine abnormalities 1242 fatigue 1243 infections 1242–3 liver disease 1241–2 skin pigmentation 1242 diagnosis 1243–6 hepatic iron concentration/iron index 1246 liver biopsy 1246 non-HFE hemochromatosis 1251 removal of iron by venesection 1245–6 serum ferritin 1244–5 transferrin saturation 1243 underdiagnosis 1243 unsaturated iron-binding capacity 1243–4 epidemiology 1239 family studies 1249–51 ferroportin disease 1248 genetic testing 1246–8 genetic testing for non-HFE iron overload 1249 and hepatocellular carcinoma 1111 history 1239 imaging studies 1246 juvenile 1248 laboratory testing 244 non-expressing homozygotes 1249 pathogenesis 1239–40 population screening 1252–3 post-transplant 998 prognosis 1253–4 secondary iron overload 1253 diagnosis 1251 transferrin receptor-2 mutation 1248–9 treatment HFE-linked hemochromatosis 1251–2 liver transplanation 1252
Index
Hemodialysis patients, hepatitis E in 701 Hemolysis 238 Hemolytic anemia 492 Hemophilia A, liver cell therapy 183 Hemorrhagic fevers 389 Hemosiderin 208 Hemosiderosis 221 Hemostasis, in liver biopsy 201 Hepadnaviruses 111 and hepatocellular carcinoma 167–8 intracellular conversion pathway 113 Heparin, sinusoidal obstruction syndrome 901 Heparin-binding epidermal growth factor 27 Hepatectomy see Liver resection Hepatic acinus 9 Hepatic angiosarcoma 272 Hepatic arterioles 220 Hepatic artery 218–19 flow measurement 354 increased flow 355 Hepatic artery thrombosis 966 Hepatic (biliary) cysts 255–6, 1364 Hepatic circulation, cirrhosis 425 Hepatic encephalopathy 311–31 and acute liver failure 393–4 ammonia hypothesis original 312–13 unifying 313 classification 312 closure of portosystemic shunts 324 as complication of TIPS 300 definition and nomenclature 311 diagnosis 314–18 blood ammonia testing 318 brain imaging 318 clinical evaluation 317 electrophysiological status 317 generalized motor abnormalities 317 mental status changes 316–17 false neurotransmitter hypothesis 313–14 GABA/benzodiazepine hypothesis 314 general pathophysiology 311–12 liver support systems 324 liver transplantation 324–5 management 401 pathogenesis 312 plasma amino acid hypothesis 313–14 precipitating factors 315 synergistic neurotoxin hypothesis 313 therapy-resistant 323–4 treatment 318–23 correction of neurotransmitter abnormalities 322–3 dietary protein restriction 321–2 disaccharidase inhibitors 322 general issues 319–20 historical review 319 probiotics 322 promotion of waste nitrogen excretion 322 reduction of ammonia production/absorption 320–1 Hepatic enzyme elevation 1273 Hepatic fibrosis 87–109 bedside diagnostic tools 96 congenital 361
Hepatic fibrosis—cont’d measurement of 96–9 natural history 88–90 and non-alcoholic steatohepatitis 1042 non-invasive markers 96–9 blood-based markers 96–8 imaging tests 98–9 liver biopsy 99 liver function tests 99 pathophysiology 90–5 disease-specific mechanisms 95 extracellular matrix 90–1 extracellular matrix-cell interactions 91–2 hepatic stellate cell activation 92–5 resolution 95 predictors of 1036, 1048 reversibility 90 risk factors 88–90 alcoholic liver disease 89 hepatitis B 89 hepatitis C 89 non-alcoholic steatohepatitis 89–90 treatment 99–103 anti-inflammatory compounds 100–1 antioxidant agents 101–2 cytoprotective agents 102 fibrogenesis inhibitors 102–3 stellate cell-specific compounds 102 therapies directed at underlying disease 100 Hepatic functional units 9–10 Hepatic granulomas, histopathology 224–31 Hepatic hemangioma see Hemangioma Hepatic hemorrhage 1019–20 Hepatic hydrothorax 336 TIPS 303 treatment 342–3 Hepatic infarct 218 Hepatic infections amebic abscess 283 bacillary peliosis 286 echinococcal disease 285–6 fungal abscess 283–5 pyogenic liver abscess 282–3 schistosomiasis 286 viral hepatitis 286 Hepatic iron buffer 13 Hepatic iron concentration 1246 Hepatic iron index 1246 Hepatic microenvironment 155 Hepatic mitogens, alcoholic hepatitis 609 Hepatic parenchymal cells 10–13 cytoplasmic inclusions 13 cytoskeleton and cytomatrix 13 endoplasmic reticulum, ribosomes and Golgi apparatus 11–12 lysosomes 12 mitochondria 12 nucleus 11 peroxisomes (microbodies) 12 plasma membrane 10–11 Hepatic pedicle, clamping of 1169–70 Hepatic rupture 1019–20 Hepatic sinusoidal dilatation 1010
Hepatic transport proteins 70–5 basolateral transport proteins 71 canalicular transport proteins 72–5 ABC-transporter proteins 73–5 basolateral anionic conjugate and bile salt transporters 73 bile salt excretory pump 73 MDR1 P-glycoprotein 72 multidrug resistance protein MRP2 73 phospholipid transporter 72 Hepatic vein 219 catheterization 351–2 thrombosis post-transplantation 966 visualization of 350 Hepatic venous pressure gradient 297, 336, 351, 905 Hepatic venules 5 Hepatitis 286 acute, histopathology 208–9 autoimmune 210, 211 laboratory testing 244, 245 bacterial and fungal 755–9 chronic histology 209–12 scoring 210, 212 chronic active 1032 drug-induced 774–5, 1102 endotoxin and inflammatory-related 755 fungal see Fungal hepatitis granulomatous 775 ischemic 219, 236, 1067–9 causes and pathogenesis 1067–8, 1102 clinical syndrome 1068–9 differential diagnosis 1069 treatment and prognosis 1069 pediatric viral see Pediatric viral hepatitis postoperative 1102 in pregnancy 1005–6 and sickle-cell anemia 1080 see also different types Hepatitis A 627 acute liver failure 385 age-specific mortality 629 in children 1433–5 clinical features 1434 diagnosis 1434 differential diagnosis 1434 epidemiology 1433 exclusion from daily activities 1435 immunoprophylaxis 1434–5 pathogenesis 1434 prognosis and natural history 1434 treatment 1434 clinical features and diagnosis 629–31 biochemical abnormalities and serology 630–1 clinical symptoms 630 fulminant hepatic failure 629–30 epidemiology 628–9 high-risk groups 631 HIV patients 783 immunization 631–2 pathogenesis 629 prevention and treatment 631–2 vaccination 1054 vertical transmission 1005 virology 627–8 Hepatitis-associated anemia 492
1499
Index
Hepatitis B 89, 111, 166–8, 635–63 acute 642 acute flares 644–6 antiviral therapy-induced 645–6 immunosuppressive therapy 645 spontaneous 645 acute liver failure 385 antiviral therapy 650–6 agents under development 653–4 choice of agents 650 entecavir 653 guidelines 650–1 HIV/HBV coinfected patients 655 innovative approaches 655–6 interferon-a 651–2 novel therapies 654–5 nucleoside analogs 652–3 virologic endpoints 650 assembly/release of virion 115–16 attachment, penetration and entry 112–13 cell lines and model systems 111–12 in children 1435–9 associated conditions 1436 clinical features 1435–6 diagnosis 1436 differential diagnosis 1436 disease complications 1436–7 epidemiology 1435 immunoprophylaxis 1439 pathogenesis 1435 prognosis and natural history 1437 treatment 1437–9 interferon-a 1438 lamivudine 1438–9 chronic 642–4 pathology 644 clinical outcome 637 core protein 111, 114 definitions 637 epidemiology 635–6 geographical distribution and sources of exposure 635–6 rates of infection in US 636 extrahepatic manifestations 643–4 arthropathy 643 cutaneous disorders 643 glomerulonephritis 643–4 polyarteritis nodosa 643 genome mutation 640–1 DNA polymerase 640–1 HBsAg mutants 640 precore, basal core promoter and core genes 640 genome variation 639–40 genomic replication 115 genotypes 639–40 global picture 635 graft-versus-host disease 867 hepatocellular carcinoma 166–8, 1110 chronic hepadnavirus infection 167–8 DNA integration 166–7 trans-activation of cellular genes 167 HIV patients 655, 783–4 incidence 783–4 natural history 784 prevalene 783 prevention 784 therapy 784
1500
Hepatitis B—cont’d immunization 656–8 escape mutants 658 high-risk groups 657 postexposure/perinatal prophylaxis 658 recent developments 658 vaccination schedule 657–8 immunopathogenesis 116–20 acute infection 116–18 anti-HBV therapy and viral clearance 120 chronic infection 118–19 disease mechanisms 119–20 HBV persistence 119 occult infection 118 laboratory testing 243–4 life cycle 113 molecular biology 637–41 natural history 641–2 occult 729 post-transplant 975–81 coinfection with hepatitis D 981 indications for liver transplantation 975 living related liver transplantation 981 natural history 975–6 nucleoside analogue resistance 980–1 pathogenesis 976 prevention of graft reinfection 976–80 hepatitis B immunoglobulins 976–7 long-term prophylaxis 979–80 oral antivirals 977–9 prevention and treatment of de novo infection 981 retransplantation 981 treatment of graft disease 980 in pregnancy 1007 serologic markers 646–50 splice protein 116 splicing 116 surface antigens 111, 112 transcription/translation of viral proteins 114–15 envelope proteins 115 hepatitis B core protein 114 hepatitis Be antigen 114 hepatitis B polymerase 114–15 hepatitis Bx protein 115 transport of viral genome to nucleus 113–14 vertical transmission 1005 viral genome 112 virus life cycle 638 Hepatitis B core antigen 211 Hepatitis Be antigen 114, 119 Hepatitis B immunoglobulins 976–7 Hepatitis B polymerase 114–15 Hepatitis B surface antigen 211 post-transplantation 979–80 Hepatitis Bx protein 115, 119 Hepatitis C 89, 125–47, 665–86 acute infection 670 acute liver failure 386 antiviral therapy and pretreatment assessments 674–5 clinical assessment 674 laboratory tests 674 liver biopsy 674–5
Hepatitis C—cont’d in children 1439–42 associated conditions 1440–1 clinical features 1440 differential diagnosis 1440 epidemiology 1439–40 exclusion from daily activities 1442 pathogenesis 1440 prognosis and natural history 1441 treatment 1441–2 clinical presentation 669–70 core antigen 243 diagnosis 673 diversity 985 epidemiology 665–6 extrahepatic manifestations 670–2 arthralgias 671 autoimmune hepatitis and HCV infection 672 diabetes mellitus 671 fatigue syndrome 672 lichen planus 671 membranoproliferative glomerulonephritis 670 mixed cryoglobulinemia 670 non-Hodgkin’s lymphoma 671–2 porphyria cutanea tarda 671 thyroid disease 671 and fatty liver disease 1049 future therapeutic approaches optimized use of interferon and ribavirin 680–1 polymerase, helicase and protease inhibitors 681–3 ribavirin analogs 681 therapeutic vaccines 683 type 1 interferons, interferon inducers, cytokines and growth factors 681 unmet treatment needs 679–80 genetic organization 126–7 5¢ and 3¢ non-coding regions 126–7 cis-acting replication elements 127 genetic variability 125 genotype 985 graft-versus-host disease 867 granuloma formation 769–70 hepatic fibrosis 95 hepatocellular carcinoma 168, 1110 HIV patients 677–8, 784–5 epidemiology 784 natural history 784–5 therapy 785 immunopathogenesis 672–3 laboratory testing 243 model systems 133–5 in vitro 133–5 in vivo 135 modes of transmission 666–8 natural history 668–9 non-structural proteins 129–33 NS2–3 protease 129–30 NS3–4A complex 130 NS4B 131 NS5A 131–2 NS5B 132–3 pathogenesis 139–40 polyprotein processing 127–8
Index
Hepatitis C—cont’d and porphyria cutanea tarda 1408 and portal hypertension 361 post-transplant 981–90 disease severity and/or progression coinfection with other viruses 985 donor-related variables 985 host-related variables 983–5 viral-related variables 985 histologic changes 986 indication for transplantation 981–2 kinetics and pathogenesis 985–6 live donor transplantation 986 natural history 982–3 treatment 986–90 alternative approaches 989–90 prevention of reinfection 986–8 recurrent disease 988–9 retransplantation 990 in pregnancy 1007–8 replication cycle 135–8 receptor candidates 135–7 replication complex 137–8 RNA assays 673–4 RNA levels 985 select patient populations 676–9 African-Americans 677 compensated cirrhosis 678 complications of cirrhosis 678–9 HIV coinfection 677–8 renal disease 679 serological assays 673 sources of infection 666 structural proteins 128–9 ARFP/F protein 128 core 128 envelope glycoproteins 128–9 p7 129 virion structure 129 taxonomy 125–6 therapeutic strategies 138–9 treatment 675–6 non-responders to previous treatment 676 treatment-naive patients 675–6 vaccine 139 vertical transmission 1005 Hepatitis D 687–92 acute liver failure 386 in children 1442–3 clinical course 690 diagnosis 689 epidemiology 688–9 HIV patients 784 laboratory testing 244 liver transplantation in hepatitis Bcoinfected patients 981 natural history 689–90 prevention and therapy 690–1 transmission 687–8 vertical 1005 virology 687 Hepatitis delta virus see Hepatitis D Hepatitis E 693–723 acute liver failure 386 antibodies 699, 700 in children 1443
Hepatitis E—cont’d clinical features 706–7 acute icteric hepatitis 706–7 acute infection 706 anicteric hepatitis and asymptomatic infection 707 prolonged cholestatic hepatitis 707 complications 709–10 acute superinfection in patients with cirrhosis 709 mortality and fulminant hepatic failure 709 pregnant women 709–10 epidemiology 696–702 incidence and prevalence 697–8 transmission 698–700 high-risk populations 700–1 chronic autoimmune liver disease 701 hemodialysis patients 701 HIV-infected persons 701 persons having contact with swine 700 persons having contact with untreated waste water 700–1 HIV patients 701 immune response 705–6 cellular 705–6 humoral 705 laboratory diagnosis 707–9 areas of high prevalence 708 areas of low prevalence 708 differential diagnosis 709 enzyme immunoassays 708 immune electronic microscopy and immunofluorescent microscopy 707 serological assays 707 virus or viral component detection 707 natural history 709 pathogenesis 702–5 course of infection in animal studies 704–5 course of infection in human studies 704 incubation period 702 viral replication 702–4 pathology 706 prevention 710–14 active immunoprophylaxis 711–14 general measures 710–11 passive immunoprophylaxis 711 perinatal and congenital infection 711 treatment 710 virology 693–6 classification 693 genome organization 694–5 genotypes 695–6 morphology 693–4 physicochemical characteristics 693 as zoonosis 701–2 animal strains 701 cross-species infection 701–2 Hepatitis flares 119–20 Hepatitis G 729 Hepatobiliary ascariasis 737–40 clinical presentation 739 diagnostic testing 739–40 importance and prevalence 737–9 treatment 740 Hepatobiliary excretion, failure of 390 Hepatobiliary function, immaturity of 1090 Hepatobiliary transport, acquired defects 1471–2
Hepatocellular adenoma 251–2, 1159–62 associated conditions 1161 clinical features 1159 diagnosis 1159 diagnostic workup 1161 epidemiology 1159 imaging 1159–61 pathogenesis 1159 pathology 1159 in pregnancy 1010 prognosis and natural history 1162 treatment 1161–2 Hepatocellular ATP production, impaired 1041 Hepatocellular autoantigens 796 Hepatocellular carcinoma 165–75, 266–9, 1109–31 and alcoholic liver disease 593–4 activation of xenobiotics 594 immunosuppression 594 lipid peroxidation and DNA mutagenesis 594 reduced DNA methylation 594 tumor necrosis-a-induced survival factors 594 angiogenesis 170 chromosomal abnormalities 1112 clinical manifestations 1114–15 diagnosis and staging 1115–16 DNA mismatch repair 170 epidemiology 1109 hepatic disorders predisposing to 948 hepatitis B 166–8 chronic hepadnavirus infection 167–8 DNA integration 166–7 trans-activation of cellular genes 167 hepatitis C 168 histology 229 imaging 266–9 incidence 1110 liver transplantation 942–3 metabolic liver diseases 168–9 and non-alcoholic steatohepatitis 1055–6 p53 tumor suppressor gene and aflatoxins 169–70 pathogenesis 1112–13 pathology 1113–14 post-transplantation 971 in pregnancy 1010 prevention 1125–6 primary biliary cirrhosis 811–12 primary sclerosing cholangitis 842 prognosis 1116–18 risk factors 1109–11 aflatoxin 1111 alcohol 1110–11 a-1-antitrypsin deficiency 1111 autoimmune hepatitis 1111 cirrhosis 1111 diabetes mellitus 1111 hepatitis B 1110 hepatitis C 1110 hormonal compounds 1111 iron and copper deposition 1111 non-alcoholic steatohepatitis 1111 obesity 1111 primary biliary cirrhosis 1111 tobacco 1111
1501
Index
Hepatocellular carcinoma—cont’d signal transduction pathways 170–3 telomerase activation 170 treatment 1118–25 palliative 1122–5 percutaneous 1121–2 surgical 1119–21 tumor suppressor genes 169 Hepatocellular injury 511–14, 1040 Hepatocellular lesions focal nodular hyperplasia 252–4 hepatocellular adenoma 251–2 nodular regenerative hyperplasia 255 Hepatocellular triglyceride accumulation 1038 Hepatocystin 1335–6 Hepatocyte growth factor 26–7, 92 Hepatocytes 154–5 bilirubin storage in 1459 copper transport/homeostasis 1223–4 Fas receptor-mediated apoptosis 40–2 ground-glass 219, 644 immune functions 155 inhibition of proliferation 32–3 size of 357 transition into mitosis 25 triglyceride accumulation in 1035–8 Hepatocyte transplantation see Stem cells Hepatoerythropoietic porphyria 1411 Hepatopulmonary syndrome 477–83 clinical features 478–9 definition 477 diagnosis 479–81 epidemiology 477–8 pathology and pathogenesis 478 prognosis and natural history 482–3 therapy 481–2 TIPS 304 Hepatorenal syndrome 422–4 alcoholic hepatitis 610 clinical and laboratory findings 422–4 definition 422 development of type 1 437–8 type 2 435–6 diagnosis 424 pathogenic mechanisms 422 precipitating factors 424 prevention of 446 treatment 443 Hepatorenal tyrosinemia see Hereditary tyrosinemia Hepatotoxic agents, and graft-versus-host disease 867 Hepatotrophic viruses 385–6 hepatitis A 385 hepatitis B 385 hepatitis D 386 hepatitis C 386 hepatitis E 386 systemic viral infection 386 Herbal preparations, hepatotoxicity 551–60 Atractylis gummifera 552–3 black cohosh 553 Callilepsis laureola 552–3 camphor oil 553–4 cascara 554 chaparral 554
1502
Herbal preparations, hepatotoxicity—cont’d Chaso and Onshido 554–5 comfrey and pyrrolizidine alkaloids 555 germander 555 greater celandine 555–6 herbal formulation 552 herb-drug interactions 552 Ju Bu Huan 556 kava 556 ma-huang 556–7 margosa oil 557 mistletoe, skullcap and valerian 557 pennyroyal 558 types of liver injury 552 Hereditary cholestasis with lymphedema see Aagenaes’s syndrome Hereditary coproporphyria 1412–13 clinical manifestations 1412 etiology and pathogenesis 1412 history, definition and prevalence 1412 laboratory evaluation and diagnosis 1412–13 treatment 1413 Hereditary fructose intolerance 168, 1270–3 biochemical characteristics 1272–3 clinical features 1271–2 laboratory features 1273 molecular basis 1270–1 Hereditary hemorrhatic telangiectasia see Osler-Weber-Rendu disease Hereditary lymphedema with recurrent cholestasis 80, 1270, 1306 Hereditary tyrosinemia 1296–8 biochemical features 1296–7 clinical features 1296 laboratory findings 1296 molecular basis 1297–8 Herpes simplex virus in children 1443 hepatitis 727 in pregnancy 1009 intrauterine infection 1368–9 High-density lipoprotein 243 Histology 205–34 bile duct integrity 207–8 guidelines for interpretation 205–7 pertinent negatives 208 pigments 208 special stains and immunohistochemistry 205 zonal changes 206–7 Histopathology biliary disease and cholestasis 212–13 cholestatic parenchymal/extrahepatic diseases 217 cirrhosis 222–4 extrahepatic inflow and outflow vascular disorders 218–19 fatty liver diseases 217–18 hepatic granulomas 224–31 hepatitis acute 208–9 chronic 209–12 intrahepatic bile duct diseases 216–17 intrahepatic vascular disorders 220 large bile duct obstructive diseases 213–16 metabolic and metal storage diseases 220–2
Histopathology—cont’d right ventricular failure, cor pulmonale and constrictive pericarditis 219–20 vascular diseases 218 Histoplasma capsulatum 760 Histoplasmosis 760 HIV anti-HIV agents 525 antiretroviral agents 783 antimycobacterial therapy 783 ascites 787–8 bacterial infections 787 biliary tract abnormalities 788–9 as contraindication to liver transplantation 937 drug-induced 782–3 fungal diseases 786–7 gallbladder disease 789 granuloma formation 770, 782 hepatitis A 783 hepatitis B 655, 783–4 hepatitis C 677–8, 784–5 hepatitis D 784 hepatitis E 701 hepatobiliary disease causes 781 diagnosis 789–90 pathologic findings 781–2 mycobacterial infections 786 neoplasms 787 protozoal infections 787 trimethoprim-sulfamethoxazole 783 HMG-CoA reductase 62–3 HMG-CoA reductase inhibitors 1054 HMG-CoA reductase inhibitors see Statins Hodgkin’s lymphoma 1080–1 Hormones, and hepatocellular carcinoma 1111 Human embryonic stem cells 177–8 Human Genome Project 53 Human herpes virus-6 728 in children 1443–4 Human herpes viruses 7 and 8 728 Human immunodeficiency virus see HIV Human parvovirus B19 728 Human telomerase reverse transcriptase 167 Hydatid cyst 285 Hydralazine 529, 531 and granuloma 774 Hydroxychloroquine, graft-versus-host disease 869 Hydroxyethyl radicals 586, 590, 591–2 Hydroxymethylglutaryl-coenzyme A reductase see HMG-CoA reductase Hyperammonemia syndromes, liver cell therapy 183 Hyperbilirubinemia 239 Hypercitrullinemia 168 Hyperdynamic circulation 453–61 clinical features 153–4 complications 460–1 management 461 pathogenesis 454–60 central neural mechanisms 457–9 endocannabinoids 456–7 heme oxygenase and carbon monoxide 457
Index
Hyperdynamic circulation—cont’d pathogenesis—cont’d increased blood and plasma volume 459–60 nitric oxide 455–6 tissue hypoxia and oxidative stress 460 Hyperemesis gravidarum 1011 Hypergammaglobulinemia 239 Hyperinsulinemia 1038 Hyperlipidemia glucose-6-phosphatase deficiency 1283–4 post-transplantation 968–9 Hypermetabolic state 583 Hypernatremia 315–16 Hyperreflexia 317 Hypersplenism 493 Hypertension 602–3 post-transplantation 968 pregnancy-induced 1017–20 Hypertyrosinemia persistent 1295 secondary to liver disease 1295 Hyperuricemia 606, 1274–5 differential diagnosis 1274–5 genetics 1274 glucose-6-phosphatase deficiency 1284–5 pathology 1274 treatment 1275 Hypervariable regions 129 Hypoglycemia 1273 alcoholic liver disease 606 galactose metabolism disorders 1278 Hyponatremia 336 dilutional 421–2 management 443 treatment 342 Hypophosphatemia 1273 glucose-6-phosphatase deficiency 1285 Hypopituitarism 1369
Ibufenac 505, 532 Ibuprofen 532 Ibutilide 529 Idiopathic adulthood ductopenia 809 Idiopathic granulomatous venulitis 879 Idiopathic hypereosinophilic syndrome 879 Idiopathic portal hypertension 359–60 Ifosfamide 536 IL-1R-associated kinase 167 Imaging studies 98–9, 251–89 benign liver tumors focal nodular hyperplasia 1153 hemangioma 1148–50 hepatocellular adenoma 1159–61 nodular regenerative hyperplasia 1163 bile duct 291–6 biliary strictures 294 common bile duct and gallbladder stones 292–3 gallstones and cholecystitis 291–2 intrahepatic stones 293–4 right upper quadrant pain 291 biliary lesions bile duct adenoma 257 biliary cystadenoma 256–7 hepatic (biliary) cysts 255–6
Imaging studies—cont’d diffuse liver disease 273–82 amyloidosis 280–2 Budd-Chiari syndrome see Budd-Chiari syndrome cirrhosis see cirrhosis iron storage disorders 275–8 sarcoidosis 225, 279–80 steatosis 278 Wilson’s disease see Wilson’s disease focal malignant lesions biliary cystadenocarcinomas 271–2 cholangiocarcinoma 271 epithelioid hemangioendothelioma 272–3 fibrolamellar carcinoma 269–71 hepatic angiosarcoma 272 hepatocellular carcinoma 266–9 liver metastases 264–6 lymphoma 272 hemochromatosis 1246 hepatic infections amebic abscess 283 bacillary peliosis 286 echinococcal disease 285–6 fungal abscess 283–5 pyogenic liver abscess 282–3 schistosomiasis 286 viral hepatitis 286 hepatocellular lesions focal nodular hyperplasia 252–4 hepatocellular adenoma 251–2 nodular regenerative hyperplasia 255 mesenchymal lesions angiomyolipoma/myelolipoma 262–4 hamartoma 258, 261 hemangioma 257–61 infantile hemangioendothelioma 261, 262 inflammatory pseudotumor 264 leiomyoma 264 lipoma 264 lymphangioma 261–2 Imipramine 517 Immune disease autoimmune hepatitis see Autoimmune hepatitis overlap syndromes 796, 800 Immune electron microscopy 707–8 Immune fluorescent microscopy 707–8 Immune response adaptive 118, 589–92 cellular 592, 704–5 genes 596 hepatitis E 705–6 humeral 705 innate 116–17, 149, 588–9 Immune suppression, HBV-specific 118–19 Immune surveillance 157–9 Immune system 149–63 acute liver failure 392–3 functions of 149–50 liver and 149 Immune tolerance 155 Immunization hepatitis A 631–2 hepatitis B 656–8 escape mutants 658 high-risk groups 657
Immunization—cont’d hepatitis B—cont’d postexposure/perinatal prophylaxis 658 recent developments 658 vaccination schedule 657–8 hepatitis C 139, 683 hepatitis E 711–14 DNA vaccine 713–14 E. coli-derived vaccine 712 insect-derived vaccine 712–13 plant-derived vaccine 713 Immunogenetics 826–7 Immunoglobulins, graft-versus-host disease 869–70 Immunohistochemistry 205 Immunosuppressants graft-versus-host disease 868–9 liver transplantation 963 primary biliary cirrhosis 812 and sinusoidal obstruction syndrome 897 Immunosuppression 594 and recurrent hepatitis C 983–4 see also Immunosuppressants Impila 553 Inborn errors of metabolism 1269–328 bile acids 1304–8 hereditary lymphedema with recurrent cholestasis 1306–8 progressive familial intrahepatic cholestasis 1304–5 carbohydrates 1269–70 cystic fibrosis 1311–14 biochemical characteristics 1312 clinical features 1311 genetics 1314 laboratory findings 1313 liver disease in childhood and adolescence 1311–12 molecular basis 1312–13 pathology 1313–14 therapy 1314 evaluation of 1269 fructose 1270–5 fructose-1,6-diphosphatase deficiency 1270 fructose diphosphatase deficiency 1275 fructose phosphate aldolase deficiency 1270 hepatic enzyme elevation 1273 hereditary fructose intolerance 1270–3 hyperuricemia and increased urate excretion 1274–5 hypoglycemia 1273 hypophosphatemia 1273 galactose 1275–81 biochemical characteristics and pathogenesis brain 1278 gonads 1278 hypoglycemia 1278 kidneys 1278 lenticular changes 1278 liver 1278 diagnosis 1279–80 diagnostic screening 1279 galactokinase-deficiency galactosemia 1281 genetics 1279 pathology 1278–9
1503
Index
Inborn errors of metabolism—cont’d galactose—cont’d prognosis 1280–1 transferase-deficiency galactosemia 1275–7 treatment 1280 uridine disphosphate galactose-4epimerase-deficiency galactosemia 1281 glycogen storage disease 221, 1269–70, 1281–93 liver cell therapy 183 type IB 1289 type I (glucose-6-phosphatase deficiency) 1282–9 type III (amylo-1,6-glucosidase deficiency) 1289–91 type IV (a-1,4 glucan-6-glycosyl transferase deficiency) 1291–3 lipids 1298–304 cholesterol ester storage disease 1301–4 intracellular metabolism of cholesterol 1298–9 lysosomal acid lipase disorders 1299 Wolman’s disease 1299–301 peroxisome biogenesis 1308–11 THCA syndrome 1310–11 Zellweger’s syndrome 1308–10 tyrosine 1293–8 disorders of metabolism 1295 hereditary tyrosinemia 1296–8 hypertyrosinemia secondary to liver disease 1295 metabolic pathway 1293–5 persistent hypertyrosinemia 1295 transitory tyrosinemia of the newborn 1295 tyrosinosis 1295 Indinavir 525 Indocyanine green 245, 354 Indomethacin 532 Infantile hemangioendothelioma 227, 261, 262 Infantile Refsum disease, liver cell therapy 183 Infection acute liver failure 398–9 as contraindication to liver transplantation 937 hemochromatosis 1242–3 and parenteral nutrition-induced liver disease 1091 and portal vein thrombosis 906–7 post-transplantation 970–1 primary sclerosing cholangitis 823 Infectious hepatitis see Hepatitis A Inferior vena cava 219 Inflammation and insulin resistance 1041 and necrosis 39–40, 47 and parenteral nutrition-induced liver disease 1091 see also Cytokines Inflammatory bowel disease 797 and primary biliary sclerosis 830–1 Inflammatory pseudotumor 264 Inflammatory-related hepatitis 755 Infliximab, graft-versus-host disease 870
1504
Informed consent, liver biopsy 201 Injection sclerotherapy 371 Innate immune system 116–17, 149 alcoholic liver disease 588–9 Inositol 1,4,5-triphosphage receptor 167 Inspissated bile syndrome 1366 Insulin, acute intermittent porphyria 1400 Insulin resistance 1038 clinical estimation of 1046–8 mechanisms of endotoxin 1042 free fatty acid toxicity 1041 hepatocellular injury 1040 impaired hepatocellular ATP production 1041 inflammation and cytokines 1041 iron 1039, 1041 lipodystrophies 1039–40 liver disease 1039 oxidant stress and lipid peroxidation 1040–1 postreceptor signal transduction 1038–9 tumor necrosis factor-a and free fatty acids 1039 see also Non-alcoholic steatohepatitis Insulin-sensitizing agents 1052–3 metformin 1052 thiazolidinediones 1052 Integrins 92 Interferon-a 120, 495 hepatitis B 651–2 pediatric 1348 pegylated 125 Interferon a/b 117 Interferon-g 31 hepatic fibrosis 102 Interferons 537 hepatitis B 650 hepatitis B flares 645–6 hepatitis C 680–1 hepatocellular carcinoma 1122 Interferon sensitivity-determining region 132 Interleukin-2 537 Interleukin-2 receptor blockers and hepatitis C recurrence 984 liver transplantation 964 Interleukin-6 30–1 Interleukin-10 hepatic fibrosis 100–1 release by Kupffer cells 151 Intermediary metabolism, failure of 390 Intermediate metabolizers 56 Intermittent acute porphyria see Acute intermittent porphyria Internal ribosome entry site 126 International Normalized Ratio 239 Intracellular conversion pathway 113 Intracranial hypertension 394 management 401–2 Intrahepatic cholestasis of pregnancy 79–80, 1011–14 clinical features 1012 etiology 1011–12 incidence 1011 natural history and prognosis 1012–14 pathology 1012 therapy 1014
Intrahepatic stones 293–4 Intrauterine infection 1367–9 cytomegalovirus 1368 herpes simplex virus 1368–9 rubella 1368 toxoplasmosis 1368 Iprindole 517 Iproniazid 505 Iridotecan 61 hepatocellular carcinoma 1122 Iron 569 and insulin resistance 1039, 1041 overload see Hemochromatosis and porphyria cutanea tarda 1408 Iron deficiency anemia 491 Iron overload 868 Iron stain 207 Iron storage disorders 275–8 Ischemic hepatitis 219, 236, 1067–9, 1102 causes and pathogenesis 1067–8 clinical syndrome 1068–9 differential diagnosis 1068 treatment and prognosis 1069 Ischemic injury 966–7 Isoniazid 524 hepatotoxicity 394, 523 metabolism 60–1 Isosorbide-5-mononitrate, prevention of variceal bleeding 370 Isospora belli 787 Isoxicam 532 Isozepac 532 Itraconazole 523
Jagged1 1371–2 Jamshidi needle 196 Janus kinases 30 Jaundice 1449–50 maternal milk 1465–6 maternal serum 1466 Jejunoileal bypass 1070–1 clinical manifestations 1071 mechanism of hepatic injury 1070 pathology of liver disease 1071 Ju Bu Huan 553, 556 Jun kinase 29
Kabuki syndrome 1357 Kalk position 198 Kallikrein-kinin system 435 Kaposi’s sarcoma 787 Kaposi’s sarcoma-associated human herpes virus-8 728 Kava 553, 556 Kayser-Fleischer rings 1221, 1225, 1226 Kepone 568 Ketanserin, prevention of variceal bleeding 370 Ketoacidosis, alcoholic liver disease 606 Ketoconazole 517, 522–3 acute intermittent porphyria 1400 Ketoprofen 532 K-ras oncogene 1135
Index
Kupffer cells 3, 15–16, 47, 91, 150–1, 178, 586 functions 150–1 hemosiderosis 221 immune response 117 phenotypes 151 soluble mediators released by 151
Lablol 529 Laboratory testing 235–50 autoimmune hepatitis 244 for fibrosis 240–1 hemochromatosis 244 hepatitis B 243–4 hepatitis C 243 hepatitis D 244 liver metabolic capacity 246 liver transport capacity 245 plasma lipids and lipoproteins 241–3 primary biliary cirrhosis 244 primary sclerosing cholangitis 244–5 serum g-globulins 245 serum liver chemistry tests 235–40 albumin 239 alkaline phosphatase 237–8 aminotransferases 236–7 bilirubin 238–9 g-glutamyl transpeptidase 238 prothrombin time and coagulation factors 239–40 serum bile acids 239 Lactate dehydrogenase 240 Lactic acid, glucose-6-phosphatase deficiency 1283 Lamivudine 120, 525 hepatitis B 652–3, 655 pediatric 1348 prevention of recurrence B 979, 980 Lamotrigine 520 Langerhans’ cells 149 Laparoscopic liver biopsy 198–9, 200 Large hepatitis B surface antigen 112 Large hepatitis B surface protein 115 Large-volume paracentesis 302, 340–1 Lead 570 Lecithin-cholesterol acyltransferase 241 Legionellosis 756 Leiomyoma 264 Leishmania 787 Lepiota helveola 572 Lepirudin 529 Leprosy, granuloma formation 775–6 Leptin 95 Leptospirosis 758–9 Leukocytes abnormalities 493 in alcoholic liver disease 606 Lichen planus 671, 797 Licorice 552 Ligand-receptor interaction 26 Linton-Nicholas tube 368 Lipid inclusions 13 Lipid metabolism disorders 1298–304 cholesterol ester storage disease 1301–4 intracellular metabolism of cholesterol 1298–9 lysosomal acid lipase disorders 1299 Wolman’s disease 1299–301
Lipid peroxidation 585–6, 594, 1040–1 Lipid rafts 68 Lipodystrophies 1039–40 Lipodystrophy 1270 Lipofucsin 208 Lipogranuloma 769 Lipoma 264 Lipopeliosis 229 Lipoprotein receptor-related protein 26 Lipoproteins 241–3 Listeria 756 Liver biliary system 8 blood flow 154–5 in pregnancy 1003 development 4–6 functional anatomy 150 gross anatomy 4, 1169, 1170 hepatic functional units 9–10 hepatic parenchymal cells 10–13 cytoplasmic inclusions 13 cytoskeleton and cytomatrix 13 endoplasmic reticulum, ribosomes and Golgi apparatus 11–12 lysosomes 12 mitochondria 12 nucleus 11 peroxisomes (microbodies) 12 plasma membrane 10–11 hepatocytes 154–5 heterogeneity 17–18 immune functions 155–9 hepatic dendritic cells 155–6 hepatic microenvironment 155 hepatocytes 155 immune surveillance 157–9 immune tolerance 155 induction of T-cell tolerance 155 liver sinusoidal endothelial cells 156 organ-resident antigen-presenting cells 157 innervation 8–9 leukocyte-liver cell interaction 154–5 microscopic anatomy 6–9 non-parenchymal cells 13–17 Kupffer cells 15–16 liver-associated lymphocytes 17 sinusoidal endothelial cells 13–15 stellate cells 16–17 in pregnancy anatomy and histology 1003 bile formation 1004 blood flow 8 drug metabolism 1003–4 plasma lipids 1003 plasma proteins 1003 sinusoidal cell populations 150–4 dendritic cells 153–4 Kupffer cells see Kupffer cells liver sinusoidal endothelial cells 152–3 natural killer T cells 151–2 stellate cells see Stellate cells structure and function 3–4 vasculature 6–8 Liver biopsy 99 Alagille’s syndrome 1373 alcoholic liver disease 597 a-1-antitrypsin deficiency 1263, 1380 BRIC 1377
Liver biopsy—cont’d compliance 201 contraindications ascites 201 biliary obstruction 201 cholangitis 201 definition 195 extrahepatic biliary atresia 1361 fatty liver disease 1044–5 hemochromatosis 1246 hemostasis 201 hepatitic C 674–5 histology see Histology informed consent 201 issues addressed by 206 laparoscopic 198–9, 200 patient selection 201–2 PFIC1 1375–6 PFIC2 1378 PFIC3 1378 recurrent hepatitis C 986 role of 205 transcutaneous 195–8 bacteremia 198 experience 196 number of passes 196 Tru-cut versus aspiration needle 196–8 transvenous 198 ultrasound-guided fine-needle aspiration 199, 201 dissemination of malignant cells 201 hemorrhage 201 specificity 201 Liver cell adenoma 228 Liver cell destruction 37–51 apoptosis/oncotic necrosis 37–40 apoptosis and secondary necrosis 39 biochemical markers 40 inflammation 39–40 severity of insult 37–9 apoptotic cell death pathways 40–4 Fas receptor-mediated apoptosis in hepatocytes 40–2 internal regulators of Fas-induced apoptosis 42 intrinsic (mitochondrial) pathway 44 modulators of TNF-induced apoptosis 43–4 TNF receptor-mediated NF-kB activation 42–3 mechanisms of oncotic necrosis 44–7 inflammation 47 intracellular proteases 46–7 mitochondrial dysfunction 45–6 nuclear DNA damage 46 Liver cell transplantation 1468 Liver decompression 889–90 Liver-directed cell therapy see Stem cells Liver disease chronic 120 pediatric 225–6 Liver failure, acute 182 Liver flukes 735–7, 740, 836, 1134 Liver function abnormal 119–20 in pregnancy 1003–4 Liver function tests 99, 235–6 in pregnancy 1004 see also individual conditions
1505
Index
Liver lobule 206–7 Liver masses 1008 Liver metabolic capacity 246 Liver metastases 264–6 Liver regeneration 23–36 acute liver failure 402–3 cell cycle regulation 24–5 changes in extracellular matrix 25–6 chemokines 31 cytokines 28–31 interferon-g 31 interleukin-6 30–1 tumor necrosis factor-a 28–30 energy metabolism 31–2 inhibition of hepatocyte proliferation 32–3 ligand-receptor interaction 26 mitogenic growth factors epidermal growth factor 27–8 hepatocyte growth factor 26–7 transforming growth factor-a 27–8 Notch/Jagged signaling 28 transition of resting hepatocytes into mitosis 25 Liver resection extended 1172–3 repeat hepatectomy 1174–5 segmental 1171–2 two-step hepatectomy 1174 Liver sinusoidal endothelial cells 152–3 immune function 156 Liver support systems 324 Liver transplantation 228–31 aims of 934 Alagille’s syndrome 1374 a-1-antitrypsin deficiency 1380 and cardiovascular function 470–1 cholangiocarcinoma 1143–4 complications 955–8, 961–3, 965–70 beyond 2 months cardiac disease 969 cardiovascular disease 968 chronic rejection 970 diabetes 969 hyperlipidemia 968–9 hypertension 968 obesity 969 renal dysfunction 969–70 biliary tract 957 donor factors 961–2 early hepatic artery thrombosis 966 portal and hepatic vein thrombosis 966 primary non-function and early graft dysfunction 966 first 2 weeks acute cellular rejection 9 bacterial and fungal infections 967 biliary complications 967 cytomegalovirus 967–8 graft dysfunction 957, 962–3 infectious Cryptococcus 970 cytomegalovirus 970–1 fungal infections 970 recurrent hepatitis see Hepatitis B, post-transplant; Hepatitis C, posttransplant
1506
Liver transplantation—cont’d complications—cont’d ischemic and preservation injury 966–7 malignancy 971–2 cutaneous malignancy 971 hepatocellular carcinoma 971 post-transplant lymphoproliferative disease 971 recipient factors 963 recurrence of primary disease 972 vascular 957–8 contraindications 935–8, 948 active alcohol or substance abuse 935 anatomic abnormalities 937–8 cardiovascular disease 936 cholangiocarcinoma 938 extrahepatic malignancy 938 human immunodeficiency virus 937 infections 937 obesity 936 older age 936 previous non-hepatic malignancy 938 prior hepatic surgery 938 pulmonary disease 937 renal failure 936 retransplantation 937 Crigler-Najjar syndrome 1468 demographics 933 diagnostic studies and consultations 939–40 donor allocation and evaluation CTP scoring system 941–2 MELD scoring system 941–3 organ demand versus supply 940 UNOS regions 942 donor pool 940–1 donor procurement 948–50 extrahepatic biliary atresia 1365 graft-versus-host disease 870 hemangioma 1151 hematologic issues 492–3 hemochromatosis 1252 hepatocellular carcinoma 1120–1 history 947 indications for 934–5, 948 acute liver failure 403 alcoholic cirrhosis 611–12 alcoholic hepatitis 612 alcoholic liver disease 611–12 comorbidity 611–12 outcome 611 post-transplant recidivism 611 preoperative abstinence 612 public opinion 611 Budd-Chiari syndrome 890–1, 935 hepatic encephalopathy 324–5 non-alcoholic steatohepatitis 1035 portal vein thrombosis 909–10 primary biliary cirrhosis 815–16 primary sclerosing cholangitis 844–6 living donor liver transplantation 940–1 living donor procurement 951–2 marginal donors 940 medical justice versus utility 933–4 non-alcoholic steatohepatitis 1053–4 parenteral nutrition-induced liver disease 1095 patient selection criteria 938–9
Liver transplantation—cont’d pediatric cholestatic syndrome 1386 polycystic liver disease 1346 post-transplantation management 963–5 antiproliferative agents 965 calcineurin inhibitors 964–5 corticosteroids 963 immunosuppressants 963 interleukin-2 receptor blockers 964 T-cell-depleting agents 963–4 pregnancy after 1008–9 pretransplant evaluation 933 recurrent non-viral conditions 995–1000 autoimmune disease 995–8 autoimmune hepatitis 996 implications of 997–8 primary biliary cirrhosis 995–6 primary sclerosing cholangitis 996–7 contraindication for transplantation 999 metabolic and other conditions 998–9 alcoholic liver disease 998–9 alveolar echinococcosis 998 amyloid 998 Budd-Chiari syndrome 998 cryptogenic cirrhosis 999 erythropoietic protoporphyria 998 giant-cell hepatitis 998 hemochromatosis 998 non-alcoholic fatty liver disease 999 sarcoidosis 998 results 953–5, 956 retransplantation 958–9 split liver procurement 950–1 split liver transplantation 940 surgery 947–60 biliary tract reconstruction 953 diagnostic implications 947 evaluation of recipient 947–8 preservation solutions 949 recipient operation 952–3 vascular reconstruction 953 timing of 938 Wilson’s disease 1232 see also Graft-versus-host disease Liver transport capacity 245 Liver tumors see Tumors Living donor liver transplantation 940–1 donor procurement 951–2 hepatitis B-infected patients 981 hepatitis C-infected patients 986 Lomustine 537 Long-Evans Cinnamon rat 179 Lopinavir 525 L-SIGN 153 Lung complications, Osler-Weber-Rendu syndrome 917–18 Lymphangioma 261–2 Lymphocyte choriomeningitis virus 118 Lymphocytes, liver-associated 17 Lymphoid aggregates 210 Lymphoma 272 granuloma formation 775–6 in HIV patients 787 Lymphoproliferative disease, post-transplant 971
Index
Lysosomal acid lipase disorders 1299 see also Cholesterol ester storage disease; Wolman’s disease Lysosomes 12
Macrocytic anemia 491–2 Macrolides 527 Magnetic resonance cholangiopancreatography 736 fatty liver disease 1045 focal nodular hyperplasia 1153 hemangioma 1149–50 hepatocellular adenoma 1160–1 Wilson’s disease 1228 Ma-huang 553, 556–7 Major histocompatibility complex 139 Mallory bodies 215 Malotilate 101 primary biliary cirrhosis 813 Mangafodipir trisodium 253 Marginal donors 940 Margosa oil 557 Masson trichrome stain 206, 207 Mastocytosis 1082–3 Maternal milk jaundice 1465–6 Maternal serum jaundice 1466 Matrix metalloproteinases 91 MDR1 P-glycoprotein 72 Mebendazole 523, 524 and granuloma 774 Meclofenamic acid 532 Medium hepatitis B surface antigen 113 Medium hepatitis B surface protein 115 Medroxyprogesterone 517 Mefenamic acid 532 MELD scoring system 941–3, 963 prioritization of hepatocellular carcinoma 942–3 Membrane permeability transition 41 Membrane transporters 62 Membranoproliferative glomerulonephritis 670 Menadione 510 Menghini aspiration device 195, 196, 197 Menkes disease 1221 Mental status, hepatic encephalopathy 316–17 Mepacrine 517 Mephenytoin, acute intermittent porphyria 1400 Meprobamate, acute intermittent porphyria 1400 6-Mercaptopurine 534–5 Mesalamine, and granuloma 774 Mesenchymal lesions angiomyolipoma/myelolipoma 262–4 hamartoma 226, 261 hemangioma 257–61 infantile hemangioendothelioma 261, 262 inflammatory pseudotumor 264 leiomyoma 264 lipoma 264 lymphangioma 261–2 Metabolic bone disease 810–11 prevention 858 Metabolic diseases with cholestasis 1379–80 histology 220–2
Metabolic liver diseases 168–9 Metabolism, failure of 390 Metallothioneins 1231–2 Metals 568–71 Metal storage diseases, histology 220–2 Metastatic liver carcinoma 361 METAVIR scoring system 241 Metformin 522, 1052 Methazolamide 521 Methionine cycle, inhibition in alcoholic liver disease 584–5 Methotrexate 534 graft-versus-host disease 869 hepatic fibrosis 101 primary biliary cirrhosis 812 Methyldopa, and granuloma 774 Methyprylon, acute intermittent porphyria 1400 Metoclopramide acute intermittent porphyria 1400 prevention of variceal bleeding 370 Metronidazole, hepatic encephalopathy 321 Mexiletine 529 Mibefradil 505 Microcirculatory dysfunction 391 Microcytic anemia 491 Microhamartoma 227 Microlithiasis 1184 Microsomal ethanol-oxidizing system 582 Microvesicular steatosis 1036–7 Minichromosomes 114 Mini laparoscopy 200 Minnesota tube 368, 369 Minocycline 527 Minoxidil 529 Mirizzi syndrome 1205–6 Mistletoe 553, 557 Mithramycin 536 Mitochondria 586 Mitochondrial b-oxidation 1037–8 Mitochondrial dysfunction 45–6 Mitogen-activating protein kinase 43 Mitogenic growth factors epidermal growth factor 27–8 hepatocyte growth factor 26–7 transforming growth factor-a 27–8 Mitosis 25 Mitoxantrone, hepatocellular carcinoma 1122 Mixed cryoglobulinemia 670 Mixed pattern injury 512–13, 515 MMPs see matrix metalloproteinases Modafinil 521 Model for End-Stage Liver Disease (MELD) 299 Molecular absorbent recirculating system (MARS) 324, 610 Molecular virology 111–16 Monoclonal antibodies graft-versus-host disease 869, 870 hepatitis B 654 Moricizine 529 Multidrug resistance protein MRP2 73 Multiorgan failure 401 Multiple myeloma 1081–2 Multipotency 177 Muromonab-DC3, graft-versus-host disease 869 Mushroom poisoning 571–2
Mutant Corriedale sheep 1474 Mycobacterial infections graft-versus-host disease 868 granuloma 772 Mycobacterium avium complex 786 Mycobacterium avium-iontracellulare 758 Mycobacterium tuberculosis 757–8 HIV patients 786 Mycophenolate mofetil graft-versus-host disease 869 and hepatitis C recurrence 984 liver transplantation 964 Mycotoxins 169–70 Myelolipoma 262–4 Myeloproliferative disorders, and portal vein thrombosis 906 Myofibroblasts 357
Nabumetone 532 Nafenopin 31 Nagase analbuminemic rat 179 Nail dystrophy 797 Nalidixic acid 527, 528 Naprosyn 532 Narcotic analgesics, acute intermittent porphyria 1400 NASH see Non-alcoholic steatohepatitis National Health and Nutrition Examination Survey (NHANES) III study 666 Natriuretic peptides 434–5 Natural killer cells 31, 47, 116, 589 Natural killer T cells 116, 117, 151–2 populations of 152 Necrosis 506–7 biochemical markers 40 and inflammation 39–40 secondary 39 Neisseria spp. 757 Nelfinavir 525 NEMO 29 Neocholangioles 209 Neomycin, hepatic encephalopathy 319, 320 Neonatal hepatitis syndrome 1261, 1367 bacterial infections 1369 chromosomal disorders 1369 etiology and pathogenesis 1367 hypopituitarism 1369 idiopathic 1369–71 clinical features 1370 histology 1371–2 management and treatment 1371 prognosis 1371 viral infections 1367–9 Neonatal hyperbilirubinemia 1465–6 bilirubin conjugation 1465–6 canalicular bilirubin excretion 1466 immaturity of hepatic bilirubin uptake 1465 increased bilirubin load 1465 increased intestinal reabsorption 1466 Neonatal sclerosing cholangitis 1366 Neonate, transitory tyrosinemia 1295 Neoplasms and alcohol 605 esophageal 601 estrogen-responsive, in pregnancy 1009–10 gallbladder carcinoma 841–2
1507
Index
Neoplasms—cont’d in HIV patients 787 pancreatic carcinoma 842 and portal vein thrombosis 906 see also Cholangiocarcinoma; Hepatocellular carcinoma; Tumors Nervous system acute liver failure 393 alcoholic liver disease alcoholic dementia 604 brainstem disease 604 cerebellar disease 604 neuropathies 604 Wernicke-Korsakoff syndrome 603–4 Nesirtide 529 Neurohumoral systems in cirrhosis 426–35 counterbalancing systems natriuretic peptides 434–5 nitric oxide 435 prostaglandins 433–4 renal kallikrein-kinin system 435 effector mechanisms adenosine 433 arginine vasopressin 430–1 endothelins 431–3 renin-angiotensin-aldosterone system 427–8 sympathetic nervous system 428–30 vasoconstrictor eicosanoids 433 Neuropathy 604 Neuropeptide Y 8 Neurotropic tyrosine receptor kinase 2 167 Neutral sphingomyelase 44 Neutrophils 47 Nevirapine 525 Niacin 529 Niemann-Pick disease 220, 221, 1270 Nifedipine 529 Niflumic acid 532 Nitrates, prevention of variceal bleeding 369–70 Nitric oxide 333, 357–8, 435, 825 in hyperdynamic circulation 455–6 neuronal 9 Nitric oxide synthase 358 inducible 586 Nitroaliphatic compounds 565 Nitroaromatic compounds 564–5 Nitrobenzene 564 Nitrofurantoin 527, 528 Nitrogen, promotion of excretion 322 Nitroglycerin, acute variceal bleeding 367 Nitroparaffins 565 Nitrosoureas 537 Nodular regenerative hyperplasia 228, 255, 362, 1162–3 associated conditions 1163 clinical features 1162 diagnosis 1162 disease complications 1163 epidemiology 1162 imaging 1163 pathogenesis 1162 prognosis and natural history 1163 treatment 1163 Non-alcoholic fatty liver disease see Fatty liver disease Non-alcoholic liver disease 361
1508
Non-alcoholic steatohepatitis 89–90, 1031–63 alcohol history 1043 associated conditions 1050–1 clinical estimation of insulin resistance 1046–8 clinical features 1043 concomitant diseases 1049 diagnosis 1042–3 differential diagnosis 1048–9 disease complications 1051 drugs causing amiodarone 1042 tamoxifen 1042 epidemiology 1033–5 liver transplantation, hepatocellular carcinoma and death from liver failure 1035 examination findings 1044 hepatic fibrosis 95 hepatitis A vaccination 1054 and hepatocellular carcinoma 1111 history 1031 imaging studies 1045–6 laboratory tests 1044 liver biopsy 1044–5 nomenclature 1031–2 pathogenesis 1035–42 accumulation of triglyceride in hepatocytes 1035–8 hepatic fibrogenesis 1042 insulin resistance 1038–42 predictors of 1048 prognosis and natural history 1054–6 cryptogenic cirrhosis 1055 hepatocellular carcinoma 1055–6 progression to cirrhosis 1055 safe alcohol intake 1054 similar disorders 1032–3 symptoms 1043 tamoxifen use 1054 treatment 1051–4 betaine 1053 exercise and lifestyle modifications 1051 insulin-sensitizing agents 1052 liver transplantation 1053–4 PPARa ligands 1053 statins 1053, 1054 ursodeoxycholic acid 1053 vitamin E 1053 weight reduction 1052 versus alcoholic steatohepatitis 1049–50 see also Fatty liver disease Non-Hodgkin’s lymphoma 671–2, 1081 Non-nucleoside reverse transcriptase inhibitors 526 Non-parenchymal cells 13–17 Kupffer cells 15–16 liver-associated lymphocytes 17 sinusoidal endothelial cells 13–15 stellate cells 16–17 Non-steroidal anti-inflammatory drugs 531–3 aspirin 531–2 diclofenac 533 diflunisal 532 indomethacin 532 sulindac 532–3 Norwegian cholestasis 80, 1270, 1306
Notch/Jagged signaling 28 Notch receptor 1371–2 NS2–3 protease 129–30 NS3–4A complex 130 NS4B 131 NS5A 131–2 NS5B 132–3 N-substituted amides 567 Nuclear DNA damage 46 Nuclear factor-kB, TNF receptor-mediated activation 42–3 Nuclear hormone receptor proteins 75 Nucleoside analog reverse transcriptase inhibitors 525–6 Nucleoside analogs hepatitis B 650, 652–3, 655–6 hepatitis B flares 646 resistance 980–1 b-L-Nucleosides 654 Nucleotide reverse transcriptase inhibitors 526 Nutritional supplementation acute liver failure 397 alcoholic hepatitis 608–9 carbohydrates 1384 energy requirements 1384 long-chain polyunsaturated fatty acids 1384 medium-chain triglycerides 1384 proteins 1384–5 structured lipids 1384
Oatp family 71 Obesity as contraindication to liver transplantation 936 and hepatocellular carcinoma 1111 post-transplantation 969 Occupational/environmental hepatotoxicity 561–77 aflatoxins 574–5 chemicals 563–71 aromatic hydrocarbons 566 chlorinated ethylenes 565–6 halogenated aliphatic hydrocarbons 566–7 halogenated aromatic hydrocarbons 563–4 metals 568–71 nitroaliphatic compounds 565 nitroaromatic compounds 564–5 N-substituted amides 567 pesticides 567–8 cocaine 574 cyanobacterial toxins 575 diagnosis 563 household products 562 management 563 mushroom poisoning 571–2 pyrroliidines 572–4 types of exposure ingestion 561 inhalation 561 mucosal absorption 563 websites 562 Octreotide acute variceal bleeding 368 hepatocellular carcinoma 1122
Index
Ofloxacin, acute intermittent porphyria 1400 Ohm’s law 354 Oncotic necrosis 37–40, 44–7 inflammation 47 intracellular proteases 46–7 mitochondrial dysfunction 45–6 nuclear DNA damage 46 Onion skin fibrosis 214, 829 Onshido 553, 554–5 Open reading frames 694 Opiate antagonist 323 Opisthorchiasis 740 Opisthorchis viverrini 740, 836, 1134 Orcein/Victoria blue stain 207 Orientia tsutugamuchi 759 L-Ornithine-L-aspartate (LOLA) 322 Ornithine transcarbamylase deficiency, liver cell therapy 183 Oropharynx, alcoholic liver disease 600–1 Orthotopic liver transplantation 177 Osler-Weber-Rendu disease 915–29 diagnostic criteria 915 epidemiology 915 genetic background 915–17 hepatic manifestations 919–25 clinical presentation 921 diagnosis 921–5 epidemiology 919 histopathology and pathophysiology 919–21 organ manifestations 917–19 central nervous system 918 gastrointestinal tract 918–19 lung 917–18 nose 917 skin 917 pathophysiology 917 therapy 925–6 Osteoporosis 606–7 Oval cells 177, 178, 179 Overflow theory 438 Overlap syndromes 796, 800, 855–61 autoantibodies 855 autoimmune hepatitis with features of primary biliary cirrhosis 858–9, 860 with features of primary sclerosing cholangitis 859, 860 autoimmune hepatitis and primary sclerosing cholangitis 855–8 sequential 860 clinical presentation 800 definition of 855 diagnostic methods 800 differential diagnosis 800, 860 management 860 pathology 800 primary biliary cirrhosis 808 treatment and prevention 800 see also Autoimmune hepatitis Owl’s eye inclusions 726 Oxacillin 527 Oxalosis, liver cell therapy 183 Oxaprofen 532 Oxcarbazepine 520 Oxidant/antioxidant imbalance 1091 Oxidant stress 93, 1040–1
Oxidative stress 460, 510–11, 585–6 in alcoholic liver disease 586 genes influencing 595–6 manipulation of 587
P7 129 P42 mitogen-activated protein kinase 1 167 P53 tumor suppressor gene 169–70 Paclitaxel, hepatocellular carcinoma 1122 Palmar erythema 1005 Pancreas, alcoholic liver disease 602 Pancreatic carcinoma 842 Papaya 552 Paracetamol see Acetaminophen Paracoccidiodes brasiliensis 760 Paracoccidioidomycoses 760 Paraquat 568 Parasitic and helminthic disease 735–45 Clonorchis senensis 735–7 Echinococcus 740–2 Entamoeba histolytica 742–3 hepatobiliary ascariasis 737–40 opisthorchiasis 740 schistosomiasis 743–4 Parenteral nutrition 1385 Parenteral nutrition-associated liver disease 1089–98 diagnosis 1093 differential diagnosis 1093 epidemiology 1089–90 histologic features 1093 pathogenesis 1090–3 bacterial overgrowth 1091 deficiencies 1091–2 gastrointestinal dysfunction 1091 immaturity of hepatobiliary function and transport 1090 infection and inflammation 1091 inherent toxicity of constituents or formulation 1092–3 oxidant/antioxidant imbalance 1091 toxic bile acids 1090–1 prognosis and natural history 1095 treatment 1093–5 antimicrobials/anti-inflammatory agents 1094 antioxidants 1094 choleretics 1093–4 cyclic infusion 1094 enteral stimulation 1094 gastrointestinal hormones 1094 transplantation 1095 Paroxysmal nocturnal hemoglobinuria 878 Parvovirus B19 1443 Pediatric cholestatic syndromes 1355–90 bile duct paucity 1371–5 Aagenaes’s syndrome 1374–5 Alagille’s syndrome 1371–4 non-syndromic 1374 progressive familial intrahepatic cholestasis 1375 BRIC 1376–7 clinical presentations 1355–6 extrahepatic disorders 1356–67 agenesis of extrahepatic bile ducts 1367 bile duct stenosis 1366–7 bile duct tumors 1367
Pediatric cholestatic syndromes—cont’d extrahepatic disorders—cont’d bile plug syndrome and inspissated bile syndrome 1366 Caroli’s disease 1365–6 choledochal cyst 1365 duplication of biliary tree 1367 extrahepatic biliary atresia 1356–65 extramural compression of common bile duct 1367 gallstones 1367 neonatal sclerosing cholangitis 1366 spontaneous perforation of common bile duct 1366 inborn errors of bile acid synthesis 1379 intestinal failure-associated liver disease 1380–3 intrauterine infection 1367–71 cytomegalovirus 1368 herpes simplex virus 1368 rubella 1368 syphilis 1369 toxoplasmosis 1368 varicella 1369 investigations 1356 management biliary diversion 1386 nutritional assessment 1383–4 nutritional support 1383, 1384–5 parenteral nutrition 1385 pruritus 1385 metabolic disorders presenting with cholestasis 1379–80 neonatal hepatitis syndrome 1367 bacterial infections 1369 chromosomal disorders 1369 hypopituitarism 1369 idiopathic 1369–71 viral infections 1367–9 PFIC1 1375–6 PFIC2 1377–8 PFIC3 1378 Pediatric liver disease 225–6 Pediatric viral hepatitis 1433–48 adenovirus 1443 cytomegalovirus 1443 enteroviruses 1443 Epstein-Barr virus 1443 hepatitis A 1433–5 hepatitis B 1435–9 hepatitis C 1439–42 hepatitis D 1442–3 hepatitis E 1443 herpes simplex virus 1443 human herpes virus-6 1443–4 parvovirus B19 1443 see also various types Pegylated interferon, recurrent hepatitis C 987 Peliosis hepatica 517, 757 HIV patients 787 Penicillamine, hepatic fibrosis 101 D-Penicillamine primary biliary cirrhosis 812 Wilson’s disease 1230–1 Penicillins 526–7 acute intermittent porphyria 1400 Pennyroyal 553, 558
1509
Index
Pentoxifylline alcoholic hepatitis 608 sinusoidal obstruction syndrome 901 Perchloroethylene 562, 567 Percutaneous transhepatic biliary decompression 1203 Pergolide 521 Pericarditis 470 Perihexilene 505 Periodic acid-Schiff 205–207 Perisinusoidal space of Disse 5 Peritoneovenous shunt, ascites 342 Pernicious anemia 797 Peroxisomal oxidation 1037 Peroxisome biogenesis disorders 1308–11 THCA syndrome 1310–11 Zellweger’s syndrome 1308–10 Peroxisome proliferator-activated receptor a 31 inhibitors 583–5 ligands 1053 Peroxisomes (microbodies) 12 Pertinent negatives 208 Pesticides 566–8 PFIC1 1375–6 clinical features 1375 genetics 1375 investigation 1375–6 management 1376 PFIC2 1377–8 clinical features 1377–8 genetics 1377 investigation 1378 management 1378 PFIC3 1378 Pharmacogenetics 53–66 HMG-CoA reductase 62–3 membrane transporters 62 phase I enzymes 54–9 cytochrome P450 2C9 57–8 cytochrome P450 2D6 55–6 cytochrome P4502C19 58 dihydropyrimidine dehydrogenase 58–9 phase II enzymes 59–62 N-acetyltransferase 2 60–1 glutathione S-transferases 61 thiopurine methyltransferase 59–60 UDP-glucuronosyltransferase 1 61–2 Pharmacogenomics 53 Phase I enzymes 54–9 cytochrome P450 2C9 57–8 cytochrome P450 2D6 55–6 cytochrome P4502C19 58 dihydropyrimidine dehydrogenase 58–9 Phase II enzymes 59–62 N-acetyltransferase 2 60–1 glutathione S-transferases 61 thiopurine methyltransferase 59–60 UDP-glucuronosyltransferase 1 61–2 Phenothiazines, acute intermittent porphyria 1400 Phenotype 56 Phenylbutazone, and granuloma 774 Phenytoin 519 acute intermittent porphyria 1400 and granuloma 774 Phosphatidylcholine 79 alcoholic cirrhosis 610–11
1510
Phospholipidosis 517 Phospholipid transporter 72 Phosphorus 562, 568 Photodynamic therapy 1142–3 Photoisomerization 1463 Photopheresis, graft-versus-host disease 870 Phototherapy 1468 Picric acid 562 Pigments 208 Pigment stones 1181 Piperacillin 527 Pipe stem fibrosis 744 Piroxicam 532 Pirprofen 532 Pit cells 17 PKD1 1332–3 PKD2 1333 Plasma amino acid hypothesis of hepatic encephalopathy 313–14 Plasma lipids 241–3 Plasmapheresis 1468 Plasma volume 459–60 Platelet plug formation 490–1 Platelets, in alcoholic liver disease 606 Plugged biopsy 198 Plumboporphyria 1396 Pluripotency 177 Pneumocystis spp. 760 Pneumocystis carinii 787 Poiseuille’s law 354 Poly (ADP-ribose) polymerase-1 41 Polychlorinated biphenyls 562, 563–4 Polycystic liver disease 1335–6, 1342–6 clinical features 1343 complications 1345 gastrointestinal symptoms 1344 mechanism of 1336 molecular diagnostics 1342 natural history 1342 PRKCSH and hepatocystin 1335–6 Sec63 and Sec63p 1336 treatment cyst fenestration 1344–5 liver resection 1345–6 liver transplantation 1346 medical 1343–4 radiological cyst aspiration and sclerosis 1344 Polycystin-1 1332–3 Polycystin-2 1333 Polyenylphosphatidylcholine 101 Polymerase chain reaction 243 Polymerase inhibitors 681–3 Polymerase protein 112 Polymorphonuclear cell count 338 Polyprotein processing 127–8 Porphyria cutanea tarda 671, 1405–11 classification and enzyme defects 1407 clinical manifestations 1405–6 etiology and pathogenesis 1406–7 experimental models 1407 history, definition and prevalence 1405 laboratory evaluation and diagnosis 1409–10 pathogenesis of porphyrin patterns 1409
Porphyria cutanea tarda—cont’d susceptibility factors 1407–9 alcohol 1408 antioxidants 1408 estrogens 1408–9 hepatitis C 1408 HIV 1408 iron 1408 smoking and cytochrome P450 enzymes 1408 treatment 1410–11 Porphyrias 168, 1391–432 acute intermittent porphyria 1397–404 g-aminolevulinic acid dehydratase porphyria 1395–7 classification and diagnosis 1394 congenital erythropoietic porphyria 1404–5 dual porphyria 1417 erythropoietic protoporphyria 1414–17 post-transplant 998 heme pathway intermediates 1394 heme synthesis 1391–3 hepatoerythropoietic porphyria 1411 hereditary coproporphyria 1412–13 laboratory testing 1394–5 acute porphyrias 1394–5 blistering cutaneous porphyrias 1395 non-blistering cutaneous porphyrias 1395 relatives and patients with subclinical porphyria 1395 total plasma porphyrins 1395 urinary porphyrin precursors 1394–5 plumboporphyria 1396 porphyria cutanea tarda 671, 1405–11 protoporphyria 1270 liver cell therapy 183 tumor-related 1417 variegate porphyria 1413–14 Porta hepatis 4 Portal hypertension 348–62, 834–5 assessment of portal venous system 350–4 hepatic artery flow and portal venous blood flow 354 identification of varices 350 portal venous pressure 351–3 variceal pressure and flow measurement 353–4 visualization of portal and hepatic veins 350–1 causes 358–62 alcoholic liver disease 360 arsenic, vinyl, chloride or copper 360 arteriovenous fistula 358–9 congenital hepatic fibrosis 361 cystic fibrosis 362 focal nodular hyperplasia 362 heart disease 362 idiopathic portal hypertension 359–60 metastatic liver carcinoma 361 nodular regenerative hyperplasia 362 non-alcoholic liver disease 361 sarcoidosis 361–2 schistosomiasis 360 splenomegaly 359 clinical features portal-systemic collaterals 349 splenomegaly 349–50
Index
Portal hypertension—cont’d definition 348 diseases causing 358 extrahepatic biliary atresia 1364–5 primary sclerosing cholangitis 834–5 pathogenesis 354–8 increased hepatic arterial flow 355 increased portal vein blood flow 354–5 increased resistance 356–8 surgical management 373–5 decompressive shunts 373–5 Portal hypertensive gastropathy 364–5 Portal lobules 9 Portal-systemic collaterals 349 Portal-systemic encephalopathy 311 Portal triads 6 Portal vein 219, 347 embolization 1173–4 endotheliitis 220 increased blood flow 354–5 neural anatomy 347 physiologic significance 347–8 visualization of 350 Portal vein thrombosis 905–9 clinical manifestations 907 diagnosis 908–9 etiology 906–7 cirrhosis 906 infection 906–7 myeloproliferative disorders 906 neoplasia 906 thrombophilias 907 liver transplantation 909–10 management 910–11 medical measures 910–11 surgical measures 910 natural history 908 pathophysiology 905–6 post-transplantation 966 Portal venous blood flow 354 Portal venous pressure hepatic vein catheterization 351–3 direct transhepatic puncture 353 Portal venous system 347, 348 Portopulmonary hypertension 483–5 clinical features 483 definition 483 diagnosis 483–4 epidemiology 483 pathology and pathogenesis 483 prognosis and natural history 485 treatment 484–5 Portosystemic shunts, in hepatic encephalopathy 324 Positron emission tomography 266 Postembolization syndrome 1124 Postoperative liver dysfunction 1101–3 acute viral hepatitis 1102 anesthetic-induced liver injury 1101–2 benign cholestasis 1102–3 bile duct obstruction 1103 drug-induced hepatitis 1102 ischemic hepatitis 1102 patient evaluation 1103 Postperfusion syndrome 725 PPARa see Peroxisome proliferator-activated receptor a
Praziquantel 736, 744 Prazosin 529 Preascitic cirrhosis 420–1 Prednisone autoimmune hepatitis 799 liver transplantation 964 Pre-eclampsia see Pregnancy-induced hypertension Pregenomic RNA 112 Pregnancy 1003–29 acute viral hepatitis 1005 after liver transplantation 1008–9 alcohol and 1006 alkaline phosphatase in 237 cholecystectomy in 1212–13 chronic liver disease autoimmune hepatitis 1006 cirrhosis 1006–7 Dubin-Johnson syndrome 1008 hepatitis B 1007 hepatitis C 1007–8 impact on pregnancy 1006 liver masses 1008 primary biliary cirrhosis 1008 primary sclerosing cholangitis 1008 Wilson’s disease 1006, 1232–3 differential diagnosis of severe liver disease in 1020 endoscopic retrograde cholangiopancreatography 1193 hepatitis E infection in 709–10 liver anatomy and histology 1003 liver blood flow 1003 liver disorders related to biliary tract disease 1009 Budd-Chiari syndrome 1010–11 estrogen-responsive hepatic neoplasms 1009–10 herpes simplex hepatitis 1009 liver disorders unique to acute fatty liver of pregnancy 1014–17 hyperemesis gravidarum 1011 intrahepatic cholestasis of pregnancy 1011–14 pregnancy-induced hypertension 1017–20 liver function changes bile formation 1004 drug metabolism 1003–4 plasma lipids 1003 plasma proteins 1003 liver function tests 1004 skin manifestations of liver disease 1004–5 Pregnancy-induced hypertension 1017–20 HELLP syndrome 1018–19 hepatic hemorrhage and rupture 1019–20 hepatic pathology 1018 incidence of pre-eclampsia 1017 liver involvement 1018 overlap with acute fatty liver of pregnancy 1019 pathophysiology 1017–18 Preoperative hepatic dysfunction 1099–101 Preservation injury 229, 966–7 Primary biliary cirrhosis 216–17, 803–20 asymptomatic 806, 814–15 clinical features 806
Primary biliary cirrhosis—cont’d diagnosis 806–8 biochemical features 806 histologic features 807–8 overlap syndrome with autoimmune hepatitis 808 radiologic features 807 serologic feature 806–7 differential diagnosis 808–9 disease-modifying therapies antifibrotic agents 812–13 combination therapies 813 immunosuppressive agents 812 novel agents 813–14 ursodeoxycholic acid 813 disease-related complications 809–12 cancer 811–12 dyslipidemia 810 fatigue 809–10 fat-soluble vitamin deficiency 811 metabolic bone disease 810–11 pruritus 810 sicca syndrome 810 steatorrhea 811 epidemiology 803–4 extrahepatic associated conditions 809 with features of autoimmune hepatitis 858 genetics 804 granuloma formation 770 and hepatocellular carcinoma 1111 laboratory testing 244 liver transplantation 815–16 natural history and prognosis 814–15 pathogenesis 804–6 immune-mediated mechanisms 804–5 non-immune mediated mechanisms 805–6 and portal hypertension 361 post-transplant 995–6 in pregnancy 1008 prognostic survival models 815 symptomatic 806, 815 Primary lobules 10 Primary sclerosing cholangitis 809, 821–54 associated diseases 832–3 inflammatory bowel disease 830–1 ulcerative colitis 831–2 and autoimmune hepatitis in adults 856–7 in children 855–6 prevention 858 specific treatment 857–8 clinical features 827–9 diagnosis blood tests 828 histology 829 imaging 828–9 differential diagnosis 829–30 disease complications 833–5 chronic cholestasis 834 portal hypertension 834–5 disease definition 827–8 epidemiology 821–2 with features of autoimmune hepatitis 859 hepatic malignancies cholangiocarcinoma 836–41 gallbladder carcinoma 841–2
1511
Index
Primary sclerosing cholangitis—cont’d hepatic malignancies—cont’d hepatocellular carcinoma 842 pancreatic carcinoma 842 prevalence 835–6 large bile duct 829 liver transplantation 844–6 natural history and prognosis 846–9 pathogenesis 822–7 animal models 822–3 immune mechanisms 824–7 autoantibodies 824 cholangiocytes 824–5 genetic factors 825–6 immunogenetics 826–7 non-immune mechanisms 823–4 post-transplant 996–7 in pregnancy 1008 small bile duct 830 treatment 842–4 endoscopic and surgical 844 Primidone, acute intermittent porphyria 1400 Pringle maneuver 1169–70 PRKCSH 1335–6 Probiotics 322 Procainamide 529 Progesterone, acute intermittent porphyria 1400 Progressive disseminated histoplasmosis 760 Progressive familial intrahepatic cholestasis 1270, 1375 liver cell therapy 183 type 1 76–7 type 2 78–9 type 3 79 Proliferative glomerulonephritis 797 Promethazine 517 Propranolol 517 acute intermittent porphyria 1400 Propylthiouracil 102 alcoholic cirrhosis 610 alcoholic hepatitis 609 Prostaglandins 433–4 Protease inhibitors 526, 680–1 Proteases, intracellular 46–7 Protein C 490 Protein S 490 Protein synthesis 3 Proteomics 98 Prothrombin time 239–40 acute liver failure 398 Protoporphyria 1270 liver cell therapy 183 Protozoal infections in HIV patients 787 Pruritus 810 management 1385 Pseudocirrhosis 920 Pseudo-Cushing’s syndrome 606 Pseudointimal hyperplasia 299 Psoralen, graft-versus-host disease 870 Psychomotor agitation 402 Pulmonary abnormalities 477 clinical features 483 Pulmonary disease, as contraindication to liver transplantation 937 Pyelophlebitis 219, 755
1512
Pyogenic liver abscess 282–3 associated conditions 751 clinical features 748–51 differential diagnosis 751 disease complications 751–2 epidemiology 747 etiology 747–8 microbiology 748 pathogenesis 747–8 prognosis and natural history 754 treatment 752–4 endoscopic drainage 754 percutaneous drainage 752–4 pharmacologic 752 surgical drainage 754 Pyrazinamide 523, 524 acute intermittent porphyria 1400 and granuloma 774 Pyrimethamine/sulfadoxine 523 Pyrrolizidine alkaloids, and sinusoidal obstruction syndrome 897 Pyrrolizidines 552, 555, 572–4 Pyrroloporphyria see Acute intermittent porphyria Pyruvate dehydrogenase, primary biliary cirrhosis 813, 814
Q fever 389, 759–60 granuloma formation 772–3 Quinidine 529 and granuloma 774 Quinine, and granuloma 774 Quinolones 527 Quinupristin/dalfopristin 527, 528
Radiation-induced liver disease 897–8 Ranitidine, acute intermittent porphyria 1400 Rapamycin, graft-versus-host disease 869 Receptor-mediated endocytosis 152 Redox cycling 510 Renal dysfunction 417–52 acute renal failure 392, 399–401 in cirrhosis circulatory abnormalities 425–6 management 441–5 neurohumoral systems 426–35 pathophysiology 438–41 time-course of renal function abnormalities 435–8 functional renal abnormalities 417 post-transplantation 969–70 renal vasoconstriction and hepatorenal syndrome 422–4 clinical and laboratory findings 422–4 definition 422 diagnosis 424 pathogenic mechanisms 422 precipitating factors 424 prevention of 446 sodium retention and ascites 418–21 assessment of renal sodium excretion 419–20 clinical consequences 419 nephron sites of sodium retention 419 preascitic cirrhosis 420–1
Renal dysfunction—cont’d water retention and dilutional hyponatremia 421–2 assessment of renal water excretion 422 clinical consequences 421–2 impaired renal water handling 421 Renal failure, as contraindication to liver transplantation 936 Renal vasoconstriction 422–4 see also Hepatorenal syndrome Renin-angiotensin-aldosterone system 333, 427–8 Repaglinide 522 Replicon system 133–5 Respiratory complications 477–88 acute liver failure 391–2, 398 hepatopulmonary syndrome 477–83 clinical features 478–9 definition 477 diagnosis 479–81 epidemiology 477–8 pathology and pathogenesis 478 prognosis and natural history 482–3 therapy 481–2 portopulmonary hypertension 483–5 clinical features 483 definition 483 diagnosis 483–4 epidemiology 483 pathology and pathogenesis 483 prognosis and natural history 485 treatment 484–5 pulmonary abnormalities 477 Respiratory failure 1070 Reteplase 529 Reticulin stain 206, 207 Retransplantation 937, 958–9, 981 Reye’s syndrome 389, 1074–6 clinical features 1076 differential diagnosis 1076 epidemiology 1074–5 histopathology and pathogenesis 1075–6 treatment 1076 Rheumatoid arthritis 797, 1072–3 Ribavirin 125, 495 recurrent hepatitis C 987 Ribavirin analogs 681 Richner-Hanhart syndrome 1295 Rickettsia 759–60 Coxiella burnetii 759–60 erlichiosis 759 spotted fevers 759 typhus 759 Rickettsia prowazekii 759 Rifampicin 523, 524 acute intermittent porphyria 1400 Rifapentine 523, 525 Rifaximin, hepatic encephalopathy 320 Right upper quadrant pain 291 Right ventricular failure 219–20 Riluzole 521 Ritonavir 525 Rocky Mountain spotted fever 759 Rofecoxib 505 Rosiglitazone, and granuloma 774 Rotavirus C 1357 Rotor syndrome 1474, 1475
Index
Roussel-UCLAF causality assessment method 506 Rubella, intrauterine infection 1368
St John’s wort 552 Salicylates 532 Salivary glands, alcoholic liver disease 600–1 Salmonella 756 Sanded nuclei 211 Saquinavir 525 Sarcoidosis 225, 279–80, 361–2, 809 granuloma formation 770–1 post-transplant 998 SARS 729 Schistosomiasis 286, 360, 743–4 clinical presentation 744 diagnostic testing 744 granuloma formation 771–2 treatment 744 Scintigraphy extrahepatic biliary atresia 1361 focal nodular hyperplasia 1153 hemangioma 1150 Scleroderma 1074 Scrub typhus 759 Sec63 1336 Sec63p 1336 Secondary necrosis 39 Segmental vascular clamping 1171 Seizures 402 Seldinger biopsy technique 198 Semliki Forest virus 129 Semustine 537 Senecio spp. 573 Sengstaken-Blakemore tube 368 SEN virus 730 Seocalcitol, hepatocellular carcinoma 1122 Sepsis, prevention 858 Serine protease 130 Serotype 243 Serum ascites albumin gradient 335, 337 Serum bile acids 239 Serum g-globulins 245 Serum hepatitis see Hepatitis B Serum liver chemistry tests 235–40 albumin 239 alkaline phosphatase 237–8 aminotransferases 236–7 bilirubin 238–9 g-glutamyl transpeptidase 238 prothrombin time and coagulation factors 239–40 serum bile acids 239 Severe acute respiratory syndrome 729, 1083 Sho-saiko-to 552, 553 Shrinkage necrosis 506 a-2,3-Sialyltransferase 167 Sicca syndrome 810 Sickle-cell anemia 1079 biliary tract disease 1079–80 hepatic crisis 1079 hepatic histology 1080 viral hepatitis 1080 Signal transduction 170–3 Silymarin 101 primary biliary cirrhosis 814
Simvastatin, primary biliary cirrhosis 813, 814 Single nucleotide polymorphisms 53 Sinusoidal endothelial cells 3, 13–15 Sinusoidal hypertension 333 Sinusoidal obstruction syndrome 897–904 clinical course and prognosis 900 clinical presentation 899 diagnosis 899–900 imaging studies 900 laboratory studies 899–900 differential diagnosis 900 epidemiology 897–8 chemotherapy and immunosuppressants 897 pyrrolizidine alkaloids 897 radiation-induced liver disease 897–8 stem cell transplantation 898 myeloablative therapy clinical features unrelated to 900–1 preventive strategies after 901 treatment after 901 pathogenesis 898–9 animal studies 898–9 clinical studies 898 Sinusoids 220 Sirolimus and hepatitis C recurrence 984 liver transplantation 964, 965 Sjögren’s syndrome 797, 1073–4 Skin cancer, post-transplantation 971 Skin complications graft-versus-host disease 86 Osler-Weber-Rendu syndrome 917 Skin manifestations of liver disease 1004–5 Skin pigmentation 1242 Skullcap 553, 557 SLC21A6 71 Small droplet cholestasis 217 Small hepatitis B surface protein 115 Small interfering RNA, hepatitis B treatment 654–5 Small intestine, alcoholic liver disease 601–2 Snowstorm sign 285 Sodium benzoate 322 Sodium restriction 340 Sodium retention 333–4, 418–21 assessment of renal sodium excretion 419–20 clinical consequences 419 management 441–3 nephron sites 419 preascitic cirrhosis 420–1 see also Ascites Solitary hepatic cysts 1348, 1350 characteristics 1348, 1350 histopathology 1341–2 treatment 1350 Somatostatin 8 acute variceal bleeding 367–8 Sotalol 529 Space of Disse 68, 90, 152, 185, 348, 357, 592 Space-occupying lesions 879 Sphingolipids 44 Spider naevi 1004–5 Spironolactone, ascites 340 Splanchnic circulation, cirrhosis 425
Spleen, as extrahepatic reservoir 182 Splenic vein thrombosis 909 Splenomegaly 349–50, 359 Splenoportography 350 Split liver transplantation 940 liver procurement 950–1 Spontaneous bacterial peritonitis 336–9 clinical picture 336 diagnosis 336–8 epidemiology 336 prophylaxis 339 treatment 338–9 Spotted fevers 759 SR-B1 135 Stains 205, 206, 207 Statins 529, 530–1 non-alcoholic steatohepatitis 1053, 1054 STAT transcription factors 30 Stauffer’s syndrome 1076–7 Stavudine 525 Steatohepatitis 218 Steatorrhea 811 Steatosis 278, 512–13, 515–16, 585 genes influencing severity of 595 macrovesicular 516 microvesicular 512–13, 1036–7 mixed micro/macrovesicular 512–13, 516 Stellate cells 3–4, 16–17, 91, 120, 150, 154, 592 activation 92–5 chemotaxis 94 contractility 94 cytokine release 94–5 fate of 95–6 fibrogenesis 94–5 immune phenotype 154 initiation 93–4 matrix degradation 94 perpetuation 94 proinflammatory responses 94–5 proliferation 94 retinoid loss 94 Stem cells 177–91 amniotic fluid 180 animal models 181 candidate populations 177–81 clinical targets 181–3 developing applications 187–8 fetal 179 human embryonic 177–8 transplanted cell engraftment 183–7 umbilical cord 180 Stem cell transplantation, and sinusoidal obstruction syndrome 898 Stomach, alcoholic liver disease 601 Stomatitis 600–1 Stones gallbladder 292–3 intrahepatic 293–4 Streptokinase 529 Streptomycin, acute intermittent porphyria 1400 Streptozocin 537 Substance abuse, as contraindication to liver transplantation 935–6 Substance P 8 Succinimides, acute intermittent porphyria 1400
1513
Index
Succinylcholine, acute intermittent porphyria 1400 Sudden cardiac death 603 Sudoxicam 532 Sulfadiazine 527 Sulfamethoxazole-trimethoprin 517 Sulfisoxazole 527 Sulfobromophthalein 245, 354 Sulfonamides 527–8 acute intermittent porphyria 1400 Sulfonylureas 522 Sulindac 532–3 primary biliary cirrhosis 814 Summerskill syndrome 77–8 Sunflower cataract 1226 Superparamagnetic iron oxide 253 Swedish porphyria see Acute intermittent porphyria Sympathetic nervous system 428–30 Symphytum officinale 573 Synergistic neurotoxin hypothesis of hepatic encephalopathy 313 Synovitis 797 Syphilis 1369 Systemic circulation, cirrhosis 425–6 Systemic lupus erythematosus 1071–2 Systemic sclerosis 797 Systemic vascular resistance 453
Tacrine 521 Tacrolimus graft-versus-host disease 869 and hepatitis C recurrence 984 liver transplantation 964 Tamarind 552 Tamoxifen 517 and non-alcoholic steatohepatitis 1042, 1054 T-cell-depleting agents, liver transplantation 963–4 T-cells autoreactive 150 cytotoxic 156 tolerance 150, 155 Telangiectasia 917 Telithromycin 527 Telomerase, in hepatocellular carcinoma 170 Temafloxacin 505 Tenecteplase 529 Tenofovir disoproxil fumarate 525 hepatitis B 653 Terazosin 529 Terbinafine 523 Terfenadine 505 Terlipressin, acute variceal bleeding 367 Tetrachlorodibenzodioxins 564 2,3,7,8-Tetrachlorodibenzo-p-dioxin 562 Tetrachloroethane 562 1,1,2,2-Tetrachloroethane 567 Tetrachloroethylene 562, 567 Tetracycline 527 acute intermittent porphyria 1400 primary biliary cirrhosis 814 Tetracyclines 528 Thalidomide 505 graft-versus-host disease 869 primary biliary cirrhosis 813, 814
1514
THCA syndrome 1270, 1310–11 biochemical features 1311 clinical features 1310 genetics 1311 laboratory findings 1310 pathology 1311 prognosis 1311 therapy 1311 Thiabendazole 523, 524 Thiazolidinediones 522, 1052 Thioguanine 517 6-Thioguanine 535 and sinusoidal obstruction syndrome 897 Thiopurine methyltransferase 59–60 Thioridazine 517 Thorotrast 517 Thrombin 490 Thrombocytopenia 493 Thrombocytopenic purpura 797 Thrombolytic agents 528 Thrombophilia, and portal vein thrombosis 907 Thrombosis 880–1 Thrombotic disorders 494 Thyroid disease 671 Thyroid hormone uncoupling protein 167 Ticardillin 527 Ticlopidine 529 Ticrynafen 505 Tight junctions 68 TIMPS 91 Tinzaparin 529 TIPS see Transjugular intrahepatic portosystemic shunt Tirofiban 529 Tissue factor pathway inhibitor 490 Tissue hypoxia 460 Tissue inhibitors of MMPs see TIMPS Tissue plasminogen activator 494 TNF see Tumor necrosis factor TNF-R-associated factor 2 28 TNF receptor-associated death domain 28 TNF-related apoptosis-inducing ligand (TRAIL) 120 Tobacco and hepatocellular carcinoma 1111 and porphyria cutanea tarda 1408 Tolcapone 521 Tolfenamic acid 532 Toll-like receptors 117, 150 Tolmetin 532 Topirimate 521 Totipotency 177 Toxic bile acids 1090–1 Toxic rapeseed oil 570 Toxins, metabolism 3 Toxoplasma 787 Toxoplasmosis, intrauterine infection 1368 Trans-activation 167 Transcutaneous liver biopsy 195–8 bacteremia 198 experience 196 number of passes 196 Tru-cut versus aspiration needle 196–8 Transferrin receptor-2 mutation 1248 Transferrin saturation 1243 Transforming growth factor-a 27–8 Transforming growth factor-b 32, 154
Transitory tyrosinemia of the newborn 1295 Transjugular intrahepatic portosystemic shunt 297–307, 445 acutely bleeding patients 301 ascites 341–2 bleeding varices 373 Budd-Chiari syndrome 303 cirrhotic ascites 302–3 complications 299–300 contraindications 298–9 hepatic hydrothorax 303 hepatopulmonary syndrome 304 hepatorenal syndrome 303 indications for 300 predictors of survival 298–9 prevention of rebleeding 302 primary prevention of variceal bleeding 300–1 procedure 297–8 secondary prevention of esphageal variceal bleeding 301–2 secondary prevention of gastric variceal bleeding 302 sinusoidal obstruction syndrome 901 veno-occlusive disease 303 Transvenous liver biopsy 198 Treponema pallidum 758 Treprostinil 529 1,1,1-Trichloroethane 567 Trientine, Wilson’s disease 1230, 1231 Trimethoprim-sulfamethoxazole 527 drug-induced liver injury 783 Trimipramine 517 2,4-Trinitrotoluene 562 Trinitrotoluene 565 Trippelennamine 517 Troglitazone 505 Trovafloxacin 527 TR- rat 1474 Tru-cut needle 196–7 TT virus 729 Tuberculosis 757–8 granuloma formation 772 HIV patients 786 Tumor necrosis factor-a 28–30, 537 alcoholic liver disease 584, 588–9 and insulin resistance 1039 survival factors 594 Tumors 226–8 benign see Benign liver tumors bile duct 1367 cholangiocarcinoma see Cholangiocarcinoma focal nodular hyperplasia see Focal nodular hyperplasia hepatocellular adenoma see Hepatocellular adenoma hepatocellular carcinoma see Hepatocellular carcinoma and porphyria 1417 surgery 1169–78 extended resection 1172–3 intraoperative ultrasound 1171 neoadjuvant chemotherapy 1174 portal vein embolization 1173–4 radiofrequency 1175–6 repeat hepatectomy 1174–5 results and prognosis 1174–6
Index
Tumors—cont’d surgery—cont’d segmental liver resection 1171–2 two-step hepatectomy 1174 vascular isolation 1169–71 clamping of hepatic pedicle 1169–70 hemihepatic vascular clamping 1169 inflow and outflow control 1170–1 segmental vascular clamping 1171 see also Neoplasms Tumor suppressor genes 169 Typhus 759 Tyrosine metabolism disorders 1293–8 hereditary tyrosinemia 1296–8 hypertyrosinemia secondary to liver disease 1295 persistent hypertyrosinemia 1295 transitory tyrosinemia of the newborn 1295 tyrosine metabolic pathway 1293–5 tyrosinosis 1295 Tyrosinemia 1270 Tyrosinosis 1295
UDP-glucuronosyltransferase 1 61–2 Ulcerative colitis and colorectal carcinoma 831–2 and primary biliary sclerosis 831–2 Ultra-rapid metabolizers 56 Ultrasound extrahepatic biliary atresia 1361 fatty liver disease 1045 focal nodular hyperplasia 1153 hemangioma 1148 hepatocellular adenoma 1159–60 high-intensity intraductal 1143 intraoperative 1171 portal venous system 350–1 Wilson’s disease 1228 Ultrasound-guided fine-needle aspiration 199, 201 dissemination of malignant cells 201 hemorrhage 201 specificity 201 Umbilical cord blood 180 Umbilical hernia 336 Umbilical vein 349 catheterization 350 Unsaturated iron-binding capacity 1243–4 Urea cycle enzyme defects 1270 Urokinase 529 Ursodeoxycholic acid graft-versus-host disease 869 hepatic fibrosis 102 non-alcoholic steatohepatitis 1053 primary biliary cirrhosis 813 with colchicine 813 with corticosteroids 813 with methotrexate 813 primary sclerosing cholangitis 843 Uveitis 797
Vaccines see Immunization Valerian 553, 557
Valproic acid 520 acute intermittent porphyria 1400 Vancomycin, hepatic encephalopathy 321 Van den Bergh reaction 238 Vanishing bile duct syndrome 229, 970 VAP-1 153 Variagate porphyria 1413–14 clinical manifestations 1413 etiology and pathogenesis 1413–14 history, definition and prevalence 1413 laboratory evaluation and diagnosis 1414 treatment 1414 Varicella zoster virus 727–8, 1369 Varices ectopic 365 esophageal/gastric see Esophageal/gastric varices Vascular endothelial growth factor 92, 185 Vascular isolation 1169–71 clamping of hepatic pedicle 1169–70 hemihepatic vascular clamping 1169 inflow and outflow control 1170–1 segmental vascular clamping 1171 Vascular resistance 356–8 Vasoactive intestinal peptide 8 Vasoconstrictors 444–5 Vasopressin, acute variceal bleeding 367 Veno-occlusive disease graft-versus-host disease 867 TIPS 303 transjugular intrahepatic portosystemic shunt 303 see also Sinusoidal obstruction syndrome Ventilatory support, acute liver failure 398 Verapamil 529 prevention of variceal bleeding 370 Veress needle 198, 199 Vertical transmission of hepatitis 1005 Very low density lipoproteins, synthesis and secretion 1037 Vinca alkaloids 536 Vinyl, causing portal hypertension 58 Vinyl chloride 517, 562, 565 Viral clearance 120 Viral infections adenoviruses 728 cytomegalovirus see Cytomegalovirus Epstein-Barr virus see Epstein-Barr virus hepatitis see Hepatitis herpes simplex virus 727, 1368 human herpes virus-6 728 human herpes viruses-7 and -8 728 human parvovirus B19 728 neonatal hepatitis 1367–9 and primary sclerosing cholangitis 823–4 SEN virus 730 severe acute respiratory syndrome 729 TT virus 729 varicella zoster virus 727–8, 1369 see also individual viral infections Viral packaging inhibitors 655 Virions hepatitis B 111, 115–16 hepatitis C 129 Virus-like particles 694 Visceral angiography 350 Vitamin A, excess 517
Vitamin E 101 non-alcoholic steatohepatitis 1053 Vitamin K 494 deficiency 240 Vitiligo 797 Volume expansion 444–5 Von Meyenburg complex 257 Von Willebrand factor 489 Von Willebrand’s disease 199 Voriconazole 523
Warfarin 529 Waterlily sign 285 Water retention 421–2 assessment of renal water excretion 422 clinical consequences 421–2 impaired renal water handling 421 management 443 see also Ascites; Sodium retention Wedged hepatic vein pressure 298, 336 Weight reduction 1052 Wernicke-Korsakoff syndrome 603–4 West Haven Scale 317 Wilson’s disease 1221–38, 1270 acute liver failure 389 animal models 179 biochemical diagnostic features 1226–8 clinical diagnostic features 1224–6 hepatic presentation 1224–5 neurological presentation 1225 ocular features 1225, 1226 psychiatric presentation 1225 complications 1333 copper disorders 1221 copper pathway 1221–2 diagnosis by mutation analysis 1228–9 diagnosis of presymptomatic sibs 1229–30 epidemiology 1222 fulminant liver failure 1232 hemolytc anemia in 492 hepatocellular carcinoma 1111 histopathology 210, 1228 history 1221 imaging 278 imaging studies 1228 liver cell therapy 183 pathogenesis 1222–4 basic defect 1222–3 consequences of copper storage 1224 copper transport and homoeostasis in hepatocytes 1223–4 and portal hypertension 361 in pregnancy 1006 pregnancy 1232–3 prognosis and natural history 1234 specific mutations and clinical features 1229 treatment 1230–2 antioxidants 1232 chelation 1230–1 dietary management 1232 gene therapy 1231 induction of metallothioneins and interference with absorption 1231–2 liver transplantation 1231 Wnt 170–3
1515
Index
Wolman’s disease 1270, 1299–301 biochemical characteristics 1300 clinical features 1299 differential diagnosis 1301 enzymatic abnormalities 1299 genetics 1301 laboratory findings 1299 pathology 1300–1 treatment and prognosis 1301
Xanthogranulomatous cholangiopathy 216
1516
Xenobiotics activation of 594 bioactivation 508–11 glutathione 508–9 oxidative stress and free radical reactions 510–11 hepatic transport 63
Zalcitabine 525 Zellweger’s syndrome 1270, 1308–10 biochemical features 1309 clinical features 1309
Zellweger’s syndrome—cont’d genetics 1310 laboratory findings 1309 molecular basis 1310 pathology 1309 prognosis 1310 Zidovudine 525 Zieve’s syndrome 492 Zinc repletion 323 Wilson’s disease 1230, 1231–2 Zone 3 of Rappaport 360 Zygomycetales 761
